PROCEEDINGS OF THE California Academy of Sciences FOURTH SERIES Vol. XVIII SAN FRANCISCO Published by the Academy 1929-1930 COMMITTEE ON PUBLICATION Col. George C. Edwards, Chairman Dr. C. E. Grunskt Dr. Barton Warren Evermann, Editor CONTENTS OF VOLUME XVIII 1. A New Species of Corambe from the Pacific Coast of North America. By Frank M. MacFarland and Charles H. O'Don- oghue. PubHshed January 29, 1929 1 Plates 1-3 2. A New Bird Family (Geospizidae) from the Galapagos Islands. By Harry S. Swarth. Published January 29, 1929 29 3. A Contribution to our Knowledge of the Nesting Habits of the Gol- den Eagle. By Joseph R. Slevin. Published January 29, 1929 . . 45 Plates 4-7 4. Marine Miocene and related Deposits of North Colombia, By Frank M. Anderson. PubHshed March 29, 1929 73 Plates 8-23 5. A New Pecten from the San Diego Pliocene. By Leo George Hert- lein. PubUshed April 5, 1929 215 Plate 24, Figs. 10-11 6. A New Species of Land Snail from Kern County, California. By G. Dallas Hanna. PubHshed April 5, 1929 217 Plate 24, Figs. 7-9 7. A New Species of Land Snail from Coahuila, Mexico. By G. Dallas Hanna and Leo George Hertlein. Published April 5, 1929. ... 219 Plate 24, Figs. 5, 6 8. Some Notes on Oreohelix. By Junius Henderson. Published April 5, 1929 221 Plate 24, Figs. 1-4 9. Notes on the Northern Elephant Seal. By M. E. McLellan David- son. Published April 5, 1929 229 Plates 25, 26 10. On a Small CoUection of Birds from Torres Strait Islands, and from Guadalcanar Island, Solomon Group. By M. E. McLellan Davidson. PubHshed April 5, 1929 245 11. The Generic Relationships and Nomenclature of the California Sardine. By Cari L. Hubbs. Published April 5, 1929 261 12. The Faunal Areas of southern Arizona: A Study in Animal Dis- tribution. By Harry S. Swarth. PubHshed April 26, 1929 267 Plates 27-32 13. The Escallonias in Golden Gate Park, San Francisco, California, with Descriptions of New Species. By AHce Eastwood. Pub- Hshed September 6, 1929 385 Plate 35 14. Studies in the Flora of Lower California and adjacent Islands. By Alice Eastwood. Published September 6, 1929 393 Plates 33, 34 15. Drepania, A Genus of Nudibranchiate MoUusks new to California. By F. M. MacFarland. Published October 4, 1929 485 16. Some Upper Cretaceous Foraminifera from Coalinga, California. By J. A. Cushman and C. C. Church. Pubhshed October 4, 1929 497 Plates 35-41 17 Report of President of Academy. By C. E. Grunsky 531-541 18. Report of Director. By Barton Warren Evermann 542-579 Report of Treasurer. By M. Hall McAllister 580-586 Index 587 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Fourth Series Vol. XVIII, No. 1, pp. 1-27, plates 1-3 January 29, 1929 A NEW SPECIES OF CORAMBE FROM THE PACIFIC COAST OF NORTH AMERICA BY FRANK M. MacFARLAND Stanford University AND CHARLES H. O'DONOGHUE University of Edinburgh The Nudibranch genus Coramhe forms a group concerning whose structure, life history, affinities and distribution, much remains yet to be learned. The general features of its or- ganization seem to ally it to the phanerobranchiate Dorids, and in that group more especially to the Goniodorididse. It has been placed by Bergh in his System (1892) in a separate family, the Corambidse, and he indicates its probable close re- lationship to the little known, older genera Hypohranchuua A. Adams, and Doridella Verrill, rather characteristically re- ducing them to synonymy with his own, later genus. While these two are unknown from an anatomical point of view, it is reasonably certain that they should be united with Coramhe in the same family at least. In doing this we recognize the pri- ority and correctness of the name proposed by P. Fischer (1883), and use it instead of the later one by Bergh. The family diagnosis given by Fischer has been slightly modified in the light of later information than was then available. Should future studies establish the generic identity of Corambe January 29, 1929 2 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. with either or both HypohranchicBa and Doridella, the name given by Bergh would, of course, be cancelled in favor of the earlier one. Family Hypobranchi^id^ P. Fischer, 1883. Fischer, P., 1883, Manuel de Conchyliologie, Fasc. VI, p. 530. Body, notaeum, and rhinophores doridiform, branchiae pos- terior, below the notaeum margin and above the foot; anus median, posterior, between notaeum and foot; reproductive openings anterior on right side; radula narrow, multiserial. Genus 1. Hypobranchiaea A. Adams, 1847. Adams, A., 1847. Proc. Zool. Soc. London, pp. 23-24. Type: Hypohranchicsa fusca A. Adams. Yellow Sea. Genus 2. Doridella Verrill, 1870. Verrill, A. E., 1870. Amer. Journal Science and Arts, I, p. 408. Type: Doridella obscura Verrill. Vineyard Sound; Long Island Sound. Genus 3. Corambe Bergh, 1871. Bergh, R., 1871. Verh. k. k. zool.-bot. Ges. Wien, XXI, pp. 1293-1294. Type: Corambe sargassicola Bergh. Sargasso Sea. Genus 4. Corambella Balch, 1899. Balch, F. E., 1899. Proc. Host. Soc. Nat. Hist., XXIX, p. 151. Type: Corambella depressa Balch. Cold Spring Harbor, Long Island. The present paper deals with the third of this list of genera, being a description with anatomical details of a new species of Corambe from the Pacific Coast of North America, for which Vol. XVIII] MACFARLAND & O'DONOGHUE—NEW SPECIES CORAMBE 3 the name Corambe pacifica is here proposed. Our grateful acknowledgments are due to Professor Walter K. Fisher, Di- rector of the Hopkins Marine Station, Pacific Grove, Cali- fornia, for the free use of the facilities afforded by that lab- oratory during the prosecution of the greater portion of this study. We are also greatly indebted to Mrs. Olive H. Mac- Farland for her generous cooperation in the preparation of the fisfures which illustrate this account. ^fc>" Corambe Bergh, 1871 Corambe Bergh, R., 1869. Bidrag til en Monographi af Phyllidierne. Naturh. Tidsskrift, 3 R, 5 B, p. 359; footnote. Corambe Bergh, 1871. Beitrage zur Kenntnis der Mollusken des Sar- gassomeeres. Verh. d. k. k. zool.-bot. Gesellschaft Wien, Bd. XXI, pp. 1293-1297, Taf. XI, Fig. 21-27, Taf. XII, Fig. 1-11. Corambe Kerbert, C, 1886. Over het Geslacht Corambe Bergh. Tijd- schrift der Nederlandsche Dierkundige Vereeniging, 2 Sen, D. 1, Afl. 2, pp. 5-6. (Abstract in Bull. Sci. du Nord, 2 Ser., 9, 1886, pp. 136-138.) Corambe Fischer, P., 1888. Note sur la presence du genre Corambe Bergh, dans le bassin d'Arcachon (Gironde). Bull. Soc. Zool. France, T. 13, No. 9, pp. 215-216. Corambe Fischer, H., 1889. Note preliminaire sur la Corambe testudinaria. Bull. Soc. Zool. France. T. 14, No. 10, pp. 379-381. Corambe Fischer, H., 1891. Sur I'anatomie du Corambe testudinaria. C R. Ac. Sci. Paris, CXII, pp. 304-307. Corambe Fischer, H., 1891. Recherches anatomiques sur un Mollusque appartenant au Genre Corambe. Bull. Sci. de la France et de la Belgique. T. XXIII (Ser. IV, Vol. II), pp. 358-398, PI. IX-XIII. Corambe Fischer, H., 1896. Note sur la distribution du Genre Corambe. Jour. Conchyl. Vol. XLIII, pp. 235-236. Corambe Bergh, R., 1892. System der Nudibranchiaten Gasteropoden. Wiesbaden. Semper's Reisen im Archipel der Philip- pinen. Wissenschaftliche Resultate. Malacologische Un- tersuchungen, Bd. Ill, H. 18, pp. 166-168. Corambe Vayssiere, A., 1901. fitude comparee des Opistobranches des Cotes Frangaises de I'Ocean Atlantique et de la Manche avec ceux de nos Cotes Mediterraneennes. Bull. Sci. France et Belgique, T. XXXIV, p. 296. Corambe Vayssiere, A., 1913. Mollusques de France et des regions voisines. T. I., Paris, p. Z62). 4 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Body doridiform, oval, depressed; notseum somewhat con- vex, its margin wide, flattened, rounded in front, deeply notched in the median line behind, everywhere extending be- yond the foot; rhinophores retractile within sheaths, the stalk bearing an inner pair of wing-like, lateral expansions, and surrounded by an outer sheath, free above, united to the stalk below, and deeply cleft or entirely free behind; foot emar- ginate in front, rounded behind, smaller than the notseum, which completely conceals it. Branchiae posterior, of a few separate, pinnate plumes sym- metrically arranged on either side of the median line between the notseum and the foot ; anus median, posterior, between the two groups of branchial plumes ; tentacles short, nearly con- cealed by the notaeum. Pharyngeal bulb armed with two lateral thickenings at the buccal aperture; radula narrow, its rhachis naked, the inner- most, lateral tooth large, bearing a denticulate hook, the outer laterals few, small, with a simple hook; buccal ingluvies con- nate with the pharyngeal bulb. Glans penis unarmed. The genus Corambe is first mentioned by Bergh in 1869 in a brief footnote in a paper upon, the Phyllidiidse. The de- scription, "a dorid-like mollusk with strong mandibles, with numerous (24) rows of teeth, with four laterals upon either side of a median series," can scarcely be taken as an adequate diagnosis of the genus, since there are neither mandibles nor median teeth present, nor could the form be identified by this statement alone. In 1871, however, the same author published a more extended diagnosis, based upon a study of a single specimen of the genotype, Corambe sargassicola Bergh, taken upon drifting seaweed in the Central Atlantic in 42° 50' N. Lat,, and 46° 20' W. Long. The description is in many de- tails quite inaccurate and incomplete, probably owing to the lack of material. A second species, Corambe batava, from the Zuider-Zee, was described by Kerbert in 1886 in a very fragmentary manner. In 1889 H. Fischer described a third species, Corambe teshidinaria from the Bay of Arcachon, and in 1891 published an excellent anatomical account, which forms the actual basis of our knowledge of the genus. In the opinion of Vayssiere (1913), these three species are Identical, forming the single species Corambe sargassicola Bergh, which Vol. XVIII] MACFARLAND & O'DONOGHUE—NEW SPECIES CORAMBE 5 is not at all unlikely, though the accurate information respect- ing the species described by Bergh and by Kerbert, necessary to certainty in this regard, is lacking. The new species of Coranihe discussed in the present paper, differs markedly from the ones previously described. It has been taken by the authors in two widely separated localities, Monterey Bay, California, and at Nanaimo, British Colum- bia. In each instance the habitat is the same : Memhranipora colonies upon the large kelps and Zoster a, from which sur- roundings the minute animal is scarcely distinguishable. Its resemblance to a young colony of the bryozoan of similar size is even more perfect. The species of the genus at present may be listed as follows : 1. Coranihe sargassicola Bergh, 1871. 2. C. hatava Kerbert, 1886. 3. C. testudinaria Fischer, 1889. 4. C. paciUca MacFarland & O'Donoghue, new spe- cies, in which summary the first three are assumed to be valid and distinct species, in the absence of positive knowledge to the contrary. Corairlbe pacifica MacFarland & O'Donoghue, new species. Animal (PI. 1, fig. 1) elliptical, flattened, disk-like, slightly arched in the central region of the body, the notseum every- where extending beyond the foot, its margin wide and thin, with a deep, median, circular notch behind, elsewhere entire. Foot rounded equally in front and behind, its anterior mar- gin, beneath the head, with a deep, median notch revealing the mouth in the angle. Head small, covered entirely by the notseum, its angles pro- longed into short, blunt tentacles, directed outward and for- ward, their tips showing beyond the notaeum margin, when the animal is crawling freely. Rhinophores retractile into low, entire, thin-margined sheaths, the blunt, tapering tip of the stalk projecting above an incomplete, inner envelope, to which it is attached in the anterior, median line below, above free, the sheath-like expan- sion sloping rapidly downward behind to the rear of the stalk, with which it merges. Within this envelope the stalk bears a 5 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. lower, plate-like expansion on either side, revolute backward, and inserted behind, above the more external sheath; a low, keel-like ridge, or plate, on the median, posterior side of the stalk. Anal opening posterior in the median line, immediately be- low the notch of the notasum margin; close to it at its right and slightly above is the single, renal opening, a minute pore. Reproductive openings three, close together, far forward on the right side, between the notseum and the foot. Branchiae a series of simple, pinnate plumes, ranging in number in mature individuals from six to 12 or 14 on each side, decreasing in size from behind forward, borne on either side of the anal opening, between the foot and the notasum, and limited to the posterior third of sides of body. A single, median plume is usually situated immediately above the anus. Lamellae of longest plumes 10 to 20 in number, opposite in arrangement upon sides of horizontally flattened shaft; at the insertion of the branchiae a series of large, simple, alveolar glands, mostly alternating with the bases of the plumes, and co-extensive with them. Color of dorsum a pale, translucent gray ground, the central area marked out by the pale, yellow-orange liver showing through the integument. Surrounding this central area is a whitish zone, determined largely by the foot showing through from below. Outside this zone and equal to it in width is the nearly transparent notaeum margin. This marginal zone is marked with irregular, continuous and discontinuous lines of clear baryta-yellow, arranged radially. Toward the center of the dorsum these lines become broken up into dots of color, and are more irregularly scattered. These radial lines with their cross connections resemble the walls of the zooecia of Meinbranipora to a very marked extent. Between the super- ficial, baryta-yellow markings are larger and smaller flecks, in general radial in arrangement, and lying deeper in the in- legument. These are largest and most numerous in the second zone, and become smaller and more rounded in the central area. The central and major portion of each fleck is terra cotta in color, and is usually edged with an incomplete line of black. Around the rhinophore bases they may form an almost continuous ring, but are usually clearly separate. Scattered Vol. XVIII] MACFARLAND & O'DONOGHUE—NEW SPECIES CORAMBE J small, black flecks may also occur in the median area. In darker specimens the terra cotta spots are larger and more numerous, especially in the median region, their borders deepening to a greenish color, where not black. Foot clear gray, with a narrow, white, marginal line. Rhinophores clear, translucent gray, the sheath either the same or with a few small spots of terra cotta, baryta-yellow, or black. Radula formula 38-40 x (4-5 -f 1 -f + 1 + 4-5). Median tooth wanting. First lateral large, compressed, consisting of a slightly curved hook rising from the anterior angle of a large, erect base, the hook bearing three to seven denticles upon its inner margin. Upper posterior angle of the base of the first lateral thickened and pointed, forming a second, minor hook directed backward. Inner face of the base with a low, recurved, wing-like lamina, arising behind and below the lowermost denticles, and curving downward to the insertion of the base. Outer, lateral teeth, usually four, decreasing in size progressively outward, each consisting of a broad, rounded base bearing a slightly curved, simple, pointed hook. Rows of teeth not exactly opposite each other in the lateral halves of the radula. Pleural ganglia not fused with the cerebral ones, but united to them by short connectives. Length in life up to 13 mm., width up to 10 mm. Habitat: Upon brown kelps, mainly Macrocystis pyrifera (Turn.) Ag. and Nereocystis luetkeana P. & R., and upon Zostera marina L., bearing incrustations of Memhranipora villosa Hincks colonies, upon which the mollusks feed. Mon- terey Bay, California. Nanaimo, British Columbia. Holotype: No. 634, Mus. Calif. Acad. Sci., collected May 21, 1928, by F. M. MacFarland, in Monterey Bay, Pacific Grove, California. Paratypes are deposited in the U. S. National Mu- seum, the British Museum, the Biological Station at Nanaimo, B. C, the Hopkins Marine Station, Stanford University, and in the authors' private collections. The careful study of Coramhe testudinaria by H. Fischer (1891) renders unnecessary a detailed account of the anatomy of this new species, save as regards certain features of pro- nounced difference found by us. Detailed dissections were 8 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. made and supplemented by serial sections from material im- bedded in paraffin and in celloidin and stained in various ways. Habitat: The animals are seldom found separated from the Memhranipora colonies, and then probably through acci- dent. They have been seen actively feeding upon the colonies of Memhranipora villosa, which seem to be their chief food. One of us (O'Donoghue, 1926) has described in detail the ravages of Coramhe upon the bryozoan colony. "When young, even less than one mm. long, this mollusk has been seen inside the zooecium, from which it has eaten all the living matter. A more common point of attack, and the only one by larger Coramhe, is the growing edge of the colony which is either not protected by a chitinous covering, or else by one so thin that it affords no protection. This method of wounding produces a very characteristic indentation of the growing edge. If of short duration, it is surrounded by the growing zooecia, and all that is left of the point of injury is an area looking like a misshapen zooecium. However, if the attack is made at one place by several small Coramhe, or the animal re- mains a long time in the same place and grows considerably, the injury will be correspondingly greater and perhaps perma- nent. So prevalent are these attacks that it is rare to find under natural conditions an uninjured colony of, say, 10 mm. in diameter." But few traces of diatoms in the alimentary canal, such as Fischer cites, have been found, though they may be present at certain times of the year. Since the hard parts of the bryozoan do not appear to be eaten, it is not sur- prising to find no trace of them. The animals are sluggish, except when removed from the surface of the host, when they tend to move around rather actively, until they find their usual surroundings again. External characters : The general color of the dorsal aspect is a clear, translucent gray, veined and dotted with pale yellow or greenish yellow. The central area of the notseum is thickly set with light garnet-red or terra cotta spots, located deep be- low the surface. The edges of these flecks usually appear deeper in color than the center, at times becoming greenish or black. Intermingled with these spots are flecks of black and baryta-yellow. Toward the margins the baryta-yellow flecks tend to unite into irregular, radial lines, sometimes in pairs, Vol. XVIII] MACFARLAND & O'DONOGHVE—NEW SPECIES CORAMBE g but usually single. Occasionally a series of irregular, longi- tudinal lines is developed in the median region. The foot is of clear gray, with a narrow, white, marginal line. Its central and posterior region is occupied by a vaguely defined, darker, greenish area, due to the denser viscera showing through the integument. The branchial plumes are transparent gray with a few scattered flecks of garnet upon them. The young forms have no color pattern, but are a pale, almost transparent gray, with the black eye spots clearly showing. Notaeum : The notaeum is very thick, slightly less so in the median area than at the sides. In general, its surface is smooth, or slightly roughened, the color markings exaggerat- ing the impression of a tuberculate surface. The low, cuboidal epithelium of the dorsal surface secretes a thick, cuticular layer, which shows distinct stratification in sections. Its thick- ness varies markedly in different specimens, sometimes being merely a moderate layer (Fig. 4), in others presenting a thickness six to 10 times the height of the cells producing it (Fig. 3). Without doubt the dorsal cuticle of the notaeum is periodically shed as a continuous sheet, and renewed, lines of cleavage parallel to the surface being shown in sections, and the detached, entire cuticle is frequently found in the aqua- riums, while animals still covered with the partially free cuti- cle are not uncommon. This phenomenon was also noted by Fischer (1891) in C. testudinaria, and appears to be without a parallel in other nudibranchs. Imbedded in this cuticular layer are abundant, conical, spine-like structures, in sections staining more strongly than the surrounding cuticle, and more resistant than it. These are the products of special, large, epithelial cells, occurring at intervals in the epidermis, each one of which secretes above it this cuticular modification. In those cases in which the general cuticle is but thin, these spines project freely above the surface, giving it a minutely rough- ened texture. Where the cuticle has become much thickened, two or three such spines may be seen in sections, one above the other, the lowermost and smallest resting upon the cell which has produced the series, the superimposed ones elevated above it in the order of their formation, and being cast off by the successive moultings of the cuticle, probably associated with growth periods (Fig. 3). A similar structure has been IQ CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. described by MacFarland (1918) for the palatal spines of a Tectibranch, Dolahella agassizii MacF. Toward the margin of the notseum these spines are increased in number, and are often closely crowded, while toward the central areas they are less numerous. Scattered among the cuticle-secreting cells are numerous, giant, mucous cells, the pear-shaped cell-body lying below the general epithelium and prolonged as a duct to its surface, from whence it is continued by a slender canal through the thickness of the cuticle. The wide notaeum margin conceals the head entirely. The angles of the latter are prolonged into short, blunt tentacles somewhat triangular in form. The tips of these tentacles may project beyond the notseum margin when the animal is crawl- ing freely, or may be entirely concealed. The same is true of the tips of the gills at the posterior end. The rhinophores (Fig. 2) are retractile within low, entire sheaths. The axis of the rhinophore is prolonged into a taper- ing, blunt tip, and bears two pairs of revolute lamellae. The outer pair of these (o) are united into a sheath-like structure, fused in front lengthwise to the greater portion of the stalk, being free only at the upper one-fourth, there encircling the rhinophore in a collar-like form, the margins curving down- ward around to the posterior face of the stalk, where they terminate a short distance apart. Within this outer invest- ment the second pair of laminae (i) are inclosed. Each arises from the side of the stalk as a thin plate curving backward, united below with the stalk, and their free, posterior margins terminate above those of the outer pair. In the median line, behind, a single, thin, keel-like ridge extends from near the tip of the rhinophore downward, dying away as the stalk of the latter enlarges toward the bottom of the inner, lateral pair of lamellae. Since these laminae are attached to the stalk of the rhinophore, are retracted with it, and bear the same rela- tion to it as the plates of the common, perfoliate clavus of the nudibranch rhinophore, they cannot be termed sheaths, that designation being restricted to the elevated margin of the opening of the notaeum, into which the rhinophore is with- drawn. This misuse of terms is committed by Bergh (1871, 1892), and also by Fischer (1891). Vol. XVIII] MACFARLAND & O'DONOGHUE—NEW SPECIES CORAMBE l\ Branchiae: The branchiae (Fig. 5) are located at the pos- terior end of the body, attached to the under surface of the notaeum above the foot, and arranged symmetrically in a sin- gle, horizontal row on either side of the anus, and usually united above it by a single plume. They vary in number on each side up to 12 or 14 in the largest individuals. They are simply pinnate plumes consisting of a flattened, tapering axis, upon either side of which is borne a series of oppositely ar- ranged, respiratory lamellae, varying in number up to 20. In the smallest, most anterior gills the plates are reduced to one or two, or the whole organ may be represented by the rhachis alone as a slight projection from the body wall. The branchiae increase progressively in length and in the number of lamel- lae toward the posterior end of the animal, the largest being usually the pair adjacent to the anal opening, or the second or third pair from it. The series extends forward not more than one-third of the length of the foot. In a specimen of 6.8 mm. body length the length of the plumes ranged from 0.25 mm. for the shortest to 0.95 mm. for the longest, which were the third pair from the posterior end of the series. In these last the lamellae reached 20 in number. In Corambe testudinaria, as described by Fischer, the number of branchiae is fewer, four to seven, and the number of lamellae on each side of the rhachis is much fewer (up to four). The lamellae are also arranged alternately upon the sides of the stalk, whereas in the present species they are opposite. The most anterior gill is stated by Fischer to be located nearly midway of the body length, which is decidedly farther forward than in our species, despite the greater number of plumes present in the latter. The tips of the posterior gills are visible at times beyond the notaeum margin, but ordinarily they are concealed, save below the median notch. A kind of respiratory movement has been noted in animals under observation in aquariums. The pos- terior end of the mantle is raised well away from the sub- stratum and the gills protruded to their fullest extent at ir- regular intervals. This reaction occurs more frequently when the water has been standing for some time. Movement of minute particles suspended in the water indicate a strong cur- rent laterally toward the sides of the body, beneath the no- January 29, 1929 12 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. taeum margin, and backward past the gills and through the elevated, median, dorsal notseum notch. Bergh (1871) describes and figures in his Figs. 23 and 24 of Plate XI and Fig. 1 of Plate XII for Corambe sargassicola Bgh., an entirely different type of gill, made up on either side of a group of thin, horizontal lamellae, 13 to 15 in number, obscurely arranged in an upper, wider and longer, and a lower, narrower and shorter set. No intimation of a pinnate arrangement is given, though later (1892, p. 165-166) he in- dicates this as a generic character, evidently following the more reliable observations of Fischer. Just above the line of insertion of the branchiae is a series of simple, alveolar glands, most of which alternate in position with the insertion of the gill stalks (Fig. 5, g). They are spherical in form, and are composed of large, clear, pyramidal cells extending from the basement membrane almost to the opening of the gland, leaving but a small lumen (Fig. 6). Each gland opens to the external surface through a very short and narrow duct near the base of the gill. No trace of the single, branched, median gland, described and illustrated by Fischer, is here present, though it is probably represented by this series of simple glands coextensive with the branchial in- sertion. When floating at the surface, the animal produces a very abundant, mucous secretion. Structurally, these glands appear to be of a mucous nature, but whether they contribute largely to this film of secretion or not has not been determined. As a rule such secretions are produced by the pedal glands to aid in adhesion or floating. Alimentary tract. The mouth is revealed in ventral view by the triangular notch in the anterior margin of the foot. The external lips are rather thick and glandular, and lead into a short, oral tube, the cuticle of which is but slightly thickened. The inner lips, surrounding the opening into the cavity of the pharyngeal bulb, are but slightly developed and show a moder- ate thickening of the cuticle on the sides, and ventrally extend- ing into the bulb. No clearly differentiated, lateral plates, such as are described by Bergh (1892), can be made out. The pharyngeal bulb (PI. 3, fig, 11) bears a thick-walled, muscular crop (c) above, such as is characteristic for the Goniodorid- idse. The posterior part of the radula sack forms a prominent Vol. XVIII] MACFARLAND & O'DONOGHUE—NEW SPECIES CORAMBE 13 median ridge {r. s.) upon the hinder face of the bulb. The radula is very small, attaining a length of but 0.25 mm. in a large specimen. Its dorsal surface is deeply grooved longi- tudinally in the median line. There are from 38 to 40 trans- verse rows of teeth present in large specimens. The half rows are not exactly opposite each other in the two sides of the radula, which, together with the minute size of the elements, renders the count of the rows difficult and often uncertain. The dental formula for the older part of the radula is 4+1+ 0+1+4, in the younger portion, in the sheath, the laterals are frequently increased by one, giving a formula of 5+1+0 + 1+5. The rhachis is very narrow and destitute of median teeth, the innermost lateral (Figs. 15, 16) is relatively large and quite different from the remaining ones. In form it somewhat resembles that of A cant hod oris. From a roughly quadrilateral, compressed, basal portion a strong, somewhat curved hook arises at the anterior, upper angle. The hook is nearly equal to the base in height, is directed obliquely inward and backward, and is terminated by a blunt point. On the lower half of its inner margin is borne a series of four to seven pointed denticles. From the upper half of the inner face of the base a narrow, recurved, wing-like extension (Fig. 16, w) projects downward, curving beneath the base as a ridge across to the opposite side. The posterior margin of the base is thickened, and its outer, upper angle (Fig. 15, a) forms a low, compressed, triangular hook, pointed backward. From the oldest, most anterior teeth of the radula backward, there is a steady increase in the dimensions of the teeth, but the rel- ative proportions remain about the same. In an average first lateral tooth the total height from insertion line on the base- ment membrane to the tip of the hook is 0.09 mm., while the height of the hook itself is 0.04 mm., and the greatest length of the base is 0.05 mm. The outer, lateral teeth, four to five in number (Figs. 12, 13, 14), consist of a rounded base, which is prolonged obliquely upward and backward as a simple, tapering, pointed hook, triangular in outline and rounded above, and below supported by a lamina, which dies away be- fore the tip is reached. The outer laterals tend progressively to be more and more compressed, and the fifth, when present, is usually reduced to a small, flattened plate. In length in an 14 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. average row the second lateral measures 0.03 mm., the third 0.026 mm., the fourth 0.025 mm., the fifth 0.022 mm., and the outermost 0.012 mm. The single pair of salivary glands form a compactly rounded mass lying upon the upper face of the pharyngeal bulb, at either side of the beginning of the oesophagus. They are alveolar in type, but slightly branched, and are composed of large cells, which leave but a small, irregular lumen, leading by a rather wide duct into the cavity of the bulb, lateral to the radula. Their staining character and general cytological structure indicate that the secretion is predominatingly mucous in nature. The strikingly thin-walled oesophagus, lined with ciliated, columnar epithelium throughout its extent, leads directly downward and backward to the anterior end of the stomach. Into it open at once the very wide, biliary passages of the liver. These are five in number, an anterior and a posterior lateral pair, and a single, posterior, median one, which bifur- cates into the posterior lobe of the liver. This organ presents a ventral, median, undivided portion from which project five lobes, an anterior and posterior one on either side, and a sin- gle, median, posterior one, which last shows a median, pos- terior notch externally, corresponding with the subdivision of its inner cavity. The right, anterior, paired lobe is quite small, its space being largely occupied by the anterior, genital complex, against the posterior face of which it extends as a narrow strip. Its fellow on the opposite side is large, nearly equalling the whole of the anterior genital mass in size. The posterior, lateral pair is likewise large and well developed. The walls of the liver are composed of a single layer of cu- boidal, granular cells lining the roomy lumen of the gland and the numerous short and wide sacculations opening into it. This cavity is strikingly large, with relatively simple ramifica- tions, and communicates widely with the cavity of the stom- ach, so freely in fact, that it is difficult to fix the boundaries of the anterior portion of the stomach, its contents passing readily into the cavity of the liver, where the main, digestive changes probably take place. What gives solidity and com- pactness of appearance externally to the organ, in fact, is the thick layer of the ovotestis, which invests the dorsal and lat- Vol. XVIII] MACFARLAND & 0'DONOGHUE~NEW SPECIES CORAMBE J 5 eral faces of the liver completely. Divested of this covering, the liver would present five, slightly ramified, broad and ir- regular tubes, resembling more the branched arrangement of the Aeolids rather than the compact liver of the Dorids. The median, dorsal surface of the ovotestis-liver mass is occupied by a wide depression passing its full length, in which are con- tained the stomach and intestine, the heart and pericardium, and the kidney. No "biliary cyst" can be distinguished. Between the adjacent liver lobes well developed, muscular septa unite the notseum and the foot and extend from the lateral body wall inward as far as the cleft between the lobes permits. Similar incomplete partitions are also found ex- tending obliquely inward between the sides of the pharyngeal bulb and the liver on the left, and the anterior, genital complex on the right. Dorsally, the stomach is clearly marked off, appearing as a retort-shaped sack, broadest at the left of the median line and narrowing into the rather slender intestine as it curves to the right, thence passing straight backward to the anus in the pos- terior, median line. Its wall is made up of cuboidal, ciliated cells, surrounded by a layer of circular muscle fibres and con- nective tissue. The epithelium of the intestine is the same, but its layer of muscle is very thin, and at times apparently absent. At the anus, however, the circular muscle is thickened into a well-developed sphincter, as noted by Fischer (1891). The anal opening is situated in the median line of the body at the posterior end, immediately below the notch in the notaeum. Close by, at the right and slightly above it, is the minute opening of the renal organ. Neither structure is con- spicuous externally. Nervous System: Close behind the salivary glands is the central, nervous system. The ellipsoidal, cerebral ganglia (Pis. 2, 3, figs. 8, 10, c), the largest of the group, are in contact along their inner faces, but are not fused, being connected by a distinct, broad, cerebral commissure above the oesophagus. Be- low the latter they are also connected by a delicate, sub-ceso- phageal commissure, recognizable in sections. From the an- terior portion of the cerebral ganglia are given off the nerves to the rhinophores and the eyes, each bearing an elliptical ganglion close to its origin (Figs. 8, 10, c.i, c.2), and three 15 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. other pairs (Figs. 8, 10, c.^, c.4, c.§) to the buccal tentacles, and the mouth and head region. From the fifth of these, as a basal branch, or very close to its origin, is given off the cere- bro-buccal connective, which passes beneath the oesophagus to the buccal ganglion (Fig. 10 c.b.c, b). The optic nerves are rather short, the eyes small and deeply buried below the in- tegument, behind and medial to the rhinophores. The nearly spherical statocysts lie close in the outer angle between the cerebral and pedal ganglia. They measure ca. 0.03 mm. in diameter, and contain many ellipsoidal statoliths, 0.002 mm. by 0.003 mm. in diameter. Lateral to the oesophagus are the spherical, pedal ganglia (Figs. 8, 9, 10, p), second in size to the cerebral pair, and joined to them by short, cerebro-pedal connectives (Fig. 10, c.p.c). The pedal ganglia are united below the oesophagus by the usual, well developed, pedal com- missure (Fig. 9, p. c), and also by a distinct, more slender, parapedal commissure (Fig. 9, pp. c) separated some distance from the main, pedal one. These commissures are very much shorter than those figured by Fischer (1891) for C. testudi- naria. From the pedal ganglia are given off the stout, an- terior, median, and posterior pedal nerves, distributed to the corresponding regions of the foot, the latter two either arising separately, or from a common stalk, which soon bifurcates. Immediately behind the cerebral ganglia and slightly below them are the distinct pleural ganglia (Figs. 8, 9, 10), united with the cerebral and pedal ganglia by the cerebro- pleural and pleuro-pedal connectives respectively (Fig. 10, c-pl. c, pl.-p. c). As a rule in the Nudibranchiata the pleural ganglia are fused more or less completely with the cerebral pair, there being varying degrees to which this fusion is indi- cated externally. In Corambe testndinaria Fischer (1891), figures (1. c. Figs. 20, 21, 22) the cerebral and pleural (pleuro- palleal) ganglia as fused in a common, supra-oesophageal com- plex, as indicated by an external, transverse groove, and by the cerebro-pedal and pleuro-pedal connectives, arising from the fused, ganglionic mass. In the present species, however, the separation of the pleural from the cerebral ganglia is equally clear and unmistakable. This difference in such fun- damental structures in two allied species of the same genus is Vol. XVIII] MACFARLAND & O'DONOGHUE—NEW SPECIES CORAMBE J 7 very remarkable, and appears without a parallel, so far as we are aware, in the Nudibranch literature. Uniting the pleural ganglia below the oesophagus is the vis- ceral loop (Figs. 8, 9, V. c), bearing a ganglionic enlargement a short distance from its right end, which gives rise to a single nerve, dividing into a stronger, right and a more slender, left branch. The left one bears two small ganglia at a short in terval apart, from each of which fine nerves arise and pass backward to the viscera. The right nerve breaks up into a number of fine rami, which apparently pass mainly to the reproductive organs. From the pleural ganglia themselves anterior and posterior nerves arise. On the left side the pos- terior, pleural nerve is usually single, on the right it arises as either two, separate roots (Figs. 8, 10, pi. 2, pi. 2a, pi. 2b), or as a single one (Fig. 9). The one on the right side sends an anastomosing branch at once to the anterior, pleural nerve (Figs. 8, 9, pi. /). The anterior pleural nerve arises on the left side from the cerebro-pleural connective (Fig. 10, pi i), close to its union with the pleural ganglion usually, but re- ceiving fibres from both cerebral and pleural ganglia. In some cases, as in Fig. 9, it may be given off from the ganglion di- rectly. It divides at once into two branches which pass to the dorsum. Excretory System : The kidney consists of a roomy, thin- walled sack, mainly lying below the heart, and above the ovo- testis and liver. The semi-diagrammatic Fig. 17 of Plate 3 represents the reno-pericardial system in outline, in its relation to the posterior portion of the body, as seen in longitudinal section. In its maximum width it extends across the full diameter of the visceral cavity. Anteriorly it narrows abruptly to about one-fourth of its greatest width, and is prolonged forward, slightly to the left of the median line, to a point ap- proximately opposite the middle of the anterior, genital com- plex, where it terminates in an irregular, blunt tip. Below, in the region of its greatest width, it sends a keel-like prolonga- tion downward (Fig. 17, v) into the dorsal, median furrow of the ovo-testis. The surface of its wall is simple and smooth, save for a small number of low, lateral and dorsal folds, which appear in front of the cardiac region. Its lining epi- thelium is made up of clear, cuboidal to columnar cells with 18 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. basal nuclei. The renal syrinx (Fig. 17, s) is relatively large and is cylindrical in form. It opens through the pericardial floor at the right of the median plane, below the ventricle of the heart, is directed downward, backward, and to the right, narrowing into the slender, reno-pericardial tube (Fig. 17, r.p. t), which recurves in a loop at the right of the median, ventral lobe of the kidney to pass forward in contact with its right ventral wall, opening into its anterior prolongation well in front of the pericardium, and a short distance behind the pharyngeal bulb. The syrinx is lined with clear, cuboidal cells bearing very long cilia. Posteriorly, the wider portion of the kidney-sack narrows abruptly into a short, narrow, renal tube which opens externally (Fig. 17, r) above, and slightly to the right of the anus (Fig. 17, a; Fig. 5). The kidney dif- fers from that of C. testudhiaria as described by Fischer (1891), chiefly in its somewhat different outline, the local folds in the renal epithelium, and in the ventro-anterior, rather than anterior opening of the reno-pericardial tube into the renal sack. Reproductive System: The ovotestis, in a mature indi- vidual, forms a thick covering completely concealing the dorsal and lateral surfaces of the liver, its main lobes cor- responding in number and outline to those of the latter organ. From each lobe a branch of the hermaphroditic duct arises by the union of several tributaries from the follicles of the ovo- testis. These unite dorsally into the main duct near the median line, which passes forward to the inner face of the anterior, genital complex, made up of the nidamental or mucus, and albumen glands and the related ducts. This com- plex occupies the right, anterior quadrant of the body cavity. It is trapeziform in shape, as seen from above; its outer, longer face is convex, conforming to the contour of the body wall ; its inner face, one-half as long, is flattened against the left, anterior lobe of the liver-ovotestis below; its posterior face is directed obliquely outward and backward in close contact with the almost rudimentary right, anterior lobe of the liver- ovotestis ; while the anterior face slopes obliquely forward and outward in contact with the vaginal duct and penis. Upon the anterior, inner face the slender, hermaphroditic duct dilates into the ellipsoidal hermaphroditic ampulla, from the upper Vol. XVIII] MACFARLAND &■ O'DONOGHUE—NEW SPECIES CORAMBE \() • extremity of which a short duct continues into the cavity of the albumen gland, giving off at right angles the vas deferens. The latter has a thick, glandular wall, loops downward around the hermaphroditic ampulla upon the median face of the com- plex to its dorsal border, thence describes a free loop obliquely backward in front of the stomach, returning in a series of close turns, caused by the varying tension of the retractor muscle of the penis, which is inserted upon it immediately at the right of the central nervous system. The penis extends obliquely forward close to the right of the pharyngeal bulb in front of and parallel with the vaginal duct and the duct of the nidamental-albumen gland complex, to its external open- ing far forward on the right side of the body near the head. In its retracted condition this organ is made up of an eversible preputium, a rather thin-walled, muscular sack, at the bottom of which arises the glans penis. In its everted position, as shown in Fig. 7, p of Plate 2, it extends from the external opening as a cylindrical structure, terminated by the bluntly conical glans {g), and usually showing a few slight, circular rugas, while near the base of the glans proper is frequently found a more prominent ring-like thickening. The organ is entirely unarmed. Immediately behind the external opening of the penis sack is the vaginal orifice, and slightly below it is that of the duct from the accessory glands. The vagina (Fig. 7, v) passes in- ward along the upper and medial border of the genital mass, curves outward and describes a loop upon its upper, posterior face, recurving to the median plane, where it opens into the thin-walled, ovoid spermatotheca (Fig 7, s). Near its en- trance the much more slender, vaginal duct (Fig. 7, vag. d) emerges, passes forward in a short, straight course, receives the duct of the quite small, ovoid spermatocyst (Fig. 7 s. c), and passes into the interior of the accessory gland complex, opening into the irregular lumen of the albumen gland close to the entrance of the oviduct. The cavity of the nidamental gland is relatively roomy and simple, is connected by a short, ciliated passage with the albumen gland lumen, and opens ex- ternally by a wide, short duct, which parallels the penis and vagina, its separate opening being slightly below them. Fischer (1891) was unable to find a spermatocyst in C. testu- y- 20 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. • dinaria, but, with this exception, our results as to the general organization of the reproductive system are in agreement. The nidosomes are common upon the Memhranipora colonies and the adjacent surface of the kelp. Each consists of a nar- row, somewhat flattened band coiled in a close spiral of from one to three turns, attached by one margin. Each nidosome contains from 500 to 1500 capsules, and each capsule contains but a single tgg. The larger the animal the more capsules there are in the nidosome. It is not known whether one ani- mal can lay more than one nidosome at a time or in a season. Blood gland : Immediately behind the pharyngeal bulb, in contact with the central, nervous system, is located the blood gland, resting on the oesophagus. It is discoidal, nearly circu- lar in outline, with quite fine lobulations. The anatomy of the heart and the vascular system does not appear to differ materially from that described for Corambe testudinaria Fischer by that author, and hence need not be re- peated here. The following comparative tabulation indicates the most significant differences between our species and that of Fischer : Corambe testudinaria Fischer . Corambe pacifica MacF. & O'D. Maximum size 4 mm. long, 3.5 mm. Maximum size 13.0 mm. long, 10.0 mm. wide. wide. Branchiae 4 to 7 on each side, the most Branchiae up to 14 on each side, the anterior situated midway of body most anterior situated at 1 /3 of body length in large specimen. length from posterior end. Branchial lamellae few, up to 4 in num- Branchial lamellae up to 20 in number, ber, alternate on shaft of gill. opposite on shaft of gill. A single, posterior, branched, median A series of simple, alveolar glands at gland opening externally above renal bases of gill plumes. pore. Radula 35-40 X (4-5 + 1+0 + 1 + Radula 30-35 X (4 + 1+0 + 1+4) 4-5) A median, cuticular plate in front of Absent. General cuticular thickening radula. only. Liver tri-lobed. Liver five-lobed. Cerebral and pleural ganglia fused. Cerebral and pleural ganglia separate. No spermatocyst. Spermatocyst present. Vol. XVIII] MACFARLAXD & O'DOSOGHUE—NEIV SPECIES CORAMBE £1 /Oo LITERATURE Adams, Arthur, 1847. Notes on certain Molluscous Animals. 1..---'- \ N i - "=^ ;. ^^^ 5 ■3gS^!5«^ras~j„,. . _^^.^^& / - ^ , " ' i . -n / ..^i^iyiS^^ O.H.MacF 24 CALIFORNIA ACADEMY OF SCIENCES L Proc. 4th Ser. Plate 2 Fig. 7. Portion of Reproductive System, v. d, Distal part of vas deferens, extending into the everted preputium, through the wall of which its terminal portion is faintly seen. The everted preputium. tipped by the conical glans, g, forms the penis, p. Close to the right of the base of the penis is the external opening of the vagina, V, which leads inward to the spermatotheca, s; the short and narrow vaginal duct, vag. d, continues on into the accessory gland complex, and receives the duct of the spermatocyst, s. c, close to its entrance. X16. Fig. 8. Central Nervous System in dorsal view, c, Cerebral ganglia; c. 1. rhinophore ganglion and nerve; c. 2, optic ganglion, optic nerve, and eye; c. 3, c. 4, c. 5, nerves to buccal tentacles and mouth re- gion; p, pedal ganglia; pi, pleural ganglia, distinctly separate from the cerebral pair, to which they are joined by the cerebro-pleural connective, c-pl. c; pi. 1, first, pleural nerve, pi. 2, second, pleural nerve of left side; pi. 2a, pi. 2b, rami of second, pleural nerve of right side; v. c, visceral commissure, imiting the pleural ganglia below the oesophagus. X122. Fig. 9. Postero-ventral view of Central Nervous System, the severed oesophagus, o, being left in place, p. c, pedal commissure, pp. c , parapedal commissure, the other abbreviations as in Fig. 8. In this dissection the first, pleural nerve of the left side, pi. 1, arises directly from the ganglion, and not from the cerebro- pleural connective, as in Fig. 8, while on the right side the second, pleural nerve arises from a single root, dividing at once into pi. 2a, and pi. 2b, with anastomosing branches to pi. 1, as in Fig. 8. X122. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVlll, No. 1 [MACFARLAND AND O'DONCGHUE] Plate 2 C.2 8 c // phi c.pl.c \\ ■-' c r\ pU pl.2a -^^s-^ %s. \ «*' O / />/.! 11 J~^'pl2a /i'j.pi2h g 7 1%, ■-'^•ik pl- *^-'- .-^> v.c ^.C^' ^^'/'Z ^^ p.c i ^^ I — ^ ^ y 'C fix ,--i^>'-' "^^^^E* »** jt / f pp.c "%.. / V L^, \\ i> V \i m %^ O.HMicF 25 CALIFORXIA ACADEMY OF SCIEXCES [Proc. 4th Ser. Plate 3 Fig. 10. Central Nervous System from the left side, b, Buccal ganglia; c-b. c, cerebro-buccal connective; c-p. c, cerebro-pedal connective; c-pl. c, cerebro-pleural connective; pl-p. c, pleuro-pedal connec- tive. Other abbreviations as in Figs. 8 and 9. X 122. Fig. 11. Pharyngeal bulb in side view, m, mouth; o, cesophagus; c, muscular crop; r. s, radula sack. X24. Fig. 12. Third, lateral tooth from above. X580. Fig. 13. vSecond, lateral tooth from below. X580. Fig. 13. Second, lateral tooth from below. X580. Fig. 14.' Second (2) to fifth, lateral teeth of radula, obliquely from above. X580. Fig. 15. Outer faces of first, lateral teeth of two, successive rows of radula. a, Small hook at upper, posterior angle of base. X580. Fig. 16. Inner face of first, lateral tooth of radula. w, winglike, basal ridge. X580. Fig. 17. Diagram of renal organ in its relation to the pericardium, as seen in longitudinal, perspective view, w, notaeum, cut lengthwise in the median line, through the posterior, median notch on the left ; b. c, body cavity; a, anus; p, pericardium, containing the heart; .s, renal syrinx, opening into the pericardium below the ventricle of the heart, and narrowing distally into the reno-peri cardial tube, r.p.t, which loops forward to open into the anterior pro- longation of the kidney sack, k,Sit b; v, median, ventral extension of kidney sack; r, external, renal pore. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIIl, No. 1 [ MACFARLAND AND O'DONOGHUE] Plate 3 10 r\ C.2..M c5 __.^>'_r^ i \w II 'I 1 bc\^ -3^v_ "^ ^ "^ J, VN x^ c p c >^^\^--^- \ 4? -%«? J- ^^j^ pip pl.2 :^ f W^^ %- |Tl^^. •% epic 75 ,::; ._:jm J ? ' t 13 ..«*£ /^ /,*{>§*»<:• ^^^ :^ a fep^ n 17 "**^^t^^"'*^ "\,-'^.4^*^^*^i O.H.Mac F PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Fourth Series Vol. XVIII, No. 2, pp. 29-43, 6 text figures January 29,|1929 II A NEW BIRD FAMILY (GEOSPIZID^) FROM THE GALAPAGOS ISLANDS BY HARRY S. SWARTH Curator, Department of Ornithology and Mammalogy The expedition that was sent by the California Academy of Sciences to the Galapagos Islands during 1905 and 1906, secured a collection of birds numbering over 8000 specimens. Gifford (1913) reported upon the species (mostly water birds) from the Columbiformes to the Pelecaniformes (as entered in Sharpe's "Hand-List of Birds"), while Loomis (1918) cov- ered the Tubinares of the expedition in his "Review of the albatrosses, petrels, and diving petrels." The remainder of the collection (nearly 6000 skins), comprising all of the land birds except the one species of pigeon, remained untouched until the middle of 1927, when I began their study. A large part of the land-bird population of the Galapagos is comprised in the "ground finches" of the genera Geospisa, Cactospisa, and Camarhynchns (with which must be included Pinaro- loxias, of Cocos Island), and the "creepers" (Certhidea), and of these there are more than 4000 specimens at hand. A pre- liminary survey of the collection sufficed to show that the ex- tensive series of specimens available would in many cases shed new light upon unsettled questions, and would probably neces- sitate the description of some new fonns. It became evident, January 29, 1929 30 CALIFORNIA ACADEMY OF SCIENCES [Pkoc. 4th Ser. too, that there were specimens in the collection representing undescribed species that were of interest and importance be- yond that attaching to mere "newness" alone. The specimens referred to are unfortunately few in number, comprising four skins representing three different forms, but they are all so trenchantly different from any bird previously discovered upon the Galapagos that their peculiar features may be dis- cussed without considering the possibility of their representing some previously unknown phase of an already described species. As regards most of the slightly differentiated and hitherto unrecognized island races that for one reason or another it may seem desirable to distinguish by name, the publication of their descriptions can await completion of the entire study. But the appearance of the exceptional birds above referred to suggests some questions that it seems to me well to have stated at once, for discussion, and, on my part, for consequent cor- rection if I have read my facts wrongly. The two most conspicuous groups of Galapagos land birds, those most abundant in species and individuals, have of late years been generally referred to two continental families. The so-called "ground finches," referred to one genus (Geospica) or to several, according to the views of different students, are regarded (and always have been, heretofore) as belonging to the Fringillidae (finches), as, curiously enough, has been also the Cocos Island Pinaroloxias inornata. The "creepers" (Ccr- thidea), after tentative assignment to the Fringillidse and Coerebidas, have lately been regarded as belonging with the Mniotiltid^e (American wood warblers), largely as the result of studies by Lucas (1894) and Rid§-\vay (1902). My own conclusions are that the "ground finches" of the Galapagos Islands and Cocos Island (Geospiza, etc.) are not of the Fringillid^e, that the "creepers" (Certhidca) are not of the Mniotiltidae, but that these two groups are very closely related to each other (far more nearly than either is to any continental family), and that the two together should be re- garded as forming one family, a family that is confined to the Galapagos Archipelago and Cocos Island. This family will assume the name Geospizidas, after Gcospiza- (Gould, 1837, p. 5 ) , the first genus described in these groups. Vol. XVIII] SWARTH—GEOSPIZIDJE 3J This opinion is contrary to most of those previously held by others, but the facts now available all point so unreservedly in one direction that I feel no hesitation in arriving at the con- clusion expressed. The characters of the several newly dis- covered forms that are here given names supply so unequivo- cally just the evidence needed to corroborate certain tentative conclusions that can be arrived at from many features found in common among the diverse species of this group, as to make the joining of these species under one family name a course that it seems to me is well-nigh inevitable. The family Geospizidae can not be defined to entire satisfac- tion at present, but the group may be roughly characterized, on the basis of external features, as follows : An assemblage of Passerine forms of small and medium size (wing 48.0 to 95.0 mm.). Wing rather short and rounded; tail rounded, much shorter than wing. Tarsus and toes long, outstretched feet extending beyond tip of tail. Rictal bristles obsolete. Bill extremely variable in relative length, depth, and width. Feathers on lower back and rump long, dense, and fluffy. Coloration unlike in adult male and female (except in Cac- tospiza and some forms of Certhidea), but with great vari- ability on different islands in the number of males of any given form that ever attain "adult" plumage. Color of bill varies seasonally and with age, being black or dusky in adults of both sexes during the breeding season, yellowish or otherwise light colored in adults at other seasons and in the young. Confined to the Galapagos Islands and Cocos Island. As a necessary preliminary to further discussion, names may here be given to the several newly discovered species to which reference is made. First, it will be seen that I am reviving here the name Cactospiza, proposed by Ridgway (1896, p. 546) as a subgenus (type, Cactornis pallida Sclater & Salvin), but, as it seems to me, deserving of full generic recognition. The species of Cactospiza are distinguished by relatively long, slender bill, with the line of the gonys slightly convex. In the slender-billed species of Geospiza the line of the gonys is straight or slightly concave. Cactospisa is further dis- tinguished by having no black in the plumage in any stage, and in that the sexes are alike in every respect. In the other genera of Geospizidse the sexes are unlike in every case except 32 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. in some forms of Certhidea. The genus Cactospiza will in- clude pallida in its several subspecific forms, heliobates, and giffordi. Intergradation between Certhidea and Cactospiza is defi- nitely shown in Cactospiza giffordi, but Cactospiza can not be said to occupy middle ground between Certhidea and Cama- rhynchus. To place the species pallida, heliobates, and gif- fordi in the genus Carnarhynchus would, therefore, in the light of their recognized leaning toward Certhidea, give a false idea of relationships, an impression that can be avoided by the gen- eric segregation of these several forms. Cactospiza giffordi*, new species Type: Male adult, No. 7522, Mus. Calif. Acad. Sci., col- lected by E. W. Gifford (orig. No. 1900), January 18, 1906, on Indefatigable Island, Galapagos Archipelago. Characters: Evidently nearly related to the pallida^helio- bates group, but much smaller and with more slender bill than any other described form in that group. Description of type and only known specimen: In rather worn plumage. Above brownish, about as in the darker ex- amples of pallida, with an olivaceous tinge. Top of head slightly darker than dorsum. A poorly defined superciliary stripe of yellowish from nostril to posterior corner of eye. Sides of head dirty brownish ; a poorly defined grayish spot on lower eyelid. Remiges and rectrices dusky, with narrow edgings of greenish olive; under wing coverts strongly tinged with yellow. Under parts of body and lower tail coverts plain, unstreaked; whitish, strongly tinged with yellow. Sides of breast and flanks grayish brown. On chin and throat irregu- lar flecks of the tawny color characteristic of the throat color in species of Certhidea. Bill black; feet dusky. "Testicles very large" (collector's notation on label). For measurements see table, page 42. •Named for Edward Winslow Gifford, Curator of the Anthropological Museum, University of California, who did a large proportion of the ornithological field work upon the California Academy of Sciences expedition of 1905-1906 to the Galapagos Islands, and who has published reports upon some of the birds collected. Vol. XVIII] SWARTH—GEOSPIZIDJE 33 Camarhynchus conjunctus, new species Type: Male adult, No. 7713, Mus. Calif. Acad. Sci., col- lected by R. H. Beck, February 28, 1906, on Charles Island, Galapagos Archipelago. Characters: Intermediate in certain outstanding features between Camarhynchus and Certhidea. In measurements and in bulk lies between the maximum reached in Certhidea and the minimum in other species of Camarhynchus. The bill in particular is intermediate in shape and size between those of typical Certhidea and typical Camarhynchus. Description of type: In fresh, unworn plumage. Upper parts generally dull olive green, feathers of pileum with dusky centers, giving a blackish appearance to top of head. Sides of head like back; eyelids and faint superciliary line pale yellow- ish. Remiges and rectrices dusky, edged with olivaceous. Greater and middle wing coverts like back, narrowly edged with yellowish, producing two poorly defined wing bars. Below greenish yellow, paler than back. Sides of breast and flanks, and lower tail coverts, tinged with brownish ; middle of belly pale yellowish. Chin and throat indistinctly marked with tawny of the same shade as is characteristic of the throat patch in species of Certhidea. Feathers of throat and upper breast black-centered, producing a streaked appearance, the general effect of which is of poorly defined black lines sur- rounding a rather nebulous tawny throat patch. "Bill black; legs dark brown; testes large" (collector's notation on label). A second specimen, also an adult male, collected by Beck on the same day, is in rather more worn plumage. Color of upper parts is about as in the type, but below it is paler colored, more whitish and with less of the greenish hue. The black streaks on the breast are obscurely indicated, and the tawny on the throat is washed out and but faintly discernible. The rufous is more widespread than on the type, though, spreading to the sides of the head and invading even the super- ciliary line. "Bill black; iris dark brown; legs dark brown; testes large." For measurements see table, page 42. 34 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th See. Camarhynchus aureus, new species Type: Male adult, No. 8121, Mus. Calif. Acad. Sci., col- lected by E. W. Gifford (orig. No. 1944), January 25, 1906, on Chatham Island, Galapagos Archipelago. Characters: Generally similar to Camarhynchus conjunctus but with slightly heavier bill and more uniform coloration. Description of type and only known specimen: In rather worn plumage. Upper parts faded, but evidently originally dull olive green. Remiges and rectrices dusky, narrowly edged with olivaceous. Closed wings, including coverts, uni- form with back. There are faint indications of light tips to the greater and middle wing coverts, and in fresh plumage there may have been discernible wing bars. Below, from bill to and including lower tail coverts, almost uniformly pale yel- low, broken only by a slightly mottled appearance on the breast, where the blackish bases of the feathers show through, and with sides of breast and flanks slightly darker. The yel- low of the under surface spreads over the sides of neck and face, over cheeks and ear coverts, to meet a broad yellow superciliary line that extends from bill and forehead back to a point well behind the eye. Bill blackish, with edges of upper and tip of lower mandible slightly paler. Feet and legs black- ish. For measurements see table, page 42. These two new forms from Charles and Chatham islands, conjunctus and aureus, appear to be closely related, and it might be that adequate series of the two would show plumage variation that would bring them even closer together than is indicated by the scanty material now available. The differ- ences apparent in the skins at hand, however, especially as two rather widely separated islands are represented, are such as to justify the present separation of the two forms. In these two puzzling species (conjunctus and aureus) re- semblance to Certhidea lies in general size and form and in certain peculiarities of markings. Resemblance to Camarhyn- chus appears in the more finch-like bill and in general colora- tion, which in conjunctus and aureus is very close to the un- streaked "immature" plumage of Camarhynchus prosthemelas. There may be significance in the fact that C. prosthemelas Vol. XVIII] SWARTH—GEOSPIZID^ 35 salvini from Chatham Island is strongly tinged with yellow, just as is the one specimen of C. aureus from that island. It is a debatable point as to whether conjunctus and aureus should not be segregated together in a separate genus. Such a genus would have to be based upon the combination of cer- tain characters, some of which in other species occur in Drawing by Mrs. Frieda Abernathy Species of Camarhynchus and Certhidea showing intergradation in bill structure between the two genera. Slightly larger than • natural size. A. Camarhynchus prosthemelas prosthemelas (No. 7756). B. Camarhynchus aureus (No. 8121). C. Camarhynchus conjunctus (No. 7713). D. Certhidea ridgwayi ? (No. 4862). E. Certhidea ridgwayi (No. 4643). F. Certhidea olivacea (No. 4538). Camarhynchus, some in Certhidea, and the genera already described in the Geospizidae are so nearly arbitrary in their nature that it seems to me undesirable to add another genus of uncertain definition. In Gould's (1837) first account of the Galapagos "finches," Geospiza is described as a new genus and Cactornis, Camar- hynchus, and Certhidea as subgenera under Geospiza, inferen- tially as of the Fringillidas, as they are spoken of collectively as "ground finches." Of Certhidea the comment is made (in ^^ CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. the third person, as written presumably by the secre- tary of the Society) "that although he confidently believed that it should also be referred to the same group with the three former, yet in its slighter form and weaker bill it has so much the appearance of a member of the Sylviadcu, that he would by no means insist upon the above view being adopted until the matter shall have been more fully investigated." Sclater & Salvin (1873, p. 16) placed Certhidea in the family Coerebidse, whence it was removed by Ridgway (1896), who, partly on the basis of anatomical studies by Lucas (1894), considered it as belonging to the MniotiltidcB, a con- viction that he (1902) has since repeated. Lucas found vari- ous points of difference between Certhidea and species of Coerebidse, but affinity with Mniotiltidae is founded mainly upon resemblances in the bones of the palatal region. Then Snodgrass (1903) published a most important paper, the results of careful comparative study of the anatomy of Geospisa, Cocornis (=Pinaroloxias) , and Certhidea, with descriptive matter and figures that merit careful scrutiny. His conclusions, reached through examination of the internal anatomy of these birds, are essentially the same as those to which I have been led by comparison of external features, but he did not push his argument to its logical outcome. His closing remarks on the structure of the skull read as follows : "All that the writer here intends is simply to call attention to the fact that there is a gradation in the skull characters of these three genera, progressing by almost equal steps from one extreme to the other. If any phylogenic theory can be based on this fact then the classification of the three genera accepted at present cannot be correct, for Certhidia is regarded as a member of the Mniotiltidae and Geospiza and Cocornis are placed in the Fringillidse. The Geospizce as birds have cer- tainly a most Fringillid appearance. The same, however, can- not be so positively asserted concerning the skull of even the least modified species." The alternatives, apparently regarded as inevitable, of placing these diverse groups either all in the Fringillidae or all in the Mniotiltidae, were so baffling as to cause Snodgrass to stop with the presentation of his really conclusive argument, and to refrain from proposing any change from the formerly Vol. XVIII] SfVARTH—GEOSPIZIDJE 37 accepted but obviously false arrangement. I do not know that anyone has followed up the matter since. Sl Now as to externals. There are of course superficial fea-V tures in which Certhidea resembles species of Mniotiltidae and of Coerebidae; and the obvious dissimilarities between Certhidea and some forms of Geospiza and Camarhynchus are such as at first sight to render apparently ridiculous any assertion of close relationship between those groups. Let us see, however, what external features they have in common. Despite con- siderable differences in size, the largest Geospha at one ex- treme, Certhidea at the other, and the host of intermediate forms between, they are all very similar in proportions. They all have rather short, rounded wings, rather short tail, and long legs (toes in every case reaching beyond tip of tail in the prepared skin); Ridgway's (1901, 1902) diagnoses of the genera Geospiza, Camarhynchus, and Certhidea read surpris- ingly alike in describing the details of those parts. The pro- portions described, too, are not commonly found, if found at all, in the Mniotiltidse or in American species of Fringillidse. Then, there is a peculiar texture of plumage that is common to the several Galapagos forms, something well nigh impossi- ble to describe but obvious to any one handling specimens, and accompanying this there is a peculiarly thick growth of long, loose feathers on the lower back and rump of all the species concerned, such as I do not find at least in North American birds of the families to which they have been relegated. The color of the bill in Geospiza and related genera, and in Certhidea, sometimes black, sometimes light colored, has been described as an irresponsibly variable feature, not to be cor- related with anything else. Without going into details, which are voluminous and complicated, it may suffice here to say that the observed facts justify the conclusion that in all these birds, Geospiza and Certhidea alike, the bill in adults of both sexes is black during the breeding season, light colored at other seasons, and light colored in the young. In Geospiza a uniformly or nearly uniformly black plumage in the male, in Camarhynchus a black-headed plumage in the male, in Certhidea a chestnut-throated plumage in the male, are regarded as the most "perfect" or "fully mature" condi- tion of plumage. In each of these groups, taking any one 38 CALIFORNIA ACADEMY OF SCIENCES [Pkoc. 4th Ser. form on the several islands on which it may occur, the "per- fect" plumage (black, black head, or chestnut throat, as the case may be) will be found in varying abundance on different islands, numerous (perhaps always present) on one, scarce on another, unknown on a third. This is a peculiar phenomenon that certainly seems like another link in the chain holding these diverse forms together. In some forms of Certhidea the juvenal plumage is plain colored and unmarked below, as in the adult, but in the young of Certhidea ridgwayi the lower parts are heavily streaked with dusky, just as in young of species of Camarhynchus. Nests and eggs of Certhidea have been described often with reservations that are significant in the light of the close rela- tionship that I believe is now demonstrated to exist between Certhidea and Caniarhynchus. Snodgrass & Heller (1904, p. 349) make the following statement: "We shot a female of C. olivacea olivacea at Iguana Cove, Albemarle, from a nest containing three eggs. The nest was exactly like that of Geospisa fuliginosa and the eggs were identical in size and coloration with those of the same species. . . Hence, since we have no other examples we hesitate in ascribing this nest to Certhidea." There are other statements in literature (see Rothschild & Hartert, 1902, p. 385) likewise bearing evidence as to the similarity in nesting habits of the two groups of birds. Gifford (1919, p. 242) says of Pinaroloxias inornata: "This species combines the habits of a ground-feeding finch with those of a tree- feeding warbler." Pinaroloxias, further, combines the bill structure of Certhidea with the coloration of Geospisa. Now, added to these suggestive characters found in com- mon in Geospiza and Certhidea, comes the discovery of the several species above described, which appear to be connecting links between the two groups. It will be noted that, curiously, there are two separate points of contact between the "creepers" and the "ground finches." At one point, through Camarhyn- chus conjunctiis, there is what appears to be close connection between Certhidea and the group comprised in the black- headed Camarhynchus; at the second, through Cactospisa gif- fordi, connection between Certhidea and the plain colored species of Cactospiza. Camarhynchus conjunctus and C. au- reus in general appearance are closely similar to C. prosthe- Vol. XVIII] SWARTH—GEOSPIZIDJE 39 melas, so much so that the type specimen of C. aureus was entered as prostheinelas in the field note book of the collector. Cactospiza giifordi, despite its small size, is obviously like C. pallida. Yet in conjunctus and giffordi both there is most un- expectedly displayed traces of the characteristically Certhidean cinnamon-tawny throat patch. As regards the type specimen of C. giffordi, it is suggestive that the note book of the col- lector, E. W. Gifford, contains the following comment: "I obtained one bird at about 350 feet elevation which seemed to be intermediate between Certhidea and Geospiza pallida. It was feeding like a Geospisa pallida on a branch of a tree." If further evidence in the shape of debatable specimens were needed it is found in a bird from Charles Island (No. 4862, Mus. Calif. Acad. Sci., female [immature?], May 29, 1906. See fig. D, p. 35. This specimen is like comparable examples of Certhidea ridgwayi of Charles Island in color and plumage, but the bill (not a variable feature in Certhidea) is larger than in that species, being as heavy as, and a little longer than, in Carnarhynchus conjunctus of the same island (see table of measurements). After careful study I do not know whether this bird is an example of Camarhynchu^ conjunctus (of which plumage stages and amount of variation are unknown) or of Certhidea ridgwayi. In other words, here is a specimen which I find myself unable to allocate, whether to the Fringil- lidae or the Mniotiltidae, as these families were formerly de- fined among Galapagos birds. Both Rothschild & Hartert (1899) and Snodgrass & Heller (1904) dissent from Ridgway's (1896, 1901) division of the "ground finches" into the several genera, Geospiza, Platyspiza, and Camarhynchus, claiming that intergradation of one sort or another necessitates the grouping of the whole aggregation under one generic name, Geospiza. The intermediates here described demonstrate further, pretty clearly it seems to me, the impossibility of drawing a line, or of expressing a clear definition of characters, dividing those genera from Certhidea. Logically, according to the standards adopted by the authors cited above as opposing Ridgway's treatment, all of these diverse forms, from the enormously large-billed Geospisa magnirostris down to the most delicate Certhidea, should be placed in the one genus, Geospiza. Furthermore, I believe that it would be possible, on the criterion of individual vari- 40 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. ation producing overlapping of characters between forms on different islands, to indicate a line of slightly differentiated subspecies under one specific name, that would include most of the described forms" of the several genera, and that would ex- tend through the extremes of bill structure and of color characters throughout the genera Geospiza, Canmrhynchus, and Certhidea. This statement is novel only in the inclusion of Certhidea in the closely linked chain of forms, for Ridgway long ago made precisely the same assertion regarding Geospi- za. In upholding the recognition of slightly differentiated local forms he says: "No other course, indeed, is practicable; for were 'lumping' once begun there could be no end to it, unless purely arbitrary limits were given to the species recog- nized, and if followed to a logical conclusion might easily end in the recognition of a single variable species, equivalent in its limits to the genus." (Ridgway, 1896, p. 468.) I feel, myself, that however logical and consistent it may be demonstrated to be to lump genera in this long list of diverse forms (fifty or more in number), it would not be desirable to do so. The course that I, personally, prefer to follow, is, first grouping the "finches" and "creepers" alike under the one family, Geospizidse, to recognize at least the genera Geospiza, Cactospiza, C amarhynchiis , and • Certhidea. It will be ad- mittedly impossible to formulate entirely satisfactory defini- tions of these genera, but their recognition will afford con- venient lines of demarcation between sections of a long list of species otherwise too unwieldy for satisfactory treatment. To group all of these diverse forms under one generic name would, it seems to me, defeat the purpose of nomenclature of giving us convenient handles to grasp. To recognize the genera indicated is admittedly indefensible on grounds of logic and consistency, and it will cause grief and indignation in the compiler of books and the arranger of "keys" for identifica- tion. It will, however, suit the convenience of whomever wishes to discuss in speech or writing the birds and the prob- lems involved, and that, to my notion, should be regarded as an important function of our nomenclature. Indication of relationships in nomenclature is of first im- fMDrtance, perhaps, but all of the known facts in the relation- ships of these birds can not be expressed in their names. To divide the Geospizidse into as many genera ;as I propose to do Vol. XVIII] SWARTH—GEOSPIZIDM ^\ may give an exaggerated impression of the taxonomic re- moteness of some species, but to lump them under a lesser number would assuredly give an even more erroneous im- pression of close connection between what are really distantly related forms. I feel that common family relationship of Geospiza, Cac- tospiza, Camarhynchus, Pmaroloxias and Certhidea is demon- strated beyond question, but the further problem as to the closest continental relative of the family Geospizidse is not so easily settled. Certhidea is sufficiently unlike any of the Frin- gillidse, and Geospiza and Camarhynchus sufficiently unlike any of the Mniotiltidae, to debar either of those groups from consideration as having supplied the immediate ancestor of the Geospizidse. The general situation is apparently much the same as we find in the Drepanididse of the Hawaiian Islands. In each case there has been wide divergence in bill structure among closely related species, and in the Hawaiian Islands, too, birds with sparrow-like bills were at first relegated to the family Fringillidae. Only after hot discussion were these ap- parent "finches" conceded to be Drepanids and listed alongside their slender-billed relatives. On the Hawaiian Islands species are mostly sharply dif- ferentiated, while on the Galapagos Islands, where we may be viewing results after a lesser period of isolation, we are troubled with innumerable intermediate stages. Strangely enough our strongest first feeling toward the existence of these equivocal races and individuals is not one of gratitude for light shed upon relationships, but of resentment at the havoc they create among our carefully ordered schemes of classifica- tion, and at the breaches they make between supposedly separated compartments in which we strive to arrange species and higher groups. In the Geospizidse of the Galapagos (as in the Drepanididse of the Hawaiian Islands) I think that we must realize that we are contemplating a group of birds that has been isolated on its island home since a remote period of time, and that has developed such distinctive group characters of its own as to have made it well nigh impossible now to recognize the nearest collateral mainland stock, if in fact there is today a corresponding terminal to a parallel line of descent upon the neighboring continent. 42 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. u g ►J H w < o w a JJ 53 u-S ^ 10 10 r>- 10 u f2 10 CS ^H d ^ Ov tS (N ts tN " :S _ 0, m «tl«*---j 4J W ^ •" ^ 10 10 10 10 10 5 _ CU 10 m c^ 00 fS cx*tr^ +j M ^02<«rt 10 10 ■0 10 VI o ■0 VO >o a a 1/5 10 •0 fO Ov 0. d "3 _, in i« 10 *3 ^ - 10 d 00 a >* rO •* fO ro bo (S m K^ ■* 00 c>^ 00 •*' ts vo "1 10 10 10 vO \0 vO >o >o a, oi Ov o ^^ t> 00 10 00" 00" oT P -^ cs cs ts (S d c ^ X: ca 1—1 I-) % ■6 '. ►—1 (U ^ 2 -d •d •6 "rt .SP S t-H h-i HH ■♦J frt w tfi W idefa hathi _4; <« CIt ca ^ Xi 43 h- w w . 3 3 -tJ •*-» w C C 3 P 3 3 1-*- tfl U "S w 3 ^ w 'c jd 0. 'u w >> ca 4) 'c 1. .« ^ M 'd w T- i-, u a cS n) ^ 4J B B 6 t: nl rt ca Gj OJ U U ,— ^ (U rv. 00 ■C •d '6 13 E y< ni rt ni ta ^ ■b 'b 'b , w . (N fo ^ . 00 r^ t^ "* Vol. XVIII] SWARTH—GEOSPIZW^ 43 Literature Cited Giflford, E. W. 1913. The birds of the Galapagos Islands, with observations on the birds of Cocos and Clipperton islands (Columbiformes to Pelecani formes). Expedition of the California Academy of Sciences to the Galapagos Islands, 1905-1906. VIII. Proc. Calif. Acad. Sci., Fourth Ser., II, pt. 1, August 11, 1913, pp. 1-132, pis. 1-7. 1919. Field notes on the land birds of the Galapagos Islands and of Cocos Island, Costa Rica. Expedition of the California Academy of Sciences to the Galapagos Islands, 1905-'1906. XIII. Proc. Calif. Acad. Sci., Fourth Ser., II, pt. II, June 16, 1919, pp. 189-258. Gould, J. 1837. [Remarks on a group of ground finches from Mr. Darwin's collection, with descriptions of new species.] Proc. Zool. Soc. London, pt. V, pp. 4-7. Loomis, L. M. 1918. A review of the albatrosses, petrels, and diving petrels. Ex- pedition of the California Academy of Sciences to the Galapagos Islands, 1905-1906. XII. Proc. Calif. Acad. Sci., Fourth Ser., II, pt. II, No. 12, April 22, pp. 1-187, pis. 1-17. Lucas, F. A. 1894. Notes on the anatomy and affinities of the Coerebidse and other American birds. Proc. U. S. Nat. Mus., XVII, pp. 299-312, 13 text figs. Ridgway, R. 1896. Birds of the Galapagos Archipelago. Proc. U. S. Nat. Mus., XIX, pp. 459-670, pis. LVI-LVII, many figs, in text. 1901. The birds of North and Middle America. U. S. Nat. Mus., Bull. SO, part 1, XXX -f- 715 pp., 20 pis. 1902. Idem, part II, XX + 834 pp., 22 pis. Rothschild, W., and Hartert, E. 1899. A review of the ornithology of the Galapagos Islands. With notes on the Webster-Harris Expedition. Novit. Zool., VI, pp. 85-205, pis. V-VI, many text figs. 1902. Further notes on the fauna of the Galapagos Islands. Notes on the birds. Novit. Zool., IX, pp. 381-418, pi. X, 2 text figs. Sclater, P. L., and Salvin, O. 1873. Nomenclator avium neotropicalium. London. Pp. i-viii + 1-163. Snodgrass, R. E. 1903. Notes on the anatomy of Geospiza, Cocornis, and Certhidia. Auk, XX, pp. 402-417, pis. XVII-XX. Snodgrass, R. E., and Heller, E. 1904. Papers from the Hopkins-Stanford Galapagos Expedition, 1898- 1899. XVI. Birds. Proc. Wash. Acad. Sci., V, pp. 231-372. PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Fourth Series Vol. XVIII, No. 3, pp. 45-71, plates 4-7 January 29, 1929 III A CONTRIBUTION TO OUR KNOWLEDGE OF THE NESTING HABITS OF THE GOLDEN EAGLE BY JOSEPH R. SLEVIN Curator, Department of Herpetology Early in the spring of 1916, my friend, the late Dr. John Van Denburgh, announced to me that he was preparing to again take up one of his boyhood hobbies, and to build up his oological collection, which at that time, contained a represen- tative series of sets of the birds around Los Gatos, the site of his father's home. He seemed to be greatly interested in securing eggs of the Golden Eagle (Aqiiila chrysaetos) , re- garding which he had carefully studied the available literature. I was asked to accompany him on his collecting trips, as he was none too good a climber, though I must confess that I am far from being one myself. Our activities commenced in April, 1916, and extended through a period of years, to and including the spring of 1922. The following notes cover our observations upon seven pairs of eagles during that time, all within Santa Clara and San Benito counties, California. For convenience I shall designate by numbers the different pairs of birds with which we became acquainted. January 29, 1929 4^ CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Pair Number One On April 30, 1916, we left San Jose quite early in the morning and motored to Almaden and then to the Uvas Creek country. Here, by the roadside, we met a small boy. Upon being asked if he knew where there were any birds' nests, he said that he did not, but did know where there was an eagle's nest ! He agreed to show it to us, and said that his sister had one of the eggs. With the boy in the tonneau of our machine we quickly reached his home, where the t.gg was soon in evi- dence. It was a very handsome eagle egg, but blown through two large, irregular holes at its ends. A price having been agreed upon. Dr. Van Denburgh was in possession of his first egg of the Golden Eagle. Taking the boy and his two brothers in the machine with us, we started for the nest. This, the boys said, had been found by their father's hired man, who climbed to it, took the two eggs, broke one in descending, and blew the remaining one. We crossed a low range of hills, and, as we were de- scending, the boys pointed out the nest, clearly visible from the well-traveled road, and but a few yards distant from it. It was indeed a surprise to us to find that the eagle had chosen a site so exposed to view and so close to human habitation. The large deciduous oak in which it was situated grows upon the side of a steeply-rounded hill, one of the first to rise above the level of the floor of the valley. Higher on the hill are a number of smaller white oaks and a little scattering sagebrush. The tree is a large one and originally had three main limbs, but one of these had fallen. Partly as a result of this mutila- tion, there are but few sheltering branches and the nest is but little hidden from view. This nest, which I shall call No. la (plate 4, fig. 1), is built upon a horizontal branch close to one of the main limbs of the tree at a height of about 40 feet. It is not a very large one, quite shallow and about 2^ feet in diameter. We had heard that eagles sometimes lay a second set when their first eggs of the year have been taken, and we hoped that these birds had done so. With much slipping and sliding on the grass we climbed the hill until we w^re level with the nest, when from the top of a small oak we could look directly into Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE 47 it, perhaps 50 yards away. Although we had seen the eagles circling about the neighboring hills, there were no eggs to reward us; the nest was empty. We next visited it on March 2, 1917, when we found it ap- parently in good repair, but empty. No eagles were seen. We did not return until March 25, when we found it still unoc- cupied. The season being now well along, we decided that the birds did not intend to use the nest this year, and that they probably had another in the vicinity, although we had seen no eagles about. We determined to make a careful search, and, separating, went in different directions, where the large trees grew. Dr. Van Denburgh went over the hill immediately be- hind the old nest and a mile or more towards the north. He had not gone far when he flushed an eagle from one of the lower branches of a large oak, but, although he searched far and wide, could find no nest. I, fortunately, was more successful. Crossing the road to a clump of large live oaks about half a mile to the south of the original nest, I found a large mass of sticks and branches which I thought might be an eagle's nest. Climbing to it, I found that it contained no eggs, but it seemed to be just ready for use, being lined with grass. This nest I shall call No. lb (plate 4, fig. 2). No eagles were seen near it. Returning on March 31, we found the old nest (la) still empty, but as we quietly passed under the new one (lb) and reached the trunk of the tree, we saw an eagle arise in the nest. When we spoke she sailed away. Climbing the tree I found one very light- colored egg. We left it, hoping for more. This nest was situated about 25 feet above the ground, and the climb was an easy one. It proved to be quite large, more than three feet in diameter, commanding a most extensive view toward the north and east. April 1 we returned to this nest just before dark. Again we found but the one egg and left it. April 6, 1917, found us back again. As the nest still contained only the one egg, we concluded that no more would be laid, and took it. Incubation was well begun. On April 24 both nests ( la, lb) were empty. Sunday, March 3, 1918, we went up to the highest point on the road. Leaving the machine there, we climbed the fence, walked across the pasture, and reached the tree which had 48 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. held nest lb. Much to our surprise nothing remained of it ex- cept some scattered rubbish on the ground. Not a stick was to be seen at the site of the nest in the tree, although this nest had been a particularly large one and so firmly built that one could stand in it with perfect safety. Returning to the auto- mobile we rode down the steep, winding road, and were soon close to nest la. Finding it unoccupied we left, having seen no eagles about. We did not return again until April 7, when we arrived at nest la at 7 :35 P. M. As we approached we heard a homed owl hooting, and soon saw it sitting in the eagle's nest. When we were quite close the owl flew. Climbing to the nest, I found nothing in it but a lining of lichen, which seemed to be fresh. No eagles were seen. On April 20 we again inspected this nest, but found it empty and saw no eagles or owls. On May 4 we found this nest still empty ; nest lb had not been rebuilt. We spent several hours thereabouts, but saw no eagles until just as we were leaving, when both birds appeared circling low over the hill behind the site of nest lb. It is probable that they constructed another nest in the vicinity. In March, 1919, nest la still remained, but lb was entirely gone. One eagle was seen. This locality was not visited again until April 25, 1922. Nest la had disappeared. Care- ful and extended search revealed no nest although one eagle was seen two or three times during this visit, and also May 4, 1922. Pair Number Tzvo We became acquainted with this pair on the Johnson Ranch, about 4y2 miles southeast of Hollister. This ranch, then farmed under lease by Joe Pacheco, lies on a series of low, rolling hills and is mostly grain fields and pasture with a few white oaks scattered about. As we approached the ranch we saw an eagle circling low over the hills perhaps a half a mile away. Through the hills winds a small stream known as Churchill Creek. At one point on this creek is a small clump of willows, and several of the oaks grow near it. We were told that many years ago the eagles had a nest (2a) in a large Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE 49 white oak at the edge of this stream and perhaps a quarter of a mile from Pacheco's house and barns, a nest that was very difficult to reach owing to the great size of the tree. Finally, the tree fell and the eagles selected another large white oak on a hillside a few hundred yards away. Here they built anew, in a situation which commanded a much more extensive view than they could have enjoyed from their former site in the creek bottom, and it was to this nest that we were led March 3, 1917. I shall call it No. 2b. Although no bird was flushed from the nest on our ap- proach, preparations were under way to climb to it, when a new one was discovered near the top of a large oak some 200 to 300 yards away. This tree grew in the creek bottom, per- haps 20 feet from the stream, and close to a clump of willows at its edge. From the point at which we stood, near nest 2b, this new one, which I shall call 2c, looked like a huge brown ball near the top of the leafless oak. We walked across the pasture, and following a fence along the edge of a grain field, approached it. When we had reached the level ground on which the nest-tree grew, and were not more than 50 to 60 feet from it, the bird arose and flew silently away. We did not see it again. This nest, 2c, is situated 45 feet from the ground. It rests firmly in a large crotch not far from the top of the tree, and is about 2}i feet in each of its dimensions. Our rope ladder was only 20 feet long, but it served to carry me past the most difficult part of the huge tree, my arms and knees carrying me up the remaining 25 feet. After some delay occasioned by the great size of the nest, the two eggs it contained were lowered, one at a time, with can and string, and when they reached the ground safely we rejoiced in the possession of a beautiful set of eggs of the Golden Eagle. With our treasures safely packed, we walked back to the machine. On the way we met Pacheco, who told us that the eagles had used another nest (2b) in 1916. He said that he had taken a young eagle from that nest and kept it some time, but, as it would eat nothing but ground squirrels and had to have a squirrel every day, he soon tired of his pet and put it back in the nest, where, under the care of its parents, it com- pleted its growth. 50 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. In the machine on the way back to town, our guide said that so far as he knew eagles never laid a second set the same year when their eggs had been taken, although they continued to use the same nest or nests during subsequent years. Notwith- standing this statement by one whose knowledge of eagles is great, we determined to investigate this matter ourselves, for we had heard rumors that eagles sometimes do depart from this rule. So, very appropriately as it turned out, the first of April found us back again on the Johnson Ranch. Leaving the machine at the house, we walked up through the fields and met Pacheco. He said that the eagles were nesting again, that he had seen the bird a couple of days be- fore on the nest where we had taken two eggs on March 3. We walked to the tree with high hopes of a second set. Armed this time with 50 feet of ladder, it was more easy to negotiate the climb. Alas for our hopes ! I found it empty. We were about to conclude that we had come too soon, when, on the ground close to the trunk of the tree, we discovered what appeared to be the contents of a fresh eagle's egg. As there was no shell to be found and as we found nearby other unmistakable evidence of his activities, we were forced to conclude that we were too late instead of too early — that some other oologist had been there just before our visit. Two photographs of this nest (2c) were taken. One (plate 5, fig. 4) shows the general location in the tree and the situ- ation of the latter, in a grain field, with a fence on one side, and Churchill Creek with its clump of willows on the other. On April first the leaf -buds were just swelling on the bare twigs and the nest was plainly visible from a distance. The second photograph (plate 5, fig. 3) was taken from the ground directly below the nest. It shows the arrangement of the great limbs and the huge nest resting on them where they fork. Leaving this nest we went up the hill to examine nest 2b. On the way we found a sparrow hawk persistently sitting in a cavity in a white oak tree which recently had been chopped into, doubtless by our unknown fellow craftsman. Nest 2b showed no signs of recent occupation by the eagles. Its ap- pearance is shown in two photographs taken April 1, 1917 (plate 5, figs. 1, 2). The large deciduous oak, with a trunk 13 feet in circumference at the base, is situated in a hillside Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE 51 pasture, near a gully. The second photograph shows the nest and the twisted, rough-barked limbs. The nest is double, a more recent portion partly covering the older platform. We have since learned from Joe Pacheco that this nest (2b) was occupied later in 1917 and that one young eagle was reared there. If his observation and memory are correct this eagle must have made three layings that year. Late in February of the following year, we again visited this region, but found that recent rains had made the roads so difficult to travel that we did not attempt to reach the Johnson Ranch at that time. Saturday, March 2, 1918, found us eat- ing an early breakfast at Hollister. Leaving town at 7 A. M., a short drive through the rolling hills in the crisp morning air brought us to Pacheco's house. Leaving the machine near his barn we walked up Churchill Creek to nest No. 2c, from which we had secured a set of eggs the previous year, and arrived under it without having seen any eagles on the way. We put up the ladder and I climbed to the nest, finding it water-soaked and without any fresh lining, but otherwise in excellent condition. We concluded that the eagles were probably using nest 2b, and, going up the hillside to examine it, reached the tree at 9 o'clock. We had been standing there talking for perhaps a minute when the bird slowly arose in the nest, seemed to step to its edge, and then sailed away. We did not see it again during this visit. With the aid of the ladder I quickly reached the nest and at half-past nine we had two nice eggs safely packed away. One, the lighter-colored egg, weighing 4^4 ounces, was either fresh or infertile, while the more heavily blotched egg, weighing 43^ ounces, contained an em- bryo so well developed that eye pigment and small bones were evident on blowing it. Still seeking to find whether or not these birds would lay a second set this year we left town on April 6, and, arriving at the Johnson Ranch at 5 :45 P. M., we visited nest 2b. We walked under the nest and talked loudly, but the bird did not leave until we threw a clod up into the tree. Again she seemed to arise in the nest and step to its edge before sailing away. Rain began falling as we put up the ladder. Having secured a second set of two eggs, we were down and away at 52 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 6:35. The whiter egg of this set weighed 4%. ounces, while the more heavily blotched one was one-eighth of an ounce heavier. Both eggs contained small embryos, apparently of about the same age. On Saturday, March 1, 1919, we again returned to the Johnson Ranch. In the distance we saw an eagle soaring. As we walked along Churchill Creek numerous mud-turtles slipped into the water from the opposite bank where they had been sunning themselves. The leaf-buds of the deciduous oaks were much more swollen than we had found them on March 3, 1917, or March 2, 1918, and altogether spring seemed somewhat earlier than in those years. We went at once to nest 2b, on the hillside; only to find it empty. There was no fresh lining and green grass six or seven inches tall was growing from it. This nest, as I have mentioned, is a double one, a newer portion resting in part upon an older one. The newer portion has diameters of about four or five feet, while the whole structure has a long diameter of more than seven feet and a depth of about five feet. The accumulation of such a mass of material must have required a great number of trips on the part of the birds. While we were examining this nest we saw an eagle flying away from the other one (2c). When first seen it was about 50 feet from the nest, but we had no doubt that it had just left it. As we were too far away to have frightened the eagle we concluded that it proba- bly was engaged in repairing the nest. However, we thought it best to investigate. We went down the hill and across the field. When nearly under the nest (2c) we whistled and shouted and clapped our hands until we felt certain that it was unoccupied. We then threw the weight over a limb about half way up to the nest and hauled up the rope ladder. Starting up the tree I reached a point about five feet below the nest, when the eagle arose, looked down at me, opened its beak widely, uttered a curious sort of hiss, stepped to the edge of the nest and flew off. In- stead of going out of sight immediately, however, as these eagles usually do, this bird circled about within one or two hundred yards of us, so that we had an excellent view of its plumage. This seemed to be in fine condition, but was pale and quite grayish, especially about the head. We concluded Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE 53 that this bird, which had just left the nest, was a very old female, but of course we could not be certain as to the sex. A few seconds later the mate appeared and both birds circled quite close to us. The second bird was much darker than the first. This was just the reverse of what we had observed at nest 3b on March 4, 1917. The eagles circled about silently for a few minutes and then disappeared. Meanwhile, I had reached the nest and found that it contained two eggs. The nest seemed larger than it was two years before, doubtless growing with repairs. It had a depth of four feet, with horizontal diameters of four and four and a half feet, the nest cavity being about 18 inches in diameter and about six inches deep. It was freshly lined with grass. Resting on the top of the nest, at one side of the cavity, was a sprig of live oak covered with fresh green leaves. On our previous visits the nests of this pair of birds had not been decorated in this manner. We have found, however, fresh leaves in those of other pairs (see 3 and 5), and this habit of nest decoration or marking seems to be a common one. The two eggs, of quite different styles of coloration, were lowered to the ground and packed away. One is heavily blotched and resembles an egg of the second set of 1918. In the other tgg the pigment is more evenly spread as a heavy suffusion about the smaller end. This egg is similar to one of the first set of 1918. The blotched egg weighed just 4y2 ounces, while the other was about one-tenth of an ounce lighter. Incubation in the blotched egg had progressed so far that the eye pigment and vertebral cartilages were evident on blowing. The fonnation of the embryo had begun in the other egg, but was much less advanced, no eye pigment or car- tilage having been formed. On March 29, 1919, four weeks after collecting the set from nest 2c, we again motored to the Johnson Ranch, where we arrived about 6 P. M. Joe Pacheco came out to meet us, to report that he had seen the eagle on nest 2b about five days before, where she remained even when he rode under the tree. Nevertheless, he thought that we would find that she had not finished laying. We walked to the hillside tree without having seen an eagle, and no bird left the nest. On climbing up, it was found to contain one egg. The nest was lined with grass 54 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. and a twig of fresh eucalyptus leaves lay on it. We left it undisturbed and returned to the machine through a gentle shower of rain, the eagles still remaining unseen. On April 10, 1919, we returned. The single egg was found on end in a somewhat mussed and apparently abandoned nest. No eagles were seen. The tgg is a very small one, weighing only 4.1 ounces, and was fresh. We did not visit this pair of eagles again until March 13, 1920, when we arrived at the Johnson Ranch at about three in the afternoon. We went at once to the nest on the hill (2b), which we found unoccupied, thoroughly wet by recent rains, and showing no renewal of its lining. While I was at the nest one of the eagles came sailing over from the south, in- spected us, and passed on toward the flat where the tree which contains the other one (2c) is situated. The eagle, however, did not visit that tree but sailed on out of sight to reappear later over the hill near nest 2b. Feeling reasonably certain that we would find something in nest 2c, we descended to the flat and walked over to the tree which contains the nest. Shouting and clapping failed to frighten any bird from it, but our experience in former seasons made us realize that eagles sometimes sit too persistently to be flushed this way, so we prepared to climb. While we were thus engaged two men rode up on horseback and said that they had seen an eagle carrying fresh green twigs to this nest two days before. We found that this observation on their part was probably correct, for on reaching it we were disappointed to find that it con- tained no eggs, although it had been freshly lined and held a number of fresh leafy twigs of eucalyptus. Only the one eagle was observed during our visit and we were in doubt as to whether the nest was about to be used or had already been robbed; or whether the old female eagle had met with some catastrophe and the green trimmings had been placed in the nest by the male, as in instances previously noted. Circumstances prevented our return until February 27, 1921, when we found the nest on the flat (2c) unrepaired, while the one on the hill (2b) contained green leaves and fresh lining of dry grass not yet pressed into position. On March 18 both nests appeared as on March 6, except that the Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE 55 green leaves in nest 2b were no longer fresh. One eagle was observed flying near on each of these visits. Returning March 3, 1922, we found conditions as on February 27, 1921. The hillside nest (2b) contained unar- ranged fresh lining material of dry grass, fresh live oak leaves and a eucalyptus twig with fresh leaves. The lower nest (2c) was unrepaired. One eagle was seen soaring near. On March 19 the lining of dry grass in nest 2b was found pressed into a well-formed cavity. The green oak and eucalyptus leaves were still present and a small branch of wild rose, with the deli- cate fresh leaves just beginning to wilt, was in the nest. While it is possible that the nest had been robbed within a day or two, we were inclined to believe that eggs had not yet been laid. The season appeared to be very late. April 4, 1922, we found both birds flying near the nest late in the afternoon. Returning to the ranch house we met Joe Pacheco, who said that on March 10 two men appeared at his house at 6 A. M. with two eggs which they had taken from this nest. April 15 we visited nest 2c and found it relined, with well-formed cavity and a branch of fresh eucalyptus leaves. This nest seemed just ready for use, but contained no eggs. On April 20 this nest (2c) was found in the same con- dition as on April 15, except that the green leaves were some- what dried. On May 13 nest 2c was still empty. Nest 2b con- tained a small branch with fresh green leaves. Pair Nimiber Three Our experience with our second pair of eagles having in- creased our desire to know more of these birds, we gladly ac- cepted the offer of our guide to lead us to other nests. On March 4, 1917, we set out in our machine for the Flint Hills. The road which we first tried was blocked by a deep mud- hole which we could not pass. Taking the main road to San Juan, we finally turned down a lane which led us to the flats by the river. The San Juan River here is quite wide. It did not, at this time, entirely cover its sandy bed. We removed our shoes and socks, rolled up our trousers and prepared to wade across. The water was quite shallow, nowhere more than a foot deep, but the sand seemed to "drop out" under our 56 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. feet, often letting us down another foot or more. As there had been a heavy frost during the night, the water was icy cold and we were indeed glad when we reached the opposite bank and could warm our aching feet. The river here runs along the edge of the hills, which are furrowed by a number of small gulches or canyons. The hills are, in the main, bare pasture lands and grain fields, but here and there are a few trees, live and white oaks, which grow singly or in small groups, usually in the hollows or canyons. The first canyon we encountered held nothing of interest, so we passed on over a low hill to the second one. Well up on the side of this canyon stands a large, solitary live oak, and in the top of this tree, perhaps 30 feet from the ground, was an eagle's nest. It seemed not to have been used for some time, but was still fairly well preserved. I shall call it nest 3a. Passing on over the hill to the next canyon, we came upon the eagle, sitting quietly on one of the posts of a wire fence, and but little disturbed by our presence. When we were quite near, it flew a short distance and lit on the ground, where it remained for some time. Our guide called our attention to its pale head and general coloration, saying that this pallor was characteristic of very old birds. On the floor of the canyon, close to the river, is a group of four or five large live oaks, and as we drew near we saw a nest well up in the tallest of them. We walked under the trees but not until we shouted and clapped our hands did the eagle leave the nest. Then she flew slowly and came to the ground near her mate on the hill- side nearby. This nest, which I shall call 3b, is one of the smallest we saw. It was about two by two and one-half feet in diameter and contained comparatively little material. It was built on the main trunk of the tree where the latter curves more or less horizontally and forks before turning upwards again. Its height above the ground was 35 feet. While preparing to climb up to it, we discovered on the ground underneath large pieces of shell of an eagle's &gg. These fragments seemed to be but a few days old, for they were glazed with albumen. Our hopes of getting a nice set were considerably lowered, and our guide said it was hardly worth while to climb for one tgg. However, I climbed to the nest and discovered that it con- Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE 57 tained two beautifully marked eggs. These were so com- pletely covered with eucalyptus leaves that I could not see them as I looked down into the nest. Two or three small eucalyptus trees growing on the bank of the river a few yards away fur- nished a ready supply of these leaves, but we were unable to understand the eagle's reason for using them in this manner. Also, we wondered about the broken egg on the ground. How did it get thrown from the nest? Should we otherwise have gotten a set of three eggs, or was the third &gg laid to take its place ? The eggs taken were both fresh. April 1, 1917, found us again approaching the home of this pair of eagles. Nest 3a, on the hillside, appeared still in its somewhat dilapidated condition as we passed it. We had thought that the eagles might repair it and lay a second set there. Walking on, we soon reached the tree which contained the other nest (3b), from which we had secured eggs just four weeks before. Thus far we had seen no eagles, but when we shouted the eagle arose in the nest and, after a momentary pause, flew off over the hill. We did not see the bird again. Going up to the nest, another beautiful set of two eggs was found. It contained fresh eucalyptus leaves, as on our fomier visit, but the eggs were not completely hidden by them. In- cubation had begun in one ^gg, while the other was fresh. The following year, on Saturday, March 2, 1918, we re- turned to these nests. Nest 3a appeared much more dilapi- dated than in 1917; probably not more than half of it re- mained. The eagles evidently did not intend to use it. Nest 3b was reached shortly after and appeared just as it had in 1917. We saw no eagles but decided to climb to it. It was found apparently ready for use and contained two eucalyptus twigs covered with fresh leaves, the larger of them about a foot long. These we left undisturbed. There were no eggs. Leaving this nest we walked over the hills and up a long canyon, toward the home of pair number four. The hills were bare and we passed very few trees on the way. A little after noon we came upon a large live oak growing on the side of the canyon, from which, when we had nearly reached it, a large eagle flew. Our first thought was that the eagles had moved up here from their old site near the river, but careful 53 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th See. search revealed no sign of a nest either in this tree or in others farther up the canyon. We returned to the nest by the river (3b) on March 16. It still contained the same eucalyptus twigs, somewhat dried, but nothing else, so it appeared that the eagles were not going to use it after all. The fresh leaves found March 2 had made us almost certain that they would. Our final visit to these nests, in 1918, was about four in the afternoon of April 6. No eagles were seen, but as we ap- proached nest 3b a pair of Western Red-tailed Hawks (Buteo horcalis calurus) circled about screaming. We found the nest had been newly lined with moss and contained two eggs of this hawk. On March 2, 1919, we returned to the home of our third pair of eagles and found that not a stick remained of either nest 3a or 3b. In March, 1920, we looked again for a nest of this pair of eagles but were unable to find one. Only the old male eagle was seen. On February 27, 1921, we visited this locality again but there was not a- trace of a nest in either tree. The old male was again seen. On March 18, 1921, we determined to make one more at- tempt to find a new nest. Nothing had been done at either of the old nesting places, but in a small canyon between them we flushed the pale old eagle and his dark mate, both of which flew silently over the hill and disappeared. We looked again in every tree and found nothing in that canyon or elsewhere, until the search seemed hopeless. As a last chance we looked on a hillside, close to the road, where there were a few trees so small that it had seemed useless to examine them, and here we found a large nest only 25 feet above the ground (plate 4, fig. 3; plate 6, fig. 1). The tree showed unmistakable signs of having been climbed recently, so we were not surprised to find the nest empty. March 3, 1922, found us again approaching this nest (3c). When distant about a third of a mile, we observed a large bird perched on the top of the tree which contained the nest. We had covered half this distance when the bird, which proved to be the pale old male, flew down close to inspect us. He then flew back over the tree and disappeared beyond. When we arrived within 50 yards of the nest the dark female arose Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE 59 from the nest and quickly flew from view. An easy climb re- vealed two beautifully marked eggs resting on a fresh lining of dry grass. The nest, constructed of oak branches and twigs, is about three by four feet in diameter and two feet deep, its cavity being about five inches below the rim. The tree is a small white oak. Both eggs were perfectly fresh. One is almost covered with red pignient and weighed 43/2 ounces. The other tgg is white with a few small blotches at one end and weighed 4^ ounces. These eggs are of the same type as those secured in 1917, and probably were laid by the same dark female. This nest may have been built in 1918 and over- looked by us because of its improbable situation. April 4, 1922, we found this nest (3c) empty and no eagles in sight. After photographing it we went on and examined the trees where the other nests had been, but found nothing. The old male flew near us when we were close to the site of the nest 3b. On April 20 we returned and again found nothing at the sites of any of the nests we had seen previously. One old eagle flew by while we were near nest 3b. We gave up the search and went on to examine a Red-tail Hawk's nest farther down the river. About an eighth of a mile beyond the location of nest 3b we found a large nest about 35 feet up in an old Cottonwood tree. On looking closely we made out the tail and wing tips of a large bird protruding over the edge. On the ground below we found a number of dead sticks and twigs and a sprig of fresh, green, live oak leaves. After we had shouted and clapped our hands the eagle left and sailed away. Returning a half hour later we found the eagle again on the nest, this time faced in the opposite direction, and, as before, she left with reluctance. We found the nest (3d) to be about three feet in diameter and a foot and a half high, lined with dry grass on which lay two eggs. The blotched egg weighed 4^ ounces, the yellowish one 4^ ounces. Both were moderately advanced in incubation, the leg bones being about a half inch in length. June 3, 1922, we again visited this nest (3d). Both this nest and No. 3c were empty, and no eagles were seen about them or at the site of nest 3b. January 29, 1929 50 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Pair Number Four Our acquaintance with our fourth pair of eagles began on the afternoon of March 4, 1917. Leaving the home of pair number three, we turned from the river and went over the hills and up a long narrow valley, where there were few trees. On went our guide up the valley, and then, turning to the left over a bare hill, he led us into a deep gulch with numerous large live oak trees. He went directly to one of the largest of these, and following him we saw a huge nest built far out on a nearly horizontal limb some 40 feet above the ground. A few moments' inspection sufficed to show us that the eagle's nest, which I shall call 4a, was not occupied. Two or three hundred feet farther up the gulch we found a still larger tree. Fifty feet from the ground was another large nest (plate 7, fig. 3), again built well out on a horizontal limb, and we could just see the eagle crouching low upon it. Soon she arose and flew away. As our ladder was not long enough to reach it we resolved to return later to this nest, 4b. Accordingly, a week later, on INIarch 11, we again tramped over the hills to this gulch, but this time from another direc- tion. On the way over, in a large live oak tree on a hillside near the lower end of this gulch, we found the remains of a still older eagle's nest, one that evidently had not been used for many years. As it was not more than a quarter of a mile from those in the gulch above, it was probably built by the same birds. I shall call this nest No. 4c. It was about 30 to 35 feet above the ground. When we arrived at nest 4b the old eagle was not at home ; the nest contained two large, pale eggs. This nest was three feet in diameter and lined with dry grass. The eggs were not covered. Incubation had been well begun in both. We saw one eagle in the distance, cir- cling over the hills. On April 15 we returned, but found the nest empty and no eagles visible. This ended our observations for the year. On March 2, 1918, we arrived at nest 4b. It seemed to be in good repair, so, although we had seen no eagles, we decided to make the climb. It was found to be empty and showed no preparation for use. Nest 4a was much more dilapidated than a year before, and of nest 4c there now remained only a few Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE 5^ Sticks. Leaving this gulch, we wandered up a long canyon running towards the northwest. After we had traveled a mile or more, we saw what we thought was a Red-tailed Hawk's nest in a small tree well up on the steep south side of the canyon. We climbed up hill until we were above the nest and could look into it. It contained no eggs, but a lot of downy feathers were sticking to the twigs and branches of which it was made. We saw no birds about and left convinced that this was a hawk's nest and would soon contain eggs. We returned to this nest (4d) March 17, 1918. Having found nest 4a still unrepaired and 4b still unoccupied, we went to the canyon towards the northwest, expecting to collect a set of Red-tailed Hawk's eggs from the nest found two weeks before. We walked up the bottom of the canyon and then straight up the side of the hill to the nest. When we had approached within about 40 feet of it, a beautiful, dark-plumaged eagle arose and sailed away. We could hardly beHeve our eyes, for it did not seem reasonable to find so large a bird on so small a structure, and we had no idea that it could be other than that of a Red- tailed Hawk. This nest was built at a height of about 18 feet, in a small deciduous oak which grew well up on the steep side of the canyon. The situation of the tree made the nest seem quite high and the view from it was very extensive (plate 4, fig. 4). Climbing higher up on the hillside, we were able to look into the nest and to see that it contained two eggs. There was much more down in and about it than we had seen in any other eagle's nest. On the ground below the tree was a lot of debris, either material wasted during its construction or the remains of some earlier platforms. The two eggs which this nest contained were unusually large. The only other eggs which we had taken before were those secured from nest 4b on March 11, 1917. Because of similarity in size of the eggs and because the nests are only about a mile apart we concluded that they belonged to the same pair of eagles. The two eggs taken March 17, 1918, are very dissimilar in appearance. One is quite heavily blotched. The other is entirely white, except for a few faint markings which may be either nest stains or very slight pigmentation. 52 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. The blotched egg weighed just five ounces, while the white one weighed just a quarter of an ounce less. Both eggs were fertile, and in both incubation was well advanced, but more so in the blotched tgg. This would indicate that in this instance the Qgg first laid was larger and more heavily pigmented than the second one. Our last visit during 1918 was on April 7, when we found all three nests (4 a, b, d) unoccupied. On the second of March, 1919, we returned and found nest 4a represented by a mere hatful of sticks. Nest 4c had en- tirely disappeared. Nest 4b, where we secured eggs in 1917, was not reduced in size, but looked ragged and deserted. Having made sure that it was not occupied, we left without climbing to it. Looking back we saw two eagles circling over the hill beyond. We did not visit nest 4d, which had con- tained eggs the previous year. On March 16 we returned for the purpose of inspecting nest 4d. As we approached the nest an eagle circled down towards us, coming quite close three or four times, and then flew farther up the canyon. Nest 4d was empty and showed no signs of occupancy. As we walked up the canyon the eagle again appeared but quickly passed from view. Two weeks later, March 30, we returned to this canyon and looked for a new nest but found none, although we saw an eagle leaving the canyon as we entered it. Nest 4d was empty and un- repaired. The following year on March 14, 1920, we climbed up over the hill from the south and entered the canyon within 100 yards or so of nest 4d. An eagle appeared from somewhere near us, and, apparently in a state of excitement, crossed to the opposite side of the canyon, where it lit upon the ground. We found nest 4d deserted and much the worse for wear, but although we searched carefully we could find no other in the canyon. We did not visit the other canyon, where nests 4b and 4c were located. In 1921, the site of nest 4d was visited and the last remnants of the nest found on the ground under the tree. Search revealed no other in this neighborhood although one eagle was seen. On March 2, 1922, we again entered this canyon from the west. Approaching nest 4b, we flushed the eagle. The nest was found to contain one tgg, which we left undisturbed. Ten Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE 53 days later, March 12, we returned to nest 4b hoping to find two eggs, but fresh marks of climbing-irons showed that we were too late. As we departed, an eagle circled about us several times, 50 to 60 feet above our heads. Pair Number Five On Sunday, March 11, 1917, we arose early and, having had breakfast, were a few minutes later on our way in the automobile, bound for the Flint Hills. In the first canyon that we entered we came upon a screaming pair of Red-tailed Hawks and soon found their nest. It was situated well up in a large oak tree in a position which made it difficult of access. We decided a visit to it would take too much time, so pro- ceeded on our way. Perhaps a mile farther on we came to another canyon with a considerable growth of live oak trees. We were walking along the edge of this canyon, seeking an easy place to cross, when an eagle suddenly flew from a small tree on the opposite bank at a distance of perhaps 40 yards from us. A second glance showed us a nest, from which the eagle had flown. The tree is a small one, and the nest only 25 feet above ground at its base, but the fact that the tree grows close to the edge of the bank of the canyon adds 30 to 40 feet to the apparent height of the nest. The eagle silently disappeared down the canyon, and did not return while we were about. We scram- bled across the canyon and around on the bank above, taking a picture from a point nearly level with the nest. The climb to the nest was an easy one. It was lined with dry grass and some gray moss, and contained two well- blotched eggs. It was very large and probably had been in use many years. I shall call it nest 5a (plate 6, fig. 2). In- cubation was well started in both eggs. On the afternoon of April 1 we returned to the nest of our fifth pair of eagles. This we found empty, though its lining was in good order, as if ready for a second set. One eagle flew down the canyon while we were there. In consequence, we made a final visit on April 15, but found the nest empty. The following year we returned to this nest (5a) on March 3, 1918. The eagle left the nest as we approached. Two 54 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. beautifully marked fresh eggs rewarded us. These are of the same type as those secured the previous year but are more heavily blotched. The one having the larger blotches weighed 4.9 ounces, the other 4.8 ounces. The freshly blown shell of the first Q:gg weighed ^ oz. On March 17, just two weeks later, we were again in this canyon. No eagle left the nest. From the hill above we saw a whitish object in the nest, but were not certain what it was. Climbing to the nest, one very dirty, weather-stained Qgg was found. The nest was wet and disordered and seemed deserted. We concluded that the tgg was part of the first set and probably had been laid soon after our visit of March third. The tgg weighed 4.7 ounces, and was fresh. On April 7, 1918, this nest was empty. We saw one eagle fly down the canyon. March 2, 1919, we arrived at this gulch in the morning during a heavy shower. The eagle was not on the nest. I climbed up to it and found that it contained a lot of fresh lining materials, dry grass and lichen, not yet arranged and packed down. There were also a few small twigs of live oak with fresh green leaves. It appeared certain that the nest would be used later. We left without seeing any eagles, but on returning late in the afternoon, when the sky had cleared, we found them both flying over the canyon. On March 16 we again visited this nest. Arriving at noon, we walked across the pasture, where for half a mile we could be seen from the nest. We crossed the canyon within 100 feet of it, shouted, and clapped our hands. Climbing up on the bank above the nest we tried to look through the branches. We concluded that the nest was empty but decided to climb up to it. Just as I reached the base of the tree, off flew the eagle and silently disappeared. I found two eggs lying in the cen- tral cavity of the nest, which was lined with lichen with an inner layer of green live oak leaves. The cavity measured about 12 by 15 inches, with a depth of about four inches. The whole nest had diameters of five and four and one-half feet, and was about two feet deep. The tgg which is more heavily marked at the small end weighed 4.25, while the other weighed 4.05 ounces. Incubation had just begun in the heavier tgg. The lighter one was fresh. Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE 55 On March 13, 1920, we again visited the canyon occupied by pair number five. An old eagle almost immediately flew up the canyon, passing over the nest on the way. We walked over toward the nest and crossed the canyon at the usual point, talking and shouting as we went. Then we climbed the hillside to a point just above the nest and perhaps 50 feet from it. Standing here, we could see the eagle sitting on the nest and watching us. As we walked closer, she arose and flew silently away. On climbing to the nest it was found to con- tain two eggs, which were considerably nest-stained and much less handsomely marked than any previously obtained from this pair. The nest was lined with grass and a few green oak leaves. In March, 1921, this nest showed no signs of repair or occupancy, and, although one eagle was seen flying in the canyon, no other nest was discovered. On March 2 and 12, 1922, careful examination of this canyon revealed no new nest, although one eagle was seen. The old nest was unrepaired and seemed deserted. Pair Number Six Our friend and guide had told us of another nest which, to his personal knowledge, had been used by the eagles for 30 years, though during this period there were some years, he believes, when they did not lay in it. He visited this nest with his son on March 11, 1917, and secured two eggs in which in- cubation was fairly well advanced. Unfortuntely one of these eggs was broken, but the remaining one he gave to Dr. Van Denburgh. This egg is of a type quite dififerent from those of any other pair of birds investigated by us. On April 15, 1917, we visited this nest (plate 6, fig. 3). It is built in a great live oak which grows well up on the south side of a deep gulch, about Ij^ miles south of Sargents. The main road, about 300 yards away, may be seen from it. It was nearly dark when we reached the foot of this tree, but we soon con- vinced ourselves that it was not occupied. The next year, 1918, we returned to this nest, which I shall call 6a, on the morning of February 22. It appeared to have 55 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. been damaged by the winter stomis, and a considerable por- tion of it was on the ground. No eagles were about, and we concluded that the birds did not intend to use it. We climbed to the top of the hill and went down another canyon, which we thought would lead us to nest 5a. We had walked per- haps half a mile when, as we had expected, Dr. Van Denburgh pointed out a nest, some 300 yards ahead of us. At that dis- tance it could not be seen clearly, but I had scarcely time to say that I thought it was the nest with which we were familiar when we saw an eagle leave it and fly off over the hill. As we drew nearer the situation looked less familiar. The trees seemed much too large and the bank too low according to our memory of nest 5a. However, it was not till we reached the base of the tree that we recognized it to be one in which we had found the nest of a Red-tailed Hawk in 1917. Nest 5a was in another canyon about a mile beyond. The hawk's nest had entirely disappeared; not a stick of it remained. The eagle's nest was a few feet higher in the tree and was built on much larger limbs. On the ground below were numerous dead oak branches and twigs, evidently dropped in constructing the nest. The structure seemed large enough to have been in use several years, yet we knew it to- be a new one, as there had been none there the year before. From the fact that the bird left while we were still so far away, we concluded that she had not laid, and that she probably was completing the lining of the nest when we discovered her. We left without climbing to the nest, which I shall call 6b. On March 3, 1918, we returned, arriving under the tree at 8 o'clock in the morning. The bird was at home and did not fly until we threw a stick up into the tree, but there were no eggs. We then walked on to the next canyon to inspect nest 5a. Returning later to nest 6b we found no eagle on it. Two weeks later, March 17, we found no eagle at nest 6b. Climb- ing up to the nest it was found to be still empty. We did not return again until April 7. Nest 6a was ragged and deserted. Nest 6b was empty, but on the ground beneath it we found the remains of a broken eagle's egg. The season now was so far advanced that we had no fur- ther expectation of adventures with eagles. We had, how- ever, found the nest of a Red-tailed Hawk near the upper end Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE 57 of this canyon and decided to visit the canyon again in hopes of getting a set of eggs. Therefore, April 20 found us again in this canyon. As we passed under the eagle's nest (6b) we noted that it was unoccupied. Some 300 yards up the canyon, \^p^ an eagle circled down towards us, and then turned and flew ^ away. We went on to the Red-tail's nest, found it empty, and returned to the place where we had seen the eagle. A little higher on the hillside is a group of large live oaks. We had examined these trees several times in 1917 and 1918 and were certain that there was no nest in them. However, we had scarcely entered this little grove when we saw a big nest well up in one of the largest trees, and, as soon as we clapped our hands, off went the eagle. On climbing to the nest I found it to contain a nice pair of eggs. Incubation was well ad- vanced. I shall call this nest 6c. These eggs, taken from nest 6c on April 20, 1918, are of the same type as those se- cured March 11, 1917, from nest 6a. The fragments of an tgg found under nest 6b also were of this type. It is probable that this pair of eagles deserted their old nest and moved a mile or more to another canyon, where they not only laid twice, but actually built two nests. Returning in 1919, we examined these three nests (6a, b, c) on March 1 and 2. All three looked deserted. I climbed to nest 6c only, which was found empty and unrepaired. It measured about five by three and a half feet over all, and about two feet in depth. We saw no eagles in either canyon. On March 16 we returned and photographed these nests. Nest 6a is shown well out on a nearly horizontal limb which hangs over the canyon. Nest 6b (plate 7, fig. 4) may be seen well up in a large live oak which grows from the side of an- other canyon. As we entered the lower end of this second canyon, on the morning of March 16, we saw two eagles soar- ing well up in the canyon near the group of trees in which nest 6c is located. This group of live oaks is shown in the photo- graph (plate 6, fig. 4). The nest does not show. It is near the top of the tree and at the extreme right of the central group. The eagles quickly disappeared, and we found these nests (6b, c) still unrepaired. After making a wide circle over the hills we entered the upper end of the first canyon, in which nest 6a is located. This gg CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. canyon near its upper end, high on the hillside, becomes broad and shallow and has but few trees. As we walked down the canyon an eagle circled to meet us and then sailed off towards the left. Lower in the canyon are many trees, growing in two main groups that are separated by an area of open pasture. Nest 6a is in the lower group of trees. We had searched through the upper group twice in 1918 and were certain there was no nest there in March of that year. We now decided to search both groups again, and Dr. Van Denburgh started for the upper grove while I set out for the lower one. We had walked only a few yards when Dr. Van Denburgh saw the eagle sweep down close to a large tree near the edge of the upper group, and a few moments later he saw a nest in the tree. As he called me, a second eagle left the nest; both birds departed silently and we did not see them again. The tree is a large one but so well provided with branches that the 40-foot climb was not difficult. I soon reached the nest, looked over the edge, and with some excitement reported three eggs. I shall call this nest 6d (plate 7, fig. 1). It was very large, about 3^ feet deep with extreme diameters of about six and five feet. Its central cavity, about a foot in diameter and seven inches deep, was lined with dry grass, and held also a cluster of fresh oak leaves. The eggs are rather small and elongate, of the same type as those taken from nest 6c in 1918, but much more beautifully blotched. They weighed 4.02, 4.01 and 3.99 ounces, respectively. The egg with the greatest weight was most heavily pigmented, and the lightest one least so. The lightest egg was infertile. Incu- bation in the other two eggs was well advanced, but had progressed further in the heavier egg, in which the bones were quite firm. From conditions in this set it would seem that the first egg laid is the largest and most pigmented. March 14, 1920, we again returned to the haunts of this pair of eagles, and ascending the canyon which contains nests number 6a and 6d, both were found unoccupied and showing no evidence of any repairs having been made. The former was very dilapidated, and the latter had been twisted out of position by the winter storms. One eagle was observed soar- ing over the top of the hill, but nowhere in the canyon did we find any other evidence of occupancy. This being so, we de- Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE gQ cided to see whether or not the eagles had moved back to the second canyon, in which nests 6b and 6c had been built during 1918. From the opposite side of the canyon nest 6b appeared to be in excellent condition, but although we shouted and clapped our hands, no bird left it until we crossed the canyon, when the eagle quietly arose and flew away. We did not see the bird again. After some delay and difficulty the nest was reached and found to contain a set of three poorly marked eggs of the same general type as those secured from this pair in previous seasons. As we were successful here, we did not visit nest 6c. None of the nests of pair six showed any signs of repair or occupancy when visited by us in 1921, and no new nests were found. On March 2, 1922, nest 6d contained dry grass not yet pressed into shape. While descending the canyon we found a nest which we were quite certain must have been built since our visit in March, 1921. On climbing to this one, 6e, I found it to be in poor shape and unlined. One eagle was seen fllying about at the lower end of the canyon. Nest 6b was found to be in a good state of preservation, but unre- paired. There was no nest at the site of nest 6c. On March 12 conditions were unchanged at nests 6b, d, and e. On March 18 there was no change in the lining of nest 6d, and 6e seemed unoccupied. On April 15 and 25, 1922, nests 6d and 6e were unoccupied and no eagles were seen. Pair Number Seven About two o'clock in the afternoon of March 3, 1922, Dr. Van Denburgh and I arrived at the foot of a steep hill three and a half miles northwest from the town of San Juan. At the top of this hill rises a huge rock about 140 feet high, the upper portion of which forms a perpendicular clifif 95 feet high, facing a little west of north. Fifty feet below the top of this cliff is a recessed ledge upon which we had seen a nest two years before. The top of the rock is nearly level, and its southern end is buried in the earth of the hill-top, so that one can easily walk out to the brink of the precipice (plate 7, fig. 2). The earth hill itself is quite steep. We spent 45 minutes 70 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. climbing to the top, and only 15 minutes returning to the automobile. Just below the top of the hill we flushed a large dark eagle from the ground. At the base of the rock, below the nest, we found a number of dead oak branches and twigs, the freshly broken ends of which showed that they had been brought to the nesting place very recently. We lowered the rope ladder from the top of the rock until it hung directly in front of the nesting ledge. When the bottom of the 50-foot ladder reached the foot of the cliff the top of the ladder was about five feet above the nest. About 15 feet up from the bottom of the cliff is a ledge upon which two men can stand. We both climbed to this ledge and Dr. Van Denburg held the ladder while I climbed the remaining 30 feet to the nest. It proved to be a large one, about four by five feet, and freshly lined with dry grass, which had not yet been arranged and pressed down to form a cavity. As I descended I noticed old holes which had been drilled in the rock near a cleft which extended up from the ledge on which we stood. Later we noticed little steps cut in the rock below this ledge. These holes and foot- holds lead us to believe that this is the same nesting place that was robbed by J. R. Chalker in 1887 and 1888, as described in 'The Ornithologist and Oologist" (XII, No. 6, 1887, pp. 86-88; XIII, No. 8, 1888, p. 120). On March 12 we visited this nest again and took photo- graphs of the rock. One eagle soared about the hill and rock as we approached, and at one time flew within a few feet of the nest. This was found to be in much the same condition as on March 3. We returned on March 18. As we drew near, an eagle left the nest and disappeared over the top of the hill. Having climbed to the nest I discovered one beautiful Ggg rather evenly covered with small red spots. The dry grass had been smoothed and pressed down, forming a slight central cavity. The nest was without any down or green decorations. We departed again without seeing the eagle. April 4 we returned to the rock. When we were about 100 yards away the eagle left the nest and silently flew straight down the valley. We found but the one egg, evidently a com- plete set. This egg weighed 4j^ ounces. Incubation was well begun, small bones being just distinguishable on blowing. Vol. XVIII] SLEVIN— NESTING HABITS OF GOLDEN EAGLE J\ On May 4, 1922, we found this nest, which I shall call No, 7a, empty, and no eagles about. This proved to be our last jour- ney in quest of the eggs of the Golden Eagle. All of the eggs mentioned in this paper and tabulated below are now in the collection of the California Academy of Sciences. Number of Eggs Museum Number Field Number in Set Date Collected 4643 la 1 March ?, 1916 4649 lb 1 April 6, 1917 4650 2b 2 March 2, 1918 4654 2b 2 April 6, 1918 4660 2b 1 April 10, 1919 4644 2c 2 March 3, 1917 4656 2c 2 March 1, 1919 4645 3b 2 March 4, 1917 4648 3b 2 April 1, 1917 4663 3c 2 March 3, 1922 4665 3d 2 April 20, 1922 4646 4b 2 March 11, 1917 4651 4d 3 Mar. 3-17, 1918 4647 5a 2 March 11, 1917 4652 5a 2 March 17, 1918 4657 5a 2 March 16, 1919 4661 5a 2 March 13, 1920 4662 6b 3 March 14, 1920 4655 6c 2 March 20, 1918 4658 6d 3 March 16, 1919 4664 7a 1 April 4, 1922 PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 3 [SLEVIN] Plate 4 Fiq.3 Fiq.4 Fig. 1. Nest la. Fig. 3. Nest 3c. Fig. 2. Tree containing nest lb. Fig. 4. Nest 4d. January 29, 19J9 PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 3 [SLEVIN] Plate 5 CM CM (U I-' CM '^ be be m CT> Li_ 1/3 cvi <1> ^ ^ to bfl PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIll, No. 3 SLEVIN] Plate 6 (.J bO lO ^ X ^ ^ .- o & .2^ y • ^— > > LL o 5 M CS 1;^ ^ •- .Ti Q O <\i ^ 'M he. fc E CO CD U_ 6 rt Nest Nest — ro bi) bi) tZ tZ PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 3 [SLEVIN] Plate 7 iL Is iZ IZ ,.~:^L^ I CO , ; jD d^ 'O -r • ^^ c/^ tn LL. r-; ro .Sf bb ti £ PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Fourth Series V^OL. XVIII, No. 4, pp. 73-213, plates 8-23 March 29, 1929 IV MARINE MIOCENE AND RELATED DEPOSITS OF NORTH COLOMBIA BY FRANK M. ANDERSON Contents Page Introduction 74 Post-Eocene Sequence 75 Poso Series 76 Structures 82 Stratigraphic relations 83 Age of the Poso series 85 The Miocene Series 86 Las Perdices group 89 The Tubera group 91 Local occturences 93 Comparison of horizons 95 Galapa-La Popa group 98 Pliocene Deposits 99 Correlations 102 Description of Species 105 Gastropoda 106 Pelecypoda 146 Foraminif era 179 March 29, 1929 74 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th See. Introduction The marine Eocene deposits of northern Colombia have al- ready been described in earlier papers^ and therefore require only general notice here. For the most part they occupy a broad synclinal area between the north coast of Colombia and the spurs of the northern Andes lying to the south. In the midst of this general syncline which extends for more than 160 miles, there are pronounced anticlinal folds extending parallel with its axis and also with the coast. On the southern border of this syncline the Eocene rocks outcrop in an irregular zone following the contours of the pre- existing ranges and spurs, while upon its northern limb they outcrop in disconnected areas along the Caribbean coast from the west flank of the Sierra Nevada de Santa Marta to the Gulf of Uraba. A large area of these rocks, for example, lies wTSt of the Rio Magdalena, extending north from Arjona nearly to the sea, and to the southwest for an unknown dis- tance. The "Arjona group" mentioned in a former paper^ occupies this area. Farther to the southwest other areas of Eocene are found in the Coloso range, in the Cerro de Cispata near Lorica, in the Cerros de las Palomas, and in other dis- tricts about the head of the Rio Sinu. Wherever they are found the Eocene rocks are highly folded and are traversed by faults. In some cases they are much compressed and distorted, but they are sufficiently fossil bearing for identification. Post-Eocene Sequence The Eocene deposits of Colombia are for the most part, especially in the central areas of the syncline, overlaid by a sequence of strata of great thickness. In some places these later beds overlap the borders of the trough and along its coastal side flank it for many miles. While the succeeding divisions of this sequence are largely the result of reconnais- sance, and only qualitative study can be claimed for them, yet it is believed that the more important series are properly dis- tinguished, and their position in the column is undoubtedly 'Anderson, F. M., Proc. Calif. Acad. Sci., Vol. 17, 1928, pp. 1-29. =< Anderson, F. M., Bull. Amer. Assoc. Petrol Gaol., Vol. 10, 1926, p. 387. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 75 correct. The maximum thickness of the post-Eocene fornia- tions is as much as 8,000 feet, of which the major part is re- ferred to the Miocene, and the remainder, some 3,000 feet, may be largely, if not wholly, Oligocene in age. There is as yet only an imperfectly defined boundary between the two, while in some localities there is evident unconformity, and this may later prove to be the general condition. In the Carmen-Zambrano section, elsewhere described,^ be- tween the proved Eocene and the fossiliferous Miocene above, there is a great body of clays, sandy shales and calcareous concretionary beds that were tentatively classed as Oligocene. Some of the shale in this interval appears to be equivalent to the "Bombo shales" of Beck,* while some of the strata may be lowermost Miocene, as described later. The lower and major part of this sequence, as it occurs here and at other points along the Colombian coast, has been given in this paper the name of "Poso series/' from the fact that in the Sinu region, where it was first recognized, and at other points on the north coast, various wells had been drilled into it for petroleum. It is well known to contain many seep- ages of oil and gas, and other evidences of having commercial possibilities as a source of petroleum. West of the area of the Arjona rocks referred to above, as to the east of Turbaco, a later series of considerable thick- ness outcrops over a wide zone, in contact with the Eocene on the east and fossil bearing Miocene on the north and west. This series is here highly folded into a succession of anticlines extending from the railroad northeasterly for some miles. The gas vents, mud volcanoes, or the "Turbacos" of von Humboldt, have their origin in this series of rocks. Its stratigraphic position is between the Eocene and the Miocene, and will doubtless find a place in the Poso series as described later, though whether the complete series is repre- sented here is not known. In the column drawn by Elfred Beck (p. 463), the "Huertas series" 1,000 feet in thickness, is divided into two nearly equal parts by the semblance of an unconformity, though it is not mentioned as such. sAnderson, F. M., Proc. Calif. Acad. Sci., vol. 17, 1928, p. 11. «Beck, Elfred, Econ: Geol., vol. 16, 1921, pp. 453-465. 76 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. The upper portion of the "Huertas series," as shown by the fossils, and as observed by the present writer, is properly Miocene, belonging to a group which will be described later. The lower portion immediately overlies the "Bombo shales," with which it appears to have stratigraphic continuity. The "Bombo shales" have been shown to be of Oligocene age, though this determination applies to not more than 500 feet of strata. The Poso series. For the purpose of recording some ob- servations made in the Tertiary districts of north Colombia in 1914-1915, and at later dates, and to call out further discus- sion of the subject, the following pages have been selected from personal notes, reports of assistants, and from various data obtained by the writer, covering the general region of the Rio Sinii and its environs, which describe in some detail the formations that, in the light of present evidence appear to in- tervene between the Eocene and the Miocene series. From a report by Bruce G. Martin (1914) on the San Sebastian district, the following is taken : "Unconformably overlying the San Sebastian chert (Eocene), is a series of arenaceous and argillaceous sediments to which the name 'Poso series' is applied. These beds consist of hard to medium soft, coarse-grained, gray sandstone, and sandy clays and a small amount of limestone. Nearly all types and colors of sandstone and clay appear to be represented in this series. The lithology and sequence of beds can be best described by giving a cross-section at right angles to the strike." Extending easterly from the Cerro de San Sebastian, Mr. Martin's condensed section follows : d. Alternating hard coarse sandstone and medium grained, sandy shale 1000 feet c. Medium soft, fine grained, bluish gray sandstone and clay, with some concretionary limestone lenses. . . . 1500 feet b. Medium coarse, hard gray sandstone, and medium soft, blue or gray sandstone 900 feet Total 3400 feet Unconformity a. San Sebastian cherts, etc. (Eocene). Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA J'J His report then continues : "In the San Sebastian section all the beds have been folded into a mono- cline which dips rather steeply toward the southeast. The sequence of beds, as here exposed, continues northward for several miles. In a general way these stratigraphic divisions will hold true for the whole area." Botli Mr. Martin and John H. Ruckman described a similar series between the villages of Cocorilla and Purissima. Mr. Ruckman says in part, concerning this district : "The oldest rocks in the district are undoubtedly the cherts and hardened sandstones of the San Sebastian series (Eocene) which also make up practically the entire mass of the Cerro de Cispata. Overlying these, and in turn hidden by later deposits, there exists a series of very considerable thickness The concretion-bearing shale and limestone on the Lorica-San Antero road represents its lower limit. Upon the limestone are sandstones containing many large, purplish concretions. They also contain considerable limestone in layers, as well as small bits of limestone, possibly representing inclusions from strata beneath This series of limestones, shales and sandstones is probably, in part at least, equivalent to the Poso series [of Martin]. Over- lying the Poso series and overlapping it unconformably upon the San Sebastian cherts, there is a rather thin deposit of chert conglomerate, gravels and poorly consolidated sands. They are not well exposed northwest of Cocorilla, but are unquestionably identical with those farther south near the San Sebas- tian hills. Fossils obtained from these beds are comparatively recent forms, suggesting correlation with the La Popa (Miocene) group." . . . . Mr. Martin, later describing the district bordering on the Cispata Bay, and about the north end of the Cerro de Cispata, says: "The rocks of this district belong to two formations; the oldest geologically is the chert formation which makes up the main mass of the Cispata hills. . . . . This formation occupies the central part of the hills and probably underlies the chert conglomerate exposed in the small hill immediately north of San Antero. Unconformably upon this chert lies a varying thickness of chert conglomerate and gravelly sandstone. This conglomerate and sandstone appear to be several hundred feet thick along the east slope of the Cispata hills. The size of the chert fragments decreases in going from the base upwards. The sandstone overlying the conglomerate consists almost entirely of small grains of chert Overlying the conglomerate and cherty sand- stone, probably conformably, is a thick series of sandstones, sandy clays, and variously colored soft shales The upper portion of the series consists mainly of medium soft, argillaceous sandstone with a small amount of thin-bedded shale interstratified with it. These latter beds are well exposed along the crest of the San Antero hills. Another belt of mediimi hard sand- stone occupies a narrow area near the central part of the map, at the gas 78 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. springs. The gas escapes from this member of the series The two largest areas of Umestone occur in the south central part of the district. The medium soft, argillaceous sandstone is represented in yellow, the clays and shales in citrine, the basal sandstone and conglomerate in brown, and the chert in red. A peculiar featiire of the series is the great lithologic varia- tion." .... The Poso series was followed southward up the valley of the Rio Sinu to above Monteria and along the east flank of the Cerros de las Palomas, where sandstones predominate in great thickness. At a locality 12 miles northwest of Cerete the formations are almost exclusively sandstones, often very siliceous, as if derived from underlying cherts. They stand at high angles with a strike of N. 70° E., and a dip of not less than 45° to the northwest. Mr. K. D. White, who visited this district after the writer's visit, says, in part : "All exposures of rock seen were phases of sandstone. In fact, no outcrops of pure shale were found. The lowest bed, forming the center of the anticline, is massive gray, micaceous sandstone, with interstratified layers of grit, also massive in bedding. The grit members have layers of conglomerate that are typically millstone grit. The entire series is ferruginous; above the grit beds the sandstones become finer grained, more compact and siliceous. Many seepages of petroleum issue from these sandstones." .... Concerning a locality some 12 miles west of Cerete, Mr. Martin says : "All the rocks observed are of sedimentary origin. They consist of shale, soft, sandy clays, fine and coarse grained sandstone, and conglomerate. . . . Thin layers of conglomerate and grit can be seen closely associated with fine sandstone and clay. The colors of these rocks vary from very light gray through blue, gray and yellowish gray to brown The rocks are well stratified in general, although in places the strata are so greatly crushed that the bedding could not be distinguished from fracture planes. The inclin- ation of the strata varies considerably Owing to [this fact] no well defined folds could be distinguished. In the vicinity of the gas and oil springs, where more detailed work was done the beds have been crushed and twisted to such a degree that it becomes impossible to recognize any definite structure The oil here appears to be seeping from greatly crushed clay shale and fine-grained sandstone. Some of the rock fragments have a strong odor of petroleum. The gas springs consist of eight or ten small vents from which small quantities of inflammable gas, water and mud are escaping Small mud cones from one to three feet high have been built up about the vents." .... Farther south and nearly west of Monteria, on the east flank of the Palomas range, the beds are less sandy and show Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 79 a disposition to become shaley, but they exhibit the same structural conditions as before. Mr. Martin, who worked in this district, reports in part : "The younger beds consist of grits, massive sandstone, soft shaley sand- stone, and soft mudstones. The grits and massive sandstone are hard and usually thick-bedded. The shaley sandstones and mudstones are thin-bedded and greatly fractured The rocks are so arranged that three or four distinct lithologic divisions can be distinguished." . , , . His report divides the strata of this district as follows : 1. An upper shale and sandstone member 1500 feet 2. A sandstone and grit member 2000 feet 3. A basal shale member 1500 feet Total 6000 feet The upper member of this section is probably later in age than the Poso series, and may be Miocene. The lower mem- bers are undoubtedly referable to this series. They are of a dark bluish or gray color, are considerably indurated, and are much folded and faulted. The strike is N. 20° E., and the in- clination is from 45° to 75°. There are two or more closely folded anticlines in the area examined and several seepages of petroleum and gas. Con- cerning these structures, Mr. Martin's report continues : "Along the axes of the folds the strata are often vertical. In going across the strike away from the axis the inclination gradually decreases, until dips as low as 10° are sometimes found The sequence of beds is similar over the entire area. Near the axis shales occur in every case. The petroletun [and gas] usually comes out with more or less acrid or sulphurous water, and accumulates on the spot as black asphaltiun, the gas springs often forming small mounds of mud, or 'mud volcans'." After an excursion made into the Palomas range, some 30 miles southwest of Monteria, Mr. Ruckman reported : "Many interesting seepages of oil and gas were found together with many mud volcanoes, characteristic of this region No igneous or schistose rocks were observed, while jasper and chert occur only as float from the Palomas range." After describing the sedimentary beds from which the oil and gas were issuing, the report continues : 8Q CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. "This series [of strata] .... is almost certainly Mr. Martin's 'Poso series.' It is made up of a highly folded series of fine, thin-bedded, or massive, micaceous sandstone, and fine, rather hard, blue-black shale con- taining calcareous concretions and occasional lenses of limestone. Fragments of chert and limestone, similar to those in the Cerro de Cispata, forming several types of conglomerate were noted along the streams draining the Palomas mountains. The petroleimi of all the seepages noted was associated with the shale On the Quebrada Matamoras there is a very fine seepage of light oil. The oil comes directly from the shale, and evaporates, leaving only a stain on the shale The seepages extend for 600 feet along the creek, issuing with some gas. The bedrock is almost entirely shale standing nearly vertical, the lowest dip being 45° toward the Palomas mountains, suggesting an overturn." Rocks of the Poso series occur also near San Andres, though not in the thickness noted in the foregoing quotations. At a point on the San Andres-Momil road, some three miles east of the former place, an outcrop of these beds was noted in 1915. They consist of thin-bedded, dark, sometimes green- ish-gray clay shales and nodular, or concretionary, limestones. In places they are gravelly, with pebbles of hard, dark, si- liceous rocks, such as occur in the underlying Tofeme member of the Eocene. These shales have a strike of N. 30° E., and dip rather steeply to the southeast. They are overlaid by a brown or rusty-colored sandstone having a similar strike and dip, which, upon further observation, appears to rest uncon- formably upon the older series. These sandstone beds are fossiliferous, and belong unquestionably to the Miocene (Tu- bera) group, later described. These two formations are probably represented by the two portions of the "Huertas series" of Beck. On the coastward side of the Palomas range, the Cerro de Cispata, and the Coloso range, the Poso series is exposed in many localities. At the west foot of the Coloso range there is a series of somewhat indurated, dark tlay shales, sand- stones, and hard conglomerate, without fossils, as far as observed, folded into a sharply compressed syncline in which the aggregate thickness of strata is not less than 2,500 feet. This section was visited by Mr. Martin and the writer in 1914, and the conclusion was reached that the series was identical with the Poso series of his earlier report. The strike of the beds is roughly parallel with the general line of the coast, or nearly northeast and southwest. Seepages of oil were found Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA gj^ here issuing from shales near the base of the series, as is usually the case. Similar beds occur about Cispata Bay to the north and west of the Cerro de Cispata, and here too are found seepages of gas smelling of petroleum. The same series outcrops near Paso Nuevo and at other points along the coast. A few miles to the southeast of Moni- tos, beneath the sandy beds of the Miocene, which here follow the coast, standing at a high angle, there are hard, dark- colored shales and sandstones, also highly inclined (60° to 75°), striking parallel to the coast line, and overlying the Eocene. Their observed thickness was estimated at 1,500 feet, though it is probably more. Beneath are fossiliferous beds of Eocene age, and above are the Miocene sandstones with molluscan fossils. The shales here described have elsewhere been called the "Monitos shales," probably representing the Oligocene. Crossing the Rio Canalete somewhat above its mouth, and extending thence into the hills to the east of Cordoba, on the Rio Cordoba, there is a series of dark clay shales and sand- stones from which issue many seepages of light oil. This series is not only highly folded and perhaps faulted, but, more- over, the strata are much crushed and crumpled and in places reduced to a structureless complex. Overlying these beds along the coast and extending to the Bay of Arboletes, there are steeply inclined Miocene sandstones and shales with many well preserved fossils. Near the Bay of Arboletes and near the contact of the two sedimentary series is the great "mud volcano" of this district, rising about 75 feet above the coastal terrace, and covering some 40 acres of area. Much gas escapes from the pool of mud at the top, smelling strongly of petroleum. Not far away outcrop the underlying shales in which are found seepages of oil, and which are probably the source of the gas. The same body of shales extends along the coast for some miles toward the Gulf of Uraba. That this series of shales and standstones from which issue the oil and gas belongs to the Poso series there can scarcely be a doubt, although no fossils were found in it. 32 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. A nearly parallel zone of the same series of strata crosses the Quebrada del Aguila, a tributary of the Rio Canalete, about 15 miles east of the Bay of Arboletes. The locality is known as El Aguila, and is on the coast side of the Palomas range. Here hard sandstones and shales are well exposed, though much broken and faulted, and standing at a high angle. Five or six miles south of El Aguila similar shales and sand- stones are exposed in the bed of a small stream, and are less broken by faulting. The strike is about N. 30° E., and the dip is not less than 75° to the northwest. About 1,000 feet of strata are exposed here, from which seepages of oil and gas are issuing. Three miles to the north are the mud volcanoes of San Diego, which cover not less than 40 acres of area. These vents have brought to the surface many fragments of hard sandstone, calcite, limonite, lignite and other mineral debris. The water escaping with the gas is slightly saline. Many other examples of these formations could be given, though they seem unnecessary. One of their chief characteris- tics is the presence in them of seepages of petroleum and gas, and the accompaniment of the well-known mud volcanoes of this region. This characteristic, together with their frequent stratigraphic position between Eocene rocks below and often fossiliferous Miocene beds above, serves for their identification even where stratigraphic evidence is not complete. The oil is believed to be largely indigenous, though in part it may have originated in the underlying Eocene formations, which contain foraminiferal and other organic strata, and in some places are bituminous, though to a less degree than the strata of the Poso series. Structures. The structural conditions of the Poso series have been already suggested in the foregoing notes and quota- tions. As a whole the series is highly folded, if not faulted, and it has been much denuded subsequent to its folding. In the range of foothills west of the Rio Sinu, where the series was most studied, there are found two or more somewhat com- pressed anticlines with intervening synclines on the east slope of the Palomas range, and as many on the westward, or coast slope of the same. Such a fold is found in the vicinity of Vol. XVIIIJ ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA §3 Arboletes Bay, and another farther inland. Still others are known in the vicinity of El Aguila and the Lorencita. Within these highly folded areas of the Poso series other strata both older and younger are involved, and in such cases the boundaries are often uncertain. In fact it would not be easy to disentangle the several series even were the country less covered with jungle and more accessible by roads than it is. The amount of faulting that has affected these Tertiary areas is not known, though there are many evidences that faulting even on a large scale has disturbed various sections of the country. One such fault has long been recognized, and appears in the section drawn by Beck (p. 465). This is probably the fault that traverses the west foot of the Coloso range, and is known as the "Bolivar fault." The full extent of this fault has not been ascertained, though it is not confined to the locality of the Coloso range. It extends from here southward toward Monteria, and northward toward San Cayetano, and may even connect with the faulting west of Arenal and of Usiacuri. Stratigraphic relations. The stratigraphic relations of the Poso series to the beds above and below have already been suggested in the foregoing paragraphs. Near Lorica in the Cerro de Cispata as well as in the Cerro de San Sebastian, the Poso series is found resting unconformably upon, or against, the cherts and other rocks of the Eocene. Along the west foot of the Coloso range the Bolivar fault complicates the problem by cutting the formations near the line of boundary between the Eocene and the Poso series, yet the lithologic contrast in the two is easily recognized. Also in the conglomerates of the latter are found many pebbles and boulders of the cherts that characterize the former. No other source than the strata of the Eocene appears to be possible for the pebbles of chert and jasper found in the conglomerates of the Poso series, and this fact, in the absence of direct evidence as to the age of the lat- ter, is sufficient to show that this series is at least post-Eocene. On the other hand, the Eocene is often richly fossiliferous in both Mollusca and Foraminifera, while the Poso series, with the exception of certain genera of the latter, is rather 84 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. poor in fossil remains. In the section drawn by Werenfels^ for the district of Toluviejo, which possibly applies equally well to that of the lower Sinu valley, the "Toluviejo series," with its fauna of Lepidocyclina and Numulites species, is ten- tatively placed by him in the upper Eocene, though most of the genera mentioned in his text seem to have been found in the middle Oligocene of Santo Domingo.^ The "Pacini shales" of his section, for which he estimates a thickness of over 3,200 feet, are possibly in part within the Poso series of the present paper, and, moreover, he assigns them to the Oligocene. The lower part of the Poso series, as found near San Antero, con- sists of calcareous concretionary shales as shown by Mr. Ruckman. The stratigraphic relations of his several "series," one to the other, are not stated by Werenfels, nor are they indicated in his section. It is not possible, therefore, to fix their position in the scale of the present plan with much confidence, though some suggestions may be offered regarding them. The cor- relation of the lower part of the Beck column with his "Pacini shales" appears to be erroneous, since the Tofeme formation of Beck is undoubtedly Eocene in age, as shown in a former paper.^ May it not be possible that the "Toluviejo series" of Werenfels is only the lower part of the Poso series, and that the "Pacini shales" correspond to the upper part? R. H. Liddle has given a "Composite geologic column" for western Venezuela,® in which the "Oligocene" strata of the Maracaibo basin are shown as having a maximum thickness of 5,500 feet, of which the Pauji shales, the lower part, con- stitute more than half. Only a few mollusks and Foraminifera (chiefly Lepido- cyclina) are mentioned to "indicate that the fomiation is of marine and not of deltaic origin." This group is followed historically by an uplift and erosion interval, while upon it, in some places, rests 1,000 feet of mas- sive coralline limestone and sandy beds, the San Luis forma- tion. Overlying this group is that of the Agua Clara shales, " Werenfels, A., Eclogae geol. Helvet., vol. 20, 1926, pp. 81-83. « Vaughan, T. W., and Woodring, W. P., Geol. Surv. Domin. Rep., Mem. vol. 1, 1921, pp. 107, 108, etc. ^Anderson, F. M., Proc. Calif. Acad. Sci., vol. 17, 1928, p. 4. « Liddle, R. A. The Geology of Venezuela, etc. 1928, pp. 54, 241, etc. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA g5 sometimes 1,500 feet in thickness. These are described as "dark-gray, sandy, micaceous, locally very fossiliferous shales, which gradually become more sandy toward the top," and passing without visible structural break into the Cerro Pelado formation (Miocene) consisting of "massive or flaggy and shaley sandstones interbedded with arenaceous lignitic shale." Each of these groups is discussed at length in the body of the book, and some indications given as to the faunas of each, to- gether with notes as to their correlations. Without offering any final judgment as to the faunas and the correctness of the correlations, it may be remarked in pass- ing that the lists of molluscan genera and species given as representing the Agua Clara formation suggest its Miocene age, rather than Oligocene, and its equivalence, in part at least, to the Tubera group described later. These remarks do not apply, however, to the whole of the San Luis formation, which, according to Liddle, seems to be conformably overlaid by the Agua Clara group. Concerning the Pauji shales, and possibly a part of the San Luis formation, with the large Foraminifera Lepidocyclina species, there should be less question as to their Oligocene age. Their stratigraphic body and their fauna both seem compar- able to the middle Oligocene of Santo Domingo, as described by Vaughan and Woodring.® Along the Colombian north coast the Poso series described in the preceding pages is regarded as directly comparable to the latter, and therefore, also to the Pauji shales and related strata of western Venezuela. Age of the Poso scries. Unconformable relations between the Poso series and the underlying Eocene have already been shown at the type locality of the former near San Sebastian, and in the Cerro de Cispata northwest of Lorica. In the con- glomerates of the Poso series on the east slope of the Cerros de Las Palomas are found the cherts and other rocks of the underlying Eocene. Such facts are noted in other parts of the country. • Vaughan, T. W., and Woodring, W. P., Geol. Surv. Domin. Rep., Mem. vol. 1, 1921, pp. 107-108. g^ CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Similar relations between the Poso series and the overlying Miocene were also pointed out in certain localities. Local evidence of such unconformity was found near Lorica as is shown in the report of Mr. John H. Ruckman, and near San Andres as noted on a preceding page. As for the definite assignment of the Poso series to the Oligocene at the present time there is some reservation. It may be in part Miocene, though there are reasons for believing that the larger part of it is older. The series as a whole is clearly post-Eocene as has been said, and in view of the occur- rence of undisputed Oligocene in other Antillean regions it should be expected to occur here also in commensurate volume. The general absence of molluscan fossils, which are abun- dant in the Miocene of north Colombia, the more varied lithology of the Poso series, as contrasted with the known Miocene, the frequent occurrence of petroleum or its indica- tions, not observed in the Miocene, and other features that could be mentioned, all suggest not only a different but older age than the Miocene of either of the groups that are de- scribed below. The Miocene Series Regarding the occurrence of Miocene deposits in Colombia, there is more satisfactory evidence than that regarding the Olig- ocene. On the geologic map of North America Willis shows later Tertiary deposits widely distributed over the northern parts of South America, particularly in the valley of the Orinoco, about Lake Maracaibo and in the valleys of northern Colombia, ex- tending far into the interior of the country, along the Magda- lena, the Cauca, the San Jorge and the Cesar rivers, about the Gulf of Uraba and along the west coast. Thence they extend into other countries bordering the Caribbean Sea. The areas actually covered by Neocene deposits in Colombia are much smaller than that shown on the map, and strictly are confined to relatively narrow zones along the coasts and along some of the larger rivers. For example, marine deposits of Neocene age do extend along the valley of the Magdalena in more or less continuity to the delta areas at the mouths of the Rios Sogamoso and Carare, where marine deposits give place Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA g/ to only partly marine in the Oponcito group. Above this the Miocene deposits are continuous but transitional in character until they connect with the non-marine deposits of the Barza- losa group of the upper Magdalena previously described. ^° A part of the marine Miocene strata of northern Colombia has already been described in earlier papers, though not the entire series. In fact, no complete statement of the marine sequence or of its distribution can be made at present. As for their distribution, the known Miocene deposits extend east- ward from the Gulf of Uraba along the Colombian coast to the Sierra Nevada de Santa Marta, and beyond this range they occur again near Rio Hacha, and according to accounts they extend from there southward into the valley of the Rio Cesar, very possibly to its mouth where it connects with the Magdalena. At any rate they are believed to fill the entire valley above its mouth. Washburn and White^^ have given a thick section of Terti- ary sediments as occurring in the valley of the Rio Cesar, a large part of which is given a position between the lower Tertiary and the late Pliocene, but as no reference is made to fossils, it is impossible to conjecture what portion of the Mi- ocene column is represented in the section. Huntley and Mason" also give an immense section of pre- sumably marine Miocene strata (after Bossier) as occurring in southwestern Colombia along the Pacific coast. Some of the sandy shales contain fossils, but there is no attempt to in- dicate what part of the Miocene they represent, if, indeed, it is known. Eastward from the Gulf of Uraba the marine Miocene de- posits are not quite continuous, and are, moreover, involved with older formations and are known only in part, as will be shown later. Along the lower stretches of the Magdalena north of Mompos fossiliferous marine Miocene deposits underlie most of the surface, but in turn are also overlaid by later deposits, partly land-laid and partly marine. From the Magdalena the "Anderson, F. M., Bull. Geol. Soc. Am., vol. 38, 1927, pp. 612, etc. " Washburn, C, and White, K. D., Tr. Am. Inst. Min. Met. Eng., vol. 68, 1923, p. 1026. " Huntley, L. G., and Mason, S., Tr. Am. Inst. Min. Met. Eng., vol. 68, 1923, p. 1018. 88 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Miocene deposits extend westward into the valley of the Rio San Jorge, and from there they pass into the valley of the Sinu, which they occupy in part. About the lower Sinu valley they enter into the composition of the lower hills near the coast, and possibly connect with the deposits along the coast about the Gulf of Morrosquillo. Miocene deposits overlie the Poso series near San Onofre and southward from this village toward Tolu and the Bay of Cispata. Along the coast to the southwest of Cispata Bay they appear again near Punta Piedras, Monitos, Bruquelles, Mangle, the Bay of Ar- boletes and farther toward the Gulf of Uraba, and along the Atrato river. In all these points beyond the Bay of Cispata the strata stand at a high angle dipping toward the sea, and with a strike nearly parallel with the coast line. For the most part they appear to be only sparingly fossiliferous, though enough fos- sils have been found for the definite determination of the mid- dle part of the Miocene. On the Ouebrado de Murindo, a tributary of the Rio Canalete, some 15 miles from the coast, fossiliferous beds occur, standing at high angles, as will be described later, from which numerous molluscan species have been obtained. In the districts about the lower Magdalena the Miocene deposits attain a great development, and a thickness much in excess of that found by the writer in other parts of the Co- lombian coast. In a former estimate of an incomplete section to the west of the river the thickness was given as 5,400 feet, or more. Other writers have given the thickness of the Miocene series in certain parts of the country as near 8,000 feet, but without detailed infonnation as to the strata or the contained faunas. Later study of the section in the district west of Barran- quilla necessitates some modification of the divisions formerly proposed, since the apparent thickness is somewhat increased by faulting. Briefly, three distinct groups of strata have been recognized here as shown below, of which the central group constitutes at least half the entire series as known at present. They are approximately, as follows: Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA gg Galapa (La Popa) group 1650 feet Tuberd group 2650 feet Las Perdices group 1000 feet Total 5300 feet Las Perdices group. In the earlier statement^' referred to above there is a brief description of some 400 feet of strata outcropping near Las Perdices; about 15 miles west of Bar- ranquilla, which appeared to be of Miocene age, but which also appeared to be separated from the overlying Tubera group by a disconformity. No definite name was proposed for these beds, but in the present paper the above name is proposed. The group as here exposed consists of clay shales, sandy shales and hard cherty, or siliceous beds and some sandstone. The shales contain at this locality a few species of Mollusca, scales of fishes and bone fragments, sponge spicules and numerous Foraminifera, as mentioned in the former account. Samples of these shales were examined by Dr. G. Dallas Hanna, and his note regarding these forms is here included for completeness : "The shales contain a very considerable number of fossils, the groups being represented about as follows in order of abundance: (1) Radiolaria; (2) Diatomaceae; (3) Foraminifera; (4) Sponges; other organisms are scarce. There has been pyritization to a considerable extent and many of the chambers of the fossils are filled with iron sulphide. A great many of the diatoms have been replaced entirely and internal casts of the frustules are abundant. Coscinodiscus was the only genus definitely identified in this group. Many of the genera and some of the species of Radiolaria are the same as have been found in the famous deposit on Barbados Island and which Payne has put definitely in the Miocene. Some of the genera are: Stylodictya, Histiaslrum, Siylosphcera and Eucyrtidium. Foraminifera are scattered rather sparingly through the mass of the material, the common genera being: Glohigerina, Orbulina, Lagena, Truncatulina, Cassidulina, Nodosaria, Anomalina, Fron- dicularia, Plectofrondicularia and Bolovina. It is believed that these organisms oflfer a means whereby a definite correlation can be made with strata of known age elsewhere. This preliminary examination indicates that the formation lies very close to the base of the Miocene, if, in fact, it is not the lowermost part of the sediments of that period." A few miles to the north of this locality and west of Puerto Colombia, similar shales are exposed along the sea cliffs for a mile or more, with a strike of nearly east to west, and a dip "Anderson, F. M., Proc. Calif. Acad. Sci., vol. 16, 1927, p. 88. March 29, 1929 90 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. toward the south of 40° to 60°, and are here overlaid by fos- siliferous sandstones of the succeeding group, which also dip southward. The underlying shales contain a variety of micro- organisms, among which are Foraminifera, scales of fishes, the following molluscan fauna and coral : Cancellaria, new species. Turris albida (Perry) Afi/ra maMry« Anderson, new species , Cassis {Phalium) dalli Anderson, Scobinella moriereiQ) (Laville) new species Polinices prolactea Anderson, new Drillia eupora Dall species Dentalium granadanum Anderson, Psammobia (Gari ?) new species Cyathomorpha sp. While most of the species are new, and therefore not at present serviceable for correlation, yet they are definitely of Miocene aspect; a few of them indicate a low position in this series. From the stratigraphic evidence they clearly belong beneath the Tubera group, and are regarded as a northward extension of the Las Perdices group. Some 10 miles to the west of Barranquilla, and extending to the southwest, the lowest beds of the Miocene are brought to the surface along the axes of a series of anticlinal folds, faulted in part, extending from near Puerto Colombia to the vicinity of Cienega de Oro, a total distance of over 100 miles. Beds believed to be Oligocene are also brought up beneath the Miocene. Near the village of Usiacuri the lowest beds exposed con- sist of clay shales, shaley sandstone, and calcareous layers, in all some 600 feet in thickness which constitute a distinct strati- graphic group. These strata are here rather poor in molluscan remains, though microscopic marine organisms have been noticed in some of them. From such remains as have been found they are believed to be Miocene in age, and in part equivalent to those exposed along the beach west of Puerto Colombia, and at Las Perdices, or in other words to represent the Las Perdices group, as described above. Near the top of this group at Usiacuri, springs of sulphur- ous water issue from the strata, which give to this village its repute as a health resort. The water is bottled and sold in the neighboring towns as a health beverage. Here the lower group terminates above by a lithologic change in the character Vot. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA gi of the sediments, which become suddenly more sandy, and at the same time they also acquire a rich fauna of marine Mollusca. The line of separation between the Las Perdices group and the succeeding group here is probably near the springs of sul- phurous water, or immediately below the village, which is situated on the east flank of the fold. No angular uncon- formity in the strata was found here, though it is suggested by the lithologic change, the abrupt appearance of the marine Mollusca, and by the springs of sulphurous water. The thickness of the Las Perdices group is not at present known, though between Usiacuri and the axis of the fold to the west the exposed thickness of strata is probably not less than 1,000 feet. In other parts of the country it is believed to be greater. From a comparison of the three localities thus far studied it can be said that a disconformity is indicated, and that it probably can be fully demonstrated by further work in this field. Olsson described a disconformity between the Uscari for- mation of Costa Rica and the overlying Gatun,^* and an over- lap of the latter upon the older rocks of the region. Similar relations exist with regard to the Tubera group as was shown by Mr. Ruckman's account of the district about the lower Sinu valley. The stratigraphic position of this disconformity in the Colombian Miocene seems to be lower than the base of the Gatun group as found in the Canal Zone. However, this dis- conformity has not been shown to exist in the Canal Zone, unless the Emperador limestone should prove to belong prop- erly to a higher horizon than has usually been conceded for it. Vaughan has suggested that it may possibly find a place among the equivalents of the Langhian (Burdigalian) of Europe.'^^ May it not also be possible that the Uscari forma- tion of Olsson and the Las Perdices group of the present paper, when fully known, will find a similar place in the se- quence of Antillean stratigraphy ? The Tubera group. In the earlier paper to which reference has been made the writer gave a brief summary of the " Olsson, A. A., Bull. Am. Pal., vol. 9, 1922, p. 784. ^^I "Vaughan, T. W., Bull. Geol. Soc. Am., vol. 3S, 1924, p. 731. '. 92 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Colombian marine Miocene deposits as found in the vicinity of the lower Magdalena valley. On the basis of its fossil zones it was divided into horizons, lettered respectively from M to T in ascending order. The name Tub era group was first suggested for this sequence of strata in 1926^"^ but without any definite delimita- tion. Later the name was employed in a more definite treat- ment/^ and while recognizing the three distinct fossil ho- rizons, namely M - N, P, and R, the faunal contents of only the lower, M - N, was given, consisting of some 64 species of Mol- lusca. A tentative correlation of this and the succeeding ho- rizons was suggested, but without elaboration, since for the two upper horizons no faunal lists were given. The sequence of strata embraced in the Tubera group has a thickness of not less than 2,650 feet. It consists for the most part of incoherent sandstones and sandy shales, divisible into some local lithologic members, though none that seems to have any great areal extent. No conspicuous and essentially organic members have been discovered. The fossil horizons probably have greater geographic range and stratigraphic value. The group is well represented about Tubera mountain and its environs, whence the name. Of the sequence forming this group, horizon M - N is, at its type lo- cality, confined to the lower 550 feet. Horizon R falls within the upper 600 feet, while horizon P occupies a position near the middle, and is probably embraced within a stratigraphic range of 300 to 400 feet. Between these several horizons the beds are somewhat bar- ren of fossils, in the immediate district about Tubera moun- tain, and in fact as far as known elsewhere along the coast. In its geographic distribution the Tubera group extends over a wide region, and it appears to represent the more usual facies of the Colombian Miocene, whereas the older group has been definitely detected only within restricted areas. Within the limits of north Colombia this group lias been recognized at such distant points as the Gulf of Uraba, Arboletes Bay, Rio Canalete, Lorica, San Andres, Zambrano, El Banco, Tur- baco, Cartagena, Punta Pua, Tubera mountain, and along the "Anderson, F. M., Bull. Amer. Assoc. Petrol. Geol., vol. 10, 1926, pp. 387 & 399. "Anderson, F. M., Proc. Calif. Acad. Sci., vol. 16, 1927, pp. 87-90. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 93 west flank of the Sierra Nevada de Santa Marta. However, it is believed to extend much farther, as into the valleys of the Rio San Jorge and the Rio Cesar. Only a few of the locali- ties in which the group occurs can be considered in detail at the present time. Local occurrences. Among the several districts in which the Tubera group has been proved is that of the upper drain- age of the Quebrada Murindo, a tributary of the Rio Canalete draining the west slope of the Las Palomas range. The district lies some 12 to 15 miles from the coast and somewhat farther from Monteria. Mr. K. D. White, who visited this district, de- scribes in detail a sharply folded anticline traversing it in a north to south direction, on the opposite sides of which he gives stratigraphic sections respectively 3,000 and 5,000 feet in thickness. Of these B - B is much the less complete, since it does not reach the axis of the fold. Section C - C crosses the axis upon which are found various seepages of oil, not found on the other. Of the latter section some 2,300 feet of the lower part is not fossiliferous. Fossils are found throughout section B - B, but through only the upper part, 2,700 feet, of section C - C. These sections are respectively represented by the numbers 354 and 355, from which were obtained the following partial lists of species : Loc. 354 (C. A. S.) Pitaria tryoniana (Gabb) Cardium dominicense Gabb Cardium venuslum (?) Gabb Chatna scheibei Anderson Pecien vaginulus (?) (Dall) Cyclinella gatunensis Dall Conus consobrinus Sowerby Conus molis Brown & Pilsbry Turritella altilira Conrad Fusinus henekeni (Sowerby) Terebra cirra Dall Serpulorbis sp. Loc. 355 (C. A. S.) Pitaria cora (Brown & Pilsbry) Cardita scabricostala Guppy Cardium lingualeonis (?) Guppy Cyclinella gatunensis Dall Tellina cibaoica (?) Maury Area trinitaria Guppy Polinices subclausa Sowerby Oliva gatunensis Toula Potamides avus Brown & Pilsbry Bullaria paupercula Sowerby Strombus proximus Sowerby Strombina sp. Many other species could be added to these lists, but the number is perhaps sufficient. The lithologic character of the strata from which they come is similar to that of the Tubera 94 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. group, and is in contrast with the underlying barren beds in which the seepages of oil occur along the axis of the fold. Near San Andres the Tubera group is represented by sun- dry localities, containing representative species, as the following : Loc. 302 (C. A. S.), four miles Loc. 303 (C. A. S.), three miles south of San Andres east of San Andres CylichneUa gatunensis Dall Chione walli Guppy Mactrella elegans (Sowerby) Tellina gatunensis (Toula) Natica guppy ana Toula Surcula servata Conrad Architectonica gatunensis (?) Toula Area sp. Loc. 350 (C. A. S.) Arboletes Bay Tivela mactroides (Bom) Bullaria paupercula (Sowerby) Cardium lingualeonis Guppy Olivella indivisa Guppy Cardium haitense Sowerby Potamides avus Brown & Pilsbry Chione mactropsis (Conrad) Bittium adele Dall At the hamlet Jesus del Monte, between Carmen and Zam- brano, near the base of the Miocene were obtained : Turris albida (Perry) Natica guppyana (?) Toula Cancellaria sp. Turritella altilira (?) Conrad Area sp. Glycymeris sp. At the village of El Banco on the Rio Magdalena, some 170 miles above Barranquilla, a zone of crystalline rocks crosses the course of the stream. On the east flank of this zone at the mouth of the Rio Cesar, and immediately beneath the village, there are soft yellowish brown sandstones overlaid by blue clay shales forming a part of a thicker series which presum- ably rests upon the pre-Tertiary crystalline rocks, which crosses the river to the west. The sandstones have a gentle dip, 6° to 8°, to the eastward. One stratum is largely com- posed of broken and decomposed marine shells, but beneath this are standstones from which better preserved fossils may be obtained. Only a few species were collected, but a number of genera were recognized in these beds, including. Area, Glycymeris, Chione, Ostrea, Anomia, Pecten, Olivella, Tur- ritella, Terebra, Phos, Polinices and many others. None of the species characteristic of the lower horizon of the Tubera group were found, while nearly all of them were such as are found abundantly in the higher beds, horizon P of this group. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 95 In view of the occurrence of the older crystalline rocks to the west, and the easterly dip of the Miocene beds, this occur- rence may be regarded as belonging to the Tertiary area of the valley of the Rio Cesar, rather than to that of the lower Magda- lena. The crystalline rocks here may be interpreted as form- ing a connecting link between the pre-Tertiary area of the Sierra Nevada and that of the Cordillera Central, as stated elsewhere. Comparison of horizons. At most places in Colombia where the Miocene beds have been noted by other writers they have been indiscriminately mentioned as representing the Gatun formation of the Canal Zone, though the definite basis for this view has not been given. However, in truth, most of the accessible exposures do represent horizons above that of M - N, the lowest part of the Tubera group. Whether this fact is due to overlap of the later horizons beyond the limits of the lower, or to other circumstances of deposition can not' now be stated. On the basis of faunal content only the middle portion of the Tubera group should be regarded as the equivalent of the Gatun formation of the Canal Zone. The expansion of the name "Gatun" to include all of the Miocene sequence, even where the sequence is a conformable series, does not appear to the writer as justifiable. The number of molluscan species obtained from the entire group by the writer has not exceeded 165, though from lists published by Dr. Pilsbry and others the total number could be considerably increased. Of the entire number obtained 38 species are added in the present contribution as new species, and doubtless many others will subsequently be found. The stratigraphic range of many of these forms is of course not known at present. Some of them doubtless range throughout the Miocene while others are of short stratigraphic duration. For the purpose of correlation a list of 86 of the better known species have been selected from the total number as being most representative. This list segregates the species as to horizons, as far as known at present. Little more than a tentative attempt is claimed for the segregated lists as they here appear. 96 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Species Terebra sulcifera Sowerby Terebra gatunensis Toula Terebra cirra Dall Terebra haitensis Dall Terebra bipartita Sowerby Conus sewalli Maury Conus imitator Brown & Pils Conus molis Brown & Pils Conus recognitus Guppy Conus planiliratus Sowerby Conus stenostomus Sowerby Turris albida (Perry) Drillia eupora Dall Cancellaria dariena Toula Cancellaria guppyi Gabb Cancellaria cossmanni Olsson Turritella altilira Conrad Turritella perattenuata Heilp Turritella fredeai Hodson Turritella mimetes Brown & Pils Turritella gatunensis Conrad Turritella cartagenensis Brown & Pils Crucibulum gatunense (Toula) Architectonica granulata (Lamarck) . , Architectonica quadriseriata (Sow.). . Natica guppyana Toula Natica cuspidata Guppy Polinices subclausa Sowerby Calliostoma grabaui Maury Calliostoma olssoni Matuy Oliva cylindrica Sowerby OUva sayana Ravenel Oliva brevispira Gabb Marginella ballista Dall Marginella conformis Sowerby Mitra dariensis Brown & Pils Mitra longa Gabb Scobinella morierei (Laville) Fasciolaria kempi (Maury) Fusinus henekeni (Sowerby) Murex domingensis Sowerby Murex mississippiensis Conrad Tuberd Group M-N R Other Regions Cer-I Ga-lTam- cado tun pa Vol, XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 97 Species Typhis siphonifera Dall Distortrix simillima (Sowerby) Cyprea henekeni Sowerby Cyprea gabbiana Guppy Malea ringens (Swainson) Sconsia laevigata (Sowerby) Strombina chiriquiensis Olsson Serpulorbis papulosa Guppy Serpulorbis granifera (Say) Petaloconchus sculpturatus Lea . . . . Area patricia Sowerby Area macdonaldi Dall Area actinophora Dall Area dariensis Brown & Pils Area lloydi Olsson Glyeymeris jamaieensis Dall Glyeymeris earbasina Brown & Pils. Glyeymeris lamyi Dall Ostrea megadon Hanley Pecten mortoni Ravenel Peeten demiurgus Dall Peeten pinulatus Toula Pecten bowdenensis Dall Spondylus bostrychites Guppy Crassatelites densus Dall Venericardia brassica Maury Cardita arata (Conrad) Cardita scabricosta Guppy Echinochama antequata Dall Cardium domingense Gabb Cardium lingualeonis Guppy Cardium gorgasi Hanna Cardium serratum Linnaeus Cardium venustum Gabb Dosinia delieatissima Brown & Pils. Dosinia acetabulum (?) Conrad. . . . Clementia dariena (Conrad) Cyclinella gatunensis Dall Cyclinella eyeliea (Guppy) Antigona caribbeana Anderson Antigona blandiana (Guppy) Callocardia gatunensis Dall Tuberd Group Other Reg Cer- Ga- M-N P * R eado tun * * * * * * * * * * * * * * * * * * * * * * * *? * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *? * * * * Tam- pa 98 CALIFORNIA ACADEMY OF SCIENCES [Phoc. 4th Seb. Tubera Group Other Regions species M-N * * * * * * * * * * * * * * * R Cer- cado Ga- tun * * * * * Tam- pa Pitaria circinata (Bom) Pitaria cercadica Maury * * * * * Macrocallista macula ta Linnaeus * Chione nuciformis Heilprin Chione mactropsis (Conrad) Chione latilirata (Conrad) Tellina dariena Conrad * Tellina gatunensis (Toula) Semele sardonica Dall * Mactrella elegans (Sowerby) Labiosa gibbosa (Gabb) Labiosa gardnerse Spieker Galapa-La Popa group. In the table of correlations here in- cluded, above the uppermost horizon of the Tubera group there is a considerable sequence of beds the exact position of which in the column may be subject to debate. On account of their apparent conformity with the Tubera group they are here regarded as of Miocene age, though they may be younger. Such beds are found in the neighborhood of Galapa to the south of Barranquilla, and also at the base of La Popa hill near Cartagena. Near Galapa they consist of little con- solidated beds of calcareous sandstone, while at La Popa hill and about the Harbor of Cartagena they consist of well- stratified but somewhat incoherent sandy shales and clays with calcareous layers of marl. In the former locality the strike is generally N. 20° E. and the dip is easterly. The thickness is not definitely known, though an estimate of 1,650 feet is believed to be conservative. They are rich in marine fossils, among which Pecten pre- dominates although Doshiia, Cardium, and various gastropod forms have been found. The La Popa formation found in the vicinity of Cartagena has an aggregate thickness of 1,000 feet, or more, though it is not well exposed. The structural condition exhibited in these deposits is at variance with those of Galapa, in that the dis- Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 99 trict is traversed by east to west faults that produce scarps of some prominence, as seen in the south face of La Popa hill itself, and in the north face of the hill of Cospique on the east side of the Harbor. In the syncline lying between the Tubera-Piojo uplift and the coast similar beds are found of which the contained fossils cannot now be given. These beds do not appear to cover the general areas of the older Miocene, but are local and are, as far as known, con- fined to districts near the present coast. None have been ob- served far inland. Not only are they conformable upon the Tubera group in the districts where they have been observed, but they participate in the structural features of the latter. From the fact that they are not coextensive with the Tu- bera group, but are local in their occurrence, it may well be sumiised that they do not form a continuous series with it, but may be separated from it by an unconformity the signifi- cance of which should not be overlooked. Possibly an uplift of the land areas near the close of the Miocene excluded the sea from the larger part of the region previously covered by it. For these reasons it would be well to reserve final judg- ment as to the proper position of the Galapa group until more data are obtained than the writer possesses at the present. Pliocene Deposits Contrary to the view expressed in the preceding paragraphs, the late Miocene epoch has often been regarded as one of up- lift for the general region of the Caribbean. This was at one time apparently the view of Dr. Vaughan,^^^ in whose conception an extensive emergence of land areas in late Miocene time was followed by warping and local submergence during the Pli- ocene, concerning which he says in part : "Subsequent to the Miocene there have been many oscillations of the West Indian area, and during perhaps Pliocene time there was profound de- formation." In the same paper Dr. Vaughan regards the Toro limestone of the Canal Zone as of Pliocene age, and with it classes "a Vaughan, U. S. Nat. Mus. Bull. 103, 1919, pp. 608-609. IQQ CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. also certain deposits of Limon, Costa Rica, and others far to the east. Concerning the district about the lower Magdalena with which the present paper deals, he says (p. 594) : "Mr. George C. Matson collected at Barranquilla, Colombia, some fossils that belong to a fauna younger than that obtained around Usiacuri, and may be of Pliocene age." The rocks classed tentatively as Pliocene by the present writer are abundant around Barranquilla and the mouth of the Magdalena. A good section is found along the railroad be- tween Puerto Colombia and Salgar. The strata here un- dulate, but on the whole dip 10° eastward along the shore. The following sequence is the result of careful study of the beds exposed here : d. Upper coral limestone 250 feet c. Incoherent sandstones 350 feet b. Lower coral limestone 160 feet a. Sandy clay shales 150 feet Total thickness 910 feet These limestones contain a great variety of corals and many Mollusca including Cyprea, two species of Codakia, many species of Pecten, oysters, and various gastropods. The coral limestones resemble that in the quarries at Barranquilla, and, in fact, their connection is not difficult to trace on the surface. In these quarries which are worked for lime, there is a greater variety of corals than on the beach, and also of Mollusca. Here and in most places the corals and shells are largely re- duced to the condition of soft marls in which are some harder layers and lenses of coral rock. These beds may be followed along the Galapa road for many miles, where they are almost always horizontal in attitude. They seem to have been at one time more extensive toward the west but have suffered much denudation, leaving the limestone more or less local in its present occurrence. Quite similar beds cover the top of La Popa hill near Car- tagena, but here rise to an elevation of some 500 feet above the sea and have an inclination of about 15° or more toward the north. They form here a distinct reef, 75 to 80 feet in Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA IQl thickness, resting upon marly shale of about equal thickness which is underlaid by the sandy shales of the La Popa forma- tion. Beds of the same character cover the top of the hill of Cospique, and occur also at Turbaco at an elevation of about 500 feet above the sea. Corals and molluscan shells are abun- dant in all of these points, and are usually reduced to charac- teristic marl. These supposed Pliocene deposits with coralline reefs of the sort here described occur at intervals along the Colombian coast, apparently not always resting upon the same horizon of the Miocene. Such beds are found on the island of Terra Bomba, Isla de Baru, Bayunca, and at points beyond the Bay of Cispata. The general attitude of these coral reefs and the associated beds does not appear to be harmonious with the underlying Miocene. They were not observed above an altitude of 500 feet, while the Miocene often rises to much greater heights. The deposits appear to be in some respects, and in some places, unconformable upon the underlying Miocene, though a clear case of unconformity was not found. With regard to age there are some general stratigraphic facts that may be mentioned. Elevated beaches and late Quaternary deposits of beach origin skirt the hills near sea level, and Quaternary gravels form old valley floors in many parts of the country and along the coasts. Such deposits are nearly always horizontal, and clearly have no relation to the supposed Pliocene deposits, except to show their distinctly more recent origin. Only a few of the fossil corals so abundant and varied in these deposits have received any attention. Three species only have been noted from the reef on La Popa hill. On a visit to Cartagena in 1898 the Princess Theresa von Bayern personally went to the summit of La Popa hill and collected four speci- mens of coral from the reef that caps the same. These corals were left at the Academy of Sciences at Munich, and were later described by Herr Johanes Felix, under the following names : Orhicella theresiana Felix IsastrcBa turbinata Duncan Stephanocmnia fairbanksi (?) Vaughan 102 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Concerning the first of these species Dr. Vaughan says that it is ''probably a synonym of Solenastrea bournoni M. Ed- wards & Haime." FeUx was unable to reach any conclusion as to age from his study of these corals, though he thought they were probably Miocene. Correlations While exact correlations of the Colombian Tertiary groups and horizons with others of the Caribbean and Central Ameri- can regions can not yet be made with complete confidence, a tentative attempt, based upon known facts may be well worth while. On the whole the Miocene series and groups seem to cor- respond fairly well with those of Santo Domingo, as for ex- ample the Yaque group, with the possible exception of its low- est member, the Baitoa formation, containing species, of Orthaiilax and associated forms. The fauna of the Las Per- dices group is not yet well known, but with further search it may well prove to be the equivalent of the Baitoa, as suggested in the accompanying table. Horizon M - N of the Tubera group lacks the species that characterize the Baitoa formation, and that are found in similar lower Miocene deposits of the Gulf Coast which have been correlated with it. On the other hand, a comparative study of its fauna shows horizon M - N of the Tubera group to be more closely related to the Cercado formation of the Yaque group than to any of the others, as the following statements will show. Of the 64 moUuscan species thus far found in this horizon, only 15 appear in the list from the Gatun formation given by Brown & Pilsbry, as enumerated by Vaughan. Of the species found in the Cercado formation, according to Maury, something more than 5 per cent are found also in the recent Antillean fauna. Of the 64 species of horizon M - N, not more than four are also found in the living faunas of the Pacific and Caribbean seas, and the number may be less. In any case it will not exceed 7 per cent of recent species, and this estimate is liberal. Horizon P of the Tubera group shows even stronger re- semblance to the Gatun fonnation of the Canal Zone. Of the Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA JQS a; a % '■+J OS 'c o o c _a3 +3 Upper Langhian Lower Langhian G OS 'c 03 '3 < G 03 o3 •n o OJ -go ^— " '£. ^ O cd :=! c •>.2 o J3 o3 ■+-( •— > 4- 1 1 oS (U G G jjnia UiTilV c3 s O 3 t o i 1 H pq 03'-' OJ b o a: o o is C o o 1 is c vo a >5 dnojto ^aaqnx o e o Q o 02 Ow 2 c3 G C O 0.2 03 1 1 pqwW og (D G ^1 dnojQ anbBj^ a ^ S 03 U3 03 c3 o c mO CL 3 1/3 foo "d ^ ^ 1 1 111 1 p-o a G o d oj O W5 'O o §^ »H r>-. 2 ^^ s s c^ 0) XJ .G S • Olsson, A. A., Bull. Amer. Pal., vol. 9, 1922, pp. 183, 188, etc. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA JQS Description of Species On the following pages are noted most of the species that have been recognized in the marine Miocene groups of north Colombia, but without any claim of supplying an exhaustive list of the same. Without the aid of large collections of ma- terial from these groups that are available for comparison in other institutions of the country, much reliance has necessarily been placed upon published figures and descriptions which pre- sumably were intended to be adequate for this purpose. Some of the Miocene forms from the Carribbean region have, un- fortunately, been illustrated by unsatisfactory figures, but where this is the case the author of such has little ground for complaint if other writers fail to recognize his species. In many such cases later writers have gratuitously supplied bet- ter figures, and where this has been done recourse has been had to them. Photographic illustrations are thus available in the valuable contributions of Miss Maury, A. A. Olsson, W. P. Woodring, Dr. Pilsbry and his co-workers, and by others, so that one need not often go astray in his determina- tions of the better known forms. As might have been expected from the backward state of paleontologic study in the marine Miocene of South America, some new species have been brought to light, and when the material has justified it these new forms have been entered in the lists with proper description. In addition, a few forms already known from other Antillean regions have been illus- trated with, or without description when this has seemed desirable. The order in which the species have been taken up is almost without regard to any scheme of taxonomy, but merely that of a convenient arrangement of the forms noted. March 29, 1929 105 CALIFORNIA ACADEMY OF SCIENCES [Peoc. 4th Ser. Gastropoda 1. Terebra sulcifera Sowerby Terebra sulcifera Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849, p. 47; Miocene, Santo Domingo. — -Guppy, (part) Quart. Jour. Geol. Soc. Lond., vol. 32, 1876, p. 525, pi. 29, fig. 8; Loc. as above. — Maury, Bull. Am. Pal., vol. 5, 1917, p. 186, pi. 3, fig. 12; Loc. as above. This species is the largest of the Terebras found in the Caribbean Miocene, one incomplete specimen of 10 whorls measuring 95 mm. in length and 22 mm. in width near the base. If complete, this specimen would have a length of over 120 mm. In size, as well as in the sculpture of the mature shell, this form resembles T. petiti Maury (not T. petitii Kiener), though the younger shells clearly have the sculpture described by Maury for T. sulcifera Sowerby, and these fea- tures are shown in the younger whorls of all the examples. This species was found at Loc. 267, C. A. S., in horizons M - N and R, and accordingly at the base and near the top of the Tubera group, and presumably its range is throughout the same. 2. Terebra clethra ( ?) Maury Terebra clethra Maury, Monog. Foss. Ter. Brazil, vol. 4, 1925, p. 198-9, pi. 10, fig. 3; Lower Miocene, Rio Pirabas. Maury's type of this species was either of a small and rare form, or it was the earlier whorls of a larger species. The figure is said to have been drawn from a cast. Two specimens found near Usiacuri, Loc. 306, both incomplete, are 65 mm. in length, if entire. In form and ornamentation they resemble Maury's type too nearly to pennit their separation at present. 3. Terebra gatunensis Toula Terebra (Oxymeris) gatunensis Toula, Jahrb. der K. K. Geol. Reichs., Bd. 58, 1909, p. 705, pi. 25, fig. 14; Gatun formation, Canal Zone. Terebra gatunensis, Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 339, pi. 22, fig. 2; Gatun formation, C. Z.— Maury, Bull. Am. Pal. vol. 5, 1917, pi. 4, fig. 5; Cercado de Mao, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 208, pi. 1, figs. 4, 5, 6; Gatun stage. Canal Zone. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA JQ? This species was found in the day shale near the top of the Tubera group, horizon R, to the west of the Tubera mountain. 4. Terebra cirra Dall Terebra (Acus) bipartita Sowerby, variety cirrus Dall, Proc. U. S. Nat. Mus., vol. 18, No. 1035, 1895, p. 38. River Amina, Santo Domingo; Miocene. Terebra (Oxymeris) bipartita Dall, Trans, Wag. Fr. Inst. Sci., vol. 3, 1903, pi. 59, figs. 13, 28, 29; Miocene, Santo Domingo. Terebra cirra Dall, Maury, Bull. Am. Pal., vol. 5, 1917, p. 189, pi. 3, fig. 17; Miocene, Santo Domingo. This species has been found at Loc. 351, C. A. S., near Punta Pua, near the middle of the Tubera group, and at Loc. 306, C. A. S., at the eastern border of Usiacuri village at about the same horizon. It has been collected also at Loc. 299, C. A. S., two miles southwest of Baranoa; Loc. 325 and 325-A, C. A. S., all representing horizon P of the Tubera group, of the Colombian Miocene. 5. Terebra haitensis Dall Terebra (Hastula) haitensis Dall, Proc. U. S. Nat. Mus., vol. 18, 1895, p. 35. — Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 35, pi. 59, fig. 30; Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 207, pi. 1, fig. 3; Gatun Stage, Costa Rica. — Maury, Bull. Am. Pal., vol. 5, 1917, p. 194, pi. 4, fig. 4; Cercado de Mao, Miocene, Santo Domingo. This species has been obtained at Loc. 299, southwest of Baranoa, and at Loc. 351, C. A. S., near Punta Pua, both near the middle of the Tubera group of the Colombian Miocene. 6. Terebra bipartita Sowerby Terebra bipartita Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849, p. 47; Miocene, Santo Domingo. — Maury, Bull. Am. Pal., vol. 5, 1917, p. 187, pi. 3, fig. 14; Miocene, Santo Domingo. — Olsson, Bull. Am. Pal. vol. 9, 1922, p. 207, pi. 1, figs. 1,2; Miocene, Costa Rica. Terebra (Acus) bipartita Sowerby, Dall, Proc. U. S. Nat. Mus., vol. 18, 1895, p. 38; not T. (Oxymeris) bipartita (Sow.) Dall, 1903, pi. 59, figs. 13, 28, 29, loc. cit. 108 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. This species has been found at Loc. 351, C. A. S., near Punta Pua, near the middle of the Tubera group, 20 miles north of Cartagena. 7. Conus sewalli Maury Conus sewalli Maury, Bull. Am. Pal., vol. 5, 1917, p. 201, pi. 5, fig. 3; pi. 6, fig. 3; Cercado de Mao, Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 220; Gatun Stage, Canal Zone, Panama. Excellent examples of this shell were obtained at the Spill- way of the Canal in 1914; it has since been found at two lo- calities in northern Colombia, namely, Loc. 304, C. A. S., four miles east of Santa Rosa, and at Loc. 351, C. A. S., near Punta Pua. Both are at central horizons of the Tubera group. 8. Conus veatchi Olsson Conus veatchi Olsson, Bull. Am. Pal., vol. 9, 1922, p. 216, pi. 2, figs. 5, 8; Gatun Stage, Canal Zone, Panama. Only a single imperfect example of this species was found, and it was obtained at Loc. 267, C. A. S., horizon M - N, near the base of the Tubera group, at the west foot of Tubera mountain. 9. Conus imitator Brown & Pilsbry Conus imitator Brown & Pilsbry, Proc. Acad. Nat. Sci., Phila., vol. 63, 1911, p. 342, pi. 23, fig. 4; Gatim formation. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 217, pi. 2, fig. 6; Gatun Stage, Canal Zone, Miocene, Costa Rica. This species was found at various localities in the Colom- bian Miocene. In many respects it resembles C. chipolanus Dall, from a low horizon of the Gulf Coast. It occurs at Loc. 267, C. A. S., horizon M-N; Loc. 325-A, C. A. S., near Cibarco; Loc. 351, C. A. S., near Punta Pua; most of these are at central horizons of the Tubera group. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA J^QQ 10. Conus molis Brown & Pilsbry Conus molis Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 343, pi. 23, fig. 1; Miocene, Canal Zone. — Maury, Bull. Am. Pal., vol. 5, 1917, p. 200, Cercado de Mao, Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 214, pi. 2, figs. 1, 2; Miocene, Costa Rica. This species occurs quite abundantly in the Tubera group of the Colombian Miocene, but can not be regarded as a horizon marker. It has been obtained at Loc. 267, C. A, S., horizon M - N ; horizon P, and horizon R ; also at Loc. 299, near Baranoa; Loc. 304, C. A. S., near Santa Rosa; and Loc. 351, C. A. S., near Punta Pua. Its occurrence is therefore at nearly all horizons of the Tubera group. 11. Conus granozonatus Guppy Conus granozonatus Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 287, pi. 16, fig. 5; Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 222, pi. 3, fig. 15; Gatun Stage, Costa Rica. A single good specimen of this species was obtained at Loc. 351, C. A. S., near Punta Pua. It is slightly larger and more robust than appears in either Guppy's or Olsson's figures, although in other respects the identification is satisfactory. 12. Conus recognitus Guppy Conus solidus Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849, p. 45; Miocene, Santo Domingo; not C. solidus Sowerby, Conch. lUust., 1841, pi. 56, No. 76.— Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 287, pi. 16, fig. 1; Miocene, Jamaica. Comis recognitus Guppy, Proc. Sci. Assn. Trinidad, 1867, p.. 171. — Guppy, Geol. Mag., vol. 1, 1874, p. 409; new name proposed. — Guppy, Quart. Jour. Geol. Soc. Lond., vol. 32, 1876, p. 527.— Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1583.— Maury, Bull. Am. Pal., vol. 5, 1917, p. 209, pi. 7, fig. 9; Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 218, pi. 2, fig. 9; Miocene, Costa Rica.— Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 73, 1921, p. 327, pi. 19, fig. 2; Mio- cene, Santo Domingo. This species is one of the most abundant in the Tubera group of the Colombian Miocene. Like C. molis, it is not reg"arded as a horizon marker, since it is found at various levels. It has been obtained at Loc. 267, C. A. S., horizons 110 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Sek. P and R, and at Loc. 325-A, and 351, C A. S., the latter of which is central in the Tubera group. 13. Conus planiliratus Sowerby Conus planiliratus Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849, p, 44. — GUPPY, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 287, pi. 16, fig. 7; Miocene, Santo Domingo. — Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1583. A single specimen of C. planiliratus was obtained at Loc. 267, C. A. S., horizon M - N, near the base of the Tubera group. It has not been found at any other horizon, as far as known. 14. Conus stenostomus Sowerby Conus stenostomus Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849, p. 44; Miocene, Santo Domingo. — Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 287, pi. 16, fig. 2.— Maury, Bull. Am. Pal., vol. 5, 1917, p. 203; Cercado de Mao, Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 214, Gatun Stage, Canal Zone. Only a single good example of this species was obtained at Loc. 267, C. A. S., horizon M - N, near the base of the Tu- bera group, at the west foot of Tubera mountain. According to Olsson, it occurs in the Gatun Stage of Port Limon, and Maury lists it from the upper Miocene of Spring^^ale, Trini- dad Island. 15. Conus concavitectum Brown & Pilsbry Conus concavitectum Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 341, pi. 23, figs. 5,6; Gatun formation, Canal Zone. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 215; Gatun Stage, Canal Zone. Three specimens of this species were obtained at Loc. 267, C. A. S., horizon P, on the north slope of Tubera mountain, which horizon is believed to closely represent the Gatun ho- rizon of the Canal Zone, Panama. 16. Conus burckhardti ( ?) Bose Conus burckhardti Bose Bull. Inst. Geol. de Mex., No. 22, 1906, p. 50, pi. 5, figs. 39, 40.— Olsson, Bull. Am. Pal., vol. 9, 1922, p. 224, pi. 3, figs. 4, 5; Miocene, Gatun Stage, Panama. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA \\\ A single specimen of Conns that seems referable to the Mexican species was obtained at Loc. 351, C. A. S., near Punta Pua. In this example the spire is distinctly different from that of Bose's species in having the upper surface of the whorls rounded, or somewhat angulated along a median line, thus forming a succession of sloping steps, rather than a smooth, regular slope. In most respects, however, the shell closely resembles the Mexican form. A number of well pre- served examples of C. hnrckhardti was obtained at the Spill- way of the Canal in 1914, though none of them show the form of spire noted in the present example. 17. Conus consobrinus (?) Sowerby Conus consobrinus Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849, p. 45. — GUPPY, Geol. Mag., vol. 1, 1874, p. 409, pi. 17, fig. 3.— Maury, Bull. Am. Pal., vol. 5, 1917, p. 203, pi. 6, figs. 5, 6; Miocene, Santo Domingo. — Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 73, 1921, p. 330, pi. 20, figs. 7, 7a, 7b; Miocene, Santo Domingo. Two examples of a Conus found at Loc. 267, C. A. S., horizon P, on the north slope of Tubera mountain, closely re- semble Sowerby's species, although there are some differences of sculpture in the last whorl. In our examples the minute beads on the spiral ribs are rounded instead of being elongated, as in Sowerby's form. The lines of growth are arcuate, and in other respects the characters are nearly identical with those of Sowerby's species. 18. Conus tortuosopunctatus Toula Conus (Cheliconus) tortuosopunctatus Toula, Jahrb. der K. K. Geol. Reichs., Wien, Bd. 61, 1911, p. 507, pi. 31, fig. 21, b.— Olsson, Bull. Am. Pal,, vol. 9, 1922, p. 226, pi. 3, figs. 6, 11; Gatun Stage, Canal Zone. It would appear from the figures given by Toula and Olsson that the height of spire in this species is variable, as is so often the case. In our examples the spire is intermediate in height between the extremes found in these figures. In other respects the identification is completely satisfactory. These samples come from Loc. 351, C. A. S., near the middle of the Tubera group, probably near the horizon of P, at Tubera mountain. 2J2 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 19. Conus tuberacola Anderson, new species Plate 9, figures 4, 5 Shell of medium size, probable height of holotype (incom- plete) 54 mm., width 3.4 mm., spire high, concavely turrited, earlier whorls coronated ; last two or three whorls smooth, but slightly excavated above; sides of older specimens smooth, in younger shells the sides are adorned with minute spirally ar- ranged beads, chiefly on the lower half of the shell ; aperture narrow. The shoulders of the last whorl sharp and abrupt; lines of growth strongly curved. This shell resembles C. consohrinus Sow., only in sculpture, but is relatively wider, has less strongly developed granula- tions on the sides. It also differs from C. toroensis Olsson in relative width and in form of spire. Holotype: No. 4623, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon M - N, near base of the Tubera group, where it appears to belong, and where several fair-sized specimens were obtained ; Miocene of Colombia. 20. Conus crenospiratus Anderson, new species Plate 9, figures 6, 7 Shell small, height of holotype 17 mm., width 10 mm., with graceful outline, low spire and somewhat rounded sides; in size, form and sculpture it recalls C. isomitratus Dall, from the Chipola beds of Florida ; upper surface of the whorls flat- tened ; sutures distinctly incised, but unlike Ball's species, the shoulders of the whorls are tuberculated, forming on the inner side of the suture a wavy, or crenulated line; body whorl or- namented by spiral lines, which become obsolete near the shoulder, but become stronger on the lower third of the whorl ; spiral threads are here flattened, or slightly concave in section, having the appearance of being double. Holotype: No. 4624, Mus. Calif. Acad. Sci., from Loc. 351, C. A. S., near Punta Pua, near the middle of the Tubera group, Colombia ; Miocene. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA \\^ 21. Turris albida (Perry) Pleurotoma albida Perry, Conch. Expl., 1811, pi. 32, fig. 4.— Dall, Bull. Mus. Comp. Zool., Harvard College, vol. 18, 1889, pp. 72-73.— Trans. Wag. Fr. Inst. Sci., vol. 3, 1890, p. 28, pi. 4, fig. 8a.— Brown & Pils- BRY, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 343; Miocene, Canal Zone. Turris albida, Dall, Bull. U. S. Nat. Mus., No. 90, 1915, p. 38, pi. 5, fig. 13; pi. 14, fig. 7; Orthaulax pugnax zone. Lower Miocene, Miss., and Santo Domingo.— Maury, Bull. Am. Pal., vol. 5, 1917.— Olsson, Bull. Am. Pal., vol. 9, 1922, p. 230, pi. 4, figs. 1, 2; Gatun Stage, Canal Zone, Panama. This species has been obtained at Loc. 267, C. A. S., horizon M - N, at Loc. 351, C. A. S., near the middle of the Tubera group, and at Loc. 266, C. A. S., on the Ouebrada Juan de Acosta, near the top of the Tubera group. Its range is, therefore, throughout the entire group, and it can not, ac- cordingly, be regarded as a horizon marker. 22. Drillia eupora Dall Drillia eupora Dall, Bull. U. S. Nat. Mus., No. 90, 1915, p. 42, pi. 5, fig. 3, Tampa Bay, Florida. Among the fossils collected from the Las Perdices shale one mile west of the pier at Puerto Colombia, Loc. 267, C. A. S., horizon L, is an incomplete example of Drillia which includes most of the spire. One whorl is missing, though the axis it- self is complete. When entire, the shell was composed of at least 13 whorls, including the nuclear portion, forming an elongated, narrow, gently tapering spire. The penultimate whorl has 18 vertical ribs of the form described by Dr. Dall, crossed by revolving threads, five in number, and a subsutural collar bordered by a carinated ridge. The resemblance of this shell to Dall's figure, reinforced by his description, permits no other identification. This species does not appear to have been recognized before in the Miocene of the Caribbean bor- ders, though doubtless subsequent work will reveal its presence in other parts of the region. 114 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 23. Cancellaria karsteni Anderson, new species Plate 10, figures 7, 8, 9 Shell of moderate size, biconic in outline, heavy ribbed on the last whorls, spinose on the shoulders; height of holotype 33 mm., width of body whorl 22 mm,; spire high and sharp, forming somewhat more than half height of shell; surface marked by heavy vertical ridges, of which there are about nine on last whorl ; these crossed by low revolving threads, with occasional intermediary lines; shoulders of whorl armed with strong spines, rising from the vertical ribs, pointing upward and outward ; upper surface of body whorl concave, rising on preceding whorl in a sort of clasping collar with wavy border; aperture . somewhat quadrilateral ; outer lip angulated near shoulder, and also midway between shoulder and terminus of canal ; inner lip thinly calloused, bearing three oblique plica- tions ; umbilicus closed. Holotype: No. 4630; paratype: No, 4631, Mus. Calif. Acad. Sci., from Loc. 267 (C. A. S. Coll.), horizon P, at the north end of Tubera mountain, Colombia; Miocene, It is also found at Loc. 305, C. A. S., near Turbaco, near the middle of the Tubera group, Colombia ; Miocene. 24. Cancellaria hettneri Anderson, new species Plate 10, figures 5, 6 Shell large, height of holotype 42 mm., width 28 mm., somewhat biconic in outline, heavily ribbed on the body whorl with irregular ridges extending to the base; spire high, sub- conic; upper slope of whorl rising in a collar, not quite clasp- ing, but slightly channelled or flattened above; surface orna- mented by revolving threads of three orders, heavy, inter- mediate and light; shoulders of whorls showing low spines directed outwardly; aperture subquadrate, narrowed above, terminating below in a straight canal ; umbilical chink distinct, but closed. This species is allied to C. harrtsi Maury, but is more coarsely sculptured, larger, and more spinose. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA H^^ Holotype: No. 4629, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., from horizon P, north slope of Tubera mountain, Colombia; Miocene. Two good specimens were obtained at this locality. 25. Cancellaria dariena Toula Cancellaria dariena Toula, Jahrb. der K. K. Geol. Reichs., Bd. 58, 1909, p. 704, pi. 28, figs. 1,2; Gatun formation, Canal Zone.— Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 345, pi. 24, figs. 1-4.— Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 69, 1917, p. 32; Gatun formation, Canal Zone.— Olsson, Bull. Am. Pal., vol. 9, 1922, p. 252, pi. 6, fig. 8; Gatun Stage, Costa Rica. This Species has not been found abundantly in the Colom- bian Miocene, but it has been obtained at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena, near the middle of the Tubera group. 26. Cancellaria scheibei Anderson, new species Plate 10, figures 1, 2, 3, 4 Shell large, robust, ovate in outline, smooth, with low spire ; height of holotype 54 mm., greatest width 40 mm. ; spire low, conical, sloping up from rounded shoulders; suture distinctly channelled; whorls about five, the younger three obscurely cancellated; aperture subovate, narrowed above, terminating below in a narrow, slightly curved canal ; outer lip sharp, lirate within near the outer edge ; inner lip strongly calloused, bear- ing three plications, the upper two being more widely sepa- rated, with three elongated beads on the pillar intervening. Holotype: No. 4627, Mus. Calif. Acad. Sci., from Loc. 306, C. A. S., from near Usiacuri, Colombia; paratype: No. 4628, Mus. Calif. Acad. Sci., from Loc. 304, C. A. S., near Santa Rosa, Colombia ; Miocene. This form remotely resembles C. IxEvescens Guppy, but is larger, smoother, more rounded, and has plications that are distinctly different from Guppy's species. It is more nearly related to C. solida Sowerby,^® found living on the Pacific " Sowerby, J. de C, Proc. Zool. Soc. Lond., vol. 2, 1832, p. SO. — Sowerby, Thes. Conchy., vol. 2, p. 440, pi. 92, fig. 4. 115 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. coast from Panama to the Gulf of California. The essential difference may be one of descent, and of senility in the living form. The earlier form is larger, more robust, has a more rugose columella, with bead-like denticles intervening between the plaits, as already described. This shell is apparently not abundant, but it has been ob- tained at Loc. 304, C. A. S., near Santa Rosa; Loc. 306, C. A. S., near Usiacuri; Loc. 325-A, C. A. S., near Cibarco. It has not yet been found at the lowest horizon of the Tubera group, though a near ally does occur there. 27. Cancellaria codazzii Anderson, new species Plate 14, figures 4, 5, 6, 7 Shell of medium size, height of holotype 30 mm., width 18 mm., biconic in outline, with numerous vertical ribs extending from suture to base; spire high, with five whorls below nuclear ones; nuclear whorls three, quite smooth; surface beautifully cancellated, with revolving threads at nearly equal intervals crossing the numerous vertical ribs in low, rounded bead-like nodes; upper surface of whorls slightly concave, rising in a collar-like expansion, not clasping; concave surface bearing a few revolving threads; suture distinctly channelled; shoulder of whorl not coronate, but bearing a wavy cord; aperture ovate, terminating in a narrow canal ; outer lip simple, lirate within ; inner lip not distinctly calloused, bearing three oblique plications on the pillar. Holotype: No. 4645; paratype: No. 4646, Mus. Cahf. Acad. Sci., from Loc. 325-A, C. A. S., near Cibarco, Colombia; Mi- ocene, near the middle of the Tubera group. This shell is named in honor of Agostino Codazzi, explorer, surveyor, writer, and author of the first authentic map of Colombia. 28. Cancellaria cibarcola Anderson, new species Plate 14, figures 1, 2, 3 Shell of medium size, resembling in most respects C. scheihei, but smaller and less rotund. Its three nuclear whorls are quite smooth; its disposition toward a truly cancellated Voi» XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA \\y sculpture in the young stages is more pronounced than in the preceding, and the spiral threads often show clearly on the fifth whorl below the nuclear ones. In outer form it recalls C. IcEvescens Guppy, from which, however, it is readily dis- tinguished by the arrangement of the plaits on the pillar. Two elongated denticles intervene between the upper and second plait which are widely separated. The internal lirations of the outer lip extend deeply into the interior, and the spiral threads become more distinct at the base of the body whorl. These features serve to distinguish this species from either of the preceding. Height of holotype 32 mm., width of body 22 mm., height of aperture 25 mm. Holotype: No. 4643; paratype: No. 4644, Mus. Calif. Acad. Sci., from Loc. 325-A, C. A. S., near Cibarco, Colombia; Miocene. This shell is found at all of the lower horizons of the Tu- bera group, and is a fairly abundant form. It has been ob- tained at Loc. 267, C. A. S., horizons M - N, P and R; Loc. 299, 304, 325-A, and 351, C. A. S., the latter representing a central horizon in the Tubera group. 29. Cancellaria cossmanni Olsson Cancellaria cossmanni Olsson, Bull. Am. Pal., vol. 9, 1922, p. 253, pi. 6, figs. 9, 1 1 ; Gatun Stage, Costa Rica. This species has not been found abundant in Colombia. A single specimen was obtained at Loc. 325-A, near Cibarco, about the middle of the Tubera group. Its range is not known. 30. Cancellaria moorei ( ?) Guppy Cancellaria moorei Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 289, pi. 17, fig. 7; Miocene, Jamaica. A single specimen that seems referable to Guppy's species was obtained at Loc, 306, C. A. S., at the east border of Usi- acuri village above the middle of the Tubera group. In spite of the fact that this species has not often been recognized in the faunas of the Caribbean region outside of the Bowden 118 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. beds, the resemblance of the sample from Loc. 306 to Guppy's original figure does not permit of any other determination at present. 31. Cancellaria guppyi Gabb Cancellaria guppyi Gabb, Trans. Am. Phil. Soc, vol. 15, 1873, p. 236; Mio- cene, Santo Domingo. — Maury, Bull. Am. Pal., vol. 5, 1917, p. 228, pi. 10, figs. 7, 8; Cercado de Mao, S. Domingo. — Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 73, 1921, p. 333, pi. 22, fig. 7; Loc. as above. In his Revision of Gabb's Tertiary Mollusca Dr. Pilsbry figures the type (or a lectotype) of this species. The rotund form and regularly cancellated sculpture are its striking characteristics. A single specimen was found at Loc. 267, C. A. S., horizon P, near the middle of the Tubera group. 32. Turritella altilira Conrad Plate 17, figures 4, 5 Turritella altilira Conrad, Pac. R. R. Repts., vol. 6, 1857, pt. 2, p. 72, pi. 5, fig. 19; Miocene, Isthmus of Panama. — Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 358, pi. 27, figs. 2, 3; Gatun formation. Canal Zone. — Olsson, Bull. Am. Pal., vol. 9, 1922, pp. 321, 322, pi. 14, figs. 4, 8, 9, 14; Miocene, Canal Zone. — Hodson, Bull. Am. Pal., vol. 11, 1926, p. 214, pi. 26, figs. 1, 4, 7, etc.; pi. 27, figs. 2-7; Miocene, North Venezuela. This shell is abundant in the Colombian Miocene. It is in- teresting to note that both Maury^° and Olsson"^ regard T. tornata Guppy, as a varietal form of this species, and that Cossmann admits-" that his T. guppyi is the equivalent of T. tornata, all of which beliefs seem to be well founded. Toula, furthermore,^^ regards his T. gabbi as being nearly related to T". altilira and T. tornata. The species occurs plentifully at Loc. 267, C. A. S., horizon M - N, of the Tubera group, and in higher horizons of the same. It has been found also at Loc. 351, C. A. S., near Punta Pua, and at Loc. 305, C. A. S., near Turbaco, and at =" Maury, C. J., Bull. Am. Pal., vol. 10, 1925, pp. 382-383. "Olsson, A. A., Bull. Am. Pal., vol. 9, 1922, p. 323. "Cossmann, M., Rev. Grit, de Pal., 1909, p. 225. » Toula, F., Jahrb. der K. K. Geol. Reichs., Bd. 58, 1909, p. 69S. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA \\g Loc. 354, Quebrada de Murindo, a tributary of the Rio Canalete. Plesiotype: No, 4658, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon M - N, base of Tubera mountain, Colombia; Miocene. 33. Turritella perattenuata Heilprin Turritella perattenuata Heilprin, Trans. Wag. Fr. Inst. Sci., vol. 1, 1887, p. 88, pi. 8, fig. 13; Pliocene, Caloosahatchie beds, Florida. — Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1900, p. 316, pi. 16, figs. 5, 9; Loc. as above; — var. prcBcellens Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 69, 1917, p. 36, pi. 5, fig. 12; Miocene, near Cartagena. This species has not often been Hsted among the forms found in the Miocene of the Caribbean region but it never- theless occurs at a number of Miocene horizons in Colombia. It has been found abundantly at Loc. 267, C. A. S., horizon M - N, of the Tubera group, at Loc. 347, C. A. S., near Tur- baco, and in the uppermost beds of the Miocene near Galapa ; it occurs also in the position of horizon P, at Loc. 306, C. A. S., near Usiacuri. It is therefore found at most of the fossil horizons of the Tubera group. 34. Turritella fredeai Hodson Plate 17, figure 1 Turritella rohusta Grzyb. var. fredeai Hodson, Bull. Am. Pal., vol. 11, 1926, p. 13, pi. 5, fig. 3; pi. 6, fig. 5; pi. 7, figs. 1, 6, 7; Miocene, Northern Venezuela. Not T. robusta Gabb, Geol. Surv. Calif., Pal. vol. 1, 1864, p. 135, pi. 21, fig. 94; Cretaceous of California. Not T. {Haustator) robusta Grzyb., upper Zorritos, Peru. Turritella abrupta Spkr., Anderson, Proc. Calif. Acad. Sci., vol. 16, 1927, p. 89; horizon M, Tuberd, group, Colombia. Not T. robusta, var. abrupta Spieker, Johns Hopkins Univ. publ., Geol. No. 3, 1922, p. 85, pi. 4, fig. 6; Zorritos formation, northern Peru. Plesiotype: No. 4175, Mus. Calif. Acad. Sci., from Loc. 351, C. A. S., near Punta Pua, 20 miles north of harbor of Carta- gena, Colombia; Miocene. Spieker' s form from northern Peru has been renamed by Hanna & Israelsky^* as T. supraconcava, as explained below. "Hanna, G. D. & Israelsky, M., Proc. Calif. Acad. Sci., vol. 14, 1925, p. 59. 120 CALIFORNIA ACADEMY OF SCIENCES [Pboc. 4th Ser. When the writer listed the Colombian form as probably identi- cal with the Peruvian of Spieker, Hodson's recent paper had not yet reached us. A comparison of the Colombian fonns with Hodson's figures clearly shows their identity, while his illustrations serve as well to distinguish the northern form from the Peruvian. The specific name "rohusta" had already been employed at the date of Spieker's writing, and Hodson's form must take the name of his supposed variety. This species has been found at Loc. 267, C. A. S., horizon M - N, the lowest horizon of the Tubera group, and at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena, also in a low horizon. It occurs, however, in higher beds, as at Galapa, near the top of the Miocene section. 35. Turritella mimetes Brown & Pilsbry Turritella mimetes Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 357, pi. 27, fig. 1; Miocene, Canal Zone. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 321, pi. 14, fig. 5; Gatun Stage, Canal Zone. This species is not uncommon in the Gatun group of the Canal Zone where it was obtained by the writer in 1914. It occurs also at Loc. 325, C. A. S., near Baranoa, near the mid- dle of the Tubera group. 36. Turritella gatunensis Conrad Turritella gatunensis Conrad, Pac. R. R. Repts., vol. 6, 1857, pt. 2, p. 72, pi. 5, fig. 20; Miocene, Isthmus of Panama. — Toula, Jahrb. der K. K. Geol. Reichs., Bd. 58, 1909, p. 694; Miocene, Canal Zone. — Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 358, pi. 27, figs. 4, 5, 9; occurrence as before. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 320, pi. 14, figs. 12, 13; Miocene, Costa Rica, etc. This shell was obtained at the Spillway in considerable numbers by the writer in 1914, and has since been found plen- tifully in the Tubera group of Colombia. It occurs at Loc. 267, C. A. S., horizons P and R. It has been obtained also at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena, near the middle of the Tubera group, and at Loc. 305, near Turbaco, about central in the group, and at Loc. 306, near Usiacuri, in a position near the middle of the section. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA ^21 37. Turritella cartagenensis Brown & Pilsbry Turritella cartagenensis Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 69, 1917, p. 34, pi. 5, fig. 13; Miocene, near Cartagena. — Maury, Bull. Am. Pal., vol. 10. 1925, p. 385, pi. 42, fig. 13; Miocene, Trinidad Island. ? Turritella bifastigata, HoDSON, Bull. Am. Pal., vol. 11, 1926, pp. 48-50, pi. 30, figs. 1-6; Miocene, northern Venezuela. This is one of the most abundant forms of Turritella found in the Tubera group. Hodson has described and figured varieties of a Turritella under the name T. bifastigata Nelson, from the Miocene of northern Venezuela. The type of Nel- son's species came from the Tertiary (probably Miocene) of Peru, but was described without any illustration whatever. Hodson's figure (pi. 30, fig. 1) is from a lectotype not sup- plied by Nelson. It should be pointed out, however, that the varieties, supposedly of this Turritella, as figured by Hodson are such as would include T. cartagenensis Brown & Pilsbry, which itself shows variations of the same character. This species has been obtained from Loc. 306, C. A. S., near the village of Usiacuri; Loc. 351, C. A. S., near Punta Pua; Loc. 353, C. A. S.. near the Bay of Cartagena; and Loc. 325-A, near Cibarco. Its vertical range is nearly central in the Tubera group. 38. Crucibulum (Dispotaea) gatunense (Toula) Plate 13, figures 4, 5, 6 Captilus ? gatunensis Toula, Jahrb. der K. K. Geol. Reichs., Bd. 58, 1909, p. 692, pi. 25, figs. 1, 2; Gatun formation, Canal Zone. — Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 360; Gatun formation, Canal Zone. Two good examples of this hitherto imperfectly known species were obtained from the Spillway of the Canal in 1914, and are now in the collections of the California Academy of Sciences. The coiled apex is smooth, showing only faint lines of growth, but two mm. below the apex the shell becomes corrugated, at first by irregular squamose vertical threads, radiating downward, interrupted by uneven lines of growth. These radial markings become more irregular with growth, March 29, 1929 122 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. forming a roughened, granular, radially marked surface. The outline of the base is sub-elliptical, with sharp, faintly crenu- lated margin. Toula's samples did not permit him to see the interior of the shell, but in ours the interior is clearly exposed in both examples. The shell possesses a well formed internal cup, semilunar in outline, attached to the wall of the shell on about one-third of its periphery, or in fact, is formed by the wall of the shell itself. This feature places it in the sub-genus Dispotcea (Say) Conrad, as has been stated by Dall. Plesiotypes: Nos. 4639, 4640, Mus. Calif. Acad. Sci., from Loc, 323, C. A. S., Gatun locks at Spillway, Panama ; Miocene. A somewhat fragmentary example of this species was found at Loc. 351, C. A. S., near Punta Pua, near the middle of the Tubera group. 39. Architectonica granulata (Lamarck) Solarium granulatum Lam., An. s. Vert., vol. 7, 1822, p. 3. — Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1892, p. 329.— Maury, Bull. Am. Pal., vol. 5, 1917, p. 295, pi. 23, fig. 3; Miocene, Santo Domingo. Cadran (=Solarium) granulatum, Kiener, Icon., vol. 1, 1873, p. 4, pi. 2, fig. 2. Solarium gatunense Toula, Jahrb. der K. K. Geol. Reichs., Bd. 58, 1909, p. 693, pi. 25, fig. 3; Miocene, Canal Zone. Architectonica granulata, Dall, Proc. U. S. Nat. Mus., vol. 37, p. 232; living, Lx)wer California to Panama. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 326, pi. 13, figs. 10-12; Miocene, Canal Zone, etc. — Maury, Bull. Am. Pal,, vol. 10, 1925, p. 388, pi. 40, fig. 1; Miocene, Trinidad Island. This species occurs abundantly in the Tubera group of the Colombian Miocene. Good examples have been obtained at the following places in northern Colombia : Loc. 266, C. A. S., Juan de Acosta creek, near Puerto Colombia; Loc. 267, C. A. S., horizon P, north slope of Tubera mountain; Loc. 299, C. A. S., near Baranoa, near the middle of the Tubera group; Loc. 305, C. A. S., near Tur- baco; and finally, Loc. 351, C. A. S., near Punta Pua, near the middle of the Tubera group. It thus appears that the vertical range of this species is confined to the middle part of the Tubera group. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA ^23 40. Architectonica quadriseriata (Sowerby) Solarium quadrisertatum Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1850, p. 51, pi. 10, figs. 8a, b, c. — Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 291.— Guppy, Geol. Mag., vol. 1, 1874, p. 438; Miocene, Santo Domingo. — Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1585; Miocene, Florida. Architectonica quadriseriata, Gabb, Trans. Am. Phil. Soc, vol. 15, 1873, p. 228.— Maury, Bull. Am. Pal., vol. 10, 1925, p. 389; Miocene, Trini- dad Island. Good examples of this species were obtained at the Spillway of the Canal in 1914, and it has since been found at Loc. 305, C. A. S., near Turbaco, and at Loc. 325-A, C. A. S., near Cibarco, in both places near the middle of the Tubera group, or at horizon P. 41. Natica guppyana Toula Natica {Stigtnaulax) guppyana Toula, Jahrb. der K. K. Geol. Reichs., Bd. 58, 1909, p. 696, pi. 25, fig.6; Miocene, Canal Zone. — Brown& Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 360.— Olsson, Bull. Am. Pal., vol. 9, 1922, p. 328, pi. 13, figs. 13-15; Gatun beds. Canal Zone, Panama. A large number of samples of this species was obtained at the Spillway of the Canal in 1914, and are now in the collec- tions of California Academy of Sciences. Equally good speci- mens have since been obtained from various localities in north- ern Colombia, as the following: Loc. 267, C. A. S., horizon M - N, west foot of Tubera mountain; Loc. 267, C. A. S., horizon P, north slope of Tubera mountain; Loc. 325, C. A. S., near Usiacuri village; Loc. 325-A, near Cibarco; and Loc. 351, C. A. S., near Punta Pua. In most of these localities the samples came from a horizon near the middle of the Tubera group. 42. Natica cuspidata Guppy Natica cuspidata Guppy, Agr. Soc. Trin. and Tobago, Ppr. No. 454, 1910, p. 5- pi. 2, fig. 4; Reprint, Bull. Am. Pal., vol. 8, 1921, p. 162, pi. 8, fig. 4; Miocene, Trinidad I.— Maury, Bull. Am. Pal., vol. 10, 1925, p. 391, pi. 40, figs. 9, 10; Loc. as before. 124 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Two examples of this seemingly rare shell were found in the lowest horizon, M - N, of the Tubera group, at Loc. 267, C. A. S., associated with many heavy shelled littoral species. 43. Polinices subclausa Sowerby Natica subclausa Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849, p. 51; Miocene, Santo Domingo. — Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 290, pi. 18, fig. 8.— Guppy, Geol. Mag., vol. 1, 1874, p. 437.— Guppy, Quart. Jour. Geol. Soc. Lond., vol. 32, 1876, p. 519; Loc. as before. Polinices subclausa, Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1585. — Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 360.— Maury, Bull. Am. Pal., vol. 5, 1917, p. 300, pi. 23, fig. 14; Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 329, pi. 13, figs. 16, 17; Miocene, Canal Zone, Panama. In 1914 the writer obtained a few samples of this species at the Spillway of the Canal. Since then others have been obtained at Loc. 266, C. A, S., Arroyo Juan de Acosta, near the top of the Tubera group, and at Loc. 267, C. A. S., hori- zon P, north slope of Tubera mountain. 44. Polinices stanislas-meunieri Maury Polinices stanislas-meunieri Maury, Bull. Am. Pal., vol. 5, 1917, p. 300, pi. 23, figs. 15, 16; Miocene, S. Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 329, pi. 13, fig. 7; Gatun Stage, Canal Zone. A large number of samples of this species has been found in the Tubera group. It occurs in the following localities : Loc. 267, C. A. S., horizon P, north slope of Tubera moun- tain; Loc. 325, C. A. S., east border of Usiacuri village; Loc. 325-A, C. A. S., near Cibarco, and Loc. 351, C. A. S., near Punta Pua. 45. Polinices prolactea Anderson, new species Plate 14, figures 8, 9 Shell of moderate size, subglobose, with low spire, open umbilicus, conspicuous callus, highly polished; aperture sub- lunar, narrowing behind to a subacute angle; callous rather heavy on the posterior part of the inner lip; narrowing to a Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 125 thin line near the anterior part of the aperture ; surface marked by Hnes of growth, and near the base of the shell by faint spiral striations, not always visible. Several good examples of this species were obtained at Loc. 267, C. A. S., in the Las Perdices beds below the Tubera group, a mile west of Puerto Colombia. Holotype: No. 4648, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon L, Los Perdices group, Puerto Colombia; Miocene. The nearest ally of this shell is Polinices lactea (Guilding), now living in the neighboring Bay of Cartagena. G. B. Sowerby has described a similar near relative from the coast of Chile as Natica solida.^^ 46. Ampullaria tuberacola Anderson, new species Plate 16, figures 1, 2, 3 Shell subovate, at least when full grown, deeply perforate, spire low in mature shells, whorls five or six, shell thin and with a deficiency of calcareous matter ; umbilicus open, funnel- form, angulated on the borders in adult shells ; suture distinct and slightly impressed ; upper surface of the whorls rounded and convex ; shoulder of last whorl sharply rounded, sides sloping toward the narrow base, making the form of the shell somewhat conical; height of holotype 52 mm., greatest wndth 48 mm. Holotype: No. 4655; paratype: No. 4656, from Loc. 267, C. A. S., from horizon R, Tubera village, Colombia; Miocene. The younger shell is more nearly sub-globose and bears a strong resemblance to A. (Pomus) canaliculata Lam., from Tropical America. Two examples of this shell w-ere found at Tubera village, Loc. 267, C. A. S., associated w^ith many strictly marine forms, such as Conns recognitus, Malea ringens, Ficiis colombiana, etc. It is quite probable that these non-marine shells were brought into this association by streams from a neighboring shore to the southwest. »Gcol. Observ. Darwin, Append, pt. 2, p. 612, pi. 3, figs. 40, 41. 126 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 47. Calliostoma grabaui Maury Calliostoma grabaui Maury, Bull. Am. Pal., vol. 5, 1917, p. 319, pi. 24, fig. 19; Zone G, Rio Gurabo, S. Domingo. This species has been obtained at Loc. 267, C. A. S., hori- zons R and M - N ; Loc. 306, near Usiacuri village, and Loc. 325, C. A. S., lower in the section; Loc. 351, C. A. S., near Punta Pua. Its vertical range is, therefore, almost through- out the Tubera group. 48. Calliostoma olssoni Maury Calliostoma (Eutrochus) olssoni Maury, Bull. Am. Pal. vol. 10, 1925, p. 399, pi. 43, figs. 6, 14; Upper Miocene, Trinidad Island. This elegant little shell has been obtained from various localities in the Colombian Miocene, as at Loc. 266, C. A. S., Arroyo Juan de Acosta; Loc. 299, C. A. S., near Baranoa; Loc. 306, C. A. S., Usiacuri village, etc. Its vertical range is confined to the upper part of the Tubera group. 49. Calliostoma tropica Anderson, new species Plate 16, figures 6, 7 Shell small, conical; height of holotype 17 mm., width 15.5 mm., finely beaded, abruptly truncate below; spire sharply conical, sloping evenly to the basal border with which it forms an angle of about 80 degrees ; whorls 7 to 8 in number, sculp- tured with 6 to 8 finely beaded threads, crossed by lines of growth ; sutures marked only by a slight depression at the bor- der of the preceding whorl; base flattened, marked by 8 to 10 flat revolving threads, also beaded, but wider than the spiral threads on the upper slope; aperture ovate in outline; umbili- cus closed. The species is characterized by its high conical spire and regular even slope, and also by its abruptly flattened base and finely beaded ornamentation. It is closely related to Calliostoma derbyi Maury from the Lower Miocene of Brazil. Holotype: No. 4168, Mus. Calif. Acad. Sci., from Loc. 267, horizon M - N, Tubera mountain, Colombia; Miocene. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 127 50. Oliva cylindrica Sowerby Oliva cylindrica Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849, p. 45; Miocene, Santo Domingo. — Maury, Bull. Am. Pal., vol. 5, 1917, p. 67, pi. 10, figs. 14, 14a; Zone G, Rio Gurabo, Santo Domingo. — PiLSBRY, Proc. Acad. Nat. Sci. Phila., vol. 73, 1921, p. 335, pi. 23, figs. 2, 3; Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 88, pi. 7, fig. 1; Gatun Stage. Oliva gatunensis Toula, Jahrb. der K. K. Geol. Reichs., 1909, Bd. 58, p. 702, pi. 25, fig. 12; Gatun formation, Canal Zone, Panama. Cossman seems to have given the first adequate description of this species in 1913 but it is not at present available. Maury has given two good figures, upon which much reliance is placed. Good examples were obtained at Loc. 299, C. A. S., three miles southwest of Baranoa, and at Loc. 267, C. A. S., hori- zon M - N, near the base of the Tubera group. 51. Oliva sayana Ravenel Oliva sayana Ravenel, Cat., 1834, p. 19. — Mazyck, Nautilus, vol, 28, 1915, p. 139. Oliva sayana var. immortua Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 69, 1917, p. 33, pi. 5, fig. 6; Miocene, near Cartagena, Colombia. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 261, pi. 7, figs. 6, 7; Gatun Stage, Costa Rica. This species and variety were obtained at Loc. 267, C. A. S., horizon R, at Tubera village, near the top of the Tubera group of the Colombian Miocene. 52. Oliva brevispira Gabb Oliva brevispira Gabb, Trans. Am. Phil. Soc, vol. 15, 1873, p. 215; Miocene. Santo Domingo. — Maury, Bull. Am. Pal., vol. 5, 1917, p. 232, pi. 10, figs. 16, 17; Loc. as above. — Pilsbry, Proc. Acad. Nat. Sci., Phila., vol. 73, 1921, p. 335, pi. 23, fig. 4 (Type); Miocene, Santo Domingo. This species has been obtained at Loc. 267, C. A. S., hori- zon M - N, of the Tubera group, and at Loc. 325-A, near Cibarco, about the middle of the same group. Its range is at least from the basal beds to the middle of the Tubera group of the Colombian Miocene. 128 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 53. Oliva tuberaetisis Anderson, new species Plate 17, figures 2, 3 Shell large, thick, robust in form, spire high and accumi- nate; height of hollotype 87 mm., width 37 mm., height of aperture 65 mm., thickness of shell at outer lip 5 mm. ; suture clean and incised; aperture expanding gradually toward the anterior end, narrowed at the top into a cleft ; outer lip smooth, simple and gently arcuate; inner border of aperture slightly calloused, a little depressed near the middle, and bearing oblique ridges below. Holotype: No. 4172; paratype: No. 4174, Mus. Calif. Acad. Sci., from Loc. 267-R of Tubera group; paratype: No. 4173, Mus. Calif. Acad. Sci., from Loc. 267-C, C. A. S., Tubera group, Colombia; Miocene. This species resembles most nearly O. couvana Maury from the Springvale group of the Miocene of Trinidad, but it has a larger, thicker and relatively heavier shell, and more ovate out- line. The external calluses are wider, and the plications are more pronounced, as judged by Maury's figures. Our species differs from O. proavia Pilsbry & Johnson in a somewhat simi- lar manner, not forgetting Maury's comparison. This shell is fairly abundant in horizon R of the Tubera group, and it was obtained also at horizon M - N, at Loc. 267, C. A. S.. and very probably it will be found at intervening horizons. 54. Marginalia ballista Dall Marginella ballista Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, p. 47, pi. 4, fig. 6; Miocene, Tampa Silex beds, Florida. This Floridan species has not before been cited from the Miocene of the Caribbean region, although beds equivalent in age and ecologic conditions probably exist at many points therein. The form and surface features of our shell are too nearly like those figured and described by Dall to warrant any other determination of it. Dall has also described a varietal form of the same which he compares to M. incrassata Nelson, with which our species was Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA ^29 for a time tentatively identified. Its identity with the Floridan form seems to be supported by the possession of four obHque plications, as well as by the thickened outer lip and low spire. A single example of this species was found at Loc. 267, C. A. S., horizon P, on the north slope of Tubera mountain, in the Tubera group. 55. Marginalia christinelladas Maury Marginella christinelladae. Maury, Bull. Am. Pal., vol. 5, 1917, p. 234, pi. 11, fig. 6; Zone B, Miocene, Rio Gurabo, Santo Domingo. More than a dozen good examples of this species were obtained at Loc. 267, C. A. S., horizon P, on the north slope of Tubera mountain, and it was also found at horizon R, at Tubera village, and therefore at the middle and near the top of the Tubera group. 56. Marginella coniformis Sowerby Marginella coniformis Sowerby, Quart. Jour. Geol. Soc, Lond., vol. 6, 1928, 1849, p. 42; Miocene, Santo Domingo. — Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 288, pi. 17, fig. 2; Miocene, Trinidad Island. — Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 348, pi. 24, fig. 12; Gatun formation, Canal Zone. — M.^ury, Bull. Am. Pal., vol. 5, 1917, p. 234, pi. 11, figs. 5, 5a; Miocene, Santo Domingo. This species was obtained at the Spillway of the Canal in 1914, and since then at Loc. 267, C. A. S., horizon P of the Tubera group, on the north slope of Tubera mountain at a horizon believed to be the equivalent of the Gatun formation. 57. Mitra dariensis Brown & Pilsbry Miira dariensis Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 346, pi. 24, fig. 9; Gatun formation, Canal Zone. — Olsson, Bull. Am. Pal. vol. 9, 1922, p. 273, pi. 6, fig. 25; Gatun Stage, Canal Zone. Several good specimens of this species were obtained at the Spillway of the Canal in 1914, and since then it has been found at Loc. 351, C. A. S., near Punta Pua, near the middle of the Tubera group. 130 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 58. Mitra longa Gabb Mitra longa Gabb, Trans. Am. Phil. Soc, vol. 15, 1873, p. 219.— Brown & PiLSBRY, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 346, pi. 24, fig. 11.— Maury, Bull. Am. Pal., vol. 5, 1917, p. 238, pi. 11, figs. 11, 11a; Miocene, Santo Domingo. — Pilsbry, Proc. Acad. Sci. Phila., vol. 73, 1921, p. 339, pi. 24, fig. 3 (Type); Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 273, pi. 6, fig. 10; Gatun Stage, Canal Zone. Good examples of this species were found at Loc. 267, C. A. S., horizon M - N, and Loc. 351, C. A. S., near Punta Pua, in the latter case near the middle of the Tubera group ; it is believed to belong to both of these horizons. It occurs at Gatun, according to Olsson. 59. Mitra mauryae Anderson, new species Plate 8, figures 4, 5 Shell moderate in size, height of holotype, incomplete, 32 mm., width 10 mm., somewhat biconic in form; spire a little longer than the body whorl; suture slightly impressed; whorls rounded above and slightly convex below the shoul- ders; body whorl obversely pyriform; spire (incomplete) con- sisting of six whorls ; sculpture cancellated ; aperture long and narrow; canal long and straight; outer lip thin, not lirate within columella bearing four plications, the larger above, slightly more distant than the others ; surface ornamented by 20 vertical ribs on the penultimate whorl, crossed by seven spiral threads, the two forming a cancellated sculpture very similar to that of M. syra Dall. This latter species, from the Silex beds of Tampa, Florida, is its nearest ally, though only one-half the length of the Colombian fonn, and with a some- what more uniform taper to the apex. Holotype: No. 4619, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon L, gray shales along the beach one mile west of Puerto Colombia; Miocene. This species is known from only a single slightly imperfect specimen obtained from the gray shales along the beach a mile west of Puerto Colombia at Loc. 267-L, C. A. S. These shales underlie the Tubera group, and probably form a part of the Las Perdices group. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 131 60. Scobinella morierei (?) (Laville) Plate 8, figures 6, 7 Euchilodon morierei (Lav.) in Cossmann, Jour. Conch., vol. 61, 1913, p. 34, pi. 3, figs. 6, 7. Scobinella morierei, Olsson, Bull. Am. Pal., vol. 9, 1922, p. 251, pi. 4, figs. 3, 4; Gatun Stage, Canal Zone. — Maury, Bull. Am. Pal. vol. 10, 1925, p. 345, pi. 34, figs. 1,8; Miocene, Trinidad Island. A single but beautifully preserved example of this interest- ing species was found in the clay shales, underlying the Tubera group a mile or more west of Puerto Colombia, asso- ciated with Turris alhida, Mitra iimuryce (n. sp.), Phalium dalli (n. sp.), Dentalium granadanitm (n. sp.), and others. This example is larger and more robust than the figures given by Olsson, but otherwise is not easily distinguished from the form found in the Canal Zone. The ratio of length to width is less, being more nearly 3 : 1 in greatest width. The aperture is relatively wider, and the columellar plications are different, though the difference seems hardly to be specific in value. The species has not been found in higher beds in Colombia, as far as known. Plesiofype: No. 4620, Mus. Calif. Acad. Sci., from Loc. 267-L, gray shales along the beach one mile west of Puerto Colombia; Miocene. 61. Fasciolaria olssoni Anderson, new species Plate 8, figures 1, 2, 3 Shell large, thick, robust, biconic in form, smooth, showing lines of growth and faint spiral markings ; length of holotype (without apex) 79 mm.; greatest width 57 mm.; paratype with six whorls ; spire high, subconic, acuminate ; upper sur- face of the whorls concave, terminating above in a clasping collar; suture distinct above the collar; shoulder of body whorl bearing five or more rounded tubercles, forming short broad ridges below, but none above the shoulder; aperture oval, terminating above in an acute angle, below in a narrow straight canal; inner margin of aperture evenly calloused, outer lip lirate within, margin unknown; umbilical chink 132 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. closed ; pillar slightly twisted, bearing three rounded plications. The largest example of this shell, although not complete, measures 113 mm. in length, and 71 mm. in width. The tubercles do not develop on the shoulders until about the fifth whorl, and become stronger on older shells. Holotype: No. 4617; paratype: No. 4618 (C. A. S. type coll.), from Loc. 267-P, C. A. S., Tubera mountain, Colombia; Miocene. The surface of the older shells become much pitted by wonn borings. This species is possibly the one listed and figured by Olsson as F. gorgasiana (Brown & Pilsbry).^® This shell is fairly plentiful in the Tubera group of the Colombian Miocene, and has been collected at Loc. 267, C. A. S., horizons P and R, and at other points which repre- sent the horizon of the Gatun formation of the Canal Zone. 62. Fasciolaria kempi (Maury) Siphonalia kempi Maury, Bull. Am. Pal., vol. 4, 1910, p. 138, pi. 5, fig. 5; Chipola marls, Florida Miocene. Fasciolaria kempi Maury, Bull. Am. Pal., vol. 5, 1917, p. 245, pi. 12, fig. 4; Miocene, Santo Domingo. This shell is not rare in the Tubera group of the Colombian Miocene, and has been collected at Loc. 351, C. A. S., horizon near M - N, and at Loc. 305, C. A. S., near Turbaco. 63. Fusinus henekeni (Sowerby) Fusus henekeni Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849, p. 49; Miocene, Santo Domingo. — Guppy, Geol. Mag. Lond., vol. 1, 1874, p. 439.— Guppy, Quart. Jour. Geol. Soc. Lond., vol. 32, 1876, p. 524, pi. 28, fig. 6; Miocene, Haiti.— Maury, Bull. Am. Pal., vol. 5, 1917 p. 242, pi. 12, fig. 1; Cercado de Mao, Miocene, Santo Domingo. Two examples of this species were found at the village of Tubera in the upper part of the Tubera group, Loc. 267, C. A. S., horizon R. The rounded longitudinal ribs are pro- nounced on every whorl, from the nuclear to the body whorl, all of which are crossed by the strong spiral cords and lines described for this species. '« Olsson, A. A.— Bull. Am. Pal. vol. 9, 1922, p. 227, pi. 8, fig. 9; Gatun Stage, Canal Zone. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA i;^;^ 64. Fusinus magdalenensis Anderson, new species Plate 15, figures 1, 2, 3 Shell large, height of holotype, incomplete, 110 mm., width 44 mm., fusiform, with high spire and long canal; spire con- sisting of nine whorls below the nuclear stage, the earlier ones only showing vertical ribs; spire sculptured by 10 to 15 strong revolving ridges, of two alternating ranks; body whorl con- taining 14 such ridges, only a few of which are of secondary rank; canal long and somewhat recurved near the terminus; pillar calloused throughout, and ornamented externally by spiral threads and cords alternating as above; spiral cords sharply ridged at the top. This shell bears some resemblance to F. henikeni (Sow.), var. veatchi Maury, but it is larger, has fewer and coarser spiral cords, longer and more recurved canal, and a clearly more calloused pillar. It is not unlike a large and strongly marked species from the Gulf of California, namely, Fusinus diipetitthouarsii (Kiener), and it may well be a precursor of the same. Holotype: No. 4651, Mus. Calif. Acad. Sci., from Loc. 267, horizon P, north slope of Tubera mountain, Colombia; Miocene. 65. Melongena propatulus Anderson, new species Plate 11, figures 1, 2 Shell large, heavy, height of holotype, incomplete, 108 mm., greatest width 91 mm., spire low and rounded, body pyriform or conical below the rounded shoulder, almost spineless, or having only few and inconspicuous spines on the shoulder of the whorl; holotype bearing two small, tubercle-like spines at the base, near aperture ; whorls five ; spire low but acuminate, rounded below the three nuclear whorls which form the apex ; suture covered by an elevated collar; aperture ovate, notched behind, slightly notched on the shoulder; outer lip showing a disposition to form crenulations ; inner lip broadly calloused ; canal broad, as in M. patulns; pillar broad, and flattened below ; surface marked by strong, flat spiral cords, crossed by 134 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. strong wavy lines of growth; spiral cords stronger near the base, one or more cords bearing a few small tubercles. This shell has its closest ally in M. patulus, living on the Pacific coast and in the Gulf of California. Careful com- parison has been made with good examples of this species in the collectioBs of the California Academy of Sciences, and with M. melongetta Linn, from the Caribbean region. It dif- fers from both. Holotype: No. 4632, Mus. Calif. Acad. Sci., from Loc. 267, horizon R, Tubera village, Colombia; Miocene; embedded in sandstone near the top of the Tubera group. 66. Solenosteira hasletti Anderson, new species Plate 16, figures 7-A, 8 Shell not large, height of holotype 48 mm., width 30 mm., thickened, biconic in outline, spiney, not nodose ; spire pagoda- like, with 5 or 6 whorls more or less concave above, the whorls culminating above in a collar clasping the preceding one; suture completely covered; surface marked by numerous revolving threads ; on the upper slope four or five of these are heavier, with interspaces occupied by three to five finer threads, all of which, under the lens, appear beaded ; lower slope ornamented in the same manner, but with more numer- ous heavy threads; periphery of each whorl supporting about seven strong spines that point upward and outward, but- tressed by a low ridge beneath and above ; aperture ovate, with narrow angle above forming a notch ; outer lip slightly angu- lated, somewhat lirate within ; inner lip symmetrically curved ; pillar calloused near the aperture, recurved without; canal long and slightly recurved ; umbilical area calloused, but show- ing a decided depression. This shell is not unlike Solenosteira alternata (Nelson) from the Zorritos formation of Peru, but it is more strongly sculptured, and considerably more spinose in its mature form. It is found in many parts of the Tubera group, and was obtained at Loc. 267, C. A. S., in horizon M - N, and horizon P. It is named in honor of Mr. Thomas D. Haslett. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA ^35 Holotype: No. 4169, Mus, Calif. Acad. Sci., from Loc. 267 — M-N, Tubera group, Colombia; paratype: No, 4170, C. A. S., from Loc. 305, C. A. S., Turbaco, Colombia; para- type: No. 4171, Mus. Calif. Acad. Sci., from Loc. 304, C. A. S., from four miles east of Santa Rosa, Colombia, on ranch of Mrs. Gomez; Miocene. 67. Solenosteira santaerosae Anderson, new species Plate 13, figures 7, 8, 9, 10 Shell of medium size, height of holotype, incomplete, 47 mm., width 35 mm., subconic in form, spinose, spiral sculp- ture pronounced; whorls five to seven below the nucleus; upper slope of whorls broad, bearing about seven strong ridges extending to the clasping sutural collar, sculptured by numer- ous revolving lines; lower slope abrupt and concave down- ward, crossed by numerous revolving lines or threads, among which appear a few stronger cords near the center of the lower surface; shoulders set with strong spines, sloping downward on the body whorl, but upward on the younger whorls ; suture concealed by a clasping collar ; pillar thick and short, reflexed ; aperture ovate, narrowed above and at the canal ; canal reflexed; umbilicus large; general appearance of the shell slouching and depressed. Holotype: No. 4641, Mus. Calif. Acad. Sci., from Loc. 304, C. A. S., 4 miles east of Santa Rosa; paratype: No. 4642, Mus. Calif. Acad. Sci., from Loc. 305, C. A. S., horizon P, near Turbaco, Colombia; Miocene. This species is not infrequent in the Tubera group of the Colombian Miocene. It has been obtained at Loc. 299-A, C. A. S., near the middle of the group, at Loc. 304, and at Loc. 305, C. A. S., lower down in the group, though not at the lowest horizon. 68. Phos tuberaensis Anderson, new species Plate 9, figures 1, 2, 3 Shell of medium size, or large; spire high, somewhat tur- rited; height of holotype 50 mm., width of body whorl 26 mm., whorls seven in number, convex, ornamented chiefly ]^36 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Sek. by spiral lines; two nuclear whorls smooth; next four whorls bearing low, rounded vertical ribs, and about 12 slightly raised spiral threads; upper slope of whorls concave; shoul- ders tuberculated ; body whorl having 10 low ribs, crossed by spiral threads, heavier on base of shell; aperture oval, nar- rowed above ; canal short, reflexed ; outer lip sharp, lirate within; pillar bearing one anterior plication, not crusted. Holotype: No. 4621, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon P, on the north slope of Tubera mountain, Colombia; paratype: No. 4622, Mus. Calif. Acad. Sci., from Loc. 305, C. A. S., near Turbaco, Colombia ; Miocene. This shell resembles Phos siibsemicostatus Brown & Pilsbry, but it is larger, has a more rugged sculpture and prominent tubercles. It is not unlikely that the two species are nearly allied, though they are not identical. This species is not rare in the Tubera group, and the type was obtained at Loc. 267, C. A. S., horizon P, on the north slope of Tubera mountain. It has been found also at Loc. 305, near Turbaco. 69. Phos turbacoensis Anderson, new species Plate 15, figures 6, 7 Shell large, heavy, strongly sculptured ; spire high, acumi- nate, heavily ribbed ; whorls nine in number, concave above, with slightly elevated collar; costate below the shoulder, hav- ing 12 heavy ribs which are crossed by five or six heavy revolving threads below the shoulder ; shoulder slightly tuber- culate; body whorl irregularly ribbed, and ornamented with strong spiral threads with wide interspaces; interspaces some- times containing intermediary lines; pillar short with one anterior plication; aperture arcuate-ovate; outer lip sharp, lirate within; pillar not calloused; canal short, reflexed; three nuclear whorls smooth; following six becoming gradually more strongly sculptured; height of holotype 55 mm., width of body whorl 27 mm., height of aperture 26 mm. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA \yj Holotype: No. 4654, Mus. Calif. Acad. Sci., from Loc. 305, C. A. S., near the village of Turbaco, Colombia; Miocene. This shell resembles Phos veatchi Olsson, but it is larger, more strongly sculptured, and has a higher spire. 70. Phos baranoanus Anderson, new species Plate 16, figures 4, 5 Shell rather large, conico-ovate, spire high, acuminate; whorls nine in number, convex; suture distinct, not impressed; two nuclear whorls smooth; next five whorls bearing small vertical ribs and four to eight spiral threads, producing a finely cancellated sculpture ; last two whorls smooth, showing growth lines, but almost no spirals, except on the base; aperture arcuate-ovate, narrow above; outer lip sharp, lirate within; inner lip not crusted ; pillar bearing a single anterior plication ; canal reflexed. Height of the holotype is 51 mm., width 21.5 mm. This shell is not rare in the Tubera group, and has been obtained at Loc. 325-A, C. A. S., near Cibarco; Loc. 299, near Baranoa; Loc. 325, near Usiacuri; and at Loc. 267, C. A. S., horizon P, on the north slope of Tubera mountain. Holotype: No. 4657, Mus. Calif. Acad. Sci., from Loc. 325-A, C. A. S., horizon P, on the north slope of Tubera mountain, Colombia; paratype: No. 4657-A, Mus. Calif. Acad. Sci., from Loc. 299, C. A. S., near Plott's well S. W. of Baranoa, Colom- bia; Miocene. 71. Murex domingensis Sowerby Murex domingensis Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849? p. 49, pi. 10, fig. 5; Miocene, Santo Domingo. — Maury, Bull. Am. Pal., vol. 5, 1917, p. 265, pi. 16, figs. 3, 4, 5, 6; Cercado de Mao, Miocene, Santo Domingo. A single specimen of this shell was found at Loc. 267, C. A. S., horizon P, on the north slope of Tubera mountain, near the middle of the Tubera group. March 29, 1929 138 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th See. 72, Murex mississippiensis Conrad Murex mississippiensis Conrad, Jour. Acad. Nat. Sci. Phila., vol. 1, 1848, p. 116, pi. 11, fig. 30.— Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1890, p. 130.— Dall, Bull. U. S. Nat. Mus. No. 90, 1915, p. 73, pi. 5, fig. 10; Tampa Silex beds, Miocene, Florida, etc. A single example of this shell was obtained at Loc. 351, C. A. S., near Punta Pua, some 20 miles north of Cartagena, in the lower part of the Tubera group. 73. Typhis siphonifera Dall Plate 9, figure 8 Typhis siphonifera Dall, Bull., U. S. Nat. Mus., No. 90, 1915, p. 77, pi. 13, fig. 9; Tampa Silex beds, Tampa, Florida. Typhis lingulifera Dall, var. coslaricensis (?) Olsson, Bull. Am. Pal., vol. 9, 1922, p. 304, pi. 10, figs. 22, 29; Miocene, Costa Rica. A single example of this interesting species was found at Loc. 325-A, C. A. S., near Cibarco, a few miles north of Usiacuri, and near the middle of the Tubera group, A careful comparison of this well preserved specimen with Ball's figure and description leaves no room for doubt as to its determina- tion, although the spire is slightly higher in our specimen. In this example the spire consists of seven whorls, including the two that form the nucleus. The specimen bears some resem- blance to T. linguifera Dall, but the latter has long and in- curved spines where the varices meet the shoulder of the whorl, giving it a decidedly spiney appearance. The tubes arising from the shoulder in the interspaces between the spines form a distinguishing mark. Plesiotype: No. 4625, Mus. Calif. Acad. Sci., from Loc. 325-A, C. A. S., near Cibarco, Colombia; Miocene, 74, Distortrix simillima (Sowerby) Triton simillima Sowerby, Quart, Jour, Geol. Soc. Lond., vol. 6, 1849, p. 48; Miocene, Island of Haiti. Persona simillima, Guppy, Quart. Joiu". Geol. Soc. Lond., vol. 22, 1866, p. 288, pi. 17, fig. 13; Miocene, Jamaica. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 139 Distortio (Distortrix, Persona) gatunensis Toula, Jahrb. d. K. K. Geol. Reichs., Bd. 58, 1909, p. 700, pi. 25, fig. 10; Gatun formation, Canal Zone.— Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 356, pi. 26, fig. 8; Gatun formation. Canal Zone. Distortrix simillima, Maury, Bull. Am. Pal., vol. 5, 1917, p. 271, pi. 17, figs. 4, 5; Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 305; Gatun Stage, Canal Zone. — Maury, Bull. Am. Pal., vol. 10, 1925, p. 368; Miocene, Trinidad Island. A good number of examples of this shell was obtained at the Spillway of the Canal in 1914, and since then it has been collected at Loc. 267, C. A. S., horizons P and R, and later at Loc. 325-A, near Cibarco, all of which represent a horizon near the middle, or in the upper part of the Tubera group of the Colombian Miocene. It has not been found in the lowest horizon of the same. 75, Cypraea henekeni Sowerby Cyprcea henekeni Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849, p. 45, pi. 9, fig. 3; Miocene, Santo Domingo. — Gabb, Am. Phil. Soc. Trans., vol. 15, 1873, p. 235.— GUPPY, Geol. Mag. Lond., vol. 1, 1874, p. 440. — GuppY, Quart. Jour. Geol. Soc. Lond., vol. 32, 1876. p. 528. — Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 356.— Maury, Bull. Am. Pal., vol. 5, 1917, p. 278, pi. 19, fig. 4; Miocene, Santo Domingo. Two good examples of this species were obtained at Loc. 267, C. A. S., horizon M - N, at the west base of Tubera mountain, in the lower part of the Tubera group, and it has not yet been found higher in the group. 76. Cypraea (Pustularia) gabbiana Guppy Plate 15, figures 4, 5 Pustularia nucleus, Gabb, Trans. Am. Phil. Soc, vol. 15, 1873, p. 236. (Not of Linnaeus). Cyprcea pustulata, Guppy, Geol. Mag. Lond., vol. 1, 1874, p. 440. (Not of Lamarck). Cyprcea gabbiana Guppy, Quart. Jour. Geol. Soc. Lond., vol. 32, 1876, p. 528, pi. 29, fig. 10; Miocene, Santo Domingo. — Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1890, p. 165, Cyprcea {Pustularia) gabbiana, Maury, Bull. Am. Pal., vol, 5, 1917, p. 280, pi. 19, fig. 12; Miocene, Santo Domingo. 140 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. A sing-le well preserved example of this species was obtained at Loc. 351, C. A. S., near Punta Pua, near the middle of the Tubera group. The species is doubtless related to C. pustularia Lam., found in the Gulf of California, though it is narrower, and has more numerous transverse bars upon the bucal surface. Plesiotype: No. 4653, Mus. Calif. Acad. Sci., from Loc. 351, C. A. S., horizon P, near Punta Pua, Colombia; Miocene 77. Ovula (Neosimnia) puana Anderson, new species Plate 9, figures 9, 10 Shell small, length of holotype, broken, 20 mm., width 10 mm., biconic, smooth, bearing a subcentral, angular hump, but little elevated; aperture narrow, outer lip apparently sim- ple, inner lip smooth and polished. Holotype: No. 4626, Mus. Calif. Acad. Sci., from Loc. 351, C. A. S., near Punta Pua, Colombia, near the middle of the Tubera group ; Miocene. This species is nearly related to Ovula emarginata Sowerby, from the Bay of Panama, but it differs in the elevation of the transverse hump. Only a single specimen of this shell was obtained at Loc. 351, C. A. S., near Punta Pua, near the middle of the Tubera group. It is herein included only for the purpose of making the record as complete as our material will permit. 78. Malea ringens (Swainson) Plate 12, figtires 1, 2, 3, 4, 5, 6 Cassis ringens Swainson, Blith. Catal. 1822, App. p. 4. — Sowerby, Tankerv. Catal., 1825, App. 21. Dolium ringens (Swains.) Reeve, Conch. Icon., vol. 5, 1849, pi. 4, fig. 5; living, Payta, Peru. Malea ringens (Swains.) Conrad, Pac. R. R. Repts., vol. 6, 1855, pt. 2, p. 72, pi. 5, fig. 22; Miocene, Gatun, Panama. This species has not recently been listed from the Miocene of the Caribbean region, although Conrad reported it from Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA ]^41 Panama Miocene beds as early as 1855. It was obtained by the writer at the Spillway of the Canal in 1914, and since then at a number of points in the Tubera group of the Colombian Miocene. The identity of the fossil Colombian species with the living form from the Gulf of California is shown in the illustrations presented herein. It differs from the more com- mon form, Malea camura Guppy, in having a higher spire, narrower and flatter revolving ribs, as is illustrated in Maury's figure of the latter, and a longer canal. The outer lip is not preserved in most of our fossil examples, but it appears to be represented in Toula's figure (pi. 30, fig. 7),^^ which agrees with some of our material from Gatun. In the Colombian Miocene it was obtained at the following localities : Loc. 267, C, A. S., horizon P, north slope of Tubera moun- tain; Loc. 267, C. A. S., horizon R, Tubera village, near top of group; Loc. 299, C. A. S., near Baranoa, near middle of the group; Loc. 305, C. A. S., southeast of Turbaco, Depart, de BoHvar; Loc. 351, C. A. S., near Punta Pua, near the mid- dle of the group. Its range is, therefore, through the upper part of the Tubera group of the Colombian Miocene. Plesiotype: No. 4633, Mus. Calif. Acad. Sci., recent shell from Bay of Panama; plesiotype: No. 4634, Mus. Calif. Acad. Sci., from Loc. 267, C. A, S., horizon P, Tubera mountain ; plesiotype: No. 4635, Mus. Calif. Acad. Sci., from Loc. 299, C. A. S., horizon P, near Plott's well, S. W. of Baranoa, Colombia ; Miocene. 79. Cassis (Phalium) dalli Anderson, new species Plate 14, figures 10, 11, 12, 13 Shell small, height of holotype, young shell, 13 mm., width 11 mm., globose, coronated, with moderate or low spire ; shell orna- mented by fine spiral sculpture covering the entire body, crossed by lines of growth; aperture lunate, outer lip thin on the two examples found ; canal short and recurved. The spire of this species consists of two smooth nuclear whorls, followed by three rapidly expanding whorls which are tabular above, ='Toula, F., Jahrb. d. K. K. Geol. Reichs., Bd. 61, 1911, p. SOO. 142 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. angulated on the shoulder and convexly rounded below. The angles of the shoulder bear 12 to 13 flattened spines, elongated laterally, forming a distinct corona. This shell bears a certain resemblance to P. moniliferum (Guppy), but has a much finer sculpture, only a single row of tubercles, a lower spire, and is of smaller size. Holotype: No. 4649; paratype: No. 4650, Mus. Calif. Acad. Sci., from Loc. 267-L, Las Perdices group underlying the Tu- bera group a mile or more west of the Pier at Puerto Colombia; Miocene. This species is represented by two examples from Loc. 267, C. A. S., horizon L, the gray shales of the Las Perdices group underlying the Tubera group a mile or more west of the pier at Puerto Colombia. It has not been found at any higher horizon. 80. Cassis (Phalium) moniliferum Guppy Cassis moniUfera Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 287, pi. 17, fig. 8; Miocene, Jamaica. Phalium moniliferum Maury, Bull. Am. Pal., vol. 5, 1917, p. 274, pis. 18, figs. 4, 5; 19, fig. 1.— Olsson, Bull. Am. Pal., vol. 9, 1922, p. 307, pi. 12, fig. 11; Miocene. This species was obtained at the Spillway of the Canal in 1914, but has not yet been certainly recognized in the Miocene of Colombia. 81. Sconsia laevigata (Sowerby) Cassidaria Icevigaia Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849, p. 47, pi. 10. fig. 2. Cassidaria sublcevigata Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 287, pi. 27, fig. 9. Cassidaria laevigata, Guppy, Geol. Mag. Lond., vol. 1, 1874, p. 439. — Quart. Jour. Geol. Soc. Lond., vol. 32, 1876, p. 525. Sconsia Iczvigata, Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 356.— Maury, Bull. Am. Pal., vol. 5, 1917, p. 275, pi. 19, fig. 2; Cercado de Mao, Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 308. This species was obtained at the Spillway of the Canal in 1914, and since then it has been collected at Loc. 267, C. A, S., Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA ^43 horizon M - N, and at Loc. 351, C. A. S., near Punta Pua, in the latter case from near the middle of the Tubera group of the Colombian Miocene. It has not been found at any higher horizon, as far as known. 82. Ficus colombiana Anderson, new species Plate 13, figures 1, 2 Shell medium or large, pyriform, graceful in outline, sculp- ture decussated, suboval; height of holotype 41.5 mm., width 29 mm., height of paratype (incomplete) 59 mm., width 42 mm. ; spire low, even in young shells ; upper slope gentle, curving gracefully to the sides; nuclear whorls smooth; sculp- ture consisting of spiral cords widely spaced, with four or five intermediary lines, the central of which is stronger than the others ; aperture wide, suboval ; pillar slightly curved. Holotype: No. 4636; paratype: No. 4637, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon P, Tubera mountain, Colombia; Miocene. The nearest ally of this species is Picas decussata (Wood) from the Bay of Panama, Magdalena Bay and the Gulf of California. The principal difference in these species seems to be in the general outline and sculpture. The fossil species is more robust, has a shorter pillar and canal, and a much coarser sculpture. It differs from F. carhasea (Guppy), in its more rounded outline as well as in sculpture. This species is represented by four good examples from Loc. 267, C. A. S., two of which came from horizon P, and two from horizon R, and accordingly from the middle and upper part of the Tubera group. Other examples have been found at other localities in the middle part of the same group. 83. Strombina chiriquiensis Olsson StromUna chiriquiensis Olsson, Bull. Am. Pal., vol. 9, 1922, p. 302, pi. 10, figs. 14, 24; Miocene, Costa Rica. This species was found abundantly at Loc. 325-A, C. A. S., near Cibarco, near the middle of the Tubera group of the Colombian Miocene. 144 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 84. Dentalium granadanum Anderson, new species Plate 13, figure 3 Shell large, subcircular in section, gently curved, tapering very gradually; both ends complete when found, but sub- sequently broken; surface sculptured by 24 rounded but irregu- lar longitudinal ribs, with no intermediate lines, the ribs con- tinuing to the basal end of the shell ; length of incomplete holo- type not less than 55 mm.; greatest width 11 mm. When complete this shell was not less than 100 mm. in length. Its nearest ally seems to be one from Costa Rica described by Olsson as D. uscarianum, coming from the Uscari stage of the Miocene. Its resemblance, however, to D. mississippiensis Conrad"** should be pointed out also. Holotype: No. 4638, Mus. Calif. Acad. Sci., from Log. 267, C. A. S., horizon L, Las Perdices group, Puerto Colombia; Miocene. A single example was obtained from the gray shales of the Las Perdices group below the Tubera group, a mile west of Puerto Colombia. 85. Serpulorbis papulosa (Guppy) Vermetus papulosa Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 292, pi. 17, fig. 3; Miocene, Santo Domingo. — Quart. Jour. Geol. Soc. Lond., vol. 32, 1876, p. 519; occurrence as above. Serpulorbis papulosa (Guppy) Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1585.— Maury, Bull. Am. Pal., vol. 5, 1917, p. 291, pi. 22, fig. 10; Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 317, pi. 12, fig. 1; Gatun Stage, Costa Rica. — Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 73, 1921, p. 376, as above. — Maury, Bull. Am. Pal., vol. 10, 1925, p. 377, etc.; Springvale horizon, Miocene, Trinidad Island. Examples of this species have been obtained at Loc. 267, C. A. S., horizon M - N, and at Loc. 351, C. A. S., near Punta Pua, in the latter case from near the middle of the Tubera group of the Colombian Miocene. «Jour. Acad. Nat. Sci. Phila., vol. 1, 1848, p. 112, pi. 11, fig. 1. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA ^45 86. Serpulorbis granifera (Say) Serpula granifera Say, Jour. Acad. Nat. Sci. Phila., vol. 4, 1824, p. 154, pi. 8, fig. 4.— Reprint, Bull. Am. Pal., vol. 1, 1896, p. 330, pi. 8, fig. 4; Miocene, Maryland. Vermelus granifera, Martin, Md. Geol. Surv., 1904, p. 232, pi. 54, figs. 14, 15. Serpulorbis granifera, Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1892, p. 303. — Maury, Bull. Am. Pal., vol. 5, 1917, p. 291, pi. 22. fig. 9; Miocene, Santo Domingo. Examples of this species were obtained at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena, near the mid- dle of the Tubera group. 87. Petaloconchus sculpturatus H. C. Lea Petaloconchus sculpturatus Lea, Trans. Am. Phil. Soc, vol. 9, 1845, p. 233, pi. 34, fig. 3. Petaloconchus domingensis Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1849, p. 51, pi. 10, figs. 8, a, b, c. — Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 359; Gatun formation. Canal Zone. Petaloconchus sculpturatus, Gabb, Trans. Am. Phil. Soc, vol. 25, 1875, p. 240; Miocene, Santo Domingo. — Guppy, Quart. Jour. Geol. Soc. Lond., vol. 32, 1876, p. 519. Vermetus {Petaloconchus) sculpturatus, Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1892, p. 305.— Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 73, 1921, p. 377; Miocene, Santo Domingo. Petaloconchus sculpturatus, Olsson, Bull. Am. Pal., vol. 9, 1922, p. 318, pi. 14, figs. 10, 15; Miocene, Canal Zone. — Maury, Bull. Am. Pal., vol. 10, 1925, p. 378, pi. 41, figs. 2, 4, 7; Miocene, Trinidad Island. This species occurs frequently in the Tubera group of the Colombian Miocene. It has been obtained at Loc. 267, C. A. S., horizon M - N, and Loc, 351, C. A. S., near the middle of the group, and at Loc. 325-A, also near the middle of the group. It occurs at higher horizons as well, at other localities. 146 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Pelecypoda 88. Yoldia pisciformis Brown & Pilsbry Yoldia pisciformis Brown & Pilsbry, Proc. Acad. Nat. Sci., Phila., vol. 17, 1917, p. 38, pi. 6, fig. 3; near Cartagena, Colombia. This species is abundant about Tubera mountain in the mid- dle part of the Tubera group, as at Loc. 267, C. A. S., hori- zons P and R, Tubera group, and it has also been found at Loc. 304, C. A. S., four miles east of Santa Rosa, near the Colombian coast. 89. Area (Scapharca) patricia Sowerby Area Patricia Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1850, p. 52; Miocene, Santo Domingo. — Woodring, Science, vol. 62, 1925, pp. 518, 519. Although Gabb was very confident that he had found and identified Sowerby 's species, Area patricia with the living Area girandis Brod. & Sowerby, it appears that his confidence was not well grounded in fact. W. P. Woodring has sum- marized the matter pertaining to the former species, including under it the following as synonymous : Area (Anadara) grandis (Brod. & Sow.), Gabb, 1873. Seapharca (Argina) tolepia Dall, 1898. Scapharca arthurpennelli Maury, 1917. Area {Argina) tolepia (Dall), Pilsbry, 1922, etc. This species has been found in the Tubera group of the Colombian Miocene at three different localities, and in fact seems to be quite common. In all of the examples the ribs number about 30, are slightly nodose, and the shell has the form and hinge characters described by Dall for his Seapharca tolepia. It occurs abundantly at Loc. 267, C. A. S., in hori- zons M - N, and P, Tubera mountain; Loc. 305, C. A. S., near Turbaco; and at Loc. 265, C. A. S., near Punta Paralillas, north of Monitos, on the Colombian coast. At the last point it was almost the only fossil found, but was sufficient to con- firm the Miocene age of the strata, determined as such on other grounds. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA \^y 90. Area (Noetia) macdonaldi Dall Area (Noetia) macdonaldi Dall, Smiths. Misc. Coll., vol. 59, 1912, p. 9. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 366, pi. 25, figs. 4-7; Miocene, Costa Rica. According to Dall this species is nearly related to Area trinitaria Guppy, from the Miocene of Trinidad Island. Examples of it were found at Loc. 323, C. A. S., at the Spill- way of the Canal in 1914, and subsequently at Loc. 267, C. A. S., horizons M - N, and P, of the Tubera group, Tubera mountain, and at Loc. 299, C. A. S., near Baranoa, Colombia. It is one of the abundant forms of this group. 91. Area (Scapharca) actinophora Dall Area {Scapharca) actinohpora Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1898, p. 647, pi. 2>i, fig. 26; Monkey Hill, Canal Zone. This species was collected at the Spillway of the Canal in 1914, and subsequently at two separate localities in the Colom- bian Miocene, as at Loc. 267, C. A. S., horizon P, Tubera group, and at Loc. 351, C. A. S., near Punta Pua, some 20 miles north of Cartagena. At the latter locality three or four good examples were obtained which agree in all essentials with those of the Gatun formation. 92. Area (Scapharca) dariensis Brown & Pilsbry Area {Scapharca) dariensis Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 362, pi. 22, fig. 10; Gatun formation. Canal Zone. This species was found abundantly at the Spillway of the Canal in 1914, Loc. 323, C. A. S., and has since been found at Loc. 267, C. A. S., horizon P, on the north slope of Tubera mountain, in the middle of the Tubera group. It appears to belong to the group of Area (Scaph.) ineqnilatcralis (Guppy) from the Miocene of Trinidad. 93. Area (Area) oeeidentalis Philippi Area {Area) oeeidentalis Philippi, Abbild. und Beschreib., vol. 3, 1847, p. 29, pi. 4, figs. 4, a, b; living, Caribbean Sea. — Maury, Bull. Am. Pal., vol. 5, 1917, p. 327, pi. 29, fig. 3; Zone H, Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 353, pi. 22, fig. 1; Miocene, Costa Rica.— WooDRiNG, Mioc. Moll. Bowden, Jamaica, Carnegie Inst. Publ. No. 1925, p. 29, pi. 2, figs. 8, 9; Bowden beds, Jamaica. 248 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. This species has been obtained abundantly in the Bay of Cartagena and has been collected from the Miocene beds of Loc. 351, C. A. S., near Punta Pua, 20 miles north of Carta- gena. It is a variable form and it would be surprising if it did not persist from the lower Miocene into the living fauna. 94. Area (Anadara) usiacurii Anderson, new species Plate 19, figure 6; plate 20, figure 6; plate 21, figure 4 Area grandis Brod. & Sow., Pilsbry (in part), Proc. Acad. Nat. Sci. Phila., vol. 73, 1921, p. 404; Miocene, near Cartagena, Colombia. Area grandis ? waringi F. & H. K. Hodson, Bull. Am. Pal., vol. 13, 1927, p. 7, pi. 7, figs. 1, 4; Miocene, N. Venezuela. Shell large, solid and heavy; nearly equivalve; length of holotype 105 mm., height 103 mm., thickness of valve from hinge plane to back 50 mm. ; radial ribs 27 in number, slightly flattened, heavy, beaded on the anterior surface, and less strongly so elsewhere ; intercostal spaces nearly equal in width to the ribs, marked by strong lines of growth; cardinal area broad, forming a nearly symmetrical triangle crossed by four to six grooves in fully grown specimens, sloping to the outer angles of the area, but not quite meeting on the median line ; hinge heavy, set with about 48-58 thin, close-set, often branch- ing teeth, which in the center are vertical, but toward the ends curve outwardly and are often broken by an oblique line; margin of shell strongly denticulate within, showing about 23 broad denticulations. Holotype: No. 4158, Mus. Calif. Acad. Sci., from Loc. 306, C. A. S., at the northeast border of the village of Usiacuri, Colombia; paratype: No. 4159, Mus. Calif. Acad. Sci., from Loc. 267 M - N, C. A. S., Tubera group, Colombia; Miocene. This species is even more nearly related to Area grandis Brod. & Sow. than is the form figured by Pilsbry as such, and by Maury as Area patrieia Sowerby^® for which the name Area patriareha is here proposed. A comparison of the hinges and cardinal areas clearly shows several marked differences. The branching of the cardinal teeth near the ends of the hinge in the Colombian species is a distinctive mark. Although Dr. 28 Bull. Am. Pal. vol. 5, p. 337, pi. 27, fig. 1. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA ^49 Pilsbry had in his collection nine specimens from the Colom- bian coast (p. 404) he seems not to have noted the points in which they doubtless differ from the Dominican species or from the form living at Panama and other Pacific points. This species is found in many parts of the Colombian marine Miocene associated with other purely marine forms. The holotype was obtained from Loc. 306, at the northeast border of the village of Usiacuri, more than 1,000 feet above the base of the group, where it is very abundant. The para- type comes from the uppermost part of horizon M - N of the Tubera group, though it is abundant in higher horizons, as P and Q, and in still higher beds near the village of Usiacuri. 95. Area (Anadara) patriarcha Anderson, new name Area grandis Brod. & Sow., Gabb, Trans. Amer. Phil. Soc, vol. 15, 1873, p. 253 (in part); Miocene, Santo Domingo. Area grandis Brod. & Sow., Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 73, 1922, p. 404, pi. 40, fig. 1; Miocene, Santo Domingo. Not Area grandis Brod. & Sow.; living, Bay of Panama, etc. Area patricia Sowerby, Maury, Bull. Am. Pal., vol. 5, 1917, p. 337, pi. 27, fig. 1 ; Caimito, Rio Cana, Santo Domingo. This species has not yet been correctly reported from Colombia, although it appears to be quite abundant in the Miocene of Santo Domingo. As shown by the figure supplied by Maury the cardinal teeth are not numerous, and are cor- respondingly very coarse. It lacks many of the details of fonn and dentition given for Area grandis Brod. & Sow., and for Area (Anadara) iisiaeurii Anderson. 96. Area (Seapharea) auriculata Lamarck Area auriculata Lam., An. s. Vert., vol. 6, 1819, p. 43; living fauna. — Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1898, p. 647; Miocene, Bowden, Jamaica.— M.\ury, Bull. Am. Pal., vol. 5, 1917, p. 339, pi. 28, fig. 3; Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 362, pi. 22, fig. 3; Miocene, Costa Rica. — Maury, Bull. Am. Pal., vol. 10, 1925, p. 201, pi. 4, fig. 2; Miocene, Trinidad Island. This species has been found living in the Bay of Cartagena, and fossil in the Tubera group of the Colombian Miocene, as at Loc. 267, C. A. S., horizon M - N, the lowest member of J50 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. the group. A careful comparison of the fossil and living examples shows the fossil form well within the range of vari- ation in the living shells. 97. Area (Scapharca?) veatchi Olsson Area veatchi Olsson, Bull. Am. Pal., vol. 9, 1922, p. 361, pi. 23, figs. 1-3; Gatun Stage, Miocene, Costa Rica. This species has been obtained from Loc. 267, C. A. S., horizon M - N of the Tubera group of the Colombian Mio- cene. The species appears to be nearly related to, though not identical with Area patricia Sowerby, as understood in this paper. 98. Area (Scapharea) medioamerieana (Olsson) Area medioamerieana Olsson, Bull. Am. Pal., vol. 9, 1922, p. 360, pi. 23, figs. 4-6; Miocene, Costa Rica. Olsson has described this species as a variety of Area golfoyaqnensis Maury, but the specific differences seem so evi- dent, both as to form and ornamentation, that it should be regarded as distinct. The species seems more closely related to Area actinophora Dall, while Maury's species seems to be nearer to Area dariensis Brown & Pilsbry. 99. Area (Scapharca) inequilateralis Guppy Area inequilateralis Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, p. 293, pi. 18, figs. 2, a, b; Miocene, Jamaica. Barhatia (Diluvarea) inequalateralis Woodring, Mioc. Moll. Bowden, Jam., Carnegie Inst. Publ. No. 366, 1925, p. 45, pi. 5, figs. 1-3; Miocene, Jamaica. This species has been obtained from Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena, from near the middle of the Tubera group, and from Loc. 299-A, C. A. S., between Cibarco and Chorrera, Tubera group, Colombian Miocene. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 15I 100. Area cacica Olsson Area cacica Olsson, Bull. Am. Pal., vol. 9, 1922, p. 362, pi, 24, fig. 1; Miocene, Costa Rica. This species occurs at Loc. 299, C. A. S., near Baranoa, Colombia, in the central part of the Tiibera group, Colombian Miocene. 101. Area (Scapharca) hispaniolana Maury Area (Scapharca) hispaniolana Maury, Bull. Am. Pal., vol. 5, 1917, p. 340, pi. 30, figs. 9, 10; Miocene, Santo Domingo. A single specimen of this species was obtained from each of the following localities : Loc. 304, C. A. S., four miles east of Santa Rosa; Loc. 306, C. A. S., near Usiacuri; and Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena, Colombia ; all of them in the central part of the Tubera group of the Miocene. 102. Area pittieri Dall Area pittieri Dall, Smiths. Misc. Coll., vol. 59, 1912, pt. 2, p. 9; Miocene, Costa Rica.— Olsson, Bull. Am. Pal., vol. 9, 1922, p. 364, pi. 24, figs. 2-6; Gatun Stage, Miocene, Costa Rica. This species has been obtained at Loc. 305, C. A. S., near Turbaco; Loc. 349, C. A. S., near Galapa; and Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena. The first two occurrences are at points low in the Tubera group, though the last is probably near the top. 103. Area (Seapharea) lloydi Olsson Area {Scapharca) lloydi Olsson, Bull. Am. Pal., vol. 9, 1922, p. 364, pi. 24 figs. 10-12; Gatun Stage, Miocene, Costa Rica. This Species was obtained at Loc. 323, C. A. S., at the Spill- way of the Canal, in 1914; and since then at Loc. 267, C. A. S., horizon P; and Loc. 306, C. A. S., Usiacuri; and also at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena; all from the central part of the Tubera group. 152 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Seb. 104. Glycymeris jamaicensis Dall Glycymeris jamaicensis Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1898, p. 608. — WooDRiNG, Carnegie Inst. Wash., Publ. No. 366, 1925, p. 24, pi. 2, figs. 1-3; Miocene, Bowden, Jamaica. This species has been found abundantly at Loc. 267, C. A. S., horizon M - N, and at Loc. 351, C. A. S., horizon P, both of the Tubera group of the Colombian Miocene. 105. Glycymeris carbasina (?) Brown & Pilsbry Glycymeris carbasina Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 363, pi. 28, fig. 9; Gatun formation. Canal Zone. This species has been doubtfully identified among the forms found in the lowest horizon of the Tubera group. It appears to be related to the preceding from the Bowden beds of Jamaica. 106. Glycymeris lloydsmithi Brown & Pilsbry Glycymeris lloydsmithi Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 69, 1917, p. 39, pi. 6, fig. 6; Miocene, near Cartagena, Colombia. Several good examples of this species were obtained from Loc. 351, C. A. S., near Punta Pua, 20 miles north of Carta- gena, near the middle of the Tubera group of the Colombian Miocene. 107. Glycymeris lamyi Dall Plate 22, figures 7, 8 Glycymeris lamyi D.\ll, Bull. U. S. Nat. Mus., No. 90, 1915, p. 122, pi. 20, figs. 11, 13; Tampa Silex beds, Tampa Bay, Fla., Lower Miocene. Glycymeris canalis, Olsson (in part, not Brown & Pilsbry), Bull. Am. Pal. vol. 9, 1922, p. 349, pi. 18, figs. 4, 5; Miocene, Costa Rica. Plesiotype: No. 4670, Mus. Calif. Acad. Sci., from Loc. 325-A, C. A. S., near Cibarco, Colombia; Miocene. Dall's description and figures are sufficiently clear to enable one to recognize the species with considerable confidence. He seems to have had, however, only the young or immature shells upon which to base his description. His figures are almost twice natural size. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA J 53 With further growth the number of primary ribs increases, and at the same time riblets appear on some of them. Super- ficially this species resembles G. trilobicosta Brown &: Pilsbry, but it is not only larger, but has a narrower, less expanded outline near the beaks, and intermediary riblets which are lacking in G. trilobicosta. Several good examples of this species were obtained at Loc. 351, C. A. S., and at Loc. 325, C. A. S., all in the central part of the Tubera group of the Colombian Miocene. 108. Glycymeris usiacurii Anderson, new species Plate 22, figures 3, 4 Shell small, sub-circular, moderately inflated; beaks small, median, a little prominent; primary ribs 15 in number, rounded, widest in the central part of the shell, separated by a groove containing a single intermediary riblet; ligamental area small, almost obsolete ; line of the cardinal teeth rounded, not angular, set with eight teeth on each side of the median line, with a few rudimentary teeth near the middle; height of holotype 24 mm., length 24 mm., depth of single valve 7 mm. This species outwardly resembles G. canalis Brown & Pils- bry, but unlike it has intermediary riblets, and not so many cardinal teeth in the hinge. This form has been found plentifully at Loc. 325, C. A. S., a mile east of the village of Usiacuri, and nearly 2,000 feet above the base of the Tubera group, of the Colombian Miocene. Holotype: No. 4668, Mus. Calif. Acad. Sci., from Loc. 325, C. A. S., horizon P, near the village of Usiacuri, Colombia; Miocene. 109. Ostrea haitensis Sowerby Ostrea haitensis Sow., Quart. Jour. Geol. Soc. Lond., vol. 6, 1850, p. 53. — Maury, Bull. Am. Pal., vol. 5, 1917, p. 346, pi. 31, figs. 1, 2; Zone D, Gurabo, Miocene, Santo Domingo. — Hodson, P., Bull. Am. Pal., vol. 13, 1927, p. 21, pi. 10, fig. 7, pi. 11, fig. 4, and pi. 12, fig. 4; Oligocene-Miocene, State of Falcon, western Venezuela. Ostrea vespertina (?), Jordan & Hertlein (not Conrad), Proc. Calif. Acad. Sci., vol. 15, 1926, p. 428; California Pliocene. March 29, 1929 154 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Sowerby's species has some marked features of resemblance to O. vespertina Conrad (=0. veatchi Gabb) from the upper Tertiary of the CaHfornia coast, though identity is not claimed. Ostrca gatunensis Brown & Pilsbry, and O. costari- censis Olsson apparently belong to the same group, and at least may be regarded as analogous, if not identical forms. Ostrea haitensis has been found at Loc. 266, C. A. S., San Juan Acosta Creek, horizon R, and Loc. 351, C. A. S., near Punta Pua. 110. Ostrea megadon Hanley Ostrea megadon Hanley, Proc. Zool. Soc. Lond., 1845, p. 106; living, west coast of Peru.— Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1898, p. 1586; Miocene, Santo Domingo, and Jamaica. — Maury, Bull. Am. Pal., Yol. 5, 1917, p. 347, pi. 34, fig. 3; Miocene, Santo Domingo. Ostrea cerrosensis Gabb, Geol. Surv. Calif., Pal. vol. 2, 1869, p. 35, pi. 11, fig. 61; Cedros Island, Pliocene. This species was found abundantly at Loc. 299, C. A. S., west of Usiacuri ; Loc. 306, C. A. S., three miles south of Baranoa; Loc. 347, C. A. S., near Turbaco; all of which are below the middle of the Tubera group of the Colombian Miocene. In this group of oysters should probably also be included Ostrea messor Maury from the Miocene of Trinidad.^" It is worthy of note that O. megadon, O. haitensis, and O. vespertina should be so often found associated in the same beds. The two former are found together in the lower Mio- cene of north Colombia, the first and last are found in the Pliocene beds of the California coast. Ostrea vespertina (= 0. veatchi Gabb) occurs in the Pliocene of Cedros Island and in contemporaneous beds in the Imperial valley, Cali- fornia, and is reported as still living in the Gulf of California. Ostrea megadon occurs with the preceding on Cedros Island and in Pliocene beds of Ventura county, and is found living at Turtle Bay, Lower California. »« Bull. Am. Pal. vol. 10, 1925, p. 233, pi. 10, figs. 3, 4. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 155 111. Pecten (Amusium) mortoni Ravenel Pecten mortoni Ravenel, Proc. Acad. Nat. Sci., Phila., vol. 2, 1844, p. 96; Miocene, South Carolina. — Tuomey & Holmes, Pliocene Foss, S. Carolina, 1857, p. 27, pi. 9, figs. 1, 2; pi. 10, figs. 1, 2. Pecten {Amusium) mortoni, Clark et al., Maryland Geol. Surv., 1904, p. 372, pi. 99, fig. 1; Miocene, Maryland. Brown & Pilsbry have described two species of Amusium from the Gatun formation of the Canal Zone, either one, or both of which may represent this species. The differences pointed out by these authors between P. mortoni Rav., and P. (Amusium) lima Brown & Pilsbry seem unimportant. Examples obtained from the Spillway of the Canal, 1914, and afterward from the Tubera group, horizon M - N, are very similar, though the Colombian forms agree better with the characters of P. mortoni than do those from Gatun. In our specimens the ears are not depressed below the plane of the valve. The external surface is smooth, or marked only by faint lines of growth, the diameter of the largest example is 143 mm., though larger specimens were seen. The angle of divergence in the dorso-lateral lines is near 123°-125°, vary- ing a little, as may be expected. The concentric growth lines nearly describe a circle, and the number of pairs of internal ribs is 22 to 24. The species is not rare in the Tubera group of Colombia. The best examples were found at Loc. 267,. C. A. S., horizon M - N, near the base of the Tubera group. It occurs also at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena. 112. Pecten (Plagioctenium) demiurgus Dall Pecten comparilis Guppy, Geol. Mag., vol. 1, 1874. (Not Tuomey & Holmes, 1855). Pecten {Plagioctenium) demiurgus Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1898, p. 718, pi. 26, fig. 3.— Maury, Bull. Am. Pal., vol. 10, 1925, p. 237, pi. 14, fig. 5; pi. 16, fig. 6; Miocene, Trinidad Island. According to Maury, the shell, when full grown, sometimes measures as much as 75 mm, in altitude, and a little more in width. Ours are not so large, though larger examples were seen at the locality from which they came. It is abundant at Loc. 267, C. A. S., horizon M - N, near the base of the 156 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Tubera group. The gravelly beds of this horizon did not permit the extraction of the larger specimens. The propor- tions maintain in all of them. 113. Pecten pinulatus Toula Pecten pinulatus Toula, Jahrb. der K. K. Geol. Reichs., vol. 61, 1911, p. 491 pi. 30, fig. 3; Miocene, Canal Zone. According to Toula's description and statement, the shell resembles that of Pecten cactaceus Dall, from the younger Tertiary of Tehuantepec. Our examples show a decided resemblance to Ball's species in surface ornamentation, although they are not so large. Two good examples were obtained from Loc. 267, C. A. S., horizon M - N, where it is not rare in the gravelly beds with the preceding. 114. Pecten atlanticola Anderson, new species Plate 19, figures 3, 7 Shell small, nearly circular, or slightly oblique, appressed, left valve a little more convex than the right ; ears long, sub- equal, the anterior right ear bearing six radial riblets, the others mostly smooth; radial ribs on the body of the shell 13 in number, rounded, with interspaces of nearly the same width as the ribs; ribs and interspaces crossed by distinct lines of growth; altitude of holotype 36 mm., length 40 mm., thickness 10.5 mm. Holotype: No. 4661 ; paratype: No. 4661-A, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon P, north slope of Tubera mountain, Colombia; Miocene. There is a strong resemblance, and evident relationship between this species and P. prcevalidus Jordan & Hertlein,^^ from the Pliocene of Turtle Bay, Lower California. Several good specimens of this species were obtained at Loc. 267, C. A. S., horizon P, north slope of Tubera mountain. As far as known this species belongs near the middle of the *> Proc. Calif. Acad. Sci., vol. 15, 1926, p. 435, pi. 29, figs. 2, 3. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA \c^y Tubera group of the Colombian Miocene, therefore near the Gatim horizon. 115. Pecten (Euvola) bowdenensis Dall Pecten {Euvola) bowdenensis Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1898, p. 713, pi. 29, fig. 1.— (?) BosE, Bol. Inst. Geol. Mex., No. 22, 1906, p. 27, pi. 1, figs. 8, 10. — WoODRiNG, Carnegie Inst. Wash., Publ. No. 266, 1925, p. 63, pi. 7, figs. 8, 9; Miocene, Bowden beds, Jamaica. A single example of this shell was obtained from Loc. 267, C. A. S., horizon P, on the north slope of Tubera mountain, from beds believed to be equivalent to the Gatun formation of the Canal Zone. 116. Pecten macloskeyi Anderson, new species Plate 19, figures 4, 5 Shell small, height of holotype 25.5 mm., length 24 mm., basal part circular, equivalve, beaks high, the borders forming an angle below 90 degrees; ears long, the anterior right ear bearing four corrugated riblets, the others nearly smooth ; sur- face ornamented by about 12 low, smoothly rounded ribs, with interspaces narrower than the ribs ; ribs on left valve very low, though not absent; all ribs more distinct on the younger shells. Holotype: No. 4662; paratype: No. 4663, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., north slope of Tubera mountain, Colombia; Miocene. This species is distinguishable from P. atlanticola by its smaller size, lower, more rounded ribs, narrower umbonal angle, and less circular outline. Several good examples of this shell were obtained at Loc. 267, C. A. S., associated with P. atlanticola, from which it is readily separated. As far as known both mark the middle of the Tubera group of the Colombian Miocene. It is named in honor of Mr. Downs McCloskey, whose active interest aided much in the study of the section and in the collections. 158 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Seb. 117, Spondylus bostrychites Guppy Spondylus bifrons Sowerby, Quart. Jour. Geol. Soc. Lond., vol. 6, 1850 (not of GoLDF. 1835); Miocene, S. Domingo. Spondylus bostrychites Guppy, Proc. Sci. Soc. Trinidad, 1867, p. 176. — Gabb, Trans. Am. Phil. Soc, vol. 15, 1873, p. 257.— Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1898, p. 758; 1903, p. 1586.— Bull. U. S. Nat. Mus., No. 90, 1915, p. 124, pi. 19, fig. 4; Silex beds, Fla.— Maury, Bull. Am. Pal., vol. 5, 1917, p. 354.— Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 73, 1921, p. 413; Miocene, Santo Domingo. A number of examples of this species were obtained at Loc. 267, C. A. S., horizon M - N, along with many other heavy shelled littoral forms, as shown elsewhere. 118. Spondylus gumanomocon Brown & Pilsbry Spondylus americanus Gabb, Trans. Am. Phil. Soc, vol. 15, 1873, p. 257 (not of Lamarck); Miocene, Santo Domingo. Spondylus gumanomocon Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 64, 1912, p. 514.— Maury, Bull. Am. Pal., vol. 5, 1917, p. 355.— Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 73, 1921, p. 413, pi. 43, figs. 4, 5; Miocene, Santo Domingo. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 379, pi. 21, fig. 1; Miocene, Costa Rica. Several examples of a Spondylus corresponding" very closely to this form were obtained at Loc. 267, C. A. S., horizon M - N, near the base of the Tubera group. They were associ- ated with the preceding form and other littoral species. The probability of their identity with the above species is very great. 119, Anomia mamillaris Anderson, new species Plate 16, figures 9, 10 Shell small, thin, smooth, translucent, circular in outline, convex; surface undulating, showing lines of growth, scaly near the umbones; umbone prominent, not quite central, inclin- ing forward; height of holotype 22 mm., length 23 mm., depth of single valve 8 mm, Holotype: No, 4165; paratypes: No, 4166 and 4167, Mus, Calif. Acad. Sci., from Loc. 267, C. A. S., from horizon M - N, Tubera group, Colombia; Miocene. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA 159 Several g^ood examples of this shell were obtained at Loc. 267, C. A. S., in the lowest horizon M - N, of the Tubera group. 120. Crassatellites berryi Spieker Crassatellites berryi Spieker, Johns Hopkins Univ. Publ. Geol., No. 3, 1922, p. 131, pi. 7, figs. 9, 10; Lower Zorritos, Peru. This species is abundant at Loc. 267, C. A. S., horizon R, Tubera village, north Colombia. As far as known it belongs only to this horizon, though its place in the Miocene of Peru is somewhat lower. 121. Crassatellites (Scambula) densus Dall Crassatellites (Scambula) densus Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1472, pi. 39, figs. 9-12; Oak Grove, Florida. This Species was found plentifully in the lowest horizon M - N, of the Tubera group at Loc. 267, C. A. S., near the western foot of Tubera mountain, and at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena. 122. Venericardia brassica Maury Venericardia lerryi, var. brassica Maury, Bull. Am. Pal., vol. 10, 1925, p. 323, pi. 30, fig. 5; Miocene, Trinidad. • Miss Maury has described this species as a variety of V. terryi Olsson, from the Miocene of Costa Rica, which it some- what resembles. In view of its larger size, more prominent ribs, exceeding those of the Costa Rican species, our samples are regarded as distinct from the latter, though identical with the Trinidad species. Three well-preserved specimens were found at Loc. 267, C. A. S., horizon M - N, of the Tubera group, Colom- bian Miocene. 150 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 123. Venericardia trinidadensis Maury Venericardia trinidadensis Maury, Bull. Am. Pal., vol. 10, 1925, p. 323, pi. 30 fig. 6; Miocene, Trinidad Island. A single valve of a venericard identifiable with the above was obtained at Loc. 305, near Turbaco, from a central hori- zon in the Tubera group. Its range is not known. 124. Cardita (Carditamera) arata (Conrad) Plate 20, figures 4, 5 Cypricardia arata Conrad, Foss. Sh. Ter. Form., 1832, p. 20, pi. 5, fig. 1; Miocene, North Carolina, etc. Cardita {Carditamera) arata, Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1413.— Maury, Monog. Serv. Geol. e Min. Brazil, 1925, p. 271, pi. 15, fig. 15; Miocene, Para, Brazil. The shell is of moderate size, length 31 mm., height 18 mm., thickness 16 mm.; elongated subquadrate, rounded before, and somewhat truncated behind; beaks near anterior end but not terminal, strongly incurved and proximate ; dorsal margin straight, ventral margin slightly arcuate; ribs 15 m number, with a tendency to become scaly, or even beaded, showing wavy lines of growth. Plesiotype: No. 4164, Mus. Calif. Acad. Sci., from Loc. 267-B, C. A. S., horizon M - N, Tubera group, Colombia; Miocene. This description is here introduced in support of the identi- fication of Conrad's species in the Miocene of north Colombia. Maury has stated that the species is found in the Chipola marls, associated with C. vaughani Dall, and in the lower Mio- cene of Para, Brazil, there is a very similar form. C. arata is said to be a widely distributed and abundant form, to which C. floridana Conrad, from the Pliocene of Florida is regarded as a successor. Several examples of this species were found at Loc. 267, C. A. S., horizon M - N, of the Tubera group of the Colom- bian Miocene. A comparison with samples of Conrad's species from Florida shows the only essential difference to be in the slightly more beaded ornamentation of the ribs in the more recent form. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA \(y\ 125. Cardita (Glans) scabricostata Guppy Cardita scabricostata Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 293, pi. 18, fig. 10; Miocene, Jamaica. Venericardia scabricostata, Dall (part), Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1428.— Maury, Bull. Am. Pal., vol. 5, 1917, p. 362, pi. iZ, fig. 1; Miocene, Santo Domingo. — Woodring, Carnegie Inst. Wash., Publ. No. 266, 1925, p. 99, pi. 12, figs. 7-9; Miocene, Jamaica. Although Dr. Woodring does not include Maury's form as coming within the range of Guppy's species, it appears that it should not be regarded as a distinct form, and that it should have at least a varietal rank there. We have several good examples from five different localities, all of which approach the form figured by Maury, more nearly than that of Woodring. It occurs at Loc. 267, C. A. S., horizons M - N, P, and R, Tubera mountain ; Loc. 306, near Usiacuri ; Loc. 355, Murindo creek; and it was obtained at Loc. 323, C. A. S., at the Spillway of the Canal in 1914. 126. Echinochama antiquata Dall Chama arcinella, Guppy, Geol. Mag., vol. 1, 1874, p. 450 (not of Linnaeus); Miocene, Bowden, Jamaica, and Santo Domingo. Echinochama antiquata Dall, Trans. Wag. Fr. Inst. Sci., vol. 3 1903, p. 1404, pi. 54, fig. 9.— Olsson, Bull. Am. Pal., vol. 9, 1922, p. 390, pi. 28, fig. 8; Miocene, Costa Rica. This species occurs abundantly at Loc. 267, C. A. S., hori- zon M - N, near the base of the Tubera group, and at Loc. 351, C. A. S., in the middle part of the group, near Punta Pua, 20 miles north of Cartagena. 127. Chama scheibei Anderson, new species Plate 22, figures 1, 2 Shell of moderate size, very inequal valves; height of holo- type 43 mm., length 37 mm.; left valve inflated, right valve nearly flat ; left valve with strongly recurved beak, right valve with smaller beak, less recurved ; surface bearing only obsolete spines, if any, and only on the posterior part of left valve ; right valve ornamented with wavy lamellae following lines of growth; anterior part and umbone of left valve somewhat 1^2 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th See. beaded. A faint depression extends from the beak near and parallel to the anterior margin. A number of samples of this species were found at Loc. 267, C. A. S., horizon M - N, near the base of the Tubera group, Colombian Miocene. Named in honor of the late Dr. Robert Scheibe of the Comicion Cientifica Nacional, Bogota. Holotype: No. 4667, Mus. Calif. Acad. Sci., from Loc. 267-B, C. A. S., horizon M - N, Tubera mountain, Colombia ; Miocene. 128. Thyasira bisecta ( ?) (Conrad) Plate 21, figure 1 Venus bisecta Conrad, Geol. U. S. Expl. Expd., 1849, p. 724, pi. 17, figs. 10, 10a; Miocene, Astoria, Oregon. Cyprina bisecta Conrad, Am. Jour. Conch., vol. 1, 1865, p. 153; locality as above. Cryptodon bisecta, Dall, Proc. U. S. Nat. Mus., vol. 17, 1895, p. 713, pi. 26. figs. 2, 5; living, Alaskan coast and southward. Thyasira bisecta, Dall, Prof. Ppr. U. S. Geol. Surv., No. 59, 1909, p. 118; Miocene, Astoria, Oregon. According to Dall this species is found living on the Alaskan coast, in Puget Sound, and occurs in the Miocene of Oregon and perhaps of California. As no reference to its occurrence in the Caribbean region has been found, it seems well to record it here, even though doubtfully recognized. The species was found by K. D. White at Loc. 350, C. A. S., near Arboletes Bay in the upper Miocene beds of the Colom- bian coast. Plesioiype: No. 4664, Mus. Calif. Acad. Sci., from loc. 350, C. A. S., Canalete Point, north coast of Colombia; Miocene. 129. Diplodonta woodringi Anderson, new species Plate 22, figures 5, 6, Shell small, circular in outline, suborbicular, moderately inflated in the umbonal area ; anterior end more abruptly slop- ing than the rounded posterior; height of holotype 26 mm., Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA ^53 length 25 mm., thickness 18 mm. ; beaks somewhat central, recurved, prominent ; lunule only faintly marked. Holotype: No. 4669, Mus. Calif. Acad. Sci., from Loc. 325-A, C. A. S., near Cibarco, Colombia; Tubera group, Miocene. Two or three samples of this species were obtained, one from Loc. 325, C. A. S., and the other, the holotype, from Loc. 325-A, C. A. S., near Cibarco, about horizon P of the Tubera group, not common. This species is named in honor of Wendell P. Woodring, whose work in the Caribbean Miocene and later formations is deserving of highest praise. 130. Erycina turbacoensis Anderson, new species Plate 22, figures 9, 10 Shell large, oval, depressed; length of holotype (incom- plete) 46 mm., height 35 mm., thickness 12 mm.; length of paratype (cast) 59 mm., height 45 mm.; beaks subcentral, a little nearer the posterior end, low, curved forward; lunular area small, impressed ; anterior dorsal margin nearly straight, anterior end produced, posterior shorter, rounded ; surface smooth, ornamented only by indistinct lines of growth. The hinge on the right valve of paratype is distinct, showing nor- mal character of Erycina. In form and general characters this species resembles Erycina fabulina Dall, from the Oak Grove Miocene, but it is many times larger. The figure of Semele sayi Toula^^ resem- bles this species somewhat, but seems to have a more decided concentric sculpture. This species was found at Loc. 305, C. A. S., near Turbaco, Colombia, in the lower part of the Tubera group. Holotype: No. 4671 ; paratype: No. 4672, Mus. Calif. Acad. Sci., from Loc. 305, C. A. S., near Turbaco, Colombia, in the lower part of the Tubera group ; Miocene. •2Jahrb. der K. K. Geol. Reichs., 1909, Bd. 58, pi. 28. 154 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 131. Cardium (Trachycardium) dominicense Gabb Cardium (Trachycardium) dominicense Gabb, Trans. Am. Phil. Soc, vol. 15, 1873, p. 250.— Gabb, Jour. Acad. Nat. Sci. Phila., vol. 8, 1874, p. 344; Miocene, Costa Rica. — Pilsbry & Brown, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 367; Gatun formation. Canal Zone. — Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 73, 1921, p. 421, pi. 25, figs. 8, 9; Miocene, Santo Domingo. A single good example of this shell was found by K. D. White in the Miocene beds of the Rio Canalete, near the mouth of the Quebrada Murindo, in the district of Arboletes Bay, Colombia. 132. Cardium (Trachycardium) puebloense Anderson, new species Plate 19, figures 1, 2 Shell of medium size, subquadrate, thick, equilateral, surface somewhat enamelled; length of holotype 40 mm., height 44 mm., thickness 36 mm. ; umbones high and prominent, only slightly angulated behind ; ribs 30 to 34 in number, nearly smooth, though showing lines of growth; margins smooth, denticulate within, the posterior margin slightly serrate. A peculiarity of the sculpture is the linear division of the rounded ribs, separated by V-shaped interspaces; the anterior 18 or 20 ribs are sometimes divided longitudinally by an elevated thread, the posterior 12 or 14 are so divided by a groove of equal strength ; in either case the ribs are marked by V-shaped incremental lines. These lines are apparent even on very young shells. This species appears to be related to C. linguu- leonis of the Jamaican Miocene, as illustrated by Woodring. Holotype: No. 4660, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon R, at the village of Tubera, Colombia; Miocene. The holotype was found at Loc. 267, C. A. S., horizon R, at the village of Tubera. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA I55 133. Cardium (Trachycardium) lingualeonis Guppy Cardium lingualeonis Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 293, pi. 18, fig. 7; Miocene, Jamaica. — Guppy, Geol. Mag., vol. 1, 1874, p. 422; (Not Guppy, vol. 32, 1876, p. 531). Cardium {Trachycardium) lingualeonis, Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1900, p. 1084; Miocene, Chipola river, Florida. — Woodring, Carnegie Inst. Wash., Publ. No. 366, 1925, p. 136, pi. 18, figs. 12, 13; Miocene, Bowden, Jamaica. This species occurs abundantly in the Tubera group, having been obtained at the following places : Loc. 267, C. A. S., horizon M - N; Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena ; in the latter of these places it occurs near the middle of the Tubera group. 134. Cardium (Laevicardium) gorgasi Hanna Cardium {Lcevicardium) dalli TouLA, Jahrb. der K. K. Geol. Reichs., Bd. 58, 1908, p. 722, pi. 27, fig. 6; Gatun formation, Miocene. — Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 367; (not C. dalli Heilprin, 1887). Cardium gorgasi Hanna, Proc. Calif. Acad. Sci., vol. 13, 1924, p. 160; new name proposed for the species. Two examples of this species, measuring respectively 53 mm. and 45 mm. in height, were obtained at Loc. 267, C. A. S., horizon M - N, near the base of Tubera group, and another from Loc. 266, C. A. S., near the top of the same group. It differs from C. ( Lcevicardiuni) serratum Linnaeus in both form and ornamentation, is larger and a thinner form in which radial ribbing is present, though not prominent; while in the living form the radial markings are faint. In the fossil form the dorsal margin is elevated into a sharp ridge, slightly arched near the hinge, and the posterior end is produced and narrowed, while the living form is here dis- tinctly rounded. 135. Cardium (Laevicardium) serratum Linnseus Cardium serratum LiNN^us, Syst. Nat. 1758, ed. 19, p. 680. Cardium (LcBvi cardium) serratum, Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1900, p. 1110; Miocene, Bowden, Jamaica. — Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 367; Gatun formation. Canal Zone. — Woodring, Carnegie Inst. Wash., Publ. No. 366, 1925, p. 145, pi. 19, figs. 14 to 16; Bowden, Jamaica. 156 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. This species has been found at Loc. 305, C. A. S., near Turbaco, and at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena, north coast of Colombia, near middle of the Tubera group. The species is still living in the Caribbean waters, and was collected in the Bay of Cartagena and neigh- boring points in 1914. 136. Cardium (Laevicardium) venustum Gabb Cardium venustum Gabb, Trans. Am. Phil. Soc, vol. 15, 1873, p. 251; Miocene, Santo Domingo. — Maury, Bull. Am, Pal., vol. 5, 1917, p. 213, pi. 36, fig. 9; as above. — Pilsbry, Proc. Acad. Nat. Sci., vol. 73, 1921, p. 421, pi. 25, figs. 2, 7; Miocene, Santo Domingo. A good example of this shell was obtained at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena, near the middle of the Tubera group. 137. Dosinia delicatissima Brown & Pilsbry Dosinia delicatissima Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 64, 1912, p. 516, pi. 26, fig. 1; Miocene, Gatun formation. Dosinia (Artemis) acetabulum (Conrad), Toula (?), Jahrb. der K. K. Geol. Reichs., Bd. 58, 1908, p. 727, pi. 27, figs. 8, 8a. Examples of this species were obtained at the Spillway of the Canal in 1914, and subsequently at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena. They are indis- tinguishable, and seem to conform satisfactorily to the figure and description of the species given by Brown & Pilsbry. 138. Dosinia (Artemis) acetabulum (?) (Conrad) Artemis acetabulum Conrad, Foss. Sh. Tert. Format., 1833, p. 20, pi. 6, fig. 1; Miocene, Maryland. Dosinia acetabulum Conrad, Foss. Med. Tert., 1838, p. 29, pi. 16, fig. 1. — Whitf., Monog. U. S. Geol. Surv., No. 24, 1894, p. 73, pi. 13, fig. 2.— Olsson, Bull. Am. Pal., vol. 9, 1922, p. 403; Miocene, Costa Rica. Dosinia {Artemis) acetabulum (Conrad), Toula, Jahrb. der K. K. Geol. Reichs., Bd. 58, 1908, p. 727, pi. 27, figs. 8, 8a; Gatun formation, Canal Zone, Panama. A fossil Species probably referable to the above was obtained at Loc. 267, C. A. S., horizon P, near the middle of the Tubera group of the Colombian Miocene. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA ^57 139. Clementia (Clementia) dariena (Conrad) Meretrix dariena Conrad, House Doc. 129, 1855, p. 18; Miocene, Isthmus of Panama. — Pac. R. R. Repts., vol. 5, 1856, p. 328, pi. 6, fig. 55; occurrence as above. Clementia dariena, Gabb, Jour. Acad. Nat. Sci. Phila., vol. 8, 1881, p. 344, pi. 44, figs. 16, 16a; Miocene, Santo Domingo. — Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1235, Sapote, Costa Rica. — Toula, Jahrb. der K. K. Geol. Reichs., vol. 58, 1908, pp. 725-727, pi. 27, figs. 9, 10; Gattm formation, Canal Zone. — Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 371, pi. 28, fig. 1.— Olsson, Bull. Am. Pal., vol. 9, 1922, p. 404; Miocene, Costa Rica. — Woodring, Prof. Ppr. U. S. Geol. Surv., No. 147-C, p. 34. Good examples of this species were obtained at the Spillway of the Canal in 1914 and it has since been collected at many localities in north Colombia, as at Loc. 267, C. A. S., horizons M-N and P; Loc. 305, C. A. S., near Turbaco; Loc. 302, C. A. S., four miles south of San Andres; Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena. 140. Cyclinella gatunensis Dall Cyclinella gatunensis Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1285, pi. 52, fig. 18; Miocene, Gatun, Panama. Several good samples of this species were obtained from Loc. 323, C. A. S., at the Spillway of the Canal in 1914, and it has since been found at various places in north Colombia. It occurs at Loc. 267, C. A. S., horizon R, Tubera village; Loc. 302, C. A. S., four miles south of San Andres, Dept. of Bolivar; upper horizon of the Miocene. As it has not hitherto been reported outside of the type locality its discovery in the Tubera group is interesting. 141. Cyclinella cyclica domingensis Pilsbry & Johnson Dosinia cyclica Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 582, pi. 26, figs. 15a, b; Miocene, Trinidad. — Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1285; probably Santo Domingo Miocene. Cyclinella cyclica domingensis Pilsbry & Johnson, Proc. Acad. Nat. Sci., Phila., vol. 69, 1917, p. 200; Miocene, Santo Domingo. — Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 73, 1921, p. 424, pi. 47, fig. 3; as above. 158 CALIFORNIA ACADEMY OF SCIENCES [Phoc. 4th Ser. Three examples of this species were obtained at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena, near the middle of the Tubera group, Colombian Miocene. 142. Antigona (Ventricola) blandiana (Guppy) Venus blandiana Guppy, Proc. Sci. Soc. Trinidad, vol. 3, 1873, pp. 85-86, pi. 2, fig. 8.— Geol. Mag., vol. 1, 1874, p. 436, pi. 17, fig. 8; Mio- cene, Trinidad. Antigona {Ventricola) blandiana, Woodring, Carnegie Inst. Wash., Publ. No. 366, 1925, p. 157, pi. 21, figs. 5-9; Miocene, Bowden, Jamaica. This species has been found in the Tubera group at Loc. 267, C. A. S., horizons M-N and P, and at Loc. 351, C. A. S., near Punta Pua, north of Cartagena. The species is closely related to Antigona fordi Yates,^^ now living on the Pacific coast from Monterey Bay to Panama (Dall). 143. Antigona caribbeana Anderson Antigona caribbeana Anderson, Free. Calif. Acad. Sci., vol. 16, 1927, p. 90, pis. 2 and 3; Loc. 267, Horizon M - N, Tuberd group of Colombian Miocene. This is perhaps the largest representative of the genus yet found in the Caribbean Tertiary deposits. It has commonly been regarded as the Miocene form of Antigona tnulticosta (Sowerby), but upon a careful comparison it can be easily dis- tinguished by various characters, among which are the crenu- lations on the inner margin of the shell. It occurs plentifully in the basal beds of the Tubera group. 144. Callocardia (Agriopoma) gatunensis Dall Callocardia {Agriopoma) gatunensis Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1260, pi. 54, figs. 1, 15; Gatun formation, Panama. — Brown & PiLSBRY, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 370; occur- rence as above. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 407, pi. 32, fig. 1; Miocene, Costa Rica. — M.\ury, Bull. Am. Pal., vol. 10, 1925, p. 298, pi. 27, figs. 5, 7; Miocene, Trinidad Island. This species has been obtained at Loc. 267, C. A. S., hori- zon P, on the north slope of Tubera mountain, and at Loc. « Yates, Santa Barbara Nat. Hist. Soc. Bull. 2, p. 46. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA I59 351, C. A, S., near Punta Pua, 20 miles north of Cartagena. It occurs, therefore, near the middle of the Tubera group. 145. Pitaria (Lamelliconcha) circinata (Born) Venus circinata Born, Test. Mus. Caes. Vind., 1778, p. 61, pi. 4, fig. 8; living in Caribbean waters. Chione circinata, Gabb, Trans. Am. Phil. Soc, vol. 15, 1873, p. 250; Miocene, Santo Domingo. Pitaria (Lamelliconcha) circinata, Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1269; Gatun formation; Cumana, Venezuela, etc. — Maury, Bull. Am. Pal., vol. 5, 1917, p. 379, pi. 37, fig. 1; Miocene, Santo Domingo. —Bull. Am. Pal., vol. 10, 1925, p. 301, pi. 27, figs. 12, 13; Mio- cene, Trinidad Island. Pilar circinata, Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 370; Gatun formation. Canal Zone. Pitaria circinata, Olsson, Bull. Am. Pal., vol. 9, 1922, p. 408, pi. 31, figs. 3, 9; Miocene, Costa Rica. Numerous examples of this species were obtained from the Bay of Cartagena in 1914, and it has since been collected at Loc. 267, C. A. S., horizon R, at Tubera village, and from Pliocene beds on the Caribbean coast of Colombia. 146. Pitaria cercadica Maury Pitaria cercadica Maury, Bull. Am. Pal., vol. 5, 1917, p. 380, pi. 37, fig. 10; Miocene, Santo Domingo. This species has been obtained at Loc. 267, C. A. S., hori- zon M - N, and horizon R, of the Tubera group, and should be found also in intervening strata. It is believed to be closely related to Pitaria alhida Gray ( ?), now living in the Bay of Cartagena. 147. Pitaria acutecostata (Gabb) Callista acutecostata Gabb, Trans. Am. Phil. Soc, vol. 15, 1873, p. 250, Miocene. Pitaria acuticostata, Maury, Bull. Am. Pal., vol. 5, 1917, p. 380, pi. 37, fig. 2; Miocene, Santo Domingo. Pilar (Lamelliconcha) acuticostatus Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 73, 1921, p. 422, pi. 47, fig. 10; occurrence as above. March 29, 1929 170 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. This species is found at Loc. 267, C. A. S., horizon R, Tubera villag-e. In size and form it approaches very near to P. affinis Sowerby, now living in neighboring waters. 148. Tivela mactroides (Born) Venus mactroides Born, Test. Mus. Caes. Vind., 1778. Cytherea mactroides, Reeve, Conch. Icon., 1863, pi. 5, figs. 18, a, b, c; living fauna, Caribbean region. Tivela mactroides, Dall, Proc. U. S. Nat. Mus., vol. 26, 1902, p. 367; occur- rence as above. — Maury, Bull. Am. Pal., vol. 10, 1925, p. 295, pi. 26, fig. 8; pi. 27, fig. 3; Miocene, Trinidad. Numerous examples of this shell were obtained from the Bay of Cartagena and near by points in 1914, and it has since been found fossil at Loc. 325-A, C. A. S., near Cibarco, a little above the middle of the Tubera group. Comparison with the living form shows no essential difference in the fossil. 149. Macrocallista (Chionella) maculata (Linnaeus) Venus maculata Linn^us, Syst. Nat. 1758, ed. 10, p. 680; living. Macrocallista (Chionella) maculata, Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1256; Chipola beds, Florida. — Olsson, Bull. Am. Pal., vol. 9, 1922, p. 406, pi. 31, figs. 6, 7; Miocene, Costa Rica. — Maury, Bull. Am. Pal., vol. 10, 1925, p. 279, pi. 25, figs. 1, 4, 5; upper Miocene, Trinidad. This species is found living in the Bay of Cartagena, and other Caribbean waters, and was found fossil at Loc. 267, C. A. S., horizon M - N, and at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena, in the lower and central parts of the Tubera group. 150. Chione (Chamelea) nuciformis (Heilprin) Cytherea nuciformis Heilprin, Trans. Wag. Fr. Inst. Sci., vol. 1, 1887, p. 116, pi. 16, fig. 61; Pliocene, Florida. Chione (Chamelea) nuciformis, Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1300; Miocene, Tampa Bay, Florida. This species has been obtained at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena, near the middle of the Tubera group of the Colombian Miocene. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA \y\ 151. Chi one (Chione) walli Guppy Venus walli Guppy, Quart. Jour. Geol. Soc. Lond., vol. 22, 1866, p. 581, pi. 26, fig. 16; Miocene, Trinidad. — Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, pp. 1291, 1587; Miocene, Trinidad, Bowden, Jamaica. — Spieker, Pal. Zorritos Format., Johns Hopkins Univ. Publ., Geol., No. 3, pp. 151, 154; Miocene, Peru. Chione {Chione) walli, Maury, Bull. Am. Pal., vol. 10, 1925, p. 311, pi. 28, figs. 2, 11, 15; Miocene, Trinidad. A species of Chione, probably referable to the above, was found at Loc. 267, C. A. S., horizons M - N and R of the Tubera group. Its occurrence at both the bottom and top of the group makes it hkely that it will be found also at inter- vening horizons. 152. Chione (Lirophora) mactropsis (Conrad) Gratelupia (?) mactropsis Conrad, House Doc. 129, 1855, p. 18; Isthmus of Panama. — Pac. R. R. Repts., vol. 5, 1856, p. 328, pi. 6, fig. 54; Miocene, Isthmus of Panama. Chione (Lirophora) mactropsis, Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1294; Gatun formation, Panama. Chione mactropsis, Olsson, Bull. Am. Pal., vol. 9, 1922, p. 417, pi. 30, figs. 7, 8; Gatun formation, Canal Zone. This species occurs abundantly in the Miocene at Gatun, and at Loc. 267, C. A. S., horizon M - N, Tubera group, and at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena. 153. Chione (Lirophora) latilirata (Conrad) Venus latilirata Conrad, Proc. Acad. Nat. Sci. Phila., vol. 1, 1841, p. 28. — Conrad, Foss. Sh. Med. Tert., 1845, p. 68, pi. 38, fig. 3; Miocene. Chione (Lirophora) latilirata. Meek, Checkl. Mio. Foss. Am., 1864, pp. 9, 30. — Dall, Trans. Wag. Fr. Inst. Sci., vol. 3, 1903, p. 1298, pi. 42, fig. 3; Miocene. Chione (Lirophora) cartagenensis F. & H. K. Hodson, Bull. Am. Pal., vol. 13, p. 63, pi. 31, fig. 4; pi. 35, fig. 6; Miocene, Colombia. This species occurs at Loc. 267, C. A. S., horizon M - N of the Tubera group, and at Loc. 351, C. A. S., near Punta Pua, 20 miles north of Cartagena. 172 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 154. Chione atlanticana Anderson, new species Plate 23, figures 5, 6 Shell of moderate size, subtriangular in outline; length of holotype 61 mm., height 51 mm., thickness 41 mm.; dorsal margin nearly straight, ventral margin broadly rounded, posterior angulated; beaks prominent; anterior slope short, projecting, forming angle with the ventral border; surface ornamented by raised concentric lamellae, fluted on the ventral side as in Chione guppyana Gabb, as described by Pilsbry.^* The lunule is relatively large and bordered by a sharply defined groove; escutcheon moderately wide, bordered by ridges ; inner border of shell finely crenulated. Holotype: No. 4676, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon P, Tubera mountain, Colombia; Miocene. This species is nearly related to Chione guppyana, but it differs from Gabb's species in being more nearly triangular in outline, straighter on the dorsal border, more prominent in front, and in lacking concentric lamellae along the ventral margin. This species has been obtained at Loc. 267, C. A. S., hori- zon P, where it was associated with Pilar ia circinata, Anti- gona caribbeana, and dementia dariena. 155. Tellina costaricana Olsson Tellina costaricana Olsson, Bull. Am. Pal., vol. 9, 1922, p. 423, pi. 26, figs. 6, 9; Gatun Stage, Costa Rica. This species is abundant at Loc. 267, C. A. S., horizons P and R, and also in the basal horizon M - N, of the Tubera group of the Colombian Miocene. 156. Tellina dariena Conrad Tellina dariena Conrad, House Doc, 129, 1855, p, 18. — Conrad, Pac. R. R. Repts., vol. 5, 1856, p. 328, pi. 6, fig. 53; Isthmus of Darien, Mio- cene. — Gabb, Jour. Acad. Nat. Sci. Phila., vol. 8, 1881, p. 343, pi. 44, fig. 13. — Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol, 63, 1911, p. 368.— Olsson, Bull. Am. Pal., vol. 9, 1922, p. 424, pi. 26, fig. 3; Gatun, Canal Zone. »*Proc. Acad. Nat. Sci. Phila. vol. 73, 1921, p. 423. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA l^-^ Tellina rowlandi Toula, Jahrb. der K. K. Geol. Reichs., Bd. 58, 1908, p. 728, pi. 28, fig. 11; Gatun, Canal Zone. This species has been found at Loc. 304, C. A. S., near Santa Rosa, and at Loc. 351, C. A. S., near Piinta Pua, 20 miles north of Cartagena. 157. Tellina gatunensis (Toula) Macoma {Tellina) gatunensis Toula, Jahrb. der K. K. Geol. Reichs., Bd. 58, 1908, p. 729, text figure 10, a; Gatun, Canal Zone. Tellina gatunensis. Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 63, 1911, p. 368; Gatun formation. Canal Zone. Macoma gatunensis, Olsson, Bull. Am. Pal., vol. 9, 1922, p. 429; Gatun Stage, Costa Rica. Several good examples of this species have been obtained from various localities in north Colombia, as at Loc. 267, C. A. S., horizons M-N and P; Loc. 303, C. A. S., about three miles north of San Andres, Dept. of Bolivar, etc. One of these examples exposes the hinge clearly, showing that it is a typical Tellina of the group T. radiata Linnaeus, found in the West Indies. The occurrence of this species with many others of the Tubera group at San Andres is to be specially noted. 158. Tellina (Eurytellina) aequiterminata (?) Brown & Pilsbry Plate 23, figure 4 Tellina cBquiterminata Brown & Pilsbry, Proc. Acad. Nat. Sci. Phila., vol. 64, 1912, p. 517, pi. 26, fig. 5; Gatun formation, Canal Zone. A rather large Tellina was found at Loc. 304, C. A. S., four miles east of Santa Rosa, which in outline and general characters conforms to the above species, though in size it agrees more nearly with T. radiata Linnaeus. The left valve is somewhat more concave in longisection than in T. radiata, and the sculpture is different. The surface is marked by undu- lations and finer concentric lines, which at the posterior end become lamellar. The growth lines form an obtuse angle on crossing the posterior angle of the shell. Approximate length 60 mm., height 35 mm., thickness 11 mm. 174 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Plesiotype: No. 4675, Mus. Calif. Acad, Sci., from Loc. 304, C. A. S., horizon P, four miles east of Santa Rosa, Colombia ; Miocene. 159. Tellina (Eurytellina) aequicincta Spieker Tellina (Eurytellina) cequicincta Spieker, Paleont. Zorritos Form., Peru; Johns Hopkins Univ. Publ. Geol., No. 3, 1922, p. 158, pi. 10, fig. 3; Zorritos group, Miocene, Peru. Two specimens of a Tellina were obtained at the village of Tubera, Loc. 267, C. A. S., horizon R, which seem to be refer- able to this species. In form and sculpture the resemblance is striking, and there appears to be no reason for doubting their identity. 160. Tellina (Eurytellina) cibaoica (?) Maury Tellina {Eurytellina) cibaoica Maury, Bull. Am. Pal., vol. 5, 1917, p. 387. pi. 38, fig. 10; Zone H, Rio Cana, Santo Domingo. A single specimen of Tellina was found at Loc. 304, C. A. S., east of Santa Rosa, that conforms to Maury's description and figure of this Dominican form. It seems to be related to Tellina striata Chemnitz, from the West Indian province. 161. Tellina protolyra Anderson, new species Plate 21, figures 2, 3 Shell small, height of holotype 25 mm., length 34 mm., thickness 12 mm., partly elliptical, truncated behind, rounded in front, more broadly rounded on the ventral margin ; peaks posterior to a central position, high, pointing forward, excavated in front forming a sort of lunule-like depression; inequivalve, the right valve being flatter and slightly concave in advance of the umbonal angle; posterior dorsal margin nearly straight, formed by a narrow carina-like ridge on either side, giving the posterior dorsal slope a groove-like character; surface ornamented by acute, elevated, concentric threads with relatively wide, concavely open interspaces, Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA J 75 almost smooth, or faintly striated, and evenly spaced from beak to ventral margin. This species is clearly related to Tellina lyra Hanley which is found living at Tumbez, Peru, which is probably a successor to our species. The examples of this species were all found at Loc. 267, C. A. S., horizon M - N, of the Tubera group of the Colombian Miocene. Holotype: No. 4163, Mus. Calif. Acad. Sci., from Loc. 267-B, C. A. S., horizon M - N, of the Tubera group, Colombia; Miocene. 162, Semele claytoni ( ?) Maury Semele claytoni Maury, Bull. Am. Pal., vol. 5, 1917, p. 391, pi. 35, fig. 9; Miocene, Cercado de Mao, Santo Domingo. A single specimen of Semele that seems referable to this Dominican species was found at Loc. 351, C. A. S., near Punta Pua, Colombia. 163. Semele sardonica Dall Semele sardonica Dall, Bull. U. S. Nat. Mus., No. 90, p. 154, pi. 20, figs. 4 and 7; Miocene, Tampa Bay, Florida. A single well preserved valve of a Semele was obtained at Loc. 351, C. A. S.. near Punta Pua, Colombia, that is identi- fiable with Ball's species from the lower Miocene of Florida. 164. Psammosolen sancti-dominici Maury Psammosolen sancti-dominici Maury, Bull. Am. Pal., vol. 5, 1917, p. 392, pi. 37, fig. 13; Miocene, Cercado de Mao, Santo Domingo. A single determinable specimen of Psammosolen was obtained at Loc. 351, C. A. S., near Punta Pua, that seems to be referable to Maury's Dominican species. 165. Mactra (Mulinia ?) atlanticola Anderson, new species Plate 20, figures 1,2, 3 Shell of moderate size, length of holotype 50 mm., height 43 mm., thickness 33 mm. ; robust, ventricose, smooth. J76 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. ornamented only by concentric growth lines; beaks rather high, nearly central or a little in advance of central, curved slightly forward ; anterior and posterior slopes straight, anterior end broadly rounded, posterior end more narrowly rounded; lunular area flattened, or somewhat concave under the beaks ; shell not gaping behind, not angulated, but for the most part regularly rounded. Holotype: No. 4161 ; paratype: No. 4162, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon M - N, of the Tubera group, at the west end of Tubera mountain, Colombia; Miocene. The nearest known related species is Mulinia densata Con- rad, in the upper Miocene of California, although it has a heavier and more solid shell than the Colombian examples here described. Several good specimens of this species were found at Loc. 267, C. A. S., horizon M - N, of the Tubera group, at the west foot of Tubera mountain. There is an outward resem- blance to other Caribbean forms, but the hinge reveals its gen- eric class. 166. Mactrella (Harvella) elegans (Sowerby) Plate 21, figures 5, 6 Mactra elegans Sowerby, Tank'v. Catal. Append. (116), p. ii, pi. (i), fig. 3; living at Panama and Pacific points. — Carpenter, Rept. Brit. Ass'n. Adv. Sci., 1857, pp. 174, 227; living at Panama and other points. Harvella pacifica Conrad, Amer. Jour. Conch., vol. 3, 1867, p. 192; vol. 5, p. 108, pi. 12, fig. 2; living at Panama. Mactrella {Harvella) elegans, Dall, Nautilus, vol. 8, 1894. Conrad described H. pacifica as living at Panama, and attempted to distinguish his supposed new form from H. elegants (Sowerby) to which he refers as a Floridan species. Dall discredits Conrad's name, on the ground of lack- ing sufficient basis, at least until further evidence was found. Although Sowerby's original description has not been seen, in view of the known variability in such forms, it appears unlikely that Conrad's discrimination is sound. Two species so similar are not likely to occur together. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA YJJ A comparison of the fossil species with representatives of the living form does not permit of any distinction that can be maintained in either form, size or sculpture. A number of good samples of this species was found at Log. 267, C. A. S., horizon M - N, of the Tubera group, at the west foot of Tubera mountain. Plesiotypes: Nos. 4665 and 4666, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon M - N, of the Tubera group, at the west foot of Tubera mountain, Colombia; Miocene. 167. Labiosa (Raeta) gibbosa (Gabb) RcEta gibbosa Gabb, Amer. Jour. Conch., vol. 5, 1870, p. 30; Miocene, Peru. — • Gabb, Jour. Acad. Nat. Sci. Phila., vol. 8, 1874, p. 264, pi. 35,' figs. 8, 8a. Two well preserved samples of this species were obtained at Loc. 267, C. A. S., horizon R, Tubera village, near the top of the Tubera group, and three of the same form were found at Loc. 351, C. A. S., near Punta Pua, some 20 miles north of Cartagena. In referring these to Gabb's Peruvian species iden- tification is based entirely upon his description and figures, as no comparative material was available from his locality. 168. Labiosa (Raeta) gardnerae Spieker Labiosa (Rceta) gardnerce Spieker, Johns Hopkins Univ. Publ. Geol., No. 3, 1922, p. 168, pi. 10, fig. 10; upper part of Zorritos, Miocene, Peru. A number of samples of this species was obtained from dif- ferent parts of the Tubera group at the following points : Loc. 267, C. A. S., horizon R, Tubera group; Loc. 299, C. A. S., central part of the Tubera group; Loc. 325, C. A. S., central part of the Tubera group; Loc. 325-A, C. A. S., mid- dle part of the Tubera group. The species seems, therefore, to range from the central to the upper part of the Tubera group. 169. Labiosa (Raeta) hasletti Anderson, new species Plate 23, figures 2, 3 Shell large, inflated in front, somewhat produced and nar- row behind; height of holotype 47 mm., length, incomplete, J 78 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 61 mm., thickness 39 mm.; test thin, somewhat nacreous; beaks a little in advance of central, prominent and rather heavy; posterior slope slightly concave; shell thickest a little in advance and above the median plane; umbonal ridges inclined forward ; surface marked by strong concentric ridges, some of which are not continuous. This species is related to L. (Rceta) gibbosa, but is thicker, less produced in front and relatively more produced behind. It has not the straight posterior slope of Gabb's species. This species is abundant at Loc. 267, C. A. S., in horizon P, on the north slope of Tubera mountain. It has been named in honor of Mr. Thomas D. Haslett, by whose courtesy and aid the investigation of this district was greatly facilitated. Holotype: No. 4674, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon P, Tubera mountain, Colombia; Miocene. 170. Periploma caribana Anderson, new species Plate 23, figvire 1 Shell sub-nacreous, large, compressed, nearly circular in outline, and nearly equivalve; beaks relatively small, umbones not prominent, sub-central, crossed by an acute transverse ridge extending downward from the beaks; anterior end short, broadly rounded, quite closed ; posterior slope straight at first, then rounded, narrower than in front ; surface marked by undulating concentric ridges and lines of growth, the former stronger near the ventral margins; hinge not well known; height of holotype 61 mm., length 71 mm., thickness 27 mm. This shell is apparently rare, though three specimens were found in the upper part of the Tubera group. Its nearest rela- tive is probably Periploma peralta (Conrad) from the St. Mary's formation at Cave Point, Maryland. Its distinctness from this species is very evident upon a careful comparison and study of Conrad's description and figure. The three samples obtained were found at Loc. 267, C. A. S., at horizon R, Tubera village. Holotype: No. 4673, Mus. Calif. Acad. Sci., from Loc. 267, C. A. S., horizon R, Tubera village, Colombia ; Miocene. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA \yg FORAMINIFERA From the lowest horizon, M - N, of the Tubera group a number of micro-organisms were obtained from the matrix of the larger mollusks which were submitted for determination to Mr. C. C. Church. His notes regarding these forms are as follows : "The few Foraminifera obtained from this material are, for the most part, so poorly preserved that specific determination is practically impossible, although genera can be distinguished easily, and in the case of the large, well preserved Amphistegina the specific characters are quite clear. "171. Amphistegina lessoni D'Orbigny "This species is known from the Tertiary to the Recent and is a common form in the Miocene and Pliocene of the Atlantic coastal plain of the United States. It is known to exist at the present time in the tropical areas of the At- lantic, Pacific and Indian oceans, and is commonest in water of less than 30 fathoms in depth, but it also occurs at greater depths. "The species is highly variable in form, and ranges from a thin complanate disc to a subspherical test. In the younger and smaller individuals the um- biUcal area is a pronounced boss of clear shell material. "In the larger and flattened forms the umbilicus is not so prominent. The largest form noted is more than one millimeter wide and very thin. The material associated with the Foraminifera shows every indication of having been deposited in shallow water. "172. Qulnqueloculina auberiana (?) D'Orbigny "There is not much doubt that this form belongs to the species here assigned, but the fact that there are no very complete, or well preserved speci- mens makes it necessary to indicate a possible error. "173. Lituotuba lituiformis (?) (H. B. Brady) ' ' This genus is represented by a single individual which is not very well pre- served. The name of the genus is after Cushman's latest classification, but it is best known as Trochammina Parker & Jones." Besides the Foraminifera listed above there are a few other microscopic forms which deserve some mention. Among these are three species, and perhaps as many genera, of Ostracoda; also several small or embrvonic forms of bivalves and gastropods. 180 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. California Academy of Sciences Localities Following is a brief description of the fossil localities referred to in the preceding text, notes, tables, etc., and are of record in the Museum of the California Academy of Sciences: Locality 265 (C.A.S.). LocaUty 266 (C.A.S.). Locality 266-A(C.A.S.). Locality 267 (C.A.S.). Locality 296 (C.A.S.). Locality 297 (C.A.S.). Locality 298 (C.A.S.). Locality 299 (C.A.S.). Locality 302 (C.A.S.). Locality 303 (C.A.S.). LocaHty 304 (C.A.S.). LocaHty 305 (C.A.S.). Locality 306 (C.A.S.). Locality 323 (C.A.S.). Locality 325-A (C.A.S Locality 325-B (C.A.S Locality 347 (C.A.S.). LocaHty 348 (C.A.S.). Punta Piedras, three miles south of Paso Nuevo, De- partment of Bolivar, Colombia; marine Miocene. Quebrada San Juan de Acosta, near Puerto Colombia, Department of Atlantico, Colombia; marine Miocene. Falls in small creek, two miles west of Tuber a mountain. Department of Atlantico, Colombia; marine Miocene. Tubera mountain, Dept. of Atlantico, Colombia; M - N, 1.5 miles west of Tuber d. village; P, 1 mile west of Tubera village; R, Tuberd. village, near summit of the mountain. East border of Usiacuri village, Dept. of Atlantico, Colombia; 2000 feet above the base of the Tuberd group. Three miles west of Barranquilla, Colombia; coralline limestone, Pliocene. One mile east of Usiacuri village, Dept. of Atlantico, Colombia; top of Tuberd group, Miocene. Three miles southwest of Baranoa, Dept. of Atlantico, Colombia; west flank of the Usiacuri anticline, near well of Wm. Plotts; Miocene. Ranch of Sr. Banda, four miles south of San Andres, Dept. of Bolivar, Colombia; Tuberd group, Miocene. Two miles east of San Andres, Dept. of Bolivar, Colombia; Miocene. Four miles east of Santa Rosa, Ranch of Sra. Gomez, Dept. of Bolivar, Colombia; Tuberd group, Miocene. Near Turbaco, 16 miles east of Cartagena, Dept. of Bolivar, Colombia; Tubera group, Miocene. Usiaciu-i village, Dept. of Atlantico, Colombia; middle of Tuberd group, Miocene. Gatun Locks, Gatun, Canal Zone, Panama, Miocene. ). Between Chorrera and Cibarco, Dept. of Atlantico, Colombia; near middle of Tubera group, on west flank of Usiacuri anticline, Miocene. ).East of Usiacuri village (same as Loc. 306), Dept. of Atlantico, Colombia, Miocene. La Popa Hill, near Cartagena, Colombia; top of Miocene. Village of Turbaco, Dept. of Bolivar, Colombia, Pliocene. Vol. XVIII] ANDERSON— MARINE MIOCENE OF NORTH COLOMBIA \^\ Locality 349 (C.A.S.). Locality 350 (C.A.S.). Locality 351 (C.A.S.). Locality 353 (C.A.S.). Locality 354 (C.A.S.). Locality 355 (C.A.S.). • Locality 356 (C.A.S.). LocaHty 357 (C.A.S.). From four to five rhiles southwest of Barranquilla Colombia; top of the Miocene. Arboletes Bay, Dept. of Bolivar, Colombia; upper Miocene. Near Punta Pua, 20 miles north of Cartagena, Dept. of Bolivar, Colombia; Tuberd group, Miocene. Near Cospique Hill, Cartagena harbor, Colombia. Quebrada de Murindo, above Pedro de Claver, Dept. of Bolivar, Colombia; Tuberd group, Miocene. Quebrada de Murindo, 30 miles west of Monteria, Dept. of BoUvar, Colombia; Tubera group, Miocene. Pedro de Claver, Quebrada de Murindo, 30 miles west of Monteria, Dept. of BoHvar, Colombia, Miocene. Emory Wood Company's camp, Rio Canalete, west of Monteria, Dept. of Bolivar, Colombia; Miocene. 182 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Seb, Plate 8 Fig. 1. Fasciolaria olssoni Anderson, new species. Holotype No. 4617 (C. A. S. type coll.), Loc. 267 C. A. S., horizon P, Tuberd group; Tuberd mountain; p. 131. Figs. 2, 3. Fasciolaria olssoni Anderson, new species. Paratype No. 4618 (C. A. S. type coll.), front and rear views; Loc. 267, C. A. S., horizon R, Tuberd group, Tuberd village; p. 131. Figs. 4, 5. Mitra maurycs Anderson, new species. Holotype No. 4619 (C. A. S. type coll.), front and rear views; Loc. 267, C. A. S., horizon L, Las Perdices group, one mile west of Puerto Co- lombia; p. 130. Figs. 6, 7. Scohinella morierei (?) (Laville). Plesiotype No. 4620 (C. A. S. type coll.), front and rear views, Loc. 267, C. A. S., horizon L, Las Perdices group, one mile west of Puerto Colombia; p. 131. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 4 [ANDERSCN ] Plate 8 March 29. 1929 184 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Skr. Plate 9 Figs. 1, 2. Phos tuberainsis Anderson, new species. Holotype No. 4521 (C. A. vS. type coll.), rear and front views; Loc. 267, C. A. S., horizon P, Tubera group, north slope of Tubera mountain; p. 135. Fig. 3. Phos tubemensis Anderson, new species. Paratype No. 4622 (C. A. S. type coll.), Loc. 305, C. A. vS., near Turbaco, 14 miles east of Cartagena; p. 135. Figs. 4, 5. Conns tuberacola Anderson, new species. Holotype No. 4623 (C. A. S. type coll.), front and top views; Loc. 267, C. A. S., horizon M - N, Tubera group, west foot of Tubera mountain; p. 112. Figs. 6, 7. Conns crenospiratns Anderson, new species. Holotype No. 4624 (C. A. S. type coll.), front and top views, X2; Loc. 351, C. A. S., horizon P, Tubera group, near Punta Pua, 20 miles north of Cartagena; p. 112. Fig. 8. Typhis siphonifera Dall. Plesiotype No. 4625 (C. A. S. type coll.), X2; Loc. 325-A, C. A. S., horizon P, Tubera group, near Cibarco, a few miles north of Usiacuri, Colombia; p. 138. Figs. 9, 10. Ovula {Neosimnia) piiana Anderson, new species. Holotype No. 4626 (C. A. S. type coll.), front and rear views, X2; Loc. 351, C. A. S., horizon P, Tubera group, near Punta Pua, north of Cartagena; p. 140. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 4 [ANDERSON] Plate 9 18(3 CALIFORNIA ACADEMY OF SCIEXCES [Proc. 4th Ser. Plate 10 Figs. 1, 2. Caiiccllaria scheibei Anderson, new species. Holotype No. 4627 (C. A. S. type coll.), front and rear views; Loc. 306, C. A. .S., horizon P, Tubera group, near Usiacuri village: p. 114. Figs. 3, 4. Cancellaria scheibei Anderson, new species. Paratype No. 4528 (C. A. S. type coll.); 3, front view, 4, slightly rotated to show plaits; Loc. 304, C. A. S., horizon P. Tubera group, near vSanta Rosa; p. 114. Figs. 5, 6. Cancellaria hetineri Anderson, new species. Holotype No. 4629 (C. A. S. type coll.), rear and front views; Loc. 267, C. A. S., horizon P, Tubera group, north slope of Tubera mountain: p. 114. Figs. 7, 8. Cancellaria karsteni Anderson, new species. Holotype No. 463:) (C. A. S. type coll.), rear and front views; Loc. 305, C. A. S., horizon P, Tubera group, near Turbaco, 14 miles east of Carta- gena; p. 114. Fig. 9. Cancellaria karsteni Anderson, new species. Paratj-pe No. 4531 (C. A. S. type coll.), view showing columellar plaits; Loc. 267, C. A. S., horizon P, Tubera group, Tubera mountain; p. 114. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 4 ANDERSON J Plate 10 ](^i^ CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Plate 11 Figs. 1, 2. Melongena propatuliis Anderson, new species. Holotype No. 4632 (C. A. S. type coll.), rear and front views; Loc. 267, C. A. S., horizon R, Tubera group, Tubera village; p. 133. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVill, No. 4 ANDERSON] Plate 11 J90 CALIFORNIA ACADEMY OF SCIENCES I Proc. 4th Skr. Plate 12 Figs. 1, 2. Malea ringens (Swainson). Plesiotype No. 4633 (C. A. S. type coll.), rear and front views of mature shell, Bay of Panama, Hemphill collection; recent; p. 140. F'gs. 3, 4. Malea ringens (Swainson). Plesiotype No. 4634 (C. A. S. type coll.), front and side views of mature shell, outer Hp missing; Loc. 267, C. A. S., horizon P, Tubera group, north slope of Tubera n:ountain; p. 140. Figs. 5, 6. Malea ringens (Swainson). Plesiotype No. 4635 (C. A. S. type coll.), rear and front views of younger shell; Loc. 299, C. A. S., near Plotts' well, southwest of Baranoa, Colombia: p. 140. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 4 ANDERSON] Plate 12 ^-im 192 CALIFORXIA ACADEMY OF SCIEXCES [Proc. 4th Skk. Plate 13 Fig. 1. Ficus iPyriila) colombiana Anderson, new species. Holotype No. 4636 (C. A. S. type coll.), aperture view; Loc. 267, C. A. S., horizon P, Tubcra grcup, north slope of Tuhera mountain; p. 143. Fig. 2. Ficus (Fyrula) cclombiana Anderson, new species. Paratvpe Xo. 4637 (C. A. S. type coll.), Loc. 267, C. A. S., horizon R, Tul^era grcup, Tuhera village; p. 143. Fig. 3. Dentuliuni gramiduuum Anderson, new species. Holotype Xo. 4638 (C. A. S. type coll.), broken shell, parts joined with clay, outer layer of shell missing in part; Loc. 267, C. A. S., horizon L, Las Perdices group, west of Puerto Colombia; p. 144. Figs. 4, 5. Criicilvlum (Dispotcea) gatunen^e (Toula). Plesiotype Xo. 4630 (C. A. S. type coll.); 4, side view showing granular surface, X2; 5, interior showing cup, X2 : Loc. 323, C. A. S., Spillway of Gatun Locks, Canal Zone; p. 121. Fig. 6. Crucibulum (Disfotcea) gatunen^e (Toula). Plesiotype Xo. 4640 (C. A. S. type coll.), side view of smaller shell, X2; Loc. 3ili, as above; p. 121. Figs. 7, 8. Solenosteira santceracF Anderson, new species. Holotype Xo. 4641 (C. A. S. type coll.), rear and front views; Loc. 304. C. A. S., horizon P, Tubera group, four miles east of Santa Rosa. Co- lombia; p. 135. Figs. 9, 10. Solenosteira santcerosce Anderson, new species. Paratype No. 4642 (C. A. S. type coll.), rear and front views; Loc. 305, C. A. S., horizon P, Tuljcra group; p. 135. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 4 [ANDERSON | Plate 13 ]^94 CALIFORXIA ACADEMY OF SCIEXCES [Proc. 4th Skr. Plate 14 Figs. 1, 2. Cancellaria cibarcola Anderson, new species. Holotype Xo. 4643 (C. A. S. type coll.); 1, front view; 2, rotated to show columellar plaits; Loc. 325-A, C. A. S., horizon P, near Cibarco, Tnbera group; p. 1 16. Fig. 3. Cancellaria cibarcola Anderson, new species. Paratype No. 4644 (C. A. S. type colh), rear view of smaller shell; Loc. 325-A, as above; p. 116. Figs. 4, 5. Cancellaria codazzii Anderson, new species. Holotype Xo. 4645 (C. A. S. type coll.), rear and front views; Loc. 325-A, C. A. S., horizon P, Tubera group, near Cibarco, north of Usiacuri village; p. 116. Fig. 6. Cancellaria codazzii Anderson, new species. Paratype Xo. 4646 (C. A. S. type coll.), broken shell showing columellar plaits; Loc. 325-A; p. 116. Fig. 7. Cancellaria codazzii Anderson, new species. Paratype Xo. 4647 (C. A. S. type coll.), young shell, X2; Loc. 325-A, as above; p. 116. Fig^-. 8, 9. Polinices prolactea Anderson, new species. Holotype Xo. 4648 (C. A. S. type coll.), rear and front views; Loc. 267, C. A. S., horizon L, Las Perdices group, west of Puerto Colombia; p. 124. Figs. 10, 11. Cassis (Phuliuni) dalli Anderson, new species. Holotype Xo. 4649 (C. A. S. type coll.), rear and front views, X2; Loc. 267, C. A. S., horizon L, Las Perdices group, west of Puerto Colombia; p. 141. Figr. 12, 13. Cassis (Phalium) dalli Anderson, new species. Paratype No. 4650 (C. A. S. type coll.), rear and front views, X2; Loc. 267, C. A. S., horizon L, Las Perdices group, as above; p. 141. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIIl, No. 4 ANDERSON! Plate 14 196 CALIFORNIA ACADEMY OF SCIEXCES [Proc. 4th Ser. Plate 15 Figs. 1, 2. Fusiniis magdalenensis Anderson, new species. Holotype No 4651 (C. A. S. type coll.), front and rear views of incomplete shell; Loc. 267, C. A. S., horizon P, Tubera group, Tubera moun- tain; p. 133. Fig. 3. Fusiniis magdalenensis Anderson. Paratype No. 4652 (C. A. vS. type coll.), fragment showing character of long canal; Loc. 267, C. A. S., as above; p. 133. Figs. 4, 5. Cyprcea {Pustularia) gabbiana Guppy. Plesiotype No. 4653 (C. A. S. type coll.), upper and basal views, X2; Loc. 351, C. A. S., horizon P, Tubera group, near Punta Pua, north of Cartagena, Colombia; p. 139. Figs. 6, 7. Phos turbacocnsis Anderson, new species. Holotype No. 4654 (C. A. S. type coll.), front and rear views; Loc. 305, C. A. S., horizon P, Tubera group, near Turbaco, 14 miles east of Carta- gena; p. 136. PROC. CAL. ACAD. SCl., 4th Series, Vol. XVIII, No. 4 [ANDERSON] Plate 15 298 CALIFORNIA ACADEMY OF SCIENCES [Proc. -Ith Ser. Plate 16 Figs. 1, 2. Ampullaria tiibcracola Anderson, new species. Holotype No. 4655 (C. A. S. type coll.), front and basal views of mature shell; Loc. 267, C. A. S.. horizon R, Tubera group, Tubera village; p. 125. Fig. 3 Ami uUaria tuheracola Anderson, new species. Paratype, No. 4656 (C. A. S. type coll.), rear view of young shell; Loc. 267, C. A. S., as above; p. 125. Fig. 4. Phns baranoaiiiis Anderson, new species. Holotype No. 4657 (C. A. S. type coll.), front view; Loc. 299, C. A. S., horizon P, Tubera group, near Plotts' well, southwest of Baranoa, Colombia; p. \?>1 . Fig. 5. Phos haranoanus Anderson, new species. Paratype No. 4657a (C. A. vS. type coll.), aperture view showing lirate interior; Loc. 299, C. A. S., as above; p. 137. Figs. 6, 7. Cnllioitoma trof.ica Anderson, new species. Holotype No. 4168 (C. A. S. type coll.), side and basal views; Loc. 267, C. A. S., horizon M- N, Tubera group, west foot of Tubera mountain; p. 126. Figs. 7- A, 8. Solenosteira hasletti Anderson. Holotype No. 4169 (C. A. S. type coll.), rear and front views; Loc. 267, C. A. S., horizon M- N, Tubera group, west foot of Tubera mountain; p. 134. Fig. 9. Anomia mamillaris Anderson. Holotype No. 4165 (C. A. S. type coll.), Loc. 267, C. A. vS., horizon M-N, Tubera group, as above; p. 158. Fig. 10. Anomia mamillaris Anderson. Paratype No. 4167 (C. A. S. type coll.), Loc. as above; p. 158. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVlll, No. 4 [ANDERSON] Plate 16 March 29, 1929 200 CAUFORXIA ACADEMY OF SCIENCES [Pkoc. 4th Ser. Plate 17 Fig. 1. Tiirritdla fredeai Hodson. Plesiotype 4175 (C. A. vS. type coll.), example from Ijasal beds of Tubera group near Punta Pua, 20 miles north of Cartagena; p. 110. Figs. 2. 3. Oliva tuheraensis Anderson. Holotype No. 4-172 (C. A. S. type coll.), front and rear views; Loc. 267, C. A. S., horizon R, Tubera group, Tubera village; p. 128. Fig. 4. Tiirritella altilira Conrad. Plesiotype No. 4658 (C. A. S. type coll.), Loc. 267, C. A. vS., horizon M - N, Tubera group, west foot of Tubera mountain; p. 118. Fig. 5. Turritella altilira Conrad. Plesiotype No. 4659 (C. A. S. type coll.), Loc. 267, C. A. S., horizon M - N, as above; p. 118. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 4 ANDERSON 1 Plate 17 202 CALIFORNIA ACADEMY OF SCIEXCBS [Proc. 4th Ser. Plate 18 Figs. 1, 2. Area {Anadara) grandis Broderip & Sowerby. Plesiotype No. 4160 (C. A. S. type coll.), interior and exterior views of right valve. Example from Bay of Panama, F. M. Anderson, col- lector. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No, 4 ANCEKSONl Plate 18 204 C.4LIF0RXIA ACADEMY OF SCIEXCES [Proc. 4th Ser. Plate 19 Figs. 1, 2. Cardium {Trachycardium) pueblcense Anderson, new species. Holotype No. 4660 (C. A. S. type coll.), left side and front views; Loc. 267, C. A. S., horizon R, Tubera group, Tubeia village; p. 164. Fig. 3. Pecten atlanlicola Anderson, new species. Holotype No. 4661 (C. A. S. type coll.), right valve; Loc. 267, C. A. S., horizon P, Tubera group, north slope of Tubera mountain. Department of At- lantico; p. 156. Fig. 4. Pecten macloskeyi Anderson, new species. Holotype No. 4662 (C. A. S. type coll.), right valve; Loc. 267, C. A. S., horizon P, Tubera group, north slope of Tubera mountain; p. 157. Fig. 5. Pecten macloskeyi Anderson, new species. Paratype No. 4663 (C. A. S. type coll.), left valve of small shell; Loc. as above; p. 157. Fig. 6. Area (Anadara) Msiaciirii Anderson. Holotype No. 4158 (C. A. S. type coll.), interior of left valve; Loc. 306, C. A. S., horizon P, Tubera group, near Usiacuri village; p. 148. Fig. 7. Pecten atlanticola Anderson, new species. Paratype No. 4661a (C. A. S. type coll.), left valve; Loc. 267, C. A. S., horizon P, as above; p. 156. PROC. CAL. ACAD. SCI., 4th Series. Vol. XVIIl, No. 4 [ANDERSON] Plate 19 206 CALIFOKXIA ACADEMY OF SCIENCES [Peoc. 4th Skr. Plate 20 Figs. 1, 2. Mactrn [Afulhna?) atlanticola Anderson, new species. Holotype No. 4161 (C A. S. type coll.), side and anterior views; Loc. 267, C. A. S., horizon, M - X, Tuberd group, west foot of Tubera mountain; p. 175. Fig. 3. Mactra {Mulinial) atlanticola Anderson. Paratype No. 4162 (C. A. S. type coll.), interior view, showing hinge parts in right valve; Loc. 267, C. A. S., horizon P, as above; p. 175. Figs. 4. 5. Cardila iCardilamera) arala (Conrad). Plesiotype No. 4164 (C. A. S. type coll.), exterior and interior views of left valve; Loc. 267, C. A. S., horizon M - N, Tubera grouj:), west foot of Tubera mountain; p. 160. Fig. 6. Area {Anadara) usiacurii Anderson. Holotype No. 4158 (C. A. S. type coll.), exterior of left valve; Loc. 306, C. A. S., horizon P, Tubera group, near village of Usiacuri; p. 148. PROC CAL. ACAD. SCI., 4th Series, Vol. XVlll, No. 4 [ANDERSON] Plate 20 208 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Plate 21 Fig. 1. Thyasira bisecta (?) (Conrad). Plesiotype No. 4664 (C. A. S. type coll.), cast, left valve; Loc. 350, C. A. S., near Arboletes Bay upper beds of the Tubera group; p. 162. Figs. 2, ?>. TeUina f^rotolyra Anderson, new species. Holotype No. 4163 (C. A. S. type coll.), side and top views; Loc. 267, C. A. S., horizon M - N, Tubera group, west foot of Tubera mountain; p. 174. Fig. 4. Area (Anadara) iisiaciirii Anderson. Paratype No. 4159 (C. A. S. type coll.), interior of left valve, showing crowding of cardinal teeth; Loc. 267, C. A. S:, upper part of horizon M - N, Tubera group, west foot of Tubera mountain; p. 148. Fig. 5. Mactrella (Harvella) elegans (vSowerby). Plesiotype No. 4665 (C. A. S. type coll.), Loc. 267, C. A. S., horizon M - N, Tubera group, west foot of Tubera mountain; p. 176. Fig. 6. Mactrella (Harvella) elegans (Sowerby). Plesiotype No. 4666 (C. A. S. type coll.), Loc. 267, C. A. S., as above, showing hinge left valve; p. 176. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 4 [ANDERSON] Plate 21 210 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4tii Skr. Plate 22 Figs. 1, 2. Chama scheibei Anderson, new species. Holutype, No. 4667 (C. A. S. type coll.), exterior of left and right valves; Loc. 267, C. A. S., horizon M - N, Tubera group, west foot of Tubera mountain; p. 161. Figs. 3, 4. Clycymeris usiacurii Anderson, new species. Holotypc No. 4668 (C. A. wS. type coll.), external and internal views of left valve; Loc. 325, C. A. vS., horizon P, Tubera group, east of Usiacuri village; p. 153. Figs. 5, 6. Diplodonta woodringi Anderson, new species. Holotyjjc No. 4669 (C. A. S. type coll.), side and top views; Loc. 325-A, C. A. S., horizon P, Tubera group, near Cibarco; p. 162. Figs. 7, 8. Clycymeris lamyi Dall. Plesiotype No. 4670 (C. A. S. type coll.), external and internal views; Loc. 325-A, C. A. S., horizon P. Tubera group, near Cibarco, Colombia; p. 152. Fig. 9. Erycina tnrbacoensis Anderson, new species. Holotypc No. 4671 (C. A. S. type coll.), Loc. 305, C. A. S., horizon P, Tubera grouj), near Turbaco, 13 miles east of Cartagena; p. 163. Fig. 10. Erycina tnrbacoensis Anderson, new species. Paratype No. 4672 (C. A. S. type coll.), Loc. 305, C. A. S., as above; p. 163. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 4 ANDERSON] Plate 22 212 CALIFORXIA ACADEMY OF SCIENCES [Proc. 4th Skr. Plate 23 Fig. 1. Periploma caribana Anderson, new species. Holotype No. 4673 (C. A. S. type coll.), exterior view of left valve; Loc. 267, C. A. S., horizon R, Tubera group, Tubera village; p. 178. Figs. 2, 3. Labiosa (RcEta) hasletti Anderson, new species. Holotype \o. 4674 (C. A. S. type coll.), top and side views; Loc. 267, C. A. S., horizon P, Tulsera group, north slope of Tubera mountain; p. 177. Fig. 4. Tellitia (Eurytellina) cequiterminata Brown & Pilsbry. Plesiotype Xo. 4675 (C. A. S. type coll.), Loc. 304. C. A. S., horizon P, Tubera group, four miles east of Santa Rosa, 12 miles north of Carta- gena; p. 173. Figs. 5. 6. Chione atlanticana Anderson, new species. Holotype No. 4676 (C. A. S. type coll.), side and anterior views; Loc. 267, C. A. S., horizon P, Tubera group, north slope of Tubera mountain; p. 172. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 4 lANDERSON] Plate 23 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Fourth Series Vol. XVIII, No. 5, p. 215, plate 24 April 5, 1929 V A NEW PECTEN FROM THE SAN DIEGO PLIOCENE BY LEO GEORGE HERTLEIN Pecten (Plagioctenium) ericellus Hertlein, new species Plate 24, figures 10, 11 Shell small, moderately convex; hinge line straight. Right valve ornamented by about 22 subrounded, fairly low radiat- ing ribs which are separated by narrower interspaces; two tiny midribs are present along the base of the sides of the major ribs; ribs and interspaces crossed by concentric lines of growth ; anterior and posterior margins ornamented by con- centric lines of growth; ventral margin rounded; ears unequal, the anterior with a well-defined byssal notch, and sculpture of about five or six radiating riblets crossed by incremental lines ; the posterior ear sculptured by about four or five radiating rib- lets crossed by lines of growth, no notch present. Altitude 28 mm. ; longitude 29. 1 mm. ; diameter of right valve approxi- mately 7.5 mm. ; apical angle in right valve approximately 94°. Holotype: No. 2998, Mus. Calif. Acad. Sci., from Loc. 1132 (C. A, S.), Pacific Beach, San Diego, California, C. H. Stern- berg collector; San Diego, Pliocene. This interesting little species differs from P. invalidiis Hanna, and P. circularis Sowerby, in its numerous low, more rounded, narrower ribs and in the possession of fine secondary ribs along the base of the sides of the major ribs. This species is named for the late Eric Knight Jordan. April 5, 1929 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Fourth Series Vol. XVIII, No. 6, pp. 217, 218, plate 24 April 5, 1929 VI A NEW SPECIES OF LAND SNAIL FROM KERN COUNTY, CALIFORNIA BY G. DALLAS HANNA Helminthoglypta berryi Hanna, new species Plate 24, figures 7, 8, 9 Shell of medium size, globose, composed of 5 3^ well- rounded whorls; suture deep; umbilicus completely closed in the holotype, almost closed in the paratype; white or pale brown, bandless (in all specimens seen) ; upper portion of whorls sculptured with irregular growth ridges, almost ribs ; lower portion of body whorl with a series of malleations, becoming pits in some cases; these pits roughly arranged in spiral order and almost obliterate the growth lines near the margin of the shell; the line of demarcation between the series of growth ridges above and the malleations below is very sharp and is approximately in the position of the color band as usually developed in this genus; aperture large and capacious; outer lip moderately reflected; terminations of peristome connected by a wash of callus over the body whorl. Diameter (holotype) 22.5 mm., height 21 mm.; diameter (paratype 1493) 23 mm., height 21 mm. Holotype: No. 1492; paratypes: Nos. 1493, 1494, Mus. Calif. Acad. Sci., collected by G. D. Hanna eight miles northeast of Bakersfield, Kern County, California. .^v^ 218 CALIFORNIA ACADEMY OF SCIENCES [Pkoc. 4th Ser. The first known specimens of this remarkable species were found in 1926 in the S. W. >4 Sec. 32, T. 27 S., R. 29 E., M. D. M., about two miles north of Poso Creek and five miles east of the mouth of Granite Creek. These were somewhat imperfect and seemed so unusual in character and habitat that better material was awaited for description. This was found in 1927, 1^ miles southeast of the top of Round Mountain, Sec. 30, T. 28 S., R. 29 E., M. D. M., about three-fourths mile north of Kern River and four miles east of Oil City ; this is the type locality. Specimens were also found further east on Sec. 34, T. 28 S., R. 29 E., M. D. M., and fragments were seen scattered in other places. It is evident that the species is fairly widely distributed in this district. All of the shells found were dead,^ but the one made the holotype has the epidermis and the pale brown color preserved. All were found on the slopes of dry, barren, ashy hills, usually, but not always, on northern slopes. No rock outcrops occur near where the shells were found, but invari- ably they were in torn up earth where cattle had trampled during wet weather. This peculiar habitat, with the pale color and absence of a band, leads to the supposition that the animal is a burrowing form. After having collected snails rather extensively in the forests and among the rocks of California, I was most astonished to find this one on soft, powdery, ashy hills. The shape is suggestive of the shell found near Monterey called calif ornicnsis, but in other characters there is little resemblance. The species is named for Dr. S. Stillman Berry in recogni- tion of his extensive studies of west American land shells. ^ Since this was written Dr. Berry has collected living specimens of what appears to be the same species in the Kern River oil field and the characters as outlined are confirmed in most respects; the living shell seems thinner than the dead ones upon which the description was based. The habitat is definitely proved not always to be the ashy hills as at first supposed. PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Fourth Series Vol. XVIII, No. 7, pp. 219, 220, plate 24 April 5, 1929 VII A NEW SPECIES OF LAND SNAIL FROM COAHUILA, MEXICO BY G. DALLAS HANNA and LEO GEORGE HERTLEIN In the autumn of 1926 several species of land shells from central, southern Coahuila, Mexico, were added to the collec- tions of the California Academy of Sciences. One species of Holospira appears to be undescribed. The specimens were collected about 16 kilometers north of Ramos Arizpe on the road to Paredon, Coahuila, Mexico. Holospira aguerreverei Hanna & Hertlein, new species Plate 24, figures 5, 6 Shell white, composed of 13.5 whorls, the earliest 2.5 smooth, the succeeding three rather indistinctly ribbed ; those following and constituting the body of the shell with only faint growth lines, almost glossy; the greatest diameter is at the fourth and fifth whorls from the last, thus producing a spindle-shaped shell ; last whorl with about 19 costse ; the number is rather indefinite because close behind the apertural expansion the ribs decrease in size and are close together; some of the later ribs are slightly sinuous below due to the constricted basal cord; imperforate; aperture projecting slightly beyond last whorl, roundly triangular in form, without lamellae; lip expanded uniformly, brilliant, glossy white; between the lip 220 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th See. and the umbibial region, the ribs continuing from the outside are much finer and somewhat indistinct close to the lip ; sutures deeply impressed in the embryonic portion constituting the first five whorls, elsewhere the whorls are much flattened; the upper and outer part of the free portion of the last whorl is very sharply angulated between the lip and the shell ; the basal portion becomes a cord through the presence on both sides of a depression. Measurements Length Diameter 21.7 mm. 6.0 mm. Holotype, No. 2848 (C. A. S.) 23.3 mm. 6.1 mm. Paratype, No. 2849 (C. A. S.) 22.5 mm. 6.4 mm. Paratype, No. 2850 (C. A. S.) 20.5 mm. 6.1 mm. Paratype, No. 2851 (C. A. S.) Holotype: No. 2848; paratypes: Nos. 2849-2853, Mus. Calif. Acad. Sci., from 16 kilometers north of Ramos Arizpe, Coahuila, Mexico; Santiago E. Aguerrevere, collector. The species apparently comes closest to H. semisculpta Stearns,^ from San Carlos, Chihuahua, but the last whorls are not so constricted as in that species. H. mesolia Pilsbry,^ from Terrell County, Texas, appears to belong to the same group of species but is even more constricted toward the base than seuiisciilpta. H. pasonis Dall^ is another similar but much more coarsely-ribbed species basally and lacks the basal keel. H. coahuilensis (Binney)* from "Cienga Grande," Coahuila, which might be expected to be closest to the shell here described, is a much larger species, being 29 mm. long, lacks the basal keel and has only about 10 ribs on the last whorl, according to Pilsbry.^ The species is named for Mr. Santiago E. Aguerrevere who made the collection. » Proc. U. S. Nat. Mus., Vol. 13, 1890, p. 208, pi. IS, figs. 1, 4. 2 Nautilus, Vol. 26, 1912, p. 89. » Nautilus, Vol. 8, 1895, p. 112. *Amer. Journ. Conch., Vol. 1, 1865, p. SO, pi. 7, figs. 4, 5. » Man. Conch., Vol. 15, sen 2, 1903, p. 92. \o PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Fourth Series Vol. XVIII, No. 8, pp. 221-227, plate 24 April 5, 1929 VIII SOME NOTES ON OREOHELIX BY JUNIUS HENDERSON Oreohelix peripherica castanea (Hemphill) has long been a puzzle to me. A great deal of material from Mr. Hemphill in many public and private collections, labelled castanea, or rather castaneus, bears no resemblance to the form that he originally designated by that name. For example, there are in the University of Colorado Museum seven such lots from the Hemphill collection. The specimens from White Bird, Idaho, labelled castaneus, are just like some that are labelled hicolor, while two others from the same place, labelled castct- neus, show some indications of the variegated colors of vari- abilis, but are more depressed and differ in sculpture and some other characters. White Bird is a locality where some exten- sive and intensive collecting should be done and the material from each station studied as a whole, before being divided into varieties, in order to comprehend the real significance of Hemphill's "varieties." I have always considered castanea a very slightly differ- entiated color form, almost an exact synonym of Oreohelix peripherica albofasciata (Hemphill)^ and still do, but that does not dispose of the whole problem. Henry Hemphill, in his » See Henderson and Daniels. Proc. Acad. Nat. Sci., Phila., LXVIII, 330-334, 1916. Henderson, Univ. Colo. Studies, XIII, 116-117, 1924. Pilsbry, Proc. Acad. Nat. Sci.. Phila.. LXVIII. 343-357, 1916. 222 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. notes published by Binney,^ gives an account of an interesting and variable fauna of snails which Binney called Patula, but now called Oreohelix, at the locality where the Bear River breaks through the low range of mountains south of Cache Junction, Utah, into the Salt Lake Valley. Among other things, he says that at the foot of a cliff he "found a colony of the ribbed variety castaneus. This spot is continually shaded, the sun never shining on it. Most of this colony are faintly marked with the broad white band of albofasciata, but a few are plain chestnut-colored." This is plainly the type locality of castanea and the vicinity is also the type locality of typical albofasciata and several other varieties of peripherica. On page 32, Binney gives the localities for castanea as fol- lows : "Box Elder County, Utah; also Celilo, 15 miles from The Dalles, Oregon. (Hemphill.)" In a footnote he says of the Celilo colony: "Probably a colony brought down by the Columbia. It was not found on a subsequent visit." Whence was it brought by the Columbia? Surely not from the Box Elder County locality, which is the type locality of castanea, for that is not in the Columbia drainage and has not been except when, during the greatest Pleistocene expansion of ancient Lake Bonneville, it established an outlet at the north- ern end of the basin. Furthermore, the Celilo colony is not the same thing at all as that called castaneus by Hemphill in his note. On Plate 2, figures 11 and 14, Binney shows castanea as a rather dark, strongly-ribbed form, one figure being quite high- spired, as in typical albofasciata, the other being depressed, but such difference in elevation is often seen in colonies of Oreohelix. It seems perfectly clear that the Utah material first mentioned by Hemphill and figured by Binney in his figure 14, if not figure 11, must be considered the typical form — the real castanea — and figure 14 the type figure. It seems also perfectly clear that this is merely a variable melanistic form of albofasciata, in which the broad, white peripheral band is more or less obscured by a wash of brown, a phenomenon not at all uncom- mon in Oreohelix, especially in 0. depressa. All of the material assignable to peripherica or any of its varieties that ^ Binney, 2nd Supplement to 5th Vol. Terr. Air-breathing Moll. U. S. and adjacent territories, p. 31, 1886. Vol. XVIII] HENDERSON— NOTES ON OREOHELIX 223 I have seen in collections, or found myself, have been from the Salt Lake Valley and its tributaries. Binney and Hemphill called all snails now placed in the genus Oreohelix varieties of Helix (or Patula) strigosa, even such very diverse things as haydeni and cooperi. In the Uni- versity of Colorado Museum three examples (No. 7140) from the Hemphill collection are labelled "//. strigosa v. castaneus, Utah." Two of them are almost typical albofasciata, but the other has the peripheral band somewhat obscured and may be considered castanea. In what Hemphill considered his "Main Collection," now in the California Academy of Sciences, of lot No. 7589, bearing a similar label, there are five specimens, all quite dark, the peripheral band showing but dimly, hence typi- cal castanea. Lot No. 7590, four specimens, bearing a similar label with the additional words "paler — longer," are light, uniform brown. I have selected the best example of No. 7589 in the Academy collection as a lectotype, which has been assigned the number 2986 in the type collection (C. A. S.). It is fully adult and has five whorls. Its size, form and sculp- ture are well represented by Binney 's figure 14; figure 1 accompanying the present paper is from a photograph of it by the author; diameter 15.5, altitude 13 mm. In the University of Colorado Museum there are two specimens of this form that I found near the tunnel at Wheelon, Utah, very close to the northern boundary of Box Elder County, and certainly but a very short distance from the type locality of castanea. The Oregon material presents greater difficulty. In order fully to understand the Hemphill material scattered through many collections, one must remember that he had a habit of dividing the specimens from a given colony into "varieties," based mostly upon slight differences in color or elevation of spire, often well marked in typical examples but grading com- pletely into one another, and the division of his material was not always made altogether consistently. Furthermore, he was very careless about his locality labels, left many of them very vague, and did not give the locality in the same language in the different "varieties" from the same colony. Thus his Oreohelix material from Oregon probably all came from the single colony at Celilo, as I concluded from an examination of the material itself, though some of it is labelled 224 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. merely "eastern Oregon," and Celilo is not in eastern Oregon,, except in the loose sense in which the term "eastern" is often used to distinguish the more arid portions of Washington and Oregon from the moist belt of the western portions of the states. Celilo is on the northern boundary of Oregon west of the middle north-south line. To reinforce the conclusion drawn from an examination of the material, we have the fact that Hemphill and Binney mention no other Oregon locality for this genus than Celilo, and the further fact that in three of five lots examined the locality is given as "eastern Oregon, near Celilo." California Academy of Sciences' Nos. 7681 and 7684 are labelled Helix strigosa var. cooperi, while Nos. 7587 and 7588 are labelled Helix strigosa var. castaneus, yet I am rather confident that these all came from one variable colony, such as are not uncommon with the genus Oreohelix, and I am equally confident that they have nothing to do with either cooperi or the form that he called castaneus from the type locality in Utah. No. 7587 carries the additional words "elevated, smooth," while No. 7588 reads "depressed, smooth, one reversed." University of Colorado Museum No. 7142, from the Hemphill collection, is labelled "Helix alternata Say var. castaneus Hemphill, eastern Oregon." I believe this lot is also from Celilo. I cannot identify any of this Oregon ma- terial with any described species and am therefore naming and describing it as new. Oreohelix variabilis Henderson, new species Plate 24, figures 2, 3, 4 Shell rather elevated, solid, whitish, variegated with small, irregular, very light-brown blotches; whorls 5^, fairly con- vex, bluntly angled at the periphery, the angulation continuing at least to beginning of last whorl, but not to the aperture: transverse sculpture rather coarse, irregular striae, about as in cooperi and depressa, crossed by very fine, obscure, irregular, incised, spiral lines. Under a lens of good power the whole surface of the last whorl appears rough and coarse. The last whorl turns more decidedly downward toward the aperture than in most species of Oreohelix, the ends of the peristome N'OL. XVIII] HENDERSON— NOTES ON OREOHELIX 225 coming rather close together and being connected by a very thick callus, thus forming an almost continuous peristome. This feature is not entirely accidental, as it is as well developed in several other specimens, though on others the callus is thin- ner and the downward turn of the whorl not quite so pro- nounced. The aperture is very oblique, somewhat wider than high, the abrupt downward turn at the base giving the ap- pearance of a strong rib within, parallel with the lip. Diame- ter 22 mm.; altitude 16 mm. The smallest example in this lot of 12 specimens has a diameter of 15 mm., altitude 11 mm. Holotype: No. 2987; paratypes: Nos. 2988, 2989, Mus. Calif. Acad. Sci., from Celilo, Oregon. Henry Hemphill collector. Some examples of this lot exhibit a few faint, narrow, spiral color bands both above and below the periphery. Four specimens of the five in lot No. 7681 exhibit one strong brown band just below the periphery, a broad band just below the suture, the two separated by a whitish band, with traces of finer bands on the base. The fifth example is coarsely ribbed, with broad, blackish bands, and does not seem to belong with the rest at all. It is not unlikely that it belongs with the Utah material and was mixed with this lot before the material was numbered. I have found much evidence of such mixtures in Hemphiirs collections. Lot No. 7587 consists of five slightly more elevated shells, each pretty well covered with a reddish- brown wash, but on the base showing the characteristic color- ing of this species and being in other ways unlike castanea. The same is true of the five examples in lot No. 7588, but they are rather depressed and one of them is reversed. The five specimens in lot No. 7142, University of Colorado Museum, are similar to No. 7587, but average a little smaller. In the more elevated examples of variahilis the spire is dis- tinctly more straightly conical than in elevated forms of cooperi or peripherica (-\- castanea, etc.), which tend more toward a dome-like outline. Dr. Henry A. Pilsbry writes me that he has found in the Hemphill material in the Academy of Natural Sciences of Philadelphia two topotypic specimens of O. variabilis which long ago had been placed with their large collection of cooperi and hence overlooked. 226 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Plate 24 Fig. 1. Oreohelix peripherica alhofasciata, color form castanea (Hemphill)' diameter, 15.5 mm.; lectotype No. 2986 (C. A. S. type coll.), from Box Elder Co., Utah; p. 221. Fig. 2. Oreohelix variabilis Henderson, new species; diameter, 22 mm.; holo type No. 2987 (C. A. S. type coU.), from near Celilo, Oregon' p. 224. Fig. 3. Oreohelix variabilis Henderson, new species; diameter, 20.1 mm.; para type No. 2988 (C. A. S. type coll.), from near Celilo, Oregon; p. 224. Fig. 4. Oreohelix variabilis Henderson, new species; diameter, 19.4 mm.; paratype No. 2989 (C. A. S. type coll.), from near Celilo, Oregon; p. 224. Fig. 5. Holospira aguerreverei Hanna & Hertlein, new species; true length 21.7 mm., diameter 6.0 mm.; holotype No. 2848 (C. A. S. type coll.), from 16 kilometers north of Ramos Arizpe, Coahuila, Mexico; p. 219. Fig. 6. Holospira aguerreverei Hanna & Hertlein, new species; side view of specimen shown in fig. 5; p. 219. Figs. 7, 8, 9. Helminthoglypta berryi Hanna, new species; diameter, 22.5 mm. J holotype No. 1492 (C. A. S. type coll.), from eight miles north- east of Bakersfield, Kem County, California; p. 217. Figs. 10, 11. Pecten {Plagiodenium) ericellus Hertlein, new species; altitude, 28 mm.; holotype No. 2998 (C. A. S. type coll.), from locality 1132 (C. A. S.), Pacific Beach, San Diego, CaUfomia. PUo- cene; p. 215. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, Nos. 5, 6, 7, 8 fHERTLEIN, HANNA, HANNA& HERTLEIN, HENDERSON] Plate 24 8 PROCEEDINGS OF THE ^ CALIFORNIA ACADEMY OF SCIENCES Fourth Series Vol. XVIII, No. 9, pp. 229-243, plates 25, 26 April 5, 1929 IX NOTES ON THE NORTHERN ELEPHANT SEAL BY M. E. McLELLAN DAVIDSON Assistant Curator, Department of Ornithology and Mammalogy After passing through various nomenclatural vicissitudes, the elephant seal of southern waters had apparently arrived at a certain permanence in Macrorhinus leoninus until the appearance, in 1909, of Lydekker's paper "On the Skull- Characters in the Southern Sea-Elephant."^ Basing his studies on the skulls of two males from Macquarie Island, a male from Chatham Island, a female from the "Antarctic Seas," a male from the Crozet group, and an old male from the Falklands. Lydekker reached the conclusion that the dif- ferences found in the palatal regions of these specimens war- ranted the recognition of the following species and subspecies : Macrorhinus leoninus typicus [=M. I. leoninus] (Juan Fer- nandez) ; M. I. falclandicus (Falkland Islands), perhaps inseparable from the typical race; M. I. uiacquariensis (Mac- quarie and PChatham islands) ; and M. crosetensis (Crozet and PKerguelen and Heard islands). In this paper no attempt was made to discuss the cranial features of the Northern Elephant Seal, but Lydekker noted that the characters exhibited by the palatines of a skull of that form were sufficient for its recognition as a distinct species. In an appended note, resulting from a communication from » p. Z. S. 1909, pp. 600-606. April 5, 19J9 230 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Rothschild, it was remarked that should the northern species prove to be identical with that from Juan Fernandez and Chile, the specific name leoniiiiis should be reserved for the animals of that region, and the elephant seals from the Falk- lands, and from Macquarie and Chatham islands should be known respectively as M. falclandicus falclandicus and M. f. macquariensis. The reason for this division is obscure in view of the statement that the Falkland Island race was perhaps inseparable from the typical leoninus. Lydekker's paper proved but the prodrome of one by Lonn- berg.^ While recognizing the probability of a widely distrib- uted species being separable into geographic races, Lonnberg found himself unable to accept Lydekker's conclusions. The characters upon which the latter's species and subspecies were founded (with the exception of the breadth of palate in the Crozet example), all fell within the range of variation exhib- ited by the series (seven adult and semi-adult males, three young males, and one adult female) from South Georgia examined by Lonnberg. With a series of five males and four females from Guada- lupe Island, one male from the Falklands, two or three from Macquarie, and two or three from Crozet Island at his dis- posal, Rothschild^ continued the discussion. Although the promised article, giving the constant characters by which the various subspecies might be recognized, has not yet appeared, Rothschild confessed his faith in the validity of Mirounga leonina leonina (coasts of California and adjacent islands, wintering on Chilean coasts), M. I. patagonica (Falkland Islands, South Georgia, and ?South Shetlands), M. I. kergue- leiisis = M. I. crosettensis [.yfc] (Herd, Kerguelen, Crozet islands, etc.), M. I. peronii (islands of Bass Straits), and M. I. macqiiarieitsis (Macquarie Islands). Apparently Rothschild was governed largely in his decision as to the unity of the species by a report from Harris "that he must reach the island [Guadalupe] before the middle of May or the Sea Elephants would have migrated to the south." Rothschild states that he "looked up the matter, and . , . found that, although a few stray individuals might formerly 2 p. Z. S. 1910, pp. 580-588. 3 Nov. Zool., XVII, pp. 445-446. Vol. XVIII] DAVIDSON— NORTHERN ELEPHANT SEAL 231 have led a pelagic life north of the Equator, the bulk of the Northern Sea Elephants migrated in the hot weather to the Chilean coast and the islands near (Juan Fernandez, Masa- fuera, etc.)," but the sources of his information are not revealed. That rookeries are more or less completely deserted sub- sequent to the breeding season is hardly sufficient for the determination of a migration in any particular direction. Breeding grounds in the southern hemisphere (Kerguelen, South Georgia, and Macquarie, etc.) are similarly vacated after the season of reproduction, and the fact that elephant seals have been found in the Antarctic pack ice (65° 08' S.)* and at Cape Royds {77° 40' S.)° in January is evidence of a movement away from, rather than across, the equator. Moreover, the information furnished by Harris was inac- curate. Scammon® found several cows and their young, the latter apparently but a few days old, on Santa Barbara Island in June, 1852. Townsend^ reports finding a pup three weeks old on Guadalupe Island, October 9, 1883; and the new-born young he met with on the Lower Californian islands in 1883-84 were dropped at various times between November 1 and February 1. In 1911, he saw a dozen or more females with very young pups on March 5 at Guadalupe Island. In the Academy's collection are skull and skeleton of a pup a few weeks old taken on Guadalupe Island, May 8, 1914. Reports from recent expeditions visiting Guadalupe Island during the summer months indicate the presence of a consid- erable herd at that season. The Tecate Expedition^ reported the presence of 264 adult animals, July 12, 1922, and 300 four days later. Mexican officials visiting the island in early Sep- tember of the same year found 150 females and an equal num- ber of pups about 30 inches in length.^ In 1923, 366^" were counted on July 16, and on August 30 of the succeeding year 'Wilkes, C, Nar. U. S. Expl. Exped,, II, p. 291. "Wilson, E. A., Geog. Jour., XXV, p. 393; Nat. Antarctic Exped., N. H., Zool., II, p. S3. "Marine Mammals of the North-western Coast of North America, p. 118. 'Proc. U. S. Nat. Mus., VIII, p. 93; Bull. Am. Mus. Nat. Hist., XXXV, p. 407. «Hanna, Proc. Calif. Acad. Sci., 4th Ser., XIV, p. 229; Anthony, Jour. Mam., V, p. 146; Proc. Calif. Acad. Sci., 4th Sen, XIV, pp. 310, 313. » Anthony, Proc. Calif. Acad. Sci., 4th Ser., XIV, p. 313. "Huey, Science, n. s., LXI, p. 406; Anthony, Jour. Mam., V, p. 148. 232 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 124" occupied the beach. A total of 465 animals was found on the island on June 23, 1926/^ This evidence of an extended breeding season and the pres- ence of a considerable herd of elephant seals in North Ameri- can waters during all seasons seem to militate against the view of a migration of these animals to Juan Fernandez, especially as Anson^^ found elephant seals with young on that island dur- ing his stay, from June 10 to September 19. It is stated that the young were born during the "winter." With a view to aiding in the determination of the status of the elephant seals of the north Pacific, the Academy's series of specimens has been examined. These examples apparently but inadequately represent the Guadalupe animals, skulls of greater length, two feet (605 mm.)^* and 556 mm.,^^ having been known. It has been deemed advisable, however, to place the measurements on record, together with notes on structural characters. In order to facilitate comparisons with previously published figures, percentages of basal length of skulls, in addition to the actual measurements, have been given in the appended table. The incomplete skulls appear to be those of adult males. The open pulp cavities of the canines and the condition of the sutures in the largest complete skull bear witness to the ani- mal's immaturity, and even the skulls of the somewhat older females show that the possibilities of additional growth had not been exhausted. Through the courtesy of Dr. Charles Anderson, Director of the Australian Museum, Sydney, and Mr. George P. Engel- hardt, Curator of the Department of Natural Sciences, Brooklyn Museum, Brooklyn, an examination of the skulls of two adult males from Macquarie Island and one adult male from South Georgia has been made possible. The measure- ments of these specimens are given below. In basal length the complete skulls of the males in the Academy's collection fall considerably short of those from " Huey, Science, n. s., LXI, p. 406. "Huey, Jour. Mam., VII, p. 160. ^*Cf. Thomas, Jour. Voy. to the South Seas in 1740-44, p. 40. ^'Townsend, Proc. U. S. Nat. Mus., VIII, p. 93. "Huey, Jour. Mam., V, p. 241. Vol. XVIII] DAVIDSON— NORTHERN ELEPHANT SEAL 233 O OfO OO »or^ C > 00 •* «o c > 10 10 -* 10 Ov C > »o 1 00 O O fs Ooo »^ »H O-^OOtOOO OOv i^CN«-»0\Or^| 0 OO to ■^ OChOfOOoo Oro i0v000>0 i O O -i* O On t^ ■* Ot^lO00'*'t~ 0»0 l^roiiOC^t^ ▼-H 00 00 vo 00 fo OS o ON»rioot---*oo (noo ost^ CO CO Ov 1-1 CS ^\0 ■*•<* O\ro r-H 10 1-1 0> C^ Ov 10 00 c- «* 1-1 rj* t*5 tN 1—) 1—1 O O^O 0-* Oi-< - 10 vo t-- CN •<* «o O iOt~ O •* O \0 O »0 CN t^ o ■<* OC vo 10 t-- «o ro ■* O tNOC 1-1 00 PO \c VO OC r- 1-- ci- ^-< u- 1—* ▼- vo lO '-I >o l^ •^ Ocs CN vO 1- Ov C^ Ov \r OV CN "t 1- »o 0 uoio o "- CN •^ 't t- On O OO or- CNTt u- 10 "* t- CO CN 00 c vo OvOC "" \0 1- t^ u- CN C rl* C tJI c- 10 I~- CO «-( O OvC TjHrt 0. c<- 1—1 vo Cv Ov CO Ov vc 00 CN rti »H f<- CN i-H *H c o\c >oc CN OC "* IT' 00 c Oi- t-» cs ^ C «0 CO OC 10 tJh 10 Tj< CN to r- oo 1-1 CN >0 r- VO 00 OC 10 '^ IT) VO CO ^H 00 ^ ^vc 1-1 r<^ CN vo »- Ov cv CN VC CO ^ 1- "^ CO fr. CN 1—1 CN 1— t c Orf c: CO -^ 00 Ov 10 CO 1- Q to 0\ c 000 CO^ iH 10 CN 10 10 r- t^ u^ r-cc Tt «o C VO VO CO CN 0\ CN t^ rj< 10 t^ tn c t^ m: 1-1 10 CN CO CO 00 C^ Cv 10 v£ CN CO 10 1- VC ■* "* cs CN 1-H 1^ CN 1-H O f 0^ t^ vo ^ c 10 vo c OO OOC IDi* 10 t- 10 «/~ t- c CN T^ 10 c to r- OS J~- 00 — 00 1- UO ir- C Ov •^ Tt< c i-H CN 1^ CO CN 00 1— t ^ OsvC CN l/~ co CO CN Ov Cv cv vO vO '^ CO «0 1-1 t-^ to 't CN CN 1^ 1— t CN 1—1 o 00 o (N) <5 1— ( t^ 00 ■* CN CO 00 vo ^ f^ o i-H 0\ 1— ( 00 00 tH CO CO 1— ( Ov CN r— - a c 13 "«■ "« 1 53 of li oj J3 03 rt 03 8 to s <-5 u a c ) bi C 1 ) 1 ) •3 G 11 o3 cS c 0) c n ^ 0^ ao ^-4 a c c c 3 u _o a, 03 w c C 11 on a level wit maxillary sutur [ on level with th larv suture in / C 1 +-> 'a VM +J +- oj c: S E c c C3 c ^ as ngth of int ngth of int lenp+.h . . . ngth of pre ngth of pre 0/0 of ha oo| TJ'd idth of skul externus. idth of sku] idth of sku terior pre idth of skul oremaxil ;ast frontal ;ast frontal length. . c c :S bj c 0) ' 1 PC > -> > -, a. 1- ^ ^ ^ ^ 0) ^ ^ ^ ^ ^ ^ t— ^ U 1 234 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4t5 Ser. OCN OCN O 0 CO vO^ OS to Tj* CJ Oi ^-H ^— 4 v—i (M *— 4 -1 1— 1 cr' o OfO OO OfS OOv O y-< Ort< Ot^ tri OpO rt o Ot- O cs OtJ. OOO O CN UO 00 to •rM 10 t^ to S in iri-*' core foo t-~ 0\ lO-*' 00 (M \0 t^ t^ to OvO to 00 t^ lO t-« lO lO t~- ro "* ^ (>q OCN r— to fO t^ »-H o\ to ■^ fO es *— ( »-H ^— t cs ^—4 o 0-* o-* OiO OOv OOn O CN 00 00 Oro o O'* Orf OvO OOO O^O too t^ 00 00 l| 1/5 ^lO O lO "T* lod J 00 o 00 t^ 00 lO t^ PO lO ^ O r-j rriCN cs \o VO CO 00 '-I to to CO fS t-H T-H T-l fO »-H J3 b40 C t— 4 03 • Cll -4- 1 :: > n! It ^ t/3 I4 4 tf 4 a 7 "> • c 1 w ) a rior edge rior edge "aS ^^ • ^ '■ c r 4 4 4 i J/ i : c ) . c ) ; C , oj a 1 a) a > -dT: 3 4) OJ > a> (/i ^ 03 ; C 4 4 4 ; c e r 1 1 ,2^ 3 rt ^ 3 6 £ 4 (4-1 U- 3 cj a 1 +-> +J < ' 04a, : -5^ 4 > : c3 ^ t .y.5: 3 0) a I -(-> +■ ; as i a ) a > -4- 5 'rt'ci J +-> +■ 3 C C 5 CTl C 3 +-> +■ . c 1 -IJ -tJ 3 Pl 3 > 03 05 : Ti : 'rt'a ; +-> +- 1 >4- 4 73 J ) a a i ^1 i -l-> -4- 3 ^X 5 oc 4 X-£ 4 -4-> -4- J O C * -4-» +■ 5 V4_4 <♦■ 4 C 3 ,c!X > ^ 4 H-4 '♦■ +3 +. 1 C ' X > +J •— ldT3 0) ClJ « 1 b 1 c : i n PC > •^ a J >- ) a 3 ^ :^ ; 4 > a ) a 3 ^ ) a 3h. i ^ ■■^ ; ■^-' :^ b" :^ a ' 1— ;^ ) a i c ) Vol. XVIII] DAVIDSON— NORTHERN ELEPHANT SEAL 235 South Georgia and Macquarie Island, but it must be borne in mind that all save one were from young animals. From the zygomatic breadth of the two incomplete skulls it may be assumed that their length would be nearly equal to the largest South Georgia ones, and the previously mentioned skulls measured by Townsend and Huey exceed in this dimension. It may be noted that the crania of two adult females from Guadalupe surpass by 60 and 35 mm. Lonnberg's South Georgia example. In the South Georgia specimens as well as in the Guadalupe ones the greatest relative zygomatic width occurred in quite young animals. It is, therefore, of significance that while the zygomatic breadth of only one adult or semi-adult from South Georgia fell below 70 per cent of the basal length, only one (an immature female) from Guadalupe Island had a zygo- matic breadth of more than 70 per cent. Six of Lonnberg's series have the relative width of skull at the posterior edge of the meatus auditorius externns more than 64.84 per cent, the highest attained by all but one young from Guadalupe. Specimens from South Georgia and Macquarie Island, measured by the author, and Turner's Heard Island skulls are 63.76, 57.57, 63.65, 64.4, and 61.2, however. The length of palate in the Guadalupe elephant seals varied in relation to the basal length from 40.80 (young) to 52.83 per cent. Even the smallest of Lonnberg's series did not fall below 45.0 per cent, and three exceeded 52.83 per cent. The Macquarie Island animals measured by the writer proved to have a relative palatal length of 53.26 and 56.02, but the one measured by Lydekker was 52.7 per cent. The width of palate in Lonnberg's series varies from 37.1 to 32.2 per cent of basal length, and the same measurement in the Guadalupe specimens is from 32.2 to 28.61 per cent. Skulls measured by Lydekker had a palatal breadth varying from 35 to 39.3 per cent, and 36.12 and 36.42 are the per- centages of the Macquarie Island skulls given in the table above. It might have been supposed that the width of skull at the level of the upper posterior premaxillary suture might bear 236 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. close relationship to the palatal breadth, but this did not mani- fest itself in the measurements. The variations in that dimen- sion in the Guadalupe examples easily fall within the limits of those of the southern seals. In the case of the least frontal width of skull, a decided difference between the northern and southern animals is evi- dent. That measurement in South Georgia and Macquarie skulls ranges from 20.5 to 14.73 per cent of the basal length, with the exception of one of 12.2 per cent, and from 15.1 to 13.6 in Heard Island examples. The percentage in Guadalupe ones is from 14.07 to 11.75. It is unfortunate that the proportionate measurements of the two larger incomplete Guadalupe Island skulls are not available for comparison, as they might have made it possible to attain a fairer estimate of the northern elephant seal. The comparisons would also be of greater value were it known how nearly similar in age were the animals whose skulls were the source of the figures. From a study of the measurements pre- sented in these tables, however, and those recorded by Lydek- ker and Lonnberg, it would appear that, although many of its cranial dimensions fall within the range of variation exhibited by the elephant seal of the southern oceans, the Guadalupe animal possesses a relatively narrower skull. Whether degeneration, due to the near approach of the northern race to extinction, is a factor involved in the reduction in breadth is a debatable point. The extent of variability manifested in the form of the skull and its component parts makes any decision based upon a limited series of slight value. One character believed to be sufficiently constant to separate the northern from the south- ern animals was discovered. In Guadalupe Island examples, it was found that in the dorsal aspect the premaxillae as they extend backward also expand laterally, the lateral outline being distinctly convex in its basal half. The southern specimens examined all appear to have the lateral margins of the pre- maxillae parallel. It would seem, therefore, that there is sufficient justification for regarding the northern elephant seal as a separate species, Macrorhinus angustirostris. Vol. XVIII] DAVIDSON— NORTHERN ELEPHANT SEAL 237 In the examination of the Academy's series of specimens certain other skeletal and anatomical characters have been noted which seem of sufficient value to record. Although it is not so pronounced a feature as in the South Georgia and Macquarie Island skulls, the premaxillary tubercle is present in all the Guadalupe specimens. In this latter series the meseth- moid has never been seen to reach the upper surface of the skull as it does in the southern specimens. The pterygoid processes of the Academy's specimens are inclined to be small and rather slender. The skull of one of the females (No. 1137) has both palatines divided into two parts by a suture. In the skull of the male pup (No. 961), probably only a few weeks old, is seen indications of the cranial element found by Cleland^® in Cystophora cristata and other Pinnipedia, and believed by him to correspond to the paroccipital of Owen in osseous fishes. There is great individual variation in the dentition of the Academy's Guadalupe series, its extent being evident in the following formulae: 2 — 2 C. 2 — 2 I. 1 — 1 2 — 2 2 — 2 I. 2-2 C. — 1 — 1 — 1 — 1 P.M. 4 — 4 4 — 3 P.M. 4 — 4 4 — 4 P.M 4 — 4 4 — 4 P.M 4 — 4 4 — 4 P.M 4 — 4 4 — 4 T) A/r 4 — 4 — re. A. s. M. ; ] No. 962 — [female 1 -; P.M. , 1 1 M. M. M. M, 1 fC. A. S. -; ]No. 1139 — I male 1 fC. A. S. -; Nos. 1136 and 879 — [males — 1 fC. A. S. j Nos. 963 and 961 (pup) — 1 [males — 1 fC. A. S. ■^No. 1137 — 2 [female — 1 M.- — 1 4 — 4 2 — 2 rc. A. S. No. 1138 young female « Rept. Brit. A. A. S., 1902, pp. 646-647. 238 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. The vertebral formula of the Guadalupe specimens appears to be: cervical, 7; dorsal, 15; lumbar, 5; sacral, 3; caudal, 9. The absence of a tenth caudal vertebra may, however, be due to mischance in the preparation of the skeleton. In compari- son with corresponding parts figured by Turner^'^ the spinous process of the cervical vertebrae is much elevated. This character is evident even in the vertebrae of a pup. The hypapophysial tubercle of the atlas is well developed, and the lateral laminae are considerably depressed apically, giving the low^r margin of that vertebra a very sinuous outline. The breadth of the anterior articular surfaces of the axis appears to be proportionately small. The spinous process of the atlas of Nos. 1139, male, and 1137, female, resembles that figured by Turner, but this process in the other males is decidedly broader. The centrum of the third cervical is more nearly oval or elliptical oval. In the seventh vertebra, the transverse processes are not depressed apically as are they in Turner's example. In no case would a straight line drawn between their lowest apices touch the lower margin of the centrum. On the ventral surface of only the anterior and posterior dorsal vertebrae is evidence of a keel discovered. The bodies of the lumbar vertebrae are slightly flattened, or, in some instances, double keeled so that a ventral groove is formed. In No. 1136, male, the epiphyses of the first and second, and the second and third sacrals are anchylosed to one another, but not to the centra. In No. 1137, female, the three sacral vertebrae and anchylosed, and in No. 962, female, four verte- brae in the sacral region are fused. In the Academy's series the first and second caudal vertebrae are possessed of a neural arch. One specimen has the arch present in three, and another specimen has the laminae of the third caudal nearly united to form an arch, and the fourth is very deeply grooved. The scapulae of the Guadalupe seals exhibit considerable variation in form, which is made evident in the following table : " Voy. Challenger, Zoology, XXVI, Seals, pis. II-IV. Vol. XVIII] DAVIDSON— NORTHERN ELEPHANT SEAL 239 Greatest Greatest Percentage of Sex depth width width in depth Females 115.00 122.00 106.08 207.00 192.00 92.75 205.00 205.00 100.00 Males 180.00 173.00 96.11 195.00 210.00 107.69 250.00 265.00 106.00 235.00 240.00 102.12 The skins in the Academy collection were examined and a count of the vibrissae made. The arrangement of the brow bristles differs, but there are usually eight to ten in the group. A single bristle is found on each side of the median line of the head about halfway between the nostril and eye. The mys- tacial bristles are arranged in seven rows, the total number varying from 46 to 49. In this regard there seemed to be such a marked difference between these numbers and those given by Allen^" that the result was verified by count of the papillae on the under surface of the hide. It appears that the number of maxillary bristles of the Guadalupe Island animals is considerably greater than that of the South Georgia ones. Murphy^*^ found that his specimens exhibited 39 maxillary bristles on each side. I am pleased to acknowledge indebtedness to Dr. G. Dallas Hanna, Curator, Department of Paleontology, and Mr. Joseph Mailliard, Curator Emeritus, Department of Ornithology and Mammalogy, for the photographs used in illustrating this paper. "U. S. Geol. Surv., Misc. Pub., XII, p. 743. " Bull. Am. Mus. Nat. Hist., XXXIII, p. 76. 240 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Plate 25 Fig. 1. Northern Elephant Seal, male. Guadalupe Island, Mexico, July 12, 1922. Photograph by G. Dallas Hanna. Fig. 2. Northern Elephant Seal, male. Guadalupe Island, Mexico, July 12, 1922. Photograph by G. Dallas Hanna. PROG. CAL. ACAD. SCI.. 4th Series, Vol. XVIII, No. 9 [ DAVIDSON ] Plate 25 FiQ.l Fiq.Z 2^2 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Plate 26 Fig. 1. Anterior surface of atlas of Macrorhinus angustirostris. Photograph by Joseph MailHard. Fig. 2. Anterior surface of axis of Macrorhinus augusiirostris. Photograpli In- Joseph MailHard. Fig. 3. Anterior surface of third cer\-ical vertebra of Macrorhinus angustiros- Iris. Photograph by Joseph Mailliard. Fig. 4. Anteiior surface of seventh cervical vertebra of Macrorhinus angustir- oslris. Photograph by Joseph Mailliard. PROC. CAL. ACAD. SCI., 4th Series, Vol. XVIII, No. 9 DAVIDSON] Plate 26 Fiq.l Fiq.2 Fig.3 Fig.4 PROCEEDINGS /f OF THE CALIFORNIA ACADEMY OF SCIENCES Fourth Series Vol. XVIII, No. 10, pp. 245-260 April 5, 1929 ON A SMALL COLLECTION OF BIRDS FROM TORRES STRAIT ISLANDS, AND FROM GUA- DALCANAR ISLAND, SOLOMON GROUP BY M. E. McLELLAN DAVIDSON Assistant Curator, Department of Ornithology and Mammalogy In the years 1920 and 1921, Mr. J. August Kusche visited Australia and the Papuan region for the purpose of assembhng general natural history collections. Among the specimens he secured was a small number of birds from Prince of Wales and Thursday islands, Australia, and Guadalcanar Island. Solomon Group. These skins, now in the museum of the California Academy of Sciences, form the basis of the present paper. None of the localities visited was an ornithological terra incognita. Torres Strait islands have been worked on sev- eral occasions by collectors. The bird life of certain of the islands has been quite thoroughly investigated, others still present opportunities for the study of their native fauna, and all are interesting because their position renders them suitable as observatories in the study of the migratory movements of Australian and Papuan birds. Both Prince of Wales and Thursday islands have already been ornithologically explored, but, despite this fact, the present assemblage of specimens includes species apparently not previously reported from Prince of Wales Island. AprU S, 1929 246 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. Prince of Wales Island (10° 40' S., 142° 10' E.) is the largest of the islands of the group bearing that name. It is about 14 miles from the mainland, and covers an area of nearly 12 square miles. Thursday Island (10° 40' S., 142° 20' E.) lies to the northward, and is only about 900 acres in extent. All the islands of the group are hilly, and on them are peaks which rise to an elevation of nearly 700 feet. In former days, Prince of Wales Island supported a native population of about 500 persons, but at the present time the numbers are greatly reduced. The desultory mining operations carried on there have resulted in no extensive settlement as there is on Thurs- day Island, where Port Kennedy has become the metropolis of the Torres Strait pearl fisheries. Lying ofif the eastern coast of New Guinea is the Solomon Group, of which Guadalcanar Island (9° 30' S., 165° E.) is one of the largest and best known. This island is about 82 miles long, and averages a breadth of 25 miles. Lofty forest- clad mountains rise in the eastern and southern portions, Mt. Lammas attaining an elevation of 8005 feet. The descent of these mountains to the sea is abrupt on the east and south, but to the north extend rolling prairies covered with high grass. From May to November the island is swept by the southeast trade wind, and during the period from December until April the inequitable northwest monsoon makes itself felt. These moisture-laden winds result in a coastal rainfall of 100 to 150 inches a year, and a precipitation in the mountains that is said to be between 400 and 500 inches annually. Several navigable rivers, flowing northward, aid in carrying off the surplus water. In spite of the reputed ferocity of the natives, the Solomon Islands have received due attention from naturalists, and the labors of nineteeth century pioneers blazed a trail for more finished work on the part of their successors. In the 20th century. Lord Rothschild's interest in the Papuan islands has resulted in the amassing of excellent collections of birds from the Solomon Islands in Tring Museum, and his studies, and those of Hartert, have aided greatly in giving us a compre- hensive knowledge of their avifauna. Nevertheless, there remain to be learned many facts in variation, distribution, and migration. It is, therefore, to be regretted that the collection Vol. XVIII] DAVIDSON— BIRDS FROM TORRES STRAIT ISLANDS 247 here under consideration lacks detailed information regarding the localities in which the collecting was done. But facts obtained from the specimens themselves seem worthy of presentation. In its entirety, Mr. Kusche's collection comprised 138 bird skins, representing 56 species and subspecies. From July 20 to August 18, inclusive, 1920, examples of 27 species were taken on Prince of Wales Island ; during five days of Septem- ber (7 to 11, inclusive), specimens of eight species were secured on Thursday Island; and between November 26, 1920, and January 30, 1921, representatives of 25 species were added to the collection on Guadalcanar Island. Ten Mega- podins eggs were obtained on Savo Island, off Guadalcanar Island, on February 18, 1921. List of Species 1. Megapodius reinwardt brenchleyi Gray Nos. 2996-3005: eggs, February 18; Savo Island. The incubation of one of these eggs had begun, the remain- der were fresh. The ground color of these eggs varies from almost pure white to light buff (Ridgway). The overlying color ranges from pinkish buff, through a pale Isabella, to avellaneous. Oates^ gives 2.8 (71.2 mm.) and 3.05 (77.5 mm.) inches in length, and 1.75 (44.5 mm.) and 1.9 (48.2 mm.) inches in breadth as the extreme measurements of his series. The extremes exhibited by the Academy's series are: 73.7 mm. and 81.0 mm. in length, and 45.2 mm. and 49.1 mm. in breadth. 2. Ptilinopus regina Swainson Nos. 24410-11: female, August 16; male, August 17; Prince of Wales Island. Both birds are in fresh plumage. The coloration of No. 24410, female, is very intense for one of that sex. On the under tail-coverts of this example, a bar of orange-red inter- venes between the narrow yellow tip of the feather and a >Cat. Birds* Eggs Brit. Mus., I, 1901, p. 16. 248 CALIFORNIA ACADEMY OF SCIENCES [Pkoc. 4th Ser. central band of magenta. The presence of this last color is apparently unusual, and it is not mentioned in the descriptions of the species given by Salvadori^ or Mathews.^ 3. Ptilinopus superbus (Temminck) No. 23312: male, August 16, 1920; Prince of Wales Island. Fresh plumage has just been assumed. "Eyes red-brown. Legs and feet blood-red. Call, a low iiuiii" (Kusche). 4. Jotreron viridis lewisi (Ramsay) No. 22425: female, January 18; Guadalcanar Island. "Bill orange-yellow. Legs and feet crimson" (Kusche). The feathers of the forehead and chin are gray tipped with green.* 5. Megaloprepia magnifica assimilis (Gould) Nos. 24407-08: male, August 18; female, August 17; Prince of Wales Island. No, 24409: male, September 10; Thursday Island. The specimens at hand differ from the Megaloprepia assimi- lis [=M. m. keri Mathews) figured by Mathews^ in having the fresh feathers of the upper parts a more golden green with bronze reflections. The breast feathers are apically Indian purple (Ridgway), and a subterminal band of dark madder violet (Ridgway) intervenes between that color and the suc- ceeding green area. The measurements (in millimeters) of the series at hand are as follows: Culmen, 15.5, 14.5, 16.5; wing, 180.0, 181.0, 183.0; tail, 148.0, 156.0, 145.0; tarsus, 23.5, 22.0, 22.5. In size these birds appear to approach M. m. poliura, but differ from individuals of that race in having the under tail-coverts washed with gamboge. » Cat. Birds Brit. Mus., XXI, 1893, p. 95. »B. Austr., I, 1910-11, pp. lOS, 107. *C/. Salvadori, Cat. Birds Brit. Mus., XXI, 1893, p. 153; Hartert, Nov. Zool., II, 1895, p. 63, footnote; and Rothschild & Hartert, Nov. Zool., VIII, 1901, p. 109. " B. Austr., I, pi. 26. Vol. XVIII] DAVIDSON— BIRDS FROM TORRES STRAIT ISLANDS £49 According to a note on the label, the oviduct of the female contained two ova. 6. Globicera rufigula (Salvadori) No. 24424 : male, December 3 ; Guadalcanar Island. A moult, involving- contour plumage as well as remiges and rectrices, is nearly complete. 7. Chrysauchoena humeralis humeralis (Temminck) No. 24414: female, July 28; Prince of Wales Island. 8. Geopelia placida placida Gould No. 24413: male, August 3; Prince of Wales Island. 9. Caloenas nicobarica nicobarica Linnaeus Nos. 24445-48: females, January 2 and 9; males, January 17 and 27; Guadalcanar Island. 10. Porphyrio indicus neobritannicus Meyer No. 24422 : male, December 2 ; Guadalcanar Island. In the specimen under examination the foreneck and breast are greenish cobalt, in distinct contrast to the remainder of the under parts. The thighs and abdomen are concolor. For birds exhibiting these characters, Hartert** has presented cogent arguments for the use of the specific name indicus Horsfield, rather than calvus Vieillot. But, unless it proves that repre- sentatives of two species of Porphyrio are resident in the Solomon Islands, it will be necessary to regard neobritannicus as a race of indicus, not of melanotus.'' •Nov. Zool., XXXI, 1924, pp. lOS-106. ^ Cf. Hartert, Nov. Zool., XXXI, 1924, p. 108; Mathews, Syst. Av. Austr., 1927, p. 101. 250 CALIFORNIA ACADEl^y QF SCIENCES [P»oc. 4th Sek. 11. Pluvialis dominicus fulvus (Gmelin) No. 24442 : female, January 1 1 ; Guadalcanar Island. The remiges are only slightly worn, but the remainder of the plumage is much abraded. 12. Actitis hypoleucus (Linnaeus) No. 24894: female, August 17; Prince of Wales Island. This is a bird in worn garb. 13. Orthorhamphus magnirostris neglectus (Mathews) No. 24423: male, January 17; Guadalcanar Island. 14. Demigretta sacra novaeguineae (Gmelin) Nos. 24416-17: male, January 12; female, January 30; Guadalcanar Island. No. 24418: specimen without data. The subspecific name has been but tentatively applied to these specimens. The female is almost pure white. A few dark streaks appear in the contour plumage, and dark tips are in evidence on some of the remiges and rectrices. The plum- age of the male is devoid of white, and the dataless example has a white line on the throat. The specimens yield the following measurements (in millimeters) : Culmen, 86.0, 85.0, 85.0; wing, 295.0, 265.0 (worn), — ; tail, 98.5, 90.0, — ; tarsus, 75.7, 70.5, 74.0. In size these specimens approach a female collected at Apia, Samoa, which measures : Culmen, 84.5; wing, 265.0 (worn) ; tail, 94.0; tarsus, 70.0. 15. Nycticorax caledonicus hilli Mathews No. 24415 : male, September 9; Thursday Island. This bird appears to be much paler than that figured by Mathews.® The mantle is fawn, but approaches mars brown (Ridgway) on the interscapulars. » B. Austr., in, pi. 193. Vol. XVIII] DAVIDSON— BIRDS FROM TORRES STRAIT ISLANDS 251 16. Anas superciliosa pelewensis Hartlaub & Finsch No. 2442 1 : male, November 26 ; Guadalcanar Island. 17. Leucospiza hiogaster pulchella (Ramsay) Nos. 24405-06, 24436-38: immature males, December 26 and 30; adult male, January 18; immature female, January 5; adult female, January 23 ; Guadalcanar Island. The adult female is in greatly worn dress. 18. Haliastur indus ambiguus Briiggemann Nos. 24419-20: female, December 22; male, December 23; Guadalcanar Island. "Iris brown. Bill ochre. Legs yellow. The stomach of the male contained a bird" (Kusche), 19. Eos grayi Mathews & Iredale Nos. 24364-68 : males, December 4 and 28, and January 8 ; female, January 8 ; Guadalcanar Island. A moult which involves all the feather tracts is in evidence in these examples. 20. Trichoglossus haematodus aberrans Reichenow Nos. 24361-63: female, December 4; males, December 4 and January 8; Guadalcanar Island. These specimens have the occiput purplish brown and the throat purple. In two individuals the unspotted dark green area of the central abdomen is quite evident. The plumage in every case is much worn, but feather renewal has begun on the forehead, crown, and flight feathers. 21. Kakatoe galerita fitzroyi (Mathews) Nos. 24371-73 : female, August 7; males, August 8; Prince of Wales Island. The auriculars of these examples are strongly tinged with yellow, in this respect differing from the type of fitzroyi as 252 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. described by Mathews." The measurements (in milUmeters) of the series are: Cuhnen, 37,0, 37.2, 42,0; wing, 305.0, 298.0, 316.0; tail, 159.0, 164.0, 175.0; tarsus, 24,0, 24.5, 25.0. 22. Ducorpsius ducorpsii Pucheran Nos. 24369-70 : female, January 5 ; male, January 6 ; Guadalcanar Island. The basal reddish orange of the feathers of the lores, sides of head, nape, throat, breast, flanks, and upper tail-coverts is quite conspicuous in the case of the male. The presence of this color is mentioned by Finsch,^° but it is ordinarily dis- regarded in descriptions of this species. 23. Lorius roratus solomonensis (Rothschild & Hartert) No. 24360 : female, January 22 ; Guadalcanar Island. The contour plumage has been recently assumed, but all but one rectrix and the outer primaries have still to be replaced. 24. Megapodargus papuensis baileyi (Mathews) No. 24316: male, July 28; Prince of Wales Island. 25. Eurystomus orientalis solomonensis Sharpe Nos. 24312-15: males, December 20 and 21; female, December 22 ; Guadalcanar Island. A white patch is present on the chin of each of the speci- mens, "Iris ruby red. Bill and legs vermilion" (Kusche). 26, Dacelo leachii kempi Mathews Nos. 24303-08: adult male, August 4; males [adult females], July 27; females [immature males], July 29 and August 4 ; Prince of Wales Island, »Nov. Zool., XVIII, 1911, p. 264. "Papag., I, 1867, p. 312. Vol. XVIII] DAVIDSON— BIRDS FROM TORRES STRAIT ISLANDS 253 An immature male (labeled "female") taken on August 4, has the tail basally dark blue, and only slight indications of brown are present on the outer webs of the lateral rectrices. A somewhat younger male, labeled "female," has the proxinal two-thirds of the tail dark blue, and the distal portion brown banded with blue. The rectrices of this specimen are quite narrow, the central ones measuring six and the lateral ones four millimeters less than the corresponding feathers of the adult bird. Two birds taken on July 27 are marked as "males" ; but, although they appear to be older than the imma- ture male dated July 29, there is no indication of an advance into the plumage of the adult male, the tails being brown to the base. In these examples the white of the throat merges into light cream buff (Ridgway) on the breast and abdomen. The amount of dark vermiculation on the under surface is vari- able. One adult female is very heavily marked, but the lower parts of the adult male are very faintly lined. 27. Lazulena macleayii macleayii (Jardine & Selby) No. 24310: female, July 29; Prince of Wales Island. 28. Sauropatis sancta confusa (Mathews) No. 24309: female, August 4; Prince of Wales Island. No. 24308 : male, September 7 ; Thursday Island. The female (apparently immature) has feather renewal in progress on the occiput, cervix, and entire under parts. 29. Sauropatis chloris alberti Rothschild & Hartert No. 24311 : female, December 4; Guadalcanar Island. This individual has the pale occipital spot, said to charac- terize alberti,^^ but there is evidence of the very narrow super- ciliary line to be found in "perplexa" (= Sauropatis chloris solomonis)}^ " Rothschild & Hartert, Nov. Zool., XV, 1908, p. 361. " Rothschild & Hartert, Nov. Zool., XV, 1908, p. 361. April 5, 1929 254 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 30. Rhytoceros plicatus mendanae Hartert Nos. 24431-35: male [not sexed], December 3; male, December 25; males [females], January 3 and 17; female [immature male], December 25; Guadalcanar Island. Two black-headed individuals collected on January 3 and 17 are clearly females, although they are labeled "males." An immature bird, with an ochraceous-tawny head, is marked "female." This example is just acquiring adult plumage, and new feathers are appearing in the contour plumage of the other birds. "Iris yellowish brown. Base of bill dull crimson. Skin below bill light blue" (Kusche). 31. Cosmaerops ornatus ornatus (Latham) Nos. 24339-44: males, July 25 and 30, August 2 and 14; female, August 4 ; Prince of Wales Island. A male taken on August 14 is undergoing extensive replace- ment of the body feathers. The remainder of the specimens appear to be in unworn garb. A few undeveloped feathers are in evidence on chin and throat. 32. Lamprococcyx russatus (Gould) Nos. 24355-56: males, July 25 and 29; Prince of Wales Island. According to the collector, the iris of the bird taken on July 25 was red, that of the one secured on July 29, dark brown. The assumption of new plumage is just begun on the head and throat. 33. Polophilus phasianinus melanurus (Gould) No. 24322 : male, July 27 ; Prince of Wales Island. This is an individual in striped plumage. "Iris black. Upper mandible straw yellow, lower mandible white. Legs and feet blue-gray" (Kusche). VpL. XVIII] DAVIDSON— BIRDS FROM TORRES STRAIT ISLANDS 255 34. Nesocentor milo milo (Gould) Nos. 24426-30: males, December 21, January 10 and 11; females, January 2 and 12; Guadalcanar Island. Fresh flight feathers are being acquired by one of the males and one female. A male in its first contour plumage has a few pin feathers still present on the rump and abdomen. 35. Kempia flavigaster terraereginae (Mathews) No. 24452: male, August 10; Prince of Wales Island. 36. Pachycephala astrolabi Bonaparte Nos. 24337-38, 24455 : males and immature female, November 30; Guadalcanar Island. The immature female has the upper surface, including the tail, bright yellowish olive, the head and interscapulars being strongly washed and pied with kaiser brown. There are faint indications of a yellow cervical collar. The inner webs of the lateral rectrices are narrowly margined with pale cinnamon buff. Externally the wing is kaiser brown, and the throat, forebreast, and auriculars are washed with the same shade. The under tail-coverts are lemon yellow, which, also, suf- fuses the lower breast and abdomen. The lower surface is obsoletely streaked by the dusky shaft stripes of the feathers. 37. Leucocirca leucophrys (Latham) No. 24451 : male, January 7; Guadalcanar Island. 38. Mastersornis rubecula yorki (Mathews) No. 24391 : male, August 10; Prince of Wales Island. This individual is an immature bird in the garb of a female. The plumage is greatly worn, and no fresh feathers are in evidence. 256 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Sep.. 39. Graucalus novaehollandiae connectens (Mathews) Nos. 24329-31: male, July 30; females, August 1 and 2; Prince of Wales Island. No. 24332 : female, September 7 ; Thursday Island, The black is wanting from the forehead of the male, and the throats of all the examples are freckled. The plumage in every case is greatly abraded, but new rectrices being devel- oped by the female taken on September 7 are the only evi- dence of feather replacement. In view of Campbell's^^ record of an unusual specimen from the Torres Strait islands, the measurements of this series may be of interest. Culmen, 26.5, 26.0, 26.0, 28.0; wing, 190.0. 192.0, 192.0, 193.0, 189.0; tail, 130.0, 143.5, 135.0, 138.0; tarsus, 28.0, 30.0, 28.5, 28.0. 40. Graucalus hypoleucus stalked (Mathews) Nos. 24333-36 : males, July 27 and August 1 ; females, July 27 and 29 ; Prince of Wales Island. The characters of hypoleucus, rather than those of papueyisis,^* appear to be exhibited by this series. In every case the throat is white, and the secondaries exhibit distinct white margins. A plumage renewal, which has already affected the secondaries, lateral rectrices, and throats of three of the specimens, is in progress. 41. Karua leucomela yorki (Mathews) Nos. 24357-59, 24392: female [male], July 29; females, August 2, 3, and 14 ; Prince of Wales Island. A specimen taken on July 29 wears the dress of the male although it is marked "female" by the collector. The under tail-coverts appear to be rather pale in this series, but the ver- miculations on the under surface of the females are quite pro- nounced. The inner secondaries have been recently acquired in all the examples, and a few pin feathers are present on the throat of the female taken on August 2. "Emu, XX, 1920-21, p. 61. "CA Hartert, Nov. Zoo!., XII, 1905, p. 224; Ogilvie-Grant, Ibis, 1915, Jubilee Suppl., No. 2, p. 128; Campbell, Emu, XX, 1920-21, p. 61; and Mathews, B. Austr., IX, 1921-22, p. 126. Vol. XVIII] DAVIDSON— BIRDS FROM TORRES STRAIT ISLANDS 257 42. Edoliisoma erythropygium erythropygium Sharpe No. 24456: male [female], December 31; Guadalcanar Island. This specimen, in newly acquired contour plumage, although marked "male," seems to possess the characters ascribed to the adult female of this species. 43. Sphecotheres flaviventris flaviventris Gould Nos. 24345-51, 24353: males, August 15, 16, 17, and 18; female [immature male], August 16; Prince of Wales Island. No. 24352: male, September 9; Thursday Island. The immature (marked "female") is just passing into the plumage of the adult male. A moult involving all the feather tracts is well advanced in two instances ; in others, the renewal of the plumage has only barely begun. 44. Artamus leucorhynchus leucopygialis Gould No. 24354 : male, July 25 ; Prince of Wales Island. ' 45. Caleya megarhyncha griseata (Mathews) Nos. 24453-54 : male, September 7 ; female, September 9 ; Thursday Island. 46. Microchelidon hirundinacea yorki (Mathews) Nos. 24381-82: males, August 12 and 17; Prince of Wales Island. Both individuals are possessed of the short tail supposed to characterize this race. This feature measures 26.5 and 26.2 mm. in the specimens in hand. Mathews^^ gives 28.0 mm. as the tail length of the type of this subspecies and 32.0 mm. for hirundinacea. "Nov. Zool., XVIII, 1911, p. 387. '^.^ 258 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ses. 47. Cyrtostomus frenatus australis (Gould) Nos. 24374-79: males, July 25, 26, and 29; female, July 25 ; Prince of Wales Island. The adult males of this series have the breast and abdomen light cadmium (Ridgway) rather than lemon yellow as has macgillivrayi. The bills measure 20 and 21 mm. An imma- ture male, collected on July 26, has metallic feathers appear- ing on the throat. The renewal of the body plumage of the remainder of the specimens is well advanced. 48. Cyrtostomus frenatus flavigastra (Gould) No. 24380 : male, January 7 ; Guadalcanar Island. This example is an immature individual. A renewal of the flight feathers, as well as of the contour plumage, is in progress. 49. Myzomela erythrocephala kempi Mathews Nos. 24320-21 : males, September 11 ; Thursday Island. 50. Melomyza obscura munna (Mathews) Nos. 24317-19: females, July 26 and 27, and August 9; Prince of Wales Island. The measurements of this series (in millimeters) are: Culmen, 19.0, 18.5, 18.5; wing, 68.0, 70.5, 68.0; tail, 52.0, 55.0, 54.0; tarsus, 18.5, 18.1, 18.0. For a female from Cape York, Mathews^^ gives the following measurements : Culmen, 15.0; wing, 60.0; tail, 45.0; tarsus, 18.0. The type of apsleyi, a male, from Melville Island measures: Culmen, 18.0; wing, 72.0; tail, 55 ; tarsus, 19.0. It would seem, therefore, that the birds from Prince of Wales Island approach more nearly in size those from Melville Island than they do those from the adjacent mainland. The feather replacement of the specimen taken on August 9 is nearly complete, and it is well under way in the other individuals. "B. Austr., XI, 1923-24, p. 331. Vol. XVIII] DAVIDSON— BIRDS FROM TORRES STRAIT ISLANDS 259 51. Dorothina lewinii ivi (Mathews) Nos. 24439-41 : males, August 10 and 18; female, August 12; Prince of Wales Island. 52. Neophilemon orientalis yorki (Mathews) Nos. 24323-28 : males, July 24, 26, and 27 ; females, July 20 and August 1 ; Prince of Wales Island. The collector has indicated that in two females the irides were "blood-red," and in one male that they were "gray." 53. Mimeta sagittata subaffinis (Mathews) Nos. 24393-95 : male, July 26; females, August 10 and 13 ; Prince of Wales Island. The extent of the white on the rectrices of these examples varies considerably. The measurements yielded are : Culmen, 30.0, 30.0, 31.5; wing, 145.0, 142.0, 144.0; tail, 107.0, 104.0. 103.0; tarsus, 25.0, 22.5, 24.5. 54. Neomimeta flavocincta kingi (Mathews) Nos. 24396-97: male and female, September 10; Thursday Island. The contour plumage of both specimens is fresh, but the replacement of flight feathers has only commenced. 55. Acridotheres tristis tristis (Linnaeus) No. 24402 : male, December 7 ; Guadalcanar Island. 56. Lamprocorax cantoroides cantoroides (Gray) No. 24443 : male, November 30 ; Guadalcanar Island. "Iris orange" (Kusche). 260 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4ih Ser. 57. Metallopsar metallicus nitidus (Gray) No. 24444: immature male, November 30; Guadalcanar Island. A single individual, in its first contour feathers, appears to belong under this head. The pileum is dark brown slightly glossed with purple. The dark brown feathers of the cervix are margined with a paler shade, producing an ill-defined, striped collar. The throat is narrowly, and the remainder of the under surface broadly, streaked with blackish brown on a white or buffy white ground. The flanks are dark brown. PROCEEDINGS OF THE / Cfcj ' CALIFORNIA ACADEMY OF SCIENCES \^^ /> Fourth Series x^^ Vol. XVIII, No. 11, pp. 261-265 April 5, 1929 XI THE GENERIC RELATIONSHIPS AND NOMENCLA- TURE OF THE CALIFORNIA SARDINE BY CARL L. HUBBS Museum of Zoology, University of Michigan Confusion has long obtained and still prevails regarding the generic relationships and nomenclature of the California sar- dine. The earlier history involved is of no distinct pertinence to the present discussion, and will not now be recounted. We shall pick up the story with Regan's 1916 contribution.^ In that paper, Regan referred the California sardine, as well as the related or identical species of Chile, Japan, Australia and South Africa, to the European genus Sardlna. Shortly thereafter, Jordan," apparently on the advice of Scale, synonymized Sardina Antipa, 1906, with Sardinia Poey, 1858. He did so because Scale had located, in the collections of the Museum of Comparative Zoology, a specimen thought to be the type of Poey's species, Sardinia pseudo-hispanica, and showing the generic characters assigned by Regan to Sardina. More recently, Thompson" pointed out a number of trenchant characters, more or less overlooked before, which ^Ann. and Mag. Nat. Hist., Ser. 8, 18, 1916, 11. 2 Copeia, 56, 1918, 46 (see also. The Genera of Fishes, Stanford Univ. Publ., Univ. Sen, pt. 3, 1919, 299, and pt. 4, 1920, 512). = Fish and Game Comm. Calif., Fish Bull., No. 11, 1926, 8-17. April S, 1929 262 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. serve to distinguish the sardines of California and Chile from those of Europe. The differences which he noted are as fol- lows : ( 1 ) in the American species there is usually a row of dark blotches behind the head, typically not apparent in the European; (2) the scales, as Regan had already observed, are arranged in a very different and in a regular order, each alter- nate row not being nearly overlapped by the one in front (the apparent number of rows, therefore, is equal to, instead of being about half as numerous as, the true number) ; (3) the ventral scutes are weaker and less keeled, and have less expanded bases; (4) the gillrakers on the lower limb, unlike those of the European sardine, become gradually and mark- edly shortened toward the angle of the arch, and they differ markedly in number at comparable sizes ; (5) the interopercle is more expanded and widely exposed behind the preopercle; and (6) the opercular ridges (and preopercular edge) are strongly oblique instead of being nearly vertical. All of these points I have completely verified. Other differences, pointed out by Thompson, involving the proportionate sizes of the parts or the position of the fins, appear less trenchant and need not be now considered. One point not specified by Thompson, nor by Regan, is that the gillrakers of the upper limb fold down over those of the lower limb near the angle, whereas they do not do so in the European species. This very character Regan* elsewhere used in the primary separation of the genera of one division of the family. Another difference in gillraker structure, equally trenchant, has just been discovered by Dr. Henry B. Bigelow, who has kindly allowed me permission to announce the interesting dis- covery. In the European sardines ( pilchardus and sardina) we find that the minute processes on the gillrakers are simple, slightly-bent, sharply pointed spines, about one-third as long as the width of the gillrakers and spaced about three in a dis- tance equal to this width. In the Californian species, and I find this equally true of the Chilean, Japanese and Australian fonns, these processes are complex, for they are composed of a flask-shaped base or stalk and a distinct, fimbriate, grooved, leaf-like terminal element. The processes are nearly half, ■•Ann. and Mag. Nat. Hist., Ser. 8, 19, 1917, 297-298. \-0L. XVIII] HUBBS— CALIFORNIA SARDINE 263 sometimes more than half, as long as the gillrakers are wide, and are more crowded, as about five occur in a space equal to this width. The appearance of the gillrakers of Californian and European sardines, under a microscope, is strikingly unlike. The complex structure and greater length and crowd- ing of these gillraker processes, as well as the longer and more numerous gillrakers, and their overfolding in the Californian and related sardines, provide a straining apparatus much finer than that possessed by the European species. This may per- haps be correlated with their living in seas in which diatoms are relatively more abundant, and crustaceans scarcer, than in European waters. Even without recourse to the "splitting" tendencies of the day, it appears necessary to divorce generically the Californian and European sardines. Their differences, particularly in scale arrangement and in gillraker structure, are too funda- mental and too trenchant to permit of their continued alloca- tion in a single genus. The question of their immediate com- mon origin is even thrown open to some doubt. The generic separation of the Californian and European sardines reopens of course the problem of the proper generic name for each. It is necessary first to consider Poey's Sar- dinia pseudo-hispanica. The specimen so labelled in the Mu- seum of Comparative Zoology, and stated to be Poey's type in Jordan's note, I have fortunately been able to reexamine. It certainly is not the type, for it is decidedly smaller than the one specimen described by Poey. Furthermore, it is not even conspecific, for it has 51 vertebrae, including the hypural, whereas Poey gives 46 as the number for pseudo-hispanica. In other respects, for instance, the lower number of dorsal rays, this alleged type fails to meet Poey's description. The specimen is probably a mislabelled example of the California sardine; at least it belongs to the same genus, for it agrees with it in every one of the characters listed above as dis- tinguishing the Californian from the European species. A main reason for thinking that the specimen in question did not even come from Cuba is that there appears to be no other indication whatever of the occurrence of a sardine of either the Californian or the European type anywhere in the western Atlantic. 264 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. It is clear from Poey's description that his Sardinia pseudo- hispanica is not closely related to either the Californian or European sardine. There is very good reason to believe that he had the common West Indian species, Sardinella anchozna Cuvier & Valenciennes, 1847, which in turn is thought by Regan"' to be identical with the European Sardinella aurita Cuvier & Valenciennes, the type-species of Sardinella. We find, for instance, that the number of vertebrae in anchovia is 46, just as in Poey's type of pseiido-hispanica. Jordan and Evermann's® Clupanodon pseudohispaiiicus is apparently the same species as their Sardinella anchovia. It is therefore impossible to refer either the Californian or the European sardine to the genus Sardinia Poey, 1858. That name should, I think, be synonymized with Sardinella Cuvier & Valenciennes, 1847. The generic name Sardina Antipa, 1906, therefore becomes available for the European species, which with Regan we may call Sardina pilchardiis (Walbaum). No generic name, how- ever, appears to be available for the California sardine. I now supply this obvious need : Sardinops Hubbs, new genus Type-species, Maletta ccerulea Girard, 1854. Diagnosis. Clupeidse with the upper jaw not notably notched on the mid-line ; the gillrakers of the upper limb folded over those of the lower limb, which become markedly and progressively shortened toward the angle; carina of glosso- hyal not denticulate;^ no bilobed dermal flap on shoulder- girdle; opercle with strong and markedly oblique ridges; pre- opercular edge strongly sloping; interopercle widely exposed behind preopercle ; scale-rows regularly spaced, the lateral scales all with subequal exposed areas; radii on the scales nearly vertical, and paired on each side of median line; keels on ventral scutes weak; last two rays of dorsal and anal fins somewhat enlarged ; a row of dark spots typically developed on upper sides behind head. =Ann. and Mag. Nat. Hist., Ser. 8, 19, 1917, 378. 'Bull. U. S. Nat. Mus., 47, pt. 1, 1896, 423 and 429. 'See Chabanaud, Bull. Soc. Zool. Fr., SI, 1926, 156-163. Vol. XVIII] HUBBS—CALIFORNIA SARDINE 265 Examples of the pilchards or sardines of Chile, Japan and Australia all agree fully with this generic diagnosis, and are clearly congeneric with Sardinops ccsriilea (Girard), as prob- ably is also the South African species ocellata, which is known to share most of the characters listed above in common with ccendea. It is, in fact, not clear whether the species of these various regions are different from one another. Pending a much needed critical comparison of good material from all these localities, I merely list the species as usually recognized : 1. Sardinops ccerulea (Girard), 1854. Californian. 2. Sardinops sagax (Jenyns), 1842. Chilean. 3. Sardinops melanosticta (Temminck & Schlegel), 1846. Japanese. 4. Sardinops neopilchardtis (Steindachner), 1879. Australian. 5. Sardinops ocellata (Pappe), 1853. South African. The distinctness of Sardinops cccrulea is particularly doubt- ful, especially since Thompson (I. c.) was unable to diflferen- tiate it specifically from 5. sagax. PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES Fourth Series Vol. XVIII, No. 12, pp. 267-383, plates 27-32, 7 text figures April 26, 1929 XII THE FAUNAL AREAS OF SOUTHERN ARIZONA: A STUDY IN ANIMAL DISTRIBUTION BY HARRY S. SWARTH Curator, Department of Ornithology and Mammalogy Introduction During the summer of 1927 the Department of Orni- thology and Mammalogy of the California Academy of Sciences conducted three field trips to southeastern Ari- zona. The region visited comprised the lowlands surround- ing the Santa Rita Mountains, from 30 to 60 miles south- east of Tucson and a short distance north of the United States-Mexico boundary line. Personnel and itineraries of the three parties were as follows: H. S. Swarth and Joseph MaiUiard, with Raymond Gilmore as assistant, left San Francisco by automobile on May 6, arriving at Pata- gonia, Santa Cruz County, Arizona, on May 10. There we were joined by David M. Gorsuch, who remained with us throughout our stay, as a volunteer aid. With Patagonia as a center, collecting was carried on along the eastern base of the Santa Rita Mountains and some distance to the east- ward, from May 10 to June 2. Camp was then shifted to the western base of the Santa Ritas, near the Florida Ranger Station, at the mouth of Stone Cabin Canon, where we remained from June 2 to 21. Return to San Francisco was accomplished on June 25. April 26, 1929 268 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Sek. Joseph Mailliard, with Floyd C. Rankin as assistant, left San Francisco by automobile on August 23 and arrived at Patagonia on August 27. They left Patagonia on October 13, reaching home on October 17. Miss Mary E. McLellan, travelling by train from San Francisco to Tucson, collected in Madera Canon, on the west side of the Santa Rita Moun- tains, September 3 to October 13. Mr. Sam Davidson was a volunteer aid in collecting mammals during part of that time. Specimens collected upon all three trips include 1127 birds and 423 mammals. For necessary permits to carry on the collecting of the above mentioned material we were indebted to the courtesy of the Arizona Fish and Game Department, through Mr. D. E. Pettis, State Game Warden. We are also under great obligations to Mr. Marshall Ashburn for permission to camp upon and to hunt over the extensive Ashburn Ranch (formerly the Pennsylvania Ranch) in the Sonoita Valley. In pursuing the study of this collection I have found it necessary to call upon various institutions and individuals for the loan of specimens and for information, all of which was most generously granted. I am under obligations for such help to Dr. Alexander Wetmore, Assistant Secretary of the Smithsonian Institution, who authorized the loan to me of numerous specimens from the collection of the United States National Museum, including the type of Agelaius phceniceus sonoriensis; to Dr. Charles W. Richmond for ad- vice upon various subjects and for specific information re- garding the above mentioned type specimen; and to Mr. Gerrit S. Miller, Jr., for identification of the specimens of Myotis we collected. To Mr. Paul G. Redington, Chief of the Bureau of Biological Survey, I am indebted for the loan of specimens and for permission to use unpublished data from the files of the Survey bearing upon the distribution of certain species of Citellus and Ammospermophilus in Ari- zona; and to Major E. A. Goldman, of the same Bureau, I am indebted for the identification of specimens of Perogna- thus, Dipodomys, and Sigmodon, and for information re- garding other species. From the Museum of Vertebrate Zoology of the University of California, through Dr. J. Grinnell, Director, I received the loan of specimens when- ever they were desired, and facilities for working at the Vol. XVIII] SWARTH—FAUNAL AREAS OF SOUTHERN ARIZONA 269 Museum whenever I chose to do so. From the Museum of Comparative Zoology, through Mr. Outram Bangs, I was permitted to borrow a series of skins of Sayornis nigricans. From the Field Museum of Natural History, and from the Museum of Leland Stanford, Jr., University, I also received the loan of specimens. From Dr. L. B. Bishop I received the loan of specimens, including an important series of Agelaius, and data upon many Arizona-taken bird skins in his collection. The half-tones illustrating this report are all from photographs taken by Mr. Joseph Mailliard. Mrs. Mary McLellan Davidson, Assistant Curator of the De- partment of Ornithology in this institution, drew the distri- bution maps and rendered important help also in other ways. In the following accounts of the species of birds and mam- mals collected I have for the most part limited my remarks to statements bearing upon distribution. Facts pertaining to nesting or other activities have been omitted in most cases where the species concerned is more or less well known. They have been included in a few cases where it seemed worth while, and, also, data pertaining to migration and molt in birds have been briefly presented, in the belief that these facts were worth placing upon record. The Region Visited and the Problem Involved Our field work in southeastern Arizona was primarily for the purpose of studying the local distribution of animal life in the section visited, which comprised the lowlands sur- rounding the Santa Rita Mountains. Years ago the writer had collected birds extensively and mammals in lesser num- bers in that general region, and he had been struck by cer- tain outstanding features in the delimitation of species there. The opportunity now presented itself of acquiring further data on the subject, and field work was pushed ac- cordingly with the object of gathering specimens and infor- mation that would bear upon the distribution of lowland forms. The several mountain ranges of southern Arizona rise much like islands from a surrounding sea of plains. Their bird and mammal faunas are peculiar and are sharply differentiated from those of the surrounding lowlands, but 270 CALIFORNIA ACADEMY OF SCIENCES [ Pboc. 4th Sbb. Fig. A. Map of southern Arizona, showing region studied and localities visited by the California Academy of Sciences expeditions of 1927. Broken lines indicate approximate boundaries of Western Desert Area and Eastern Plains Area. they are quite well known and in any event have no bearing upon the peculiar differentiation of faunas that distin- guishes different lowland areas. So, while the Santa Rita Mountains, as a conspicuous boundary line between two lowland differentiation areas, formed a center for our field work, and were even, perforce (through lack of camping fa- cilities elsewhere), the site of our base camps for work on their west side, little attention was paid to the typically high zone species of birds and mammals, and only one or two brief trips were made to high altitudes. In southern Arizona, from the Colorado River on the west, east to the Santa Rita Mountains, the general appear- ance of the lowlands is everywhere about the same. Except for limited areas along the river bottoms it is desert of the most arid type, covered with a fairly dense growth of desert plants, a chaparral composed of many different shrubs, bushes and cactuses. This chaparral, as in desert regions elsewhere, is in the shape of isolated clumps of vegetation of Vol. XVIII 1 SWARTH—FAVNAL AREAS OF SOUTHERN ARIZONA 271 Fig. B. Map showing distribution in southern Arizona oiLophortyx gambelii gainbelii and Callipepla squamata pallida. Symbols indicate published record stations; broken line indicates approximate northern and southeastern boun- daries of L. g. gambelii; solid line indicates approximate northern and western boundary of C. s. pallida. greater or less extent, separated by areas of bare ground. Cactus of several species are important plants, there being thickets of low-grovi^ing cholla almost everywhere, and in places scattered individuals or extensive "forests" of the tall and conspicuous giant cactus. The cactus plants are an im- portant factor in the economics of birds and mammals, so much so that the very existence of several bird species in a region is dependent upon the presence of the giant cactus. The few river beds are marked by rows of tall cottonwoods, with a lesser growth of willows and arrow- weed, the latter sometimes forming dense jungles of considerable extent. Mesquite, catclaw, ocotilla and the creosote bush are all present in abundance, and each occurs in almost pure stands over large areas, and there are many other species of trees and bushes that enter into the composition of the plant covering of this area. It is desert, but well covered with shrubby or tree-like vegetation. There is relatively little grass anywhere. 272 CALIFORNIA ACADEMY OF SCIENCES [ Pnoc. 4th Ser. ..<:^s \ VV''ii,' '^■ %: t'. RN DESERT AREA Wl ^\ 1 .«v--"\ X^ ,0 ^ n O Corvus cryptoleucns + Colaptes c. mearnsi I'H, CiujUi Fig. C. Map showing distribution in southern Aiizona of Colaptes chrysoides mearnsi and Corvus cryptoleucns. S>Tnbols indicate published record stations; broken line indicates approximate northern and eastern boundaries of Colaptes c. mearnsi; solid line indicates approximate northern and western limits of Corims cryptoleucus. East of the Santa Rita Mountains is an entirely different sort of region, and the transition from one to the other is abrupt. Desert chaparral is there replaced by grassy plains. In some rocky foothill sections there may be found small tracts of "brush" or a few scattered cholla cactuses, and in places there are extensive stands of creosote, but for the most part there are illimitable stretches of rolling hills or gently sloping plains covered with grass and with almost nothing else. In some low-lying swales the shorter prairie ("grama") grass is replaced by growths of "sacaton," a coarse bunch grass eight or ten feet high. In parts of the foothill country tree yuccas form the most conspicuous plant growth, and there are places on the grassy plains where small mesquites cover many miles, spaced so regu- larly and so uniformly of a size as to give the impression of a young peach orchard. In the western desert area the elevation of the lowlands Vol,. XVIII] SWARTH—FAVNAL AREAS OF SOUTHERN ARIZONA 273 rises from a little less than 100 feet above sea level on the lower Colorado River to nearly 4,000 feet at the vv^estern base of the Santa Rita Mountains. On the eastern grassy plains the average elevation is probably between 4,200 and 5,000 feet. From the south-central portion of Arizona southward and westward and along the western border the summers are long and intensely hot, while the winters are mild. In the southeast the heat of summer is not so intense and the winters are somewhat colder. The annual mean temperature at Tucson is 68° Fahrenheit, at Fort Hua- chuca, 61°. Table op Temperatures in the Western Desert Region (at Tccson) and in the Eastern Plains Region (at Fort Huachuca) WINTER SPRING SUMMER FALL Mini- mum Maxi- mum Mean Mini- mum Maxi- mum Mean Mini- mum Maxi- mum Mean Mini- mum Maxi- mum Mean Tucson op 10 op 90 79 op 52 4.5 op 22 16 op 106 97 op 66 60 op 40 37 op 112 104 op 85 77 op 21 15 op 107 99 op 70 Fort Huachuca. . 63 There is considerable difference in the rainfall and hum- idity of the two regions. The valley of the Colorado in southwestern Arizona, with an annual rainfall of less than three inches, represents the extreme conditions as to aridity in the United States. Such conditions prevail along the southern boundary of Arizona eastward over most of Pima County, but in the eastern portion of that county, as the higher mountains are approached, the precipitation in- creases, the average annual rainfall at Tucson being 9.8 inches. Farther east it becomes still higher, being 16.2 inches at Fort Huachuca.* It is thus seen that the two sections of southern Arizona that are contrasted in the present study (the boundary line between indicated by the Santa Rita Mountains) present cer- tain slight differences of altitude, of temperature, and of rainfall, that are correlated with different types of vegeta- *The meteorologic data cited is taken from Climatology of the United States, by A. .1. Henry (U. S. Dept. Agric, Weather Bureau, Bull. 2, 1906), in which publication see also plate XXVl, showing normal annual precipitation in the United States. 274 CALIFORNIA ACADEMY OF SCIENCES [ Phoc. 4th Ser. o + Otocoris a. adusta Myiarchus t. magister *' "■'-■ Fig. D. Map showing distribution in southern Arizona of Myiarchus lyran- nulus magister and Otocoris alpestris adusta. Symbols indicate published record stations ; broken line indicates approximate northern and eastern boundaries of Myiarchus t. magister; solid line indicates approximate boundaries of Otocoris a. adusta. tion and with well marked differences in the faunas of the two regions. To define the two as occupying different life zones, the western Lower Sonoran, the eastern Upper Sonoran, does not seem satisfactory. The western section is, of course, emphatically Lower Sonoran in every respect. The eastern section is slightly higher altitudinally, of slight- ly greater rainfall, and of slightly lower temperature, and may be conceded to present some Upper Sonoran aspects. At the same time, wherever the eastern grassy plains are invaded bj^ limited growths of shrubs, bushes, or trees, these are in most cases Lower Sonoran desert species, such as mesquite, cholla cactus, ocotilla, etc. In the mountains of this section the foothill regions immediately above the plains possess characteristic Upper Sonoran assemblages of plants and animals which do not descend any lower. In some parts of the plains there are limited numbers of charac- teristic Lower Sonoran desert birds (Scaled Quail, White- Vol. XVIIII SWARTH—FAU.^AL AREAS OF SOUTHERN ARIZONA 275 o C^ c.. p. "^5 5^ WEST DESERT AREaVU\ 'X ''•■ ^■ r ^^ .* -.■^ tl X\«^£V 'IT % ^^^S^^^"^^^ .0 W 'S^' Citelhis t. neglectvis Citellus s. caneseens •ii, ^-EASTERN PLAINS AREA Fig. E. Map showing distribution in southern Arizona of Citellus terelicaudus neglectus and C. spilosoma caneseens. Symbols indicate record stations, mostly from hitherto unpublished data supplied by the United States Bureau of Bio- logical Survey. winged Dove, Phainopepla, Vermilion Flycatcher, and others) and mammals (species of Peromyscus, Onychomys, Lepus, and others) associated with such species as the Prong-horn, Prairie Dog, Horned Lark, and others, that occur elsewhere in Upper Sonoran and higher. The two sections, on the whole, do not seem to me to show differences of life zones in their contrasting characteristics, but to be comparable rather to the "faunal areas" described by Grinnell (1915, pp. 9-12) in his treatment of the distri- bution of birds in California. The extreme southeastern corner of Arizona appears to be definable as a faunal area distinct from the regions to the westward and to the north- ward. The western boundary of this faunal area is the sub- ject of the present study. Of the boundary line elsewhere I can speak with less assurance, but on the northwest the Santa Catalina Mountains may perhaps mark the dividing line. Of the extent of this faunal area eastward into or through southern New Mexico, and southward into Mexico 276 CALIFORNIA ACADEMY OF SCIENCES I Proc. 4th Sbr. Fig. F. Map showing distribution in southern Arizona of Ammospermophilus harrisii. Symbols indicate record stations, mostly from hitherto unpublished data supplied by the United States Bureau of Biological Survey. I know nothing, but my impression is that the faunal area I am describing in the southeastern corner of Arizona, forms the northwestern portion of a much more extensive area over the regions mentioned. Aside from Mearns' (1907) divisions along the United States-Mexico boundary hue, there has been no previous attempt to indicate in Arizona any faunal divisions other than life zones, but it seems feasible now to outline, though in loose terms and with rather indefinite boundaries, at least five faunal areas into which the state can be divided. The Western Desert Area and the Eastern Plains Area, with which this paper is mainly concerned, are capable of fairly exact definition, and the boundary between these two can be closely indicated. To the northward of these areas is the Central Plateau Area, with the Mogollon Plateau as a center and extending diagonally nearly across the state, from the Grand Caiion at the northwest, to the White Mountains at the southeast. In extreme northeastern Ari- zona, centering about the Painted Desert and the Little Vol. XVIII] SWARTH—FAUNAL AREAS OF SOUTHERN ARIZONA 277 Fig. G. Map showing approximate boundaries of habitat of Lepus alleni alleni in Arizona. Symbols indicate known stations of occurrence. Colorado River, is what may be designated as the North- eastern Desert Area. In the northwest, north of the Colo- rado River, is a region concerning which I have no first hand knowledge, but which presumably is faunally related to the Great Basin. The boundary line I have indicated between the Western Desert Area and the Eastern Plains Area does not accord with that described by Mearns (1907, pp. 73-74, pi. II) in his study of the mammals of the Mexican boundary. I can not appreciate any reason for the dividing line he draws across the desert midway between Tucson and Yuma, with the "Western Desert Tract" to the westward, the "Ele- vated Central Tract" to the eastward. Neither is there any general division of forms in mammals, birds or plants along that line, nor is there any marked change in altitude or climate. The same species and subspecies of mammals and birds, with few exceptions, and the same sorts of vege- tation range from the Colorado River eastward to the west base of the Santa Ritas. Grinnell (1914) has shown how potent a barrier the Colorado River is as regards the mam- 278 CALIFORNIA ACADEMY OF SCIENCES I Proc. 4th Seb- mals of the deserts on either side. In the bottom lands the same species of birds and mammals occupy both sides of the stream, forming a characteristic river-bottom fauna; this fauna as a whole is distinctly that of the Arizona valleys to the eastward. My conception of the deserts of southwestern Arizona are as comprising one faunal area, extending from, and including, the bottom lands of the west bank of the Colorado River east to the western base of the Santa Rita Mountains. The northern boundary of this desert area may be very roughly indicated as extending from the vi- cinity of Fort Mohave on the Colorado River, in a south- easterly direction toward Phoenix and Tucson, following the bases of the mountains northeast of those cities. Faunal conditions at the western boundary of this area, along the Colorado River, are presented by Grinnell (1914) in fullest detail. It has been my aim in the present paper to give as exact a statement as circumstances permit of conditions at the eastern border of this faunal area. Con- siderably more collecting of small mammals is necessary, however, for the filling out of details. There are certain conspicuous diurnal mammals whose restriction to one or the other of the two areas here consid- ered is apparent to even rather casual observation. Fore- most of these is Lepus alleni, as detailed beyond. The re- striction of this species east and west within the wider habi- tat of Lepus calif ornicus is one of the most peculiar delimita- tions among North American animals. In former years the Prairie Dog {Cynomys ludovicianus arizonensis) was abun- dant in southeastern Arizona. Whether or not it has sur- vived persecution by governmental rodent control activities I do not know, but until at least 1907 there were large numbers on the plains between the Huachuca Mountains and Bisbee, and a small and singularly isolated colony some 30 miles farther north, between Fort Huachuca and Fair- bank. It is a curious fact that the species did not extend farther north and west over apparently suitable ground. Whether or not it ever reached as far west as the Santa Rita Mountains I do not know; it probably never went beyond. The Prong-horn (Antilocapra americana) was fairly nu- merous in southeastern Arizona in years past. Upon my first visit to the region, in 1896, there were still herds of Vol. XVIII] SWARTH—FAUNAL AREAS OF SOUTHERN ARIZONA 279 15 or 20 to be found in the San Pedro Valley, and single ani- mals or two or three together were seen by me near the Huachucas and near the east base of the Santa Ritas as late as 1902 and 1903. In 1907 I was told by cattlemen that none remained in that section. The species occurred also in places west of the Santa Ritas, and may still do so here and there, but I do not believe ever in such numbers as on the plains to the eastward. The small ground squirrels, Ammos'permophiliis and Citel- lus, afford good examples of delimitation of range, and re- placement of one species by another in the two regions. Of the smaller nocturnal mammals too little is known to com- pile any long or exact list of species confined to one section or the other. The pocket gopher (Thomomys) , prone as this genus is to become differentiated into local forms, apparent- ly is not to be divided in the two regions here considered. Over most of the country no gophers occur, being entirely absent from the hard, dry uplands, distribution taking place along riparian surroundings of the river beds. So division of races of Thomomys in this section is apparently entirely altitudinal. Among birds there are many striking replacements of species or subspecies in the two regions. Some of the most conspicuous are the Gambel Quail and Scaled Quail, and Western Meadowlark and Texas Meadowlark. Some less noticeable replacements are found in subspecies of the Red- winged Blackbird, Cliff Swallow, and Curve-billed Thrasher. It will be noted that although the Western Raven is com- mon in the Lower Sonoran zone of southeastern California and southwestern Arizona, it is rare and mostly an Upper Sonoran species in southeastern Arizona, being replaced on the Lower Sonoran plains of that region by the White- necked Raven. There is a longer list of bird species from southwestern Arizona than from the southeast, the varied vegetation of the southwest affording congenial surroundings to many that do not occur on the grassy plains. On the other hand, there are certain conspicuous bird species of the southeast- ern plains that are pretty closely confined to that region, such as the Swainson Hawk, Scorched Horned Lark, and White-necked Raven. There are in the southeast some 280 CALIFORNIA ACADEMY OF SCIENCES 1 Phoc. 4th Seb. Upper Sonoran species characteristic of the region that oc- casionally descend as far as the upper edge of the plains, and that form one of the several factors tending to give an Upper Sonoran aspect to the lowlands. Some of these are the Western Nighthawk, Western Yellow-wing Sparrow, and the Azure Bluebird. The last mentioned species was not encountered by us, but information recently received by me from Dr. L. B. Bishop, and from Mr. Edward C. Jacot, of Prescott, Arizona, justifies its inclusion in my statement. The accompanying lists of mammals and birds may serve to convey an idea of the two contrasting faunas. It must be borne in mind, though, that these are not hard and fast divisions and that in many cases species mainly confined to one of the two regions may extend more or less into the other territory. This is especially true of certain birds of the river bottoms, which, occurring in greatest abundance in southwestern Arizona, penetrate in lesser numbers along the more sparsely brush-margined streams of the southeast. This applies to such species as Song Sparrow, Pyrrhuloxia, Least Vireo, and Yellow Warbler. The same is true of certain species of the chaparral of the mesa. A striking feature of our findings along the dividing line between the two opposed faunal areas is the manner in which many species from either side extend short distances beyond the normal boundary. As a basis for our work the Santa Rita Mountains were a convenient line of demarca- tion, and forming as they do a colossal wall across the plains, they might easily be supposed to be a barrier in fact as they are in appearance. Again and again, though, we found western species ranging clear around the mountains in a ribbon-like habitat below the eastern foothills, and, con- versely, eastern species extending around to the western base of the range. The Allen Jack Rabbit, in small num- bers, occurs eastward as far as Sonoita and Patagonia, but at that point finds some insuperable obstacle to its farther extension over the open plains beyond, an obstacle that has no existence for the more widely spread Black-tailed Jack Rabbit. The Scaled Quail ranges westward around the mountainous wall, to be stopped below the western foothills by some impalpable barrier that absolutely forbids farther progress. So, at any point around the base of the moun- Vol. XVIII 1 SWARTH—FAUNAL AREAS OF SOUTHERN ARIZONA 281 tains one may find in greater or less abundance an infiltra- tion of species that properly belong on the opposite side, with assurance that within a short distance east or west, as the case may be, those species will be found to disappear. The elucidation of this feature in the distribution of spe- cies along this boundary line entails in the case of many of the small nocturnal mammals more extensive trapping than we were able to accomplish. With such an animal as Dipodomys spectahilis the conspicuous mounds and burrows are sufficient to advertise its presence, but with many others it is not usually safe to generalize as to their status in either of the faunal areas upon the basis of a limited number of specimens from a few localities. With diurnal mammals and birds the facts are more readily apparent. Another interesting aspect of distribution in this part of Arizona is found in the manner of occurrence of certain mi- gratory birds. The McCown Longspur, Chestnut-collared Longspur, and Baird Sparrow are all common migrants on the eastern grass-lands, but they do not occur on the west- ern deserts. The Lark Bunting, however, which might be expected to adhere as closely to the open prairie, is far more abundant in western Arizona. There are certain bird species that have almost or en- tirely disappeared from Arizona in recent years, exact infor- mation regarding which would be of great value and inter- est in this connection. I refer to the Masked Bob-white (Colinus ridgwayi), the Rufous-winged Sparrow {Aimophila carpdlis), and the Botteri Sparrow (Peuccea hotterii). In all likelihood these three birds were mainly inhabitants of grasslands, and there seems little reason to doubt that their disappearance was due entirely to the overstocking of the ranges with cattle. When years of drought came every vestige of their natural cover was destroyed. This explana- tion has been advanced by Brown (1904) to account for the disappearance of the Bob-white, and it probably ex- plains also the nearly complete extinction locally of the two species of sparrows. The Cassin Sparrow, with similar habitat predilections, is migratory, if, in fact, it breeds in this region at all. So it survives in Arizona and is to be found, we may assume, in the same sort of surroundings that were formerly shared with its vanished relatives. 282 CALIFORNIA ACADEMY OF SCIENCES (Pboc. 4th Seb. The delimitation of the ranges of species, east or west, as described in this paper, must be understood to apply to a relatively narrow area bordering the Arizona-Mexico boun- dary line. Thus, certain of the birds here ascribed to a western habitat are known to occur farther east into New Mexico and Texas, but this eastern extension of range occurs either north or south of the region here under dis- cussion. Western Desert Area Lophortyx g. gambelii Melopelia a. trudeaui Scardafella inca Asturina plagiata Micropallas w. whitneyi Dryobates s. cactophilus Colapies c. meamsi Myiarchus t. magister Corvus c. sinuatus Agelaius p. sonoriensis Sturnella neglecta Melospiza m. saltonis Cardinalis c. superbus Pyrrhuloxia s. sinuata Guiraca c. interfusa Piranga r. cooperi Petrochelidon I. lunifrons Vireo b. arizonse Vermivora luciae Dendroica a. sonwana Toxostoma c. pahneri Toxostoma bendirei Polioptila ni. melanura Birds Eastern Plains Area Colinus ridgwayi Callipepla s. pallida Buteo swainsoni Otocoris a. adnata Corvus cryptoleucus Agelaius p. nevadensis Sturnella m. hoopesi Ammodramus s. bimaculalus Peucasa botterii Aimophila carpalis Petrochelidon I. melanogastra Toxostoma c. curviroslre Vol. XVIII 1 SWARTH—FAVNAL AREAS OF SOUTHERN ARIZONA 283 Western Desert Area My Otis c. pallidus Citellus t. neglectus Ammospermophilus harrisi Thomomys f. toltecus Perognathus amplus Perognathus h. baileyi Perognathus p. pricei Dipodomys s. spectdbilis Dipodomys m. merriami Onychomys t. torridus Reithrodontomys m. megalotis Peromyscus e. eremiais Peromyscus m. sonoriensis Sigmodon h. cienegx Neoloma a. albigula Lepus a. alleni Lepiis c. eremicus Sylvilagus a. arizonse Mammals Eastern Plains Area Myotis c. californicus Citellus s. canescens Cynomys I. arizonensis Thonwmys f. toltecus Perognathus p. pricei Dipodomys m. olivaceus Dipodomys o. ordii Onychomys t. torridus Reithrodontomys m. megalotis Peromyscus e. eremicus Peromyscus m. sonoriensis Peromyscus I. arizonx Sigmodon h. cienegx Neotoma a. albigula Lepus c. eremicus Sylvilagus a. arizonx April 26, 1929 284 CALIFORNIA ACADEMY OF SCIENCES I Pboc. 4th Seb. Check-List of the Birds 1. Chlidonias nigra surinamensia (Gmelin) 53. 2. Netlion carolinense (Gmelin) 3. Querquedula cyanoptera (Vieillot) 54. 4. Dafila acuta tzitzihoa (Vieillot) 55. 5. Ardea herodias tregamai Court 56. 6. Butorides virescens anthonyi (Mearns) 57. 7. Rallus virginianus Linnseus 58. 8. PoTzana Carolina (Linnseus) 59. 9. Fulica americana Gmelin 10. Gallinago delicata (Ord) 60 11. Pisobia minulilla (Vieillot) 61. 12. Tringa solitaria Wilson 62. 13. Aclitis macularia (Linnaeus) 63. 14. Oxyechics vociferus tociferus (Linnseus) 64. 15. Callipepla squamata pallida Brewster 16. Lophortyx gambelii gambelii Gambel 65. 17. Cyrtiinyx montezumx mearnsi Nelson 66. 18. Columba fasciata fasciata Say 67. 19. Zenaidura macroura marginella 68. (Woodhouse) 69. 20. Melopelia asiatica trudeaui (Audubon) 70. 21. Chxmepelia passerina pallesceris Baird 71. 22. Scarda/ella in.ca (Lesson) 72. 23. Catharies aura septentrionalis Wied 73. 24. Accipiter celox (Wilson) 74. 25. Accipiter cooperii (Bonaparte) 75. 26. Parabuteo unicinctus harrisi (Aadubon) 76. 27. Buteo borealis calurus Cassin 77. 28. Buteo swainso7ii Bonaparte 78. 29. Asturina plagiata Schlegel 79. 30. Aquila chrysaetos (Linnaeus) 80. 31. Cerchneis sparveria pkalsena (Lesson) 81. 32. Polyborus cheriway (Jacquin) 82. 33. Asio wilsonianus (Lesson) 83. 34. Otus asio eineraceus (Ridgway) 84. 35. Bubo virginianus pallescens Stone 85. 36. Speotyto cunicularia hypogsea (Bona- parte) 86. 37. Micropallas whitneyi whiineyi (J. G. 87. Cooper) 88. 38. Geococcyx californianus (Lesson) 89. 39. Coccyzus americanus occidentalis Ridgway 90. 40. Ceryle alcyon caurina Grinnell 91. 41. Dryobates scalaris cactophilus Oberholser 92. 42. Dryobates arizonx arizonse (Hargitt) 93. 43. Sphyrapicus varius nuchalis Baird 44. Melanerpes formicivorus aculeatus 94. Mearns 95. 45. Centurus uropygialis uropygialis Baird 46. Colaptea cafer collaris Vigors 96. 47. Colaptes chrysoides mearnsi Ridgway 97, 48. Phalxnoptilus nuttallii nuttallii 98, (Audubon) 99, 49. Chordeiles virginianus henryi Cassin 100, 50. Chordeiles acutipennis texensis Lawrence 101, 51. Aero7iautes saxatalis (Woodhouse) 102, 52. Eugenes fulgent (Swainson) 103, Archilochus alexandri (Bourcier & Mul- sant) Calypte costse (Bourcier) Cynanthus latirostris Swainson Tyranrnis verticalis Say Tyrannus vociferans Swainson Myiarchus tyrannulus magister Ridgway Myiarchus cineraacens cinerascens (Lawrence) Myiarchus tuberculifer olivascens Ridgway Sayornis sayus sayus (Bonaparte) Sayornis nigricans nigricans (Swainson) Nutlalloniis mesoleucus (Lichtenstein) Myiochanes richardsonii richardsonii (Swainson) Empidonax difficilis difficilis Baird Empidonax iraillii brewsteri Oberholser EmpidoiMX hammondii (Xantus) Empidonax griseus Brewster Pyrocephalus rubimis mexicanus Sclater Camptostoma imberbe Sclater Otocoris alpestris adxista Dwight Otocoris alpestris occidentalis McCall Cyanocilta stelleri diademata (Bonaparte) Aphelocoma sieberi arizonx (Ridgway) Corvus corax sinuatus Wagler Corvus cryptoleucus Couch Molothrus ater obscurus (Gmelin) Tangavius xneus xneus (Wagler) Agelaius phoeniceus nevadensis Grinnell Sturnella magna hoopesi Stone Sturnella neglecta Audubon Icterus parisorum. Bonaparte Icterus cucullatus nelsoni Ridgway Icterus bullockii (Swainson) Euphagus cyanocephalus cyanocephalus (Wagler) Passer do7nesticus (Linnaeus) Carpodacus cassinii Baird Carpodacus mexicanus frontalis (Say) Astragalinus psaliria hesperophilus Oberholser Spimis pinus pinus (Wilson) Calcarius ornatus (J. K. Townsend) Pooecetes gramineus confinis Baird Passerculus sandwichensis nevadensis Grinnell Ammodraimis bairdii (Audubon) Ammodramus savannarum bimaculatua Swainson Chondesles grammacus strigatus Swainson Zonotrichia leucophrys (Forster) Zonotrichia gambelii (Nuttall) Spizella passerina arizonx Coues Spizella breweri Cassin Junco phxonotus palliatus Ridgway Amphispiza bilineata deserticola Ridgway Peucxa cassinii (Woodhouse) Vol. XVIII 1 SW ART H— FAUN AL AREAS OF SOUTHERN ARIZONA 285 104. Aimophila ruficeps scottii (Sennett) 134. 105. Melospiza melodia sallonis Grinnell 1-35. 106. Melospiza melodia fallax (Baird) 136. Melospiza lincolnii lincolnii (Audubon) 137. Pipilo fuscus mesoleucus Baird Oberholseria chlorura (Audubon) 138. Cardinalis cardinalis superbus Ridgway 139. Pyrrhuloxia sinuata sinuata (Bonaparte) 140. liedymeles melanocephalus nielanocepha- 141. lus (Swainson) 142. 113. Guiraca cxrulea interfusa Dwight & 143. Grisoom 144. Passerina amcena (Say) 145. Spiza americana (Gmelin) 146. Calamospiza melanocorys Stejneger 147. Piranga ludoviciana (Wilson) 148. Piranga hepatica oreophasma Oberholser Piranga rubra cooperi Ridgway 149. Pelrochelidon lunifrons melanogastra 150. (Swainson) 151. Hirundo erythrogasira Boddaert 152. Tachycinela thalassina lepida Mearns Stelgidopleryx serripennis (Audubon) 153. Bombycilla cedrorum Vieillot 154. Phainopepla niiens (Swainson) 155. Lanius ludovicianus excubitorides Swainson 156. Vireosylva gilva swainsonii (Baird) 157. 128. Lanivireo solitarius cassinii (Xantus) 158. 129. Lanivireo solitarius plumbeus (Coues) 1.59. Vireo huttoni slephensi Brewster 160. Vireo belli arizonse Ridgway 161. Vermivora lucix (J. G. Cooper) 162. Vermivora ruficapilla gutturalis (Ridg- 163. way) 164. 107. 108. 109. 110. 111. 112. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 130 131 132 133, Vermivora celata lutescens (Ridgway) Dendroica xstiva sonorana Brewster Dendroica xstiva brewsteri Grinnell Dendroica auduboni auduboni (J. K. Townsend) Dendroica nigrescens (J. K. Townsend) Dendroica lownsendi (J. K. Townsend) Oporornis tolmiei (J. K. Townsend) Geothlypis Irichas scirpicola Grinnell Geothlypis trichas occidentalis Brewster Icteria virens longicauda Lawrence Wilaonia pusilla pileolata (Pallas) Wilsonia pusilla chryseola Ridgway Setophaga picta Swainson Mimus polyglottos leucopterus (Vigors) Toxostoma curvirostre curvirostre (Swain- son) Toxostoma curvirostre palmeri (Coues) Toxostoma bendirei (Coues) Toxostoma crissale crissale Henry Heleodytes brunneicapillus couesi (Sharpe) Salpinctes obsoletus obsoletiis (Say) Catherpes mexicanus conspersus Ridgway Thryomanes bewickii eremophilus Oberholser Troglodytes aedon parkmanii Audubon Sitta carolinensis nelsoni Mearns Bxolophus wollweberi annexus (Cassin) Psaltriparus plumbeus (Baird) Auriparus flaviceps flaviceps (Sundevall) Regulus calendula calendula (Linnceus) PoUoptila cxrulea amcenissima Grinnell Polioptila m,elanura melanura Lawrence Hylocichla uslulata ustulata (Nuttall) General Accounts of the Birds 1. Chlidonias nigra surinamensis (Gmelin) Two specimens (Nos. 29822-29823), birds of the year, were collected six miles north of Patagonia, September 8. There are few records of the occurrence of this species in Arizona, but it was collected bj^ Henshaw in August in Cochise County (Henshaw, 1875, p. 487; Saunders, 1896, p. 22) and is probably a fairly regular late summer migrant in the southeastern section of the state. 2. Nettion carolinense (Gmelin) Small flocks were seen on cattle ''tanks" near Patagonia, on September 22, when an immature male (No. 29824) was collected, and on the 23rd, when a female (No. 29826) was shot. 286 CALIFORNIA ACADEMY OF SCIENCES [ Pkoc. 4th Seb. 3. Querquedula cyanoptera (Vieillot) A few birds, paired or singly, appeared on the several reservoirs and "tanks" on the Ashburn ranch, May 11 to 20. We were told that prior to our arrival ducks of several species had been of fairly common occurrence there. Pre- sumably the few we saw were the last straggling migrants. A female Cinnamon Teal (No. 29825) was collected Sep- tember 23, and others were seen. 4. Dafila acuta tzitzihoa (Vieillot) A flock of ten or twelve seen near Patagonia, September 1, and one specimen (No. 29827) preserved. Ducks that may have been of the same species were seen later, in September and Oct