ZOOLOGICA SCIENTIFIC CONTRIBUTIONS of the NEW YORK ZOOLOGICAL SOCIETY VOLUME 34 '3 c> 1949 -s1" NUMBERS 1-20 Published by the Society The Zoological Park, New York December 30, 1949 NEW YORK ZOOLOGICAL SOCIETY General Office: 30 East Fortieth Street, New York 16, N. Y. Publication Office: The Zoological Park, New York 60, N. Y. OFFICERS Fairfield Osborn, President Alfred Ely, Vice-president Laurance S. Rockefeller, Vice-president Donald T. Carlisle, Vice-president Harold J. O’Connell, Secretary Cornelius R. Agnew, Treasurer SCIENTIFIC STAFF General 3 C\ , QG Q 1 A ^ John Tee-Van, Executive Secretary William Bridges, Editor and Curator of Publications Sam Dunton, Photographer Zoological Park Lee S. Crandall, General Curator Grace Davall, Assistant to General Curator Brayton Eddy, Curator of Reptiles and Insects Leonard J. Goss, Veterinarian Robert M. McClung, Assistant, Mammals and Birds Aquarium Christopher W, Coates, Curator and Aquarist James W. Atz, Assistant Curator Ross F. Nigrelli, Pathologist Myron Gordon, Geneticist C. M. Breder, Jr., Research Associate in Ichthyology G. M. Smith, Research Associate in Pathology Homer W. Smith, Research Associate in Physiology Department of Tropical Research William Beebe, Director Jocelyn Crane, Research Zoologist Henry Fleming, Entomologist William K. Gregory, Associate John Tee-Van, Associate Gloria Hollister, Associate Mary VanderPyl, Associate Scientific Advisory Council A. Raymond Dochez Caryl P. Haskins Alfred E. Emerson K. S. Lashley W. A. Hagan John S. Nicholas George M. Smith Editorial Committee Fairfield Osborn, Chairman William Beebe Lee S. Crandall William Bridges Brayton Eddy Christopher W. Coates Leonard L. Goss John Tee-Van CONTENTS Part 1. May 16, 1949. PAGE 1. Paradilepis simoni n. sp., a Cestode Parasitic in the Osprey. (Ces- toda: Dilepididae) . By Robert Rausch. Text-figure 1 1 2. A Contribution to the Study of North American Cestodes of the Genus Paruterina Fuhrmann, 1906. By Robert Rausch and Everett L. Schiller. Text-figures 1-12 5 3. Behavioral Interactions in a Herd of Barbary Sheep ( Ammo- tragus lervia) . By Irwin Katz 9 4. The Pericopidae (Moths) of Kartabo, British Guiana, and Cari- pito, Venezuela. By Henry Fleming 19 5. Report on a Collection of Phalangids from Rancho Grande, Vene- zuela. By Clarence and Marie Goodnight. Text-figures 1-4 21 6. Fresh-water Crabs of the Genus Pseudothelphusa from Rancho Grande, Venezuela. By Jocelyn Crane. Text-figures 1-3 25 Part 2. August 10, 1949. 7. Comparative Biology of Salticid Spiders at Rancho Grande, Vene- zuela. Part III. Systematics and Behavior in Representative New Species. By Jocelyn Crane. Text-figures 1-8 31 8. The Swifts of Rancho Grande, North-central Venezuela, with Special Reference to Migration. By William Beebe. Plate I; Text-figures 1-3 53 9. Eastern Pacific Expeditions of the New York Zoological Society. XL. Mollusks from the West Coast of Mexico and Central Amer- ica. Part VII. By Leo George Hertlein & A. M. Strong. Plate I.. 63 10. Fishes That Rank Themselves Like Soldiers on Parade. By E. W. Gudger. Plate I ; Text-figures 1 & 2 99 11. Notes on Seasonal Changes in Creatophora cinerea, the Wattled Starling. By Lee S. Crandall. Plate 1 103 12. Insect Migration at Rancho Grande in North-central Venezuela. General Account. By William Beebe. Plates I & II ; Text-figure 1 107 Part 3. November 30, 1949. PAGE 13. The Behavior of Two Captive Specimens of the Lowland Gorilla, Gorilla gorilla gorilla (Savage & Wyman). By B. F. Riess, Sherman Ross, S. B. Lyerly & H. G. Birch. Plates I & II; Text-figures 1 & 2 Ill 14. Migration of Papilionidae at Rancho Grande, North-central Venezuela. By William Beebe. Plate I; Text-figure 1 119 15. Notes on Ergasilus Parasites from the New Brunswick, New Jersey, Area, with a Check List of All Species and Hosts East of the Mississippi River. By Roland F. Smith 127 16. An Analysis of Reproductive Behavior in the Mouth-breeding Cichlid Fish, Tilapia macrocephala (Bleeker). By Lester R. Aronson. Plates I-III; Text-figures 1-10 133 Part 4. December 30, 1949. 17. Comparative Biology of Salticid Spiders at Rancho Grande, Ven- ezuela. Part IV. An Analysis of Display. By Jocelyn Crane. Plate I; Text-figures 1-9 & 159 18. Differential Effects of Estradiol, Estradiol Benzoate and Preg- neninolone on Platypoecilus maculatus. By Margaret Cordsen Tavolga. Plates I-V ; Text-figures 1-5 215 19. Eastern Pacific Expeditions of the New York Zoological Society. XLI. Mollusks from the West Coast of Mexico and Central America. By Leo George Hertlein & A. M. Strong. Plate 1 239 20. Tettigellidae and Gyponidae (Homoptera) of Kartabo, Bartica District, British Guiana. By Z. P. Metcalf. Text-figures 1-8.... 259 Index to Volume 34 2wl Rausch: Paradilepis simoni, n. sp., a Cestode Parasitic in the Osprey 1 1. Paradilepis simoni n. sp., a Cestode Parasitic in the Osprey. (Cestoda: Dilepididae).1 Robert Rausch2. Department of Veterinary Science, University of Wisconsin, Madison 3 (Text-figure 1). The knowledge of helminths parasitic in North American birds is very incomplete, especially for the region west of the Miss- issippi. Not only is this true in regard to parasite-host ecology, but a little work with almost any group of birds discloses unde- scribed forms or species unreported from North America, as well. In fact, anyone wish- ing to carry out host-parasite studies must devote considerable time to describing spe- cies-time which could be more profitably spent otherwise. The osprey might be men- tioned to illustrate this situation. Of this bird the writer has examined but three speci- mens, one each from Ohio, Wisconsin and Wyoming. From this small series of birds, four species of helminths were collected; of these, two species were undescribed and two had never been recorded from North Amer- ica. While it is true that a comparable situa- tion is not to be expected in every case, it soon becomes obvious from work with a given host-group that much remains to be done before the helminths encountered in birds can be readily identified. The cestodes with which this paper is con- cerned were obtained from the small intes- tine of an osprey, Pandion haliaetus carolin- ensis (Gmelin), collected on June 3, 1948, near Moran, Wyoming. This osprey was one of 267 birds collected by the writer for hel- minthological study from the Jackson Hole region of Wyoming. A total of more than 75 worms was ob- tained. Whole-mounts were prepared of specimens stained with Semichon’s acetic carmine and Delafield’s haematoxylin. Serial sections, cut at 15/x, were also studied. This cestode is named in honor of Mr. James Simon, Director of the Jackson Hole Wildlife Park, whose generous cooperation contributed much to the success of the field work in Jackson Hole. 1 Contribution of the 1948 Research Program of the New York Zoological Society at Jackson Hole Wildlife Park. 4 Now with U. S. Public Health Service, Anchorage. Alaska. 3 Section on Parasitology, B. B. Morgan. Project Leader. Paradilepis simoni n. sp. (Text-fig. 1 A-E). Diagnosis : Strobila from 50 to 90 mm. long; greatest width, up to 450 p, attained in terminal gravid segments. External seg- mentation absent ; strobila very delicate and translucent in the living worm. Scolex large and distinct from neck; from 470 to 596 p in diameter. Suckers from 180 to 220 p in diameter. Well-developed rostellum slightly over 100 p long; armed with about 36 hooks arranged in a double row. Large hooks from 98 to 102 p long ; small hooks 68 to 72 p long. Hook shape typical for genus. Neck from 250 to 270 p wide, narrowing gradually to a distance of about 2 mm. posterior to scolex ; from this point the strobila widens to reach greatest width at posterior enu. Musculature consists of two layers; the first layer, of lon- gitudinal fibers, is from 1 to 3 bundles deep; directly beneath it is a layer of transverse fibers. Excretory canals typical in arrange- ment; the ventral longitudinal canal meas- ures from 6 to 20 p in diameter; the dorsal and transverse canals about 3 p in diameter. Internal segmentation best recognized by the arrangement of the transverse excretory ca- nals, which divide the strobila into about 30 “segments” per mm. of length in the mature region. Genital organs not confined entirely to space between transverse canals, but over- lap into adjacent segments. Genital Anlagen appear about 2 mm. posterior to scolex. Geni- tal pores unilateral and sinistral; genital atrium about 16 p deep. Genital canals pass dorsal to longitudinal excretory canals. Five spherical to ellipsoidal testes, not all in same plane, in each segment; testes measure from 26 to 33 p in diameter in mature segments. Usually 4 testes are aporal, and one is poral of female organs; at times 2 may be poral. Flask-shaped cirrus sac extends to middle of mature segments, or beyond, dorsal to testes; it measures from 100 to 132 p long by 30 to 40 p wide. Cirrus heavily spined. Internal and external seminal vesicles absent; ductus ejaculatorius coiled within bulb of cirrus sac. Text-fig. 1. The morphology of Paradilepis simoni n. sp. The drawings were made in part with the aid of a projector. A. Ventral view of a typical mature segment. B. Typical scolex. C. Dorsal view of a section of gravid segments. D. Hooks from rostellum. E. Cross- section of genital atrium region, showing relation of cirrus to vagina. 1949] Rausch: Paradilepis simoni, n. sp., a Cestode Parasitic in the Osprey 3 Well-developed vas deferens with numerous convolutions situated in dorsal part of seg- ment. Cirrus sac provided with strongly- developed retractor muscles. Thin-walled va- gina opens ventral to cirrus sac; it enlarges gradually, attaining greatest diameter near place where longitudinal excretory canals are crossed ; it narrows after this point and runs medially to join small seminal receptacle dor- sal to ovary. Ovary rather variable in shape and position; usually 4-lobed, situated near middle of segment. Spherical to ellipsoidal vitelline gland dorsal to posterior part of ovary; it increases in size toward posterior end of strobila, attaining a maximum diam- eter of about 40 p. Uterus develops as two lateral, spherical sacs situated ventral to ovary, and connected by a narrow neck. Grav- id uterus fills entire segment; unlobed and sac-like when completely gravid. Cirrus sac and vagina persist into terminal gravid seg- ments. Eggs, from 27 to 33 p in diameter, are arranged in 3 to 4 rows across the segments. Embryonic hooks about 6 p in length. Host: Pandion lialiaetus carolinensis (Gmelin) (Osprey). Habitat: Small intestine. Locality: Moran, Wyoming. Type : Three slides of cotype material have been deposited in the Helminthological Col- lection of the U. S. National Museum, No. 46403. Discussion. As far as could be determined, the genus Paradilepis Hsii, 1935, has not been previ- ously recorded from North America. Neither has the writer discovered any record of ces- todes parasitic in the osprey. Cestodes of the genus Paradilepis are typi- cally parasitic in pelicaniform birds, particu- larly in cormorants, Phalacrocorax spp. The genus Paradilepis was established (Hsii, 1935) for cestodes from a Chinese cormor- ant, with P. duboisi as type. Hsii also assigned Dilepis scolecina (Rudolphi, 1819) to the genus Paradilepis. According to Joyeux and Baer (1935), P. duboisi is identical with P. scolecina; consequently P. scolecina (Syn. P. duboisi ) becomes type species. The examina- tion of the original preparations of Oligor- chis delachauxi Fuhrmann, 1909, led Joyeux and Baer (1935) to place it in the genus Paradilepis. It had been earlier assigned by the same writers (1930) to the genus Dilepis Weinland. Further study of their African material disclosed that they were deaiing with two species, referred to as Dilepis de- lachauxi (Fuhrmann, 1909). As a result, a new name, P. macracantha, was proposed (Joyeux and Baer, 1935) for Dilepis dela- chauxi Joyeux and Baer, 1930 nec Fuhrmann, 1909. Burt (1940) described Paradilepis brevis from a Ceylon cormorant, apparently without referring to the work of Joyeux and Baer (1935). It is possible that P. brevis is iden- tical with P. scolecina. Joyeux and Baer (1935) suggested that Oligorchis longivaginosus Mayhew, 1925, might also belong to the genus Paradilepis. This is of particular interest in connection with the present paper, since O. longivagino- sus was collected from a white pelican from Yellowstone Park, Wyoming. Apparently this species has a single crown of hooks, in- stead of a double row as seen in Paradilepis ; external segmentation also seems evident. The number of species of the genus Para- dilepis is at present indefinite, and must re- main so until some of the material is studied further. Regardless of this situation, P. simoni is readily differentiated from any others previously assigned to the genus in that it possesses 5 testes in each segment, instead of 4. Although the presence of 4 testes is con- sidered a generic character by Joyeux and Baer (1935, 1936), we do not consider it justifiable to erect a new genus for P. simoni on the basis of this character alone. It is otherwise very similar to the other members of the genus. Since the previously known species have been described from pelicani- form birds, it is not strange that they are morphologically similar. If, in addition to P. simoni, cestodes of this genus are recorded from other host groups, a much better con- cept of morphological variation within the genus may be had. It is possible that P. simoni is an “acci- dental” parasite of the osprey, and occurs naturally in cormorants. It would be of in- terest to examine cormorants from the colony north of Jackson Hole in order to determine whether they are parasitized by any species of Paradilepis. Since all the hosts are pis- civorous, presumably species of fish might act as the intermediate hosts of cestodes of this genus. At present, there is no reason to doubt that the osprey is the natural host of P. simoni. References. Burt, D. R. R. 1940. New species of cestodes from Chara- driiformes, Ardeiformes, and Pelicani- formes in Ceylon. Ceylon Jour. Sci., sect. B, 22 1-63. Hsu, H. F. 1935. Contributions a l’etude des cestodes de Chine. Rev. Suisse Zool., 42: 477-570. Joyeux, Ch. and Baer, J. G. 1930. Mission saharienne Augieras-Draper, 1927-1928. Cestodes. Bull. Mus. Nat. Hist., second ser. 2: 217-223. 1935. Notices helminthologiques. Bull. Soc. Zool. France, 60: 482-501. 1936. Faune de France 30. Cestodes. Paris, 613 pp. Mayhew, R. L. 1925. Studies on the avian species of the cestode family Hymenolepididae. III. Biol. Monogr., 1 0 : 7-125. . . . - . . * .... ■ Rausch & Schiller: North American Cestodes of the Genus Paruterina 5 2. A Contribution to the Study of North American Cestodes of the Genus Paruterina Fuhrmann, 19061. Robert Rausch2 and Everett Schiller. Department of Veterinary Science, University of Wisconsin, Madison 3 (Text-figures 1-12). Two of the 16 apparently valid species of Paruterina Fuhrmann, 1906, have been re- corded from North American birds. Par- uterina similis (Ransom, 1909) occurs in the yellow-billed cuckoo, Coccyzus a. americanus (L.), and P. candelabraria (Goeze, 1782) is the most frequently encountered cestode parasitic in owls. The latter occurs in Europe as well as in North America, and infects sev- eral species of owls (Wolffhiigel, 1900; Rausch, 1948). Evidence to the present would indicate that a high degree of host specificity has been developed in the cestodes of this genus. It is the purpose of this paper to describe two species of Paruterina, and to include some remarks concerning the two previously- recorded Noi’th American species. The un- described species were collected by one of us (R. R.) from birds in the Jackson Hole region of Wyoming. Both were taken from hosts whose parasites probably have not been previously studied. The Wyoming birds parasitized by ces- todes of the genus Paruterina were a rock wren, Salpinctes o. obsoletus (Say), and a green-tailed towhee, Chlorura chlorura (Au- dubon), which were collected from the same area, along with numerous birds of other species. The wren was collected from the southeast slope of a hill, at an altitude of about 7,000 feet. Sandstone outcroppings were numerous here, and rock wrens were rather commonly observed among them. A marmot, Marmota flaviventris nosophora Howell, was the characteristic mammal of this zone. The towhees were common a few hundred feet lower, where a sage, Artemesia tridentata Nutt., was the characteristic plant. Brewer’s sparrow, Spizella b. breweri Cassin, was also characteristic of this habi- tat. In view of the fact that representatives of the genus Paruterina have not been often reported in North America, it seemed un- 1 Contribution of the 1948 Research Program of the New York Zoological Society at Jackson Hole Wildlife Park. - Now with U. S. Public Health Service, Anchorage, Alaska. 3 Parasitology Section, B. B. Morgan, Project Leader. usual to collect two undescribed species from so small an area. However, since the mor- phological differences are quite distinct, there can be no doubt as to their specific validity. As pointed out earlier (Rausch, 1948), the helminth parasites of the North American avifauna are only poorly known. Paruterina ch lorurae n. sp. (Text-figs. 1-4). Diagnosis: Strobila from 35 to 50 mm. long; greatest width, attained in terminal gravid segments, about 1 mm. Strobila con- sists of about 140 segments; margins of latter not serrate. Segments wider than long in mature segments, with a gradual increase in length as segments become older; terminal gravid segments, in well relaxed strobilae, considerably longer than wide. Scolex about 550 g in diameter, not set off from neck. Suckers rather weakly developed, about 180 g in diameter. Rostellum armed with a dou- ble row of hooks, from 40 to 42 in number; large hooks 20 g long, and small hooks 16 g long. Handle of larger hook about same length as guard and blade; guard of smaller hook relatively shorter, with blade and guard of nearly equal length. Ventral longitudinal excretory canals measure about 33 g in diameter; dorsal canals about 10 g in diameter, median to ventral canals. Transverse canals about 3 g in diameter. Genital pores irregularly alter- nate ; genital ducts pass between longitudinal excretory canals. Musculature well devel- oped; longitudinal bundles numerous. Cirrus sac anterior to vagina, from 105 to 119 g long by 23 to 29 g wide. Cirrus sac does not extend to level of ventral longi- tudinal excretory canal. Internal and ex- ternal seminal vesicles absent. Vas deferens well developed and strongly coiled in area between poral ventral excretory canal and ovary. Testes spherical, from 10 to 12 in number; about 50 g in diameter in mature segments. Testes lateral and posterior to female genital organs, not extending anterior to vagina on poral side, nor antei'ior to ovary on aporal side. 6 Zoologica: New York Zoological Society [34: 2 1949] Rausch & Schiller: North American Cestodes of the Genus Paruterina .7 Vagina runs directly from genital pore toward ovary; poral to latter it enlarges to form a well-developed seminal receptacle. Ovary slightly lobed, about 50 by 60 y in mature segments; situated on mid-line near middle of segment. Vitelline gland spherical to ellipsoidal, about 20 y in diameter; situ- ated at posterior margin of segment on mid- line, directly posterior to ovary. Uterus appears as a crescent-shaped organ ventral to ovary; the arms lengthen until the organ assumes an inverted V-shape. In terminal gravid segments, arms of uterus become somewhat sinuous. Parauterine organ de- velops slowly from anterior margin of early uterus; it becomes elongate and finally at- tains anterior margin of segment. Spherical eggs, observed only in the uterus, measure from 43 to 50 y in diameter. Host : Chlorura chlorura (Audubon). (Green-tailed towhee). Locality : Near Moran, Wyoming. Habitat-. Small intestine. Type : Cotype material has been deposited in the Helminthological Collection of the U. S. National Museum, slide number 46421. Paruterina chlorurae is differentiated from the other species of the genus by shape, size and number of rostellar hooks. Differ- entiation of this species is considered more fully under the discussion below. Paruterina morgani n. sp. (Text-figs. 5-8). Diagnosis: Strobila about 40 mm. long; maximum width, attained in gravid seg- ments, about 500 y. Strobila consists of about 150 segments; margins of latter strongly serrate. Mature segments wider than long ; they increase gradually in length as they become older, with gravid segments being slightly longer than wide. Scolex about 250 y wide, distinctly set off from neck; suckers about 100 y in diameter. Rostellum armed with a double row of 34 to 36 hooks ; large hooks measure 66 y long; short hooks measure 40 y long. Blade of large hook, slightly longer than handle, curves down- ward abruptly at end; guard, near middle of hook, inconspicuous. Blade and handle of small hook nearly equal in length; guard, at middle hook, about Yz as long as blade. Ventral longitudinal excretory canals about 13 y in diameter; dorsal canals about 4 y. Transverse canals about 3 y in diam- eter. Genital pores irregularly alternate; genital ducts pass between longitudinal ex- cretory canals. Musculature well developed ; two rows of longitudinal and a single row of transverse fibers occur in close contact. Longitudinal muscle fiber bundles not numer- ous; accurate count not obtained. Cirrus sac clavate, anterior to vagina; it extends beyond poral ventral excretory canal, and measures from 86 to 105 y long by 16 to 20 y wide. Internal and external seminal vesicles absent. Vas deferens well developed and strongly coiled; convolutions fill greater part of poral half of segment, from end of cirrus sac to level of mid-line of ovary. Testes spherical, from 15 to 18 in number; about 50 y in diameter in mature segments. Testes lateral and posterior to female genital organs; not extending an- terior to vagina on poral side, nor anterior to ovary aporally. Vagina runs directly from genital pore, without convolution, to form a well-devel- oped seminal receptacle posterior and poral to ovary. Ovary unlobed, ellipsoidal; about 120 y long by 80 y wide in mature segments; situated in posterior half of segment, at mid-line. Vitelline gland ellipsoidal; about 60 y long, situated directly behind ovary some- what anterior to posterior margin of seg- ment. Uterus appears as an elongate organ lying transversely in the posterior part of the segment, ventral to ovary. It enlarges gradu- ally, and finally forms an elongate, irregular sac, situated at posterior margin oi com- pletely gravid segments. Parauterine organ grows gradually from anterior margin of uterus; it does not reach anterior margin of segment. Eggs spherical, observed only in uterus; from 36 to 43 y in diameter. Host: Salpinctes o. obsoletus (Say). (Rock wren). Locality: Near Moran, Wyoming. Habitat: Small intestine. Type : Cotype material has been deposited in the Helminthological Collection of the U. S. National Museum, slide number 46422. Paruterina morgani is differentiated from the other members of the genus by size, shape and number of hooks, as well as by other, less obvious details. This cestode is named in honor of Dr. B. B. Morgan, Depart- ment of Veterinary Science, University of Wisconsin. Discussion. At least 18 species have been assigned to the genus Paruterina; of these, 2 species, P. fuhrmanni Baczynska, 1914, and P. meli- erax (Woodland, 1929) have been trans- ferred to other genera. Three of the remain- ing species, P. angustata Fuhrmann, 1906, P. guineensis Joyeux and Baer, 1928, and P. southwelli Hilmy, 1936, have unilateral genital pores, and are immediately separated by this character from the species described in the present paper. Of the North American species, Paru- terina similis (Ransom, 1909) has been re- described by Linton (1927). This species was placed in the genus Paruterina by Jones (1929). Certain morphological details of this species have never been completely de- scribed; Linton (1927, page 50) stated “There is a short rostellum surmounted by a double circle of very short hooks. Their exact number was not satisfactorily made out, but there appear to be in the neighbor- hood of 40 . . .” Jones (1929) examined both Linton’s material, and that of Ransom, but did not give further details concerning the hooks of P. similis. We found that P. similis possesses from 50 to 52 hooks, arranged in 8 Zoologica: New York Zoological Society [34: 2: 1949] a double row. The large hooks measure 13 g long, while the small hooks measure 11 g long. They are essentially the same in form, except that the larger hooks have a larger, more rounded guard (Text-fig. 12). Our ob- servations were made under oil immersion on hooks which had been removed from the scolices, and were lying flat on the slide. Paruterina chlorurae and P. morgani are differentiated from P. similis and P. candel- abraria by hook size, shape and number (Text-figs. 2, 7, 10, 12). It is of interest to note that the North American species can also be separated by differences in the ar- rangement of the parauterine organ and the uterus in the fully-gravid segments (Text- figs. 4, 8, 9, 11). In fact, differences here are more obvious than are those seen in the mature segments. It might also be mentioned here that cestodes of this genus can easily be recognized as such macroscopically, at the time they are removed from the intestine of the host, by the appearance of the gravid segments. The remaining 11 species, widely dis- tributed geographically, are best separated by hook characters. All of these (P. bovieni Iliibscher, 1937; P. bucerotina Fuhrmann, 1909; P. cholodkowskii Skrjabin, 1914; P. daouensis Joyeux, Baer, and Martin, 1936; P. javanica Htibscher, 1937; P. meggitti Johri, 1931; P. otidis Baczynska, 1914; P. parallelipida (Rud. 1809) ; P. purpurata (Dujardin, 1845) ; P. septotesticulata Moghe and Inamdar, 1934; P. vesiculigera (Krabbe, 1882) all differ appreciably in hook size, shape and number. There are also differences in testes number in most cases. Where there is overlapping of this character, hook differences serve to separate the species involved. Other taxon- omic details need not be discussed here in order to separate the species described in the present paper. References. Baczynska, H. 1914. fitudes anatomiques et histologiques sur quelques nouvelles especes de ces- todes d’oiseaux. Bull. Soc. Neuch&t. Sci. Nat., .40:187-239. Fuhrmann, 0. 1906. Die Tanien der Raubvogel. Centralbl. Bakt. Parasit. (orig.), -41:212-213. Johri, L. N. 1931. A new cestode from the grey hornbill in India. Ann. Mag. Nat. Hist., 8, ser. 10, pp. 239-242. Jones, M. F. 1929. Tapeworms of the genera Rhabdo- metra and Paruterina found in the quail and yellow-billed cuckoo. Proc. U. S. Nat. Mus., 75, Art. 20, pp. 1-6. Linton, E. 1927. Notes on cestode parasites of birds. Proc. U. S. Nat. Mus., 70, Art. 7, pp. 1-73. Moghe, M. A. and Inamdar, N. B. 1934. Some new species of avian cestodes from India with a description of Biuterina intricata (Krabbe, 1882). Rec. Ind. Mus., 36: 7-16. Rausch, R. 1948. Observations on cestodes in North American owls, with the description of Choanotaenia speotytonis n. sp. (Ces- toda: Dipylidiinae) . Amer. Midi. Nat. 40 (2) : 462-471. 1949. Paradilepis simoni n. sp., a cestode parasitic in the osprey. Zoologica, 5-4 (1) : 1-3. Skrjabin, K. I. 1914. Vogelcestoden aus Russisch Turkestan. Zool. Jahrb. (syst.) 57:411-492. W OLFFHUGEL, K. 1900. Beitrag zur Kenntnis der Vogelhel- , minthen. Inaug. Diss. Basel. 204 pp. Katz: Behavioral Interactions in Barbary Sheep 9 3. Behavioral Interactions in a Herd of Barbary Sheep ( Ammotragus lervia).1 Irwin Katz. The University of Buffalo. Introduction. Studies of social behavior in animals have generally been of two types, the naturalistic field investigation and the laboratory experi- ment. Field studies of ungulates have been made by Darling (5) on the red deer, and Mills (10), Davis (6) and Spencer (13) on the Rocky Mountain bighorn sheep. The ex- perimental method has produced an exten- sive literature on dominance relationships and aggressive behavior in many species. Collias (4) has reviewed the work on aggres- sive behavior among vertebrates up to 1944. Studies on dominance have been too numer- ous even to be mentioned briefly in the pres- ent paper. Carpenter (2) has pointed out that the development of a science of comparative so- cial behavior requires that the results of field investigations and those of the laboratory should be systematically co-related. He also has stated that the standards of scientific research which apply in the laboratory can and should be applied in the field. Recogniz- ing the research potentialities of an inte- grated approach to animal behavior, Scott (11) recently combined systematic observa- tion and experimentation in a study of a small flock of domestic sheep living under semi-natural conditions. The methods and aims of the following study of a herd of Barbary sheep were sug- gested in large part by the work of Scott and the theoretical discussions of Carpenter (2, 3). Carpenter (2) has listed 11 types of behavioral interactions found in primate societies. It was hoped that the first six of these might be studied in the Barbary sheep. They are: 1, Interactions among adult males of organized groups ; 2, among adult females of organized groups ; 3, between adult males and adult females; 4, between adult males and young; 5, between adult females and 1 A report submitted to the New York Zoological Society on research performed as a Summer Research Fellow of the Society during July, August and September, 1947. The writer is indebted to his colleagues. Dr. N. E. Collias and Dr. B. F. Riess, who made many valuable sug- gestions and participated in the observation and experi- mentation from time to time. Special gratitude is due a third associate, Mr. D. Lehrman, who did most of the work with the Bristol Recorder, and who shared equally in some of the experiments. The Summer Research Fellows worked under the general direction of Professor C. R. Carpenter. young; and 6, among the young. The data were to be compared with information on the domestic sheep and the Rocky Mountain big- horn. In addition, the investigator sought to obtain data relevant to the hypothesis that deprivation is an effective instigator of ag- gressive behavior. The Herd. The Barbary sheep, or aoudad ( Ammotra - gus lervia ) is very distinct in appearance from all other wild sheep, its most unique features being a mane of long hairs over the fore-quarters, the length of tail and the large size of the female’s horns. Its color is uni- form rufous tawny. The habitat of the Barbary sheep is the arid southern slopes and foothills of the mountains of North Africa, extending from near the Atlantic seaboard to Egypt. Lydekker (8) quotes re- ports that the animals go about in groups of four or five and may drink as seldom as once in four or five days. The herd at the New York Zoological Park is descended from stock brought to the Park during the years 1901-1906. No new stock has been introduced since then. During the summer of 1947 the herd consisted of four rams, four ewes and four lambs. One of the lambs was a yearling, while the others were first-season. The sheep lived on an enclosed field of about two acres. They shared the field with two elands and a zebra. Human regula- tion of the activity of the sheep has been kept at a minimum by the Park authorities. Under normal circumstances the herd is fed about one and one-half buckets of grain, which is spread out on a large, flat rock at about 9 :30 every morning. The elands and the zebra usually feed from a box some distance away, although they sometimes wander over to the rock and feed with the sheep. The grain ration is supplemented occasionally with hay, which is placed in one corner of the field. Vending machines in the Park provide spe- cial food pellets which visitors may throw through the fence. During the warm months the sheep regularly gather at the north fence in the afternoon to receive these pellets. The sheep tend generally to avoid the two elands and the zebra. In recent years a newborn lamb was killed by the male eland, and another by the zebra. 10 Zoologica: New York Zoological Society [34: 3 Methods. Casual observations were made almost daily from mid-July until mid-September, and during all hours of the day, so that a complete picture of the daily routine could be obtained. Colored dyes were used to mark the individual animals until the observer could recognize them easily by differences in appearance and behavior. The following ex- periments were performed repeatedly: (1) tossing of bread between pairs to ascertain relationships of dominance-subordination; (2) placing of daily grain ration in a small box to study dominance-subordination rela- tionships in a complex herd situation, as well as to provide observations on food sharing, fighting, and related phenomena; (3) fright- ening of the herd to elicit leadership and timidity. These experiments were carried out from July 22 to September 13. In addition, on two days a Bristol multi-pen recorder was employed in connection with the feed box experiment to ascertain the amount of time each animal actually fed from the box. Mo- tion pictures were taken of types of behavior which had been previously recognized and described. The study of social relationships was limited by the fact that the age and par- entage of adult individuals could not be established with certainty. Although the Park maintains complete records of births and deaths, there is no provision made for identifying individual members of herds. It was possible, however, to know three ewe- lamb relationships on the basis of observed behavior. Daily Routine. The daily pattern of behavior of the herd was marked by fairly regular periods of alternating activity and rest. But this was greatly modified by changes in the weather, experimentation and marked variations in the supply of food from visitors. Usually in the early morning the sheep wandered about the field. Grazing was desultory, since other food was available. At about 9:00 A.M. the sheep generally gathered on an outcropping of broad, flat rocks situated on a hillock in the center of the field, and there they rested until 9:30 A.M., when grain was scattered on a nearby rock by the keeper. The sheep ate the grain peacefully, with very little butting or shoving. By 10:15 A.M. the sheep were either back at their earlier places on the rocks or were under a tree, where they re- mained until after the noon hour. Shortly afterwards children and adults would usually begin to gather along the north fence. The movement of the sheep to the fence seemed to be associated with the size of the gathering of people rather than the hour. On days when very few people visited the Park the sheep might remain entirely away from the fence throughout the after- noon. Once at the fence, the herd remained there as long as pellets were given to them, usually until about 5:30 P.M. On hot after- noons the males made occasional trips to a nearby water hole. Here they cooled them- selves by sinking down into the shallow water and rolling in the mud. In the evening the sheep wandered and rested until dark. The lambs played actively at this time by running and leaping on the rocks. During late August and September fighting and attempted breeding occurred among the males, and most frequently in the evening. At dusk the herd gathered inside or near a shed and bedded down for the night. Sometimes the sheep moved as a group, but consistent leadership was not apparent. In general, there was much independent move- ment among the ewes, rams and lambs. Matching Tests. The matching tests were conducted every few days from July 22 until September 13 to determine dominance-subordination inter- actions between individual animals. Usually the tests were made in the afternoon, when the sheep were gathered at the north fence. By supplying several willing children with bread, and placing them along the fence, it was possible to disperse the sheep so that all or most of the possible pairings could be made among the rams, the ewes and the lambs. The matching of adults and lambs, or of rams and ewes, was not attempted after the first day because of practical difficulties. The matching test was simple. The ex- perimenter stood at the fence and held a small piece of bread in his extended hand. When two sheep, which were not more than ten feet apart, looked in the direction of the experimenter, the bread was tossed so that it landed approximately equidistant between them. No score was recorded unless both animals moved toward the bread. The one which obtained possession of the bread by causing the other to withdraw was consid- ered dominant. The behavior elicited in this situation was clear and unambiguous ; if both sheep advanced toward the food, one always threatened or butted and the other always withdrew. Sometimes the bread landed much closer to the animal known to be subordinate. In such a case the subordinate sheep might obtain the bread, but this was usually fol- lowed by vigorous butts from the dominant animal. Often, however, a quick dash by the dominant sheep caused the other to retreat, even when the bread lay directly at its feet. On three occasions the sheep appeared to be uniformly unmotivated with regard to the bread. Two of these days were extremely hot, and the third was marked by a morning of heavy feeding. At all other times compe- tition was keen and sustained. The method of scoring is somewhat defective in that no scores were recorded for those tests in which only one animal moved toward the food. The assumption here is that of “no contest” and this is, of course, questionable, since the ani- mal’s lack of a positive overt response to the food might be due to the presence of the dominant animal. However, the almost per- 1949] Katz: Behavioral Interactions in Barbary Sheep 11 feet consistency of the results presented in Table I, and the agreement between these re- sults and behavior observed in other situa- tions strongly suggest that the method is highly valid. In the majority of contests, dominance was decided by a sudden tivisting movement of the dominant sheep’s head in the direction of the other sheep. At this “signal” (or sign) the subordinate sheep stopped advancing. Sometimes a token butt was delivered, but seldom was a more forceful attack necessary to effect retreat. Counter attacks by subordi- nates occurred rarely and were never suc- cessful. Among the rams, the ewes and the lambs straight line dominance orders were revealed on the first day and remained al- most stable during the 53-day period of test- ing. Only two instances of reversals occurred during a total of 272 matching tests. Al- though interactions between rams and ewes, and between adults and lambs, were not for- mally tested, it was apparent that all rams were dominant over all ewes, and all adults over all lambs. The results of the matching tests are pre- sented in Table I. The dominance order is as follows: Ram 1> Ram 2> Ram 3> Ram 4> Ewe 1> Ewe 2> Ewe 3> Ewe 4> Lamb 1> Lamb 2> Lamb 3> Lamb 4>. The attempt was to test at least twice a week every pos- sible combination of individuals within each of the three subgroups. But this could not always be done because of the difficulty of bringing certain of the sheep together. For example, matchings between lambs were often disrupted by the sudden approach of one or more adults. Feed Box Experiments. The feed box experiments were intended to furnish information on social behavior in a competitive group situation. The matching tests had indicated the existence of a clear, stable relationship between any two animals which were made to compete for a small food object while in relative isolation from the other members of the herd. But it could not be assumed that these relationships would hold in all types of competitive situa- tions, especially in those where more than two animals are interactive. Maslow (9) found that stable dominance-submission re- lationships which were established between monkeys by the method of paired matching tests broke down when three or more indi- viduals were placed together. The food incentive box was heavy and made of wood, typical of those used in the Park for the feeding of large animals. Its sides were about two feet long and about one and one-half feet high. The box was modified so that the interior sides measured 15" by 12", with the depth remaining un- altered. The interior was large enough to hold more than a bucket of grain without spilling by the feeding animals. The size of the opening was such that two adults could not feed simultaneously without frequent contact, while simultaneous feeding by three Table I. Results of the Matching Tests. Rams* Number of matchings 1-2 25 1-3 20 1-4 22 2-3 22 2-4 21 3-4 12 4-2f 1 2-lf 1 Eives* 1-2 22 1-3 18 1-4 15 2-3 15 2-4 13 3-4 9 Lambs* 1-2 7 1-3 6 1-4 5 2-3 12 2-4 10 3-4 14 Groups Total Rams 124 Ewes 94 Lambs 54 * Numbers indicate sheep according to position in domi- nance order. Number of dominant animal precedes that of subordinate, t Reversal. adults would result in almost constant con- tacts. It was hoped that food sharing, and the conditions surrounding this behavior might result as well as competition for food. The feed box experiment was conducted 12 times. On mornings when the experiments were performed, the experimenter moved the elands and the zebra from the field to adja- cent pastures. At about 9:30 or 10:00 the box was placed on the rock where grain nor- mally was scattered by the keeper. Then the experimenter emptied one bucket of grain into the box and withdrew behind a gate about 20 yards away. The sheep were ob- served by means of binoculars and their be- havior was recorded immediately in a note book. Usually at the end of an hour it was necessary to place more grain in the box. The general pattern of social interaction at the feed box was similar throughout the entire series of experiments. During the se- ries of group tests an order of dominance was formed which conformed closely to that observed during the matching tests. Rams 1 and 2 always dominated the other animals at the box during the first 15 or 20 minutes of feeding. The other sheep milled around the feed box but were not permitted to eat. The two dominant rams ate alternately. Whenever Ram 1 raised his head to chew or rest, Ram 2 ate from the box. As Ram 1 again lowered his head, Ram 2 usually withdrew his head. A high degree of orderliness usu- ally characterized the feeding of these two sheep. Often Ram 2 did not withdraw until 12 Zoologica: New York Zoological Society [34: 3 ha was threatened or mildly butted by Ram 1. In the main, Ram 1 butted and shoved Ram 2, and the latter in turn kept the other sheep from the box. Upon being forced from the box by the more dominant animal, Ram 2 might circle the box and butt all the sheep in his path.2 After the first few minutes the other animals ceased to crowd around the box. Some of them moved to a nearby tree and others formed a wide circle about the feeding place. When he had completed his first feeding, Ram 1 left the feeding area. Ram 2 would either leave at the same time, or continue feeding. Then Ram 3, or Rams 3 and 4 to- gether, moved up to the box, and interactions very similar to the previous ones were ex- hibited. During the first 30 or 40 minutes the box was controlled constantly by a pair of rams. But after the initial feeding of Rams 1 and 2 the pairings shifted frequently due to the movements to and from the box of dominant rams. From time to time ewes and lambs attempted to feed, usually with little success. The subordinate ram of a pair did most of the butting and chasing of the other members of the herd. During the second hour the rams spent less time at the box, and when there mani- fested increasing tolerance toward the lambs and ewes. The order of feeding among the ewes was also determined mainly by domi- nance status, while a lamb’s ability to feed depended on the tolerance of its own ewe. Often Ewe 4 and Lamb 1 (the yearling) ob- tained little or no food during an entire ex- periment. Being of low dominance status, the ewe was excluded, and the lamb likewise because it lacked high dominance maternal protection. The experiment usually ended shortly after the noon hour, when the sheep began to move toward the north fence for pellets offered by visitors. Dominance. In Table II are presented the butts and threats given and received by each sheep during the series of 12 feed box ex- periments.3 Except for two instances, threats were always directed by dominant animals against subordinate ones. The butt more frequently was directed by a subordinate sheep against a dominant one. Nineteen butts, of a total of 198, fall in this category. The data on rams in Table II indicate that the dominance-subordination relationships among these animals were somewhat less rigid and involved more behavioral inter- action among individuals than in the match- ing tests. However, it must not be assumed that relationships at the feed box were less stable. Stability cannot be inferred from the ratio of butts given and received. Nor would 2 These attacks by Ram 2 against subordinate sheep appeared to be clear instances of displaced aggression, and will be discussed later on in this paper. 3 A threat is defined as an aggressive movement or pat- tern of movements which one sheep directs at another, but which does not end in physical contact. The typical threat consisted of a sudden lowering of the head and slight movement toward the other animal. But sometimes a mere lowering and twisting of the head composed the pattern. The object of a threat usually withdrew imme- diately or modified his behavior in some observable way. a mere tabulation of instances of food shar- ing and food hoarding provide a valid basis for inferring dominance status. For example, Rams 1 and 2 usually ate together with little overt indication of dominance-subordination. Often they fed alternately for three or four minutes without observable conflict. But upon closer examination it could be seen that Ram 1 'permitted, Ram 2 to eat with him, and even to shove him occasionally. Over-vigor- ous shoving or persistent crowding on the part of Ram 2 always elicited a sharp attack from Ram 1. Exchanges of butts might be equal in number but they always ended with Ram 1 in control of the box. Wide variations in the “social distance” between different rams are apparent. Reci- procity of aggression was relatively high between Rams 1 and 2 and between Rams 2 and 4, but Ram 3 never aggressed against Ram 1 or Ram 2, and was almost never ag- gressed against by Ram 4.4 The meaning of these differences in “social distance” will become clearer at a later point in the dis- cussion. Among the sheep included in Table II, frequencies of butts and threats decrease in almost perfect rank order. The data on ewes and lambs has not been analyzed in detail because of the low frequency of ag- gressions. Lambs 2, 3 and 4 are not included in the Table only because their scores on both items were zero. Feeding time. Scores for feeding time were computed from the Bristol recordings of an experiment on August 11 according to the method described in the first footnote to Table III. The time score for each sheep is the number of 10-second periods during which the animal had its head in the box for two seconds or longer. In Table III individual scores are given for each of nine consecutive periods. The periods are 16.6 minutes in length. Individual totals for the whole experi- ment indicate a lack of positive relationship between dominance status and feeding time, although the differences between rams and ewes, and adults and lambs, are on the whole substantial. The lack of relation between feeding time and dominance status may be due to wide individual differences both in rate of food intake and nutritional needs. For most of the sheep there is a single period during which much more feeding oc- curred than during other periods. This would seem to justify a comparison of the periods in which individuals made their highest scores. Such scores have been indicated in the Table by a small circle. The circles follow a line which gradually descends from left to right, indicating that time-of-maximum- feeding is closely related to the dominance order. The near-zero scores of Ewe 4 and Lamb 1 have already been discussed. Coordinate-feeding. An important type of social interaction among the sheep was that of coordinate-feeding, or food sharing. The concept and the unit of measurement em- ployed are described in the first footnote to •* See footnote to Table II. 1949] Katz : Behavioral Interactions in Barbary Sheep 13 Table II. Total Butts and Threats for 12 Feed Box Experiments*. Threats Butts Ram #1 Ram #2 Recipients Ram #3 Ram #4 Ewes & Lambs Total Ram #1 Ram #2 Recipients Ram #3 Ram #4 Ewes & Lambs Total Ram #1 X 21 1 2 21 45 X 42 5 5 15 67 Ram #2 0 X 4 2 16 22 9 X 2 13 16 40 Ram #3 0 1 X 8 11 20 0 0 X 14 19 33 Ram #4 0 0 1 X 18 19 0 8 2 X 15 25 Ewe #1 0 0 0 0 5 5 0 0 0 0 17 17 Ewe #2 0 0 0 0 6 6 0 0 0 0 12 12 Ewe #3 0 0 0 0 0 0 0 0 0 0 8 8 Ewe #4 0 0 0 0 0 0 0 0 0 0 3 3 Lamb #1 0 0 0 0 0 0 0 0 0 0 4 4 * Intermittent fighting was observed between Ram 2 and Ram 4. and between Ram 3 and Ram 4, during the second week in September. Butts exchanged during these fights have not been included in the table because on the two days that these fights occurred the experimenter was occupied with taking motion pictures. However, only one Table IV. The unit of measurement is, of course, arbitrary, but it has the merits of being easily computed from the raw data and of yielding quantitative relationships which are in close agreement with observa- tions made during 12 experiments. Table IV contains data on coordinate-feeding among the rams. Both absolute frequencies and co- ordinate-feeding ratios indicate that among those pairings in which the possibility of co- ordinate-feeding existed relatively often its frequency varied widely. The ratios are high for Rams 1 and 2 and Rams 2 and 4, and low for Rams 2 and 3 and Rams 3 and 4. Parallel differences among pairs have already been noted with regard to butts and threats. Ag- gressive interactions and food sharing vary together. But although they are associated, it would be wrong to assume that one is a primary cause of the other. Rather, both are directly related to the amount of social dis- tance between individuals. Social distance is psychological rather than spatial, and is de- fight might be said to have ended in favor of Ram 4. This fight was between Ram 4 and Ram 2. Since these fights were associated with sexual excitement in Ram 4 the table as presented is representative of social relations before the onset of rutting behavior. Fighting and rutting behavior will be discussed in a later section. termined by the willingness-for-contact, or tolerance, of the dominant sheep with regard to the subordinate. Lack of unilateral or bi- lateral aggression between two sheep may be an indication of very low tolerance on the part of the dominant animal. Thus, on the day for which coordinate-feeding data is pre- sented, Rams 3 and 4 were the only rams to- gether at the box for a longer time than Rams 1 and 2 were present togther, yet the only aggression between the former individ- uals was a single threat by Ram 3, and the coordinate-feeding ratio was 4/25. In con- trast, Ram 1 threatened or butted Ram 2 six- teen times and received three butts from the latter, while the coordinate-feeding ratio was 16/21. Quantitative data on coordinate-feeding among the ewes was not obtained because during the experiment on August 11 one or more rams were almost always present at the box. The recorded observations of all 12 ex- periments show that when no ram was pres- Table III. Feeding Time Scores of Sheep in Experiment Lasting Two and One-half Hours.* Periods! Ram #1 Ram #2 Ram #3 Ram #4 Ewe #1 Ewe #2 Ewe #3 Ewe Lamb #4 #1 Lamb #2 Lamb #3 Lamb #4 I 87° 82° 2 3 1 5 0 3 0 0 0 0 II 27 18 54° 8 6 3 0 0 0 0 0 0 III 0 13 28 18 26° 15 14 3 0 9 5 8 IV 10 26 6 39° 7 0 11 0 0 18 0 0 V 0 0 15 12 0 7 7 0 0 11 0 0 V I 9 9 28 25 14 22° 13 0 0 21 4 0 VII 12 0 5 15 12 0 35° 0 0 36° 0 11 VIII 0 28 1 28 0 16 0 0 0 8 11° 0 IX 0 43 19 19 15 7 3 0 2 1 6 14° Totals 145 219 158 167 81 75 83 6 2 104 26 33 * This was the only occasion upon which a Bristol re- cording was made for an entire experiment. The experi- ment was conducted on August 11, and was the sixth in the series of 12. Time scores were computed from the recorded data by counting for each sheep the number of 10-second periods during which the animal had its head in the box for two seconds or longer. This method of scoring, while less accurate than the very laborious pro- cedure of counting actual time in seconds, does not intro- duce a serious bias, in the opinion of the experimenter. t The total time of two and one-half hours was divided into nine periods of 16.6 minutes each. ° Maximum score for a single period. 14 [34: 3 Zoologica: New York Zoological Society Table IV. Incidence of Coordinate-feeding Between Rams During Experiment Lasting Two and One-half Hours.* Pairings Frequency of coord-feeding Highest frequency possiblef Coord-feeding ratio Rams 1 and 2 16 21 16/21 Rams 1 and 3 1 5 1/5 Rams 1 and 4 1 4 1/4 Rams 2 and 3 3 14 3/14 Rams 2 and 4 8 11 8/11 Rams 3 and 4 4 25 4/25 * Coordinate-feeding between two rams is said to occur when both of these rams obtain feeding scores of 3 or more during a 100-second period, while the other two rams obtain scores of zero (c.f. first footnote to Table III for unit of scoring) . The coordinate-feeding data were obtained from the Bristol recordings and synchronized field notes of the feed box experiment on August 11. t The highest frequency possible is the total number of 100-second periods during which a given ram could have engaged in coordinate-feeding with the dominant ram by virtue of the fact that the dominant ram was (a) the only other ram at the box, or (b) accompanied at the box by a ram subordinate to the ram in question. ent Ewes 1 and 2 dominated other animals at the box. Food sharing was common among Ewes 1, 2 and 3. The lambs almost never had exclusive possession of the box, and when at the box they showed no overt aggression.5 6 The amount of time that they fed depended on the tolerance of the lambs by the ewes. Hence, differences in the ability of individ- ual lambs to feed are related to differences in ewe-lamb relationships. Field observa- tions furnished ample evidence of the follow- ing mother-young relationships: Ewe 3 and Lamb 2, Ewe 1 and Lamb 3, and Ewe 2 and Lamb 4.G The yearling was not associated with a ewe in any observable way, either at the box or in the field. There is reason to believe that the very high feeding score of Lamb 2 was due to an especially close and permissive relationship with its mother. Lamb 2 usually stood very close to its own ewe at the box, and fed when- ever she did. Ewe 3 never butted her lamb, although she did not tolerate other lambs. Lambs 3 and 4, on the other hand, did not stay close to their ewes, and often were not at the box when their ewes were feeding. It will be seen in Table III that the feeding scores of Ewe 3 and Lamb 2 are very similar from period to period. The average difference in scores for the same period is only 2.8. Anal- ysis of the Bristol recordings reveals that, of a total of 20 100-second periods during which Ewe 3 fed, her lamb also fed during 17 peri- ods. In contrast with this, Ewe 1 and Ewe 2 fed about as frequently as Ewe 3, but Ewe 1 shared with her lamb only twice, and Ewe 2 never shared with hers. The lack of agreement between the domi- nance status of ewes and their lambs may have been noted. Dominance order among the first-season lambs shows no relation to dom- inance order among the mothers. Of course the number of sheep is much too small to war- rant generalizing, but there are two possible correlates of dominance order among first- season lambs that might be mentioned. First, 5 The lambs often shoved each other, but never threatened or butted when at the box. 6 Sucking and following were the principal behavioral indications of mother-young relationships during July, August, and September. it is possible that dominance among these lambs is related to order of birth, so that lambs born in February tend to be dominant over those born in March or later. There is no way of checking this hypothesis in the pre- sent study, since the birth dates of individu- als are not known. Stewart and Scott (14) have found that age is favorable to domin- ance in a herd of goats. A second hypothesis is that the amount of social distance between a ewe and her lamb will have a direct bearing on the dominance status of the lamb. In the case of Ewe 3 and Lamb 2 extreme social closeness is associated with dominance of this lamb over two other lambs born in the same season. On the other hand, the relationship between Ewe 2 and Lamb 4 (the lamb of lowest dominance sta- tus) was the weakest of the three mother- young relationships. Ewe 2 was the least will- ing to share food with her lamb or to be sucked. Further credence is given to this sug- gestion by Scott’s (11) observations of two orphan lambs which were placed with a flock of domestic sheep. He noted that, “Both or- phans appeared to show less fighting than the other sheep . . . and the ram was not aggres- sive toward other males even in the breeding season.” The hypothesis could be tested in a large herd by testing dominance inter- actions among the ewes and among the lambs over a period of time beginning shortly after the birth of the lambs, and making frequent observations of each ewe with her lamb in isolation from the other sheep. Leadership Recent studies indicate that leadership may be a behavioral characteristic quite un- related to dominance status maintained by fighting. The reports of Darling (5) on red deer, Mills (10) and Davis (6) on Rocky Mountain bighorn sheep, and Scott (11) on domestic sheep, all mention that the usual leader in a herd is an old female. Lack of cor- relation between leadership and dominance interactions has been noted by Allee et al (1) in a flock of ducks, and Stewart and Scott (14) in a herd of goats. In the present study, clear instances of 1949] Katz: Behavioral Interactions in Barbary Sheep 15 leadership occurred only when the sheep were in a conflict situation involving both a source of attraction and a source of danger in close proximity to each other. The source of attrac- tion was the feed box; danger usually was represented by the presence of a strange per- son, such as the experimenter. The experi- menter discovered that if he stood a few yards behind the box after filling it, the sheep would flock into the center of the field and not advance to feed for several minutes. Finally, Ewe 1 slowly moved forward about ten yards and then halted. Her lamb imme- diately ran to her. Then the other ewes and lambs, and finally the rams, moved up to her advanced position. If the experimenter with- drew further from the box the process of ad- vancing and halting under the leadership of Ewe 1 might be repeated several times, until the herd finally reached the box. The pattern of advance might vary from day to day, but Ewe 1 always led the others. Sometimes she advanced 15 or 20 yards in front of the herd before they followed her. Sometimes the en- tire herd moved in single file, with Ewe 1 in the lead and the rams bringing up the rear. The rams rarely came up to the box until Ewe 1 had begun to eat. Leadership was observed 15 times, always in connection with the feed box experiments. Once the rams dashed ahead of Ewe 1 when she was about five yards from the box, and on two occasions Ewe 2 took the lead after Ewe 1 had led most of the way. But at all other times Ewe 1 moved in advance of the others. There is ample evidence that the sheep were following Ewe 1, rather than just moving toward the feed box. Seven times Ewe 1 did not take the most direct path to the box, but turned and walked at right angles to it for several yards. When she did this, the other sheep continued to follow her just as as though she were approaching the box. When no person was in the field, the sheep moved toward the box more or less independ- ently of each other, but the rams still tended to stay behind the ewes. On general grounds it would be expected that boldness and leadership were related. In the present study there were no opportu- nities for observing differences in boldness among the ewes. However, the rams did ap- pear to be more timid than the ewes. A lone ram rarely stayed at the box for more than a few seconds. In all likelihood, when he raised his head from the box and saw that the others had left he would quickly run to where they were. On the other hand, a single ewe might continue feeding alone indefi- nitely. When the zebra, which had been placed in an adjacent enclosure, suddenly galloped toward the fence, the rams were the first to run from the box and the last to return. This was also true when the experimenter inten- tionally frightened the herd. The zebra frightened the sheep away from the box about eight times, and each time Ewe 1 led them back. Thus there is considerable evi- dence that the role of Ewe 1 as leader was not due merely to a greater familiarity with humans. In the field studies cited above, leadership usually was an important factor in the normal moving about of the sheep and deer. But among the Barbary sheep, instances of leadership were quite l'are outside of the spe- cial conflict situations described. In wander- ing and grazing the herd often was scattered widely over most of the field. The rams, ewes and lambs often formed into separate and dispersed sub-groups, yet no consistent lead- ership was apparent in any one of the sub- groupings. What little leading and follow- ing there was occurred between lambs and ewes, and between rams and ewes with the onset of sexual activity. Ewes and lambs. During approximately 80 hours of observation from mid-July to mid- September, each first-season lamb attempted to suck its ewe about 20 or 30 times. The usual duration of sucking was only a few sec- onds, and often the attempt consisted of a single, brief thrust at the udder. Generally the ewe was passive while the attempt was made. Lamb 4, however, often was rejected by its ewe, even though it tried to suck less often than the others. For a while, in fact, its maternal origin was not clear. The lamb alternated between following Ewe 2 and Ewe 4, and twice tried to suck from Ewe 4. Re- peated observations confirmed its relation- ship with Ewe 2. Sucking was accompanied by a certain amount of following of ewes by their lambs. When the herd rested a lamb often lay beside its ewe. The yearling asso- ciated wth the other lambs and joined in the general movements of the herd, but did not favor a particular ewe. Rams and ewes. As the rams began to manifest sexual interest in the ewes they gradually spent more time in the company of the ewes and tended to follow them closely during early morning and evening. Before August 15 relatively little following occurred. Perhaps the following of the ewes by the rams in the conflict situations is, in part, the result of conditioning which develops during the rutting season. Fighting. The Barbary sheep would seem to fall about half way between the domestic sheep and the bighorn with regard to amount of fighting. Scott (11) mentioned pushing and shoving among domestic sheep competing for food in winter, and some butting between rams following the same ewe in heat. In con- trast, Mills (10) stated that fights between big-horn rams in rut might result in “bleed- ing noses, splintered horn tips, limping, and skull fractures.” No complete comparison can be made with the bighorn because the Bar- bary sheep were not observed during the height of rutting. There were no observable injuries, and it is probable that the fighting witnessed was far less serious than that as- sociated with breeding. Two main types of fighting occurred 16 Zoologica: New York Zoological Society [34: 3 among the Barbary sheep. One type con- sisted of a series of head-on charges, usually between rams. The sheep walked away 10 or 15 yards, turned, and walked rapidly toward each other, gradually picking up speed and breaking into a run shortly before they col- lided. Just before impact their heads were lowered and turned slightly to opposite sides. They attempted to meet squarely with their noses crossed. At the beginning of the charge, the sheep got in step and then tried to keep in step until they struck. If one got out of step they broke off the charge and walked away to charge again. Spencer (13) described bouts between bighorn rams which were very similar in detail. He termed such fights “playful” because one ram would not attack if the other was off balance or not pre- pared. A fight of this type between Ram 1 and Ram 2 continued intermittently for 25 minutes on August 25. Before that time fights had never lasted more than five min- utes. The second type of fighting consisted of close butting, and locking and twisting of horns. Usually the sheep stood head to head, facing in the same or in opposite directions, and engaged their horn tips. Each tried to twist the head of the other by pulling down- ward and away. Also, attempts were made to hook the belly or the flank. Fighting of this sort might continue for several minutes. It sometimes started at the feed box as a kind of maneuvering for position. Until mid-August fighting was almost as common among the ewes as among the rams. The usual fight between ewes consisted of a brief exchange of butts, perhaps with lock- ing of horns and twisting. Charges were very rare. Ewes did not fight with rams. From mid-August until the termination of the study fighting among the rams in- creased in frequency, duration and vigor, and was connected with sexual excitement. Rams 2 and 4 were the first to show in- creased pugnacity and sexual behavior. These rams fought with each other and with the other two rams. Sometimes the penis of a fighting ram emerged briefly from its sheath. By September 10 all four rams had reached a high level of sexual arousal and aggressiveness. Most of the fighting took place in the early morning and in the evening, with relatively little aggressive in- teraction at the feed box or during the matching tests. Even when Ram 4 was be- ginning rut and displayed strong aggres- sion toward Rams 2 and 3, he remained rather submissive at the feed box. During the final two days at the box, Ram 4 started fights with Rams 2 and 3. He fought with each male about three times, the average duration of fighting being about two or three minutes. But only once, in a contest with Ram 2, did it appear that he had achieved temporary dominance over his opponent.7 And only once did Ram 4 achieve dominance ~ The butts exchanged have not been included in the data because the experimenter was engaged in taking motion pictures, and so could not take notes. in a paired matching test. His opponent in this test was Ram 2. Possibly in the two experimental situations Ram 4 was inhibited by previous experiences of defeat and sub- ordination. Seward (12) with rats, and Ginsburg and Allee (7) with mice, have shown that an animal could be conditioned to defeat much more readily than to victory. Although they did not study the factor of physical environment, it seems reasonable to expect that aggression would be most strongly inhibited in the place where sub- ordination had been experienced most fre- quently. Play fighting was frequent among the lambs. Often in the evening they scrambled about, pushing and butting each other, in order to gain a position on top of the rocky hillock in the center of the field. This ac- tivity was similar to the children’s game, “king-of-the-hill,” and has been reported by Darling (5) as occurring among red deer fawns. Sometimes fighting in lambs had a more serious appearance. Two lambs might butt head-on forcefully and in rapid suc- cession until both seemed quite exhausted. There was one instance of a lamb fighting with an adult. Late in August Ewe 4 butted Lamb 3 and the lamb immediately butted back. A short fight ensued, ending with the lamb’s retreating and then attacking Lamb 4. Sexual Behavior of the Rams. The first witnessed attempt to mount a female was by Ram 2 on August 10. The ram reared on his hind legs and his penis emerged about three inches from its sheath for a few seconds. The ewe ran off. The ram then tried to mount another ram. Toward the end of August attempted mount- ings by rams of both rams and ewes were common. This behavior was not observed to occur in the ewes or lambs. Female urine had an excitatory effect upon the rams. The ram sniffed the urines, then curled the up- per lip, extended the neck, and tilted the nose in the air. This pattern, according to Spencer (13) is found in bighorns, as well as in other ungulates. Rams sometimes lay on their backs and sucked their penises for short periods. Ejaculation of semen was not observed to occur. Often in the evening, sex- ual activity and fighting occupied the rams continuously until dark. At no time during observation was a ewe receptive. A ram did not persist in attempting to mount the same ewe. Two tries were usually enough to dis- courage him. There was no chasing about the field. A ewe had only to walk or run a few yards to get rid of a ram. Displaced Aggression. One of the most striking behavior pat- terns observed in the course of this study was that in which a sheep responded to a butt, threat or attack from a dominant sheep by delivering in kind to the nearest sub- ordinate. Such “displaced aggression” might continue chain-wise through three or four 1949] Katz: Behavioral Interactions in Barbary Sheep 17 individuals, each successive one being lower in dominance. Thus a ram might butt a sub- ordinate ram away from a piece of bread, the latter might butt a ewe nearby, and the ewe in turn a lamb. At the feed box Ram 2 often responded to a sharp butt from Ram 1 by circling the box and butting all the sheep in his path. The examples of displaced ag- gression are too numerous to be listed. The pattern appeared in rams, ewes and lambs with great frequency. There seemed to be only two factors determining which sheep was to receive a displaced attack, physical proximity and lower dominance status. No special relationships between individuals were apparent, other than the usual ones of dominance-subordination. Winslow (15 > found displaced aggression in cats made to compete for food. Summary and Conclusions. 1. This study represents an attempt to anaiyze social behavior and group organi- zation in a small herd of Barbary sheep. 2. Observations and experiments were made on the herd of four rams, four ewes, and four lambs at the New York Zoological Park during the summer of 1947, under con- ditions with a minimum of human care and interference. 3. In two types of tests it was found that stable relationships of dominance-subordina- tion existed between all individuals, and that the dominance order of all eight sheep was one of straight descent through rams, ewes and lambs. 4. When grain was placed in a small feed box it was found that individual differences in total feeding time at the box were not related to the dominance order. These in- dividual variations probably were due to different rates of food intake and differences in nutritional needs. 5. In the feed box experiments it was found that there was an order of time-of-maxi- mum-feeding which was very similar to the order of dominance. 6. Differences were found in the “social distance” between any two rams when at the feed box. These differences were re- flected in the amount of food sharing that occurred and in amount and reciprocity of aggression. Food sharing and aggressive in- teraction were positively related, and both appeared to be manifestations of the domi- nant animal’s tolerance, or willingness-for- contact with regard to the subordinate. 7. There were individual differences in amount of food sharing and amount of suck- ing among three ewe-lamb pairs. The domi- nance status of the lamb seemed to be asso- ciated with the social distance between the lamb and its mother. But there appeared to be no relation between the dominance status of the ewe and the dominance status of her lamb. 8. Consistent leadership appeared only in conflict situations characterized by a locus of attraction and a locus of danger in close proximity to each other. In conflict situa- tions a ewe always led, and with only two exceptions it was always the same ewe. The rams were more timid than the ewes in strange and potentially “dangerous” situa- tions. 9. Fighting occurred between rams, ewes and lambs. With one exception, there were no fights between ewes and rams, or between adults and lambs. Ewes fought less than rams, while among lambs play-fighting often was observed. 10. As sexual activity appeared in the rams, fighting became more frequent and vigorous. 11. Dominance relationships between rams remained stable throughout the study and from the time of the first appearance of sex- ual interest until the study terminated five weeks later. 12. Sexual activity in the rams consisted of attempted mounting of ewes and rams and sometimes of incomplete masturbation. Ewes were not receptive up to September 14, when the study ended. 13. Instances of displaced aggression were very numerous. The recipient was usually the nearest subordinate animal. Bibliography. 1. Allee, W. C., Allee, M. N., and Castles, E. E. Concerning leadership in a flight of white Pekin ducks (Ab- stract). Bull. Ecol. Soc. of America, 1946, 27, 15-16. 2. Carpenter, C. R. Societies of monkeys and apes. Biol. Symp., 1942, 8, 177- 204. 3. Carpenter, C. R. Concepts and prob- lems of primate sociometry. Socio- metry, 1945, 8, 56-61. 4. Collias, N. E. Aggressive behavior among vertebrate animals. Physiol. Zool., 1944, 17, 84-123. 5. Darling, F. F. A herd of red deer: a study in animal behavior. Oxford University Press, Humphrey Milford, London, 1937. 6. Davis, W. B. Summer activity of moun- tain sheep on Mt. Washburn, Yellow- stone National Park. J. Mammal., 19, 88-94, 1938. 7. Ginsburg, B. and Allee, W. C. Some effects of conditioning on social dom- inance and subordination in inbred strains of mice. Physiol. Zool., 1942, 15, 485-506. 8. Lydekker, R. Wild oxen, sheep and goats of all lands. Rowland Ward, London, 1898. 9. Maslow, A. H. The role of dominance in the social and sexual behavior of infrahuman primates: IV. The de- termination of hierarchies in pairs and in a group. J. Genet. Psychol., 1936, 49, 161-198. 18 Zoologica: New York Zoological Society [34: 3: 1949] 10. Mills, H. B. A preliminary study of the bighorn of Yellowstone National Park. J. Mammal., 1937, 13, 205-212. 11. Scott, J. P. Social behavior, organiza- tion and leadership in a small flock of domestic sheep. Comp. Psychol. Monogr., 1945, 18. 12. Seward, J. P. Aggressive behavior in the rat: II. An attempt to establish a dominance hierarchy. J. Comp. Psychol., 1945, 30, 213-224. 13. Spencer, C. C. Notes on the life history of Rocky Mountain bighorn sheep in the Tarryall Mountains of Colorado. J. Mammal., 1943, 24, 1-11. 14. Stewart, J. C. and Scott, J. P. Lack of correlation between leadership and dominance relationships in a herd of goats. J. Comp. Psychol., 1944, 37, 297-314. 15. Winslow, C. N. Social behavior in cats, 1. J. Comp. Psychol., 1944, 37, 297- 314. Fleming : Pericopidae of British Guiana and Venezuela 19 4. The Pericopidae (Moths) of Kartabo, British Guiana, and Caripito, Venezuela.1 Henry Fleming. Entomologist, Department of Tropical Research, New York Zoological Society. [This contribution is the result of various expeditions of the Department of Tropical Re- search of the New York Zoological Society to British Guiana and to Venezuela, all under the direction of Dr. William Beebe. The Guiana expeditions were made during the years 1917, 1919, 1920, 1921 and 1924. The expeditions were arranged so that each month of the year is represented in the collections. The Venezuelan expedition, in 1942, during which field work was carried on from February 19 to September 2, was sponsored by grants from the Committee for Inter-American Artistic and Intellectual Relations and from four trustees of the Zoologi- cal Society, George C. Clark, Childs Frick, Laurance S. Rockefeller and the late Herbert L. Satterlee, and by invaluable assistance from the Standard Oil Companies of New Jersey and Venezuela.] A total of eight species of Pericopidae were collected at Kartabo and four at Cari- pito. One species from British Guiana and two species from Caripito are new locality records for their respective countries. One species common to both British Guiana and Venezuela is represented by a new race in Venezuela. Eucyane bicolora (Sulzer). Sulzer, Gesch. Ins., t. 22, f. 6 (Expl. Tab.) (1776) (Noctua). Three specimens taken at Kartabo, two on October 11 and one on December 2. The species has been reported from the Guianas, South Brazil and Peru. Eucyane temperata Walker. Walker, List. Lep. Ins. Brit. Mus., 7, p. 1656 (1856). One specimen taken at Caripito on July 11. The species has been reported from South Brazil, Guianas and Colombia, so this is a new record for Venezuela. Pericopis c atilina catilina (Cramer), new status. Cramer, Pap. Exot., 1, t. 79. f. E. F. (1775) (Attacus) . In my opinion Dysschema brotes (Druce) Ann. Mag. Nat. Hist., 15 s. 6, p. 48 1895 ( Anthomyza ) is only the male form and thus a synonym of Pericopis catilina catilina. The males in collections are usually named brotes 1 Contribution No. 827, Department of Tropical Research, New York Zoological Society. and the females catilina. The male specimens match Druce’s description better since the males are usually blackish-brown rather than the cinnamon brown typical of the females. Furthermore, the character given in the lit- erature to separate the genus Dysschema from Pericopis is not valid. This character, the length of the pinnae of the antennae double the width of the shaft, is a male sexual character typical of many of the species of Pericopis. I am not synonomizing the genus Dysschema since the genotype tiresias is not available, but the present generic character does not justify the genus. Thus, on the basis of this character the male specimens of catilina are assignable to Dysschema, hence brotes, and the females to Pericopis, hence catilina. One male captured at Kartabo on May 24. Recorded from Brazil, Guianas and Colombia. Pericopis catilina angustilineata, new sub-species. Length of Specimen Sex Date forewing Type 42487 male March 15 35 mm. Holotype 42488 female March 16 38 mm. Allotype 4239 male March 11 38 mm. Paratype 42489 male April 15 35 mm. Paratype 42490 male March 15 37 mm. Paratype 42491 male June 5 35 mm. Paratype 42492 female April 8 38 mm. Paratype Head as in c. catilina. In the male the length of the pinnae of the antennae is twice the width of the antennal shaft, while in the female the pinnae are barely as wide as the antennal shaft. Ground color of both the fore and hind- wings blackish-brown to cinnamon brown with bands of greenish-yellow. The forewing with two semi-hyaline greenish-yellow bands, one median and the other apical as in c. catilina with brown or brownish-black veins crossing the bands. The bands differ from those of c. catilina in being much reduced; little more or not more than half the width of the bands of c. catilina. If one assumes the nomenclatural type to be ancestral, the reduction of the bands has been caused by the encroachment of the brown or blackish-brown scales on both sides of the bands. This is most easily discerned in the median band. The inner side of the band in c. catilina crosses the wing nearer to the 20 Zoological New York Zoological Society [34: 4: 1949] point where vein C112 forks from the cubital stem than it does in c. angustilineata. Similar results are obtained if one measures basally from the point where vein Cui forks from the cubital stem. In both c. catilina and c. angus- tilineata, but particularly in the latter, the encroachment of the ground color on the yellow bands may be seen. The brown scales on the margins of the bands are duller and lighter than the surrounding ground color. This is variable, being more evident in some specimens than others and occurring indis- criminately along the length of the bands. It is most noticeable and frequent on the me- dian bands. The white spots along the outer margin of the forewing are evident to vary- ing degrees. Their place is taken by the brown or blackish-brown scales making up the background. However, all specimens have a streak of white scales on the outer side of the yellow spot which terminates the apical band in cell Ms. The spots in cells Mi and M2 appear to vanish first since they are faintly discernible in only one specimen. The hindwings of angustilineata, as in catilina, are yellow hyaline for half the length of the wings from the base with a large yellow spot beyond in cells Mi and M2. Peri- copis c. angustilineata differs in that this large yellow spot is separated from the basal patch by a distance approximately twice that of catilina. The part of the basal facies that extends into the proximal part of cell Cui is much smaller in angustilineata than in cati- lina. The ground color along the inner mar- gins of the hindwings encroaches more on the yellow basal area in angustilineata than in catilina. The specimens were all collected during the day while flying. No specimens of Danaidae or Heliconiinae were captured or seen in the same area, though the general appearance and flight of angustilineata is suggestive of various members of either of the above groups. The specimens were captured in an area of about one hundred meters’ diameter near the end of an unmaintained road going to an abandoned oil well “No. 1”. The locality is approximately ten miles west of Caripito, State of Monagas, eastern Venezuela. The area the specimens came from is parched during the dry season, flooded during the wet and characterized by numerous palms and small to very moderate-sized seasonal trees. Pericopis tricolora tricolora (Sulzer). Sulzer, Gesch. Ins., t. 22, f. 5 (1776) (Noctua) . Three specimens taken at Kartabo; two females on March 22 and one male on Novem- ber 24. Recorded from eastern Peru, Ama- zons and Guiana. Dysschema heliconides (Swainson). Swainson, Zool. 111. (2), 3, t. 124, f. 2 (1833) (Anthomyza). One specimen collected at Kartabo in 1920. Recorded from the Amazons, Guianas, Co- lombia and Peru. Hyalurga fenestra (Linnaeus). Linnaeus, Syst. Ent. (ed. 10), 1, p. 505, n. 41 (1758) (Phalaena). One specimen collected at Kartabo which represents a new record since the species has only been reported from Brazil and Peru. Hyalurga sixola Schaus. Schaus, Ann. Mag. Nat. Hist., (8), 6, p. 206 (1910). A total of seven specimens was taken at Caripito; the males on March 20 (two speci- mens), April 5, May 12, and May 16 and the females on April 15 and July 2. Recorded from Venezuela and Colombia. Hyalurga m ysis (Erichson). Erichson in Schoenburgk, Brit. Guiana, 3, p. 606 (1848) (Glaucopis). A female from Kartabo on May 25. Has only been found in British Guiana. Hyalurga modesta Moschler. Verh. Zool.-bot. Ges. Wien, 27, p. 663, t. 9, f. 29 (1877). One female at Kartabo on August 19. Recorded from Guiana and Colombia. Hyalurga partita (Walker). Walker, List. Lep. Ins. Brit. Mus., Het. 2, p. 335, n. 27 (1854) (Dioptis). Four female specimens at Kartabo; two with no date and the remaining two speci- mens on November 21 and December 21. One female at Caripito on March 21. Recorded from Brazil, Guianas, Venezuela and Peru. Goodnight : Phalangids from Rancho Grande, Venezuela 21 5. Report on a Collection of Phalangids from Rancho Grande, Venezuela.’ Clarence and Marie Goodnight. Department of Biology, Purdue University. (Text-figures 1-4.) [This is one of a series of papers resulting from the 45th and 46th Expeditions of the De- partment of Tropical Research of the New York Zoological Society, made during 1945 and 1946 under the direction of Dr. William Beebe, with headquarters at Rancho Grande in the National Park of Aragua, Venezuela. The expeditions were made possible through the generous coop- eration of the National Government of Vene- zuela and of the Creole Petroleum Corporation. [The characteristics of the research area are in brief as follows : Rancho Grande is located in north-central Venezuela (10° 21' N. Lat., 67° 41' W. Long.), 80 kilometers west of Caracas, at an elevation of 1,100 meters in the undisturbed montane rain forest which covers this part of the Caribbean range of the Andes. Adjacent ecological zones include seasonal forest, sa- vanna, thorh woodland, cactus scrub, the fresh- water lake of Valencia and various marine lit- toral zones. The Rancho Grande area is gener- ally subtropical, being uniformly cool and damp throughout the year because of the prevalence of the mountain cloud cap. The dry season ex- tends from January into April. The average humidity during the expeditions, including parts of both wet and dry seasons, was 92.4% ; the average temperature during the same period was 18° C; the average annual rainfall over a five-year period was 174 cm. The flora is marked by an abundance of mosses, ferns and epi- phytes of many kinds, as well as a few gigantic trees. For further details, see Beebe and Crane, Zoologica, Vol. 32, No. 5, 1947. Unless other- wise stated, the specimens discussed in the pres- ent paper were taken in the montane cloud for- est zone, within a radius of one kilometer of Rancho Grande.] This paper is a report on the phalangids collected during the 45th and 46th Expedi- tions of the Department of Tropical Re- search. The species showed relationships to those of Trinidad (Goodnight and Goodnight, 1947), many of the specimens representing the same species. The Cosmetidae, one of the most typical of neotropical families, is here represented by two species, one of which is new. Among the Phalangodidae, the Stygnom- matinae are represented by the wide-ranging Zygobunus rufus (Petrunkevitch) . This was formerly known only from Panama. The sub- family Phalangodinae is represented by Kalina tuberculata Goodnight and Goodnight l Contribution No. 828, Department of Tropical Research, New York Zoological Society. known formerly only from Trinidad. The Triacommatinae are represented by one new species, Vima plana. Among the family Gonyleptidae, three spe- cies are represented, one of which is new. The writers wish to express their appre- ciation to Dr. Beebe and Mr. H. Fleming for making this material available for their study. Types are deposited in the collections of the Department of Tropical Research, New York Zoological Society, New York 60, New York. SUBORDER LANIATORES THORELL. Phalangodidae Simon. Phalangodinae Roewer. Kalina tuberculata Goodnight and Goodnight. Reference: Kalina tuberculata Goodnight and Goodnight, 1947, p. 1, fig. 4. Record: Zone 28, Rancho Grande, Vene- zuela, 1945. Stygnommatinae Roewer. Zygobunus rufus (Petrunkevitch). References: Stygnomma rufum Petrunke- vitch, 1925, p. 62. Zygobunus barronus Chamberlin, 1925, p. 245; Roewer, 1928, p. 546. Stygnommatiplus rufus Roewer, 1928, p. 544. Zygobunus barronus Goodnight and Good- night, 1942, p. 4, figs. 10, 11, 12. Record : Rancho Grande, Venezuela, March 4, 1945. Tricommatinae Roewer. Mima plana sp. nov. (Text-figs. 1 & 2) . Male : Dorsum with five areas, very small paired tubercles on the 1st, 2nd, 3rd and 4th areas. These tubercles are very small and vary in size in different individuals. Cephalo- thorax smooth, with a low tubercle at the posterior lateral portion. Eye tubercle wider than long, with low tuberculations across the median portion. First area of the abdomen without a median line. Boundaries of areas indistinct, not parallel. Lateral margin of abdominal scute smooth, without median ar- mature. 5th area and free tergites each with a few small granulations. Anal operculum smooth, free sternites each with a transverse 22 Zoologica: New York Zoological Society [34: 5 Text-fig. 1. Vima plana sp. nov. Dorsal view of male holotype. row of minute tubercles. Spiracle visible. Coxae with a few small granulations; 1st coxa with a transverse row of spines. First leg slender, unarmed ; 2nd to 4th legs heavier, a few scattered tuberculations on the trochanters. Femora with longitudinal rows of spines, remainder of legs only with scattered hairs. 4th patella with a few apical tubercles. Double claws smooth, without scopula or false claw. Tarsal segments: 9-18- 8-9. Distitarsus of 1st tarsus with 3 seg- ments, 2nd with 3 also. Length of Legs. I. II. III. IV. mm. mm. mm. mm. Trochanter 0.3 0.7 0.8 0.8 Femur 6.3 14.4 11.5 16.2 Patella 0.7 1.2 1.4 1.5 Tibia 4.0 10.8 6.1 8.5 Metatarsus 8.5 17.4 12.4 20.5 Tarsus 1.5 4.5 2.9 3.2 Total 21.3 49.0 35.1 50.7 Palpus with the trochanter 0.8 mm. long, femur 1.1, patella 0.7, tibia 0.6, and tarsus 0.7. Total length, 3.9 mm. Femur armed retrolaterally as in figure. Prolaterally fe- mur and patella each with a median apical spine. Tibia and tarsus armed as on retro- lateral margin. Proximal segment of chelicera with a dor- sal elevation on which are scattered tuber- cles. Distal segment greatly enlarged. Body, chelicerae, and palpi light yellowish with scattered black mottlings. First leg uni- formly colored ; second leg with a white mark at the distal end of the femur; patella black, a white patch at the distal portion of the tibia; third leg with a darker patella but no white markings; fourth leg with a white band following a black band on the distal portion of the tibia. Legs otherwise uni- formly dark brown to dusky. Female: Similar in appearance to male. Measurements in mm.: Male, total length 3.7 ; cephalothorax 1.5; width at widest por- tion 2.6. Female, total length 7.2; cephalo- thorax 1.7 ; width at widest portion 4.2. Record : Male holotype from Rancho Grande, Venezuela, July 22, 1945; paratypes from same locality, March 22, 1945, and July 22, 1945. Remarks: Vima plana is related to Vima insignis Hirst. It differs from this latter spe- cies by lacking the raised area of the dorsum and the paired low tubercles over the eye. Cosmetidae Simon. Cosmetinae Cambridge. Cynorta estebana Roewer. Reference: Cynorta estebana Roewer, 1947, p. 18, pi. 18, fig. 66. Record: Rancho Grande, Venezuela, Au- gust 9, 1945. Cynorta bromeliaca sp. nov. (Text-fig. 3). Male: Eye tubercle wider than long. 1st area with a pair of enlarged tubercles, 3rd area with a pair of robust spines which are short and heavy at the base. Remaining areas and free tergites without median armature. Entire dorsum covered with small white ele- vations. These are more numerous on the lateral posterior portions of the scute. Each free tergite with a transverse row of these same tubei’culations. Anal operculum with only a few granulations, free sternites each with a transverse row of hair-tipped granu- lations. Coxae and genital operculum smooth except for scattered hairs. A few teeth on the anterior margins of the 3rd and 4th coxae, and a transverse row of granulations across the 1st coxa. Text-fig. 2. Vima plana sp. nov. Retrolateral view of palpus of male holotype. 1949] Goodnight: Phalangids from Rancho Grande, Venezuela 23 Text-fig. 3. Cynorta bromeliaca sp. nov. Dorsal view of male holotype. Legs clothed only with hairs except for a few tuberculations at the apical portion of the 4th patella; femora straight. Tarsal seg- ments: 6-14-11-12. Distitarsus of both 1st and 2nd tarsi with 3 segments. Proximal por- tion of 1st tarsus enlarged. Length of Legs. I. II. III. IV. mm. mm. mm. mm. Trochanter 0.6 0.8 0.8 0.8 Femur 4.4 6.6 6.6 9.3 Patella 1.0 1.6 1.6 1.6 Tibia 2.8 8.3 3.8 5.3 Metatarsus 4.3 10.3 5.5 9.4 Tarsus 2.6 5.2 3.6 4.4 Total 15.7 32.8 21.9 29.8 Palpus with the trochanter 0.8 mm. long, femur 1.4, patella 0.9, tibia 1.5, and tarsus 0.8. Total length, 5.4 mm. Palpus character- istically flattened with a ventral row of teeth on the femur. Proximal segment of chelicera with a dor- sal elevation on which are a few granulations. Distal segment somewhat enlarged. Dorsum reddish-brown, thickly covered with white spots which are more numerous on the lateral and posterior portions of the scute. These form an irregular band of white spots, with a few scattered ones in the median area. Several white spots on the eye tubercle. A transverse row of white spots across each free tergite. Anal operculum without markings. Venter, coxae, and cheli- cerae reddish-brown with darker markings. Legs yellowish, trochanters, and bases of fe- mora, patellae, and tibiae reddish-brown. Metatarsi white. Measurements in mm.: Male, total length 6 ; cephalothorax 1.9 ; width of body at widest portion 4.1. Record : Male holotype from bromeliads, Rancho Grande, Venezuela, August 8, 1946. Remarks: This species is most nearly re- lated to Cynorta catenulata Roewer. It differs from this latter species by having an entirely different pattern of white on the dorsal scute. Gonyleptidae Sundevall. Cranainae Roewer. Poecilocranaus gratlosus Roewer. Reference: Poecilocranaus gratiosus Roewer, 1943, p. 54, pi. 7, fig. 63. Record : Rancho Grande, March 22, 1945. Santinezia albilineata Roewer. Reference : Santinezia albilineata Roewer, 1932, p. 290, fig. 7. Record: Rancho Grande, Venezuela, Au- gust 1, 1945. Stenostygninae Roewer. Bunistygnellus beebei sp. nov. (Text-fig. 4) . Male: Dorsum smooth, cephalothorax without a median eye tubercle. Eyes widely separated near the posterior portion of the cephalothorax. Between the eyes a large rounded elevation with a short apical spine. Elevation granular. Anterior margin of the cephalothorax with a short anterior projec- tion between the chelicerae and palpi. A large vertical spine at the anterior margin near the coxa of the palpus. A small tubercle at the anterior lateral margin. Abdomen with five areas. 1st area constricted in the middle, 2nd area narrow, 3rd area with a pair of large spines, 4th and 5th areas unarmed. Free tergites smooth, unarmed. Lateral mar- gins of scute smooth, anal operculum smooth. Free sternites each with a transverse row of hair-tipped tubercles which are enlarged into spines at the lateral margin. Spiracles widely open. Coxae covered with hair-tipped tubercles. A transverse row of spines across the 1st coxa. 3rd coxa with anterior and pos- terior teeth. 4th coxa only slightly projecting and with a large dorsal apical spine. Trochanters globular. 2nd and 3rd tro- chanters tuberculate, 2nd with two small dorsal apical spines, 3rd with a posterior apical spine. 4th trochanter very heavy, with a large dorsal apical spine and a lateral spine on each side ; covered with smaller tubercles. 1st and 2nd femora clothed only with hairs. 3rd femur covered with hairs and tubercles, ventrally with 2 long rows of spines and with 2 dorsal apical spines. 4th femur tuber- culate, ventrally with 2 rows of very large spines, dorsal-apically with 2 large spines. Remaining segments of 1st and 2nd legs clothed only with hairs. Patella of 3rd leg tuberculate with a large ventral apical spine. 3rd tibia with 2 ventral rows of tubercles at the distal third, remainder of 3rd leg un- armed. 4th patella heavily tuberculate and with large apical spines. 4th tibia clavate, with 2 ventral rows of spines at the apical third. Remainder of leg clothed only with hairs. 3rd and 4th tarsi with heavy scopulae, double claws toothed. Tarsal segments: 7- 24 Zoologica: New York Zoological Society [34: 5: 1949] 16-8-9. Distitarsi of both 1st and 2nd tarsi with 3 segments. Length of Legs. I. II. III. IV. mm. mm. mm. mm. Trochanter 0.7 0.8 1.0 1.0 Femur 3.0 4.8 3.8 4.1 Patella 0.9 1.3 1.8 2.0 Tibia 2.0 4.3 2.6 3.3 Metatarsus 3.2 5.0 4.0 5.0 Tarsus 1.4 4.8 2.7 3.1 Total 11.2 21.0 15.9 18.5 Palpus with the trochanter 1.2 mm. long, femur 3.6, patella 1.6, tibia 1.9, and tarsus 1.9. Total length, 10.2 mm. Coxa with scat- tered granulations. Trochanter globular with a dorsal elevation, with a small dorsal and a small ventral spine. Femur curved, un- armed except for a small basal ventral tu- bercule. No dorsal apical or median apical spine. Patella unarmed, tibia and tarsus each with 5 hair-tipped spines on either side. Tarsal claw long and curved back against the tarsus. Chelicera greatly enlarged, proximal seg- ment with a dorsal elevation, with several small tubercles dorsal and ventral. A large retrolateral spine at the apical portion. Distal segment huge, elevated considerably over the proximal segment. Distal segment smooth. Dorsum reddish-brown. Spines and eleva- tion of the cephalothorax likewise reddish- brown. Eyes black. Lateral portions of 1st \i to 3rd segments with a large white blotch. Within these U II white areas 1 or 2 circles of ' / reddish-brown. 4th area with / / a transverse line of white, end- ing in a wider area at the ^/ / lateral edge. 4th area with a / / posterior margin of white. A // narrow line of white at the y lateral margin of the scute, ex- tending from the anterior por- tion of the cephalothorax to the 2nd area. A large white spot on the lateral margins of the 3rd and 4th areas. Each free tergite with the poster- ior margin irregularly lined in white. Last free sternite lined with white. Venter and coxae reddish-brown, basal portions of 3rd and 4th legs likewise reddish-brown. Palpus, 1st and 2nd legs, and distal portion of 3rd and 4th legs yellowish, penciled with gray. These markings give an annulate appearance on the metatarsi. Chelicera dark reddish- brown, with netted dark markings. Measurements in mm. : Male, total length of body, 5.8 ; cephalothorax, 3 ; width of body at widest portion, 4. Record : Male holotype from Rancho Grande, Venezuela, 1945. Remarks : This species is related to Buni- stygnellus macrochelis Roewer, differing principally in the color markings on the dor- sum. References Cited. Beebe, W., and Crane, J. 1947. Ecology of Rancho Grande, a Sub- tropical Cloud Forest in Northern Venezuela. Zoologica, Vol. 32, No. 5, pp. 43-60. Chamberlin, R. V. 1925. Diagnoses of New American Arach- nida. Bull. Mus. Comp. Zool., Harvard Coll., Vol. 67, pp. 211-248. Goodnight, C. J. and Goodnight, M. L. 1942. Phalangida from Barro Colorado Island, Canal Zone. Amer. Mus. Novi- tates, No. 1198, pp. 1-18. 1947. Studies of the Phalangid fauna of Trinidad. Amer. Mus. Novitates, No. 1351, pp. 1-13. Petrunkevitch, A. 1925. Arachnida from Panama. Trans. Conn. Acad, of Arts & Sciences, Vol. 27, pp. 51-248. Roewer, C. Fr. 1927. Weitere Weberknechte II. Abh. Naturw. Verein Bremen. Vol. 26, no. 3, pp. 527-632. 1932. Weitere Weberknechte VII. Archiv. filr Naturg. n.s., Vol. 1, no. 2, pp. 275- 350. 1943. Uber Gonyleptiden, Weitere Weber- knechte XI., Senckenbergiana, Vol. 26, pp. 12-68. 1947. Diagnosen neuer Gattungen und Arten der Opiliones Laniatores (Arach.), Senckenbergiana, Vol. 28, pp. 7-67. Crane: Crabs of the Genus Pseudothelphusa from Venezuela 25 6. Fresh-water Crabs of the Genus Pseudothelphusa from Rancho Grande, Venezuela.1 Jocelyn Crane. Research Zoologist, Department of Tropical Research New York Zoological Society. (Text-figures 1-3). [This is one of a series of papers resulting from the 45th, 46th and 47th Expeditions of the Department of Tropical Research of the New York Zoological Society, made during 1945, 1946 and 1948 under the direction of Dr. William Beebe with headquarters at Rancho Grande in the National Park of Aragua, Venezuela. The expeditions were made possible through the gen- erous cooperation of the National Government of Venezuela and of the Creole Petroleum Cor- poration. [The characteristics of the research area are in brief as follows: Rancho Grande is located in north central Venezuela (10° 21' N. Lat., 67° 41' W. Long.), 80 kilometers west of Cara- cas, at an elevation of 1,100 meters in the undis- turbed montane cloud forest which covers this part of the Caribbean range of the Andes. Ad- jacent ecological zones include seasonal forest, savanna, thorn woodland, cactus scrub, the fresh water Lake Valencia, and various marine littoral zones. The Rancho Grande area is generally subtropical, being uniformly cool and damp throughout the year because of the prevalence of the mountain cloud cap. The dry season ex- tends from January into April. The average humidity during the expeditions, including parts of both wet and dry seasons, was 92.4%; the average temperature during the same period was 18° C.; the average annual rainfall over a 5-year period was 1 14 cm. The flora is marked by an abundance of mosses, ferns and epiphytes of many kinds, as well as a few gigantic trees. For further details, see Beebe & Crane, Zoologica, Vol. 32, No. 5, 1947. Unless otherwise stated, the specimens discussed in the present paper were taken in the montane cloud forest, within a radius of 1 kilometer of Rancho Grande.] General Account. Two species of Potamonidae, both belong- ing to the genus Pseudothelphusa, live within the Rancho Grande area and are common near the laboratory at an altitude of about 3,500 feet. One, P. garmani, occurs also farther down, near stream-beds in semi- evergreen seasonal forest, at least to 2,000 feet. The other, the apparently new P. chacei, is confined to the cloud forest. Each occurs on both the Caribbean and Valencia sides of the Cordillera. 1 Contribution No. 829, Department of Tropical Research, New York Zoological Society. . The general habits are similar in both. Each is occasionally found walking along the damp forest floor, at considerable distances from the small mountain torrents. More rarely the crabs are seen submerged in the streams themselves, clinging tightly to the rocks and moss with their spiny feet. In both species the young are carried principally dur- ing the dry season, in March and early April, three females with young having been taken during that season, as well as all of the very small free-living young in the collection. In contrast only one female taken during the rains, in June, carried young. The single female with spermatophores was captured in September; by inference the eggs are laid between October and February, the season during which we have not visited the labora- tory. The food of both species consists largely of insects, especially beetles. The remains of eaten crabs are found frequently, and it seems certain that these crustaceans form an important food item in the diet of such local animals as tayras and opossums. My sincere appreciation goes to Dr. Fenner A. Chace of the United States National Mu- seum for his kindness in determining the identity of P. garmani and the systematic status of P. chacei. The specimens are in the collections of the Department of Tropical Research of the N. Y. Zoological Society, ex- cept for two examples of each species which have been deposited in the United States National Museum. Text-figs. 3A and 3B are the work of Miss Pamela Marmont; the re- mainder are by Miss Louise Moore. Field Key to the Rancho Grande Species of Pseudothelphusa. A. Manus of cheliped with a large tubercle at base of fingers ; anterior part of cara- pace slightly rough to touch; adults large, measuring at least three inches across, females 2% inches across still having narrow abdomens; marsupial young numbering between 200 and 300, their front distinctly bilobed and con- vex ; semi-evergreen seasonal and cloud forests garmani [34: 6 26 Zoological New York Zoological Society A D Text-fig. 1. Pseudothelphusa garmani. Adult male, length 47 mm. A, right first abdominal appendage, extero-posterior view; B, same, extero-anterior view; C, same, postero- internal view; D, major cheliped, external view. AA. Manus of cheliped without tubercle at base of fingers; anterior part of cara- pace smooth to touch ; adults small, ma- ture females measuring about IV2 inches across, males less; marsupial young numbering between 20 and 40, their front scarcely lobed, truncate; cloud forest only chacei, sp. nov. Pseudothelphusa garmani Rathbun, 1898. (Text-fig. 1, 2A) . Reference. Pseudothelphusa garmani Rath- bun, 1898, p. 522, text-fig. 14. Color in life. Adults of both sexes : variable shades of dark brown ; color of carapace uni- formly distributed, the chelipeds, ambula- tories and abdomen slightly lighter. Marsu- pial young : carapace dark brown anteriorly, paler behind; chelipeds apricot buff (Ridg- way), deepest on upper merus, carpus and upper half of manus ; chelae creamy or white ; sides of carapace light brown to buff ; ster- num and abdomen white; ischium and tro- chanter of ambulatories buff, other segments dark brown. Development. Two females carried 258 and 260 young, respectively. An example, illus- trated in Text-fig. 2A, measures 3.5 mm. in length, 4.7 in breadth. The general form is closely similar to that of the adult, but the front has each of the two distinct lobes more convex and there is no trace of a tubercle outside the manus at the base of the fingers; manus not swollen. No very small free-living examples of this species were taken, but a young female 29 mm. long has the tubercle distinct. Food. Three large examples all contained comminuted black chitin, showing in one case unmistakable beetle elytra; in addition one stomach held soft animal matter, probably worm tissue. 1949] Crane: Crabs of the Genus Pseudothelphusa from Venezuela 27 Habitat and Range. Taken at Rancho Grande from semi-evergreen seasonal and cloud forests, between 2,000 and 3,800 feet. Previously known also from near Caracas, Venezuela, and from Trinidad. Material. A total of six specimens, not counting marsupial young, have been pre- served. Department of Tropical Research Cat. Nos. 4635, male, length 47 mm.; cloud forest, Mai*. 28, 1946; 4626, female, 55 mm., Rancho Grande verandah, June 24, 1946, with 260 marsupial young, No. 4626a; No. 45449, 2 immature females, 29, 42.5 mm., cloud forest, April 1, 1945. U. S. Nat. Mus. Nos. 82379, male, length 38.5, cloud forest, Mar. 18, 1946; 82380, female, with spermato- phores, length 42.5, semi-evergreen seasonal forest (2000 feet), Sept. 6, 1946. Pseudothelphusa chacei sp. nov. (Text-figs. 2B, 2C, 3). Diagnosis : Superior margin of front dis- tinct and tuberculate but not carinate; bi- lobed, with median suture present; cervical suture nearly straight; carapace scarcely convex, smooth; exognath of maxillipeds re- duced to a stump; manus somewhat swollen; no tubercle at base of fingers; front low; male matures at length of 17 mm., female at about 23. Description. Carapace slightly convex, regions scarcely elevated ; gastric region slightly less elevated than branchial; ante- rior margins of protogastric lobes distinct but not prominent; depressions defining an- terior part of mesogastric region scarcely indicated; median furrow less well defined than in garmani ; cervical groove nearly straight, deep, continued practically to lateral margin. Carapace almost completely smooth to the touch, with only a hint of microscopi- cally fine granules in antero-lateral regions. Antero-lateral margins with a small orbital tooth followed by a slight gap; behind this are about 23 to 25 fine teeth, similar and close-set. Front low; superior margin dis- tinct but not carinate, almost truncate, bi- lobed, finely tuberculate, scarcely or not at all projecting over the vertical surface; in front view slightly depressed in middle; lower margin sinuous, the tubercles tending to be obsolete. Orbits nearly filled by eyes, margins almost smooth, the superior slightly sinuous. Maxillipeds substantially as in garmani and simoni. Merus of chelipeds finely rugose above, the inner margin armed with a single line of stout, graduated teeth, the lower and ventro- distal margins by beaded granules. Carpus smooth except for dorso-inner surface which is armed by a crest of small, distinct teeth, of which the usual large tooth is only one exaggerated element. Major and minor manus smooth in both sexes, save for scat- tered microscopic granules and punctations on outer surface and faint rugosities dor- sally. Major manus, especially in male, defi- nitely swollen ; upper and lower margins in all slightly convex; sinus at base of pollex C Text-fig. 2. Young crabs taken from ab- dominal pouches of female Pseudothelphusa. A, P. garmani, carapace, length 3.5 mm.; B, P. chacei, carapace, length 3.65 mm. (drawn to same scale as A) ; C, same, right cheliped, outer view, length 2.9 mm. practically lacking; no tubercle at base of fingers, although there may be a very slight swelling of the margin at that point; fingers moderately broad, slightly and irregularly punctate, and near dorsal surface of dactyl, tuberculate; prehensile edges practically in contact, the teeth broad, irregular and vari- able. Merus of all ambulatories flattened, with upper margins convex and finely den- ticulate ; superior margin of carpus and both margins of manus microscopically spinulous; dactyli slender and spined. Male abdominal appendage illustrated (Text-fig. 3). Color in Life. Adults variable and capable to a certain extent of color change, the cara- pace ranging from a dull red to dark brown. A female with marsupial young had the cara- 28 Zoologica: New York Zoological Society [34: 6 Text-fig. 3 [Part]. Pseudothelphusa chacei male holotype, length 17.5 mm. A, dorsal view; B, major cheliped, external view; C, right first abdominal appendage, extero- posterior view; D, same, extero-anterior view; E, same postero-internal view. pace snuff brown (Ridgway) , slightly lighter posteriorly. Chelipeds chiefly cinnamon buff to clay color with snuff brown on dorsal ridge; fingers pale buff; ambulatories snuff brown with darker segment markings. Her young were apricot buff in general coloring; front much darker, almost black; ambula- tories cinnamon rufous; dorsal ridges of chelipeds apricot buff ; underparts pale buff. Measurements. Holotype male, No. 461197 : Length of carapace 17.5 mm.; breadth 28.5; depth 11.5; width of front, lower margin, 7 mm.; major manus (measured along lower margin) plus pollex 26. Para type female, No. 45105: Length of carapace 23 mm.; breadth 38.5 ; depth 15 ; width of front, lower margin, 8.4; major manus (measured along lower margin) plus pollex 30. Development. Two females carried 22 and 40 marsupial young, respectively. An ex- ample, illustrated in Text-fig. 2B, measures 3.65 mm. long by 5.0 mm. broad. Compared with the corresponding stage of P. garmarvi the front is scarcely lobed and strongly trun- cate, the eyes are larger and the cheliped manus is distinctly swollen; the latter dif- ference is carried through into the adult. Behavior of Young. Samples of the young, when removed from one of the mothers and placed close to her, were almost helpless ; they could move feebly, but did not try to climb back into the pouch ; one, however, attempted to grip one of her legs as she moved slowly past. Food. A female with young, when captured in an open patio of Rancho Grande, was hold- ing a large, black tenebrionid beetle (D.T.R. No. 45,107) in her major chela, and pulling off the legs with her left. The beetle was uncrushed, being strong and active when liberated. Two of three stomachs examined contained black chitin, including small beetle elytra, in addition to indeterminate soft ani- mal matter; the third was empty. Habitat and Range. Known only from Rancho Grande, in the National Park of Aragua, Venezuela, in the montane cloud forest, between about 3,000 and 3,800 feet. Affinities. Dr. Chace writes as follows re- garding this species : “It is very close to Miss Rathbun’s P. simoni. The male abdominal appendages agree very well with her figure of that species, but your specimens have a somewhat less convex carapace and a sharply carinate upper frontal margin which is com- pletely lacking in P. simoni. The general ap- pearance of your material would indicate full specific distinction from P. simoni, but the similarity in the male appendages sug- gests that possibly it deserves only subspe- cific rank.” P. simoni is known only from the types, taken from Colonia Tovar and Caracas, Venezuela, and from the “Antilles” (Clau- dius). Material. A total of 13 specimens was taken, not including marsupial young. The following have been designated as types: Holotype male, Department of Tropical Re- 1949] Crane: Crabs of the Genus Pseudothelphusa from Venezuela 29 Text-fig. B [Part]. Pseudothelphusa chacei male holotype, length 17.5 mm. A, dorsal view; B, major cheliped, external view; C, right first abdominal appendage, extero- posterior view; D, same, extero-anterior view; E, same postero-internal view. 30 Zoologica: New York Zoological Society search Cat. No. 461197, length 17.5 mm., Rancho Grande court, March 5, 1946; para- type female, D.T.R. No. 45105, 23 mm., with 21 young, Rancho Grande court, March 29, 1945; paratype male, United States National Museum No. 87067, 17 mm., Water Trail, March 15, 1945; paratype female, U.S.N.M. No. 87067, 24 mm., Water Trail, March 7, 1945. In addition, the following were taken, all retained in the collections of the Depart- ment of Tropical Research: No. 45106, 1 fe- male, length 22 mm., and 3 young, 7-16 mm., cloud forest, March 20-April 20, 1945; No. 4636, 1 female, length 22 mm., with 40 young, cloud forest, March 14, 1946; No. 461198, 4 [34: 6: 1949] young, 7-9 mm., cloud forest, March 1-15, 1946. It gives me great pleasure to name this species for Dr. Fenner A. Chace, Jr. References. Rathbun, M. J. 1898. A contribution to a knowledge of the fresh-water crabs of America. — The Pseudothelphusinae. Proc. U. S. Nat. Mus., Vol. 21, pp. 507-537, (No. 1158). 1905. Les crabes d’eau douce (Potamonidae) . Nouv. Arch. Mus. Paris, ser. 4, Vol. 7, pp. 159-321. Crane: Salticid Spiders; Systematics and Behavior in New Species 31 7. Comparative Biology of Salticid Spiders at Rancho Grande, Venezuela. Part III. Systematics and Behavior in Representative New Species.1 Jocelyn Crane. Research Zoologist, Department of Tropical Research, New York Zoological Society. (Text-figures 1-8). [This is one of a series of papers resulting from the 45th, 46th and 47th Expeditions of the Department of Tropical Research of the New York Zoological Society, made during 1945, 1946 and 1948, under the direction of Dr. William Beebe, with headquarters at Rancho Grande in the National Park of Aragua, Venezuela. The expeditions were made possible through the generous cooperation of the National Govern- ment of Venezuela and of the Creole Petroleum Corporation. [The characteristics of the research area are in brief as follows: Rancho Grande is located in north-central Venezuela (10° 21' N. Lat., 67° 41' W. Long.), 80 kilometers west of Cara- cas, at an elevation of 1,100 meters in the undis- turbed montane cloud forest which covers this part of the Caribbean range of the Andes. Ad- jacent ecological zones include seasonal forest, savanna, thorn woodland, cactus scrub, the fresh water Lake Valencia, and various ma- rine littoral zones. The Rancho Grande area is generally subtropical, being uniformly cool and damp throughout the year because of the preva- lence of the mountain cloud cap. The dry season extends from January into April. The average humidity during the expeditions, including parts of both wet and dry seasons was 92.4%; the average temperature during the same period was 18° C.; the average annual rainfall over a 5-year period was 174 cm. The flora is marked by an abundance of mosses, ferns and epiphytes of many kinds, as well as a few gigantic trees. For further details, see Beebe & Crane, Zoolog- ica, Vol. 32, No. 5, 1947. Unless otherwise stated, the specimens discussed in the present paper were taken in the montane cloud forest zone, within a radius of 1 kilometer of Rancho Grande.] Contents. Page Introduction 31 Subfamily Lyssomaninae Lyssomanes bradyspilus sp. nov 31 Subfamily Salticinae Semorina brachychelyne sp. nov 35 Semorina megachelyne sp. nov 38 Ashtabula furcillata sp. nov 39 Sassacu8 flavicinctus sp. nov 41 Sasmcus ocellatus sp. nov 44 Phiale flammea sp. nov 47 Mago dentichelis sp. nov 49 References r, 1 Contribution No. 840, Department of Tropical Research. New York Zoological Society. Introduction. The eight species described in the present paper have been selected from among other Rancho Grande salticids for two reasons. First, they represent a number of different stages and directions in salticid evolution; and, second, special experimental display data and/or examples of the earliest instars have been assembled in each. Part I of this series (Crane, 1948.1) dealt monographically with several species of Corythalia, while Part II (1948.2) described the methods of study. In the succeeding parts, which will be based largely on Corythalia and the present group of species, it is proposed to discuss the re- leasing mechanisms of display, to compare post-embryological development and, finally, to evaluate evolutionary trends. With the exception of Text-figure 8F, which was drawn from life at Rancho Grande by Mr. Kenneth Gosner, all the illustrations are the work of Miss Louise A. Moore. The types are deposited in the collections of the Department of Tropical Research, New York Zoological Society, New York 60, N. Y. Lysso manes bradyspilus sp. nov. (Text-fig. 1), Diagnosis : Retromargin of fang groove with 6 teeth, the 2 proximal minute, none crowded toward fang base ; basal segment of chelicera in male with cluster of 3 to 7 dorsal distal spines; fang toothless; no fringes on first metatarsi, which are straight; no clus- ter of dorsal tibial spines on palp, its distal apophysis very small, blunt; bulb with three strong, spinous, distal processes; epigynum with two pairs of large rounded bodies dis- tinct, the anterior pair the smaller and prac- tically contiguous. Abdominal black spots present or absent. Color. Color in Life : Adult male. Cephalothorax : Integument of carapace translucent green, without dark pigment, varying from a yel- lowish-green, especially in recently molted 32 Zoologica: New York Zoological Society [34: 7 examples, to apple green ( Ridgway) . Ocular quadrangle including black eye tubercles with varying amounts and proportions of yellowish- or silvery white and orange-red scale hairs, the latter usually placed anteri- orly. AME rimmed with silvery-white; the eyes themselves clear apple green, shifting to black (see under Behavior) ; other eyes black. A narrow submarginal clypeal band of orange-red scale hairs, directed downward. Chelicerae fangs brown. Palpal bulbs pink- ish to orange. Integument of legs translucent apple green, without dark pigment except for black tarsal pads. Abdomen : Integument translucent green, sparsely covered with short hairs, ranging from apple green to dull green-yellow, usually with a short, me- dian basal stripe of darker green. Hairs short, rather sparse, of same color as integu- ment. Paired, subdermal black spots on pos- terior half of abdomen present or absent, strong or weak, rarely appearing — if at all — until three or four days after final molt; any number up to four pairs may develop. A patch of white hairs often present at distal end of dorsum. Adult female. Differs from male as fol- lows: Scale hairs of ocular quadrangle, in- cluding eye tubercles, tend to be more uni- formly yellowish- or silvery white, with the orange-red reduced or absent, except for a variable, sometimes conspicuous, crest be- hind AME ; subdermal clypeal band of orange-red absent, replaced by a band of scant white hairs; palps completely green; legs usually with some dark subdermal pig- ment concentrated near joints ; this often is confined to a single spot in antero-distal part of first tibia. As in the case of the abdominal spots, it develops, if at all, after the final molt. No female seen with more than two pairs of abdominal spots ; as in the male, they develop slowly or not at all. Color in Alcohol : All green fades promptly, as usual in the genus, to yellowish-white; no black leg or abdominal pigment remains; on the other hand, the orange-red clypeal band of males and the crest of females are strongly persistent and even intensified. Structure. Characteristics below apply to both males and females unless otherwise specified; per- centages approximated; measurements of types given on p. 34. Carapace: Height, including tubercle of PLE, scarcely more than half length; short anterior part of thoracic slope almost level, descent of posterior part moderate; width greatest midway between PLE and pedicel, wider in male (1.5 times height, 79% of length), narrower in female (1.35 times height, 71% of length) ; longitudinal thora- cic .groove well developed, lying midway be- tween PLE and pedicel. Eyes: Eight eyes in four distinct rows; all except AME elevated on low black tu- bercles, the PME on same tubercle as ALE. First row 87% as wide as second; length of Text-fig. 1. (Part). Lyssomanes bradyspilus. A-D holotype $: A, dorsal view; B, chelicera, ventral view; C, left palp, ventral view; D, same, ectal view. E, paratype ?: epigynum. ocular quadrangle including AME 42% of carapace length, length from ALE to PLE 27 % ; breadth at ALE much wider than at PLE, 46% and 34% of length respectively; ocular quadrangle length from ALE to PLE only 66% of its breadth at ALE. Diameter of AME 21% of carapace length : ratio of eyes : AME: ALE: PME: PLE: :100: 42: 11. 5: 35. AME practically contiguous, separated from ALE by about a tenth of their diam- eter; PME slightly closer to ALE than to PLE. Clypeus: Height in males 38% of AME diameter; 54% in females. Chelicerae: In males, strongly produced but of variable length, porrect, robust, diver- gent. Length of basal segment in best devel- oped more than half carapace length, in least developed about half. Each with 1-2 prs. of overlapping spines near base on medial front 1949] Crane: Salticid Spiders; Systematica and Behavior in New Species 33 Text-fig. 1. (Part). Lyssomanes bradyspilus. A-D, holotype $: A, dorsal view; B, cheli- cera, ventral view; C, left palp, ventral view; D, same, ectal view. E, paratype $: epigynum. margin, and a group of 3 to 7 strong distal spines, the number and arrangement varying even on two sides of same individual. Fang slender and sinuous, toothless; groove weak; promargin with three small teeth near base, the smallest proximal, it and the next closer together than second and third; no tooth at base of fang; inferior margin typically with 6 teeth in a straight row, increasing in size distally, along entire edge of groove. The basal one or two, however, although appar- ently constant, are minute, delicate and easily destroyed ; they are separated consid- erably from each other and the distal 4, which are quite evenly spaced. In females the chelicerae are, of course, much shorter; distal spine group absent; teeth closer to- gether, tending to be evenly spaced through- out and of more nearly equal size. Maxillae: Parallel; width 60% of length; distal dilation slight; external angle evenly rounded without tubercle. Lip: Breadth 90% of length; basal exca- vation extending 25% of length; distal end reaching slightly beyond middle of maxillae; sternal suture straight. 34 Zoologica: New York Zoological Society [34: 7 Sternum : Broadly scutiform; width 85% of length in males, slightly less in females; equally wide between second and third coxae ; base of lip 60% as wide as anterior border in males, 50% in females; posterior end a bluntly rounded lobe extending about half- way between fourth coxae, which are sepa- rated by two-thirds of their diameter. Legs : Tibial indices : Holotype male, fii’st leg 12, fourth leg 11; paratype female, first and fourth legs, 13. First legs of male con- siderably elongated and enlarged. See Table I for formula. TABLE I. Lyssomanes bradyspilus : Leg Formula. 12 4 3 Male holotype 3.8 3.0 3.0 2.9 1 4 2 3^ Female paratype 3.2 2.9 2.7 2.6 All legs with little hair; hairs on metatarsi arranged clearly in dorsal and ventral rows, but in no sense profusely enough to be called fringes. Spines: (Male holotype and female para- type). First and second legs: Femur dorsal 1-1-1; prolateral and retrolateral 0-1-1. Pa- tella 0 but with a long, slender dorsal distal bristle. Tibia: Prolateral 1-1; retrolateral 1-1 in male, and on second female leg, 0-1 on first female leg; ventral 2-0-2-2, not opposite, the distal ones not terminal. Metatarsus ven- tral only 2-2-2, not terminal. Third leg: Fe- mur as in first and second. Patella dorsal distal only 1. Tibia, dorsal 1-0-0-1; pro- and retrolateral, as in first and second male; ven- tral 0-0-2-0. Metatarsus, prolateral and re- trolateral 1-1-0; ventral, male, 2-0-0, female none. Fourth leg: Femur dorsal 1-1-1; pro- and retrolateral male 0-0-1, female none. Pa- tella as in third. Tibia dorsal as in third; pro- and retrolateral as in first and second male, except fourth female prolateral is 0-1; ventral none. Metatarsus ventral only 1 (re- tro) -0-0. In addition, there are rudiments on third and fourth legs of distal metatarsal spines, 2 prolaterals, 2 retrolaterals and 2 ventrals, all minute and very weak. Palpal spines: Femur dorsal 0-1-1; pro- and retro- lateral distal 1 ; patella, dorsal distal 1 ; tibia, prolateral male, 0-1, female, 1-1 ; metatarsus female, dorsal 1-0; pro and retrolateral 1-1. Abdomen: About 3 times longer than broad in males and young females, tapering from level of genital groove; anal tubercle not pronounced; vestigial colulus not indi- cated. Palp: Femur slightly curved; patella and tibia nearly equal; tibia without dorsal spine cluster; tibial apophysis scarcely more than a truncate tubercle opposing basal ridge of tarsus; bulb with three pointed distal pro- cesses, variously shaped, and a distal tubercle (see Text-fig. 1), the whole complex struc- ture differing only in proportions and details from Chickering’s description of the palp in L. banksi (1946, p. 12). Epigynum : No median notch. Two pairs of large, rounded bodies; members of anterior pair smaller, apparently contiguous; pos- terior pair separated by less than half their own diameter. Measurements. Male holotype. Total length in alcohol 4.7. mm. ; carapace length 2.1 ; carapace breadth 1.6; carapace height 1.1; ocular quadrangle length, AME to PLE .89, ALE to PLE .58; ocular quandrangle breadth, at ALE .96, at PLE .72; diameter AME .45, ALE .19, MLE .05, PLE .15; clypeus height .17; basal seg- ment chelicera 1.1; patella breadth, 1st leg .38, 4th .24. Leg Measurements, Male. Leg Femur Pat. Tib. Metat. Tarsus Total 1st 2.3 .96 2.1 2.0 .55 7.9 2nd 1.9 .75 1.6 1.7 .41 6.4 3rd 1.8 .68 1.4 1.7 .41 6.0 4th 1.7 .65 1.5 1.9 .44 6.2 Palp 1.0 .44 .41 — .58 2.4 Female paratype. Total length in alcohol 4.7 mm.; carapace length 2.1; carapace breadth 1.5; carapace height 1.1; ocular quadrangle length, AME to PLE .89, ALE to PLE .65; ocular quadrangle breadth, at ALE .99, at PLE .72; diameter AME .45, ALE .19, AME .05, PLE .15; clypeus height .24; basal segment chelicera .75; patella breadth 1st leg .34, 4th .24. Leg Measurements, Female. Leg Femur Pat. Tib. Metat. Tarsus Total 1st 2.0 .79 1.8 1.7 .41 6.7 2nd 1.7 .75 1.4 1.5 .38 5.7 3rd 1.6 .68 1.2 1.5 .38 5.4 4th 1.9 .55 1.4 1.8 .41 6.1 Palp .82 .44 .44 — Behavior. .68 2.4 Locomotion: This species is a typical run- ner; I have never seen it jump, except in a final short pounce upon prey. The spider runs in brief spurts, during which the palps hang down practically touching the ground ; during the pause they palpate the surface. No special use is made of the first legs, which take an active part in running. Courtship Display: In Stage I, the cara- pace is held high, the first three pairs of legs braced somewhat forward, obliquely, and the fourth pair back; the palps hang over chel- icerae, now and then tapping ground, while the abdomen hangs straight down. To super- ficial observation, the display consists only of posing in this position, varied with occa- sional bobbing of the carapace and twitching of the abdomen during rising excitement. Not until Stage II is reached, within touch- ing distance of the female, are the first legs raised; they are then extended to the front, while the carapace sinks low and the abdo- men is swung back in the horizontal position. When the spiders are observed from their own level, however, in a straight front view, it is obvious that during display the rate of 1949] Crane: Salticid Spiders; Systematics and Behavior in New Species 35 activity of the muscles controlling the an- tero-median eyes is considerably increased; this gives rise to a much accelerated color “change” of the eyes, from green to black to green again. Similar eye color shifts have been known for many years in a few other salticids (e.g. Bristowe 1941, p. 419 ff. and I references ) . It is apparently caused by slight motions of the long, cone-shaped optic “cups,” possibly concerned with a change in focus, or in the lateral range of vision, al- I though the exact mechanism does not seem to have been worked out. In Lyssomanes the shifts take place slowly but continuously during ordinary daily activity, and may be observed at close range under a bi- nocular microscope. The mechanism works independently in the two eyes, and at a given instant either or both eyes show any propor- tion of green or black. To human beings, at least, the asymmetrically rolling effect is startling. In a dorsal view, the slight motions of the elongate “cups” may be simultaneously viewed through the translucent cuticle of the carapace. Bristowe suggests the possibility that the color shifts may be useful in enticing prey. However that may be, after the Rancho Grande observations it seems to me highly probable that acceleration of muscular ac- tivity during display should be considered as a definite part of the behavior pattern, probably with an adaptive significance; its relative value among the various sign stim- uli has not yet been established. This entire subject will be further considered in subse- quent papers. Once her attention has been attracted, the female usually sits quietly, sagging to one side on several folded legs; during the male’s display, the rate of her ocular muscular ac- tivity also is increased. Threat Display : Males usually took no no- tice of one another, and were induced to dis- play only three times. During these periods, eye color shift was not especially noted. No differences were observed between threat position and activity from those of court- ship, except that the carapace and abdomen were neither bobbed nor twitched. I never saw the long chelicerae unsheathed, although twice there was a brief, butting skirmish be- fore one opponent retreated. Habitat : Known only from the cloud forest near Rancho Grande. Shaken from green herbs, shrubs and low trees; one example taken from an epiphytic bromeliad growing twenty feet from the ground. Affinities : This species holds its chief characteristics in common with a number of Lyssomanes, although their combination seems quite distinct. L. quadrinotatus Simon, (1900), from nearby mountains, has only three teeth on inferior margin of fang groove. Material : A total of 7 adult males and 4 females have been preserved in addition to a number of young. The following have been designated as types: HOLOTYPE: Male. Cat. No. 461199, De- partment of Tropical Research, New York Zoological Society; Portachuelo, Rancho Grande, near Maracay, National Park of Aragua, Venezuela; 1136 meters; cloud for- est; March 20, 1946. PARATYPE: Female. Cat. No. 45450, Department of Tropical Research, New York Zoological Society; same locality as holo- type; July 9, 1945. The name bradyspilus is proposed in ref- erence to the delayed development of the black markings after the final molt. Semorina brachychelyne sp. nov. , (Text-fig. 2). Diagnosis : Small, brown, scale-less salt- icids, carapace low, abdomen long and nar- row with a very slight constriction near mid- dle, first legs greatly elongated and enlarged, extended forward and scarcely used in walk- ing, while the abdomen is frequently ele- vated. Chelicerae in male scarcely a fourth length of carapace; tibial apophyses of palp both curved. Color. Color in Life : Adult male. Carapace in- tegument dark brown, without scales and al- most without hairs, except around eyes. AME clear ochraceous brown shifting to black. Palps dark. First legs brown, the fe- mur and tibia almost black, the tarsi and sometimes the metatarsi translucent horn- color. Other legs translucent horn. Abdomen covered with fine dark brown hairs with a pair of small spots of white hairs (not scales and not shiny or iridescent) three-fifths of distance from base to tip. In one male there was a pair of faint pale spots near tip of abdomen in addition to the distinct more anterior pair. Adult female. Carapace integument yel- lowish-brown except sternum which is faint- ly pinkish. Eyes surrounded by a few yel- lowish hairs. Eyes themselves as in male. Tibia and tarsus of palps shiny silvery white, very conspicuous when vibrated. Swollen tibia of first legs with a ventral dark spot extending laterally; entire first leg darker than the others, which are pale translucent yellow-brown. Abdomen with a median, slightly darker stripe giving off three pairs of dark cross bars reaching mid- dle of side. A median dark spot immediately before tip of abdomen. Color in Alcohol: The white spot(s) of the male abdomen are pi'actically invisible, the pattern now resembling closely that of the female, which is little altered from life. Structure. The characteristics below apply to both males and females unless otherwise speci- fied; percentages approximated; measure- ments of types given on p. 37. Carapace: Height only 30% of carapace length; postocular plateau long; thoracic Text-fig. 2. Semorina brackychelyne. A-G, holotype $ : A, dorsal view; B. first leg, anterior view; C, carapace, lateral view; D, palp, ventral view; E, same, ectal view; F, same, tibial apophysis; G, chelicera, ventral view. H-I, paratype $: H, first leg, anterior view; drawn to same scale as B; I, epigynum. slope slightly concave; width of carapace greatest at level of PLE, about twice height, and 60% of carapace length. Longitudinal groove well defined, in middle of postocular plateau. Eyes : Eyes occupying slightly less than one-half length of carapace. Ocular quad- rangle only a third as long as broad, the sides practically parallel but with PLE very slightly closer together than ALE. Carapace extending moderately beyond PLE at their level; PME median, or slightly nearer ALE than PLE. Diameter of AME about 21% of carapace length; ratio of eyes, holotype: AME : ALE : PME : PLE : : 100 : 41 : 7 : 3 : 41. AME practically contiguous, separated from ALE, which are recurved, by about one-third diameter of ALE. Clypeus : Height in male only 5 to 6% of AME diameter, in female 11 to 12%. Chelicerae : Short, divergent, 25% of cara- pace length in male, slightly shorter in fe- male. Two small teeth on superior, one larger on inferior margin. Maxillae: Length 54% of width in male, 64% in female ; outer distal margin a blunted, obtuse angle, not produced. Lip : Width 55% of length in male, 78% in female. Sternal suture straight. Sternum: Width 56% of length in males, 53% in females. Anterior margin straight, a little narrower than lip base, greatest width between posterior margins of first legs; posterior end tapering, blunt-tipped extending between fourth coxae; the latter separated by less than a quarter of their thickness. Legs : Tibial indices : Holotype male, first leg 17, fourth leg 14; paratype female, first leg 23, fourth leg 12. First leg in both sexes much elongated and enlarged with the femur and tibia especially deep (tibia depth of first leg in male 30% of its length, in female 45%) . See Table II for formula. Hair scanty 1949] Crane: Salticid Spiders; Systematics and Behavior in New Species 37 except as follows. In male, first tibia and metatarsus with a short, moderately dense ventral fringe of dark hairs, and a scantier dorsal one of pale hairs; second tibia with a very scant pale fringe, dorsally and ventrally, metatarsus with a similar, slightly longer one ventrally only; third and fourth legs with very scanty ventral metatarsal fringes only. Fringes of negligible development in female. TABLE II. JL 4 2 3 Male holotype 2.5 1.8 1.6 1.3 1 4 2 3^ Female paratype 1.6 1.6 1.3 1.1 Spines : (From male holotype and female paratype). Femur, dorsal 0-1-1-1 through- out, the proximal two weak, bristle-like, es- pecially in female. Patella spineless through- out. Spines otherwise as follows: First leg: Tibia, ventral only 0-2-2-2, the latter not terminal; metatarsus, ventral only 0-2-2. Second leg: Tibia, retro-ventral only 1-1-0; metatarsus, male 0, female 0-2. Third leg: Tibia and metatarsus 0. Fourth leg: Femur, retrolateral distal in male 1, in female 0; tibia and metatarsus 0. Abdomen : Very elongate and tapering in both sexes, the breadth about a third of length, a very slight constriction near mid- dle. Palp: Femur practically straight; tibia more than one-half length of patella; two lateral tibial apophyses, the more dorsal longer, tapering, recurved at tip, the more ventral shorter, strongly curved antero-in- wardly. Embolus slender and tapering. Dis- tal part of bulb with a conspicuous, chiti- nized, knob-like protuberance directed out- ward. Epigynum: An anterior pair of kidney- shaped bodies, diverging posteriorly ; a pos- terior pair, smaller and closer together, fol- lowed by a pair of conspicuous small dark spots; a broad and shallow marginal notch. Measurements. Male holotype. Total length in alcohol 5.3 mm. ; carapace length 2.2, breadth 1.3, height .68; ocular quadrangle length .79, breadth 1.2; diameter AME .46, ALE .19, PME .03, PLE .19; clypeus height .02; basal segment of chelicera .55 ; sternum length .86, breadth .48; patella breadth, 1st leg, .38, 4th .21; ab- domen length 3.2, breadth .99. Leg Measurements, Male. Leg Femur Pat. Tib. Metat. Tarsus Total 1 1.7 .89 1.5 .96 .48 5.5 2 1.1 .51 .79 .65 .31 3.4 3 .82 .38 .58 .68 .31 2.8 4 1.1 .51 1.0 .89 .38 3.9 Palp .72 .24 .14 — .62 1.7 Female paratype. Total length in alcohol 5.3 mm.; carapace length 2.2; carapace breadth 1.3; carapace height .65; ocular quadrangle length .79; ocular quadrangle breadth 1.2; diameter AME .45; ALE .19 PME .03, PLE .19; clypeus height .05; basal segment of chelicera .50; sternum length .96, breadth .50; patella breadth, 1st leg, .32, 4th .15; abdomen, length 3.2, breadth 1.1. Leg Measurements, Female. Leg Femur Pat. Tib. Metat. Tarsus Total 1 1.1 5.8 .92 .65 .38 3.6 2 .82 .48 .62 .51 .44 2.9 3 .79 .34 .44 .55 .34 2.5 4 1.0 .48 .79 .75 .48 3.5 Palp .65 .27 .24 — .44 1.6 Behavior. Locomotion: The movements of this spider in the field are absurdly reminiscent of those of scorpions or pseudoscorpions, and bear little resemblance to ant behavior. Their small size, however, makes the existence of an adaptive mimetic function extremely questionable. They are to be counted among the runners in the family, their progress being a rapid sort of scurry, with short jumps reserved for crossing gaps in the ter- rain, or, of course, for the final stage in catching prey. During running the palps are vibrated continually up and down, while the first legs are held straight out in front, the metatarsi and tarsi curved inward; these legs are often vibrated, scarcely or not at all touching the ground, almost as rapidly as the palps. Meanwhile the abdomen is frequently elevated and waved slightly, also in the ver- tical plane. Immature specimens show all these characteristics in progress, and they are typical of locomotion whether or not another individual is present. Both abdomen and first legs are invariably raised whenever any obstacle is encountered. Courtship Display: Indistinguishable from ordinary locomotion except that the first legs are extended at a wide angle (more than 90%) and slightly more elevated, the tarsi usually bent down; often the palps are held still; there is the usual pursuit with sidling, and the abdomen, with increasing excitement, tends to remain elevated. Mo- tionless posing with abdomen up and first legs extended at the usual angle, also occurs with excitement. In Stage II the first legs are brought close together in front, about as in simple locomotion. During courtship the female vibrates her white palps rapidly, once her attention has been gained. Threat Display: No threat displays were seen, although a number of attempts were made to induce them. Habitat: Known only from the montane cloud forest (about 3,600 feet) around Ran- cho Grande. Always shaken from shrubs and low trees. Affinities: This species differs from Simon’s Venezuelan species, known only from females (S. seminuda and S. iris, 1901), in the complete lack of shining abdominal scales in any specimens. It likewise appears dis- tinct from Mello-Leitao’s S. lineata (1945) 38 Zoologica: New York Zoological Society [34: 7 from the Argentine. No other species seem to have been referred to this genus. It differs clearly from the other Rancho Grande spe- cies (see below) in details of the chelicerae, palp and epigynum. Material : A total of 5 adult males and 4 adult females have been preserved, in addi- tion to a number of young. The following have been designated as types : HOLOTYPE: Male. Cat. No. 481558, De- partment of Tropical Research, New York Zoological Society ; Portachuelo, Rancho Grande, near Maracay, National Park of Ar- agua, Venezuela; 1136 meters; cloud forest; July 15, 1948. PARATYPE: Female. Cat. No. 461200, Department of Tropical Research, New York Zoological Society; Limon Gorge, Rancho Grande, near Maracay, National Park of Ar- agua, Venezuela; 1100 meters; lower edge of cloud forest; April 20, 1946. The name brachychelyne is proposed in reference to the relatively short chelicerae. Semorina megachelyne sp. nov. (Text-fig. 3). Diagnosis: Very similar to S. brachychel- yne in general appearance. Chelicerae elon- gated, about half carapace length in male; tibial apophyses of palp slender and straight. Color. Color in Alcohol: Both sexes scaleless, brown except for pale second, third and fourth legs; no distinct and unvarying spots or other markings. Structure. Does not differ significantly from S. brach- ychelyne except as follows: Height of cara- pace slightly more in male (33% of length, instead of 30%) ; thoracic groove less dis- tinct, transverse rather than longitudinal; ALE and PLE slightly larger, almost one- half diameter of AME. Ratio of eyes, holo- type: AME : ALE :PME :PLE : : 100 : 48 :8 : 48. Clypeus even narrower, in both sexes, about 4% of AME in male, 5.4% in female. Maxillae and sternum both narrower with little sexual difference in breadth. Chelicerae: These form a major specific difference, being long in males, the length of the basal segment 50% of carapace length; in females it is only 30%. They are held al- most horizontally in both sexes, but are more divergent in males than in females. Tooth on inferior margin relatively larger in males of present species than in brachy- chelyne. Legs : Tibial indices : Holotype male, first leg 12, fourth leg 12; paratype female, first leg 20, fourth leg 15. General form, propor- tions and fringes similar to those in brachy- chelyne. The leg formula is given in Table III. TABLE III. Semorina megachelyne: Leg Formula. 14 2 3 Male holotype 2.3 1.7 1.5 1.2 14 2 3 Female paratype 1.5 1.4 1.1 1.1 Text-fig. 3. Semorina megachelyne. A-E, holotype $: A, carapace, lateral view; B, chelicera, ventral view; C, palp, ventral view; D, same, ectal view; E, same, tibial apophysis. F, paratype female: epigynum. 1949] Crane: Salticid Spiders; Systematics and Behavior in New Species 39 Spines : (From male holotype and female paratype). As in brachychelyne, except for second leg, as follows: In male, metatarsus retroventral 1-0, not 0; female, as in brachy- chelyne male, except metatarsus is lr-2. Palp : Differs from brachychelyne as fol- lows : Both tibial apophyses are straight, the tarsus along with its bulb is more slender, and the coiling of the tubule within the bulb is different. Epigynum : The structure differs dis- tinctly in the two species, as shown in the figure; the more nearly spherical shape of the four bodies is especially noticeable. Measurements. Male holotype. Total length in alcohol 4.7 mm. ; carapace length 2.2, breadth 1.3, height .75; ocular quadrangle length .79, breadth 1.2; diameter AME .43, ALE .21, PME .03, PLE .21; clypeus height .02; basal segment of chelicerae 1.1 ; sternum length .96, breadth .46; patella breadth, 1st leg .27, 4th .17; abdomen, length 2.5, breadth .82. Leg Measurements, Male. Leg Femur Pat. Tib. Metat. Tarsus Total 1 1.5 .82 1.4 .96 .41 5.1 2 .96 .51 .75 .62 .34 3.2 3 .82 .38 .55 .68 .27 2.7 4 1.1 .48 .92 .85 .38 3.7 Palp .72 .31 .17 — .58 1.8 Female paratype. Total length in alcohol 4.0 mm.; carapace length 1.7, breadth 1.0, height .55; ocular quadrangle length .75, breadth .96; diameter AME .36, ALE .16, PME .03, PLE .17 ; clypeus height .09; cheli- cera, basal segment .52 ; sternum length .79, breadth .36; patella breadth 1st leg .21, 4th leg .14; abdomen, length 2.3, breadth .79. Leg Measurements, Female. Leg Femur Pat. Tib. Metat. Tarsus Total 1 .79 .48 .58 .44 .27 2.6 2 .62 .34 .41 .31 .24 1.9 3 .55 .31 .34 .41 .27 1.9 4 .79 .31 .62 .55 .27 2.5 Palp .44 .17 .14 — .34 1.1 Behavior : Locomotion as in brachychel- yne. No displays observed. Habitat : Known only from lower edge of montane cloud forest, about 3,500 feet, near Rancho Grande. Collected from tree trunks and shrubs. Affinities : See remarks under brachy- chelyne. Material: A total of 2 adult males and 5 adult females were taken, along with a num- ber of young. The following have been des- ignated as types: HOLOTYPE: Male Cat. No. 461201, De- partment of Tropical Research, New York Zoological Society; Water Trail, Rancho Grande, near Maracay, National Park of Ar- agua, Venezuela; 1100 meters; lower edge of cloud forest; May 5, 1946. PARATYPE: Female. Cat. No. 461202. Same data as holotype. The name megachelyne is proposed in ref- erence to the long chelicerae of the male. Ashtabula furcillata sp. nov. (Text-fig. 4) . Diagnosis: Color in life above entirely ir- idescent green with white dorso-lateral band encircling carapace and abdomen ; dorsal ab- dominal spots lacking, although sometimes faintly indicated in alcohol; carapace low; abdomen elongate; tibial apophysis of male forked. Color. In Life: Adult male. Carapace above en- tirely covered with iridescent scales, rich green with bronze reflections. A white stripe starting behind ALE, bordered narrowly on ventral margin with black, passing imme- diately below PME and PLE, and extending along thorax almost to pedicel. Sides of cara- pace naked, black with a narrow white sub- marginal border of scales. AME narrowly rimmed with yellowish. Clypeus black, naked. Palps and first legs black (except pale 1st tarsi), other legs translucent buff. Sternum black. Abdomen covered with green scales like those of carapace, outlined dorso-later- ally with white, which either continues to tip of abdomen or stops short of the tip; a white distal median spot present or absent. Moderate green iridescence on lower abdom- inal sides, below white stripe; venter black. Adult female. Like male, except sides of carapace brown, not black ; palps light green- yellow, not black ; first legs dark brown, not black, the distal metatarsus and entire tar- sus paler; other legs pale as in male, but with greenish tinge. In Alcohol: The green iridescence is al- most or completely lacking, and the scales may be largely missing, especially on the ab- domen, where there may be faint traces of median spots or other markings. The white dorso-lateral bands, however, are very per- sistent. Structure. Essentially as in Chickering’s description of A. dentata Cambridge, 1901 (Chickering, 1946, p. 248). The only significant differ- ences are as follows : Chelicerae : Large pro- lateral tooth of basal segment of chelicera straight, not curved; enlargement at base of fang less distinct, a tubercle rather than a tooth. Fringe on first leg continues onto metatarsus. Spines: Very similar in the two species ; the femoral prolateral distal spines tend to be more numerous than in dentata (first leg 2, not 1; 4th leg, n>ale, 1 not 0, but 0 in female) ; metatarsal prolaterals tend to be fewer than in dentata (second leg 0, not 1; third leg 1, not 2) ; a weak fourth meta- tarsal ventral distal is present in furcillata, absent in dentata. Female furcillata as in male, except that femoral distal spines are reduced, about as in male dentata, and tibials are completely absent. Palp: Tibial apophy- sis differs radically from that of all pre- Zoologica : New York Zoological Society [34: 7 Text-fig. 4. Ashtabula furcillata. A-E, holotype 3 : A, carapace and abdomen, dorsal view; B, carapace, lateral view; C, chelicera, ventral view; D, palp, ventral view; E, same, ectal view; F, courtship display. G, paratype $: epigynum. 1949] Crane: Salticid Spiders; Systematics and Behavior in New Species 41 viously known males — zonura Peckham, 1894, dentata Cambridge, 1901, and of den- tichelis, sexgutta and glauca, all of Simon, 1902; in furcillata alone it is not simple, but distally forked. Measurements. Male holotype. Total length in alcohol 4.2 mm. ; carapace length 1.9, breadth 1.4, height .75; ocular quadrangle length .82, breadth 1.2; diameter AME .34; ALE .17; PME .04; PLE .17; clypeus height .05; basal segment of chelicera 2.4; patella breadth, 1st leg, .19, 4th .21; length of abdomen 2.3, breadth 1.1. Leg Measurements, Male. Leg Femur Pat. Tib. Metat. Tarsus Total 1 1.2 .79 .99 .79 .41 4.2 2 .82 .44 .55 .48 .31 2.6 3 .79 .38 .51 .51 .31 2.5 4 .99 .51 .72 .58 .34 3.1 Palp .68 .14 .10 — .62 1.5 Female paratype. Total length in alcohol 3.7 mm.; carapace length 1.7; carapace breadth 1.1; carapace height .68; ocular quadrangle .79 ; ocular quadrangle breadth 1.1; diameter AME .33; ALE .17 ; PME .04; PLE 17; clypeus height .03; basal segment of chelicera .36 ; patella breadth, 1st leg .26, 4th .21 ; length of abdomen 2.0, breadth 1.0. Leg Measurements, Female. Leg Femur Pat. Tib. Metat. Tarsus Total 1 .82 .55 .62 .48 .31 2.8 2 .68 .38 .44 .38 .27 2.2 3 .68 .38 .38 .41 .31 2.2 4 .85 .44 .65 .51 .31 2.8 Palp .44 .21 .41 — .38 1.4 Tibial indices: Holotype male, first leg 11, fourth leg 17 ; paratype female, first leg 22, fourth leg 19. See Table TV for formula. TABLE IV. A. furcillata : Leg formula. 1 4 2 3 Male holotype 2.2 1.6 1.4 1.3 1 4 2 3 Female paratype 1.6 1.6 1.3 1.3 Behavior. Locomotion : A scurrying run, the first legs held flat and low, straight in front of body; both they and the palps palpate the surface almost constantly during progress. During pauses the first legs are usually elevated, and they and the palps jerked rapidly up and down. Both Ashtabula and Sassacus are mas- ters of backward running, and both can jump well, although they never resort to it except in crossing gaps and in the final stage of prey capture. Courtship Display : Stage I. Male carapace well elevated, abdomen swung to one side (usually the left) , where it is held low, prac- tically resting on ground; the spider sidles back and forth, raising the front legs at a wide angle and waving them up and down in unison. The palps occasionally jerk up and down, but hang quietly during height of dis- play. The white abdominal stripe and its bounding iridescence show clearly, little im- peded by the short, pale, posterior legs. When the attention of a female has been gained, her pale, greenish-yellow palps jerk up and down rapidly and almost continuously, being conspicuous against her dark brown clypeus and mouthparts. Stage II. Not seen. Threat Display : Inter-male display seems feebly developed in this species; three dif- ferent pairs of males at various times, all in display condition, judging by their behavior toward females, paid little or no attention to each other, except for some brief elevation of the forelegs, which frequently takes place in any situation and appears to be of an ex- ploratory nature. Habitat : Known only from the montane cloud forest (about 3,600 feet) around Rancho Grande. Always taken on herbs, shrubs and low trees. Affinities: Close to A. dentata; see re- marks under Structure. It seems likely that dentata, dentichelis and furcillata will prove to be no more than subspecies of zonata. Material : A total of 5 adult males and 1 adult female have been preserved. The fol- lowing have been designated as types: HOLOTYPE : Male. Cat. No. 461203, De- partment of Tropical Research, New York Zoological Society; Portachuelo, Rancho Grande, near Maracay, National Park of Aragua, Venezuela; 1,136 meters; cloud for- est; June 15, 1946. PARATYPE: Female. Cat. No. 481559, Department of Tropical Research, New York Zoological Society ; same locality as holotype ; July 21, 1948. The name furcillata is proposed in refer- ence to the characteristic forked tip of the palp’s tibial spine. Sassacus ttavicinctus sp. nov. (Text-fig. 5). Diagnosis : Male black with yellow on cly- peus, in paired stripes and a submarginal band on carapace, and in transverse mark- ings on abdomen. Female brown with obscure ochraceous markings. Chelicera of male strongly produced, the promargin with two teeth, far separated, the retromargin with a single strong tooth near distal end. Tibial apophysis of palp strong, simple, tapering, tip slightly recurved ; embolus curved. Color. Color in Life : Adult male. Cephalothorax : Integument of carapace black, with a mod- erate number of long bristles in ocular re- gion, and with lemon yellow (Ridgway) scales arranged in dense bands as follows: A pair on carapace just below dorsal eyes, converging slightly behind them and ending, without meeting, halfway down thoracic slope; a narrow submarginal band; a well- developed band of scales and scale-hairs 42 Zoological New York Zoological Society [34 : 7 Text-fig. 5. Sassacus flavicinctus. A-E, holotype $: A, carapace and abdomen, dorsal view; B, carapace, lateral view; C, chelicera, ventral view; D, palp, ventral view; E, same, ectal view. F, paratype $: epigynum. completely covering and slightly pendent from the narrow clypeus. Mouthparts and legs black except as noted below; all tarsi brown ; tibia and metatarsi of all except first legs banded brown and black in varying pro- portions ; all legs with small anterior patches of yellow and white hairs and scales on some or all of the following segments: Femur, patella and tibia ; these markings are highly variable. Sternum black with white hairs, which occur also on underside of coxae. Ab- domen : A basal semi-circular band of lemon yellow scales continuing backward a third of abdominal length; behind this two pairs of short, curved bars, concave posteriorly, of which the posterior pair may join in the midline ; at tip of abdomen a tiny round spot, or a short bar concave posteriorly, may be present or absent. Center black with a tri- angular patch of white hair, the apex pos- terior. Adult Female. Cephalothorax : Carapace black with rather weak markings of ochra- ceous brown scale-hairs as follows: Across clypeus and completely encircling sides of carapace and thoracic slope, absent only in j middle of ocular quadrangle. Palps dark with yellowish hairs. Legs banded light and dark brown. Sternum black. Abdomen : Dorsum with an indistinct, in- terrupted reticulated pattern which consists basically of an anterior basal band, followed by several pairs of hollow bands ; the latter do not meet in midline, but join with the pre- ceding band by a narrow stripe just before the center; tip of abdomen covered with 1949] Crane: Salticid Spiders; Systematics and Behavior in New Species 43 ochraceous hairs. Venter black with a few scattered light hairs. All scale-hairs easily removed, and fre- quently absent in preserved specimens. Structure. The characteristics below apply to both males and females unless otherwise specified ; percentages approximated ; measurements of types given below. Carapace : Height about half (female) or slightly less than half (male) of carapace length; anterior part of thorax flat, with a very gentle slope, rounding into rounded sides of cephalic part; descent of posterior part (less than half postocular length) abrupt, slightly concave; width of carapace greatest a little behind PLE, 1.5 times height, 67% (male) to 75% (female) of carapace length ; thoracic groove scarcely indicated. Eyes: Length of ocular quadrangle about 58% as long as broad, its sides almost par- allel, though very slightly wider at ALE than at PLE ; carapace extending slightly beyond PLE at their level, PME slightly nearer ALE than PLE. Diameter of AME about 20% of carapace length; ratio of eyes, holotype; AME: ALE: PME :PLE : : 100 :48 :8 :44. AME practically contiguous, separated from ALE, which are slightly recurved, by about an eighth of their diameter. Clypeus : Height 12% of AME diameter. Chelicerae : In males strongly produced, held almost parallel to ground, divergent; length of basal segment about three-fifths of carapace length. Promargin with one slender tooth at proximal end of groove and one, long, robust, triangular, far removed, near base of fang; slightly proximal to this on retromargin a single large tooth. Fang slen- der, slightly sinuous. Chelicerae of females much shorter with a very short groove flanked on promargin by two teeth close to- gether, the proximal the smaller, and one large tooth on retromargin. Maxillae: Width about 75% of length; outer distal edge in male more dilated and obtusely angled than in female. Lip: Breadth more than 90% of length; distal end reaching slightly beyond middle of maxillae; sternal suture curved, especi- ally in male. Sternum: Width 62% of length in males; wider, about 73%, in females. Anterior mar- gin concave, narrower than base of lip; greatest width between first and second legs ; posterior end bluntly pointed, extending slightly between fourth coxae; the latter separated by less than half their diameter. TABLE V. S. flavicinctus : Leg Formula. 1 4 2 3 Male holotype 1.8 1.5 1.4 1.35 2 1 3 jl 1.4 Legs : Tibial indices : Holotype male, first leg 16, fourth leg 17.5; paratype female, first leg 26, fourth leg 23. First femur in both sexes enlarged, and entire first leg somewhat thickened and elongated in male. See Table V for formula. All legs with little hair. Spines: (From male holotype and female paratype). Patella without spines through- out. First leg : Femur, dorsal 3 in distal half ; prolateral distal 1 in male, 2 in female; tibia ventral only lr-2-2, the two distal pairs close together, the proximal at beginning of sec- ond quarter of segment; metatarsus, ventral only, 0-2-2. Second leg differs from first in having tibia ventral lr-lr-2, (male) or lr-0-2 (female) ; tibia prolateral, male only, 1-1 (both small) ; metatarsus, female only, with 1 prolateral distal. Third leg, femur, dorsal 0-1-1-1, prolateral distal 2 (male), or 1 (female) ; tibia prolateral 0-1 (male) or none (female) ; retrolateral 0-1; ventral 0-0-2 (male) or lp-lp-2 (female) ; meta- tarsus prolateral distal 2, retrolateral distal 2, ventral distal 2. Fourth leg, femur as in third; tibia prolateral none (male), or 0-1 (female) ; retrolateral 0-1 or none (variable on two sides) ; ventral lp-0-2 or lr-lr-2 or 0-0-2 (variable on two sides) ; metatarsus prolateral distal 0-1, sometimes in female only 0-2, the second weak; ventral distal only 2, on one side of female 0-2-2. Abdomen: Ovate in both sexes, the breadth about 70-75% of length, widest near middle. Palp: Femur strongly curved; tibia more than one-half length of patella; tibia with a retrolateral apophysis which tapers to a blunt, slightly recurved point. Embolus tap- ering from a broad base to a curved and slen- der tip. Epigynum : An anterior pair of bodies well separated, a posterior pair contiguous; mar- ginal notch deep and narrow. Measurements. Male holotype. Total length in alcohol 4.51 mm.; carapace length 2.4, breadth 1.6, height 1.0; ocular quadrangle length .79, breadth 1.4; diameter AME .43, ALE .21, PME .03, PLE .19; clypeus height .05; basal segment of chelicera 1.37 ; patella breadth, 1st leg, .31, 4th .24. Leg Measurements, Male. Leg Femur Pat. Tib. Metat. Tarsus Total 1 1.3 .85 1.0 .72 .44 4.3 2 1.1 .58 .62 .62 .41 3.3 3 1.1 .48 .55 .68 .34 3.2 4 1.2 .55 .82 .75 .34 3.7 Palp .79 .27 .17 — .58 1.8 Female paratype. Total length in alcohol 4.68 mm.; carapace length 2.05, breadth 1.54, height 1.03; ocular quadrangle length .83, breadth 1.4; diameter AME .43, ALE .21, PME .03, PLE .19; clypeus height .05; basal segment of chelicera .72; patella breadth, 1st leg, .34, 4th .26. Female paratype 1.8 1.65 1.7 44 Zoologica: New York Zoological Society [34: 7 1 Leg Measurements, Female. Leg Femur Pat. Tib. Metat. Tarsus Total 1 1.1 .65 .68 .55 .38 3.4 2 1.2 .51 .85 .72 .38 3.7 3 1.2 .51 .79 .68 .34 3.5 4 .92 .55 .55 .48 .34 2.8 Palp .55 .24 .24 — .38 1.4 Behavior. Locomotion : Compared with Ashtabula, this Sassacus is somewhat more a jumper and walker, less a scurrier; also it palpates the ground far less with the first legs and palps. Compared to the spiders of the Plexip- pus group, however, it is a poor and reluctant jumper. Courtship Display. Stage I. Male follows female about, the carapace moderately ele- vated and the first legs raised at a wide angle to each other; frequently lowered; the ab- domen hangs down and is trailed inconspicu- ously from side to side with sideling. Display tends to be in a wide semi-circle around fe- male, once her attention has been attracted. The long chelicerae are folded but held out laterally (when not displaying they are held at right angles to each other) , and the palps extend straight out also, in contrast to their usual resting position when they hang over chelicerae. With increasing stimulation, zig- zagging becomes more pronounced and a slow rocking is involved, the carapace and abdomen held stiffly and rocking as a unit. Stage II is usually attained within three to five minutes by couples of low threshold to display stimuli, and consists of the first legs thrust out in front, clear of the ground. Threat Display: Stage I. Indistinguish- able from Stage I of courtship, except that no rocking is involved. Stage II. It is only in the rare occurrence of this stage that the chelicerae blades are unsheathed ; when two opponents are practically touching the first legs are brought upright, from the obliquely outward display position, and simultaneously the chelicerae blades are extended straight out in front, at right angles to the basal seg- ment, which is maintained in the horizontal position typical of display. In each of the dozen or so observed encounters that reached this stage, one or the other male usually backed off promptly at this point; more rare- ly there was a brief tangle which ended with- out apparent injury. Usually one or both males retreated before reaching Stage II. Habitat: Known only from the montane cloud forest (about 3,600 feet) around Rancho Grande. Always shaken from herbs, shrubs and low trees. Affinities: This species appears exceed- ingly close to S. arcuatus Simon, 1902, from Teffe, in the Amazon region. From the brief description, the only apparent differences are slight distinctions in the abdominal mark- ings and the absence, in the present form, of a yellow spot on the palp femur. Material: A total of 5 adult males and 5 adult females have been preserved in addi- tion to a number of young. The following have been designated as types: HOLOTYPE: Male. Cat. No. 45451. De- partment of Tropical Research, New York Zoological Society; Portachuelo, Rancho Grande, near Maracay, National Park of Aragua, Venezuela; 1,136 meters; cloud for- est; June 1, 1945. PARATYPE: Female. Cat. No. 45452 De- partment of Tropical Research, New York Zoological Society ; same locality as holotype (with which she mated) ; July 1, 1945. The name flavicinctus is proposed in ref- erence to the yellow bands characterizing the male. Sassacus oce/fatus sp. nov. (Text-fig. 6). Diagnosis: Both sexes iridescent green above, with a pair of black spots, each crossed by a white bar, near tip of abdomen. Chelicera of male strongly produced, the pro- margin with two well-separated teeth in proximal half, opposed by a single large tooth on retromargin. Spines of first tibia 2-2-2. Tibial apophysis of palp strong, sim- ple, tapering, tip straight; embolus tip straight. Color. Color in Life : Adult male. Cephalothorax : Integument of carapace black; ocular region with a number of long bristles and completely covered with iridescent green scales which extend a little below it on sides and thoracic region. A broad band of white hairs, starting below PME on side of carapace, extends for- ward across clypeus. Palps, mouthparts and first legs jet black; other legs brown; two narrow, conspicuous stripes of white scales extend along anterior and posterior sides of first patella, tibia and base of metatarsus. These scales, although progressively fewer posteriorly, are present on anterior sides of all other legs, as well as on posterior sides of second legs. Sternum black. Abdomen en- tirely covered above, except as hereafter noted, with iridescent green scales, larger than those on carapace. On dorso-lateral sur- face on each side of posterior third is a large spot of velvety black scales, each with a nar- row cross-bar of white scales from one-third to two-thirds of the way to its posterior edge. Around the entire abdomen laterally is a narrow band of iridescent green, confluent except in region of spot, with the dorsal green. Venter black. Adult female. Cephalothorax : carapace as in male, with the addition of a narrow sub- marginal border of white scales continuing almost as far as pedicel. Entire face, around eyes, with more white scales and hairs than in male. Chelicerae black with a few white hairs basally; palps translucent brown barred narrowly with darker on joints, and with a few white hairs on patellae. All legs translucent brown except first femora, which are almost black. Sternum black. Abdomen as in male, except that there is a faint an- 1949] Crane: Salticul Spiders; Systematics and Behavior in New Species 45 Text-fig. 6. Sassacus ocellatus. A-E, holotype $: A, carapace and abdomen, dorsal view; B, carapace, lateral view; C, chelicera, ventral view; D, palp, ventral view; E, same, ectal view. F, paratype $: epigynum. terior band of white scales, dying out later- ally in variable faint spots, while the white cross-bars on the posterior black spots tend to be on the latter’s anterior margin. In alcohol the iridescent green completely vanishes, the scales appearing dull yellowish or brownish ; the abdominal black spots with white cross-bars are discernible, but far less distinct than in life, the anterior part of the spot tending to disappear altogether. As us- ual, the black integumentary areas fade to brown. Structure. Essentially as in S. flavicincta except in the following respects: carapace lower, its height less than half carapace length in both 46 Zoologica: New York Zoological Society [34: 7 sexes, lower in male than in female. Cheli- cera of male even longer in some specimens, but varying in individuals; basal segment in holotype is 5/6 of carapace length; pro- margin with two small teeth well separated, along proximal half of groove; opposite their interspace, on retromargin, is a single, much larger, conical tooth. Tibial indices : Holo- type male, first leg 21, fourth leg 19; para- tvpe female, first leg 25, fourth leg 19. See Table VI for formula. TABLE VI. S. ocellatus: Leg Formula. 1 4 2 3 Male holotype 1.9 1.5 1.4 1.2 14 2 3 Female paratype 1.7 1.7 1.3 1.3 Spines : As in flavicinctus, but with first tibial ventral 2-2-2, not lr-2-2, and with spines on posterior legs somewhat fewer, viz.: Second leg: Male, tibia prolateral 0, not 1-1 ; female as in flavicinctus. Third leg : Male, femur prolateral distal 1, not 2; fe- male, prolateral 0, retrolateral 1; tibia 0 in both sexes, not with a few pro- and retro- laterals and ventrals; metatarsus (both sexes) pro- and retrolateral distals each 1, not 2. Fourth leg (both sexes) : Femur pro- lateral 1 not 2 ; tibia, ventral distal only lp in male, 0 in female; metatarsus, as in third leg, but with traces of another lateral distal pair (very weak), similar to those in flavi- cinctus; especially noticeable in female. Palp: Tibial apophysis and embolus both straight, not curved. Epigynum : Radically different from that of S. flavicinctus (see figure) ; marginal notch broad and shallow. Measurements. Male holotype. Total length in alcohol 3.3 mm.; carapace length 2.1, breadth 1.5, height .79; clypeus height .07; basal segment of chelicera 1.8; patella breadth, 1st leg .34, 4th leg .22; length of abdomen 2.2, breadth 1.4. Leg Measurements, Male. Leg Femur Pat. Tib. Metat. Tarsus Total 1 1.2 .72 .92 .65 .44 3.9 2 .89 .55 .62 .58 .38 3.0 3 .82 .44 .51 .48 .38 2.6 4 .99 .51 .68 .65 .38 3.2 Palp .82 .55 .14 — .58 2.1 Female paratype. Total length in alcohol 5.0 mm. ; carapace length 1.7, breadth 1.3, height .72 ; clypeus height .10 ; basal segment of chelicera .58; patella breadth, 1st leg .31, 4th leg .21 ; length of abdomen 3.3, breadth 2.1. Leg Measurements, Female. Leg Femur Pat. Tib. Metat. Tarsus Total 1 .85 .62 .62 .48 .34 2.9 2 .68 .48 .41 .38 .31 2.3 3 .68 .41 .41 .44 .31 2.3 4 .89 .48 .65 .55 .37 2.9 Palp .44 .21 .21 — .34 1.2 Behavior. Locomotion: About midway between Ash- tabula and S. flavicinctus. Its usual progress is a rapid scurry, jumping only when neces- sary, the first legs held forward, usually scarcely touching the ground, the palps held just clear of it. During the infrequent pauses, the first legs and palps are raised in the air and waved up and down ; after which both sets of appendages sometimes palpate the ground itself. Courtship Display: Stage I. Carapace scarcely elevated, first legs held up at about right angles to each other, and brought to ground again during the jerking, zig-zag approach to female. The long chelicerae are sheathed, the palps hanging quietly over them in the normal resting position, except for occasional vibration. Approach to the fe- male is often quick and direct after the pre- liminary zig-zags. The most interesting phase may or may not be included; it con- sists of posing for a few moments, motion- less, the legs elevated, and the abdomen twisted slightly to one side or the other; once the female was seen to perform the same mo- tion, although that courtship was not com- pleted. The relatively short abdomen was never swung far to the side as in the elon- gate Ashtabula, and the black, white-barred terminal spot could not have been in full view. In the single courtship which ended in actual mating, this phase was altogether omitted. Stage II. This was often reached within three minutes ; in one case mating fol- lowed five minutes after display began. It did not differ from that of flavicinctus. Threat Display: True fighting frequently takes place in this species and even when inter-male display ends in mere threat, the chelicerae are always more or less un- sheathed, which never happens in courtship. The behavior otherwise is similar except that I observed little or no trace of the side- swinging of the abdomen. During actual bat- tle the first legs are raised directly overhead, and the palps extended laterally, wide- spread, out of the way; the wide-open cheli- cerae are opposed to those of the opponent. The two may then push back and forth for seconds, until one of the pair retreats or is bitten. Habitat: Known only from the montane cloud forest (about 3,600 feet) around Rancho Grande. Always shaken from herbs, shrubs and low trees. Affinities: The abdominal markings are somewhat similar to those of S. aurantiacus Simon, 1902, from Para, Brazil, known only from the briefly described female. The pres- ent species has a full set of 2-2-2 spines on the first tibia, instead of lp-2-2, in both sexes. Material: A total of 11 adult males and 9 adult females have been preserved in addi- tion to a number of young. The following have been designated as types: HOLOTYPE : Male. Cat. No. 461204, De- partment of Tropical Research, New York 1949] Crane: Salticid Spiders; Systematics and Behavior in New Species 47 Zoological Society; Portachuelo, Rancho Grande, near Maracay, National Park of Aragua, Venezuela; 1,136 meters; cloud for- est; March 27, 1946. PARATYPE: Female. Cat. No. 481560, ■ Department of Tropical Research, New York Zoological Society; same locality as holo- type; July 17, 1948. The name ocellatus is proposed in refer- ence to the eye-like abdominal markings. Phiale Homme a sp. nov. (Text-fig. 7). Diagnosis : All carapace bands in both sexes creamy yellow. Male : Carapace mark- ings broad, including submarginal and cly- peal bands and mid-dorsal stripe; no spots near PME. Abdomen above bright rufous with white markings; median spot absent, although a faint cross-bar may be present or absent beneath rufous scales; no terminal hook on antero-lateral band; three terminal spots. Palp with tibial apophysis stout, trun- cate; bulb strongly bilobed; lateral process of embolus shorter than and widely separated from embolus proper. Female: Carapace markings less extensive than in male. Ab- domen with reddish scales ranging almost to black; anterior abdominal band as in male; strong, post-median cross-bar and terminal spots present. Epigynum with two strongly chitinized, external cross-bars. Color. Color in Life: Adult male. As in Chick- ering’s description of P. aliceae in alcohol (1946, p. 207), except as follows: Cephalo- thorax : Integument of carapace, mouthparts, palps and first legs (except metatarsus and tarsus) black, not dark brown; integument of other legs translucent, medium brown. All carapace scale-hair bands distinctly buffy yellow; anterior eyes rimmed with rust; cly- peus with a strong band of creamy yellow scalerhairs, instead of only “a fringe of yel- lowish bristles;” palp femur with dorsal scale-hair patch as in aliceae; a patch of white-scale hairs on proximal anterior face of first metatarsus and tarsus; variable num- bers and arrangements of similar scales, diminishing posteriorly, on other segments of other legs. Abdomen: Dorsum in full sun- light often matches the flame scarlet of Ridg- wayj other individuals tend to orange rufous. As in aliceae, white markings consist of a simple anterior band extending dorso-later- ally more than halfway to spinnerets, and ending without a hook-shaped inward curve Text-fig. 7. Phiale fiammea. A-E, holotype $: A, carapace and abdomen, dorsal view; B, carapace, lateral view; C, palp, ventral view; D, same, ectal view; E, same, tibial apophysis. F, G, paratype ?: F, carapace and abdomen, dorsal view; G, epigynum. 48 Zoologicn: New York Zoological Society [34: 7 (as is characteristic of P. dybowskii, for ex- ample) ; usually it ends abruptly; sometimes there is a very slight inward curve. The “narrow, light-colored central bar” of aliceae is invisible in live specimens though it some- times shows in preserved examples, beneath the rufous scales. Three small white terminal markings, in the form of spots or short bars, as in aliceae; carapace stripe easily rubbed, often small in preservative. Adult female. Exceedingly variable, both in the pattern of white and dark scales, and in the vividness of the reddish abdominal markings ; the individuals are separated with difficulty in pattern from at least two other species occurring typically on the lower slopes of the same mountain range. They dif- fer from the male as follows : Cephalothorax : buff stripe and bands of carapace — median, submarginal and clypeal — much less exten- sive; sparse rusty hairs usually present on and around ocular quadrangle; anterior eyes rimmed with yellowish-white, not rust; some buff hairs on face below ALE; palps trans- lucent buffy yellow, not black, and lacking buff scales; first legs black only on femur and patella; white hairs and scales of all legs reduced or absent. Abdomen : Red of dorsum exceedingly variable, practically always less bright than in male, sometimes almost black. A strong post-median, black-bordered cross- bar of white scales always present, but of variable length and breath, sometimes con- fluent with ends of anterior dorso-lateral band, which is as in male; posterior spots present as in male, but of more variable size and shape, sometimes partly confluent. Structure. This species is so close to P. aliceae (known only from holotype male) that no significant structural differences emerge from a com- parison of Chickering’s description with our species, except for minor spine and palp dif- ferences as given below. The females are closely similar to the males in structure, ex- cept for the usual leg differences, and for the absence of the small hooked maxillary process. Spines (both sexes) : Differ from aliceae as follows: First leg, Female: Patella pro- lateral 0, not 1. Second leg, both sexes : Tibia prolateral as in first (1-0-1, not 1-1-1), ven- tral apparently consistently lr-2-2, not vari- able; metatarsus male, prolateral distal 0, not 1, but this spine present in female. Third leg (female only) : Femur prolateral distal only 2, not 1-2, retrolateral 1, not 2; tibia dorsal 0, not 1 ; metatarsus with slight ir- regularities on one side of paratype female only, retrolateral 0-1-2, not 1-1-2, ventral lp-lp-2, not 0-2-2. Fourth leg: Femur (both sexes) prolateral and retrolateral distal re- spectively 0 and 1, not each 2; male tibia as on right side of aliceae holotype, female dor- sal 0, not 1. Palp\ Differs from that of aliceae in its relatively greater breadth and in the char- acter of lateral process of embolus; in flammea the two parts are much farther apart, though connected by a thin, horny plate ; also, the lateral process is much short- er than embolus proper, and scarcely curved distally. Epigynum : Confusing, as usual in this genus, on account of the frequent secretion of gummy matter which obscures and dis- torts the structure. Always distinct, how- ever, are two strongly chitinized transverse, lip-like structures, one between the two pairs of subdermal bodies and one near posterior border. Measurements. Male holotype. Total length in alcohol 4.6 mm. ; carapace length 2.7, breadth 1.9, height 1.1; clypeus height .19; basal segment of chelicera .89; patella breadth. 1st leg, .41, 4th .28; length of abdomen 1.9, breadth 1.4. Leg Measurements, Male. Leg Femur Pat. Tib. Metat. Tarsus Total 1 1.8 1.1 1.7 1.1 .68 6.4 2 1.3 .75 .89 .79 .48 4.2 3 1.5 .79 .85 1.1 .55 4.8 4 1.6 .75 1.2 1.3 .55 5.4 Palp .85 .24 .24 — .82 2.2 Female paratype. Total length in alcohol 5.1 mm.; carapace length 2.5, breadth 1.7, height 1.1 ; clypeus height .07 ; basal segment of chelicera .85; patella breadth, 1st leg .40, 4th .31 ; length of abdomen 2.6, breadth 1.7. Leg Measurements, Female. Leg Femur Pat. Tib. Metat. Tarsus Total 1 1.2 .85 .89 .65 .51 4.1 2 1.1 .65 .65 .62 .48 3.5 3 1.3 .72 .79 .82 .55 4.2 4 1.4 .72 .99 1.1 .62 4.8 Palp .65 .27 .31 — .55 1.8 Tibial indices : Holotype male first leg 15, fourth leg 14; paratype female, first leg 23, , fourth leg 18. See Table VII for formula, i TABLE VII. P. flammea: Leg Formula. 4 3 2 Male holotype 2.4 2.0 1.8 1.5 J. 1 3_ 2 Female paratype 1.9 1.6 1.7 1.4 Locomotion : Primarily a runner, although jumps are undertaken over gaps without hesitation. The first legs take little part in locomotion and are habitually waved up and down during the pauses. Courtship Display : Stage I. Carapace ele- vated high; abdomen hangs down, usually touching ground and leaving a silk thread. First legs raised at 45° angle with each other and the ground. Female approached in zig- zag spurts, as the carapace is rocked from side to side, sinking alternating almost to the ground, from right to left. Palps irregu- larly vibrated up and down. Pursuit of fe- male plays an important part in early stages, 1949] Crane: Salticid Spiders; Systematics and Behavior in Neiv Species 49 but once female’s attention is gained, she usually watches with first legs elevated and palps vibrating rapidly. Stage II. Male abruptly crouches almost on ground, when two inches or less from female; his legs far outstretched in front, almost parallel, he approaches her directly with crawling motion, the palps vibrating in unison and entire body quivering. The re- markable point about Stage II in this species is that it begins at such a relatively long distance from the female. Threat Display. As in Stage I of court- ship, except that the palps are held quiet most of the time, the creamy yellow patch of the curved femur continuing that of the cly- peus in an unbroken line. When approach is very close the chelicerae are opened and the first legs spread more widely, often actually touching those of the opponent. The bouts are always brief and I have never seen dam- age inflicted. Habitat : Known only from the montane cloud forest (about 3,600 feet) around Rancho Grande. Always taken on herbs, shrubs or small trees. Affinities : The closeness of this species to P. aliceae has already been noted. When ade- quate material is taken from intermediate localities, it seems likely that the distinc- tions will prove to be of only subspecific im- portance. Material : A total of 14 adult males and 20 adult females have been preserved. The following have been designated as types : HOLOTYPE: Male. Cat. No. 481561, De- partment of Tropical Research, New York Zoological Society; Portachuelo, Rancho Grande, near Maracay, National Park of Aragua, Venezuela; 1,136 meters, cloud for- est; July 25, 1948. PARATYPE : Female. Cat. No. 45453, De- partment of Tropical Research, New York Zoological Society; same locality as holo- type; July 26, 1945. The proposed name flammea refers to the color of the male dorsum. Maqo dentichelis sp. nOV. (Text-fig. 8). Diagnosis: Carapace of unrubbed individ- uals with a median white stripe enclosing a central black spot. Male chelicera with tooth on external border; four or five teeth on in- ferior margin; two or three teeth, plus a series of denticles, on superior margin ; tibia of palp with three unequal apophyses; epigynum with a median, rounded, super- ficial, pale anterior body. Color. Color in Life: Adult male. Cephalothorax : Carapace integument black, practically naked except for a conspicuous median stripe of white scales enclosing, behind level of PLE, a central black spot. The stripe begins behind AME, or near level of PME, widens to encompass the spot, then narrows once more, ending at or behind middle of thorax. White of spot region sometimes extending laterally as a short cross-bar. Sparse chest- nut and black hairs scattered on ocular quadrangle near dorsal eyes, and around AME. The wide clypeus is black and com- pletely naked; palps, mouthparts and first pairs of legs black, except for leg tarsi. These and entire third and fourth legs translucent brown, variably and faintly banded with darker near ends of segments. Palps and all legs, especially first two, with inconspicuous white scale-hairs on antero-dorsal surfaces near joints. Sternum black. Abdomen: Pat- tern of dorsum very variable, formed chiefly of short hairs or scale hairs, brown mixed with gray and white areas. Usually a white lyre-shaped anterior marking — a strongly curved band with a short median basal stripe — is distinct; this is followed by several pairs of faint chevrons and some white lateral streaks and spots. The most constant mark- ings are a pair of white terminal spots. Ven- ter black with a pair of pale faint longi- tudinal stripes in middle; buff hairs rather thickly scattered over entire surface. Adult female. Dorsal markings very simi- lar to those of male, but posterior abdominal spots less distinct and more variable. Palps pale, translucent horn; first and second legs banded, not black; white scale-hairs on ap- pendages almost or completely absent, though short yellowish hairs sometimes pre- sent near joints. In alcohol, the distinctive markings usu- ally disappear from both sexes. Structure. The characteristics below apply to both males and females unless otherwise speci- fied ; percentages approximated ; measure- ments of types given on p. 51. Carapace : Height 57 % of carapace length ; profile rises behind AMF, gently convex, to PLE; anterior half of thorax descends very gently, posterior half abruptly; widest at level of PLE, 1.3 times height, 73% of length; total length of eye group slightly more than half carapace length. A distinct longitudinal thoracic groove, centering at level of posterior margin of PLE. Eyes: Length of ocular quadrangle about two-thirds of breadth, its sides almost paral- lel but width at ALE slightly greater than at PLE ; carapace extending well beyond PLE at their level; PME slightly closer to ALE than to PLE. Diameter of AME 23% of carapace length; ratio of eyes, holotype: AME : ALE : PME : PLE :: 100:46:14:40. AME practically contiguous, separated from ALE, which are slightly recurved, by about a tenth of their diameter. Clypeus : Height 52% of AME diameter in male, 28% in female. Chelicerae: Not produced, vertical, paral- lel. Length of basal segment less than 30% of carapace length. Male with a strong tooth about middle of external border. Promargin 50 Zoologica: New York Zoological Society [34: 7 Text-fig. 8. Mago dentichelis: A-E, holotype $: A, carapace and abdomen, dorsal view; B, carapace, lateral view; C, chelicera, ventral view; D, palp, ventral view; E, same, ectal view; F, threat display. G, paratype $: epigynum. with two (rarely three) moderate-sized teeth at proximal angle, the distal the larger; distal to these is a series of minute granular teeth, numbering three or more. Inferior margin usually with four, sometimes five, contiguous, well developed teeth. Maxillae : Less than twice as long as wide, outer distal angle little dilated. Lip: Length and breadth similar; pos- terior margin slightly convex, about equal in breadth to anterior margin of sternum. Sternum: Breadth three-fourths of length in male, two-thirds in female, widest at an- terior margin of third leg. Anterior border concave, posterior broad and convex, ending before anterior half of fourth coxae; pos- terior half of latter separated by about an eighth of their diameter. Legs: Tibial indices: Holotype male, first leg 17, fourth 29; paratype female, first leg 23, fourth 16. First femur, patella and tibia moderately enlarged, less so in second leg. See Table VIII for formula. All legs with little hair. 19491 Crane: Salticid Spiders; Systematics and Behavior in New Species 51 TABLE VIII. M. dentichelis : Leg Formula. 1 4 3 2 Male holotype 2.1 2.0 1.9 1.9 4 3 1 2 Female paratype 1.9 1.8 1.7 1.6 Spines: First leg: Femur dorsal 0-1-1-1, prolateral distal only 2; patella prolateral only 1 or 0; tibial prolateral 1-0-1 (both weak), or 0-0-0; retrolateral 0; ventral, lr-lr-2, or 2-2-2; metatarsus ventral only 2-2. Second leg: Femur dorsal 0-1-1-1, pro- lateral distal only 2, retrolateral female only 1; patella prolateral 1 or 0; tibia prolateral 1-1-1 or 1-0-1, retrolateral 0, ventral lr-2-2; metatarsus ventral only 2-2. Third leg: Femur dorsal 0-1-1-1, prolateral 1 or 2, retro- lateral 1 or 0; patella prolateral 1, retro- lateral 1 ; tibia prolateral 1-1, retrolateral 1-1-1; ventral lp-0-2; metatarsus prolateral 1-1, retrolateral 1-2, ventral 2-2. Fourth leg: Femur dorsal 0-1-1-1, prolateral 1, re- trolateral 1 ; patella prolateral 1, retrolateral 1 ; tibia prolateral 1-1-1 or 1-1, retrolateral 1-1-1, ventral lp-2; metatarsus prolateral 1-1, retrolateral 1-1-2, ventral lp-2. Abdomen: Rather narrowly ovate, widest near middle. Palp: Femur slightly curved, tibia about 70% length of patella; tibia with three apo- physes, one small and ventral, one long and tapering, external to the first, and the third still larger, sinuously tapering, dorso-lateral. Embolus short and simple. Epigynum: A large, rounded, median, whitish anterior area, followed by a variable arrangement of four or five subdermal, near- median tubules, related to two less distinct, well separated oval bodies. Measurements. Male holotype: Total length in alcohol 5.2 mm.; carapace length 2.6, breadth 1.9, height 1.5; total length of eye group 1.4; ocular quadrangle length 1.1, breadth 1.7 ; diameter AME .60, ALE .28, PME .09, PLE .24; cly- peus height .31 ; basal segment of chelicera .99; sternum length .99, breadth .75; ab- domen length 2.6, breadth 1.5; patella breadth, 1st leg, .41, 4th .39. Leg Measurements, Male. Leg Femur Pat. Tib. Metat. Tarsus Total 1 1.6 .96 1.4 .99 .58 5.5 2 1.4 .89 1.1 .92 .58 4.9 3 1.6 .75 1.1 1.1 .48 5.0 4 1.5 .68 1.2 1.2 .62 5.2 Palp .89 .38 .27 — .68 2.2 Female paratype: Total length in alcohol 5.3 mm.; carapace length 2.5, breadth 1.8, height 1.4; total length of eye group 1.4; ocu- lar quadrangle length 1.1, breadth 1.6; diameter AME .55, ALE .26, PME .09, PLE .24; clypeus height .15; basal segment of chelicera .79; sternum length .96, breadth .65; abdomen length 2.8, breadth 2.0; patella breadth, 1st leg .40, 4th .28. Leg Measurements, Female. Leg Femur Pat. Tib. Metat. Tarsus Total 1 1.3 .79 .96 .72 .38 4.2 2 1.3 .79 .85 .65 .44 4.0 3 1.4 .79 .92 .89 .51 4.5 4 1.5 .65 1.1 1.1 .58 4.9 Palp .68 .41 .34 — .38 1.8 Behavior. Locomotion: Not specially observed in this species; however, another Mago (unde- scribed) as well as Hypaeus sp. are both ex- cellent jumpers. In these the repeated pat- tern of ordinary progress is a deliberate walk for two or three centimeters followed by a series of short jumps; the first legs take ac- tive part in the walking and jumping, and are never raised except during display. Courtship Display: Stage I. Carapace ele- vated only enough so that the motionless, hanging palps clear the ground; first legs raised at a wide angle to each other (about 135°), the other legs extending far side- wards, the second pair slightly forward. Pos- ing in this attitude is extended, but at in- tervals the first legs wave alternately up and down. Meanwhile the abdomen, which is held horizontally clear of the ground, is occasionally vibrated briefly up and down. Stage II. First legs extend to front, usu- ally not befoi-e female thrusts her first legs momentarily forward. Carapace and legs of male, in addition to the abdomen, twitch and jerk before he touches her. Threat Display: Much more active than courtship, and in several respects quite dis- tinct. Stage I : Carapace held moderately low, the abdomen either straight out as in courtship, or relaxed downward for silk at- tachment. First legs held with femur bent obliquely up, the other segments out; from that joint the two legs are waved up and down, usually in unison with each other, sometimes alternately. The palps hang down outside the closed chelicerae, as in courtship. Stage II. The tempo and span of waving increases, the first legs almost meeting over- head at peak of display. Series of waves are punctuated by the rapid rubbing together of the first and second tarsi of each side, the second legs are braced somewhat forward, as in courtship, and are occasionally lifted briefly from the ground during waving. Stage III. The two males oppose each other closely, the first legs straight overhead, prac- tically or completely touching, the palps swung obliquely out, and the chelicerae opened wide and knocking against each other for seconds at a time. I have seen this stage reached only twice, no injury being inflicted either time. Only when one was retreating did the abdomen twitch very briefly, as in courtship. Habitat: Known only from the montane cloud forest (about 3,600 feet) around 52 Zoologica: New York Zoological Society [34: 7: 1949] Rancho Grande, taken from vines on tree trunks, herbs and shrubs. Several specimens collected on upper Rancho Grande verandah, many yards from vegetation. Affinities: Apparently related to Simon’s briefly described longidens and acutidens from Brazil, although distinct in details of white markings, distal dentition of cheli- cerae and presence of three apophyses on palpal tibia. Material: A total of 6 adult males and 11 adult females have been preserved in addi- tion to a number of young. The following have been designated as types: HOLOTYPE: Male. Cat. No. 45454, De- partment of Tropical Research, New York Zoological Society; Portachuelo, Rancho Grande, near Maracay, National Park of Aragua, Venezuela; 1,136 meters; cloud for- est; June 6, 1945. PARATYPE: Female. Cat. No. 45455. Taken near holotype, same locality and date. The name dentichelis is proposed in refer- ence to the large outer tooth of the chelicera. References. Beebe, W., and Crane, J. 1947. Ecology of Rancho Grande, a Sub- tropical Cloud Forest in Northern Venezuela. Zoologica, Vol. 32, No. 5, pp. 43-60. Bristowe, W. S. 1941. The Comity of Spiders. London, printed for the Ray Society. Vol. 2. Cambridge, F. 0. P. 1901. Arachnida. Araneidea and Opiliones, Vol. II, in Biologia Centrali- Ameri- cana. Chickering, A. M. 1946. The Salticidae (Spiders) of Panama. Bull. Mus. Comp. Zool., Harvard Coll., Vol. 97. Crane, J. 1948.1 Comparative biology of salticid spiders at Rancho Grande, Venezuela. Part I. Systematics and life histories in Cory- thalia. Zoologica, Vol. 33, No. 1, pp. 1-38. 1948.2 Comparative biology of salticid spiders at Rancho Grande, Venezuela. Part II. Methods of Collection, Culture, Obser- vation and Experiment. Zoologica, Vol. 33, No. 9, pp. 139-145. Mello-Leitao, C. de 1945. Aranas de Misiones, Corrientes y Entre Rios. Rev. Mus. La Plata (n.s.), 4, Zool., 29, pp. 213-302. Peckham, G. W. and E. G. 1894. Spiders of the Marptusa group of the family Attidae. Occ. Pap. Nat. Hist. Soc. Wisconsin, Vol. 2, 1892-1895, pp. 85-141. Simon, E. 1900. Descriptions d’especes nouvelles de la famille des Attidae. Ann. Soc. Entom. France, Vol. 69, pp. 27-61. 1901. Descriptions d’especes nouvelles de la famille des Salticidae. Ann. Soc. Entom. France, Vol. 70, pp. 66-76. 1902. Descriptions d’arachnides nouveaux de la famille des Salticidae (Attidae). Ann. Soc. Entom. Belgique, Vol. 46, pp. 24-54. Beebe: Swifts of Rancho Grande, and Their Migration 53 8. The Swifts of Rancho Grande, North-central Venezuela, with Special Reference to Migration.1 William Beebe. Department of Tropical Research, New York Zoological Society. (Plate I; Text-figures 1-3). [This is one of a series of papers resulting from the 45th, 46th and 47th Expeditions of the Department of Tropical Research of the New York Zoological Society, made during 1945, 1946 and 1948, under the direction of Dr. William Beebe, with headquarters at Rancho Grande in the National Park of Aragua, Venezuela. The expeditions were made possible through the generous cooperation of the National Govern- ment of Venezuela and of the Creole Petroleum Corporation. [The characteristics of the research area are in brief as follows: Rancho Grande is located in north-central Venezuela (10° 21' N. Lat., 67° 41' W. Long.), 80 kilometers west of Cara- cas, at an elevation of 1,100 meters in the undis- turbed montane cloud forest which covers this part of the Caribbean range of the Andes. Ad- jacent ecological zones include seasonal forest, savanna, thorn woodland, cactus scrub, the fresh water lake of Valencia, and various ma- rine littoral zones. The Rancho Grande area is generally subtropical, being uniformly cool and damp throughout the year because of the preva- lence of the mountain cloud cap. The dry season extends from January into April. The average humidity during the expeditions, including parts of both wet and dry seasons, was 92.4% ; the average temperature during the same period was 18° C.; the average annual rainfall over a 5-year period was 174 cm. The flora is marked by an abundance of mosses, ferns and epiphytes of many kinds, as well as a few gigantic trees. For further details, see Beebe & Crane, Zoolog- ica, Vol. 32, No. 5, 1947. Unless otherwise stated, the specimens discussed in the present paper were observed or taken in or over the montane cloud forest zone, within a radius of 1 kilometer of Rancho Grande.] Contents. Page Introduction 53 Streptoprocne zonaris albicincta (Cabanis, 1862) .... 64 Chaetura brachyura brachyura (Jardine, 1846) 57 Chaetura cinereiventris lawrencei Ridgway, 1893 58 Chaeturella rutila brunneitorques Lafresnaye, 1844 ... 58 Cypseloides cherriei Ridgway,1893 59 Cyp8eloide8 cryptus Zimmer, 1945 60 Aeronaute g montivagus montivagus (d’Orbigny and Lafresnaye, 1837) 61 Panyptila cayennensis (Gmelin, 1789) 61 1 Contribution No. 841, Department of Tropical Research, New York Zoological Society. Introduction. In all of South America there have been recorded (Peters, 1940) nine genera of swifts, divided into twenty species and a total of thirty-three kinds, including subspecies. In Venezuela Mr. William H. Phelps informs me there are six genera, of thirteen species of twenty kinds, if we include subspecies. Of these Venezuelan birds, within an area of less than one square kilometer with its cen- ter at Rancho Grande, I have recorded eight species of five genera; roughly eighty per cent, of the genera and sixty per cent, of the total Venezuelan species of swifts. Late in the year 1937 Dr. Alexander Wet- more (Wetmore, 1939) spent some time col- lecting birds near Rancho Grande, and I quote the following notes concerning the swifts. “In Tropical America swifts are tantaliz- ing birds usually seen out of range ... On November 4 at Rancho Grande several (Chae- tura brachyura ) circled out of range. This species appears very black as it flies over- head, so that at first glance it suggests the black swift ( Nephoecetes niger), but a sec- ond look distinguishes it by the shorter, light- colored tail. The specimen taken, a male, measures as follows: Wing 118.7, tail 29.0, culmen from base 5.8, tarsus 11.8 mm.” Concerning Streptoprocne zonaris albi- cincta, he writes, “While I was collecting in Portachuelo above Rancho Grande on Novem- ber 3, 6 and 10, groups of these large swifts dashed at intervals through the pass at light- ning speed with a great rushing of wings. Occasionally I observed them circling in air.” These are, I believe, the only published notes on swifts in this restricted area. My thanks go to Mr. William H. Phelps for the loan of skins of rare swifts, to Dr. Neal Weber for names of ants in the food of birds taken in 1948, and to Dr. J. Bequaert for the name of the feather fly found on Aeronautes. The three text-figures are the work of Miss Louise A. Moore. The photographs were taken by Miss Jocelyn Crane. 54 Zoological New York Zoological Society [34: 8 Streptoprocne zonaris albicincta (Cabanis, 1862). Giant White-collared Swift. Species Range : Southern Mexico and the Greater Antilles, south over northern South America to British Guiana, north Matto Grosso and Peru; vertically to more than ten thousand feet in the Andes. Subspecies Range: Five subspecies are recognized, of which albicincta occurs at Rancho Grande. Its range is extensive, from Honduras south to British Guiana, northern Matto Grosso and Peru, together with the islands of Granada and Trinidad. In Vene- zuela, Mr. Phelps records it as inhabiting the northern mountains. Field Characters for Sight Identification : The most unmistakable species, distin- guished by great size and white nuchal col- lar. It measures eight to nine inches in length, as compared to the five-inch average of the seven other species. Panyptila is the only other Rancho Grande swift with a white collar, but is about half the size of albicincta, and has a deeply forked tail. In young giant swifts the collar is reduced and indistinct in flying birds. Occurrence: February 22 and September 9 are the earliest and the latest dates of our occupancy of Rancho Grande throughout three years. On both dates I recorded giant swifts within sight of the laboratory. Seldom did a day pass between these extremes when one or more did not come into view. Soon after we opened the station I ceased keeping detailed notes on these birds, as their visits seemed governed by no regularity. They commanded attention under four sep- arate conditions: (1) Almost daily either singly, but usually in small flocks, they hawked in the sky after insects, or (2) they flew headlong through Portachuelo Pass, low over the trees. (3) They entered rarely into the diet of a pair of resident bat falcons, Falco albigularis, and (4) on nights of storm, rain or neblina they occasionally struck against the windows of our lighted labora- tory. Throughout the seven months during which we carried on our observations, there was no marked period of absence or extreme scarcity of these swifts. The breeding period must have occurred throughout part of this time but it was not noticeable in the rai'ity or abundance of individuals or flocks. When it became evident that Portachuelo Pass was used as a migrating flyway on an unprecedented scale by other birds and by insects, I watched and noted these passing swifts for a period of several weeks, to see if there was any definite factor or sequence in their numbers or movements. The daily, circling, feeding birds whose general direction was indefinite, varied their elevation, high or low, according to the volant stratification of edible insects. This proved to be definitely associated with the southward migration of insects of many orders through the pass. The swifts often joined flocks of swallows and even of large dragonflies where, on clear days, the migrants offered rich feed- ing in the area of the pass. At times of dense fog, high winds or lowering of temperature, the lessening or cessation of migration was correlated with a total absence of giant swifts. At Kilometer 15, a few kilometers south of the pass, I frequently saw flocks of these birds feeding high in air as I passed in the car; and to the north at Kilometer 30, six or eight pairs of the swif ts were occasion- ally seen hawking about. Beyond these limits I saw no swifts. Giant swifts are supposed to be normal in- habitants of strictly tropical regions. At Kartabo, British Guiana, at practically sea- level, I found them commonly in good-sized flocks, feeding on flying insects, especially in June, July and early August. During this season, mating flights of ants and termites were frequent. From March 14 to July 17 I noted the fol- lowing groups of giant swifts passing on twenty-three days at full speed south through the pass, all between 7 and 8:30 A.M. 1, 16, 6, 2, 11, 4 and 16, 3 and 7, 21, 12, 8, 16, 4, 1, 3, 1, 2, 14, 5, 4, 22, 19, 7, 5. All were in a terrific hurry, flying headlong, mostly low, their whistling wings just clearing the upper branches of bushes and trees. Throughout this period there were only five records of birds going north in early morning and few in numbers, 2, 6, 1, 9, 1. On June 24 at 3 P.M. 64 swifts rushed past over my head, headed full speed northward through the pass, just ahead of the onrolling fog. On June 6, at five in the afternoon, a com- pact flock of 200 to 210 birds, at a moderate height, circled northward, giving the impres- sion of a leisurely, non-feeding migration. On August 1, closely intermingling with about five thousand Argentine martins, Phaeoprogne tapera fusca, about 300 giant swifts accompanied the other birds, all at high speed. On August 8, 24 swifts passed low, going north through the pass. The assumption of the northward return every afternoon and evening of these swifts through Portachuelo Pass seems justified be- cause of the number of birds which long after dark on nights of storm or fog struck against the windows of our laboratory. These acci- dents occurred from 7:20 to 10:45 P.M. Fourteen birds struck in this way on eleven nights, April 9, 12, 18, May 4, 16, 23, June 10, 27, July 3, 4 and 6. On three nights two birds appeared. Four of the swifts which crashed the windows were skinned, three others were sexed, and the remaining seven escaped. All examined were males, and. of those examined, only the two birds which struck on April 9 were in full breeding con- dition. Reviewing the records through the pass, it seems reasonable to assume a daily migra- tion from some more northerly sleeping or breeding place, south to a feeding area, with the return very late in the afternoon or in the evening. 1949] Beebe: Swifts of Rancho Grande, and Their Migration 55 On May 4, four of these swifts fearlessly attacked a bat falcon, the male of the pair whose nesting we were watching. In connec- tion with the attacking and repulsing of this hawk by the the swifts, we were interested to see the same individual falcon on three separate occasions return to his lofty perch with a dead swift. This is a remarkable feat when we realize that the latter is only about one-quarter less in size than the hawk. In an active flight dive the falcon could strike and capture any small bird it selected, but on a level the swifts were superior. In the Santa Marta mountains of Colom- bia, about 575 kilometers west of Rancho Grande and at an altitude of 1,500 meters, this giant swift has been found nesting (Todd and Carriker, 1922). The account is as follows (p. 245) : At the coffee plantation of Cincinnati, “on March 19, 1917, a colony of this large swift was discovered nesting in a shallow cavern behind a waterfall. The place was absolutely inaccessible, so that no idea of the number of nests could be had. Only one nest, which happened to be near the top, was secured, together with the occu- pants, . . . which had been stunned by the blasting, and proved to be an adult female and two recently hatched young. The nest re- sembled very closely that of the Chimney Swift, being composed of twigs fastened to- gether with saliva. The birds entered and left the cavern by dashing through the cur- tain of water falling over the front of it. The altitude of the site was about 4,300 feet.” Data on Collected Specimens. For comparison I have included data con- cerning a female of this species taken many years ago at Kartabo, British Guiana. hundred winged females of small ants. 31188 : Stomach crammed with ants. (Weber). Homoptera: Cicadellidae. Diptera: fly fragments. Hymenoptera: parasitic sp. Hymenoptera: Pheidole sp. Hymenoptera: Atta sexdens Linn. Four gasters and a hind wing frag- ment. A species known from Ciudad Bolivar, Venezuela, and south. Found in Eastern British Guiana, but ap- parently not in Venezuelan Guiana, the Orinoco Delta, N. W. District (B.G.) ; in these places replaced by A. cephalotes Linn. Aztecal : wings. Camponotus ( Myrmobrachys ) sp. Same as I took at 1,020 meters in Rio Porce, Colombia. Camponotus ( Tanaemyrmex ) sub- stitutaus Emery. Distribution : Cen- tral America to Paraguay. I have the same form from Kartabo, B. G. Camponotus ( Myrmobrachys ) crassus Mayr. Distribution: South America. 529 : Three beetles, three wasps, one hem- ipteron, three membracids, one tipu- lid, and upwards of two hundred female ants of six species. It is significant that although swifts 30447 and 30452 struck the laboratory within six minutes of each other, yet their food was quite distinct, indicating very different feed- ing territories. Yet they were headed for the pass, focusing upon a sixty-foot-wide bottle- neck. 30447 male Lgth. 205 Wing 194 30452 male 196 188 31135 male 202 195 31188 male 210 200 529 female 190 190 The relative discrepancies between length and weight are accounted for by the food. The stomach, with contents, of No. 31188 weighed 22.6 grams. Gross food content may be expressed as follows: 30447, crammed with ants; 30452, moderately filled; 31135, empty, after a day in cage; 31188, crammed with ants; 529, only about one-fourth filled with insects. Without exception, all the ants in the food were winged females. Detailed Food. 30447 : At least 800 ants of an undetermined species of Azteca. 30452 : Several hundred females of Dolicho- derus ( Monads ) debilis Emery, and Crematog aster ( Orthocrema ) sp. A single female Solenopsis geminata edwardi Forel. 31135: Five female Atta sp. More than four Tail Grams Weight Extent Date 64 105.8 501 July 3, 1945 61 96 482 July 3, 1945 67 68.5 500 April 9, 1948 70 109.5 July 6, 1948 60 77.2 July 10, 1919 Individual Characters. I find the following recorded concerning the Kartabo female, No. 529: Parasites: Only a few bete rouge on the head feathers. Colors: Bill black, face pale medici blue, iris light brownish-olive, legs and feet vina- ceous slate. Eyelid: Quite bare above. Below, a line of fifteen small feathers along rim. At posterior end of eye a small group of a dozen feathers, arranged in several rows. Oilgland: Elongated, blunt, tapering, bare. Wing Graph: Primaries 10th — 152 mm. 5th — 111mm. 9th— 155 “ 4th— 98 “ 8 th— 149 “ 3rd— 86 “ 7th— 139 “ 2nd— 75 “ 6th— 126 “ 1st— 62 “ 56 Zooloffica: New York Zoological Society [34: 8 Secondaries 1st — 49 mm. 2nd— 50 “ 3rd— 52 “ 4th— 54 “ 5th — 53 mm, 6th— 53 “ 7 th— 49 “ 8th— 42 “ Scalation : Front of tarsus with an indis- tinct irregular line of ten, fleshy scales down the inner aspect. Inner, rear and outer sides of tarsus, bare, wrinkled skin, with no trace of scales. Text-fig. 1. Streptoprocne zonaris albicincta Text-fig. 2. Streptoproc7ie zonaris albicincta (Cabanis). Scalation of tarsus, front and side (Cabanis). Tongue. views. Palate : The palatine fissure begins well toward the front of the roof of the mouth, and divides in front. It is rather long (12.5 mm.) and is guarded by about a dozen pairs of teeth. Four-fifths of the way back there is a double-curved or angular transverse row of teeth, extending out at right angles on each side, with about twenty teeth on each side. The guardian, denticulated flaps end openly posteriorly, and just beyond is the very small tubal fissure. At the extreme pos- terior of the roof of the mouth is a transverse row of about twenty weak teeth. Tongue : Narrow for a swift, and small for the size of the mandibular area. Greatest width of tongue 5.7 mm., length 10; greatest width of mandible 20 mm., length 26. Tongue fleshy, channelled toward the tip, sides some- what sinuate, tapering slowly to two blunt tips. The two cornua are lined along the edge, both on inner and outer sides, with strong teeth. Smaller ones are scattered over the posterior surface of the tongue itself. Glottis : A narrow ellipse on a low flat area, with inconspicuous unarmed rim. Pos- teriorly, there is an irregular transverse row of teeth, all large, flattened, sub-equal in size and numbering about fifteen on each side. Beyond these there arises a second irregular row. Most of the teeth, in a cleared condition, show stout, parallel-sided bases, and slender tips. Syrinx : Swift No. 31135 (KOH No. 2589) . Male. April 9, 1948. There is little change in the posterior tracheal rings except that the last eight are slightly narrowed with more even edges. The syrinx is a wide tracheo- bronchial collar of bone. Anteriorly the upper margin is level except in the center where an irregular, rounded projection overlies a segment of the last tracheal ring. This, like the rest of the syringeal collar is ossified and coarsely fenestrated. The anterior vertical width of the collar is 1.6 mm., its lateral, front to back, length is 3.3 mm. The lower border of the collar is formed by the closely- applied, strongly arched, upper bronchial 1949] Beebe: Swifts of Rancho Grande, and Their Migration 57 Text-fig. 3. Streptoprocne zonaris albicincta (Cabanis). Syrinx, front and side views. semiring. This semiring dips far down in front, forming an acute angle with the pes- salus. The triangular space within this angle, as far anterior as the collar, is ossified, but without fenestration. Posteriorly the tracheal rings are appre- ciably wider, with little more intervening membrane showing than the median, open notches. Two of the rings anastomose. The two lowermost rings are narrow and even as to outline. Posteriorly, the syringeal collar is similar to its anterior half, with the difference that this aspect is flat and the median anterior projection is less pronounced. There is a faint but distinct indication that the present ossi- fied syringeal collar was originally composed of two rings. The entire ventral syringeal aspect, bounded by the lower border of the collar, the pessalus and the first semiring is of course membranous. The free ends of the anterior semirings, joined by the tympani- form membrane, narrow rapidly posteriorly, until the ninth onwards become almost com- plete rings, thus forming the end of the membrane. There are about twenty-three bronchial rings, the second, third and fourth being somewhat longer, projecting slightly into the inner profile of the bronchi. From the eighteenth bronchial ring onwards there is a gradually increasing disintegration of the rings within the lung tissue, a thinning and irregular anastomosing of adjoining rings. A drawing and description of the syrinx of the female No. 529, made thirty years ago, are similar to that of the present male except that posteriorly, the median protuberance is considerably larger, and extends forward over the last three tracheal rings. Chaetura brachyura brachyura (Jardine, 1846). Short-tailed Swift. Species Range-. Same as that of the sub- species below, with the addition of the Lesser Antilles. Subspecies Range: Northern Venezuela and the Guianas to Trinidad and Tobago, south through eastern Ecuador and Peru to Matto Grosso and Para. Field Characters for Sight Identification : This is the smallest of the Rancho Grande swifts (length 100 mm.). On the wing it ap- pears totally black, with conspicuous pale brownish-gray rump, tail-coverts and tail. The absence of gray on the underparts dis- tinguishes it from the slightly larger cinere- iventris. Occurrence : By far the commonest swift at Rancho Grande, becoming really abundant after the rains began. On clear days numbers were often seen feeding with swallows and other swifts, espe- cially when migration of various orders of insects was in full swing. On partly cloudy days or when fog drifted up the lower val- leys, these short-tailed swifts would swing through the pass in small or larger numbers. There was none of the rather regular, southward, morning shift of the giant swifts. We saw this species every month from March to August, and on almost every clear day. It often flew in twos and threes, or again in flocks of considerable size. For three con- secutive days, we were able to identify an individual trio. Two of the birds had recog- nizable gaps in their primaries due to molt, and these marked birds hawked on the north side of the pass, low in the gorge, throughout a three-day flight of termites, Coptotermes testaceus. On the third day an onrushing mass of dense fog drove these swifts away in the wake of eight turkey vultures has- tening toward the upper zone of clear sun- shine. The swifts did not return, and the next day the insect flight had ceased. On July 9, 1948, we observed an unusual flocking, a migration of sorts ; a fairly com- pact mass of considerably more than four 58 Zoologica: New York Zoological Society [34: 8 hundred of these swifts, circling, not feeding. As they approached the pass from the south, the flight changed to a more direct movement, and when siphoning through, all circling ceased, and the rush of wings was like a loud wind as the birds passed low and at great speed. The northern valley was partly filled with fog and the birds rose slowly above it, and before they passed from view, were again circling high in air as before. The gen- eral effect was of a maze of inorganic units, without volition, sucked by a wind through the narrow notch, and then sprayed out in a slower movement as the draught lessened. As a matter of fact, there was little or no breeze in the pass itself. A spectacular coincidental sight of the same type of flocking of the same species is recorded in my notes on an identical July 9, but thirty-two years before, at Kalacoon, British Guiana. The note reads as follows: “An enormous flock of short-tailed swifts appeared over the forest at 9:30 this morn- ing. There were certainly more than a thousand birds, all flying in a great circle, gradually attaining higher and higher alti- tude. They were massed so closely together that there seemed hardly room for any move- ment of the outspi’ead, crescentic wings. The relative slowness of movement and the fre- quent effortless gliding indicated an upward surge of air. Through the glasses a scatter- ing of equally small white-rumped swifts, Chaetura spinicauda, was clearly seen. The whole company vanished very high up and drifting southward.” Although these swifts showed little fear of the pair of bat falcons nesting near the laboratory, yet at least seven individuals fell victims to these hawks, and an eighth mangled swift found in the road near the nesting tree indicated an additional dropped item of diet. In this, as in other species, the swifts were able to evade the raptores when on the same aerial level. It was only when the hawks could go into a vertical dive that their speed made them almost unavoidable. Two short-tailed swifts crashed against the laboratory windows on successive eve- nings, May 16 and 17, but one was able to fly away. The other was a male, breeding, with the stomach quite empty. Chaetura cinereiventris lawrencei Ridgway, 1893. Gray-breasted Swift. Species Range: The range of the eight recognized subspecies extends from Nicara- gua, Grenada, Tobago and Trinidad south to Venezuela, Colombia, Ecuador, Peru, cen- tral Brazil and Bolivia. Subspecies Range: lawrencei occurs in Grenada, Tobago, Trinidad and the moun- tains of northern Venezuela. Field Characters for Sight Identification: A small swift. Black, except for rump and all underparts, which are pale gray. When as- sociated with brachyura and seen from above at a distance, the birds are much alike, the brownish shade of the rump of brachyura being hardly distinguishable. From below, the gray of cinereiventris instantly sets it apart. Occurrence : This is not common at Rancho Grande, but was observed occasionally throughout May, June and July. Only once, on June 9, did an individual come to the lab- oratory windows. It clung for five minutes to the sill but evaded all efforts at capture. On May 9, the male bat falcon caught a gray-breasted swift and plucked it. The fe- male then gave it to her nestlings who tore it apart and ate it. As they pulled it apart I could distinctly see the mass of small ants which filled the stomach, together with two large abdomens of Atta queens which were eaten by the young birds. Chaeturella rutlla brunneitorques Lafresnaye, 1844. Chestnut-collared Swift. Species Range : Central Mexico, south through northern South America to Peru, the Guianas and Trinidad. Subspecies Range: Southeastern Mexico, south to Colombia, Venezuela, Ecuador and Peru. Field Characters for Sight Identification: A medium-sized swift, about five inches in length. It is the only species marked with chestnut; throat, breast and collar. This color is especially distinct when the birds are silhouetted against the foliage of the moun- tain jungle, but with glasses is conspicuous even when they are high in the sky. Occurrence: The chestnut-collared swift was third in order of abundance at Rancho Grande, surpassed only by Streptoprocne and Chaetura brachyura. It was frequently seen shuttling back and forth through the pass, or hawking about on days of insect mi- gration, associating with giant swifts or with swallows, mostly single birds or in small flocks. Occasionally they would race back and forth through the pass, yet the diurnal ob- servations I was able to make showed no cer- tain regularity of north or southward shift. On the other hand sixteen birds struck against the laboratory windows and on three occasions, when no swifts actually flew against the glass, individuals were seen flut- tering about among the bats within the area of illumination. This would indicate a daily, crepuscular northward migration, as in Streptoprocne, also perhaps to some roosting or breeding colony. Support of the probability of such a mi- gration is furnished by a chestnut-collared swift (31129) taken eight kilometers east of Rancho Grande. A reliable assistant, Pe- dro Infante, shot this bird January 8, before my ai'rival, on the Choroni road which par- allels that from Maracay to Ocumare. He re- ported this swift as shot from a group of thirty to forty which, throughout October, November and December assembled every evening and spent the night clinging in a 1949] Beebe: Swifts of Rancho Grande, and Their Migration 59 ' compact mass to the vertical side of a rocky , cliff near the road. All left at dawn. Thus we have evidence of a roosting colony at the same elevation as the pass, and, from the point of view of a volant swift, only a short distance away. At Rancho Grande the meteorological con- ditions which induced the appearance of the swifts at the lighted windows were, high but dense clouds; low neblina fog with or without wind ; precipitation, whether drizzle or pelting rain and with or without light- ning. No birds ever came on clear nights, whether moon or starlighted, and no bird after 10 P. M. There was considerable variation in the amount of chestnut on the plumage of these swifts, but the typical pattern was rich chestnut throat and breast with a wide col- lar extending over nape and hind neck. Two adult males had the chestnut reduced to a pectoral tinge, and several swifts on the wing were intermediate between these ex- tremes. The only female examined (31143), an adult, showed no pigmental difference from a full-plumaged male, except that the under tail coverts were strongly edged with white. Coleoptera spp., including a ceram- bycid. Hymenoptera : Camponotus sp., fragmentary remains of several hundred. 31132: Half the meal composed of ants (Weber) . Coleoptera spp. Hymenoptera: Camponotus sp., fragmentary. 31143: Dominant food, hundreds of small flying ants (Weber). Hemiptera spp. Coleoptera: cucurlionid. Hymenoptera: Pheidole sp. Hymenoptera: Very small dolicho- derine fragments. Cypseloides cherriei Ridgway, 1893. White-spotted Swift. Former Records and Species Range : Two swifts taken on Volcan de Irazu, central Costa Rica, were described by Ridgway in 1893. The type was thought to be a male; the second bird was uncertainly sexed as a female. The most noticeable character was “a large, sharply defined spot of silky white Data on Chestnut-collared Swifts Lgth. Wing 30382 male 126 123 30382a male 30448 male 133 128 Four swifts escaped 31128 male 115 120 One swift escaped 31131 male 130 120 31132 male 128 120 Four swifts escaped 31143 female 122 125 Tail Grams Weight Extent Date 45 25 May 14, 1945 43 22.9 320 May 14, 1945 July 3, 1945 36 292 July 3, 1945 Mar. 23, 1948 43 21.5 295 Mar. 23,1948 April 3, 1948 45 21.5 302 April 3, 1948 43 19.5 305 April 3, 1948 April 24, 1948 Detailed Food. 30382: Many winged females of Campono- tus (T anaemyrmex) coruscus F. Smith; and Solenopsis geminata edwardi Forel. 30448: A series of winged females of Cre- matogaster ( Orthocrema ) sp. 31128: Stomach crammed with a mass of winged ants and small beetles. (Weber) . Homoptera: Cicadellidae. Coleoptera spp., including a carabid. Hymenoptera: parasitic sp. Hymenoptera : Crematogaster ? wing. Hymenoptera: Solenopsis geminata Fabr., widespread in the northern neotropics and replaced in the south by saevissima F. Smith, from the interior of British Gui- ana (Courantyne) and Brazil. Camponotus sp. 31131: Food dominantly flying ants (Weber). Hemiptera: wing. on each side of the forehead, immediately over the lores, and a short streak of the same color immediately behind the eye.” A third specimen was reported (Zimmer, 1945) in a collection of birds from Colombia. This was taken at San Gil, Santander, and was also questionably sexed as a female. This then, on February 26, 1948, was the summation of our knowledge of the white- spotted swift. The Costa Rican birds came from the same 10th degree of north latitude as Rancho Grande, but 1,800 kilometers west; whereas the Colombian swift, from 6 degrees, 33 minutes north latitude, was 750 kilometers to the southwest of our labora- tory. Field Characters for Sight Identification : The white-spot is a five-inch swift, appear- ing uniformly black, with a conspicuous, round, white spot between beak and eye. These spots stand out strongly whenever these birds are seen head-on in flight or from the side, giving a rather fantastic im- pression of a slightly misplaced pair of bril- liant eyes. Occurrence : At 9.30 o’clock in the evening 60 Zoologica: New York Zoological Society [34:8 of February 26, 1948, a white-spotted swift came to the windows of Rancho Grande and was caught. From this time until June 13 we captured or recorded eight others, making a total known of this unusual species of twelve individuals. The details of the nine Rancho Grande birds are as follows. 31125: Female not breeding. February 26, 1948. Length 132, wing 123, tail 132, ex- tent 308 mm. Black above, sooty brown be- low; supra-loral spot and small post-ocular patch white; small feathers along edge of wrist and front of wing white-edged; trace of white on chin. Ovary small but distinct. At 10 P.M. this bird fluttered against a bedroom window. It was later found and cap- tured on the Rancho Grande porch, fluttering confusedly around the electric light. The eve- ning was one of dense fog, with a strong breeze blowing from the southwest. Food: A mass of rather comminuted flying ants. (Weber). Coleoptera spp. Hymenoptera: Highly fragmentary Cam- ponotus sp., forming most of the con- tents. 31133: Male, not breeding, April 4, 1948. Length 120, wing 120, tail 39, extent 305 mm. Weight 22.5 grams. Frontal spots large and pure white, begin- ning on lores with only a few feathers be- tween them and nostrils, and extending back over eye frame, and on a narrow line to mid- way over eyes, thus approaching the post- ocular spot. This latter forms the posterior border of the feather circle around the bare area on the lower lid. When the eye is closed and this lid drawn up, the spot is directly behind the eye. When the eye is open it is behind and below eye. Wrist edge of wing and rim featherlets all have distinct white edges. The chin is grayish-white. The bird was caught at 8:30 P.M. as it clung to the vertical electric light wire de- pending from the ceiling of the porch outside the laboratory. There was sufficient fog to hide the stars and the recent wind had died down. The air was cool, 62 degrees Fahren- heit, sufficient to keep all moths away. Food: Flying ants. (Weber). Coleoptera spp. Hymenoptera: Camponotus sp. Syrinx: C. cherriei, No. 31133 (KOH No. 2588) differs from Streptoprocne zonaris al- bicincta in there being three, instead of two, rows of post-glottid teeth. The syrinx proper, although completely ossified, shows distinctly its composition of three rings. The median anterior protuberance is directly connected with a slight, posterior, cartilaginous projec- tion of the lowermost free tracheal ring. 31134: Female, not breeding. April 5, 1948. Length 137, wing 127, tail 51, extent 310 mm. Weight 25.5 grams. Large white preocular spots almost join white chin. Postocular and white wing edges well developed. Flew against laboratory win- dows at 7 :30 P.M. in dense cold fog. Food : Stomach crammed with insects, one- half of which were ants (Weber). Hemiptera spp. Coleoptera spp. Hymenoptera: parasitic sp. Hymenoptera: Camponotus sp. fragmen- tary. Hymenoptera: dolichoderine wings. April 5. At 7 :45 two more white-spots came to the porch but both escaped. At 8:10 another bird came and went. In all, the men- tal white was almost absent. April 11. In dense, drenching fog a white- spot came to my bedroom window at 9:30 P.M. and clung out of reach to the rough surface of a cement pillar. Eye-white as usual, with more on the chin than in any bird hitherto seen. May 10. Male bat falcon caught a white- spot, held it for three minutes, with the dangling head in full view. Hardly any men- tal white, but very large and fluffed out eye spots. After plucking it he gave it to the female who fed her young. June 13. Female bat falcon brought a swift to her perch, and had begun plucking it when the young male flew up, took it and ate it. Cypseloides cryptus Zimmer, 1945. Tropical Black Swift. Former Records and Species Range: In 1945 a new species of swift was described (Zimmer, 1945). The type came from the Rio Tavara, Peru. Only four other specimens were known, taken at the following localities : British Guiana (Kaieteur Falls), Venezuela (Mt. Auyan-tepui, and Saroropan-tepui) , and Costa Rica (San Pedro). Only a single individual of this species was seen at Rancho Grande in 1948. The two which were taken in 1946 have already been reported by me (Beebe, 1947), and I here repeat several paragraphs. Field Characters for Sight Identification : This five-inch black swift would show no definite characters in flight except the gen- eral black coloration. The grizzled and vari- able dull whitish of the lores and chin could hardly be detected. Occurrence: On April 20, 1946, a female of this swift crashed against the laboratory windows at Rancho Grande at 8 :30 o’clock in the evening and was stunned. It is No. 30,634, female, not breeding, fairly fat, weight 40.2 grams. Length 120, wing 137, tail 48, extent 355 mm. The stomach was filled with winged female Azteca ants. On April 21, the following evening, at the same time, a second bird killed itself against the identical window. This is No. 30,640, fe- male, not breeding, considerable fat, weight 35.8 grams. Length 138, wing 130, tail 50 mm. First primary in each wing half grown. Food: A great quantity of Crematogaster and Azteca flying ants. April 12, 1948. At 10:15 A.M. the male bat falcon swung up to his perch in the top of the candelo tree, with a swift. Through the 1949] Beebe: Swifts of Rancho Grande, and Their Migration 61 20-power glasses I could see every detail and in every respect of the cephalic pale color it seemed to be this species. There was no trace of the supraloral white spots of cherriei, and the area around the base of the beak showed the pale grizzled appearance so apparent in both of the specimens taken in 1946. I could have had no more certain evidence if the bird had been in my hand. It was slowly and thoroughly plucked and as the female did not appear, the male proceeded to eat the eighth known individual of Cypseloides cryptus. Aeronautes montivagus mantivagus (d’Orbigny and Lafresnaye, 1837). White-breasted Swift. Species and Subspecies Range : Mountains of northern Venezuela, Peru and Bolivia. Field Characters for Sight Identification: A small five-inch swift. Easily distinguished by great extent of ventral white, no nuchal collar, and almost square tail. Occurrence : Known at Rancho Grande laboratory from a single specimen which flew into the porch in dense neblina, and at about 9:15 P.M. was caught as it crouched in a corner. 31142: Adult male, breeding, testes 8.5 mm. April 23, 1948. Length 120, wing 110, tail 40, extent 266 mm. Weight 20 grams. Food : Many flying ants and small cucurli- onid beetles (Weber). Hymenoptera: ponerine and Camponotus ant fragments. Parasites: This swift was strongly in- fested with parasites, one of which (48375) was a giant feather fly with bright green ab- domen. In addition, there were several Mallo- phaga, and a number of bete rouge. Dr. J. Bequaert has kindly identified the large feather fly, and sends me the following note: “The fly is Brachypteromyia neotropica J. Bequaert. This was described from a single male, taken from the same host species, at Galipan, close to Pico Avila, Estado Mirando, 2,000 meters elevation, Venezuela.2 Yours is the second specimen known, also a male. The description, with figure, is in Psyche, 49. (1942) published in 1943, p. 113. The only other species of the genus, Brachypteromyia fimbriata (Waterhouse), is North American, on the swifts Aeronautes saxatilis and Nephoecetes niger.” In addition to the single captured specimen of this swift, we have three other records. June 19, 1948. Six white-breasted swifts flew, one after the other, through the pass at 10 A.M. They were headed south and not flying very fast. While still in sight two of the birds veered aside from their direct flight and caught insects. June 21, 1948. A compact flock of twelve of these swifts swung south through the pass at 8:05 in the morning. They flew very low, just skimming the trees. 2 Pico Avila is in the immediate neighborhood of Caracas, about 100 kilometers due east of Rancho Grande. Three of this species were caught by the male bat falcon; on April 4, June 10 and 19, 1948. Panyptila cayennensis (Gmelin, 1789). Fork-tailed White-collared Swift. Species Range: Southeastern Nicaragua, south over Colombia, Ecuador, Venezuela, Tobago, Trinidad and the Guianas to Bahia and Sao Paulo. Field Characters for Sight Identification: A five-inch swift, unmistakably fork-tailed, black except for white eye-spots and flank- spots, chin, throat and collar. Occurrence: 30439: A male flew against the windows of the laboratory and was badly injured. July 1, 1945. Length 123, wing 125, tail 57 mm. Food: Small species of flying ants. (This specimen was overlooked in the paper on Avian Migration at Rancho Grande, Beebe, 1947). June 8, 1948. Six fork-tailed swifts hawk- ing about early in the morning with three blue and white swallows, over the compound of Rancho Grande. June 12, 1948. A swift of this species caught and eaten by male bat falcon. Summary. At Rancho Grande a total of eight species of swifts were collected or observed, out of the thirteen species recorded from Venezuela as a whole. The types of observation resolve into: flocks feeding at various altitudes, others migrating through Portachuelo Pass usually low down, or striking against the lighted windows of the laboratory on nights of fog or rain. Finally a number of the birds were caught by a male bat falcon. Throughout twenty months of residence during three years no swift was seen to alight, nor was there at Rancho Grande first- hand proof of breeding or roosting colonies, although the latter were indicated as a result of various activities. Observations in adjoining areas, both higher up the surrounding mountains, and down to four hundred and forty-five meters on the Maracay plain, showed a relative dearth or absence in comparison with their numbers in the square kilometer whose cen- ter was Rancho Grande and the pass. The obvious explanation of this concentration of swifts, by day and night, must be the same as that of many other organisms, both ver- tebrates and invertebrates; viz., the contin- ual procession on migration of countless numbers of insects representative of almost every order, traversing the sixty-foot-wide pass, from north to south, on every clear day throughout the rainy season. This abun- dant and ever renewed source of food was obviously a focusing factor of prime impor- tance. A second reason for the abnormal numbers of species and individuals was the use of the pass by several of the species on daily mi- gration from a presumed breeding or roost- 62 Zoologica : New York Zoological Society [34: 8: 1949] ing place to a trans-pass feeding-area. It is difficult otherwise to account for the forty- four specimens of all eight species taken or observed on black nights of poor visibility. Of twenty-two specimens sexed, seventeen were males, five females. The presence throughout their breeding season of a pair of bat falcons, close to Rancho Grande, revealed an interesting rela- tionship between these birds and the swifts. The latter showed little fear of the small fal- cons when these were perched, and giant swifts did not hesitate to attack and drive off the male hawk. Yet I recorded nineteen individuals and seven out of the eight spe- cies of swifts as entering into the diet of the hawks. When high in the sky, a power dive attack of the male falcon rendered es- cape impossible on the part of the swifts, but when the birds met at horizontal levels, the speed and dodging ability of the small birds rendered them safe. The flocking habits of these Venezuelan swifts are, in some ways, suggestive. Single birds were very rare, and pairs were not often seen. But, especially in mid rainy sea- son months, trios were common. Even in flocks of twenty to fifty, feeding in midair, subdivision into trios was often evident. If the same rule of a single surviving young holds in tropical swifts as in many other tropical birds, these trios probably repre- sented the season’s families. Larger flocks in rapid movement were too infrequent to war- rant definite classification or object. Literature Cited. Beebe, William & Jocelyn Crane 1947. Ecology of Rancho Grande. Zoologica, 32: 43-59. Peters, J. L. 1940. Check-list of Birds of the World. IV: 1-291. Ridgway, R. 1893. Description of Two Supposed New Species of Swifts. Proc. U. S. Nat. Mus., XVI: 43-44. Todd, W. E. C. & M. A. Carriker, Jr. 1922. The Birds of the Santa Marta Region of Colombia. Annals of the Carnegie Museum, XIV : 3-582. Wetmore, A. 1939. Observations on the Birds of Northern Venezuela. Proc. U. S. Nat. Mus., 87: No. 3073, 173-260. Zimmer, J. T. 1945. A New Swift from Central and South America. Auk, 62 : 586-592. EXPLANATION OF THE PLATE. Plate I. Two migrant swifts which came to the electric lights of Rancho Grande on nights of rain or fog. Fig. 1. Giant White-collared Swift. Strepto- procne zonaris albicincta (Cabanis). Fig. 2. Chestnut-collared Swift. Chaeturella rutila brunneitorques Lafresnaye. BEEBE. PLATE I. FIG. 2. THE SWIFTS OF RANCHO GRANDE. NORTH-CENTRAL VENEZUELA, WITH SPECIAL REFERENCE TO MIGRATION. Hertlein & Strong: Mollnsks of Mexico and Central America 63 Q Eastern Pacific Expeditions of the New York Zoological Society. XL. Mollusks from the West Coast of Mexico and Central America. Part VII.1 Leo George Hertlein & A. M. Strong. Calif ornia Academy of Sciences. (Plate I) . [This is the fortieth of a series of papers deal- ing with the collections of the Eastern Pacific Expeditions of the New York Zoological Society made under the direction of William Beebe. The present paper is concerned with specimens taken on the Templeton Crocker Expedition (1936) and the Eastern Pacific Zaca Expedition (1937- 1938) . For data on localities, dates, dredges, etc., refer to Zoologica, Vol. XXII, No. 2, pp. 33-46, and Vol. XXIII, No. 14, pp. 287-298.] Contents. Page Introduction 63 Superfamily Tellinacea 64 Family Tellinidae 64 Genus Tellina Linnaeus 64 Subgenus Tellinella Morch 64 Tellina ( Tellinella ) cumingii Hanley 65 Tellina (Tellinella) zacae Hertlein & Strong, sp. nov 65 Subgenus Scrobiculina Dali 66 Tellina ( Scrobiculina ) ochracea Carpenter ... 66 Tellina ( Scrobiculina ) viridotincta Carpen- ter 66 Subgenus Moerella Fischer 67 Tellina ( Moerella ) amianta Dali 67 Tellina ( Moerella ) arenica Hertlein & Strong, sp. nov 68 Tellina ( Moerella ) erythronotus Pilsbry & Lowe 69 Tellina (Moerella) felix Hanley 70 Tellina ( Moerella ) macneilii Dali 70 Tellina (Moerella) paziana Dali 71 Tellina (Moerella) recurvata Hertlein & Strong, sp. nov 71 Tellina ( Moerella ) suff usa Dali 72 Tellina (Moerella) tabogensis Salisbury 72 Subgenus EuryteUina Fischer 73 Tellina (EuryteUina) eburnea Hanley 73 Tellina (EuryteUina) inaequistriata Donovan . 74 Tellina (EuryteUina) laceridens Hanley 75 Tellina (EuryteUina) mantaensis Pilsbry & Olsson 75 Tellina ( EuryteUina ) panamanensis Li 76 Tellina (EuryteUina) planulata Sowerby .... 76 Tellina (EuryteUina) prora Hanley 77 Tellina (EuryteUina) regia Hanley 78 Tellina (EuryteUina) rubescens Hanley 78 Tellina (EuryteUina) simulans C. B. Adams.. 79 Subgenus Tellinidella Hertlein & Strong, subgen. nov 79 Tellina (TeUinidella) purpureus Broderip & Sowerby 80 Subgenus Macaliopsis Cossmann 81 Tellina (Macaliopsis) lyra Hanley 81 Tellina ( Macaliopsis ) lyrica Pilsbry & Lowe . . 81 Subgenus Merisca Dali 82 TeUina (Merisca) crystalling Spengler 82 Tellina (Merisca) proclivis Hertlein & Strong, sp. nov 83 Tellina (Merisca) reclusa Dali 84 Subgenus Scissula Dali 84 TeUina ( Scissula) cognata C. B. Adams 84 TeUina (Scissula) nicoyana Hertlein & Strong, sp. nov 85 Tellina (Scissula) virgo Hanley 86 Subgenus Phyllodina Dali 86 Tellina (Phyllodina) pristiphora Dali 86 Subgenus Phyllodella Hertlein & Strong, subgen. nov 87 1 Contribution No. 842, Department of Tropical Research, New York Zoological Society. . Tellina ( Phyllodella ) insculpta Hanley 87 Subgenus Elliptotellina Cossmann 87 TeUina (Elliptotellina) pacifica Dali 87 Genus Tellidora Morch in H. & A. Adams 88 Tellidora burneti Broderip & Sowerby 88 Genus Macoma Leach 88 Subgenus Macoma s.s 88 Macoma (Macoma) nasuta Conrad 88 Subgenus C ymatoica Dali 89 Macoma (Cymatoica) undulata Hanley 89 Subgenus Psammacoma Dali 89 Macoma (Psammacoma) elongata Hanley. ... 89 Macoma (Psammacoma) lamproleuca Pilsbry & Lowe 90 Macoma (Psammacoma) panamensis Dali .... 91 Macoma (Psammacoma) panamensis spectri Hertlein & Strong, subsp. nov 91 Subgenus Psammotreta Dali 92 Macoma (Psammotreta) aurora Hanley 92 Macoma (Psammotreta) pads Pilsbry & Lowe 92 Subgenus Macoploma Pilsbry & Olsson 93 Macoma (Macoploma) medioamericana Olsson 93 Genus Apolymetis Salisbury 93 Apolymetis cognata Pilsbry & Vanatta 93 Apolymetis dombei Hanley 94 Genus StrigiUa Turton 95 Strigilla cicercula Philippi 95 StrigiUa costulifera Morch 95 StrigiUa disjuncta Carpenter 96 Strigilla lenticula Philippi 96 Introduction. This is the seventh of a series of papers dealing with collections of mollusks taken on the Templeton Crocker Expedition (1936) and the Eastern Pacific Zaca Expedition (1937-1938). The general plan of presenta- tion followed in the present contribution is that mentioned in Part II of this series of papers2. Formal headings and keys are given for 51 species and subspecies of the Tellini- dae collected by the expeditions of 1936 and 1937-1938. Occasionally additional species are included in the keys for convenience but in such cases it is indicated which species do not occur in the present collection. Acknowledgment is due Dr. G. Dallas Hanna, Curator, Department of Paleontology of the California Academy of Sciences, Mr. A. G. Smith, Research Associate of the same institution, and Dr. A. Myra Keen, Stanford University, California, for assistance and suggestions. Acknowledgment is also due Miss Viola Bristol, Curator of Mollusks, San Diego Society of Natural History, for the loan of specimens. The photographs used for illustrations on the plate were prepared by Mr. Frank L. Rogers. 2 Hertlein, L. G., and Strong, A. M. Eastern Pacific Expeditions of the New York Zoological Society. XXIII. Mollusks from the West Coast of Mexico and Central America. Part II. Zoologica, New York Zool. Soc., Vol. 28, Pt. 3, December 6, 1943, pp. 149-168, pi. 1. See especially pp. 149-150. 64 Zoologica: New York Zoological Society [34:9 Superfamily Tellinacea. Family Tellinidae. Papers by Dali3 and Salisbury4 dealing with the Tellinidae are very useful in a study of West American members of this family. Key to the Genera of the Family Tellinidae. A. Shell with lateral teeth in one or both valves a. Exterior with oblique, flexuous, divari- cating striae; orbicular Strigilla aa. Exterior usually without, sometimes with, oblique but not flexuous, divari- cating striae b. Dorsal margins serrate; trigonal; very inequivalve and compressed Tellidora bb. Dorsal margins not serrate or oc- casionally so only posteriorly; usu- ally elongate Tellina B. Shell without lateral teeth a. Suborbicular to subtrigonal; subequi- lateral; moderately inflated; broad submedian concavity in right valve Apolymetis aa. Subtrigonal or elongate; posterior end produced and narrowed ; moderately compressed; sometimes inequivalve Macoma Genus Tellina Linnaeus. The present collection from the tropical eastern Pacific contains a good representa- tion of the species of Tellina which occur in that region. Many of these shells are conspic- uous because of their beautiful red color, especially when observed upon the beach. Morch long ago mentioned that red Tellinas are characteristic of tropical American waters. Key to the subgenera of Tellina. A. Posterior area with simple concentric lamellae or smooth a. Surface obliquely grooved Scissula aa. Surface not obliquely grooved b. Beaks anteriorly directed Macaliopsis bb. Beaks not anteriorly directed c. Right anterior lateral distant from the beak d. Smooth, polished Tellina s.s.5 dd. Strong concentric sculpture Tellinella 3 Dali, W. H. Synopsis of the Family Tellinidae and of the North American species. Proc. U. S. Nat. Mus., Vol. 23, [No. 1210], November, 1900, pp. 285-326, pis. 2-4. 4 Salisbury, A. E. On the Nomenclature of Tellinidae, with Descriptions of new species and some remarks on Distribution. Proc. Malacol. Soc. London. Vol. 21, Pt. 2, July, 1934, pp. 74-91, pis. 9-14. 5 Not represented in the present collection. cc. Right anterior lateral extendi close beneath or near beak e. Shell usually exceeding 25 mm. in length f. Resilium external; shell elongate g. Thick; right pos- terior lateral strong Eurytellina gg. Thin; right pos- terior lateral weak; fine reticu-ji late sculpture Tellinidella ff. Resilium internal; shell thin, high Scrobiculina ee. Shell not exceeding 25 mm. in length (usuallyi not exceeding 20 mm.) h. Sculpture' chiefly of con- centric lines of growth (ex- cept on poste- rior area) Moerella hh. Sculpture chiefly of con-, centric lamel- lae (over en- tire shell) ; tri- gonal; strong, very narrow posterior flex- ure Merisca B. Posterior area with plate-like foliations \ or posterior end with strong radial sculp- ture a. Pallial sinus free or confluent with ' pallial line for not more than one-third 1 its length b. Posterior area with plate-like fo- liations Phyllodina bb. Posterior end of shell with strong radial sculpture ; very small Elliptotellina aa. Pallial sinus confluent with pallial line for entire length Phyllodella Subgenus Tellinella Morch. Tellinella Gray, Morch, Cat. Conch. Yol- di, Fasc. 2, 1853, p. 13. [Species originally cited under Tellinella include antoni Phil- ippi, interrupta Solander, pulchella Lamarck, rostrata Linnaeus, virgata Linnaeus, and several others]. — Dali, Bartsch & Rehder, Bernice P. Bishop Mus., Bull. 153, July 25, 1938, p. 187. Type : Tellina virgata Linnaeus. Type (designated by Stoliczka, Mem. Geol. Surv. India, Palaeont. Indica, Ser. 6, Vol. 3, 1870, pp. XVII, 116) : Tellina virgata Linnaeus [Syst. Nat., ed. 10, 1758, p. 674. “Habitat in O. Indico.” Illustrated by Hanley, )| .949] Hertlein & Strong: Mollusks of Mexico and Central America 65 a Thes. Conch., Vol. 1, 1846, p. 228, pi. 63, fig. 212. Indian Ocean]. Key to the species of TellineUa. A. Ornamented by radial stripes or spots of chocolate or purple; distance separating pallial sinus from anterior adductor im- pression 5 mm. or more cumingii B. Ornamented by radial stripes or bands of golden-orange ; distance separating pal- lial sinus from anterior adductor impres- sion not exceeding 2 mm. zacae Tellina I Tellinella I cumingii Hanley. Tellina cumingii Hanley, Proc. Zool. Soc. London, September, 1844, p. 59. “Hab. Gua- comayo, Central America; in coral sand.” — Hanley, Thes. Conch., Vol. 1, 1847, p. 223, pi. 58, fig. 72. “Guacomayo and America.” — M. Smith, Panamic Mar. Shells (Tropical Photogr. Lab., Winter Park, Florida), 1944, p. 64, fig. 847. Lower California to Panama. [Not the record “Red Sea?”]. Type Locality : Guacomayo, Central Amer- ica. Range : Magdalena Bay, Lower California, to the Gulf of California and south to Gor- gona Island, Colombia. Collecting Stations : Mexico: Tangola- Tangola Bay (196-D-14, 15), 5 fathoms, crushed shell; Costa Rica; Culebra Bay; Cedro Island (213-D-4), 5 fathoms, mud; off Ballena Bay, Gulf of Nicoya (213-D-15), 40 fathoms, mud. Description: Shell elongate, narrow, rather compressed, subrostrate, somewhat biangulated and bent to the right posterior- ly; yellowish-white with radiating brown or chocolate or purplish streaks or spots; sculptured with moderately fine, close, con- centric lamellae which become coarser and elevated on the posterior portion of the shell; hinge with two cardinal teeth in each valve, the right posterior and left anterior card- inals grooved, and two equidistant laterals in each valve; pallial sinus rather wide (in young shells rounded but in the adult oblique- ly pointed at the end), projecting forward about two-thirds the distance between the two adductor impressions and for about two- thirds its length confluent with the pallial line ; interior white or pale yellow or a com- bination of the two. Large specimens of this species attain a length of 55 mm. or more. Tellina interrupta Wood6, which ranges from North Carolina to Brazil, is a similar species. Tellina strophia Dali, in the Mio- cene of Florida, also is somewhat similar to T. cumingii. a Tellina interrupta Wood, General Conch., 1815, p. 146, 2 , J}g" 3- Inhabits the Indian and American Seas.” (Also edit. 1835. I Regarding the dates of issue of this book see Pritchard & Gatliff, Proc. Roy. Soc. Victoria, £°1- 16 (N.S.) , Pt. 1, September, 1903, p. 114; Iredale, roc. Malacol. Soc. London, Vol. 15, Pts. 2 and 3, December, 1922. p. 91], Not Tellina interrupta Solander, Portland Cat., 1786. DP. 31, 72, 105. Nomen nudum. Distribution: A few specimens of this species were taken by the expedition off western Mexico and Costa Rica. The species also is known to occur in the Pleistocene of Magdalena Bay, Lower California. Tellina ITellinellal zacae Hertlein & Strong, sp. nov. Plate I, Figs. 12, 13, 17. Shell of moderate size, elongately ovate, umbos a little posterior to the center, white, with golden-orange radiating bands of vary- ing width; anterior end elliptically rounded, posterior end rather pointed (but acutely rounded at the extremity) and slightly bent toward the right, the ventral margin is broadly curved; right valve with an elevated, rounded, curved ridge radiating from the posterior side of the umbo to the posterior ventral margin and there is a correspond- ing depression in the left valve; posterior to the ridge the shell is depressed and an- terior to the ridge there is a slight sinus; the early part of the shell is ornamented with fine, rounded, concentric threads, these on the adult shell become stronger and slight- ly irregular and somewhat lamellated pos- teriorly; ribs separated by interspaces of about the same width or in some instances slightly narrower, very fine concentric threads are present in the interspaces; a short but rather stout ligament on a narrow nymph is present posterior to the umbos; hinge with two cardinals in each valve, the right posterior and the left anterior ones grooved, the other two are more slender, there are two laterals in each valve, those on the right valve are strong, those on the left low and fused with the margin; pallial sinus long, rounded at the end and extending about three-fourths the length of the shell; interior whitish, the external rays showing through the shell. Dimensions of the holo- type: length, 33.4 mm.; height, 15.2 mm.; convexity (both valves together) , 7.8 mm. ; pallial sinus extends anteriorly 25 mm. from the posterior end of the shell. Holotype (California Acad. Sci. Paleo. Type Coll.), from Station 136-D-l, Arena Bank in the Gulf of California, Lat. 23° 29' N., Long. 109° 25' W., dredged in 45 fathoms (82 meters), mud. Paratypes were dredged in the same general region at Station 136-D- 31-32, Lat. 23° 24' 30" to 23° 28' N., Long. 109° 24' to 109° 23' 30" W., in 35-42 fathoms, in sand, calcareous algae and weeds. Other specimens were dredged at Station 150-D- 12, Gorda Banks in the Gulf of California, Lat. 23° 02' N., Long. 109° 28' W., in 80-90 fathoms, sand. The shell of this new species is in general features similar to that of Tellina cumingii but it differs in several details. In the pres- ent species the posterior area on the left valve possesses a deep well-developed groove corresponding to a ridge in the opposite valve, while in T. cumingii the correspond- ing area on the left valve is somewhat flat- 66 Zoologica: New York Zoological Society [34: ! tened and dorsally bounded by a fine incised radial line. The pallial sinus of the new spe- cies is more evenly rounded at the anterior end and extends much nearer (about three- fourths the length of the shell) the anterior adductor impression than that of T. cum- ingii. Furthermore in the color pattern the beautiful radial orange stripes on a white ground attain greater width than the purple stripes which occur on a yellowish-white ground on T. cumingii. The specimens of the new species in the present collection do not attain the size of Tellina cumingii. Tellina zacae bears a resemblance to Tel- lina crassiplicata Sowerby as illustrated by Dali, Bartsch & Rehder7 from Hawaii, but is narrower anteriorly and less broadly trun- cated posteriorly. The original illustration of Tellina crassiplicata8 does not show any radial stripes. Subgenus Scrobiculina Dali. Scrobiculina Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, November, 1900, p. 290. “Type, Scrobicularia viridotincta Carpen- ter.” Schumacheria Cossman, Rev. Crit. de Pa- leozool., Vol. 6, No. 1, January, 1902, p. 52. New name for Scrobiculina Dali, not Scro- biculinus Monterosato, 1884. Type (by original designation) : Scrobi- cularia viridotinca Carpenter. Cossman proposed the name Schumacheria to replace Scrobiculina Dali, 1900, because of the prior name Scrobiculinus Montero- sato, 1884. A strict interpretation of the present In- ternational Rules of Zoological Nomencla- ture (Article 36), allows the retention of Scrobiculina Dali. Key to the species of Scrobiculina. A. Color of umbos ochraceous ochracea B. Color of umbos yellowish-green viridotincta Tellina IScroblculinal ochracea Carpenter. Tellina ( Peronaeoderma ) ochracea Car- penter, Ann. & Mag. Nat. Hist., Ser. 3, Vol. 13, April, 1864, p. 312. Cape St. Lucas. Re- print in Smithson. Miscell. Coll., No. 252, 1872, p. 210. Tellina ( Scrobiculina ) ochracea Carpen- ter, Dali, Proc. U. S. Nat. Mus., Vol. 23, 1900, p. 302. “Cape St. Lucas to the Gulf of Cali- fornia.” Type Locality : Cape San Lucas, Lower California. Range-. Cape San Lucas to the Gulf of California. Collecting Station: Mexico: Arena Bank, 7 Tellina crassiplicata Sowerby, Dali, Bartsch & Rehder, Bernice P. Bishop Mus., Bull. 153, July 25, 1938, p. 187, pi. 48, figs. 5-8. Various localities in Hawaii cited, also Midway Island and Ocean Island. 8 Tellina crassiplicata Sowerby, Conch. Icon., Vol. 17, Tellina, April, 1869, species 332, pi. 56, figs. 332a, 332b. “Hab. Sandwich Islands.” Gulf of California (136-D-6), 45 fathoms ; mud, Area conglomerate. Description: Two somewhat worn ant broken specimens in the present collectioi answer to the description of Tellina ochracea They are colored a light sulphurous yellow strongest near the beaks and fading gradu ally toward the margins. Dali (1900), statec that the species was very similar to Telliru viridotincta Carpenter, differing only ii color. From the ranges given by Dali it woulc seem that Tellina viridotincta is a mort southern shell while T. ochracea is a specie^ of the Gulf of California. The green tip of the beak of T. viridotincta may not be a con- stant character because Stearns and Pilsbrj & Lowe have recorded it well within tht range of T. ochracea. If the color differences do not prove to be constant, viridotincta being the older name, should take precedence Distribution: A few specimens referred to this species were dredged by the expedi- tion on Arena Bank in the Gulf of California in 45 fathoms. Tellina IScroblculinal viridotincta Carpenter. ? Scrobicularia virido-tincta Carpenter, Proc. Zool. Soc. London, November 11, 1856, p. 160. “Hab. in Sinu Panamensi, una cum. ?S. producta; legit T. Bridges. Sp. un. in Mus. Cuming.” Macoma viriditincta Carpenter, Stearns) Proc. U. S. Nat. Mus., Vol. 17, 1894, p. 156. La Paz; various localities in the Gulf of California. Tellina ( Scrobiculina ) viridotincta Car- penter, Dali, Proc. U. S. Nat. Mus., Vol. 23, i 1900, p. 302. Lower California to Panama., Tellina viridotincta Carpenter, Pilsbry & Lowe, Proc. Acad. Nat. Sci. Philadelphia, Vol. 84, 1932, p. 133. Espiritu Santo Island; La Paz. Type Locality: Bay of Panama. Range: Gulf of California to Panama. Collecting Station: Mexico: Port Guatulco (195-D-6), 3 fathoms, sand, algae, crushed shell. Description : Shell large, high, oval, beaks posterior to the center, anterior end ellip- tically rounded, ventral margin broadly rounded, posterior end tapering, subtrunc- ately rounded and bluntly pointed at the ex- tremity; white, umbos greenish-yellow; pos- terior area set off by a low umbonal ridge and anterior to this there is a low, broad, radial depression; valves ornamented with fine, rather regular, concentric threads and striae, which are more crowded and irregular on the posterior area on which, especially on the right valve, there is usually one and some- times more fine radial ridges ; the concentric sculpture is crossed by very fine, faint, ra- diating striae over much of the disk; the pallial sinus extends about three-fourths the length of the shell, highest somewhat an- terior to the posterior adductor impression then sloping downward anteriorly where, at about 5-8 mm. from the pallial line, it is sub- 949] Hertlein & Strong: Mollusks of Mexico and Central America 67 ngularly or bluntly rounded and then joins he pallial line; a large sunken ligament is ituated upon a nymph; hinge with two car- inals in each valve, the right posterior and jft anterior ones grooved ; right valve with wo well-developed laterals, left valve with . weak anterior lateral and the posterior ateral very slight or obsolete ; interior white Hth blotches of greenish-yellow, especially oward the dorsal portion of the shell. A specimen from the Gulf of California n the Henry Hemphill collection of the Cali- 'ornia Academy of Sciences, measures: ength, 62 mm.; height, 44 mm.; convexity both valves together), 14.5 mm.; pallial ;inus extends anteriorly 46 mm. from the )osterior end of the shell. The species attains i greater size than this. As mentioned by Dali, about the only dif- ference between Tellina viridotincta and T. )chracea is in color. Distribution: One rather worn pair of calves of this species was dredged in 3 ?athoms at Port Guatulco, Mexico. Subgenus Moerella Fischer. Moerella Fischer, Man. de Conchyl., Fasc. LI, June 15, 1887, p. 1147. Sole species, rellina donacina Linnaeus. Type (by monotypy) : Tellina donacina Ldnnaeus. Recent, seas of Europe. Illustrated >y Bucquoy, Dautzenberg & Dollfus, Moll. Mar. Roussillon, Vol. 2, Fasc. 25, March, L898, p. 648, pi. 91, figs. 13, 14, and vars. L5-19. Mediterranean. Also other localities :ited. In the present paper nine species and sub- species have been referred to the subgenus Xloerella. Some of these, in earlier publica- ;ions, have been referred to Angulus Megerle /on Miihlfeld. The type of Angulus desig- lated by Gray, 1847, is Tellina lanceolata Linnaeus, a species in which the hinge is said to possess a right anterior lateral but acking all other laterals. Salisbury, 1934, stated that only two Recent species, T. lan- :eolata, the type, and T. armata Sowerby, were referable to Angulus. However, in the explanation to his plates five species are re- ferred to Angulus. It is unfortunate that this well known supraspecific group name should be applicable to so few species. All the species in the present paper which have been referred to Moerella, although varying somewhat in shape, possess a right posterior lateral tooth of varying strength. This lateral occurs just below a socket which is present just below the ventral end of the nymph upon which the ligament is situated. The results of our studies which have led us to place the following group of species under the subgenus Moerella, are in agree- ment with the conclusions reached by Gard- ner9 with regard to the east American Miocene and Pliocene species formerly re- ferred to Angulus. 9 Gardner, J., U. S. Geol. Surv., Prof. Paper Ht-E, 1928, p. 195; U. S. Geol. Surv., Prof. Paper 199-A, 1943, p. 94. Key to the species of Moerella. A. Posterior end triangular or bluntly pointed a. Posterior end the longer, tapering, pointed; white, yellowish or pinkish suffusa aa. Anterior end the longer b. Posterior dorsal margin sinuous; ivory white tabogensis bb. Posterior dorsal margin straight or slightly curved c. Ventralmargin strongly curved ; posterior end blunt; white paziana cc. Ventral margin very gently curved d. Length more than twice the height; white, occasionally yellowish amianta dd. Length less than twice the height e. Area anterior to poste- rior umbonal angulation strongly depressed; usu- ally pink on dorsal mar- gins erythronotus ee. Area anterior to poste- rior umbonal angulation not depressed or only faintly so; red or pink zoned with white f. Umbos moderately inflated; posterior end very short macneilii ff. Umbos rather com- pressed; posterior end more attenuated ; rose red felix B. Posterior end broad, obliquely truncated a. Pale rose color arenica aa. White ; posterior end more abruptly truncated recurvata Tellina l Moerella I amianta Dali. Tellina ( Moerella ) amianta Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, November, 1900, pp. 303, 317, pi. 3, fig. 12. “Dredged in 14 fathoms, sand, off Cape Tepoca, Lower California, near the head of the Gulf, by the U. S. Fish Commission at station 3019.” Type Locality: Off Cape Tepoca, Lower California, near the head of the Gulf of California, in 14 fathoms, sand. Range: Gulf of California to Colombia. Collecting Station: Mexico: Santa Inez Bay, Gulf of California (145-D-l, 3), 4-13 fathoms, sand; Nicaragua: Corinto (200-D- 19), 12-13 fathoms, mangrove leaves, sand, also on shore in beach drift; Costa Rica: Port Parker (203-D-l, 3), 12-15 fathoms, sandy mud, shelly mud, crushed shell. Description: Shell small, elongated, ante- riorly produced and rounded, the posterior end shorter, obliquely truncated and rather 68 Zoologica: New York Zoological Society pointed at the extremity; sculptured with fine, close, concentric thi-eads which toward the posterior angulation become somewhat irregular, sharper, and in the adult shell two or three ribs coalesce to form low raised lamellae on the posterior area; hinge with two cardinals in each valve, the right pos- terior and left anterior ones grooved, right valve with a strong fairly close anterior lateral and a small posterior lateral; pallial sinus extends near to, but does not touch, the anterior adductor impression, confluent with the pallial line below; color white or partly salmon yellow. A large right valve dredged in Santa Inez Bay in the Gulf of California in 4-13 fath- oms, measures: length, 14 mm.; height, 6.8 mm.; convexity (one valve), 2.2 mm. The small size, elongate form which is much produced anteriorly, and fine, close, concentric sculpture are characteristic fea- tures of this species. Specimens dredged by the expedition in Santa Inez Bay in the Gulf of California are typical of the species. Specimens collected at Corinto, Nicaragua, in beach drift, and dredged in 12-13 fathoms agree in general characters with T. amianta except that they are thinner. Many small specimens dredged in 12-13 fathoms off Port Parker, Costa Rica, in which the hinge, concentric sculpture and other general characters agree well with T. amianta, appear to represent the young of that species. Some are less elongate in proportion to the height as compared to typical T. amianta but this appears to be somewhat variable among these young shells. Many small specimens in the collections of the California Academy of Sciences, dredged in the Gulf of California, are similar to T. amianta. Some are white, some white with pink radial streaks or dots, others yellowish- white. These bear a similarity to one of the illustrations given by Sowerby10 (pi. 47, fig. 278d) under the name of Tellina silicula Deshayes. That species was originally de- scribed by Deshayes11 with the type locality “W. Columbia.” Salisbury12 stated that the type specimen of T. silicula Deshayes is referable to Tellina rhomboides Quoy & Gaimard, a species which occurs in the wes- tern Pacific region. According to Iredale13 Tellina clathrata Deshayes is the correct name for that species. Sowerby’s figures 278a, b, c, appear to be referable to it but it seems possible that his figure 278d might be referable to a young T. amianta. Distribution : This species was dredged in Santa Inez Bay, Gulf of California, in 4-13 fathoms, at Corinto, Nicaragua, in 12-13 10 Sowerby, G. B., Conch. Icon., Vol. 17, Tellina, October, 1868, species 278, pi. 47, fig. 278d. "Hah. W. Columbia.” [Not figs. 278a, b, c.]. 11 Tellina silicula Deshayes, Proc. Zool. Soc. London for 1854, (issued May 16, 1855), p. 363. "Hab. Columbia. Coll. Cuming.” 12 Salisbury, A. E., Proc. Malacol. Soc. London, Vol. 21, Pt. 2, 1934, p. 89. See also Lamy, E., Bull. Mus. Nat. Hist. Nat. (Paris), Vol. 24, No. 2, 1918, p. 116. 13 Iredale, T., Mem. Queensland Mus., Vol. 9, Pt. 3, June 29, 1929, p. 266. [34: fathoms, also in beach drift on shore, and a | ,>rred Port Parker, Costa Rica, in 12-15 fathoms I. eT It also has been recorded as occurring in th Pleistocene at Magdalena Bay, Lower Cali | fornia. Tellina IMoerellal arenica Hertlein & Strong! tie The sp. nov. Plate I, Figs. 5, 11. Tellina carpenteri Dali, Packard, Univ\ Calif. Publ. Zool., Vol. 14, No. 2, Septembe: 12, 1918, p. 276, pi. 25, figs. 10a, 10b. , “in 68 fathoms just south of the Faralloi Islands.” —I. S. Oldroyd, Publ. Puget Sount Biol. Sta., Vol. 4, 1924, pi. 41, figs. 10a, 10b (Copies of Packard’s figures). —I. S. Old royd, Stanford Univ. Publ. Univ. Ser. Geol Sci., Vol. 1, 1924, pi. 44, figs. 10a, 10b (Copies of Packard’s figures). Not Tellina carpenteri Dali, 1903. Description : Shell of moderate size, trans- versely ovate, thin, compressed, the anterioi end slightly the longer, color creamy white salmon pink and rose in concentric zones of irregular width ; anterior dorsal margin sloping gently convexly from the beaks, an- terior end rounded, ventral margin broadly| ! rounded, posterior dorsal margin gently slop- ing, slightly concave, posterior end obliquely truncated; posterior area defined only by a rounded umbonal angulation ; sculptured with concentric lines of growth and by some- what irregularly spaced fine, shallow, con- centric grooves which on the posterior area become deeper and give rise to sublamellate sculpture; hinge of right valve with two cardinal teeth, the posterior one bifid, an anterior lateral is fairly close to the cardinals and there is a distant posterior lateral below a socket; left valve (paratype) with a bifid anterior cardinal and traces of a posterior cardinal (some valves with a thin posterior cardinal lamella) ; pallial sinus highly tri- gonal back of the beak then sloping some- what irregularly to a position below and well separated from the anterior adductor im- pression, the end rounded then bending pos- teriorly for a very short distance where it becomes confluent with the pallial line; in terior white and pink. Dimensions of the holotype : length, 24.5 mm. ; height, 15 mm. ; convexity (one valve) , 2.5 mm. ; pallial sinus extends anteriorly 19 mm. from the posterior end of the valve. Holotype, a right valve (Calif. Acad. Sci Paleo. Type Coll.), dredged at Station 136- D-20 in Lat. 23° 30’ N., Long. 109° 26’ W., in 43 fathoms, mud, on Arena Bank, at the south end of the Gulf of California. One ad ditional specimen, a paratype, was dredged at the same locality. Three small single valves were dredged in the channel east of Cedros Island at Station 126-D-17, in 40 fathoms. Paratypes were dredged by the Templeton Crocker Expedition in 1932, near Puntarenas, Costa Rica. Range: Farallon Islands, California, to Panama. 1949] Hertlein & Strong: MollusJcs o) Mexico and Central America 69 This beautiful species has often been re- ferred to Tellina carpenteri Dali14, a distinct species, which has a smaller shell and a gen- erally more northern distribution. Dali15 cit- ed T. carpenteri as occurring in the Gulf of Panama at a depth of 182 fathoms. It seems probable that that record may be referable to the present species. The form illustrated under the name of T. carpenteri by Packard, 1918, which was dredged in 68 fathoms just south of the Farallon Islands, the illustrations of which were reproduced by I. S. Oldroyd, appears to be identical with the present specimens. This new species differs from T. carpenteri in the much greater size, irregular concen- tric zones of color and in the stronger con- centric grooves. It differs from Tellina re- curvata Hertlein & Strong [_—T. recurva Dali, 1900, not of Deshayes, 1844], in the much more obliquely truncated posterior end and in the pink coloration. The shell of Tel- lina arenica differs from that of T. tabo- gensis in the larger size, pink color and in that the posterior dorsal margin is slightly concave or nearly straight rather than flexu- ous. Tellina IMoerellal erythronotus Pilsbry & Lowe. Tellina ( Angulus ) erythronotus Pilsbry & Lowe, Proc. Acad. Nat. Sci. Philadelphia, Vol. 84, May 21, 1932, p. 94, pi. 12, fig. 7. “Panama, east of the city” (type) . Also from Montijo Bay, Panama. Type Locality : Panama, east of the city. Range : Magdalena Bay, Lower California, to the Bay of Panama. Collecting Stations: El Salvador: Mean- guera Island, Gulf of Fonseca (199-D-l), 16 fathoms, sand, mud, crushed shell ; La Union, Gulf of Fonseca (199-D-8-16, 19-25), 3-16 fathoms, mud, mangrove leaves ; Nicaragua: Monypenny Point, Gulf of Fonseca (199-D- 2-6), 4-7 fathoms, mud. Description: Shell donaciform, moderately thin, somewhat compressed, creamy white often tinged with pink around the margins, opalescent; anteriorly elongated and round- ed, ventrally gently rounded and posteriorly slightly embayed due to the presence of a shallow depressed area which occurs anterior 14 Angulus variegatus Carpenter, Rept. Brit. Assoc. Adv. Sci. for 1863 (issued August, 1864) , p. 611. “Mont., Cat. Is., 20-60 fm. ; rare (Neeah Bay, Swan).” Also pp. 627, 639. Reprint in Smithson. Miscell. Coll., No. 252, 1872, pp. 97, 113. 125. —Carpenter, Ann. & Mag. Nat. Hist., Ser. 3, Vol. 14, December,1864, p. 423. “Hab. Neeah Bay (Swan) ; Monterey and Catalina Island, 20-60 fathoms, rare (Cooper).” Reprint in Smithson. MisceU. Coll., No. 252, 1872, p. 235. Not Tellina variegata Gmelin, Syst. Nat., ed. 13, 1790, p. 3237. Tellina (Angulus) carpenteri Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, November, 1903, pp. 303, 320. New name for Angulus variegatus Carpenter, 1864, not Tellina variegata Gmelin, 1790. “Strait of Juan de Fuca to Lower California.” Illustrated by I. S. Oldroyd, Publ. Puget Sound Biol. Sta., Vol. 4, 1924, p. 51, pi. 10, fig. 4. —I. S. Oldroyd, Stanford Vniv. Publ. Univ. Ser. Geol. Sci., Vol. 1, 1924, p. 166, pi. 29, fig. 2. San Pedro, California. 15 Tellina ( Angulus ) carpenteri Dali, Dali, Bull. Mus. Comp. Zool., Vol. 43. No. 6, October, 1908, p. 421. “U. S. S. ‘Albatross’, station 3355, Gulf of Panama, in 182 fathoms, mud, bottom temperature, 54°. 1 F. U. S. N. Mus. 122, 934.” to the posterior umbonal ridge; posteriorly triangular and somewhat bluntly pointed, with nearly straight dorsal margin; pallial sinus long not quite touching the anterior adductor impression, along the base wholly confluent with the pallial line; hinge of right valve with a grooved anterior and a bifid posterior cardinal, the anterior lateral strong and close to the cardinals, the posterior lat- eral farther removed but not distant; left valve with a grooved anterior and a thin posterior cardinal lamina, anterior lateral represented by a triangular projection of the margin, a similar posterior projection faintly developed or absent. One of the largest specimens in the present collection measures : length, 25 mm. ; height, 14.5 mm. ; convexity (both valves together) , 7 mm. The specimens in the present collection show all the characters of Tellina erythrono- tus mentioned by Pilsbry & Lowe. The shell of this species appears to be very similar to that of Tellina hiberna Han- ley10 which also has been cited as occurring at Panama. However, the shell of T. ery- thronotus appears to be somewhat more elongate in comparison to the figures of T. hiberna given by Hanley and Salisbury. The posterior dorsal margin of Hanley’s species is said to be first convex then concave, but there is no marked convexity on the margin of T. erythronotus. Furthermore there is nothing in Hanley’s description regarding pink or prismatic colors such as often can be observed on T. erythronotus. Tellina puellula Salisbury17 (=T. puella C. B. Adams, not Hanley) , is another closely re- lated species. According to Pilsbry & Lowe it is larger, relatively higher and thicker than T. erythronotus. Tellina hiberna was described in 1844 and it seems possible that either T. erythronotus or T. puellula, both described later, may be referable to it, but we have not the type spec- imens available to enable us to make any def- inite decision on this question. Tellina pananiensis Philippi18, from Pana- ma, was described as white with the umbos red both exteriorly and interiorly but it was said to lack lateral teeth. Distribution: Tellina erythronotus was dredged quite abundantly in the Gulf of Fon- seca at a depth of 3 to 16 fathoms on a bottom of mud and mangrove leaves. The present 16 Tellina hiberna Hanley, Proc. Zool. Soc. London, De- cember, 1844, p. 148. “Hab. Panama and Bay of Quayaquil ; six to eleven fathoms, in sandy mud : Cuming.” —Hanley, Thes. Conch., Vol. 1, 1846, p. 282, pi. 57, fig. 53. Original localities cited. Tellina ( Angulus ) hiberna Hanley, Salisbury, Proc. Malacol. Soc. London, Vol. 21, Pt. 2, 1934, p. 91, pi. 13, figs. 7, 8, 9. [Illustrations of holotype and paratypes]. 17 Tellina puellula Salisbury, Proc. Malacol. Soc. London, Vol. 21, Pt. 2, July, 1934, p. 86. A new name for Tellina puella C. B. Adams, Ann. & Lyceum Nat. Hist. New York, Vol. 5, July 1852, pp. 507, 646 (separate, pp. 283, 322). “Panama.” Not Tellina puella Hanley, Proc. Zool. Soc. London, February, 1845, p. 165. “Hab. Senegal. Cuming, Metcalfe.” 18 Te'linu yanamensis Philippi, Zeit. f. Malakozool.. Jahrg. 5, No. 11, 1848, p. 176. "Ad Panama legit frater E. B. Philippi.” 70 Zoologica: New York Zoological Society [34:9 records of occurence furnish a long extension north of the known range of the species. The noi’thernmost occurrence of this species known to us is that based upon a specimen in the collections of the California Academy of Sciences which was collected by C. R. Orcutt at Magdalena Bay, Lower California. Tellina IMoerellal felix Hanley. Plate I, Fig. 1. Tellina felix Hanley, Proc. Zool. Soc. Lon- don, September, 1844, p. 71. “Hab. Panama; sandy mud, six to ten fathoms.” — Hanley, Thes. Conch., Vol. 1, 1846, p. 281, pi. 57, fig. 52. “Panama.” Type Locality : Panama, in 6 to 10 fathoms, sandy mud. Range: Mazatlan, Mexico, to Zorritos, Peru. Collecting Station: Nicaragua: Potosi and Monypenny Point, Gulf of Fonseca; Corinto (200-D-8, 9, 17, 19), 6-13 fathoms, sand, mangrove leaves, also beach drift. Description: Shell small, elongate, anteri- or end much the longer, rounded, the posteri- or end very short and obtusely obliquely sub- truncated, anterior dorsal and ventral mar- gins nearly parallel, exterior and interior of a glossy rose red color; a posterior area is set off by an umbonal angulation; valves ornamented with fine, regular, concentric striae; hinge of right valve with two cardi- nals, the posterior one grooved, a strong, high, elongate anterior lateral extends al- most to the beak and a small posterior lateral is present just beyond and below the posteri- or end of the ligamentary area; left valve with a well-developed anterior cardinal, a thin posterior cardinal lamella is almost fused to the posterior margin and a pointed projection of the nymph represents an an- terior lateral; pallial sinus not quite reach- ing the anterior adductor impression and along the base, except for a short distance, it is confluent with the pallial line. A specimen from the Gulf of Fonseca in the Henry Hemphill collection in the Cali- fornia Academy of Sciences measures ap- proximately : length 17.2 mm. ; height, 9.4 mm.; convexity (both valves together), 4.3 mm. The shell of this little species is character- ized by the beautiful glossy rose red color, weakly inflated valves, nearly parallel anteri- or dorsal and ventral margins, strong right anterior lateral, and by the short, obliquely subtruncated posterior end. The shorter posterior end, more gently sloping anterior dorsal margin and deep red color are features which serve to separate Tellina felix from T. tabogensis Salisbury. The longer, stronger, right anterior lateral and more steeply sloping posterior dorsal area just below the beaks are features sepa- rating T. felix from T. carpenteri Dali. Carpenter19 mentioned that Tellina puella 19 Carpenter, P. P., Proc. Zool. Soc. London, 1863, p. 366. Reprint in Smithson. Miscell. Coll., No. 262, 1872, p. 202. C. B. Adams20 [ —T . puellula Salisbury] is “not unlike T. felix.” The remarks of Pilsbry & Lowe seem to indicate that the species de- ; scribed by Adams possesses a higher, heavier shell than T. felix. Distribution: Specimens of this species were dredged off Nicaragua in 6-13 fathoms and also were taken in the beach drift. It has been recorded with doubt, as occurring in the Miocene21 of Peru, and definitely in the Pliocene of Ecuador. This species has been cited as occuring in the Red Sea but, as men- tioned by Lamy22, it does not occur in that region. Tellina IMoerellal macneilii Dali. Tellina ( Angulus ) macneilii Dali, Proc. U . S. Nat. Mus., Vol. 23, No. 1210, November, 1900, pp. 303, 318, pi. 3, fig. 7. “Obtained at Guaymas, Mexico, by W. H. Dali.” Type Locality: Guaymas, Mexico. Range : Guaymas, Mexico, to the Gulf of Nicoya, Costa Rica. Collecting Station: Costa Rica: Cedro Is-, land to off Ballena Bay, Gulf of Nicoya (213- ' D-11-17), 35-40 fathoms, mud. Description : Shell small, solid, inequilat- eral, the anterior end longer, rounded, the posterior end quite short, depressed, bluntly pointed ; color deep rosy, slightly zoned, and paler toward the basal margin ; surface closely, sharply concentrically striated, the posterior dorsal area feebly imbricate, with a little obscure radial striulation ; valves moderately full, flatfish toward the middle of the disk; hinge strong, normal; internal ray obscure; pallial sinus long, nearly reach- I ing the anterior adductor scar, wholly con- j fluent below. Lon. 12.5, alt. 7.6, diam. 3.5 mm. (Original description). One pair and two single valves dredged in the Gulf of Nicoya agree well with Dali’s de- scription and illustration. The largest speci- men measures approximately: length 11.5 mm. ; height, 7.3 mm. ; convexity (both valves together) , 4.2 mm. These are identical with specimens identified as T. macneilii in the Lowe collection in the San Diego Society of Natural History. The hinge of this species is similar to that of T. felix Hanley. Compared to T. felix, the valves of T. mac- neilii are more inflated in proportion to their size, less elongate, the posterior end is more abruptly truncated and the anterior dorsal margin slopes more steeply. These same char- acters (except the slope of the anterior dor- sal margin) as well as the pink color likewise serve to separate this species from T. tabo- gensis. Tellina guaymasensis Pilsbry & Lowe23 is 20 See footnote No. 17, p. 69. 21 Tellina (Eury tellina) cf. felix Hanley, Olsson, Bull. Amer. Paleo., Vol. 19, No. 68, June 30, 1932, p. 123, pi. 14, fig. 8. “Tumbez formation, Que Tucillal at Zorritos." Peru. Miocene. 22 Lamy, E., Bull. Mus. Nat. Hist. Nat. (Paris), Vol. 24, No. 2, 1918, p. 119 (footnote). 23 Tellina ( Angulus ) guaymasensis Pilsbry & Lowe, Proc. Acad. Nat. Sci. Philadelphia, Vol. 84, May 21, 1932, p. 94, pi. 16, fig. 7. "Guaymas.” Mexico. 1949] Hertlein & Strong: Mollusks of Mexico and Central America 71 quite similar to T. macneilii but the shell appears to be a little more depressed medially and is in greater part white with streaks of pink. The present specimens agree well with the brief description of Tellina deshayesii Car- penter24 from the Bay of Panama. It was said to resemble Tellina similis Sowerby but much more inequilateral. However the spe- cific name proposed by Carpenter is unten- able in any case because of the prior use of that combination of names, Tellina deshaye- sii, by Hanley25. Distribution : Specimens of Tellina mac- neilii were taken by the expedition in the Gulf of Nicoya in 35-40 fathoms. This record represents an extension south of the known range of this species. Tellina IMoerellal paziana Dali. Tellina ( Moerella ) paziana Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, November, 1900, pp. 303, 318, pi. 3, fig. 8. “Dredged in 26 V2 fathoms, near La Paz, Lower Califor- nia, by the U. S. Fish Commission, at station 2823.” Type Locality : Near La Paz, Lower Cali- fornia, Mexico, in 26*4 fathoms. Range : Gulf of California to Cedro Island, Costa Rica. Collecting Stations : Mexico : Port Guatulco (195-D-2, 7, 10, 11), 3-5 fathoms, sand, gr. sand, crushed shell, dead coral ; Tangola-Tan- gola Bay (196-D-6, 7, 13, 14, 15), 5-12.8 fathoms, sand, crushed shell; Nicaragua: Corinto (200-D-8, 9, 19,) , 6-13 fathoms, man- grove leaves, also on the beach ; Costa Rica : Cedro Island, Gulf of Nicoya (213-D-1-10) , 4-10 fathoms, mud, sand, crushed shell. Description : Shell small, thin, white, con- vex, the anterior end slightly longer, round- ed, the posterior end bluntly pointed ; surface finely concentrically sculptured by the incre- mental lines, covered with a very delicate de- hiscent pale straw-colored epidermis; hinge well-developed, a minute but distinct anteri- or left lateral present; interior polished, only about half the lower portion of the pallial sin- us confluent, the anterior part not reaching the adductor. Lon. 10.2, alt. 7, diam. 3.5 mm. (Original description). The shell of this species is usually small, about 10-15 mm. in length. A large right valve in the present collection from Corinto, Nicaragua, measures approximately: length, 17.5 mm.; height 14 mm., convexity (one valve) , 3.3 mm. ; pallial sinus extends for- ward 12.8 mm. from the posterior margin. The pallial sinus in this species ascends to a high rounded point slightly posterior to a line vertical with the beaks, then descends obliquely. The end is rounded and well sepa- rated from the anterior adductor impression. 24 Tellina deshayesii Carpenter, Proc. Zool. Soc. London , November 11, 1856, p. 160. “Hab. in Sinu Panamensi ; legit T. Bridges. Sp. un. in Mus. Cuming.** 25 Tellina deshayesii Hanley, Proc . Zool. Soc . London, December, 1844, p. 148. “Hab. Red Sea? Mus. Cuming, Deshayes.” Tellina paziana somewhat resembles T. meropsis Dali. It differs from that species in the more elongate outline, the anterior end is longer in proportion to the length, the pos- terior end is much more bluntly pointed and the pallial sinus is separated from the an- terior adductor impression by a much wider space. According to Dali, “This differs from the young of Scrobiculina viridotincta Car- penter, which in outline it resembles, by be- ing less polished, more inflated, and without the deep-set resilium.” Distribution: This species was taken by the expedition from off western Mexico to the Gulf of Nicoya, Costa Rica, in 3-13 fathoms, also on the beach. This is an extension south of the known range of this species. Tellina IMoerellal recurvata Hertlein & Strong, sp. nov. Plate I, Figs. 2, 3, 4, 8. Tellina ( Angulus ) recurva Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, November, 1900, pp. 304, 320, pi. 3, fig. 4. “Dredged near the head of the Gulf of California in 24 fath- oms, mud, off Point Fermin.” Not Tellina recurva Deshayes, Proc. Zool. Soc. London for 1854 (issued May 16, 1855), p. 361. “Hab. Australia.”— Hedley, Proc. Linn. Soc. New South Wales, Vol. 38, Pt. 2, 1913, p. 272. Hedley stated: “I failed to find this unfigured Australian species in the Brit- ish Museum. It is recommeded that the name be treated as lost and unrecognizable.” Description : Shell small, translucent white, polished, rather compressed, beaks very low, the anterior end the longer; anteri- or dorsal margin gently curved, ventral mar- gin gently rounded, posterior dorsal margin depressed below the beak, sloping gently along the ligamentary area (about 2 mm. on the type), the end obliquely sloping and roundly truncated ; a weak posterior umbonal angulation present; surface of valves with concentric, chiefly incremental sculpture, the posterior area with low lamellae ; hinge plate arched anterior to the beaks; right valve with two cardinals, the posterior one grooved, anterior lateral strong and fairly close to the cardinals, a posterior lateral oc- curs below a socket at about the end of the ligamental area; left valve with two car- dinals, the posterior one sloping posteriorly, a faint projection of the margin represents an anterior lateral; pallial sinus subtri- angular, extending forward about three- fourths the length of the shell but well sepa- rated from the anterior adductor impres- sion, along the base confluent with the pal- lial line. Dimensions of the type: length, 12 mm.; height, 7.5 mm.; convexity (both valves together), 2.9 mm. Holotype (Calif. Acad. Sci. Paleo. Type Coll.), from Loc. 23802 (C.A.S.), San Luis Gonzaga Bay, Gulf of California. Paratypes from the same locality. Dredged by the ex- pedition at the following stations: Mexico: 4 miles SSW. of Maldanado Point (192-D- 72 Zoologica: New York Zoological Society [34: 9 1), 26 fathoms, mud; Port Guatulco ( 195- D-20), 23 fathoms, mud; Costa Rica: Port Parker (203-D-l), 15 fathoms, sandy mud, crushed shell. Range: Point Firmin, Lower California, near the head of the Gulf of California, to Port Parker, Costa Rica. Some of the shells of this species are iri- descent. Dali pointed out that the shell of this species somewhat resembles that of young Macoma yoldif ormis but is more blunt pos- teriorly and the hinge of course is different. The more abruptly truncated posterior end and white color are features separating the present species from T. arenica. Compared to T. tabogensis the outline of T. recurvata is less elongate, the anterior dorsal margin is more arcuate and the posterior dorsal margin is not flexuous. Compared to T. buttoni Dali, the shell of the present species is less attenuated both anteriorly and pos- teriorly, the posterior dorsal margin just below the beaks slopes more gently and it lacks the strong anterior ray internally which is so conspicuous in Dali’s species. Distribution: A few specimens of this species were dredged by the expedition off western Mexico in 23-26 fathoms and off Port Parker, Costa Rica, in 15 fathoms. The present record of the occurrence of this shell off Costa Rica is an extension south of the known range of the species. Tellina IMoerellal suffusa Dali. Tellina ( Angulus ) suffusa Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, November, 1900, pp. 303, 319, pi. 3, fig. 10. “Collected at San Ignacio Lagoon, Lower California, by Henry Hemphill.” Type Locality: San Ignacio Lagoon, west coast of Lower California. Range: San Ignacio Lagoon, Lower Cali- fornia, to Corinto, Nicaragua. Collecting Station: Nicaragua: Corinto, beach. Description: Shell cuneate, very thin, con- vex, blunt in front, pointed behind, the pos- terior end slightly longer, pinkish, yellowish, or translucent white in color ; surface rather strongly, closely, and irregularly concen- trically striate, with an unusually large and wide lunular impression, but no escutcheon to speak of ; hinge normal, delicate ; interior polished; the pallial sinus high, well sepa- rated from the anterior adductor, though there seems to be no trace of a ray in the specimens examined. Lon. 13.5, alt. 9.2, diam. 4.7 mm. (Original description). The unusually large lunular area, short, blunt anterior end and pointed posterior end are features characteristic of this little species. Several single valves from Corinto, Nica- ragua, the largest measuring 11 mm. in length, agree well with Dali’s description and illustration of Tellina suffusa. They like- wise appear to be identical with specimens of that species from Magdalena Bay in the Hemphill collection of the California Acad- emy of Sciences. Tellina pumila Hanley,26 described from Chile, is somewhat similar in outline but the posterior end is narrower. Distribution: Specimens of this species were taken by the expedition only in beach drift at Corinto, Nicaragua. This is a con- siderable extension south of the known range of this species. Tellina IMoerellal tabogensis Salisbury. ' Tellina ( Angulus ) panamensis Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, November, 1900, p. 319, pi. 3, fig. 3. “Types.— No. 108557, U.S.N.M., dredged in 30 fathoms in Panama Bay by the U. S. Fish Commission, at station 2799.” Not Tellina panamensis Philippi, Zeit. f. Malakozool., Jahrg. 5, No. 11, 1848, p. 175. “Ad Panama legit frater E. B. Philippi.” Tellina tabogensis Salisbury, Proc. Mala- col. Soc. London, Vol. 21, Pt. 2, July, 1934, p. 86. A new name for Tellina ( Angulus ) panamensis Dali, 1900, not Tellina panamen- sis Philippi, 1848. Type Locality: Panama Bay, in 30 fath- oms. Range : Gulf of California to Santa Elena Bay, Ecuador. Collecting Stations: El Salvador: Mean- guera Island, Gulf of Fonseca (199-D-l), 16 fathoms, sand, mud, crushed shell ; La Union (199-D-13), 6 fathoms, mud; Nicaragua: Corinto (200-D-19), 12-13 fathoms, man- grove leaves ; Costa Rica : Cedro Island, Gulf of Nicoya (213-D-1-10) , 4-10 fathoms, mud, sand, crushed shell. Shell small, thin, ivory white, polished, rather compressed, flexuous behind, the an- terior end much the longer, produced and rounded, posterior end with the ligament rather deeply inset, margin obliquely de- scending to a rather blunt point; surface smooth or marked only by incremental lines, except near the basal margin, where there are a few incised lines with wider inter- spaces, not quite in harmony with the lines of growth; posterior dorsal area minutely concentrically rippled; hinge normal, deli- cate ; pallial sinus large, not reaching the ad- ductor, mostly confluent below; the elevated ray absent or obsolete. Lon. 9, alt. 5.25, diam. 2.5 mm. (Original description of Tellina panamensis Dali). Fresh specimens exhibit on the surface a lovely iridescent glow (Dali). A large series of specimens of this species dredged off El Salvador by the expedition have been compared with a series dredged at Acapulco, Mexico, in the H. N. Lowe Col- lection in the Museum of the San Diego So- ciety of Natural History and with specimens 26 Tellina pumila Hanley, Proc. Zool. Soc. London, Sep- tember, 1844, p. 69. “Hab. Valparaiso: sandy mud, from seven to thirty fathoms.” —Hanley, Thes. Conch., Vol. 1, 1846, p. 279, pi. 57, fig. 41. “Valparaiso.” Tellina (Angulus) pumila Hanley, Salisbury, Proc. Malacol. Soc. London, Vol. 21, Pt. 2, July, 1934, p. 91, pi. 13, figs. 3 and 4. [Illustrations of holotype and para- type]. 1949] Hertlein & Strong: Mollusks of Mexico and Central America 73 in the California Academy of Sciences. It appears from this study that Dali’s type, 9 mm. in length, is a juvenile shell and that the species reaches a much larger size. A large right valve in the present collection dredged off El Salvador, measures : length, 18.3 mm. ; height, 10 mm.; convexity (one valve), 2.6 mm. The most striking features of this species are the ivory white color, iridescence in fresh shells and the flexuous posterior dorsal mar- gin. This flexuosity is caused by the convex- ity of an area on the posterior dorsal margin just postei'ior to the ligamentary area. Some specimens are faintly tinted with pink on the umbonal area. Variation in the thickness and in the flexuosity of the posterior dorsal margin can be observed in a series of speci- mens. These features are most pronounced in the larger shells. The present specimens as well as a series in the Collections of the California Academy of Sciences from Pana- ma Bay, a series in the same collection from Santa Elena Bay, Ecuador, collected by Woodbridge Williams, and a series in the H. N. Lowe collection from Acapulco, Mexico, show gradation from small, thin shells with a slightly flexuous posterior dorsal margin to fairly thick shells with strong flexuous pos- terior margins. The hinge of this species is similar to that of Tellina felix except that the right anteri- or lateral is shorter and in the left valve the inner margin back of the beak is slightly thickened and some large specimens show traces of a left posterior lateral. The ivory color, longer posterior end and more convex posterior dorsal margin easily serve to sepa- rate this species from T. felix. Tellina hiberna Hanley27 appears to be a very similar shell. The illustrations of that species published by Hanley and Salisbury indicate that the shell is more abruptly slop- ing posteriorly, that there is a constricted area just anterior to the posterior umbonal ridge and anterior to this the shell is more expanded than in T. tabogensis. Distribution : This species was dredged by the expedition off Meanguera Island, El Sal- vador, in the Gulf of Fonseca, where it oc- curred abundantly in 6 to 16 fathoms, off Corinto, Nicaragua, in 12-13 fathoms, and in the Gulf of Nicoya, Costa Rica, in 4-10 fath- oms. It occurs north to the Gulf of California and south to Ecuador. Subgenus Eurytellina Fischer. Key to the species of Eurytellina. A. Shell white or brownish a. Pallial sinus touching the anterior ad- ductor impression b. Very elongate; adult valves with a median depressed area ventrally planulata 27 See TeUina hiberna Hanley, Thes. Conch., Vol. 1, 1846, p. 282, pi. 67, fig. 63. “Panama ; Bay of Guayaquil.” —Sal- isbury, Proc. Malacol. Soc. London, Vol. 21, Pt. 2, 1934, p. 91, pi. 13, figs. 7, 8, 9. [Illustrations of holotype and para types]. bb. Shell high; without a median de- pressed area ventrally laceridens aa. Pallial sinus not touching anterior ad- ductor impression c. Sculpture coarse, about 1 rib per millimeter panamanensis cc. Sculpture fine, about 3 ribs per millimeter eburnea B. Shell entirely or partly some shade of pink or red a. Concentric sculpture of even strength over shell b. Pallial sinus confluent with a por- tion of the posterior margin of the anterior adductor impression rubescens bb. Pallial sinus not touching anterior adductor impression c. Concentric sculpture decussated by radial striae princeps2S cc. Concentric sculpture not decus- sated, radial striae very fine or absent d. About 10 concentric grooves per millimeter prora dd. About 2 or 3 concentric grooves per millimeter e. Right valve with de- pressed area anterior to posterior umbonal angu- lation simidans ee. Right valve without de- pressed area anterior to posterior umbonal angu- lation f. Posterior dorsal area with strong concen- tric sculpture mantaensis ff. Posterior dorsal area with weak concentric sculpture or of growth lines only ecuadoriana 28 Tellina I Eurytellina I eburnea Hanley. Tellina eburnea Hanley, Proc. Zool. Soc. London, September, 1844 p. 61, “Hab. Tum- bez, Peru ; in soft sandy mud, five fathoms.” —Hanley, Thes. Conch., Vol. 1, 1846, p. 241, pi. 58, fig. 91. Tumbez, Peru. — Sowerby, Conch. Icon., Vol. 17, Tellina, 1867, species 60, pi. 13, fig. 60. Original locality cited. 28 Not represented in the present collection. aa. Concentric sculpture of unequal strength over shell g. Posterior third of shell with coarse concentric lamellae giving way to fine striae anteriorly inaequistriata gg. Posterior third of shell smooth or nearly so in one or both valves, an- teriorly sculptured with distant concentric grooves regia 74 Zoologica: New York Zoological Society [34: 9 Tellina ( Peronaeoderma ) eburnea Hanley, Morch, Malakozool. Blatter, Bd. 7, 1860, p. 186. Sonsonate, El Salvador. Type Locality : Tumbez, Peru, in 5 fath- oms, sandy mud. Range : Gulf of California to Tumbez, Peru. Collecting Stations: Guatemala: 7 miles west of Champerico (197-D-l, 2), 14 fath- oms, mud; El Salvador: La Libertad (198-D -1,2), 13-14 fathoms, mud ; Nicaragua : Cor- into, beach; Panama: Gulf of Chiriqui (221- D-l-5) , 35-40 fathoms, sandy mud. Description : Oblong, solid, opaque, rather inequivalve, convex, whitish, glossy, clearly inequilateral, with strong deep concentric sulci, which usually become obsolete in one of the valves, and which diverge and become elevated on passing the flattened space at the upper edge of the more convex valve ; ventral edge very slightly convex, curving obliquely upward anteriorly; posterior side much the shorter, subcuneiform; the ligamental edge straight, and abruptly sloping; ligament shoi't and prominent ; fold and umbonal ridge almost obsolete; inside pure white, teeth as in punicea. (Hanley, Thes. Conch., 1846). The largest specimen in the present collec- tion measures : length 28.3 mm. ; height, 17.8 mm. The present specimens agree well with the younger stages of a specimen of T. ebur- nea in the collection at Stanford University which was collected in Ecuador by Stanley Herold, which measures : length, 47.5 mm. ; height, 29.8 mm.; convexity (both valves to- gether), 12.6 mm. The concentric sculpture on the posterior dorsal areas becomes stronger and raised after crossing the umbonal ridge. This fea- ture is emphasized by Hanley and Sowerby but it is not very pronounced on the present specimens. Tellina eburnea is very similar to T. alter- nata Say, an east American species. Tellina laplata Pilsbry & Olsson29, describ- ed from the Pliocene of Peru, is also very similar to T. eburnea. Compared to T. laplata the present specimens do not show such strong raised sculpture on the posterior dor- sal areas nor is the sculpture as strong on the left anterior dorsal area as that shown in the illustrations by Pilsbry & Olsson. The shell of Tellina eburnea is higher in proportion to the length as compared to that of T. simulans, furthermoi'e the color is pure white both exteriorly and interiorly. The general character of the pallial sinus is simi- lar to that of T. simulans. It is somewhat higher behind and extends forward almost to but does not quite touch the anterior adduc- tor impression and is confluent with the pal- lial line below. The hinge is similar to that of T. simulans except that the right posterior lateral is less distant from the cardinals. The sculpture is much finer and the pallial sinus 29 Tellina (Eurytellina) laplata Pilsbry & Olsson, Proc. Acad. Nat. Sci. Philadelphia, Vol. 93, September 9, 1941, p. 67, pi. 15, figs. 1-5. "Canoa formation, Punta Blanca.” Ecuador. Pliocene. extends nearer the anterior adductor impres- sion than that of T. panamanensis Li. Distribution: Specimens here referred to Tellina eburnea were dredged by the expedi- tion in 13-40 fathoms, from off Guatemala and El Salvador, in the Gulf of Chiriqui, Panama, and were taken on the beach at Cor- into, Nicaragua. Tellina I Eurytellina I inaequistriata Donovan. Plate I, Fig. 18. Tellina inaequistriata Donovan, Nat. Hist. Brit. Shells, Vol. 4, 1802, pi. 123 [two figs.]. “A very rare species of Tellina communi- cated to Da Costa after his Conchology was published, and therefore not noticed in that work. It has been found by the late Dr. Pul- teney we believe on the coast of Dorsetshire.” — Chenu, Bibl. Conchyl., Ser. 1, Vol. 1, 1845, p. 82, pi. 32, figs. 7, 8. [French translation of Donovan’s work on Shells]. — Hanley, Thes. Conch., Vol. 1, 1846, p. 238, pi. 57, fig. 58; pi. 58, fig. 80. “Bay of Guayaquil; (Cuming).” — Forbes & Hanley, Hist. Brit. Moll., Vol. 1, | 1853 (issued August 1, 1848), p. 314. “In- ' habits the Bay of Guayaquil; was introduced by Donovan, who only surmised that it had been taken by Dr. Pulteney on the Dorset coast.” Tellina sanguinea Wood, Gen. Conch., 1815, p. 159, pi. 44, fig. 2. “This shell is in the cabi- net of Dr. Coombe.” [No locality cited]. — Wood, Index Test., 1825, p. 18, pi. 4, fig. 27. Also ed. 1828. Locality unknown. Also ed. by Hanley, 1842-1856, p. 23, pi. 4, fig. 27. “Guayaquil.” [States that fig. 80 in Thes. Conch, represents T. sanguinea ]. — Hanley, Cat. Rec. Biv. Shells, 1843, p. 67. Tellina ( Eurytellina ) leucogonia Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, No- vember, 1900, p. 317, pi. 4, fig. 5. “Type. — No. 102182, U.S.N.M., from the Gulf of Cali- fornia, Stearns collection.” Type Locality : Bay of Guayaquil, Ecuador ( according to Hanley and Forbes & Hanley) . [Erroneously cited from the coast of Dorset- shire, England, by Donovan]. Range: Gulf of California to the Bay of Guayaquil, Ecuador. Caribbean (Dautzen- berg) . Collecting Stations : Mexico : Santa Cruz Bay (195-D-19-21) , 17-18 fathoms, mud, gr. mud, crushed shell; Tangola-Tangola Bay (196-D-13), 10 fathoms, gr. sand, crushed shell; Costa Rica: Port Parker (203-D-3), 12 fathoms, shelly mud. Description: Shell ovate, compressed and rather flattish, rosy, very finely striated transversely : the striae fewer and larger at the anterior [posterior] end (Donovan). Shell elongate, moderately thick, some- what compressed, glossy, subequilateral, red or orange-red; a well-defined fairly broad posterior area is set off by a rounded post- umbonal ridge; posterior dorsal margin slightly rounded, sloping downward and slightly expanded along the ventral half ; the ornamentation consists of concentric striae which, especially on the right valve, are very 1949] Hertlein & Strong: Mollusks of Mexico and Central America 75 strongly developed and widely spaced on about the posterior third of the shell then change abruptly to fine striae anteriorly; hinge normal for the subgenus; the pallial sinus does not quite touch the anterior ad- ductor impression but is separated by a nar- row area, wholly confluent with the pallial line below and, in general features, similar to that of T. simulans and T. prora. A right valve in the present collection from Santa Cruz Bay, Mexico, measures : length, 23 mm. ; height, 12.5 mm. It agrees well with Hanley’s plate 50, figure 80. Other specimens in the collection are similar but show the strong sculpture only slightly developed an- terior to the angulation. Some of the small shells, especially left valves, almost lack strong concentric sculp- ture anterior to the posterior area; in such cases they may be ornamented only by faint grooves slightly out of harmony with the incremental lines. A series of specimens in the H. N. Lowe collection of the San Diego Society of Natu- ral History, varying in length from approxi- mately 8.5 mm. to 26.2 mm. and identified by Lowe as Tellina leucogonia Dali, are identical with the present specimens. The present ser- ies together with Lowe’s specimens show all variations from young smooth shells up to a large right valve with the typical sculpture of T. inaequistriata. We are therefore inclin- ed to consider Dali’s species as identical with T. inaequistriata. The very distinct concentric sculpture, changing from coarse to fine at about the posterior third of the shell, serves to sepa- rate this form from similar west American species of the genus. Tellina waylandvaughani Maury30, de- scribed from the Miocene of Santo Domingo, is a similar species. Distribution : Specimens of this species were dredged by the expedition in 10 to 18 fathoms from off western Mexico and Costa Rica. The present records of occurrence show the range of this species to extend from the Bay of Guayaquil north to the Gulf of Cali- fornia. Dautzenberg31 cited Tellina inaequi- striata as occurring in the Caribbean region at the Island of Martinique and in Venezuela. He stated that he could detect no differences which would serve as a basis for separating the Caribbean shells from those illustrated under that name from the Bay of Guayaquil. Tellina I Eurytellina I laceridens Hanley. Tellina laceridens Hanley, Proc. Zool. Soc. London, September, 1844, p. 61. “Hab. Tum- bez, Peru; soft sandy mud, five fathoms.” “Var. testa, magis trigona . . .” “Hab. chir- iqui, West Columbia ; sandy mud, three fath- oms.” —Hanley, Thes. Conch., Vol. 1, 1846, P- 243, pi. 61, figs. 168, 176. [Not. pi. 66, SO Tellina waylandvaughani Maury, Bull. Amer. Paleo., Vol. 5. No. 29, 1917, p. 386 (222) , pi. 64 (38) , figs. 7 and 8. ‘‘Zone G. Rio Gurabo at Los Quemados.” Santo Domingo. Miocene. 81 Dautzenberg, P., Mem. Zool. Soc. France, Vol. 13 1900, p. 260. fig. 258.]. Original localities cited. — Sower- by, Conch. Incon., Vol. 17, Tellina, 1867, spe- cies 104, pi. 20, fig. 104. Tumbez, Peru. Type Locality : Tumbez, Peru, in 5 fath- oms, soft, sandy mud. Range: Realejo [near Corinto], Nicara- gua, to Tumbez, Peru. Collecting Stations: Nicaragua: Corinto, beach ; Costa Rica : Port Culebra ; Colombia : Gorgona Island. Description: Shell elongately trigonal, beaks nearly central but slightly anteriorly placed, posterior end obliquely truncated; sculpture of fine, concentric grooves which in places are irregular, the umbonal region, and often the posterior area, relatively smooth ; white, sometimes with a pinkish spot on the umbonal region ; ligament large, exterior; hinge with cardinals grooved, es- pecially the right posterior cardinal which appears ragged due to 7 to 10 grooves; the pallial sinus touches the anterior adductor impression just above the base; interior white and yellow and in large specimens with somewhat granular areas and salmon pink spots. A specimen collected at Gorgona Island, Colombia, measures: length, 53 mm.; height, 34.8 mm.; convexity (both valves together), 11.6 mm. Some of the characters in which this spe- cies differs from Tellina panamanensis Li are: the finer concentric sculpture, larger smooth umbonal area, regular sculpture along the posterior dorsal margin and in that the pallial sinus touches the anterior adductor impression. Distribution: A few specimens of this species were collected by the expedition in the beach drift at Corinto, Nicaragua, one at Port Culebra, Costa Rica, and one at Gor- gona Island, Colombia. It also has been re- corded as occurring in beds of Pliocene age in Panama. Tellina lEurytellinal mantaensis Pilsbry & Olsson. Tellina ( Eurytellina ) mantaensis Pilsbry & Olsson, Nautilus, Vol. 56, No. 3, January, 1943, p. 80, pi. 8, figs. 1-4. “Manta, Ecuador.” Type Locality: Manta, Ecuador. Range : Gulf of Chiriqui, Panama, to Man- ta, Ecuador. Collecting Station : Panama : Gulf of Chir- iqui (221-D-1-5), 35-40 fathoms, sandy mud. Description : Shell elongated, rather com- pressed, the anterior end slightly the long- er, posterior end obliquely subtruncated ; sculpture of flat concentric ridges which are separated by narrower grooves; the poster- ior area is usually ornamented by waved raised threads and with faint radial lines on one or the other valve; hinge of right valve with a strong anterior lateral adja- cent to the cardinals and a more distant pos- terior lateral, the laterals are smaller in the left valve; the pallial sinus reaches almost to the anterior adductor impression and in this character is similar to T. simulans, 76 Zoologica: New York Zoological Society [34: 9 and is confluent with the pallial line below; fresh specimens are rose colored, more deeply on the umbos, and somewhat brown- ish colored ventrally. Two valves in the present collection agree well with the illustrations of T. mantaensis Pilsbry & Olsson. The larger specimen meas- ures approximately 21 mm. in length and 15 mm. in height. The shell of this species differs from that of Tellina simulans in that it is more elon- gated and the dorsal margins slope more gently, especially posteriorly where the area is somewhat flattened. There also are dif- ferences in the details of the hinges of the two species. The right posterior cardinal of T. mantaensis reaches almost to the ventral margin of the hinge plate which beneath that tooth is strongly indented. In T. simu- lans the corresponding tooth reaches little more than halfway to the ventral margin of the hinge plate which at that point is gently rounded. The strong concentric sculpture on the posterior dorsal area and more gently slop- ing anterior dorsal margin are features which serve to sepai'ate this species from T. ecuadoriana Pilsbry & Olsson. Distribution: Two single valves of this species were taken by the expedition in 30- 40 fathoms on a bottom of sandy mud in the Gulf of Chiriqui, Panama. This is an exten- sion north of the known range of the species. Tellina I Eurytellinal panamanensis Li. Tellina panamanensis Li, Bull. Geol. Soc. China, Vol. 9, No. 3, October, 1930, p. 262, pi. 5, fig. 32. “ ‘Brought up by marine dredge from depths varying from 10. ft. to 40. ft. in the mud at the mouth of the Rio Grande near La Boca about one mile from the main- land in Panama Bay.’ ” “Horizon : Gatun formation.” —Pilsbry, Nautilus, Vol. 58, No. 4, April, 1945, p. 145. Tellina ( Eurytellina ) panamanensis Li, Pilsbry, Proc. Acad. Nat. Sci. Philadelphia, Vol. 83, November 13, 1931, p. 436, pi. 41, figs. 4, 5, 6. A Recent shell from Panama Bay. Tellina liana Hertlein & Strong, Nautilus, Vol. 58, No. 3, January, 1945, p. 105. “Dredged off Meanguera Island, El Salva- dor, in the Gulf of Fonseca, in 16 fathoms.” Type Locality : Mouth of Rio Grande near La Boca about 1 mile from the mainland in Panama Bay, 10-40 feet, mud. Range: Tenacatita Bay, Mexico, to the Bay of Panama. Collecting Stations : Mexico : Tenacatita Bay (183-D-2), 30 fathoms, muddy sand; Port Guatulco (195-D-20, 21), 18-23 fath- oms, mud; Tangola-Tangola Bay (196-D-17, 18), 23-30 fathoms, mud; El Salvador: Me- anguera Island, Gulf of Fonseca (199-D-l), 16 fathoms, sand, mud, crushed shell. Description: Shell similar in general out- line to that of Tellina laceridens but more steeply truncated posteriorly ; the left valve is more convex and overlaps the right along the right anterior dorsal margin; a faint / broad median concavity is present toward the ventral margin of the valves ; color gray- ish-white covered by a thin ochraceous per- i iostracum; the concentric sculpture con- sists of ridges (about 1 per millimeter) which on their upper portions are flat, on the lower sloping, crossed by fine weak ra- dial striae; on the right valve a ridge or angulation occurs from beak to base pos- teriorly; on the posterior dorsal area the concentric sculpture is usually sinuated, sometimes bent back, due to the presence | (although sometimes absent) of a median radial convexity; hinge with two cardinals and two laterals in each valve, the right an- terior cardinal grooved, the posterior with about four sulcations, the anterior lateral close to the cardinals, the posterior lateral distant about one-third the length of the i posterior dorsal margin ; left valve with the anterior cardinal grooved, the posterior car- i dinal a thin lamella, laterals weak; pallial sinus highest beneath the beaks, descending to a broadly rounded or blunt point which is well separated from and lower than but j posterior to the anterior adductor impres- , sion, along the base confluent with the pal- t lial line. Dimensions of a typical specimen : I length, 44.9 ; height, 29 mm. ; convexity .} (both valves together), 13 mm.; pallial sinus extends anteriorly 34.5 mm. from pos- terior end of shell. The shell of this species differs from that of Tellina laceridens in the more steeply 't sloping posterior dorsal margin, more con- 1 vex valves, much coarser sculpture which, ■ especially on the left valve, is sinuated on the posterior dorsal area, in the smaller smooth area at the beaks, less crenated car- u dinal teeth and in that the pallial sinus does ' not touch the anterior adductor impression r but is separated from it by a considerable distance. The much coarser sculpture and the much greater distance between the pallial sinus and the anterior adductor impression are features separating it from T. eburnea. Distribution : Specimens of Tellina pan- I amanensis were dredged by the expedition from Tenacatita Bay, Mexico, to Meangu- era Island, El Salvador, in 16-30 fathoms. This record of Tenacatita Bay, Mexico, is an extension north of the known range of the species. It also has been recorded as oc- curring in the Pliocene of Ecuador. Tellina (Eurytellinal planulata Sowerby. Plate I, Fig. 22. Tellina planulata Sowerby, Conch. Icon., i Vol. 17, Tellina, June, 1867, species 186, pi. 33, fig. 186. “Hab. — ?” Type Locality: Gulf of Dulce, Costa Rica (here designated as type locality). No lo- cality originally cited. Range: La Libertad, El Salvador, to the Gulf of Dulce, Costa Rica. Collecting Stations: El Salvador: La Lib- ertad (198-D-2), 14 fathoms, mud; Costa 1949] Hertlein & Strong: Mollusks of Mexico and. Central America 77 Rica: Gulf of Dulce; Golfito Bay, Gulf of Dulce. Description: Shell elongate, the posterior end the longer, fairly thick, white, similar to Tellina laceridens in general features but longer in proportion to the height; some- times with a low broad medial depression toward the ventral margin; posterior dor- sal area set off by a slight angulation ; sculp- ture consisting of very fine, shallow, some- what irregularly spaced concentric grooves ; right valve with two grooved cardinals, an- terior lateral close to the cardinals but the posterior lateral distant about half the length of the posterior dorsal margin; left valve with a grooved anterior cardinal and posterior to this two small, thin, laminae, latex'als small; the pallial sinus projects an- teriorly and barely touches the base of the anterior adductor impression, along the base it is confluent with the pallial line. A left valve measures: length, 59.2 mm.; height, 33 mm. ; convexity (one valve) , 6 mm. Several single valves from the Gulf of Dulce agree so closely with Sowerby’s de- scription and figure of Tellina planulata that we have assigned our specimens to that spe- cies. This species was originally described without information as to the locality from which it came. Paetel32 cited the species as occurring at “Sitka,” Alaska, but we have not seen any specimens from that region which appear to be referable to it. Other than Pae- tel’s record the species apparently has not been recognized as occurring elsewhere. We therefore have designated the Gulf of Dulce as type locality. Sowerby stated that it is “A much larger and flatter shell than Tel- lina eburnea, with closer grooves and no transverse ridges on the dorsal margin of the overlapping valve.” Those differences are true with regard to the present specimens. The pallial sinus in the present specimens touches the anterior adductor impression whereas in T. eburnea it is separated from the corresponding impression by a narrow space. Sowerby stated with regard to the posterior end: “terminal margin notched.” This latter feature is not pronounced on our specimens but they are not perfectly pre- served. The present specimens closely resemble Tellina ecuadoriana Pilsbry & 01sson32a de- scribed from Ecuador. A single valve of that species collected by the senior author at Cor- into, Nicaragua, is in the collections of the California Academy of Sciences. Compared to Tellina ecuadoriana the pos- terior dorsal margin of T. planulata slopes a little more steeply and the posterior area, especially on the right valve, appears to be a little narrower than the corresponding 32 Paetel, Fr„ Cat. Conchyl.-Samm]., Vierte Neube- arbeitung (Berlin: Verlag von Gebruder Paetel), Abt 3, 1890, p. 49. 32a Tellina ( Eurytellina ) ecuadoriana Pilsbry & Olsson, Proc. Acad. Nat. Sci. Philadelphia, Vol. 93, September 9, 1941, p. 67, pi. 15, figs. 6, 7, 8. Canoa formation, Punta Blanca. Ecuador, Pliocene. Also Recent, Santa Elena. Ecuador (type), also at Canoa and Callo, the port of Jipijapa. area on the species described by Pilsbry & Olsson. The shell of the present species is white rather than rose red with whitish zones. Furthermore the pallial sinus in the present specimens touches the anterior ad- ductor impression while in T. ecuadoriana the two are separated by a narrow space. Distribution : Several valves of this spe- cies were collected by the expedition on the beach in the Gulf of Dulce, Costa Rica. One small specimen with both valves was dredged in 14 fathoms off La Libertad, El Salvador. Tellina lEurytellina) prora Hanley. Tellina prora Hanley, Proc. Zool. Soc. Lon- don, September, 1844, p. 61. “Hab Porto St. Elena, West Columbia; sandy mud, six fath- oms; and Salango, West Columbia, sandy mud, nine fathoms.” — Hanley, Thes. Conch., Vol. 1, 1846, p. 243, pi. 60, fig. 152. Original localities cited. — Sowerby, Conch. Icon., Vol. 17, Tellina, 1866, species 90, pi. 18, fig. 90. Original localities cited. — Salisbury, Proc. Malacol. Soc. London, Vol. 21, Pt. 2, July, 1934, p. 86. “The shell ranges through the south of the Panamic and north of the Peruvian areas.” Tellina cibaoica Maury, Li, Bull. Geol. Soc. China, Vol. 9, No. 3, 1930, p. 261, pi. 4, fig. 30. Dredged in Panama Bay. Referred to the Gatun formation, Miocene. According to Pilsbry {Proc. Acad. Nat. Sci. Philadelphia, Vol. 83, 1931, p. 430), Li’s record was based upon “A left valve of Tellina prora Hanley”. . . Not Tellina cibaoica Maury, 1917. Santo Domingo, Miocene. Type Locality: Santa Elena, Ecuador, in 6 fathoms, sandy mud (here designated as type locality). Salango, Ecuador, in 9 fath- oms, sandy mud, also originally cited by Hanley. Range: Mazatlan, Mexico, to the Bay of Guayaquil, Ecuador. Collecting Stations: Mexico: Tangola- Tangola Bay (196-D-17), 23 fathoms, mud; Guatemala: 7 miles west of Champerico ( 197-D-l, 2) , 14 fathoms, mud ; El Salvador : La Libertad (198-D-2), 14 fathoms, mud; Meanguera Island, Gulf of Fonseca (199- D-l), 16 fathoms, sand, mud, crushed shell; La Union, Gulf of Fonseca (199-D-15), 6 fathoms, mud ; Nicaragua : Potosi and Mony- penny Point; Costa Rica: Port Parker (203- D-3), 12 fathoms, shelly mud; Panama: Ba- hia Honda (222-D-5), 11 fathoms, mud, shells, leaves. Description: Shell ovately trigonal, smooth, polished, a posterior area set off by an angulation, colored rose pink with whitish concentric bands; sculpture of very fine closely spaced (about 10 per millimeter) incised concentric striae; hinge with two grooved cardinals in each valve, the left pos- terior one very narrow, two laterals in each valve, those in the left valve much the smaller; pallial sinus highest in middle part of shell, usually separated from the anterior adductor impression by about a millimeter 78 Zoological New York Zoological Society [34:9 but sometimes almost, but not quite, touch- ing, the end blunt and almost in line verti- cally below the posterior side of the adductor impression, along the base confluent with the pallial line. A large specimen dredged off Guatemala measures approximately: length, 46.4 mm.; height, 17.5 mm.; convexity (both valves together), 12 mm. The shell of this species differs from that of Tellina rubescens Hanley in that the pos- terior slope is more gently inclined, the con- centric incised striae are much finer and much more closely spaced and the pallial sinus does not touch the anterior adductor impression. The very much finer and more closely spaced concentric sculpture easily serves to separate the species from T. sim- ulans C. B. Adams. The record of Tellina prora from the Cape Verde Islands in the Atlantic cited by Roche- brune33 can be referred to some other spe- cies. Tellina (Eury tellina) trinitatis Tom- lin34, described from Colon Harbor on the east side of the isthmus of Panama, is said to be similar to T. prora. Distribution: This species was collected by the expedition off western Guatemala, El Salvador and Nicaragua, in 6-14 fath- oms on a muddy bottom. It also has been re- corded as occurring in the Pliocene of Ecua- dor. Tellina I Eurytellina! regia Hanley. Tellina regia Hanley, Proc. Zool. Soc. Lon- don, September, 1844, p. 61. “Hab. Real Lle- jos, Central America; in coarse sandy mud, seven fathoms.” — Hanley, Thes. Conch., Vol. 1, 1846, p. 240, pi. 60, fig. 140. Original locality cited. Type locality: Real Llejos [near Corinto], Nicaragua, in 7 fathoms, coarse, sandy mud. Range : Known only from the type locality and vicinity. Collecting Station: Nicaragua: Corinto, beach drift. Description: Oblong, thin, rather com- pressed, almost inequivalve, subequilateral; extremely glossy, both externally and in- ternally of a deep subpellucid purplish crim- son; the surface marked with distant con- centric grooves, which posteriorly become obsolete in one or both of the valves; the ventral edge nearly straight, subretuse in the middle; anterior side slightly shorter, its extremity obtusely rounded ; posterior extremity almost biangulated; dorsal mod- erately and almost equally sloping on either side of the beaks, nearly straight pos- teriorly; umbonal ridge and flexure nearly obsolete; the ligament rather prominent; teeth as in punicea. (Hanley, Thes. Conch., 1846). S3 Peronaeoderma prora Hanley, Rochebrune. Nouv. Arch. Mus. d’Hist. Nat. (Paris), Ser. 2, Vol. 4, 1881, p. 258. “Hab.— Rade de Saint-Vincent.” Cape Verde Islands. 34 Eurytellina trinitatis Tomlin, Jour. Conch., Vol. 18, No. 11, July, 1929, p. 310. “Hab. Colon Harbour, not un- common, dead but very fresh in 6 f.” A left valve from Corinto, Nicaragua measures approximately: length, 17 mm. height, 10 mm.; convexity (one valve), 1 mm. The outline as well as the other features of the present specimen are similar to thost described for Tellina regia which came from the same general vicinity. The concentric ornamentation of the shell is like that de- cribed for T. regia, namely, distant concen- tric grooves which become obsolete poste riorly. Tellina regia differs from Tellina rubes cens in that the concentric grooves become obsolete posteriorly, the dorsal margins do not slope so steeply, and the pallial sinus does not touch the anterior adductor impres- sion. It differs from Tellina princeps in lack ing radial striae. Distribution: Only one valve referred to this species was found in the beach drift at Corinto, Nicaragua. kPr liter Ike i Lov Tellina I Eurytellina} rubescens Hanley. Tellina rubescens Hanley, Proc. Zool. Soc.\ m London, September, 1844, p. 60. “Hab. Pan- PP ama and Tumbez; in sandy mud.” — Hanley, !IB Thes. Conch., Vol. 1, 1846, p. 242, pi. 60, fig. jj1 153. Tumbez, and Panama (Cuming). — $!’ Sowerby, Conch. Icon., Vol. 17, Tellina, 1866, v species 93, pi. 18, fig. 93. “Hab. Tumbez and A Panama, Peru.” Type Locality: Panama in sandy mud (here designated as type locality). Tumbez, Peru, in sandy mud, also originally cited. S( Range: Tenacatita Bay, Mexico, to Turn- ; bez, Peru. Collecting Stations: Mexico: Tenacatita ‘ Bay; El Salvador: La Union, Gulf of Fon- seca (199-D-12), 5 fathoms, mud; Nicara- gua: Potosi and Monypenny Point, Gulf of D Fonseca; Costa Rica: Port Parker. Description: Shell trigonally ovate, smooth, polished, colored exteriorly and in- teriorly by light and darker concentric bands ; of rose pink; posterior dorsal margin slop- ing steeply, the area set off by an angulation ; | sculpture consists of fine concentric grooves (about 2 per millimeter) and between these finer concentric striae, the whole crossed by very fine submicroscopic radial striae; hinge with two cardinals and two laterals in each valve, the right anterior cardinal usually grooved, the posterior bifid, left an- terior grooved, the posterior one thin, lat- erals in left valve weak ; anterior end of pal- lial sinus confluent with the lower posterior margin of the anterior adductor impression from a point just above the base to approx- imately the middle of the base of the im- pression, confluent with the pallial line be- low. A large specimen from the Gulf of Fon- seca measures : length, 43 mm. ; height, 37 mm.; convexity (both valves together), 9.3 mm. The shell of Tellina rubescens differs from that of T. prora Hanley in that it is higher 1949] Hertlein & Strong: Mollusks of Mexico and Central America 79 in proportion to the length, the posterior dorsal margin slopes more steeply, the in- cised concentric sculpture is more widely spaced and the pallial sinus is confluent with a portion of the posterior side of the ante- rior adductor impression. These same char- acters of proportionate height to length and that of the pallial sinus serve to separate T. rubescens from T. simulans C. B. Adams, | a species in which the concentric sculpture is much more strongly developed. Distribution : This species was taken by the expedition along the west coast of Mex- ico and Central America but at no place abun- dantly. It also has been recorded by Arnold, 1903, as occurring in the upper Pleistocene of San Pedro, California. Some of the rec- ords of the occurrence of this species at San Ignacio Lagoon and at Magdalena Bay, Lower California, are referable to T. simu- lans. Tellina I Eurytellina I simulans C. B. Adams. Tellina simulans C. B. Adams, Ann. Ly- ceum Nat. Hist. New York, Vol. 5, July, 1852, pp. 508, 546 (separate pp. 284, 322). “Pan- ama.” Also cited from Xipixapi, Ecuador, in sandy mud at 10 fathoms. — Romer, Neues Syst. Conchyl.-Cab. Martini-Chemnitz, Bd. 10, Abt. 4, Tellina, 1872, p. 99, pi. 25, figs. 4, 5. Tellina punicea Born, Carpenter, Cat. Mazatlan Shells, August, 1855, p. 35. Maz- atlan, Mexico. Also earlier records cited. — Sowerby, Conch. Icon., Vol. 17, Tellina, 1866, species 53, pi. 12, fig. 53. “Hab. Xipixapi, W. Columbia; in sandy mud, ten fathoms; H. Cuming.” Not Tellina punicea Born, Test. Mus. Caes. Vind., 1780, p. 33, pi. 2, fig. 8. “Patria ig- nota.” [Now believed to be a Caribbean spe- cies. See Gardner, U. S. Geol. Surv., Prof. Paper 142 -E, 1928, p. 193]. Tellina costaricana Olsson, Li, Bull. Geol. Soc. China, Vol. 9, No. 3, October, 1930 p. 262, pi. 4, fig. 31. Dredged in Panama Bay at mouth of Rio Grande River in 10-40 ft. “Occurrence : Gatun Stage, Banana River, Costa Rica.” “Horizon: Gatun formation.” This record is based upon “Two valves of Tellina simulans C. B. Ad.,” Panama, Recent, according to Pilsbry ( Proc . Acad. Nat. Sci. Philadelphia, Vol. 83, 1931, p. 430). Type Locality : Panama. Range : Scammon Lagoon, Lower Califor- nia, to the Gulf of California and south to Tumbez, Peru. Collecting Stations: Mexico: Santa Inez Bay, Gulf of California; Tenacatita Bay; 17 miles SE. X E. of Acapulco (189-D-3), 13 fathoms, mud; Nicaragua: Corinto (200- D-ll, 19), 8-13 fathoms, sand, mangrove leaves; Costa Rica: Culebra Bay; 1 mile south of Golfito. Description: Shell elongately oval, pos- terior dorsal margin obliquely sloping, the end obliquely truncated and slightly bent to the right, rose pink with whitish concentric bands; posterior area set olf by an angula- tion anterior to which is a shallow broad depressed area on the right valve, the pos- terior area bears a narrow, shallow, radial furrow which is especially noticeable on the left valve; sculpture of regular, deep, con- centric grooves (about 2 or 3 per millimeter) and very fine submicroscopic radial striae; hinge with anterior laterals close to the car- dinals, posterior laterals distant, weaker in left valve; the pallial sinus does not quite touch the anterior adductor impression, the end is blunt and almost in line vertically below the posterior margin of the adducto** impression, along the base it is confluent with the pallial line. A large specimen of this species in the Henry Hemphill Collection of the California Academy of Sciences from Magdalena Bay, Lower California, measures: length, 48.8 mm.; height, 29 mm.; convexity (both valves together), 11 mm. This species, in some instances, has been cited in the earlier literature under the names of Tellina punicea Born and T. ru- bescens Hanley. The shell of this species differs from that of T. rubescens in the greater length in pro- portion to the height, in the much deeper, sti'onger and more widely spaced concentric sculpture and in that the pallial sinus does not touch the anterior adductor impression. Tellina simulans is similar to T. angulosa Gmelin ( T . punicea of some authors), an east American species ; in fact Carpenter and others considered the two to be identical. According to Adams T. simulans differs from the east American species in that “its fur- rows are deeper and are continued over the flexure, without change of depth ; the inter- spaces are less flattened, and the lateral teeth are nearly obsolete.” A comparison of spec- imens of T. simidans with a series of T. an- gulosa collected by F. M. Anderson at Car- tagena Bay, Colombia, shows differences be- tween the two. The west coast shells are more pointed posteriorly, there is a low depressed area anterior to the posterior angulation on the right valve, and the concentric grooves along the posterior dorsal margin bend more acutely upward than on the east coast shells. Tellina princeps Hanley, described from Tumbez, Peru, is a distinct species posses- sing a large, red, subequilateral shell with a gently sloping posterior dorsal margin and the concentric sculpture is crossed by strong radial striae. Distribution: This species was taken by the expeditions although not abundantly, from Santa Inez Bay, in the Gulf of Califor- nia to Culebra Bay, Costa Rica. It ranges south to Peru. It also is known to occur in the Pleistocene at San Ignacio Lagoon and at Magdalena Bay, Lower California. Subgenus T ellinidella Hertlein & Strong, subgen. nov. Type: Tellinides purpureus Broderip & Sowerby. Shell elongate, compressed, very thin, 80 Zoologica : New York Zoological Society [34 : 9 with a strong posterior angulation; orna- mented with fine concentric granulated ridges which are crossed by impressed ra- dial striae forming reticulate sculpture; hinge as in Eurytellina but with a very small right anterior lateral and a weak pos- terior lateral, in the left valve the laterals are faint or obsolete. The general outline and hinge of this new subgenus are similar to those of Eury- tellina. The weak distant right posterior lat- eral and the very thin shell, which is orna- mented by both concentric and radial sculp- ture, are characteristic features of the type species of T ellinidella. Dali, 1900, placed Tellina purpureus Brod- erip & Sowerby in the subgenus Tellinides Lamarck33. That name was proposed by La- marck for a genus of Tellina with the sole species T. timorensis as type. No illustra- tions accompanied that work. Dubois36 dis- cussed Lamarck’s genera but he did not il- lustrate the type species of Tellinides. Nei- ther did Delessert, 1841, include T. timor- ensis among his illustrations of the types of Lamarck’s shells. Hanley37 later gave il- lustrations showing two views of the ex- terior of right valves of T. timorensis. Phil- ippi38 also illustrated a species under that name and gave views of both the exterior and interior of the right valve and an um- bonal view of both valves. His illustrations agree well with Lamarck’s description of the species. Bertin39, 1878, stated that 4 type specimens of Lamarck’s species were present in the collections of the Museum of Natural History in Paris. In his synonymy of the species he included references to Han- ley’s figures 158 and 172, Philippi’s figure 3, as well as illustrations of Romer’s, 1812, plate 34, figures 4-6. It appears then that these figures may be considered to represent authentic speci- mens of T. timorensis. The right valve has two diverging cardinal teeth, a close ante- rior lateral and a weak close posterior lat- eral. The left valve is said to lack laterals. The valves are not flexed posteriorly and the posterior umbonal fold or ridge is broad- ly rounded and nearly obsolete. The hinge of the right valve of Tellina purpureus has two diverging cardinal teeth, the posterior one strong and bifid or grooved. The posterior lateral is distant from the cardinals. A small lamina sometimes occurs near the margin above the small anterior lateral and sometimes the margin is some- 35 Tellinides Lamarck, Hist. Nat. Anim. s. Vert., Vol. 5, July, 1818, p. 535. Type, Tellina timorensis, p. 536. “Habite l'ocean des grandes Indes ou austral, pres de Timor.” 36 Dubois, C., Epit. Lamarck’s Arrang. Test., 1824, p. 68. 37 Tellina ( Tellinides ) timorensis Lamarck, Hanley, Thes. Conch., Vol. 1, 1846, p. 292, pi. 61, figs. 158 and 172. Isle of Negros, Philippines, etc. 38 Tellina timorensis (Tellinides) Lamarck, Philippi, Abbild. u. Beschreib. Conchyl., Bd. 2, Heft 4, Tellina, August, 1846, p. 90 (22), pi. 4, fig. 3. Timor, Philippine Islands, Sumatra, etc. 39 Bertin, V., Nouv. Arch. Mus. Hist. Nat. (Paris), Ser. 2, Vol. 1, 1878. p. 283. [He mentioned that T. timorensis is represented in the collection by specimens including “par 4 individus ordinaires de Timor (types de Lamarck)."] what projecting just over the posterior lat- eral. Furthermore the shell has a strong, angular posterior umbonal fold exteriorly and the concentric sculpture on the shell is decussated by radial grooves giving it a fine- ly granulose character. These characters on the present shell are so different from those of T. timorensis that we propose a new sub- genus Tellinidella with Tellina purpureus Broderip & Sowerby as type. Tellina ITellinidellal purpureus Broderip & Sowerby. Tellinides purpureus Broderip & Sowerby, Zool. Jour., Vol. 4, No. 15, January, 1829, p. 363. “Hab. ad littora Oceani Pacifici.” — Sowerby, Zool. Beechey’s Voy., 1839, p. 153, pi. 42, fig. 2. “Inhabits the sandy shores of the Pacific Ocean.” Tellina ( Tellinides ) purpurascens Brod- erip & Sowerby, Hanley, Thes. Conch., Vol. 1, 1846, p. 295, pi. 62, fig. 194. “Real Leijos [Llejos] Central America (Cuming).” Not Tellina purpurascens Gmelin, Linn, i Syst. Nat., ed 13, Vol. 1, Pt. 6, 1790, p. 3237. Habitat not cited. Ref. to Lister, Conch., pi. 391, fig. 230. Also “B” and “Y,” ref. to Gual- tieri, Test., pi. 77, figs. L? and M? Tellina purpurata is described on p. 3243. Hab. not cited. Ref. to Gualtieri, Test., pi. 77, fig. L. Tellina broderipii “Desh. ms. (teste Cum.)” Carpenter, Cat. Mazatlan Shells, Au- gust, 1855, p. 32. “Mazatlan.” Tellina purpurascens Broderip & Sowerby, Sowerby, Conch. Icon., Vol. 17, Tellina, 1867, species 103, pi. 20, fig. 103. Same locality as given by Hanley. Type Locality : Real Llejos [near Corinto], Nicaragua (here designated as type local- ity). Shores of Pacific Ocean originally cit- ed. Range : Altata, Mexico, in the Gulf of Cal- ifornia, to Colombia. Collecting Stations: Mexico: Tenacatita Bay; Sihuatanejo Bay; Nicaragua: Corinto, beach drift. Description: Shell ovately oblong, sub- equilateral, thin, compressed, colored a beau- tiful purplish-rose with the dorsal margins white; sculpture of close, decussating con- centric and radial striae; posteriorly a low radial furrow is present near the dorsal mar- gin which anteriorly is bounded by a low carina; right valve with two cardinals, the posterior one the larger and bifid, and a very close, small, anterior lateral above which there is sometimes a small lamina near the margin, and a posterior lateral distant near- ly half the length of the posterior dorsal margin, above this tooth the margin is sometimes somewhat projecting; left valve with laterals obsolete; the end of the pal- lial sinus is considerably lower than and pos- terior to the anterior adductor impression, the base is confluent with the pallial line. A right valve from Tenacatita Bay, Mex- ico, measures : length, 49.4 mm. ; height, 26.4 mm.; convexity (one valve), 4 mm.; pallial 1949] Hertlein & Strong: Mollusks of Mexico and Central America 81 sinus extends forward 36 mm. from the pos- terior end of the valve. The species attains a greater size than this specimen. The original name for this species, given by Broderip & Sowerby in 1829, was Tellin- ides purpureas. Dali40 apparently considered that combination of names preoccupied due to the fact that Tellinides is now considered to be a subgenus of Tellina, also the fact that Dillwyn41 had referred to a Tellina pur- purea. However, Dillwyn in a footnote, in re- marking on certain of Gmelin’s species, re- ferred to Gmelin, page 3243, and on that page of Gmelin’s work the species was cited as Tellina purpurata .. It thus appears that Dillwyn’s spelling of “purpurea” was a mis- print and not a renaming of Gmelin’s spe- cies. If this view is accepted, there then ap- pears to be no valid reason for rejecting the original name given the west American spe- cies by Broderip & Sowerby. The name Tel- lina purpurascens which Hanley used for this species is preoccupied by Tellina pur- purascens Gmelin, 1790. The name Tellina broderipii attributed to Deshayes was ap- plied to the west American form by Carpen- ter. The shell of this species bears a resem- blance to that of Tellina princeps Hanley42 but is thinner, narrower, the posterior dor- sal margin slopes more gently, the radial striae are stronger and the right posterior lateral is weaker. Distribution: A few specimens of this species, nearly all right valves, were taken by the expedition along the coast of west Mexico and in the beach drift at Corinto, Nicaragua. Subgenus Macaliopsis Cossmann. Key to the species of Macaliopsis. A. Beaks directed strongly anteriorly; con- centric lamellae about 1 per mm. .... lyra B. Beaks directed only slightly anteriorly; concentric lamellae, lower, fewer, usually about 3 per mm. lyrica Tellina I Macaliopsis! lyra Hanley. Tellina lyra Hanley, Proc. Zool. Soc. Lon- don, September, 1844, p. 68. “Hab. Tumbez, Peru.”— Hanley, Thes. Conch., Vol. 1, 1846, p. 271, pi. 62, fig. 187. Tumbez, Peru. — Sow- erby, Conch. Icon., Vol. 17, Tellina, Septem- ber, 1867, species 203, pi. 36, fig. 203. Tumbez, Peru. Type Locality: Tumbez, Peru. Range : Lower California to Tumbez, Peru (Dali). 40 Dali, W. H., Proc. U. S. Nat. Mus., Vol. 23, 1900, p. 302. ■U Dillwyn, L. W., Descript. Cat. Rec. Shells, Vol. 1, 1817, p. 72, footnote. 42 Tellina princeps Hanley, Proc. Zool. Soc. London, Sep- tember, 1844, p. 62. “Hab. Tumbez, Peru ; soft sandy mud. five fathoms.” —Hanley, Thes. Conch., Vol. 1, 1846, p. 238, pi. 63, fig. 206. “Tumbez, Peru; (Cuming).” —Sowerby, Conch. Icon., Vol. 17, Tellina, 1867, species 135, pi. 25, fig. 135. Tumbez, Peru. —Salisbury, Proc. Malacol. Soc. London, Vol. 21, Pt. 2, 1934, p. 91, pi. 9, fig. 4. (Illustration of type] . Collecting Stations: Guatemala: 7 miles west of Champerico (197-D-l, 2), 14 fath- oms, mud; El Salvador: La Libertad (198- D-2), 14 fathoms, mud, Description: Shell elliptic, thin, compress- ed, slightly longer anteriorly, dull white exteriorly and interiorly, beaks curved for- ward ; anterior dorsal margin excavated be- low the beaks, anterior end rounded, ventral margin broadly rounded, the posterior dorsal margin nearly straight or very broadly curv- ed and sloping posteriorly directly from the beaks ; a deep smooth lunule and escutcheon present; the sculpture consists of regular, thin, sharp, raised, concentric ribs which are separated by much wider interspaces (about 1 mm. wide) but which become narrower toward the ventral margin, a narrow pos- terior area is set off by an umbonal carina; hinge of right valve with a strong grooved triangular posterior and a thin anterior car- dinal near the margin, a strong anterior and posterior lateral present; left valve with a grooved anterior and a thin posterior cardi- nal tooth and lateral triangular projections of the nymph ; the pallial sinus extends for about three-fifths the length of the shell, sub- trigonal above and highest just posterior to a line vertical with the beaks, then descend- ing and narrowly elliptically rounded anteri- orly then bending posteriorly and for about two-thirds of its length confluent with the pallial line. The largest specimen in the present collec- tion, a left valve, measures : length, 50 mm. height, 35 mm.; convexity (one valve), 6 mm.; pallial sinus extends anteriorly 29 mm. from the posterior end of the valve. The strongly anteriorly directed beaks, more convex anterior dorsal margin which is excavated beneath the beaks, higher and more widely spaced concentric sculpture and shorter and more trigonal pallial sinus are features separating this species from Tellina lyrica Pilsbry & Lowe. Tellina protolyra Anderson43 is a similar species but it is smaller, less elongate, more inflated and the posterior dorsal margin is straighter than in the present species. Tellina ( Macaliopsis ) aequizonata Pilsbry & Olsson44, described from the Pliocene of Ecuador, is said to be much larger, more strongly sculptured and more circular in out- line than T. lyra. Distribution: A few specimens, mostly single valves, of this interesting species were dredged in 14 fathoms off Guatemala and El Salvador on a mud bottom. Tellina I Macaliopsis! lyrica Pilsbry & Lowe. Tellina ( Macaliopsis ) lyrica Pilsbry & Lowe, Proc. Acad. Nat. Sci. Philadelphia, 43 Tellina protolyra F. M. Anderson, Proc. Calif. Acad. Sci., Ser. 4, Vol. 18, No. 4, March 29, 1929, p. 174, pi. 21, figs. 2, 3. "From Loc. 267-B, C. A. S., horizon M-N, of the Tubera group, Colombia; Miocene.” 44 Tellina ( Macaliopsis ) aequizonata Pilsbry & Olsson, Proc. Acad. Nat. Sci. Philadelphia, Vol. 93, September 9, 1941, p. 68, pi. 14. fig. 7. “Jama formation, Puerto Jama.” Ecuador. Pliocene. 82 Zoologica: New York Zoological Society [34 : 9 Vol. 84, May 21, 1932, p. 94, pi. 10, figs. 4, 4a. “Guaymas in about 20 fathoms.” Type Locality : Guaymas, Mexico, in about 20 fathoms. Range: Gulf of California to the Gulf of Chiriqui, Panama. Collecting Stations: Mexico: Santa Inez Bay in the Gulf of California (143-D-3, 4), 25-35 fathoms, mud, crushed shell, sand ; El Salvador: La Libertad (198-D-2), 14 fath- oms, mud ; Panama : Gulf of Chiriqui (221-D- 1-5) , 35-40 fathoms, sandy mud. Description : Shell transversely oval, beaks turned slightly forward, anterior end broadly rounded, posterior dorsal margin broadly curved and rather steeply sloping; a slight angulation sets off a narrow posterior area ; lunule smooth, slightly sunken ; escutcheon deeply sunken and bounded by a high keel; sculpture consists of fine, close, concentric ribs (about 3 per millimeter) which are nar- rower than the interspaces ; right valve with a simple anterior and grooved posterior cardinal and well-developed laterals, the nos- terior one more distant; left valve with a grooved anterior and thin posterior lamella- like cardinal, anterior lateral weak, posterior lateral represented by a triangular projec- tion of the margin ; the pallial sinus is high- est posteriorly, rounded at the end and extends forward about four-fifths the length of the shell and is separated from the ante- rior adductor impression bv a considerable distance, for over half its length along the base it is confluent with the pallial line. A large right valve from Santa Inez Bay in the Gulf of California measures; length, 40.2 mm.; height, 29.6 mm.; convexity (one valve). 6.5 mm.; pallial sinus extends for- ward 31 mm. from the posterior end of the valve. The exterior of fresh valves shows a bril- liant iridescence of spectral colors due to the fine growth lines between the concentric ridges acting as a diffraction grating. The specimens in the present collection show the differences pointed out by Pilsbry & Lowe between this species and Tellina lyra. They stated : “Related to T. lyra Hanley, but relatively longer, with the beaks less, only very slightly, turned forward, the dorsal margin in front of them less convex, and the concavity runnning to the posterior basal extremity wider.” The concentric ribbing on T. lyrica is much finer and more closely spaced than that on T. lyra. Distribution: This species was dredged in Santa Inez Bay in the Gulf of California, in 25-35 fathoms, off La Libertad, El Salvador, in 14 fathoms and in the Gulf of Chiriqui, Panama, in 35-40 fathoms, in mud and sandy mud. It also has been recorded as occurring in the Pliocene of Ecuador. Subgenus Mer/sco Dali. Key to the species of Merisca. A. Interspaces with minute radial striae a. Rostrum attenuated posteriorly; con- centric lamellae about 1 per mm. crystallina aa. Rostrum not attenuated posteriorly ; concentric lamellae about 5 per mm. reclusa B. Interspaces without radial striae proclivis Tellina l Merisca I crystallina Spengler. Tellina crystallina Chemnitz, Neues Syst. Conchyl.-Cab. von Martini-Chemnitz, Bd. 11, 1795, p. 210, pi. 199, figs. 1947, 1948. “Es ist diese Muschel an der Nordamericanischen Kiiste bey Newport auf Rhode-Island ge- funden worden.” (Spengler). — Spengler, Skr. Nat. Selsk. (Copenhagen), Vol. 4, No. 2, 1798, p. 113. “Fra Newport Long-Island.” Ref. to Chemnitz, pi. 199, figs. 1947, 1948.— Wood, General Conch., 1815, p. 149. [No lo- cality cited]. Ref. to Chemnitz, pi. 199, figs. 1947, 1948. — Hanley, Thes. Conch., Vol. 1, 1846, p. 270, pi. 57, fig. 43. “St. Elena, West Columbia.”— Olsson, Bull. Amer. Paleo., Vol. 9, No. 39, 1922, p. 421 (249), pi. 29 (26), fig. 10. Banana River, Costa Rica. Miocene. Also Recent. Tellina ( Merisca ) crystallina Wood, Dali, Proc. U. S. Nat. Mus., Vol. 23, 1900, pp. 293, 302, 311, pi. 2, fig. 10. Sullivans Island, South Carolina, to Cartagena, Colombia. Lower California to Panama. — M. Smith, Panamic Mar. Shells (Tropical Photogr. Lab., Winter Park, Florida), p. 64, fig. 834. Lower Cali- fornia to Guayaquil, Ecuador. Also West Indies. Tellina schrammi Recluz, Joum. de Con- chyl., Vol. 4, December, 1853, p. 152, pi. 6, figs. 7, 8. . . “habite la rade de la Pointe-a- Pitre (Guadeloupe) : elle a ete draguee sur un fond vaseux au fond de cette rade.” Type Locality: Newport, Rhode Island. Range: Scammon Lagoon, Lower Cali- fornia, to the Gulf of California and south to Guayaquil, Ecuador. Also Atlantic, from Charleston, South Carolina, to Cartagena Bay, Colombia. Collecting Stations: Mexico: Tangola- Tangola Bay, on beach; Nicaragua: Corinto (200-D-19), 12-13 fathoms, mangrove leaves, also on beach. Description: Shell rather small, roundly trigonal, posteriorly rostrate, somewhat at- tenuated and with a sharp flexure, thin, white; sculpture of strong, raised distant, concentric ribs separated by much wider in- terspaces (about 1 mm. wide) although the spacing varies with various specimens, and where the lamellae cross the flexure they are V-shaped; the interspaces contain fine con- centric striae (sometimes about 8) and faint radial striae; right posterior and left ante- rior cardinals grooved, lateral teeth in right valve well developed, in the left obsolete or nearly so; pallial sinus high behind then descending and sometimes touching the base of the anterior adductor impression, wholly confluent with the pallial line below. The specimens in the present collection are 1949] Hertlein & Strong: Mollusks of Mexico and Central America 83 small but a large valve in the collections of the California Academy of Sciences, from Mazatlan, Mexico, measures : length, 23 mm. ; height, 17.4 mm. ; convexity (one valve) , 3.4 mm. Chemnitz originally described and illus- trated Tellina crystallina and cited Spengler as the authority for the locality, Newport, Rhode Island. The International Committee on Zoological Nomenclature has recently ruled against acceptance of the specific names proposed by Chemnitz. Spengler next described the species. He referred to Chem- nitz’s description and illustrations and gave the same locality as Chemnitz. However, the species is not known with certainty to occur north of Charleston, South Carolina. One might venture the opinion that since Spengler was a citizen of Denmark, the type specimens secured by him originally might have come from the Danish West Indies. Salisbury45 recently stated that the east and west American shells referred to this species should be considered as separate spe- cies. So far as we have been able to deter- mine, there is no reason to differ from Dali’s46 conclusion that “The specimens from the two oceans are absolutely similar, and differ no more than individuals from either sea among themselves.” Tellina schrammi Recluz, described from the island of Guade- loupe in the West Indies, appears to be iden- tical with T. crystallina. Tellina errati Pilsbry & Johnson47, de- scribed from the Miocene of Santo Domingo, is identical or a very similar form. Maury,48 1917, stated that fossils from Santo Domingo were identical with Recent shells found on the beach on that island. Tellina ( Merisca ) sancti-dominici Maury,46 also described from the Miocene of Santo Domingo, is a similar species. Tellina martensi Lynge,50 described from the East Indies, was compared to T. crystallina, but the west American species is longer with a more rostrate form. Distribution : A few specimens of this spe- cies were collected by the expedition on the beach at Tangola-Tangola, Mexico, and others were dredged in 12-13 fathoms at Corinto, Nicaragua. It also occurs from Mio- cene to Recent in the Caribbean region and has been recorded as occurring in the Plio- 45 Salisbury, A. E., Proc. Malacol. Soc. London, Vol. 21, Pt. 2, July, 1934, p. 83. 40 Dali, W. H., Proc. U. S. Nat. Mus., Vol. 23, 1900, p. 302. 41 Tellina ( Merisca ) errati Pilsbry & Johnson, Proc. Acad. Nat. Sci. Philadelphia, Vol. 69, May 5, 1917, p. 201. Santo Domingo, Oligocene [Miocene]. —Pilsbry, Proc. Acad. Nat. Sci. Philadelphia, Vol. 73, Pt. 2, 1922, p. 425, pi. 41, fig. 7. [Figure of type]. 48 Maury, C. J., Bull. Amer. Paleo., Vol. 5, No. 29, 1917, pp. 387 (223) -388 (224), pi. 64 (38), fig. 4. 49 Tellina ( Merisca ) sancti-dominici Maury, Bull. Amer. Paleo., Vol. 5, No. 29, Pt. 1, April 7, 1917, p. 388 (244), pi. 64 (38), fig. 11. “Bluff 3. Cercado de Mao.” Santo Domingo, Miocene. 50 TeUina ( Merisca ) martensi Lynge, Kyi. Dansk. Vidensk. Skr., Ser. 7, Nat. og Math., Bd. 5, 1909, p. 195, pl. 3, figs. 40-42. "South of Koh Kut, 17-20 fathoms, mud (%). Gulf of Siam. Singapore, 2-3 fathoms, coral reef; 5/2." cene of Ecuador and Pleistocene of Magda- lena Bay, Lower California. Tellina IMeriscal proclivis Hertlein & Strong, sp. nov. Plate I, Figs. 6, 7, 14. Tellina declivis Sowerby, Conch. Icon., Vol. 17, Tellina, March, 1868, species 261, pl. 44, fig. 261. “Hab.— ?”— I. S. Oldroyd, Stanford Univ. Publ. Univ. Ser. Geol. Sci., Vol. 1, 1924, p. 165 (under section Merisca). Catalina Island, California, to Panama.— Strong & Hertlein, Allan Hancock Pac. Ex- ped., Vol. 2, No. 12, 1939, p. 184. Bahia Honda, and off Taboga Island, Panama. Not Tellina declivis Conrad, Jour. Acad. Nat. Sci. Philadelphia, Ser. 1, Vol. 7, 1834, p. 131. “Locality, Yorktown, Va.” Tertiary. [Miocene]. Tellina ( Merisca ) declivis Sowerby, Dali, Proc. U. S. Nat. Mus., Vol. 23, 1900, p. 301. “Cerros Island, Lower California, to the Gulf of California.” Type Locality : Magdalena Bay, Lower California, Mexico. Range : Cedros Island, Lower California, to Panama. Collecting Stations: Mexico: 4 miles SSW. of Maldanado Point (192-D-l), 26 fathoms, mud; PortGuatulco (195-D-20), 23 fathoms, mud; Tangola-Tangola Bay (196-D-17), 23 fathoms, mud; Costa Rica: Port Parker (203-D-l, 3), 12-15 fathoms, sandy mud, crushed shell, shelly mud. Description: Shell small, subtrigonal, white, beaks subcentral and elevated; ante- rior dorsal margin sloping, anteriorly rounded and inflated, posterior dorsal mar- gin steeply sloping, the end pointed, wedge- shaped, subcompressed; a well-developed and sunken lunular area and long escutcheon present; right valve with a broad flexure, the left with a shallow radial depression fol- lowed by an angulation; sculpture of fine, regular, concentric lamellae, about 3 or 4 per millimeter on adult shells; hinge of right valve with two strong cardinals, the poste- rior one grooved, and two laterals, left valve with a strong grooved anterior and weaker posterior cardinal, the latter close to the margin, the posterior cardinal in each valve grooved, pallial sinus high in the middle then descending near to but not touching the anterior adductor impression, then bending posteriorly and for about half its length con- fluent with the pallial line. Dimensions of the type: length, 9 mm.; height, 7.8 mm.; convexity (both valves together), 4.8 mm. Holotype (Calif. Acad. Sci. Paleo. Type Coll.), from Loc. 20299 (C.A.S.), Magdalena Bay, Lower California, Mexico; Charles R. Orcutt collector. Paratype, from station 196-D-17, Lat. 15°45' N., Long. 96°05'34" W., Tangola-Tangola Bay, Mexico, dredged in 23 fathoms, mud. One of the largest valves in the present collection measures 9.3 mm. in length. A large right valve dredged about 5 miles west of Mazatlan, Mexico, by the Templeton 84 Zoological New York Zoological Society [34: 9 Ci'ocker Expedition of the California Acad- emy of Sciences in 1932, measures: length, 11.4 mm.; height, 9.8 mm.; convexity (one valve, 3 mm. The use of the combination of names, Tel- lina declivis by Conrad, 1834, makes it nec- essary to pi'opose a new name which is based on a new type specimen for the west Amer- ican shell described under that name by Sow- erby in 1868 which is here named Tellina proclivis. The shell of Tellina proclivis differs from that of T. reclusa in the more steeply sloping dorsal margins, in lacking radial sculpture, in that the pallial sinus does not touch the anterior adductor impression and along the base it is confluent with the pallial line for only about one-half its length. It differs from T. meropsis in the more steeply sloping dorsal margins, more pointed posterior end and in the narrower, radially depressed area posteriorly. Distribution : Specimens of this species, mostly single valves, were dredged by the expedition off western Mexico and Costa Rica in 12-26 fathoms, mostly on a mud bot- tom. Dali51 cited this species as occurring north to Catalina Island, California, but Burch52 indicated that it is questionable whether it is a member of the fauna of Cali- fornia. Tellina IMeriscal reclusa Dali. Tellina ( Merisca ) reclusa Dali, Proc. U.S. Nat. Mus., Vol. 23, No. 1210, November, 1900, pp. 301, 315, pi. 3, fig. 2. “Types.— No. 105513, U. S. N. M., from San Ignacio La- goon, Lower California, Hemphill. Also off Lower California, in lat. 30° 28', by the U. S. Fish Commission, at Station 3019, in 14 fathoms, Gulf of California.’’ Tellina reclusa Dali, E. K. Jordan, Contrib. Dept. Geol. Stanford Univ., Vol. 1, No. 4, 1936, p. 145. Magdalena Bay and San Ignacio Lagoon, Lower California, Pleistocene. Re- cent from San Ignacio Lagoon to the Gulf of California. Type Locality : San Ignacio Lagoon, Lower California. Range : San Ignacio Lagoon, Lower Cali- fornia, to the Gulf of California and south to Bahia Honda, Panama. Collecting Stations: Mexico: Arena Bank, Gulf of California (136-D-18, 21, 22), 40-45 fathoms, mud ; Santa Inez Bay, Gulf of Cal- ifornia (142-D-3; 145-D-l, 3), 4-40 fathoms, sand, weed; Port Guatulco (195-D-2, 3), 3-3.5 fathoms, sand, crushed shell, Tangola- Tangola Bay (196-D-14-15) , 5 fathoms, crushed shell; Nicaragua: Corinto (200-D- 8, 9), 6-6.6 fathoms, mangrove leaves, also on beach. Description : Shell small, moderately con- vex, subtrigonal, posteriorly wedge-shaped and flexed, the posterior end keeled dorsally; 51 Arcopagia declivis Sowerby, Dali, U. S. Nat. Mus., Bull. 112, 1921, p. 45 (under section Merisca). Catalina Island, California, to Panama. 52 Burch, J. Q., Min. Conch. Club South Calif., No. 45, sculpture of fine, close (about 5 per mm.), little elevated, sharp, concentric lamellae which are separated by wider interspaces which are faintly, radiately striated; lunu- lar area smooth, escutcheon long, narrow, deep; hinge normal for the subgenus; pallial sinus high behind then descending and usu- ally, although not always, touching the base of the anterior adductor impression and wholly confluent with the pallial line below. A specimen from Tangola-Tangola Bay, Mexico, measures: length, 20 mm.; height, 15.5 mm.; convexity (one valve), 3.6 mm. Dali stated that “this species is notable for the rasp-like quality of its surface to the touch.” Compared to Tellina proclivis the shell of T. reclusa is longer in proportion to the height, the dorsal margins slope less steeply and the interspaces are ornamented by fine radial sculpture. The shell of T. reclusa is more elongate in outline but the posterior end is less attenuated and flexed and the con- centric sculpture is more closely spaced than that of T. crystallina. The coarser sculpture and the fact that the pallial sinus usually touches the anterior adductor impression are features separating T. reclusa from T. mer- opsis Dali53. Tellina ( Merisca ) lintea Turton, which occurs in the Caribbean region from Florida to Brazil, is very similar to T. reclusa. Pris- tipagia gemonia Iredale,54 an Australian species, also is a somewhat similar shell. Distribution: This species was collected by the expeditions from Santa Inez Bay in the Gulf of California, to Corinto, Nicara- gua, on the beach and dredged at depths of 3-45 fathoms. It is also known to occur in the Pleistocene of Lower California. Subgenus Scissula Dali. Key to the species of Scissula. A. Shell large, length exceeding 25 mm. a. Posterior end obliquely truncated; white or tinged with brown cognata aa. Posterior end tapering and roundly pointed ; pale rose color nicoyana B. Shell small, length not exceeding 25 mm.; very thin, glassy a. Oblique striae widely spaced, coarse, very oblique virgo aa. Oblique striae closer, finer, less oblique and approaching lines of growth varilineata 55 5S Not represented in the present collection. Tellina IScissulal cognata C. B. Adams. Tellina cognata C. B. Adams, Ann. Lyceum Nat. Hist. New York, Vol. 5, July, 1852, pp. 503, 545 (separate pp. 279, 321). “Habitat. — Panama.” 53 Tellina ( MoereUa ) meropsis Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, November, i900, p. 317, pi. 3, fig. i. “San Diego, California.” 3- Pristipagia gemonia Iredale, Rec. Australian Mus.. Vol. 19, No. 5, April 7, 1936, p. 281, pi. 21, fig. 6. “Habitat. Hertlein & Strong: Mollusks of Mexico and Central America 85 1949] Psammobia casta Reeve, Conch. Icon., Vol. 10, Psammobia, June, 1857, species 55, pi. 8, fig. 55. “Hab. Guatemala.” Not Tellina casta Hanley, Proc. Zool. Soc. London, September, 1844, p. 63. “Hab. Singa- pore ; sandy mud.” Tellina tenuilineatus Li, Bull. Geol. Soc. , China, Vol. 9, No. 3, October, 1930, p. 262, pi. 5, fig. 33. “Brought up by marine dredge from depths varying from 10. ft. to 40. ft. in 1 the mud at the mouth of the Rio Grande near La Boca about one mile from the mainland in Panama Bay.” “Horizon : Probably Gatun formation.” According to Pilsbry this record is based upon “Two pieces of a left valve of ‘Tellina’ cognata C. B. Ad. Compared with Adams’ type” {Proc. Acad. Nat. Sci. Phila- delphia, Vol. 83, 1931, p. 431). Type Locality : Panama. Range : Mazatlan, Mexico, to Panama. Collecting Stations: Guatemala: 7 miles west of Champerico (197-D-l, 2), 14 fath- oms, mud; El Salvador: La Libertad (198- D-l, 2), 13-14 fathoms, mud; Meanguera Island, Gulf of Fonseca (199-D-l), 16 fath- oms, sand, mud, crushed shell; Nicaragua: Monypenny Point, Gulf of Fonseca (199-D- 4, 5, 6), 4-7 fathoms, mud; Costa Rica: 13 miles S. X E. of Judas Point (214-D-l), 42 fathoms, mud, shell; Panama: Gulf of Chiri- qui (221-D-1-5), 35-40 fathoms, sandy mud. Description: Shell oblong, anterior end obliquely rounded, posterior end obliquely truncated, gaping, a depressed posterior area is delimited by a rounded umbonal angula- tion, white or with a reddish or pale brown- ish tinge; sculpture consists of minute un- equal radiating striae and concentric grooved striae which cross the lines of growth obliquely and become weak or absent on the posterior third of the shell; lateral teeth obsolete; pallial sinus highest posteriorly then descending gently and extending ante- riorly for a little more than three-fourths the length of the shell but well separated from the anterior adductor impression; basally, for about one-half its length, it is confluent with the pallial line. A specimen from the Gulf of Fonseca, in the present collection, measures : length, 45.3 mm. ; height, 30 mm. ; convexity (both valves together), 13.5 mm.; pallial sinus extends forward 36.4 mm. from the posterior end of the shell. A large left valve dredged in the Gulf of Chiriqui, Panama, measures: length, 49.5 mm. ; height, 32 mm. ; convexity (one valve) , 7.5 mm. This species is unlike any other along the Pacific Coast of the Americas. It bears only a general resemblance to the east American species T. similis Sowerby. The present species appears to be referable to the subgenus Scissula Dali, 1900, type Tellina decora Say, rather than to other supraspecific groups in which the shell bears oblique striations such as Scissulina Dali, 1924, type, T. dispar Conrad; Jactellina Iredale, 1929, type, T. obliquaria Deshayes; Obtellina Iredale, 1929, type, T. bougei Sowerby; and Loxoglypta Dali, Bartsch & Rehder, 1938, type, T. obliquilineata Conrad. Distribution: This species was dredged by the expedition from Guatemala to Panama in 4-42 fathoms. It occurred rather abund- antly off Champerico, Guatemala, in 14 fath- oms, and in the Gulf of Chiriqui, Panama, in 35-40 fathoms, mostly on a muddy bottom. It also has been recorded as occurring in the Pliocene of Ecuador. Tellina IScissula) nicoyana Hertlein & Strong, sp. nov. Plate I, Figs. 23, 24, 25, 26. Shell elongately ovate, thin, gently inflated, nearly equilateral, color, pale rose; anterior dorsal margin slightly rounded, gently slop- ing anteriorly then rounding abruptly to the elliptically rounded anterior end which merges into the broadly rounded ventral margin; posterior end tapering, roundly pointed; the posterior dorsal margin slopes gently from the beaks with a convexity in the middle portion ; the ornamentation con- sists of fine concentric lines of growth ; these are crossed by oblique striae which begin on the anterior end and continue for about two- thirds the length of the shell but are absent on the posterior end where there are a few vague submicroscopic radiating striae; hinge of right valve with two grooved cardinals and, close by, an anterior lateral, and distally a posterior socket below which is a weak lateral; left valve with a grooved anterior cardinal and a thin lamella-like posterior cardinal, also a short anterior lateral or pro- jection of the nymph and a faint posterior lateral; pallial sinus extending about four- fifths the length of the shell and separated from the anterior adductor impression by a considerable distance, high and rounded beneath the beaks then sloping and broadly undulating downward then turning abruptly downward to the pallial line with which it is confluent below; interior beautifully pale yellowish-rose and with faint submicroscopic radiating striae. Length, 34.4 mm. ; height, 19 mm.; convexity (both valves together), 7.8 mm. ; pallial sinus extends forward about 27 mm. from the posterior end of the shell. Holotype and paratypes (Calif. Acad. Sci. Paleo. Type Coll.), dredged in Ballena Bay, Gulf of Nicoya, Costa Rica; also dredged in the same vicinity at Station 213-D-ll, 17, in Lat. 9° 44' 52" to 9° 42' 00" N., Long. 84° 51' 25" to 84° 56' 00" W., in 35 fathoms, mud. This beautiful species appears to be dis- tinct from any described shell. It apparently has some characters in common with Tellina delicatula Deshayes36, the type of which has never been illustrated, but differs in that the shell is nearly equilateral, rather than strongly inequilateral, and the hinge is quite different from that described by Deshayes. It also lacks the dark irregular lines crossing the oblique striae on that species mentioned 5tJ Tellina delicatula Deshayes, Proc. Zool. Soc. London for 1854 (issued May 16, 1855), p. 363. “Hab. Mazatlan. Coll. Cuming.”— Bertin, Nouv. Arch. Mus. Hist. Nat. (Paris), Ser. 2, Vol. 1, 1878, p. 290. Coast of Mazatlan. 86 Zoologica: New York Zoological Society [34:9 by Carpenter57. It is interesting to note that Maury has described Tellina ( Scissula ) cercadica 58 from the Miocene of Santo Do- mingo, which species, she stated, is very similar to a Recent shell from Panama Bay in the Newcomb collection which was labeled Tellina delicatula Deshayes. The Miocene shell illustrated by Maury is not at all similar to the present species but is more similar to Tellina ( Scissula ) cognata C. B. Adams from Panama. Tellina IScissulal virgo Hanley. Tellina virgo Hanley, Proc. Zool. Soc. Lon- don, December, 1844, p. 143. “Hab. — ? Mus. Cuming.” “Allied to the Iris of Say, but much larger.” — Hanley, Thes. Conch., Vol 1, 1846, p. 284, pi. 57, fig. 42. “Chiriqui, West Columbia.” — Sowerby, Conch. Icon., Vol. 17, Tellina, 1867, species 207, pi. 37, fig. 207. “Hab. Chiriqui, West Indies.” — Pilsbry & Olsson, Nautilus, Vol. 56, No. 3, January, 1943, p. 79 (in text), pi. 8, fig. 5. Panama and west coast of northern South America to Puerto Pizarro, Peru. Tellina ( Fabulina ) virgo Hanley, Salis- bury, Proc. Malacol. Soc. London, Vol. 21, Pt. 2, July, 1934, p. 91, pi. 13, figs. 5 and 6. [Illus- trations of holotype and paratype]. Type Locality: Chiriqui, west Panama (here designated as type locality). No lo- cality cited originally. Range: Magdalena Bay to the Gulf of California and south to Puerto Pizarro, Peru. Collecting Stations: Nicaragua: Corinto (200-D-11-13, 15), 1-8 fathoms, mangrove leaves, also on beach. Description: Shell small, ovately elongate, very thin, glassy, transparent, obtusely an- gulated posteriorly, colored pink or white; sculpture consists of fine striae which cross the shell obliquely but are absent on the pos- terior area; the pallial sinus projects ante- riorly about four-fifths the length of the shell but does not quite touch the anterior adductor impression, along the base it is con- fluent with the pallial line. One of the largest valves collected on the beach at Corinto, Nicaragua, measures: length, 20.2 mm.; height, 12 mm.; convexity (one valve), 2 mm. The spacing of the oblique lines on this shell seems to vary somewhat but on the specimens which we have observed these lines are always more widely spaced and cross the shell at a greater inclination than those of the similar Tellina ( Scissula ) varilineata Pilsbry & Olsson59. Distribution: This species was collected by the expedition only at Corinto, Nicaragua, on the beach and dredged in 1 to 8 fathoms. 57 Carpenter, P. P., Cat. Mazatlan Shells, September, 1855, p. 37. 58 Tellina ( Scissula ) cercadica Maury, Bull. Amer. Paleo., Vol. 5, No. 29, Pt. 1, April 7, 1917, p. 388 (224), pL 64 (38), fig. 9. "Bluff 3, Cercado de Mao.” Santo Domingo, Miocene. 59 Tellina ( Scissula ) varilineata Pilsbry & Olsson, Nau- tilus, Vol. 56, No. 3, January, 1943, p. 79, pi. 8, fig. 6. Type, "from Puerto Bucaro, Province of Los Santos, Panama.” Also ranges south to Puerto Pizarro, Province of Tumbez, Peru. Subgenus Phyllodina Dali. Tellina IPhyllodinal pristiphora Dali. Tellina ( Phyllodina ) pristiphora Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, No- vember, 1900, pp. 302, 316, pi. 4, fig. 14. “Dredged near La Paz, Lower California, in 26V? fathoms.” Tellina pristiphora Dali, Pilsbry & Lowe, Proc. Acad. Nat. Sci. Philadelphia, Vol. 84, 1932, p. 132. Dredged in 20 fathoms, Man- zanillo; Acapulco, Mexico. Type Locality: Near La Paz, Lower Cali- fornia, in 26V2 fathoms. Range : Santa Inez Bay, Gulf of California, to Puntarenas, Costa Rica. Collecting Stations: Mexico: Arena Bank, Gulf of California (136-D-4, 7, 9, 15, 18, 20- 22, 28, 32), 40-85 fathoms, mud, crushed shell, muddy sand, sand; Santa Inez Bay (146-D-l), 35 fathoms, mud, crushed shell; Manzanillo (184-D-2), 30 fathoms, gravelly sand ; Port Guatulco ( 195-D-21 ) , 18 fathoms, mud; Costa Rica: Port Parker (203-D-3), 12 fathoms, shelly sand. Description: Shell of medium size, ovate, compressed, nearly equilateral, beaks low, anterior end rounded, posterior dorsal mar- gin sloping, the end blunt, surface yellowish- white, somewhat chalky; sculpture consists of evenly spaced concentric lamellae which are separated by wider and finely concen- trically striated interspaces, the lamellae are somewhat obsolete medially; on the pos- terior area, which on the right valve is set off by an angulation and on the left valve by a groove, the lamellae develop small squarish or rounded elevated foliations, along the anterior dorsal margin similar foliations present a serrated character; lunule and escutcheon elongate; two cardinal teeth in each valve, the right posterior and left an- terior ones grooved, two large laterals in the right valve and two small ones in the left, the anterior ones closest to the cardinals; pallial sinus highest behind, then tapering, the end pointed, extending toward the an- terior adductor impression for more than one-half the length of the shell, free and ascending from the pallial line except for a short distance posteriorly; interior white and in large specimens yellowish which be- comes pale salmon in the anterior dorsal area. A large specimen dredged on Arena Bank in the southern portion of the Gulf of Cali- fornia measures: length, 35.8 mm.; height, 23 mm.; convexity (both valves together), 10.2 mm. ; pallial sinus extends forward 21.5 mm. from the posterior end of the shell. Tellina fluctigera Dali60, described from the Gulf of Panama, the type of which has not been illustrated, appears to be a very similar form. Tellina dodona Dali and T. leptalea Gard- ner are representative of the subgenus Phyl- lodina in the Miocene of Florida. 60 Tellina l Phyllodina) fluctigera Dali, Bull. Mus. Comp. Zool., Vol. 43, No. 6, October. 1908, p. 419. Dredged "in the Gulf of Panama, in 182 fathoms, mud, bottom tem- perature 54°.l F.” 1949] Hertlein & Strong: Mollusks of Mexico and Central America 87 Distribution : Specimens of Tellina pristi- phora were dredged from Santa Inez Bay, in the Gulf of California, to Port Parker, Costa Rica, in 12-85 fathoms. The present record of occurrence at Costa Rica is an extension south of the known range of this species. Subgenus Phyllodella Hertlein & Strong, subgen. nov. Shell elongate, compressed, moderately thin, equilateral, both sides sloping nearly equally, anterior end rounded, posterior end roundly obliquely truncated, a posterior area is set off by a weak posterior umbonal angu- lation; sculpture consists of fine close con- centric threads which are crossed by fine radial striae; on the posterior area the con- centric sculpture becomes squamose with fine plate-like scales; right valve with two grooved cardinals, a close anterior lateral and a small distant posterior lateral; left valve with a grooved anterior cardinal, a thin posterior cardinal and a weak anterior lat- eral, no posterior lateral ; pallial sinus long, almost touching the anterior adductor im- pression, along the base wholly confluent with the pallial line. The character of the hinge, the pallial sinus, and concentric sculpture anterior to the posterior angulation in this subgenus are similar to those of Eurytellina. The character of the posterior area, ornamented with squa- mose foliations, resembles that of Phyllodina Dali. Tellina I Phyllodella I insculpta Hanley. Tellina insculpta Hanley, Proc. Zool. Soc. London, September, 1844, p. 70. “Hab. Chiri- qui, West Columbia; sandy mud, three fath- oms.” — Hanley, Thes. Conch. Vol. 1, 1846, p. 289, pi. 60, fig. 136. “Chiriqui, W. Colum- bia.” — Sowerby, Conch. Icon., Vol. 17, Tell- ina, 1867, species 208, pi. 37, fig. 208. “Hab. Chiriqui, West Columbia.” Type Locality. Chiriqui, west Panama, in three fathoms, sandy mud. Range : Champerico, Guatemala, to Santa Elena Bay, Ecuador. Collecting Stations: Guatemala: 7 miles west of Champerico (197-D-2), 14 fathoms, mud ; El Salvador: La Libertad (198-D-l, 2), 13-14 fathoms, mud. Description: Shell elongate, compressed, thin, white, equilateral, dorsal margins slop- ing, anterior end rounded, posterior end an- gulated; sculpture of fine, close, equidistant, concentric ribs (about 4 per mm.), both ribs and interspaces are crossed by fine submicro- scopic radiating striae; posterior area set off by a weak angulation and on this area the surface is roughened by small scales or inter- rupted delicate lamellae; right valve with two grooved cardinals and an anterior lateral so closely situated near the beaks as to re- semble a cardinal, and a small posterior lat- eral ; left valve with a grooved anterior car- dinal, a thin posterior cardinal and a faint anterior lateral; pallial sinus rather high behind then descending, end blunt and al- most, but not quite, touching the posterior basal margin of the anterior adductor im- pression, along the base it is confluent with the pallial line. The largest specimen in the collection measures: length, 33.6 mm.; height, 18.3 mm.; convexity (both valves together), 5.8 mm. The only other shell described from west American waters that bears much resem- blance to this species appears to be Tellina ( Phyllodina ) fluctigera Dali, the type of which has not been illustrated. According to Dali’s description the concentric sculpture of his species is more widely spaced (about 2 lamellae per mm. whereas there are about 4 per mm. in the present species) and the pallial sinus is entirely free from the pallial line rather than confluent with it. Distribution: Specimens of Tellina in- sculpta were dredged by the expedition west of Champerico, Guatemala, in 14 fathoms, and at La Libertad, El Salvador, in 13 fath- oms, on a muddy bottom. A small specimen of this species in the collections of the Califor- nia Academy of Sciences was collected by Woodbridge Williams in Santa Elena Bay, Ecuador. These records extend the known range of the species both to the north and to the south. Subgenus Elliptotellina Cossmann. Tellina lElliptotellinal pacihca Dali. Tellina ( Elliptotellina ) pacifica Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, November, 1900, pp. 302, 316, pi. 3, fig. 9. . . . “dredged in Panama Bay, in 18 fathoms, sand.” Type Locality: Panama Bay, in 18 fath- oms, sand. Range: Santa Inez Bay, in the Gulf of California, to the Bay of Panama. Collecting Stations: Mexico: Santa Inez Bay, Lower California, in the Gulf of Cali- fornia (145-D-1-3), 4-13 fathoms, sand; Manzanillo (184-D-2), 30 fathoms, gravelly sand; Port Guatulco (195-D-9, 19), 7-17 fathoms, gr. mud, gr. sand, crushed shell; Tangola-Tangola Bay (196-D-6, 7), 6-7 fath- oms, sand, crushed shell. Description: Shell small, oval, both ends rounded, the anterior the longer, moderately convex, yellowish-white with a rose colored spot near each end of the hinge margin ; sculpture of fine concentric grooves sepa- rated by wider interspaces, these on the pos- terior half (and sometimes faintly to the an- terior third) of the shell are crossed by deep, angular, radial grooves which serrate the posterior ventral margin and between which are narrow interspaces ; hinge of right valve with two cardinals, the anterior one a large subtriangular mass and the posterior one smaller and faintly grooved, two laterals present; left valve with two cardinals, the anterior one grooved, the laterals obsolete; pallial sinus ascending, extending anteriorly about one-half or more the length of the shell, rounded at the end and, except for a short 88 Zoologica : New York Zoological Society [34: 9 distance posteriorly, free from the pallial line. A specimen from Tangola-Tangola Bay, Mexico, measures approximately : length, 7.8 mm. ; height, 4.4 mm. ; convexity (both valves together) , 2.5 mm. A left valve from Port Guatulco, Mexico, measures 8.2 mm. in length. This species differs from Tellina ameri- cana Dali, 1900, a similar east American species, in that the radial sculpture is strong- er and is present farther anteriorly, the pallial sinus is longer and the color is said to be somewhat brighter. Tellina cymobia WoodTing''1', described from the Miocene of Jamaica, is a very simi- lar species. Distribution-. The discovery of the occur- rence of Tellina pacifica in Santa Inez Bay in the Gulf of California is a long extension north of the known range of the species. Genus Tellidora Morch in H. & A. Adams. Tellidora burneti Broderip & Sowerby. Tellina burneti Broderip & Sowerby, Zool. Jour., Vol. 4, No. 15, January, 1829, p. 362, pi. 9, fig. 2. “Hab. ad Mazatlan, in Aestuario.” “Found in the Estuary of Mazatlan, among the shoals of large Pinnae which are left dry at low water.” — Hanley, Thes. Conch., Vol. 1, 1846, p. 271, pi. 58, fig. 99. “Salango, W. Columbia.” — Sowerby, Conch. Icon., Vol. 17, Tellina, 1867, species 199, pi. 35, figs. 199a, 199b. West Colombia. — M. Smith, Panamic Shells (Trop. Photogr. Lab., Winter Park, Florida), 1944, p. 65, fig. 843. Lower Cali- fornia to Ecuador. Type Locality: Mazatlan, Mexico, in the estuary among shoals of large Pinnae at low water. Range: Soledad, Lower California, to the Gulf of California and south to Salango, Ecuador. Collecting Stations: Mexico: Port Gua- tulco (195-D-17), 6 fathoms, sand; El Sal- vador: Meanguera Island, Gulf of Fonseca (199-D-l), 16 fathoms, sand, mud, crushed shell; Nicaragua: Corinto, beach. Description: Triangular, rather thin, in- equivalve, subequilateral, compressed, sub- nacreous, white ; the convex valve with obso- lete distant dilated concentric grooves, which are most visible in front and towards the elevated acute and curving beaks (one of which projects over the other) ; the flat valve with the elevated fine rather irregular and close concentric striae, which become obso- lete posteriorly ; ventral edge arcuated ; dor- sal edges strongly sloping on either side, and armed with large tooth-like projections, the front one greatly incurved, the hinder nearly straight; ligament minute; dorsal slopes ex- cavated; fold and flexure distinct; lateral teeth distinct, rather remote, and subequi- distant. (Hanley, Thes. Conch., Vol. 1, 1846, p. 271). 61 Tellina. ( Elli ptotellina ) cymobia Woodring, Carnegie Inst. Washington, Publ. 366, May 20, 1925, p. 174, pi. 24, figs. 14-16. Bowden, Jamaica, Miocene. The pallial sinus is rather high and pro- jects anteriorly more than half the length of the shell. A large valve in the present collection from the Gulf of Fonseca, measures: length, 49 mm. ; height, 40.5 mm. ; convexity ( one valve), approximately 3.4 mm. There is some variation in the concentric sculpture of this species but the shape and ornamentation are so characteristic that it cannot be confused with any other west American shell. Tellidora cristata Recluz62 is a similar spe- cies which occurs from Miocene to Recent in the Caribbean region. Tellidora ( Tellipiura ) peruana Olsson, 1944, has been described from the Cretaceous of Peru. Distribution: This species was collected by the expedition off west Mexico, in the Gulf of Fonseca off El Salvador and at Co- rinto, Nicaragua. It also has been recorded as occurring in the Pleistocene at Magdalena Bay, Lower California, and Panama. Genus Mocomo Leach. Key to the subgenera of Macoma. A. Shell subtrigonal, beaks subcentral a. Shell ornamented with slightly oblique corrugations; small; thin Cymatoica aa. Shell ornamented with concentric striae only Macoma s.s. B. Shell elongate, beaks posteriorly situated ; posterior end much the shorter a. Resilium internal and shorter than the ligament Psammotreta aa. Resilium external and about as long as the ligament b. Posterior area granulated Macoploma bb. Posterior area not granulated Psammacoma Subgenus Macoma s.S. Macoma I Macoma l nasuta Conrad. Tellina nasuta Conrad, Jour. Acad. Nat. Sci. Philadelphia, Ser. 1, Vol. 7, 1837, p. 258. “Inhabits coast of California near Sta. Diego.” Macoma nasuta Conrad, I. S. Oldroyd, Stanford Univ. Publ. Univ. Ser. Geol. Sci., Vol. 1, 1924, p. 174, pi. 45, figs, la, lb, lc, Id. Kodiak Island and Cook Inlet, Alaska, to Scammon Lagoon, Lower California. Also Miocene, Pliocene and Pleistocene of Cali- fornia. — Grant & Gale, Mem. San Diego Soc. Nat. Hist., Vol. 1, 1931, p. 365, pi. 20, figs. 11a, lib. Earlier records cited. ? Oligocene and Miocene to Recent. 62 Lucina cristata Recluz, Rev. Zool., Soc. Cuvierienne , 1842, p. 270. “Hab.: — Trouvee sur le banc de Campeche i par M. Cosmao, commandant la station navale du Mexique.” —Recluz, Mag. de Zool. (par Guerin-Meneville) , 1843, Moll., | pi. 60, p. 1, figs. 1-5. Original locality cited. TeUina cristata Recluz, Sowerby, Conch. Icon., Vol. 17, Tellina, 1868, species 291, pi. 49, figs. 291a, 291b. “Hab. — V* — M. Smith, East Coast Mar. Shells (Edwards Bros., Ann Arbor, Michigan), 1937, p. 60, pi. 54, figs. 6a, 6b. West Florida to Trinidad, West Indies. 1949] Hertlein & Strong: Mollusks of Mexico and Central America 89 Type Locality : Coast near San Diego, Cali- fornia. Range: Kodiak Island and Cook’s Inlet, Alaska, to Cape San Lucas, Lower California. Collecting Station : Mexico: Cape San Lucas, Lower California. Description: A single right valve of this well known strongly flexed shell, measuring 81 mm. in length and 42.8 mm. in height, was taken by the expedition at Cape San Lucas, Lower California. The present specimen might be referable to Macoma kelseyi but it appears to be inter- mediate in characters between that form and M. nasuta. Dali described Macoma kelseyi™ from the “Pleistocene” of San Diego, Cali- fornia. It was said to differ from M. nasuta by its greater size, thicker, flatter shell and in that the pallial sinus in the right valve bends pos- teriorly before coalescing with the pallial line rather than joining it at a right angle. Study of a series of specimens of Macoma nasuta and of M. kelseyi reveals that there is some variation in the pallial sinus and it may be open to question whether Macoma kelseyi is a distinct species, subspecies, or merely a very large M. nasuta. Distribution: The discovery of the occur- rence of Macoma nasuta at Cape San Lucas, Lower California, is an extension south of the known range of the species. It also is known to occur as a fossil in California from Oligocene or Miocene to Recent. Subgenus Cymatoica Dali. Macoma t Cymatoica) undulata Hanley. Tellina undulata Hanley, Proc. Zool. Soc. London, September, 1844, p. 72. “Hab. St. Elena, West Columbia; sandy mud, six fath- oms.” — Hanley, Thes. Conch., Vol. 1, 1846, p. 310, pi. 59, figs. 107, 107*. “St. Elena, West Columbia.” — Sowerby, Conch. Icon., Vol. 17, Tellina , 1867, species 119, pi. 23, figs. 119a, 119b. “St. Elena, West Columbia.” Cymatoica occidentalis Dali, Proc. U. S. Nat. Mus., Vol. 12, No. 773, 1889 (issued March 7, 1890) , p. 272, pi. 10, fig. 11. Dredged “ ... in latitude 24° 18' N., longitude 110° 22' W., off the coast of Lower California, in 26V2 fathoms, fine sandy mud.” Tellina ( Cymatoica ) undulata Hanley, Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, 1900, p. 309. “Gulf of California, south to St. Elena, West Colombia.” Type Locality: Santa Elena, Ecuador, in 6 fathoms, sandy mud. Range : Off the west coast of Lower Cali- fornia in Lat. 24° 18' N., Long. 110° 22' W., to the Gulf of California and south to Santa Elena, Ecuador. Collecting Stations: Mexico: Santa Inez Bay, Gulf of California (145-D-l, 3), 4-13 fathoms, sand; Port Guatulco (195-D-19, 63 Macoma kelseyi Dali, Trans. Wagner Free Inst. Sci., Vol. 3. Pt. 5, December, 1900, p. 1052, pi. 49, fig. 7. “Pleis- tocene of San Diego, California, obtained in the City Park by Dr. R. E. C. Stearns.” According to Dr. W. P. Wood- ring (oral communication ) , the type of this species came from Pliocene beds. 20), 17-21 fathoms, gr. mud, crushed shell, mud; Santa Cruz Bay (195-D-21), 18 fath- oms, mud; Tangola-Tangola Bay (196-D-6, 7), 6-7 fathoms, sand, crushed shell; El Sal- vador: Meanguera Island, Gulf of Fonseca (199-D-l), 16 fathoms, sand, mud, crushed shell; Costa Rica: Port Parker (203-D-3), 12 fathoms, shelly mud. Description: Shell small, thin, oblong, beaks subcentral, bluntly pointed and flexed to the right posteriorly, ornamented with small, rounded, undulating riblets which do not coincide with the incremental lines of growth except partially so on the posterior fold. A right valve from the Gulf of Fonseca, measures: length, 16.4 mm.; height, 9 mm.; convexity (one valve), 2.6 mm. Macoma orientalis Dali04, described from the Antilles, is a very similar species. Distribution: Specimens of this species were dredged from the Gulf of California to Costa Rica, in 4 to 21 fathoms. Subgenus Psammacoma Dali. Key to the species of Psammacoma. A. Shell moderately thick; telliniform lamproleuca B. Shell thin; sometimes iridescent a. Posterior end narrow; length usually not exceeding 35 mm. b. Very narrow and thin panamensis bb. Higher and thicker spectri aa. Posterior end wide; length usually exceeding 35 mm elongata Macoma (Psammacoma) elongata Hanley. Tellina elongata Hanley, Proc. Zool. Soc. London, December, 1844, p. 144. “Hab. Chi- quiqui [Chiriqui], West Columbia; in sand at three fathoms.’-Hanley, Thes. Conch., Vol. 1, 1846, p. 302, pi. 62, fig. 199. “Chiriqui, West Columbia.”— Sowerby, Conch. Icon., Vol. 17, Tellina, 1867, species 137, pi. 25, fig. 137. “Hab. Chiriqui, West Columbia.” Type Locality: Chiriqui, Panama, in 3 fathoms, sand. Range : Lower California (Lat. 30° 36' N.) to Panama (Dali). Caribbean region (Daut- zenberg) . Collecting Stations: Mexico: Santa Cruz Bay (195-D-21), 18 fathoms, mud; Acapulco (189-D-4), 28 fathoms, mud; El Salvador: Meanguera Island, Gulf of Fonseca (199-D-l), 16 fathoms, sand, mud, crushed shell; La Union (199-D-8-10, 12-14, 22), 3-6 fathoms, mud, mangrove leaves; Nicaragua: Monypenny Point (199-D-2-6) , 4-7 fathoms, mud; Costa Rica: Cedro Island, Gulf of Nicoya (213-D-l, 10), 8-10 fathoms, mud; 14 miles S. X E. of Judas Point (214-D-l, 4), 42-61 fathoms, mud, shell, rocks. Description : Shell elongate, thin, smooth, 64 Cymatoica orientalis Dali, Proc. U. S. Nat. Mas., Vol. 12, 1889, No. 773, (issued March 7, 1890), p. 273, pi. 10, fig. 12. "Hab.-Samana Bay, Santo Domingo, in 16 fathoms! mud .... Also found at the same depth at Cardenas, Cuba. 90 Zoologica: New York Zoological Society [34:9 white, sometimes iridescent outside, white within ; anterior end rounded, posterior end produced and bluntly truncated; a subme- dian radial, depressed area present on the ventral half of the shell; ornamented with fine concentric lines of growth which are coarser along the posterior dorsal area, and along the posterior umbonal ridge there are faint irregular oblique striations that do not coincide with the lines of growth ; hinge with two cardinals in each valve (on a large speci- men the right anterior one sometimes some- what roughened anteriorly), the posterior tooth slightly cleft, the left anterior cardinal slightly cleft, sometimes slightly indented at the base, on some specimens there are grooves along the margin back of the pos- terior tooth; pallial sinus higher behind and rounded in front, projecting forward about two-thirds the length of the shell and along the base for about half its length confluent with the pallial line. A specimen dredged southwest of Maldan- ado Point, Mexico, measures approximately: length, 47.4 mm.; height, 25.5 mm.; convex- ity (both valves together), 13 mm.; pallial sinus extends forward 30 mm. from the pos- terior end of shell. Tellina lamproleuca Pilsbry & Lowe is a somewhat similar species but the median de- pressed area is less developed, the shell is thicker and more telliniform and the poste- rior end is generally narrower. Macoma gatunensis Toula, 1908, described from the Miocene Gatun beds at Panama, is very similar but is more excavated beneath the beaks. The shell illustrated by Olsson65 under the name of Macoma guatunensis Toula appears to be very similar to the pres- ent species; in fact Olsson mentioned that he could detect no differences between the fossil form and Recent specimens of M. elon- gata. Macoma falconensis H. K. Hodson, de- scribed from the Miocene of Venezuela, also is a somewhat similar form as is Macoma hosfordensis Mansfield66 which was de- scribed from the upper Miocene of Florida. The posterior area of Macoma elongata is smooth in comparison to somewhat similar species assigned to Macoploma Pilsbry & Olsson in which the posterior area is orna- mented with granules. Distribution : Specimens of Macoma elon- gata were dredged by the expedition from off western Mexico to Costa Rica, in 3 to 61 fathoms, mostly on a muddy bottom. Dautzenberg67 cited this species as occurring at the island of St. Lucie in the West Indies and in the Gulf of Paria and Gulf of Mara- caibo, Venezuela. We have not seen specimens from east American waters. 65 Olsson, A. A., Bull. Amer. Paleo., Vol. 5, No. 39, Pt. 2, June 21, 1922, pi. 29 (26), fig. 13. Banana River, Costa Rica. Miocene. Macoma I Psammacoma! lamproleuca Pilsbry & Lowe. leen ill Tellina lamproleuca Pilsbry & Lowe, Proc. n, Acad. Nat. Sci. Philadelphia, Vol. 84, May 21, 1932, p. 93, pi. 11, figs. 6 and 7. “Corinto, Nicaragua.” Also Panama. Macoma parthenopa Pilsbry & Lowe, Proc. ■' Acad. Nat. Sci. Philadelphia, Vol. 84, May 21, 1932, p. 144, pi. 11, figs. 6 and 7. “Corinto,” , type. Also Panama. Macoma lamproleuca Pilsbry & Lowe, Pils- bry & Olsson, Proc. Acad. Nat. Sci. Philadel- phia, Vol. 93, 1941, p. 69. Jama formation, Puerto Jama, and Canoa formation, Punta Blanca, Ecuador, Pliocene. Also Recent from Panama to Zorritos, Peru. Type Locality: Corinto, Nicaragua. Range : Santa Inez Bay, Gulf of Calif or- , nia, to Zorritos, Peru. | jj Collecting Stations: Mexico: Santa Inez [j Bay, Gulf of California (142-D-4), 40-50 p0! fathoms, sand; Guatemala: 7 miles west of m, Champerico (197-D-l, 2), 14 fathoms, mud; p.] El Salvador : La Libertad (198-D-l, 2) , 13-14 pa, fathoms, mud; Meanguera Island (199-D-l), in 16 fathoms, sand, mud, crushed shell. Description: Shell elongate, telliniform, I |j fairly thick, white under a buff perio- ja stracum, somewhat inequilateral, the ante- ,: rior end the longer, dorsal margins sloping j .j and nearly straight, anterior end broadly rounded, posterior end bluntly truncated; ; surface ornamented with irregular lines of growth and wrinkles, stronger on the poste- j rior area; sometimes in the right sometimes in the left valve the posterior area bears a median radial depression; a medial flatten- ing from the umbos to the ventral margin often present; lunular area lanceolate and concave; hinge with two cardinals in each valve, the right posterior and left anterior ones bifid, no laterals present; the pallial sinus extends about four-sevenths the length of the shell, it is usually higher in the middle, rounded at the end and joins the pallial line at an acute angle and for a little over half its length is confluent with the pallial line. The largest specimen in the present col- lection, a left valve, measures : length, 72.8 !> mm.; height, 40.5 mm.; convexity (one valve), 10.5 mm.; pallial sinus extends for- ward 41 mm. from posterior end of shell. The shell of this species is similar to that of Macoma elongata but the dorsal margins i slope more steeply, especially anteriorly, the posterior extremity is usually narrower, the shell is thicker and the hinge is heavier. Macoma falconensis H. K. Hodson68, de- scribed from the Miocene of Venezuela, is a similar species. Distribution: This species was taken off Guatemala and El Salvador in 13 to 16 fath- oms, usually on a muddy bottom. It also has 66 Macoma ( Psammacoma ) hosfordensis Mansfield, Flor- ida Geol. Surv.. Bull. 8, 1932, p. 142, pi. 30, figs. 6, 10, 12, 13. “Type locality: Station 3671, 2 miles north of Hosford, Liberty County, Fla.” Choctawhatchee, upper Miocene. 67 Dautzenberg P., Mem. Zool. Soc. France, Vol 13, 1900, p. 263. 68 Macoma ( Psammacoma ) falconensis H. K. Hodson, Bull. Amer. Paleo., Vol. 16, No. 59, October t, 1931, p. 16, pi. 6, figs. 1, 6, 7. “Holotype Locality—. 5 kilometers north and 350 meters west of Urumaco, in Rio Codore, District of Democracia, Falcon.” Venezuela. Miocene. Also other locali- ties. 1949] Hertlein & Strong: Mollusks of Mexico and Central America 91 been recorded as occurring in the Pliocene of Ecuador. Macoma f Psammacoma! panamensis Dali. Macoma ( Psammacoma ) extenuata var.? panamensis Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, November, 1900, p. 310. “Pan- ama.” Macoma ( Psammacoma ) panamensis Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, November, 1900, p. 324, pi. 4, fig. 3. . . . “dredged in 33 fathoms, sand, in Panama Bay.” Type Locality : Panama Bay, in 33 fath- oms, sand. Range : Gulf of California to Panama. Collecting Stations: Mexico: Tenacatita Bay (183-D-3), 40 fathoms, sandy mud; 4 miles SSW. of Maldanado Point (192-D-3), 38 fathoms, mud; Tangola-Tangola Bay (196-D-18), 30 fathoms, mud; Costa Rica: Port Parker (203-D-3), 12 fathoms, shelly mud; 14 miles S. X E. of Judas Point (214- D-l, 4,), 42-61 fathoms, mud, shell, rocks; Panama: Gulf of Chiriqui (221-D-l, 5), 35- 40 fathoms, sandy mud. Description : Shell very elongated, slender, thin, moderately convex, right valve slightly flattened posterior to the center, inequilat- eral, the anterior end the longer, posterior end produced and subrostrate, yellowish- white ; ornamented with fine concentric striae which are heavier on the posterior area; hinge with two teeth in each valve, the right posterior and left anterior ones cleft; pallial sinus rounded at the anterior end which projects forward about five- eights the length of the shell, and for about one-half its length confluent with the pallial line; interior of valves sometimes show ob- scure striations near the margins. A right valve from the Gulf of Chiriqui, Panama, measures: length, 31 mm.; height, 14.4 mm.; convexity (one valve), 3.5 mm.; pallial sinus extends forward 20 mm. from the posterior end of the valve. Macoma extenuata Dali,69 described from the Gulf of Mexico, is a very similar species. Macoma panamensis canalis Olsson, de- scribed from the Miocene of the Canal Zone, is higher in proportion to the length. Compared to Macoma elongata Hanley, the shell of M. panamensis is generally nar- rower and more elongate and the posterior end is narrower. Distribution: This species was dredged by the expedition from off Mexico to the Gulf of Chiriqui, Panama, in 12-61 fathoms, mostly on a muddy or sandy mud bottom. Macoma I Psammacoma I panamensis spectri Hertlein & Strong subsp. nov. Plate I, Figs. 9, 10, 16. Shell elongate, fairly thick, white, some- 69 Macoma ( Psammacoma ) extenuata Dali, Proc. XJ. S. Nat. Mus., Vol. 23, No. 1210, November, 1900, p. 314, pi. 2, fig. 7. Dredged “between the delta of the Mississippi and Cedar Keys, Florida, in 32 fathoms, sand.” Cited on p. 300 as “Macoma (Cydippina) extenuata.” what flexed posteriorly ; left valve moderately inflated, the right flattened a little posterior to the middle ; posterior dorsal margin slop- ing, anterior end the longer, the dorsal mar- gin slightly convex and gently sloping, that of the right overlapping the left for about a third of its length ; anterior end well rounded, ventral margin only slightly curved, posterior end produced, roundly truncated, on each valve a distinct angulation extending from near the beaks to the lower end of the truncation ; exterior surface showing distinct resting stages between which there are many very fine concentric striations and micro- scopic radial striae most distinct just ante- rior to the posterior angulation ; fresh speci- mens are often iridescent; interior shining white; two cardinal teeth in each valve, the right posterior and left anterior ones bifid, lateral teeth lacking; pallial sinus subangu- late above, highest at a point nearly vertically below the beaks, end elliptically rounded and extending forward about two-thirds the length of the shell and along the base for more than half its length confluent with the pallial line. Holotype, a left valve, measures : length, 34.4 mm.; height, 18.5 mm.; con- vexity, 5.3 mm. ; pallial sinus extends ante- riorly 21.5 mm. from the posterior end of the shell. Holotype, left valve, and paratype, a right valve (Calif. Acad. Sci. Paleo. Type Coll.), dredged in Santa Inez Bay, Gulf of Cali- fornia, Station 143-D-3, Lat. 26° 57' N., Long. 111° 56' W., in 35 fathoms (64 meters) , mud, crushed shell. Three additional specimens were dredged at the same locality. Other specimens were dredged in the same general area in Santa Inez Bay at Station 143-D-l, 4, in 25-29 fath- oms, mud, crushed shell, weed, sand; speci- mens were dredged on Arena Bank, Gulf of California, at Station 136-D-2, 45 fathoms, mud, Area conglomerates. About a dozen single valves, probably from the Gulf of California, are without information as to locality. One valve from the Gulf of Chiriqui, Panama, Station 221-D-1-5, 35-40 fathoms, sandy mud, appears to belong to this sub- species. The growth lines of these specimens are so fine that when fresh specimens are ex- amined at an angle they act as a grating and produce beautiful iridescent spectral colors. The shell here described as a new subspe- cies is similar to that of Macoma panamensis, but it is higher in proportion to the length and so far as known it is generally restricted to a more northern range. Some young speci- mens are very similar to Dali’s species and perhaps additional specimens may show the form here described as new to be without significance, but the greater height, irides- cent color and generally more northern dis- tribution appear, at the present time, to be sufficient reasons for separating it as a dis- tinct subspecies. This new subspecies is very similar to 92 Zoological New York Zoological Society [34: 9 Macoma panamensis canalis Olsson,70 de- scribed from the Miocene Gatun beds of Panama, which also is higher in proportion to the length as compared to M. panamensis. The shell of the present subspecies is higher and a little less elongate than Olsson’s sub- species. Furthermore adult shells are mod- erately thick whereas those of Olsson’s sub- species were described as very thin and fragile. Subgenus Psammotreta Dali. Key to the species of Psammotreta. A. Umbos orange-red; shell subrectangular, narrow aurora B. Umbos white shading to ochraceous ; shell proportionately larger and higher pads Macoma (Psammotreta) aurora Hanley. Tellina aurora Hanley, Proc. Zool. Soc. London, December, 1844, p. 147. “Hab. Pan- ama ; soft sandy mud, ten fathoms : Cuming.” — Hanley, Thes. Conch., Vol. 1, 1846, p. 301, pi. 58, fig. 76. Panama. Macoma ( Psammacoma ) aurora Hanley, Salisbury, Proc. Malacol. Soc. London, Vol. 21, Pt. 2, July, 1934, p. 91, pi. 11, fig. 4. Fig- ure of lectotype. Type Locality: Panama, in 10 fathoms, soft sandy mud. Range: Gulf of California to Boca de Pan, Peru. Collecting Stations: Mexico: Santa Cruz Bay (195-D-21), 18 fathoms, mud; Tangola- Tangola Bay (196-D-14, 15), 5 fathoms, crushed shell; Nicaragua: Corinto (200-D-8, 9), 6-6.6 fathoms, mangrove leaves. Description: Shell elongate, longer ante- riorly, the end rounded, posterior end bluntly truncated, posterior area set off by an um- bonal angulation; resilium internal and partly separated from the ligament; shell white or yellowish-white with the umbonal area both exteriorly and interiorly orna- mented with orange red which grades into yellow anteriorly; two cardinals in each valve, the right posterior and left anterior ones faintly cleft; on some large right valves there is a small denticle-like projection of the nymph at the ventral end of the resilium; the pallial sinus extends forward about two- thirds the length of the shell to the anterior edge of the orange red area but is separated from the anterior adductor impression by considerable space ; along the base for about half its length it is confluent with the pallial line. A left valve of this species from Tangola- Tangola Bay, Mexico, in the present collec- tion, measures 28 mm. in length. A specimen from Panama in the collections of the Cali- fornia Academy of Sciences, measures: length, 27.5 mm.; height, 17.3 mm.; convex- ity (both valves together), 8.2 mm.; pallial TO Macoma panamensis Dali var. canalis Olsson, Bull. Amer. Paleo., Vol. 9, No. 39, Pt. 2, June 21, 1922, p. 429 ( 2B7 > , pi. 29 (26), fig. 11. “Gatun Stage; Mt. Hope, C. Z." sinus extends forward 19 mm. from the pos- 1 terior end of the shell. The elongate, subrectangular form and :< orange red radial area on the umbonal region are characteristic features of this Macoma. Macoma hesperus Dali,71 an unfigured spe- fii cies described from the Gulf of Panama, was said to bear a resemblance to M. aurora but with a narrower shell and a different hinge. Distribution: A few specimens of this spe- ^ cies were dredged in 5-18 fathoms off west Mexico and Corinto, Nicaragua. 0 Macoma ( Psammotreta ) pads Pilsbry & Lowe. 0 Macoma pads Pilsbry & Lowe, Proc. Acad. Nat. Sci. Philadelphia, Vol. 84, May 21, 1932, ' p. 95, pi. 10. figs. 1, la, 2, 3. “La Paz, Lower California.” Type Locality : La Paz, Lower California. Range : Gulf of California to Golfito, Gulf t of Dulce, Costa Rica. Collecting Stations : Mexico : Port Gua- tulco; Guatemala: 7 miles west of Champer- S ico (197-D-2) , 14 fathoms, mud ; Costa Rica : Cedro Island, Gulf of Nicoya; Golfito. Description: Shell resembling that of Ma- n coma aurora but larger and higher in pro- portion to the length, usually whiter and t tinted with yellow and on some specimens .1 the umbonal area is of an ochraceous salmon color. Two cardinal teeth in each valve, the right posterior and left anterior ones bifid, the left posterior cardinal narrow and near the posterior margin ; pallial sinus high be- neath the beaks then descending to a bluntly rounded end and for about half its length confluent with the pallial line. A specimen in the present collection from Golfito, Costa Rica, measures: length, 34.6 mm.; height, 21.8 mm.; convexity (both valves together) , 9.9 mm. A large right valve from Cedro Island in the Gulf of Nicoya, measures : length, 52.6 mm. ; height, 34 mm. ; convexity (one valve), 6.6 mm.; pallial sinus extends forward 35 mm. from the posterior end of the shell. Small specimens of this species are very similar to those of Macoma aurora but seem to differ constantly in the particular men- tioned above. As pointed out by Pilsbry & Lowe the relation of height to length varies considerably in different specimens. Illustrations of Macoma plebeia Hanley72 indicate a shell which is more roundly oval with a more rounded venti'al margin poste- riorly. According to the illustration of that species given by Pilsbry & Lowe the pallial sinus is more broadly rounded at the end and is confluent with the pallial line for a shorter distance than that of Macoma pads. Tl Macoma ( Psammacoma ) hesperus Dali, Bull. Mus. Comp. Zool., Vol. 43, No. 6, October, 1908, p. 421. Dredged in the “Guif of Panama, in 182 fathoms, mud, bottom temperature 54°. 1 F.’’ Tellina plebeia Hanley, Proc. Zool. Soc. London, De- cember, 1844, p. 147. “Hab. Real Llejos, Central America; sandy mud, seven fathoms.” —Hanley, Thes. Conch., Vol. 1, 1846, p. 299, pi. 60, fig. 151. Original locality cited. [Not pi. 59, fig. 129. “Senegal”]. Macoma plebeia Hanley, Pilsbry & Lowe, Proc. Acad. Nat. Sci. Philadelphia, Vol. 84, 1932, p. 95 (in text), pi. 10, fig. 5. “Real Llejos ( = Corinto).“ 1949] Hertlein & Strong: Mollusks of Mexico and Central America 93 Macoma mazatlanica Deshayes73 is very similar to M. pads but is more tapering pos- teriorly. Distribution : Specimens of Macoma pads were taken by the expedition off west Mexico, Guatemala and Costa Rica. This is an exten- sion south of the known range of the species. Subgenus Macoploma Pilsbry & Olsson. Macoma I Macoploma ) medioamerlcana Olsson. Macoma ( Macoploma ) medioamericana Olsson, Bull. Amer. Paleo., Vol. 27, No. 106, December 25, 1942, p. 196 (44), pi. 17 (4), fig. 8. “Pliocene. Quebrada Pehitas, Costa Rica.” Type Locality : Quebrada Penitas, Costa Rica. Pliocene. Range: Arena Bank, Gulf of California, to Panama. Collecting Stations: Arena Bank, Gulf of California (136-D-21), 45 fathoms, mud; El Salvador: La Libertad (198-D-2), 16 fath- oms, mud; Costa Rica: off Ballena Bay, Gulf of Nicoya (213-D-11-17), 35-40 fathoms, mud. Description: Shell elongate, moderately thin, general characters much like those of Macoma elongata but with the anterior dor- sal margin more steeply sloping and with the posterior area ornamented with strong concentric laminae and earthy granules. The largest specimen in the collection from the Gulf of Nicoya, measures approximately : length, 101 mm. ; height, 54 mm. ; convexity (both valves together), 24 mm. Some of the specimens in this collection agree so closely with Olsson’s description and illustration of Macoma medioamericana that we have referred them to his species. There is variation in the specimens and some might be equally well referred to Macoma ( Maco- ploma) ecuadoriana Pilsbry & Olsson74 de- scribed from the Pliocene of Ecuador. According to Olsson, M. medioamericana is proportionately longer and has coarser and more earthy granulations on the posterior submargins in comparison to M. ecuadoriana. It is obvious from a study of a series of specimens that there are several very closely related variable species, including Macoma elongata, M. panamensis, M. lamproleuca, M. ecuadoriana and M. medioamericana. The presence of granulation on the posterior areas appears to be the only certain criterion separating the species of Macoploma from some of the others. The granules are present on specimens in the present collection after they have attained a length of about 20 to 25 mm. It is very difficult to separate speci- mens of a smaller size from those of M. elongata or M. lamproleuca. ,3 Tellina mazatlanica Deshayes, Proc. Zool. Soc. London for 1854 (issued May 16, 1855), p. 359. ”Hab. Mazatlan. Coll. Cuming:.” — Sowerby, Conch. Icon., Vol. 17, Tellina, October, 1868, species 320, pi. 54, fig. 320. “Hab. Mazatlan. Coll. Cuming.” 74 Macoma ( Macoploma ) ecuadoriana Pilsbry & Olsson, Proc. Acad. Nat. Sci. Philadelphia, Vol. 93, September 9, 1941, p. 69, pi. 19, fig. 5. ‘‘Canoa formation. Punta Blanca.” Ecuador, Pliocene. Distribution: Specimens of this species were dredged by the expedition at depths of 16 to 45 fathoms from Arena Bank, Gulf of California, La Libertad, El Salvador, and off Costa Rica in the Gulf of Nicoya, mostly on a muddy bottom. The species also is known to occur in the Pliocene of Costa Rica. The present records of occurrence reveal for the first time that this species is living at the present time in west American waters. Genus Apolymeth Salisbury. Key to the species of Apolymetis. A. Posterior end broadly rounded a. Shell thick, hinge broad biangulata 73 aa. Shell thin, hinge narrow cognata B. Posterior end tapering, truncated a. Hinge fairly broad; anterior adductor impression elongately oval; pallial si- nus for a little less than half its length confluent with pallial line dornbei aa. Hinge very narrow; anterior adductor impression very long and narrow; pal- lial sinus confluent with pallial line for only a very short distance asthenodon':> Apoly metis cognata Pilsbry & Vanatta. Lutricola cognata Pilsbry & Vanatta, Proc. Washington Acad. Sci., Vol. 4, September 30, 1902, p. 556, pi. 35, fig. 5. “From Tagus Cove, Albemarle.” Galapagos Islands. Apolymetis cognata Pilsbry & Vanatta, Pilsbry & Lowe, Proc. Acad. Nat. Sci. Phila- delphia, Vol. 84, 1932, pp. 96 (in text), 133. Galapagos Islands, also Taboga Island, Pan- ama, and Corinto, Nicaragua. Type Locality : Tagus Cove, Albemarle Island, Galapagos Islands. Range : Magdalena Bay, Lower California, to Paita, Peru. Collecting Station: Nicaragua: Corinto (200-D-19) , 12-13 fathoms, mangrove leaves. Description: Shell rounded-quadrate, mod- erately compressed, bent to the right poste- riorly, rather thin, gray white. Sculptured with irregular growth wrinkles and low ra- dial striae, covered with an extremely minute secondary radial striation. Beaks median, worn at the tip. Anterior margin rounded; posterior margin subtruncate, the basal margin straightened, sinuous; pallial sinus ample. Length of left valve, 41 mm., alt. 34 mm., diameter 8 mm. (Original description) . Closely related to L. alta Conr., but higher, shorter, of a more quadrate shape. The anterior end of the pallial sinus is more rounded, and its upper margin is not sinu- ous. L. excavata Sowb. is a more wedge- shaped shell (Pilsbry & Vanatta). This species differs from the generally more northern A. biangulata Carpenter in the much narrower, weaker hinge, in the more flaring and less steeply sloping poste- rior area, and in the generally thinner shell. 75 Not represented in the present collection. 94 Zoologica : New York Zoological Society [34:9 Specimens referable to the present species from the Gulf of California and southward have in some instances, formerly been re- ferred to “Tellina” excavata Sowerby76 which was originally described without informa- tion as to locality. As pointed out by Pilsbry & Vanatta the illustration given by Sowerby indicates a shell with the posterior end wedge-shaped. Pilsbry & Lowe later pointed out that there seemed to be no characters by which “Tellina” excavata could be separated from Apolymetis dombei Hanley. They rec- ognized only four west American species of the genus, Apolymetis alta Conrad [=bian- gulata Carpenter], A. asthenodon Pilsbry & Lowe, A. cognata Pilsbry & Vanatta and A. dombei Hanley. However for some unex- plained reason, probably an oversight, they cited (p. 195) A. excavata Sowerby from Mazatlan, Mexico. In later publications, Lowe77 cited “Metis excavata Sowerby” as occurring at Punta Penasco, Mexico, in the Gulf of California, and Pilsbry & Olsson78 recorded “Apolymetis excavatus Sowerby” from the Pliocene of Ecuador. No illustra- tions were given of the shells representing those records so we are uncertain which species was represented. Specimens from the coast of the mainland appear to be identical with those from the Galapagos Islands, the type locality of A. cognata. Distribution : A single small right valve of this species, measuring about 20 mm. in length, was taken by the expedition at Corinto, Nicaragua, in 12-13 fathoms. Apolymetis dombei Hanley. Tellina dombei Hanley, Proc. Zool. Soc. London, December, 1844, p. 144. “Hab. Pan- ama ; twelve fathoms, sandy mud.”— Hanley, Thes. Conch., Vol. 1, 1846, p. 323, pi. 62, fig. 182. Panama. Variety, pi. 64, fig. 222. Tum- bez, Peru. Tellina dombeyi Hanley, Proc. Zool. Soc. London for 1844 (issued February, 1845) , p. 195 (index).— Sowerby, Conch. Icon., Vol. 17, Tellina, 1867, species 169, pi. 30, fig. 169. “Hab. Panama, Tumbez, Peru.”— Romer, Syst. Conchyl.-Cab. von Martini und Chem- nitz, Bd. 10, Abt. 4, Tellina, 1871, p. 205, pi. 39, figs. 7-9. Various localities cited from the Gulf of California to Tumbez, Peru. Psammobia sp., Li, Bull. Geol. Soc. China, Vol. 9, No. 3, 1930, p. 262, pi. 5, fig. 32. Dredged in Panama Bay in 10-40 ft. “Prob- ably Gatun formation.” Pilsbry {Proc. Acad. Nat. Sci. Philadelphia, Vol. 83, 1931, p. 431) stated that Li’s record was based on “A good but bleached valve of Apolymetis dombei ( Hanley).” Type Locality. Panama, in 12 fathoms, sandy mud. Range : Gulf of Fonseca to Tumbez, Peru. Collecting Stations: Nicaragua: Potosi 76 TeUina excavata Sowerby, Conch. Icon., Vol. 17, Tellina, March, 1867, species 138, pi. 26, fig. 138. "Hab.— ?” 77 Lowe, H. N., Trans. San Diego Soc. Nat. Hist., Vol. 8, No. 6, 1936, p. 28. 78 Pilsbry, H. A., and Olsson, A. A., Proc. Acad. Nat. Sci. Philadelphia, Vol. 93, 1941, p. 70. and 5 miles SSW. of Monypenny Point, Gulf of Fonseca, beach; Costa Rica: 1 mile S. of entrance to Golfito Bay, Gulf of Dulce, beach. Description: Shell ovately subtrigonal, moderately thick, rather smooth, white but often with umbonal area reddish-orange ex- teriorly and sometimes interiorly; the ante- rior end the longer, rounded and somewhat obliquely produced; a flexure or depressed radial area is present anterior to the poste- rior umbonal ridge; posterior end sloping, subtriangular and subtruncated, area set off by an umbonal angulation; hinge with two cardinals, the right posterior bifid, the left anterior grooved, no laterals present; the pallial sinus is high and subangulate in the middle then sloping down even with but well separated from the anterior adductor im- pression and for a little less than one-half its length confluent with the pallial line. The shell of this species is somewhat vari- able in outline. Some specimens agree almost exactly with Hanley’s original figure, others are more trigonal. A large right valve from the Bay of Panama in the collections of the California Academy of Sciences, measures : length, 66 mm. ; height, 51 mm. ; convexity (one valve), 14 mm. Romer, 1871, and Stearns, 1891, pointed out the resemblance between “Tellina” dombei and “ Tellina ” excavata Sowerby79 which was described without information as to the locality from which it came. Later Pilsbry & Lowe, 1932, stated that they knew of no difference separating these two species. Compared to Apolymetis dombei, A. asthe- nodon Pilsbry & &Lowe80 was described as possessing a more elongate shell with a very narrow hinge, a narrower, longer anterior adductor impression. Judging from the illustrations, the pallial sinus is confluent with the pallial line for a very short distance if at all. The shell of Apolymetis dombei differs from that of Macoma grandis Hanley81 which was originally described from Tumbez, Peru, and was taken at Corinto, Nicaragua, by the Templeton Crocker Expedition, 1932, in the stronger, broader, posterior umbonal fold and much longer pallial sinus. Judging from the illustration of Macoma gubemaculum Hanley,82 originally described from Real 79 TeUina excavata Sowerby, Conch. Icon., VoL 17. TeUina, March, 1867, species 138, pi. 26, fig. 138. “Hab. — T” Soot-Ryen ( Nyt. Mag. for Naturvid., Bd. 70 (Meddel. Zool. Mus. Oslo. No. 27), 1932, p. 321, pi. 2, fig. 10) illus- trated a shell under the name of Apolymetis excavata Sow- erby from Floreana (Charles) Island, Galapagos group. He remarked on the variability of the outer form revealed by different specimens, some of which were elongated, other shorter and higher. 80 Apolymetis asthenodon Pilsbry & Lowe, Proc. Acad. Nat. Sci. Philadelphia, Vol. 84, May 21, 1932, p. 96, pi. 11, figs. 1-3. "Panama, on the beach (D. E. Harrower, J. Zetek).” Type. Also at La Union, Gulf of Fonseca, El Salvador. 81 TeUina grandis Hanley, Proc. Zool. Soc. London, De- cember, 1844, p. 141. "Hab. Tumbez, Peru.” —Hanley. Thes. Conch., Vol. 1, 1846, p. 327, pi. 65, fig. 247. Tumbez, Peru. 82 TeUina gubemaculum Hanley, Proc. Zool. Soc. London, December, 1844, p. 142. “Hab. Real Llejos, Central Amer- ica ; in sandy mud, seven fathoms.” —Hanley, Thes. Conch., Vol. 1, 1846, p. 326, pi. 62, fig. 186. "Real Lejos, Central America (Cuming).” 1949] Hertlein & Strong: Mollusks of Mexico and Central America 95 Llejos, Nicaragua, the anterior dorsal mar- gins slope more abruptly than those of A. j dombei. Distribution-. Specimens of Apolymetis dombei were taken by the expedition on the beaches in the Gulf of Fonseca, Nicaragua, and Gulf of Duke, Costa Rica. We have not seen specimens from north of Nicaragua. It has been reported as ranging south to Peru. It also has been recorded as occurring in beds of upper Pliocene age in the Galapagos Islands. “Apolymetis cf. A. dombei (Hanley)” has been cited by Stewart83 as occurring in upper Pliocene beds in the Kettleman Hills, San Joaquin Valley, California. Genus Strigilla Turton. Key to the species of Strigilla. A. Shell equilateral; small, white lenticula B. Shell inequilateral, elongated posteriorly a. Striae much more widely spaced an- teriorly cicercula aa. Striae equally or more closely spaced anteriorly b. Shell thick; convex; roundly trun- cated anteriorly costulifera bb. Shell thin ; flatter ; expanded an- teriorly disjuncta Strigilla cicercula Philippi. Plate I, Fig. 19. Tellina cicercula Philippi, Zeit. f. Malako- zool., Jahrg. 3, February, 1846, p. 19. “Patria: Mazatlan.” Tellina dichotoma Philippi, Zeit. f. Mala- kozool., Jahrg. 3, February, 1846, p. 20. “Patria: Mazatlan.” Tellina ervilia Philippi, Zeit. f. Malako- zool., Jahrg. 3, February, 1846, p. 20. “Patria: Mazatlan.” Strigilla maga Morch, Malakozool. Blatter, Bd. 7, December, 1860, p. 189. “Sonsonate.” El Salvador. Strigilla interrupta Morch, Malakozool. Blatter, Bd. 7, December, 1860, p. 190. “Son- sonate.” El Salvador. Strigilla circercula Philippi, Dali, Proc. U. S. Nat. Mus., Vol. 23, No. 1210, 1900, p. 305. Gulf of California to Panama. Type Locality: Mazatlan, Mexico. Range : Gulf of California to Ecuador. Collecting Station: Nicaragua: Corinto (200-D-ll, 19, also beach), 8-13 fathoms, sand, mangrove leaves. Description : Shell small, usually less than 1 cm. in length, rounded, elongated poste- riorly, polished, white, with pink color usu- ally confined to the umbonal region; sculp- tured with fine radial striae which on the an- terior area are curved and very much more widely spaced. A large right valve of this beautiful little Apolymetis cf. A. dombei (Hanley), Stewart, U. S. Geol. Surv., Prof . Paper 195, 1940 (issued June 7, 1941), p. 93, pi. 32, fig. 2. Siphonalia zone in North Dome, Kettle- man Hills, San Joaquin Valley, California. Etchegoin formation, upper Pliocene. species, dredged off Corinto, Nicaragua, in 13 fathoms, measures: length, 9.5 mm.; height, 8.5 mm. ; convexity (one valve) , 2.5 mm. The smaller size, pink colored umbonal area, and much wider spaced sculpture on the anterior area, are characters which serve to separate the shell of Strigilla cicercula from that of the young of S. costulifera Morch. Strigilla pisiformis Linnaeus, which oc- curs in the Caribbean region, is a very simi- lar species. Distribution: This species was collected by the expedition at Corinto, Nicaragua, on the beach and dredged at depths of 8 to 13 fathoms. Strigilla costulifera Morch. Plate I, Fig. 15. Tellina carnaria Linnaeus, Hanley, Thes. Conch., Vol. 1, 1846, p. 260 (in part), pi. 56, fig. 38 [West Colombia record only]. Not Tellina carnaria Linnaeus. Caribbean region. Tellina ( Strigilla ) fucata Gould, Proc. Boston Soc. Nat. Hist., Vol. 4, November, 1851, p. 91. Pacific coast [No exact locality cited].— Gould, Boston Jour. Nat. Hist., Vol. 6, October, 1853, p. 399, pi. 16, fig. 4. “In- habits Mazatlan.” Not Tellina fucata Hinds, Zool. Voy. Sul- phur, Moll., Pt. 3, 1844, p. 67, pi. 21, fig. 4. Strigilla costulifera Morch, Malakozool. Blatter, Bd. 7, December, 1860, p. 189. “Son- sonate.” El Salvador. Tellina chroma Salisbury, Proc. Malacol. Soc. London, Vol. 21, Pt. 2, July, 1934, p. 84. New name for Tellina ( Strigilla ) fucata Gould, 1851, not Tellina fucata Hinds, 1844. Type Locality : Sonsonate, El Salvador. Range : Magdalena Bay, Lower California, to the Gulf of California and south to Ecuador. Collecting Stations: Mexico: Tenacatita Bay, beach; Sihuatanejo Bay, beach; Nica- ragua: Corinto (200-D-16), 4-7 fathoms, mangrove leaves. Description: Shell suborbicular, the ante- rior side much the shorter, moderately thick, glossy, pink, rose or white ringed with pink or carmine ; umbos smooth ; the ornamenta- tion consists of striae, anteriorly flexuous, forming an angle with those on the central area, which usually radiate obliquely poste- riorly; on the posterior slope the striae are fine and usually meet at an acute angle; a lunule present; hinge with a bifid cardinal and two laterals in each valve; one or two thickened rays often present interiorly; the pallial sinus touches the anterior adductor impression and is confluent with the pallial line below ; the interior is often of a beautiful deep red or carmine color. Dali pointed out that the shell of this spe- cies is very inconstant in details of sculpture, color, and in the presence or absence of a smooth radial streak on each valve. A very large specimen of this species from Magdalena Bay, Lower California, in the 96 Zoologica : New York Zoological Society Henry Hemphill collection of the California Academy of Sciences, measures approxi- mately: length, 25 mm.; height, 23 mm.; convexity (both valves together), 10.3 mm. The specimens in the present collection are smaller. Strigilla costulifera is very similar to the east American S. carnaria Linnaeus. The west American shell is often somewhat more rounded and the striae appear to be less numerous. Distribution: Specimens of Strigilla cos- tulifera were taken by the expedition on the beaches along western Mexico and dredged in 4 to 7 fathoms at Corinto, Nicaragua. Strigilla disjuncta Carpenter. Plate I, Fig. 20. Strigilla disjuncta Carpenter, Proc. Zool. Soc. London, November 11, 1856, p. 160. “Hab. In Sinu Panamensi ; legit. T. Bridges.” Type Locality : Bay of Panama. Range : Corinto, Nicaragua, to Panama. Collecting Station: Nicaragua: Corinto (200-D-ll), 8 fathoms, sand. Description: S. testa satis magna, alba, tenui, planata; inaequilaterali, postice pro- ducta; marginibus dorsalibus subrectis, ad angulam 120°, aliis bene arcuatis; lineis in- crementi vix monstrantibus ; lineis undulatis exillimis, antice concentricis, umbones versus ascendentibus, sinu angustiore; dein ad mar- ginem ventralem rapide descendentibus; dein subito, angulo acuto, circiter 20° postice rursus ascendentibus; lineis angularum in valva utraque haud convenientibus ; margine postico sinuato, sculptura postea fortiore; margine antico quoque sinuato; lunula dis- tincta, sinuata ; ligamento subelongato ; dent card, valva altera uno parvo et uno magno bifido; altera uno parvo bifido; dent. lat. acutioribus, haud distantibus. Long. 1.35, lat. 1.54, alt. .54 poll. (Original description). “Allied to S. sincera Hanl. ; remarkable for its large size and very fine markings, and named from the lines of markings in the two valves not agreeing at the edges.” (Car- penter) . Compared to Strigilla costulifera Morch, the shell of S. disjuncta is usually larger and the shell is thinner, flatter, more produced anteriorly, the beaks are more projecting and beneath them the dorsal margin is more con- cave. It is generally white or white tinged with pink. A large specimen collected at Panama by James Zetek measures: length, 36 mm.; height, 31 mm.; convexity (both valves to- gether) , 13.8 mm. [34:9:1949] This species has been cited from wesl American waters under the name oi Strigilla sincera Hanley84 which, according to Hedley85, is an Australian species. Salis- bury80 in 1934 stated that Strigilla sincera occurs in the Panamic area but he did not mention Hedley’s remarks on that species. Distribution: Two single valves of this species were dredged by the expedition at Corinto, Nicaragua, in 8 fathoms, and one valve was taken on shore. This is an exten- • sion noi'th of the known range of the species. Strigilla lenticula Philippi. Plate I, Fig. 21. Tellina lenticula Philippi, Zeit. f. Malako 3. zool., Jahrg. 3, February, 1846, p. 19. “Pa- tria: Mazatlan.” Strigilla serrata Morch, Malakozool. Blat- ter, Bd. 7, December, 1860, p. 189. Central t’ America. Strigilla lenticula Philippi, Dali, Proc. V. S. Nat. Mus., Vol. 23, No. 1210, 1900, p. . 305. Cape San Lucas to Central America. Type Locality: Mazatlan, Mexico. Range : Cape San Lucas, Lower California, to Corinto, Nicaragua. Collecting Stations: Nicaragua: Corinto (200-D-ll, 17, 19, also on beach), 7-13 fath- oms, sand, mangrove leaves. Description: Shell small, about 8 mm. in length, rounded, inflated, white, sculptured with well-developed, oblique, incised striae which posteriorly develop small chevron- shaped sinuations. A large right valve of this species, dredged off Corinto, Nicaragua, in 13 fathoms, meas- ures approximately: length 8 mm.; height, 7 mm. ; convexity (one valve) , 2.7 mm. The rounded form, white color, and slight development of V-shaped sinuations in the sculpture posteriorly are characters which easily serve to separate this species from others of the genus in west American waters. Strigilla flexuosa Say, an east American species, is a very similar form. Distribution: This little species was taken by the expedition only at Corinto, Nicaragua, on the beach and at depths of 7-13 fathoms. 84 Tellina sincera Hanley, Proc. Zool. Soc. London, April, 1844, p. 68. “Hab.— ? Mus. Cuming, Metcalfe.” —Hanley, Thes. Conch., Vol. 1, 1846, p. 261, pi. 60, fig. 144. [Not the record “N. W. Coast of America, (Dr. Sinclair) ,” according to Hedley]. 85 Strigilla sincera Hanley, Hedley, Proc. Linn. Soc. New South Wales, Vol. 38, Pt. 2, 1913, p. 272. Strigilla grossiana Hedley, 1908, was considered by Hedley to be a synonym of S. sincera Hanley. 86 Salisbury, A. E., Proc. Malacol. Soc. London, Vol. 21, Pt. 2, July, 1934, p. 89. Hertlein & Strong: Mollusks of Mexico and Central America 97 949] EXPLANATION OF THE PLATE. Plate I. i # "IG. 1. Tellina ( Moerella ) felix Hanley. Hy- potype, right valve, from Monypenny Point, Gulf of Fonseca, Nicaragua. Length, 17 mm.; height, 10 mm. View of exterior. P. 70. 'hG. 2. Tellina ( Moerella ) recurvata Hertlein & Strong, sp. nov. Holotype, right valve, from Loc. 23802 (C.A.S.), San Luis Gonzaga Bay, Lower California, in the Gulf of California. Length, 12 I mm., height, 7.5 mm. P. 71 Fig. 3. Tellina ( Moerella ) recurvata Hertlein & Strong, sp. nov. Holotype. View of the exterior of the specimen shown in Fig. 2. Fig. 4. Tellina ( Moerella ) recurvata Hertlein & Strong, sp. nov. Holotype, left valve. View of the exterior of the left valve of the specimen shown in Figs. 2 and 3. Fig. 5. Tellina ( Moerella ) arenica Hertlein & Strong, sp. nov. Holotype, right valve, from Station 136-D-20, Lat. 23° 30' N., Long. 109° 26' W., Arena Bank, Gulf of California, dredged in 43 fathoms (78 meters), mud. View of interior. P. 68. Fig. 6. Tellina ( Merisca ) proclivis Hertlein & Strong, sp. nov. Holotype, left valve, from Loc. 20299 (C.A.S.), Magdalena Bay, Lower California, Mexico. Length, 9 mm.; height, 7.8 mm. View of in- terior. P. 83. Fig. 7. Tellina ( Merisca ) proclivis Hertlein & Strong, sp. nov. Holotype. View of the exterior of the specimen shown in Fig. 6. Fig. 8. Tellina ( Moerella ) recurvata Hertlein & Strong, sp. nov. Paratype, right valve, from the same locality as the holotype shown in Figs. 2, 3 and 4. Length, 11.5 mm.; height, 7 mm. Fig. 9. Macoma ( Psammacoma ) panamensis spectri Hertlein & Strong, subsp. nov. Paratype, right valve, from Station 143-D-3, Lat. 26° 57' N., Long. 111° 56' W., Santa Inez Bay, Lower California, in the Gulf of California, dredged in 35 fathoms (64 meters), mud, crushed shell. Length, 33.8 mm. ; height, 18 mm. P. 91. Fig. 10. Macoma ( Psammacoma ) panamensis spectri Hertlein & Strong, subsp. nov. Paratype. View of the interior of the specimen shown in Fig. 9. riG. 11. Tellina ( Moerella ) arenica Hertlein & Strong, sp. nov. Holotype. View of the exterior of the specimen shown in Fig. 5. riG. 12. Tellina ( Tellinella ) zacae Hertlein & Strong, sp. nov. Holotype, right valve, from Station 136-D-l, Lat. 23° 29' N., Long. 109° 25' W., Arena Bank, Gulf of California, dredged in 45 fathoms (82 meters), mud. Length, 33.4 mm.; height, 15.2 mm. P. 65. *'ig. 13. Tellina ( Tellinella ) zacae Hertlein & Strong, sp. nov. Holotype, left valve. View of the interior of the left valve of the specimen shown in Fig. 12. Fig. 14. Tellina ( Merisca ) proclivis Hertlein & Strong, sp. nov. Holotype. Right valve of the specimen shown in Figs. 6 and 7. Fig. 15. Strigilla costulifera Morch. Hypotype, right valve, from Loc. 4798 (C.A.S.), Lower California; Henry Hemphill Coll. Length, 25 mm.; height, 22 mm. P. 95. Fig. 16. Macoma ( Psammacoma ) panamensis spectri Hertlein & Strong, subsp. nov. Holotype, left valve, from the same locality as the paratype shown in Figs. 9 and 10. Length, 34.4 mm.; height, 18.4 mm. P. 91. Fig. 17. Tellina ( Tellinella ) zacae Hertlein & Strong, sp. nov. Holotype. View of the exterior of the specimen shown in Fig. 13. P. 65. Fig. 18. Tellina ( Eurytellina ) inaequistriata Donovan. Hypotype, right valve, from Station 195-D-21, Lat. 15° 44' 45" N., Long. 96° 06' 55" W., Santa Cruz Bay, Mexico, dredged in 18 fathoms (33 meters), mud, crushed shell. Length, 23 mm.; height, 12.5 mm. P. 74. Fig. 19. Strigilla cicercula Philippi. Hypotype, right valve, from Station 200-D-19, Lat. 12° 28' 03" N., Long. 87° 12' 39" W., Corinto, Nicaragua, dredged in 12- 13 fathoms (22-24 meters), mangrove leaves. Length, 9.6 mm.; height, 8.5 mm. P. 95. Fig. 20. Strigilla disjuncta Carpenter. Hypo- type, right valve, from Panama; James Zetek Coll. Length, 36 mm.; height, 31.3 mm. P. 96. Fig. 21. Strigilla lenticula Philippi. Hypotype, right valve, from the same locality as the specimen shown in Fig. 19. Length, 7.9 mm.; height, 7.0 mm. P. 96. Fig. 22. Tellina ( Eurytellina ) planulata Sow- erby. Hypotype, left valve, from 1 mile south of entrance to Golfito Bay, Costa Rica. Length, 51.8 mm.; height, 29.5 mm. P. 76. Fig. 23. Tellina ( Scissula ) nicoyama Hertlein & Strong, sp. nov. Holotype, right valve, from Ballena Bay, Costa Rica. Length, 34.4 mm. ; height, 19 mm. View of hinge. P. 85. Fig. 24. Tellina ( Scissula ) nicoyana Hertlein & Strong, sp. nov. Holotype, left valve. View of hinge. Fig. 25. Tellina ( Scissula ) nicoyana Hertlein & Strong, sp. nov. Holotype, right valve. View of the exterior of the speci- men shown in Fig. 23. P. 85. Fig. 26. Tellina ( Scissula ) nicoyana Hertlein & Strong, sp. nov. Holotype, left valve. View of the exterior of the specimen shown in Fig. 24. P. 85. All the specimens illustrated on this plate are in the type collection of the Department of Paleontology of the California Academy of Sciences. - HERTLE1N & STRONG. PLATE I. MOLLUSKS FROM THE WEST COAST OF MEXICO AND CENTRAL AMERICA. Gudger : Fishes in Ranks 99 10. Fishes That Rank Themselves Like Soldiers on Parade. E. W. Gudger. American Museum of Natural History. (Plate I; Text- INTRODUCTION. I have read that some mammals, such as the American bison and the antelopes of South Africa, advance, wheel and deploy in something like military order. But it is doubtful if they form in ranks with heads in a real military formation. We all know that migrating birds fly in fair formation and some writers have alleged that some aquatic birds will fish in synchronous order. But so far as known, no one seems to have produced photographic evidence of this lat- ter alleged behavior. For reptiles we apparently have no photo- graphic evidence, but there are at least two written accounts that surely establish the matter of ranked formation. The first of these is from the pen of the old naturalist, William Bartram. In 1774 (Travels through North and South Carolina, Georgia, and . . . Florida, etc. London, 1792, p. 118), while ascending the St. Johns River in eastern Florida, he found great numbers of huge and very aggressive alligators. Some of these threatened attacks on his little boat, when he sought to go into a lagoon off the river to catch some fish for his supper. He says — “I . . . made good my entrance into the lagoon, though not without opposition from the alligators who formed a line across the entrance but did not pursue me into it.” Here the alligators were ranked in a line, ap- parently waiting for the fishes to try to get out into the main stream, when the alligators too would get their supper. This account is counterbalanced by a par- allel description of what C. R. S. Pitman (A Game Warden Among his Charges, London, 1931, p. 248) saw just below Murchison Falls on the Nile River in East Central Africa, where crocodiles are found in incredible numbers. “Looking from above on a still evening, one will be struck by the regular formation taken up by row after row of crocodiles, like ships of war, with intervals of about 50 feet between each crocodile [and those on either side] and 300 feet between the rows, which extend from bank to bank and for about two miles down stream.” But let a fish come down and all is wild confusion figures 1 & 2). and struggle of the neighboring crocodiles to get it or at least a portion of it. The more we know about animals, the more we find them doing unusual and unexpected things. It is not safe offhandedly to contra- dict accounts by non-scientific observers of unusual behavior not otherwise physically impossible — in fishes, as well as in other ani- mals. No article has been found in this search describing regimented fishes and bearing such a title as that at the head of this report. However, there is widely scattered evidence that fishes do “fall in and form ranks.” Un- fortunately, I have never had the opportunity to see fishes take on a military formation but various reputable observers have, and their cumulative evidence will now be set forth chronologically. Fishes Ranked Like Soldiers on Parade. The earliest account of ranked fishes found in this search is by a writer in The New Monthly Magazine, 1820, part II, p. 137, who signs himself “Amateur.” This account is also found in Thomas Boosey’s “Anecdotes of Fish and Fishing,” London, 1887, p. 123. “Amateur”, in writing of the exploits of one Darcey of Oxford, an expert swimmer and diver, who caught fishes with his hands in a deep hole well-known to Oxonians, makes the following statement : — The report that Darcey made, was that many of these fish [barbel] lay with their heads against the bank, in parallel line, like horses in their stalls. They were not disturbed at his approach, but allowed him to come close and select the finest. In talking over this unusual matter with my long-time friend, the late Dr. John L. Peters, an upstate New York man, he told me that in his boyhood he had seen this very thing in a stream in Ulster County. At my request he prepared a statement of his ob- servations, which is the earliest American evidence that has come to hand. In 1907 or 1908, while trout-fishing in the headwaters of Woodland Stream in Ulster Co., New York, my attention was called to the 100 Zoologica : New York Zoological Society [34: 10 peculiar formation of some brook trout after a disturbed pool had quieted down. They seemed to line up in a formation as if some military officer among them had got them ready for a parade. This I saw more than once since I used to go out of my way to watch the trout in this pool. When disturbed, they would scatter, but when things quieted down, they would again take on their military formation. This pool was just below a little rapid in a narrow stream. It was about 8 feet wide, 3 feet deep at its head and about 6 or 8 inches at its shallow end. Dr. Peters drew a little diagram (Text-fig. 1) to show how the fishes were ranked. /I I I I II I II Text-fig. 1. Diagram of trout on parade, in Woodland Stream, Ulster County, New York, 1907 or 1908. Sketch by Dr. John L. Peters. Chronologically our next evidence is in a personal communication from Mr. Joshua W. Atlee of Riverton, New Jersey. He wrote that in October, 1911, he saw ranked fishes in a pool in a rivulet flowing into the Bay of Chaleur, Gulf of St. Lawrence. Carefully pushing aside the shrubbery on the bank of the pool, he had a clear view of it and its piscine inhabitants, which he thought were getting ready for spawning. Of these fishes he noted that: An interesting feature of the sight was the fact that in the slowly moving water, due to the [small] volume of the pool, the fish lay in “sar- dine fashion” closely packed with heads up- stream, stemming the current so as to retain a similar relative position by the slight move- ment of their tails and fins. Finally on being disturbed, some left the pool in various directions, mostly upstream; but my guide, detouring and getting into the stream above them, actually drove many of the trout back into the pool, where they finally settled down again as we had first found them. My next evidence is from Mr. Howard B. MacDonald of Yonkers, N. Y., a traveler and lecturer of wide experience. A photograph taken by him at Rotorua, New Zealand, in 1925, is reproduced as Plate I, Fig. 1. Of it he wrote (personal communication) as fol- lows: Unfortunately, the photograph does not show the fish in quite such straight lines as the other picture you have. However, these fish I saw did act in the same manner as the ones you are studying. Each fish had a certain definite posi- tion in relation to the others of this company; and if the fish were disturbed by throwing a stone into the water near them (as we all did) then they would scatter, but each would return almost immediately to his same position in the group. This was checked and verified by obser- vation many times and there is no doubt but that each fish knew his correct position and always went to it. Here is the word of another reliable ob- server, backed by photographic evidence. As Mr. MacDonald says, this is not such as may be seen in Plate I, Fig. 2, but discounting shadows, the fish are in pairs and they are lined up fairly well in ranked rows. And now follows an excerpt from a per- sonal communication from Dr. Louise M. Perry, long a winter resident of Sanibel Island off Fort Myers, southwest coast of Florida. Dr. Perry, an acute observer for many years of the habits of marine fishes in that region, writes as follows under date of July 26, 1926: Naples, Florida [below Fort Myers], has a fine pier for still fishing, and while waiting for bites, I have repeatedly watched small schools of snook (rovalle) 8 or 12 in a group, lying on the sandy bottom, close together and parallel with each other, all heading the same way and all their tails gently moving to right and left in perfect unison. Suddenly with a rush they would dart into a school of minnows and play havoc for a moment, then each would gently settle down in its former place and position. This performance would be repeated at fairly regular intervals for a long time, and always made me wonder how separate individual ac- tions could be so perfectly synchronized. How do they do it? What is the stimulus that keeps all the tails waving to marching time and starts the snook off in a simultaneous dash after the little fish? Specific attention is called to the fact that all the tails of these marine fishes moved to right and left in perfect unison. And so did the tails of the freshwater fishes observed by Mr. Atlee in 1911 in the rivulet flowing into the Bay of Chaleur, Gulf of St. Lawrence. Presumably the same purpose activated both lots of fishes— to maintain position. And now, also in 1926, come three accounts of this behavior of other marine fishes at the Galapagos Islands, from the pen of that vet- eran observer, William Beebe, in his “Arc- turus Adventure” (New York, 1926). On p. 54 he states that they paid out strings with pieces of bait and enticed three sharks along- side their boat. Here follows his description of the behavior of a large shark and its at- tendants. ... by pulling in the tempting morsel two feet in front of the eager blunt snouts, we brought them to the surface directly under our feet, so that we could watch the movements of the brilliant blue pilotfish, that . . . anticipated every movement of their huge patrons. One of the big fellows had three of these little satellites that unfailingly held their formation, one just above his head, the other two in perfect align- ment a few inches in front of his jaws. So ex- actly synchronized are the movements . . . that 1949] Gudger : Fishes in Ranks 101 it is impossible to tell whether the shark fol- lows the pilotfish or the pilotfish the shark. Again Beebe (1926, p. 183) notes that: Two mighty schools of Xesurus laticlavius [the yellow-tailed surgeonfish] passed me graz- i ing slowly. When within six feet, they left off i their eternal feeding and formed up into more or less orderly ranks which flowed like some enormously long sea-serpent around the iden- tical corners of rocks where had passed the i leaders, yards and yards in advance. Invariably the formation of an irregular line led very close 1 to me, the closing up of ranks evidently being connected with the presence of danger or at least something suspicious or strange. Further, Beebe (1926, pp. 290-291) makes note of another synchronous action of the ! yellow-tailed surgeonfish: “Several hundred approached swimming slowly along, when, as if at a signal, all would stop, and over a rather flat bottom would up-end like ducks and begin to graze [“on the plant and animal fodder which covers the rocks,” p. 290]. On February 27, 1929, the late Prof. M. M. Metcalf wrote — “I am sending you some quo- tations [copied] from an old letter. These . . . aroused my interest at the time the observa- tions were made some years ago.” These observations were by another man, and lack- ing a name and date, will be entered under date (1929) of the covering letter. The per- tinent quotation reads as follows : ... I saw in the clear pool below Trick Falls in the Two Medicine River in Glacier National Park seven trout behaving in a way that seemed interesting. They were headed into the current and were lying motionless in two perfectly straight rows, four in the front row and three in the back row, aligned as accurately as a squad of well drilled soldiers. A moth came flip- ping over the pool, touching the water now and then. All the trout remained quiet, except that the right trout in the rear row turned to the right and backed around the left end of the squad, caught the moth, returned around the left end of the squad to his place at the right end of the rear row again, and they all remained in perfect formation for the several minutes I watched them. On the margin of the typed sheet is a pencil sketch of the movements of the trout at the right hand end of the second rank. This is reproduced herein as Text-fig. 2, and is a graphic presentation of the interesting ac- tion of this particular fish. The well-known sports magazine, Field and Stream, for November, 1929, p. 104, has reproduced the splendid photograph shown in Plate I, Fig. 2. It is also reproduced (in larger size) in the same journal for June, 1935, p. 44. But in neither issue is there any account of the phenomenon, marvellous as it is. The 1929 figure has this caption : “A most extraordinary photograph of resting trout in the Brule River, 40 miles from Duluth, Min- nesota. Note the very unusual formation — like soldiers on parade.” The 1935 issue has a caption which remarks that “When great schools of fish lie in still water, it takes a Text-fig. 2. Trout in two ranks in Two Medicine River, Glacier National Park. The right trout in row two, backed out and followed the course indicated to catch a moth dipping in the water, and then returned to his position. Sketch fur- nished by Prof. M. M. Metcalf, 1929. skillful angler to interest them” — and nothing about military formation in the ar- ticle in which the figure is set. Here is a priceless photograph showing eight rows of “trout on parade.” It is appar- ently the only one ever published of this unique, indeed phenomenal, behavior of fishes, and there is no word of comment be- yond the caption. It seems incredible, but such is the fact. However, the figure splen- didly illustrates the accounts quoted above. But before going further, the present writer submits the following remarks. Trout at rest in running water always face upstream. In pools, especially small ones, they are likely to do the same. In “trout water,” such a pool always has a riffle or rapid at its head, and just below this the water is cooler and has more oxygen than ordinary. Gill-breathing is much easier in fishes facing upstream. Also, in such posi- tion, the fish can readily snap up any edibles coming down with the current. These would seem to explain, in part at least, the heads- upstream of this regiment of trout. At first glance, in Plate I, Fig. 2, we see scores of trout in right-left ranks — fishes on parade — and the ranks separated by right- left stretches of gravel swept clean of fine detritus. Now let us recall that Mr. Atlee found ranked trout in the stream leading into the Bay of Chaleur, Gulf of St. Lawrence, maintaining their positions by moving their tails right and left in unison. Also Dr. Perry saw marine fishes at Naples, southwest coast 102 Zoologica: New York Zoological Society [34: 10: 1949] of Florida, acting in similar fashion. So we must conclude that the trout in Plate I, Fig. 2, were doing this very thing. Any given rank of trout fans out the fine detritus under the tails of its members. This is checked and some of it precipitated by the bodies of the rank of fishes just behind it — and so all the way from the foremost rank to the hindmost. Probably these rows of trout lie on gently backward sloping ridges of the detritus. This cleaning action holds best for the center of the stream but fades out somewhat on the edges where the current is weaker. The collecting of data for an article on military fishes was begun more than 20 years ago. But the work went slowly and presently press of other work — particularly the editing of the Bashford Dean Memorial Volume — led to the filing away of all material till a more convenient time — which has just come. Dur- ing this period of inactivity in this study, letters came in from a few persons who had heard of my interest in this problem. But their statements were in very general terms, and quite unclear. Had I taken the time to ask for more specific accounts of what they saw, I might have gotten additional valuable data. Now it is too late. However, abundant evidence is to be found in the written accounts and in the sketches and photographs herein to establish the fact that various fishes, but especially trout, do rank themselves in parade order. These data certainly justify the title of this article. For trout in running water, some tenable expla- nation has been advanced. But for trout and all other fishes, where behavior has been de- scribed, there must be a more fundamental universal reason. This no one has attempted — the explanation must be left to the animal behaviorists. Finally, it may be said that, from the com- ments of several friends, who know of the work on this article, I am satified that this curious behavior of ranked fishes, “fishes on parade,” is not at all uncommon. In fact, it is probably far better known than the present writer realizes. However, it is a curious thing that in this study there has not been found a single article with such an indicative title as this paper bears and it is hoped that others, who have witnessed this curious be- havior, will publish their observations and thus establish this parade behavior as a normal procedure. EXPLANATION OF THE PLATE. Plate I. Fig. 1. Ranked trout in a pool at Rotorua, New Zealand. Discounting the shadows, the trout are seen to be roughly ranked in pairs. Photograph by Howard B. Mac- Donald. Fig. 2. Resting trout in eight ranks, like sol- diers on parade, photographed in the Brule River, 40 miles from Duluth, Minnesota. From Field & Stream, 1929. GUDGER. PLATE I FIG. 2. FISHES THAT RANK THEMSELVES LIKE SOLDIERS ON PARADE. vSZa Crandall: Seasonal Changes in Creatophora cinerea 103 11. Notes on Seasonal Changes in Creatophora cinerea, the Wattled Starling. Lee S. Crandall. (Plate I). The Wattled Starling ( Creatophora ci- nerea), is an African member of the family Sturnidae. It has a rather wide distribution, extending from southwestern Arabia through East Africa to the Cape. An inves- tigation of seasonal changes in males of this species is outlined in the following notes. Sequence 1. A young specimen of the Wattled Starling ( Creatophora cinerea) of undetermined sex, and unknown locality, was received at the New York Zoological Park on October 17, 1927. During the entire period of observation this bird, as well as the others noted, was kept in an indoor, heated aviary from Octo- ber to May, with access to an outdoor cage during the intervening months. At the time of arrival, the head and throat were completely feathered and only the usual narrow, bare malar streaks were visible. In the spring of 1928, the bird showed itself to be a male by an enlargement of the throat wattle, although there was no evi- dence of crown wattles and no loss of plum- age of the head. In the spring of 1929, the throat wattle again became enlarged, a small area of the forehead became bare and the crown wattles appeared, reaching an upright maximum of about %"• On April 19, 1930, the feathers of the forehead were dropping out. By May 9, the crown and face were black and entirely bare, except for a small tuft behind each nostril. The occipital region was bare and yellow. The crown wattles, completely sessile, had no power of erection. The posterior wattle, attached longitudinally, measured 9 mm. at the base, expanding to a width of 11 mm. and reaching a length of 20 mm. The anterior wattle, set at an approximate right angle to the longitudinal center line, overhung the base of the bill and was overhung, in turn, by the posterior wattle. Its width at the base was 7 mm., its greatest width was 15 mm. and its greatest length 11.5 mm. The throat wattle, along the anterior mar- gin, measured 40 mm. when drawn out with the fingers. (As this wattle is slightly re- tractile, an accurate measurement could not be made.) It is bifurcated at the tip, the right division measuring 9 mm., the left 12 mm. At this time, the bird indulged in mild courtship maneuvers, tossing his head so that the crown wattles might fall on one side or other, and singing a broken and guttural song. On August 25, the wattles were seen to be shrinking and feathers of head and face growing. This process continued until, on September 14, the wattles were entirely with- drawn and feathering was complete. How- ever, the nodular crown wattles, normally invisible, remained discernible when the plumage was tightly compressed. Also, the malar streaks, from which the throat wattle had sprung, remained slightly more pro- nounced than in an immature male or a female. During the following years, changes took place as follows : 1931. April 11. Feathers dropping. May 2. Change complete. September 26. Feathers growing. October 17. Change complete. 1932. March 1. Feathers dropping. March 22. Change complete. September 19. Head feathering, wat- tles shrinking. October 10. Change complete. 1933. January 4. Feathers dropping. January 25. Change complete. August 18. Head feathering. September 12. Change complete. December 1. Feathers dropping. December 26. Change complete. 1934. September 12. Head feathering. October 1. Change complete. 1935. February 27. Feathers dropping. March 22. Change complete. September 30. Head feathering. October 22. Change complete. December 30. P’eathers dropping. 1936. January 13. Change complete. January 20. Feathers re-growing, wattles shrinking. 104 Zoologica: New York Zoological Society [34: 11 February 3. Head entirely re-fea- thered, wattles partly retracted. February 10. Feathers dropping again, wattles enlarging. February 24. Change complete, wattles fully extended, bird singing. September 21. Feathers growing. October 14. Change complete. December 21. Feathers dropping. 1937. January 8. Change complete. September 20. Feathers growing. October 7. Change complete. December 20. Feathers dropping. 1938. January 3. Change complete, wattles fully enlarged. February 7. Feathers re-growing, wattles shrinking. February 21. Feathering complete. March 7. Feathers dropping again, wattles enlarging. March 28. Change complete. September 12. Feathers growing. October 1. Change complete. December 19. Feathers dropping. 1939. January 6. Change complete. March 5. Observations ended by death of subject. Sequence 2. Two Wattled Starlings which proved to be males, were received on December 14, 1934, from a dealer. The locality from which they had come was unknown. These birds were kept together, under identical condi- tions, during the course of the following observations. They are designated as #1 and #2. Both were fully feathered on arrival. 1935. #1. #2. 1936. #1. #2. May 1. Feathers dropping. May 21. Change complete, crown and throat wattles well devel- oped. October 21. Feathers growing, wattles shrinking. November 13. Change complete. June 5. Throat wattle enlarged, feathers dropping. June 24. Face and crown bare, throat wattle pendant, no evi- dence of crown wattles. October 28. Feathers growing. November 11. Change complete. April 6. Feathers dropping. April 26. Change complete. October 26. Feathers growing. November 13. Change complete. April 27. Feathers dropping. May 16. Change complete, crown wattles minute. November 2. Feathers gi'owing. November 20. Change complete. 1937. #1. March 15. Feathers dropping. March 31. Change complete. October 18. Feathers growing. November 6. Change complete. #2. April 12. Feathers dropping. April 29. Change complete, crown wattles minute, throat wattle deeply pendant. October 25. Feathers growing. November 14. Change complete. 1938. #1. February 28. Feathers dropping. March 16. Change complete. October 31. Feathers growing. November 18. Change complete. #2. March 28. Feathers dropping. April 18. Change complete, crown wattles minute. October 2. Feathers growing. October 21. Change complete. 1939. #1. March 6. Feathers dropping. March 25. Change complete. October 23. Feathers growing. November 10. Change complete. #2. March 20. Feathers dropping. April 7. Change complete, crown wattles minute. October 23. Feathers growing. November 15. Change complete. 1940. #1 February 26. Feathers dropping. March 15. Change complete. September 23. Feathers growing. October 9. Change complete. #2 April 8. Face feathers dropping. April 27. Change complete, crown wattles minute. October 28. Feathers dropping. November 13. Change complete. 1941. #1. March 3. Feathers dropping. March 23. Change complete. October 13. Feathers growing. October 29. Change complete. #2. April 14. Feathers dropping. May 4. Change complete, crown wattles minute. October 27. Feathers growing. November 14. Change complete. 1942. #1. #2. 1943. #1. #2. March 9. Feathers dropping. March 31. Change complete. September 28. Feathers growing. October 19. Change complete. March 30. Feathers dropping. April 20. Change complete, crown wattles minute. November 9. Feathers growing. November 29. Change complete. February 15. Feathers dropping. March 8. Change complete. March 15. Feathers dropping. April 7. Change complete. Sequence 3. Two Wattled Starlings which appeared to be male and female, were purchased from Christoph Schulz on August 9, 1935. They were reported by Schulz to have been col- lected in Kenya. The male was in breeding condition at the time of arrival, with head bare and wattles well developed. 1949] Crandall: Seasonal Changes in Creatophora cinerea 105 Throughout the course of observations on this pair, the female showed no plumage change and no enlargement of the bare malar streaks, at the time the male was coming into breeding condition. During October and November of each year she went through a complete body molt. Changes in the male were noted as follows : 1935. October 22. Feathers growing. November 13. Change complete. 1936. April 6. Feathers dropping. April 24. Change complete, wattles well developed. October 5. Feathers growing. October 27. Change complete. 1937. January 18. Feathers dropping. February 4. Change complete. March 1. Feathers re-growing. March 15. Face almost completely feathered, wattles shrunken. March 22. Feathers dropping again, wattles re-enlarging. April 2. Head entirely bare, wattles large. October 25. Feathers growing. November 12. Change complete. 1938. February 14. Feathers dropping. March 3. Change complete but wattles small. March 28. Face re-feathering. April 11. Face almost entirely re- feathered. April 18. Feathers dropping again. May 5. Change complete, wattles much enlarged. October 31. Face feathers growing. November 19. Change complete. 1939. March 27. Feathers dropping. April 17. Change complete. October 9. Feathers growing. October 30. Change complete. 1940. February 26. Feathers dropping. March 20. Change complete. September 23. Feathers growing. October 10. Change complete. 1941. March 24. Feathers dropping. April 12. Change complete. September 29. Feathers growing. October 18. Change complete. 1942. March 2. Feathers dropping. March 20. Change complete. October 5. Feathers growing. October 23. Change complete. 1943. February 15. Feathers dropping. March 6. Change complete. Summary. Seasonal changes in four male and one female specimens of the Wattled Starling ( Creatophora cinerea) have been tabulated and described. All of these birds were kept under identical conditions, as far as caging, food and temperatures were concerned. It is shown that in the males there is a seasonal loss of the plumage of the head, accompanied by enlargement of the crown and throat wattles. The single female showed no en- largement of the bare gular tracts and had only a single annual change of plumage, which took place at the period of regression in the accompanying male. The males of Sequence 1 and Sequence 3 showed occasional “false starts,” in which newly bared heads almost immediately re- feathered, quickly followed by a resumption of the bare condition. Neither of the two males described in the second series of ob- servations showed this phenomenon. Recorded dates of changes were estab- lished on a visual basis. However, standards of judgement were the same in all cases, so that periods indicated are properly com- parable and variations would be small. 104 Zoologica : New York Zoological Society [34: 11 February 3. Head entirely re-fea- thered, wattles partly retracted. February 10. Feathers dropping again, wattles enlarging. February 24. Change complete, wattles fully extended, bird singing. September 21. Feathers growing. October 14. Change complete. December 21. Feathers dropping. 1937. January 8. Change complete. September 20. Feathers growing. October 7. Change complete. December 20. Feathers dropping. 1938. January 3. Change complete, wattles fully enlarged. February 7. Feathers re-growing, wattles shrinking. February 21. Feathering complete. March 7. Feathers dropping again, wattles enlarging. March 28. Change complete. September 12. Feathers growing. October 1. Change complete. December 19. Feathers dropping. 1939. January 6. Change complete. March 5. Observations ended by death of subject. Sequence 2. Two Wattled Stai’lings which proved to be males, were received on December 14, 1934, from a dealer. The locality from which they had come was unknown. These birds were kept together, under identical condi- tions, during the course of the following observations. They are designated as #1 and #2. Both were fully feathered on arrival. 1935. #1. #2. 1936. #1. #2. 1937. #1. May 1. Feathers dropping. May 21. Change complete, crown and throat wattles well devel- oped. October 21. Feathers growing, wattles shrinking. November 13. Change complete. June 5. Throat wattle enlai'ged, feathei's dropping. June 24. Face and crown bai'e, throat wattle pendant, no evi- dence of crown wattles. October 28. Feathers growing. November 11. Change complete. April 6. Feathers dropping. Api’il 26. Change complete. October 26. Feathers growing. November 13. Change complete. April 27. Feathers dropping. May 16. Change complete, crown wattles minute. November 2. Feathers growing. November 20. Change complete. March 15. Feathers dropping. March 31. Change complete. October 18. Feathers growing. November 6. Change complete. #2. Api'il 12. Feathers dropping. Api’il 29. Change complete, crown wattles minute, throat wattle deeply pendant. October 25. Feathers growing. November 14. Change complete. 1938. #1. Februai’y 28. Feathers dropping. March 16. Change complete. October 31. Feathers growing. November 18. Change complete. #2. March 28. Feathers dropping. April 18. Change complete, crown wattles minute. October 2. Feathers growing. October 21. Change complete. 1939. #1. March 6. Feathei's dropping. March 25. Change complete. October 23. Feathers growing. November 10. Change complete. #2. March 20. Feathers dropping. April 7. Change complete, crown wattles minute. October 23. Feathers growing. November 15. Change complete. 1940. #1 February 26. Feathers dropping. March 15. Change complete. September 23. Feathers growing. October 9. Change complete. #2 April 8. Face feathers dropping. April 27. Change complete, crown wattles minute. October 28. Feathers dropping. November 13. Change complete. 1941. #1. March 3. Feathers dropping. March 23. Change complete. October 13. Feathers growing. October 29. Change complete. #2. April 14. Feathers dropping. May 4. Change complete, crown wattles minute. October 27. Feathers growing. November 14. Change complete. 1942. #1. March 9. Feathers dropping. March 31. Change complete. September 28. Feathers growing. October 19. Change complete. #2. March 30. Feathers dropping. April 20. Change complete, crown wattles minute. November 9. Feathers growing. November 29. Change complete. 1943. #1. February 15. Feathers dropping. March 8. Change complete. #2. March 15. Feathers dropping. April 7. Change complete. Sequence 3. Two Wattled Starlings which appeared to be male and female, were purchased from Christoph Schulz on August 9, 1935. They were reported by Schulz to have been col- lected in Kenya. The male was in breeding condition at the time of arrival, with head bare and wattles well developed. 19491 Crandall: Seasonal Changes in Creatophora cinerea 105 Throughout the course of observations on this pair, the female showed no plumage change and no enlargement of the bare malar streaks, at the time the male was coming into breeding condition. During October and November of each year she went through a complete body molt. Changes in the male were noted as follows : 1935. October 22. Feathers growing. November 13. Change complete. 1936. April 6. Feathers dropping. April 24. Change complete, wattles well developed. October 5. Feathers growing. October 27. Change complete. 1937. January 18. Feathers dropping. February 4. Change complete. March 1. Feathers re-growing. March 15. Face almost completely feathered, wattles shrunken. March 22. Feathers dropping again, wattles re-enlarging. April 2. Head entirely bare, wattles large. October 25. Feathers growing. November 12. Change complete. 1938. February 14. Feathers dropping. March 3. Change complete but wattles small. March 28. Face re-feathering. April 11. Face almost entirely re- feathered. April 18. Feathers dropping again. May 5. Change complete, wattles much enlarged. October 31. Face feathers growing. November 19. Change complete. 1939. March 27. Feathers dropping. April 17. Change complete. October 9. Feathers growing. October 30. Change complete. 1940. February 26. Feathers dropping. March 20. Change complete. September 23. Feathers growing. October 10. Change complete. 1941. March 24. Feathers dropping. April 12. Change complete. September 29. Feathers growing. October 18. Change complete. 1942. March 2. Feathers dropping. March 20. Change complete. October 5. Feathers growing. October 23. Change complete. 1943. February 15. Feathers dropping. March 6. Change complete. Summary. Seasonal changes in four male and one female specimens of the Wattled Starling ( Creatophora cinerea) have been tabulated and described. All of these birds were kept under identical conditions, as far as caging, food and temperatures were concerned. It is shown that in the males there is a seasonal loss of the plumage of the head, accompanied by enlargement of the crown and throat wattles. The single female showed no en- largement of the bare gular tracts and had only a single annual change of plumage, which took place at the period of regression in the accompanying male. The males of Sequence 1 and Sequence 3 showed occasional “false starts,” in which newly bared heads almost immediately re- feathered, quickly followed by a resumption of the bare condition. Neither of the two males described in the second series of ob- servations showed this phenomenon. Recorded dates of changes were estab- lished on a visual basis. However, standards of judgement were the same in all cases, so that periods indicated are properly com- parable and variations would be small. 106 Zoologica: New York Zoological Society [34: 11: 1949 EXPLANATION OF THE PLATE. Plate I. Fig. 1. Adult $ Wattled Starling (Sequence 1) , photographed on June 13, 1933. The head is completely bare and wattles fully developed. Fig. 2. The same bird, photographed on No- vember 20, 1934. He has completed regressive changes and is in resting condition. The nodular crown wattles, not normally visible at this time, are seen because the feathers are tightly compressed. The throat wattle has re- ceded to the lateral malar patches. CRANDALL. PLATE I. FIG. I. FIG. 2. NOTES ON SEASONAL CHANGES IN CREATOPHORA C1NEREA, THEWATTLED STARLING. Beebe : Insect Migration at Rancho Grande 107 Insect Migration at Rancho Grande in North-central Venezuela. General Account.1 William Beebe. Director, Department of Tropical Research, New York Zoological Society. (Plates I & II; Text-figure 1). [This is one of a series of papers resulting from the 45th, 46th and 47th Expeditions of the Department of Tropical Research of the New York Zoological Society, made during 1945, 1946 and 1948, under the direction of Dr. William Beebe, with headquarters at Rancho Grande in the National Park of Aragua, Vene- zuela. The expeditions were made possible through the generous cooperation of the Na- tional Government of Venezuela and of the Creole Petroleum Corporation. [The characteristics of the research area are in brief as follows ; Rancho Grande is located in north-central Venezuela (10° 21' N. Lat., 67° 41' W. Long.), 80 kilometers west of Caracas, at an elevation of 1,100 meters in the undisturbed montane rain forest which covers this part of the Caribbean range of the Andes. The migra- tion flyway of Portachuelo Pass, which is also the water-shed between the Caribbean and Lake Valencia, is 200 meters from Rancho Grande. Adjacent ecological zones include seasonal forest, savanna, thorn woodland, cactus scrub, the fresh-water lake of Valencia and various marine littoral zones. The Rancho Grande area is generally subtropical, being uniformly cool and damp throughout the year because of the prevalence of the mountain cloud cap. The dry season extends from January into April. The average humidity during the expeditions, in- cluding parts of both wet and dry seasons, was 92.4%; the average temperature during the same period was 18° C; the average annual rain- fall over a five-year period was 174 cm. The flora is marked by an abundance of mosses, ferns and epiphytes of many kinds, as well as a few gigantic trees. For further details see Beebe and Crane, Zoologica, Vol. 32, No. 5, 1947. Un- less otherwise stated, the specimens discussed in the present paper were taken in the montane cloud forest zone, within a radius of one kilo- meter of Rancho Grande.] Contents. General Account 107 Migrating Insects 108 Migration Factors 109 Rainy Season 109 Inhibiting Conditions 109 Favoring Conditions 109 Recurrent Waves 109 Between Waves 109 Wing Condition 109 Models and Mimics 109 Specific Characteristics 109 Sexes and Breeding 109 Diurnal Sequence 109 Interpretation HO Sight Identification 110 t Contribution No. 843, Department of Tropical Research, New York Zoological Society. General Account. Throughout the first year of our occupa- tion of the station at Rancho Grande in north-central Venezuela, we had no idea of the importance of Portachuelo Pass as a migration flyway for birds and insects. Even later on, when we came to compile a list of thirteen life zones within our visual radius, a fourteenth, the Aerial Zone, was added with hesitation, having in mind the inclusion of organisms such as humming- birds, swifts and mayflies which spend the major part of their lives in midair. Almost immediately, however, the value of and need for such a niche in our phenological pro- gram became apparent. If for no other reason, an Aerial Zone was needed to accommodate the volant or- ganisms which passed and repassed, or oc- casionally were detected soaring in air, and which were never to be found resting or flying in the jungle of our immediate area of research. Many of these creatures were essentially tropical, occupying our sub- tropical elevation only as a temporary route of passage. About 200 meters beyond Rancho Grande, the road leads through a narrow notch in the east-west, coastal Andean range. This is Portachuelo Pass with an elevation of 1,136 meters, about 36 meters higher than Rancho Grande. The flattened floor of the pass is only about 20 meters wide, and the shoulders on either side rise in sharp ridges, 389 meters to the summit of Pico Periquito on the west, and 764 meters on the east to the top of Pico Guacamayo. The pass is at the 22.5 kilometer mark on the road from Maracay, and, as has been stated, is at an elevation of 1,136 meters. Kilometer 31, well to the north of the pass, is 770 meters above the sea. Kilometer 15, equally distant to the south of the pass and close to the beginning of the lowland savan- nas, is 760 meters above the sea. At both of these lower stations many migrants have been taken, en route to or on their way from the pass. The pass is on the real divide, shunting the waters on its northern slope into the 108 Zoologica: New York Zoological Society Caribbean Sea, and those on the south side ultimately into Lake Valencia. Fifteen orders of insects have already been collected or observed as they passed southward on migration, singly, in few or in enormous numbers. Of other possible . migrants this leaves only three orders, Ephemerida, Embiidina and Trichoptera. The four parasitic groups, Anopleura, Si- phonoptera, Mallophaga and Strepsiptera, are of course absent, although the two latter doubtless hitch-hike through the pass on birds and bees respectively. The migrating orders, arranged in three columns of relative numbers, are as follows : Rare Isoptera Neuroptera Plecoptera Corrodontia Thysanoptera Dermaptera Mecoptera Moderate Orthoptera Odonata Homoptera Hemiptera Abundant Coleoptera Lepidoptera Diptera Hymenoptera Up to the date of this publication/ .Mr. Henry Fleming has identified two hundred and forty-five species of butterflies, and fifty-two species of day-flying moths. Of the single family of nocturnal moths, Sphing- idae, we have recorded seventy-six species, either directly migrating through the pass, or flying about our lights at Rancho Grande. Mr. Fleming has found only two of these which may be classed as breeding in the cloud forest surrounding the pass and our laboratory. Except for a temporary, limited, north- ward drift of individual Phoebis ( Catopsilia ) in the early part of the rains, the movement of all orders of insects was invariably from north to south through the pass. Migrating Insects. Examples of extremes in migration will serve to point up succeeding papers dealing with families and species. On April 29, 1946, I caught a solitary butterfly at the rim of the pass. Its wing spread was small, and it was an inconspic- uous purplish-brown with five spots of dull white near the tip of each fore wing. It belonged to the family Nymphalidae. I gave it the reference field name of Ten-spot Brown and later found its technical name was Eunica monima (Cramer). On May 4, five days later, I caught sight of several butterflies passing overhead and at the pass itself I entered a dense swarm of the Ten-spots. Mingled with them were tailed nymphalids, Marpesia chiron chiron (Fab.) in large numbers, and now and then a swarm of large sulphurs, Phoebis eubule marcellina (Cramer). In the dis- tance I could see myriads of Ten-spots con- verging on the pass. One swoop of the net captured seven, five of which were tat- tered and torn, the remaining two freshly emerged. [34: 12 Jfll T J'r- Text-fig. 1. Map showing location of Rancho Grande, Portachuelo Pass and surrounding territory. bg to j indwf Two of us climbed a mound giving a view of about half the width of the pass, and here, facing in opposite directions, at eye level, we averaged thirteen hundred butter- flies in several counts of four minutes each. At intervals throughout an hour and a half this insect content of a limited time and space remained fairly constant, and when we left we knew that at the very least, two hundred and eighty-six thousand Ten-spots had passed close to us. An hour later the insects were going full strength and now I brought to bear my giant binoculars, first twelve and then twenty powers. I began about twenty-five feet overhead and then refocussed slowly upward until the limit of vision of the small insects was reached. This, judged by horizontal tests of objects of similar size would be about a half mile zenithwards, and at every fractional turn of the screw, more and more smaller-ap- pearing butterflies fluttered into clarity. Throughout the entire extent of vertical- ity there was no lessening of denseness of flying insects, and it was almost a pure culture of Eunica and Marpesia. For many days this particular phase of migration con- tinued, millions upon millions coming from some unknown source, travelling due south to an equally mysterious destination. Three weeks later, on May 24, there was a resurgent migration of the same species, all fresh insects. Their numbers far ex- ceeded the first wave. Four of us lined up across the entire width of the pass, with stop-watches and counters, completely failed ami . rent o m. ■ lion :: Bid bs my ne two V,' ind : warm K CJ butte; ing v dean April tort rery Se] wr 1 day r leai icon m\ Fen Dole teaj Dial T1 liar cfa 1949] Beebe: Insect Migration at Rancho Grande 109 to keep up with fast enough estimate of 1 numbers, but at the minimum clocked a thousand a second going past in the face of a gentle breeze. In the narrow trail above the gorge it was necessary to put on glass- es, so dense were the crowds impinging upon our faces. As the other extreme, I may mention a half hour of collecting when many species in fewer numbers were passing. Twenty successive specimens of butterflies resolved into eleven species of pierids. On another occasion thirteen individual butterflies proved to be thirteen separate species of ithomiids. Non-recognition of the pass as a fly way accounts for the small number of observa- tions in the year 1946, and the still more meagre and casual notes in 1945. Some time passed before we realized that all of the host of moths which came in windrows to the roof lights and laboratory windows of Rancho Grande were Portachuelo Pass mi- grants, deflected by confusion of fog or rain. On clear nights of star and moonlight ohr torches and portable ultra-violet machines revealed unbroken streams of moths of all sizes headed up and through the pass. Other indirect evidences were the wings, belong- ing to great numbers of species and indi- viduals of moths, found glued in early morn- ing to the dew-moistened leaves of shrubs and weeds in the pass; the remains of noc- turnal feasts of marauding bats. Migration' Factors. One definite factor, which seems the dominant stimulus of migration, is the ad- vent of the rainy season. For example, in 1948, there was no hint whatever of migra- tion in February. On March 1 a single torn and bedraggled nymphalid, Marpesia chiron chiron, struggled up to the pass and into my net. Hardly another insect appeared for two weeks, throughout a period mostly cold and overcast. Then, on March 15, a day of warm sun after several days of heavy rain, we caught or recorded twenty species of butterflies in considerable numbers. Succeed- ing weeks of cold resulted in a complete dearth or mere scattering of insects, until April 15. From this date until August 1 there was no cessation of numbers pouring through, varied only by irregular fluctua- tions due to occasional days of cold rain or very high wind. September 9 is the latest date of any of our three years of residence, and on that lay migration was in full swing. From what I can learn, the passing insects gradually decrease throughout the succeeding two weeks. On the authority of Dr. Francisco Fernandez, Venezuelan Government Ento- mologist, diurnal migration at the pass ceased for the year by October 1. The an- nual picture thus seems clear cut. The following applies more particularly to diurnal Lepidoptera, but in general is true of all orders : Inhibiting Conditions: Very high winds, from twenty-five miles per hour upward; chilly temperatures, 62° Fahr. down; dense fog (neblina) or heavy rain; darkness. Favoring Conditions : Calm, up to a twen- ty mile per hour following wind; 64° Fahr. plus; sun or thin neblina. Recurrent Waves: These last from twenty minutes to three weeks, and usually com- prise few species (two to twenty), but often large numbers of individuals. These waves are occasionally independent of favorable conditions, the hosts of insects banking up in the lee of brush, waiting for good flying weather. Between Waves: At these times insects tend to fly singly, and in great variety of species. Wing Condition: Worn and fresh speci- mens may be present in the same wave, but as a rule all are worn or all are freshly emerged. Models and1 Mimics: Some of the more generally accepted categories of models and mimics may appear mingled together in the waves, or, very interestingly, there are not infrequently pure cultures of each, confined to waves of considerable magnitude. Specific Characteristics: Normal specific characteristics of flight and of choice of habitat are maintained throughout migra- tion. Rapid or slow flyers do not alter their relative speeds, nor change their dodging, zigzag or direct 'flight. The same applies as well to high or low habitual levels of flight, fast or slow flapping of wings. Species which prefer to wind their way through low, thick brush adhere to this habit en route up to and through the pass. Sexes and Breeding: The general rule is the presence of both sexes, and many of the females captured alive deposit eggs within twenty-four hours. Rarely, attempts at mat- ing on migration are observed, or pairs ap- pear to be fighting as they circle rapidly in midair. Very few loiter to feed at blossoms. Few worn individuals stop to rest. Diurnal Sequence: A few insects, belong- ing to various orders other than Lepidoptera, appear very early in the morning, for months on end, flying past singly, but in the aggre- gate in great numbers. Especially noticeable among these are two species of cockchafers ( Cyclocephala spp.), a chrysomelid ( Dia - brotica quindecimpunctata) , a small vespid ( Stalopolybia areata ) , a giant hairy scoliid ( Campsomeris ianthina ) , and a bee ( Euglos - sa fasciata ) . The numbers of these solitary migrants passing on the morning of June 19, 1948, typifies the numbers on every day for the preceding two months: cockchafers, 200; chrysomelids 150; small vespids 150; giant scoliids 140; rufous bees 90. Following these there comes, for an hour or longer, a steady procession of day-flying moths, also singly. Butterflies dominate the remainder of the day as far as relatively large insects are concerned. Throughout the 110 Zoologica: New York Zoological Society [34: 12: 1949] 0- daylight hours there is a continuous passing of migration nekton, hosts upon hosts of minute winged insect life. When dusk gives way to darkness, moths and other nocturnal insects appear, and surge through the pass. If the night is fine, with clear moon or star- light, all continue down Limon Gorge. If the sky is overcast, thick with neblina or rain, the moths leave their direct southern route and detour in tens of thousands to our lighted laboratory windows or white roof walls. Interpretation: At present, before a de- tailed study has been made of the mass of specimens and data, and further explorations undertaken of places of origin and destina- tion in surrounding country, no reasonable explanation of this wholesale annual emigra- tion is possible. Unlike the migration of hosts of Phoebis males which I have observed in British Gui- ana and elsewhere, the Portachuelo hosts are represented by both sexes. Many of the fe- males are ready to deposit eggs, although they are headed away from areas rich in a variety of plants, toward less luxuriant savannas. The known distribution of many species, or especially subspecies, of butterflies such as the Papilios, shows that their place of origin cannot be very far away to the north and west, but as yet we have no hint of whence the fifteen orders are derived, or whither they are headed. From our own ob- servations, reinforced by the reports of reli- able govenment official entomologists, we are certain that little or no migration occurs during the dry season, and that not an insect ever returns to its natal haunts. Hence the phenomenon is really an emigration of the cross section of a considerable volant inverte- brate fauna of this part of Venezuela. Observations during three years confirm the fact that this migration is a regular an- nual event. It presents the inexplicable prob- lem of a regular abstraction of an appreci- able percentage of non-returning population from the area of origin. This implies the leaving behind of an adequate residual num- ber of non-migrants to carry on the race and to sustain future migration. Sight Identification: When there came recognition of migration on a relatively great scale, our first and indeed continued impulse was to capture as many specimens of as many different kinds as was humanly possible. On an early day of observation a butterfly was taken which, in our mind, was instantly label- led a Monarch, Danias archippus, or, if you prefer, Danaus plexippas, a familiar of our northern fields. Within an hour eighteen of the same species flapped slowly past, two of them alighting for a few seconds. Within ten minutes there passed eight smaller, darker red butterflies, two of which we took, which vividly l'eminded us of our northern Danaus berenice, commonly called the Queen. Ul- timately these two resolved into Danaus plexippus megalippe and Danaus eresimus eresimus, but throughout my notes, before . identification, they were nostalgically record- ' ed as Monarch and Queen. The important thing is that, being easily and accurately recognizable at a distance in flight for what they are, I am able to record, without fear of error, all the numbers that came within my purview, in all my hours of observation at the pass. This is all by way of introducing the im- portant question of sight identification, which, in any research such as the present, must play a dominant part. I based all my field naming first, on captured and ultimately precisely named specimens, and secondly, on characters in flying or resting insects which left no shadow of doubt. Although hundreds and tens of thousands of insects passed with only the vaguest hints of family or genus, yet day after day familiarity introduced to the perception characteristics of flight, pat- tern, color and shape of wings, and general facies, which materially increased range and certainty of recognition. Viewing from a distance of ten meters, groups of species mounted in open cases, proved an excellent check on visual aware- ness of distinctions. A brief treatment of sight identification will be added to each paper dealing with families of insects. This and following papers are intended only as factual presentations of notes made during three seasons of observations of mi- gration from north to south through Por tachuelo Pass. Hence no attempt has been made at correlation or even mention of mi- grations of the same or related species ob- served by entomologists elsewhere. The in- sects themselves will be considered group by group in successive papers, with a final sum mary in detail of the migration as a whole. We hope, in future expeditions in this same field, that data will be obtained which will clarify the place of origin and ultimate destination of the insect hosts, as well as the initiating stimuli and directive factors of their migration. An account of the bird migration through the pass has already appeared, treating of sixty species divided among ten types of migration.2 Fig. Fig. EXPLANATION OF THE PLATES. Plate I. 1. Looking south toward Portachuelo Pass from Kilometer 31. 2. Portachuelo Pass, the notch in the distant sky-line, from farther north, near the sea. Plate II. Fig. 3. Migrant insects alive but quieted by refrigeration. Fig. 4. Migrant moths deflected in great num- bers, on nights of storm, from their migration through the pass, to the electric lights on Rancho Grande roof. 2 Zoologica. 32 (18), 1947, pp. 153-168. BEEBE. PLATE I FIG. I. FIG. 2. INSECT MIGRATION AT RANCHO GRANDE IN NORTH-CENTRAL VENEZUELA. GENERAL ACCOUNT. :ebe. PLATE II. FIG. 4. INSECT MIGRATION AT RANCHO GRANDE IN NORTH-CENTRAL VENEZUELA. GENERAL ACCOUNT. Riess, Ross, Lyerly & Birch: Behavior of Captive Lowland Gorillas 111 13. The Behavior of Two Captive Specimens of the Lowland Gorilla, Gorilla gorilla gorilla (Savage & Wyman).1 B. F. Riess,2 * Sherman Ross,2 S. B. Lyerly,4 and H. G. Birch." Behavior Research Fellows (1948), New York Zoological Society. (Plates I & II; Text-figures 1 & 2). I. Introduction. The field of comparative behavior research has long been subject to two methodological factors which have to a certain extent pre- vented the attainment of its goal, namely the securing of information on behavioral pro- cesses of representative species of a wide range of living organisms. The first factor has been the concentration of research work- ers on those animals which are adaptable to the limited conditions of laboratories. This emphasis on a few selected species has led to the second factor, the acceptance of the conventional laboratory as the prototype of habitat for the species under investigation. Both these methodological limitations have arisen in part from the same set of circum- stances, the relative lack of availability of less adaptable organisms and the expense of field studies. The increasing number of in- vestigations under naturalistic field condi- tions and of studies on rare specimens under favorable conditions has provided additional evidence of the fruitfulness of the extension of both laboratory and field methodology to specimens other than those generally used in comparative behavior laboratories. It is one of the purposes of this paper to point to a source of data which can facilitate not only the collection of more information on a wider variety of animals but which can also serve as a training facility for field- workers and others. Within the reach of re- searchers in most large cities there exist collections of living animals in great variety of species and under varied living conditions. The reference is to the zoological parks and exhibition areas. In many of these, natural habitat conditions are approximated and even the differences can be fertile sources of comparative psycho-ecological studies. 1 The success of this project was due in large part t the co-operation of the staff and keepers of the New Yor Zoological Park. Particular thanks are due Mr. Fairfiel Osborn Mr. John Tee-Van, Mr. Lee S. Crandall, Dr. L . Goss, Keepers Reilley and Quinn, and Mr. Sara Duntoi photographer. 2 Hunter College of the City of New York, N. Y. C. 3 Bucknell University, Lewisburg, Penna. 4 University of North Carolina, Chapel Hill N C 5 City College of New York, N, Y. C. In 1937, Carpenter (2) published a study of two young male mountain gorillas, Gorilla gorilla berengei, resident at the San Diego Zoo. In this paper he pointed to the oppor- tunities presented by the fourteen specimens of the largest of the great apes which he listed as available in the zoological parks of the world. Gorillas offer a dramatic field for this type of research since they are, with the exception of the orang-utan, the least studied of the anthropoid apes. The present paper seeks to compare Carpenter’s data with those obtained in a study of male and female pre- adolescent lowland gorillas, Gorilla gorilla gorilla, in the New York Zoological Park (Bronx Zoo). Such comparative data as can be assembled will be helpful to workers who seek a base line for similar investigation elsewhere. That these investigations are pos- sible is demonstrated in Table I which lists the location and over-all biological indices of the specimens now resident in various collec- tions. The number of gorillas in the United States is increasing. In 1937 there were only eight specimens in this country; today there are twenty-four. Other than Carpenter’s study, material on the behavior of the gorilla is found only in a limited number of papers. Yerkes’ pioneer work (6) with the pre-adolescent female mountain gorilla, Congo, is known to all students of comparative psychology. Bing- ham’s 1932 observations (1) on gorillas in their native habitat and Valker’s similar study in the Gaboons (5) in 1931 are the only other relevant research. The two lowland gorillas in the New York Zoological Park are included in a report of the external genitalia published by Goss (3). II. Subjects. The gorillas studied at the Bronx Zoo are Oka, female, and Makoko, male. Since this study was made the Bronx Zoo has acquired a young female mountain gorilla, but no reference to it will be made in this report, except to list it in Table I. Little is known of the early history of Oka and Makoko. Both animals arrived in the Zoo on September 7, 1941, at which time they weighed 20 and 28 112 Zoologica: New York Zoological Society [34: 13 Text-fig. 1. Rate of growth of Oka and Makoko from time of arrival at the New York Zoological Park on September 7, 1941, until summer of 1948. pounds respectively. Ages were estimated as one and three years on the basis of weight and bone structure. Text-figure 1 gives the data on the growth of the specimens from the time of arrival until the summer of 1948. The female was weighed, up to June, 1948, while on the keeper’s back, the male’s weight being merely estimated. In 1948 a Toledo balance was installed, on which the animals were weighed when they voluntarily mounted a platform. At the present time, Oka and Makoko are magnificent specimens. Makoko has a par- ticularly brilliant coat and a notably promi- nent supra-orbital ridge. Oka is less im- pressively marked but is a splendid female example of the species. Both animals are active, healthy and strong. Indeed, Makoko’s grip is so powerful that it has been necessary to replace the %-inch steel bars of the cage front, which he bent repeatedly. A steel hori- zontal ladder had to be removed from his cage when he tore it loose from its mooring. Oka is less destructive and retains her ladder. III. Environmental Conditions. The gorillas are housed in individual cages separated by a partition consisting of a solid metal access door and a double grill of steel bars. At the ends of each cage are doors leading to shift cages into which the animals are chased when the exhibition compart- ments are cleaned or repaired. The back wall of the living space is solid masonry with a recessed access door. Between the glass par- tition through which the public views the animals and the metal barred cage fronts is a passageway for the keepers. The internal features of the living compartments include a platform and a three-step staircase, with the platforms raised two inches from the floor, facing each other in front of the grill between the two cages. The staircases are at the ends near the shift cages. Oka’s cage contains a metal horizontal ladder slung between the rear wall and the cage front at a height of three feet. Text-figure 2 shows the floor plan of the gorilla exhibition space. IV. Daily Schedule of Animal Care. Although the animals are on exhibition during the hours of 10:00 A.M. to 5:00 P.M. (except on Sundays when the visiting time is extended to 6:30 P.M.), the daily routine is more extensive. The overhead fluorescent lights are turned on at 8:00 A.M. when the keepers enter the building. Between 8 :30 and 9:00 in the morning, the gorillas, each of whom has been in its own cage all night, are given their morning meal of skimmed milk and raw eggs. The ingredients are mixed and fed to each gorilla by tilting a can containing the mixture into the subject’s mouth as the animal protrudes its lips through the bars of the cage front. Some half-hour to three- quarters of an hour later, the keepers move the gorillas to the shift cages. Separation and enclosure in the individual cages are effected by means of a stream of water. With the gorillas out of the way, the cages undergo thorough cleaning. Shortly before visiting time, the animals are released and are fre- quently permitted to remain together in one of the two cages for periods up to one hour. This opportunity for association occurs five A . c • , — r — ' B 21' 9" X 8'9" t i B' _ , h-, »n_ jstci U“ * A' Text-fig. 2. Floor plan of gorilla compartments in New York Zoological Park. A, A', passageways for keepers. B, B’, exhibition cages for Oka (female) and Makoko (male), respectively. C, C', shift cages. D, three-step stairs. E, platforms; right-hand platform used for weighing. Thin line indicates solid wall; dotted line indicates grill-front walls; double line indicates glass partition. 1949J Riess, Ross, Lyerly & Birch: Behavior of Captive Lowland Gorillas 113 TABLE I. Physical Characteristics of Gorillas in the United States.* 1 Name. 1 1 Location. 1 Form. Sex. Estimated age on September 1, 1949. Known | Weight. | 1 Estimated Weight. Oka N. Y. Zool. Park | gorilla F 9 yrs. 282 lbs. | Makoko | N. Y. Zool. Park gorilla M 11 yrs. 408 lbs. | Sumaili | N. Y. Zool. Park berengei F 20 mos. 20 lbs. 1 Joanne | Central Park Zoo, N. Y. C. | gorilla F 10 yrs. | 190 lbs. Carolyn | Central Park Zoo, N. Y. C. | gorilla F 10 yrs. j 190 lbs. Bamboo | Philadelphia Z. G. gorilla M 23 yrs. | 435 lbs. Massa Philadelphia Z. G. gorilla M 18 yrs. 1 400 lbs. Bushman | Lincoln Park Zoo, Chicago | gorilla M 21% yrs. 542 lbs. | Sinbad | Lincoln Park Zoo, Chicago | gorilla M 20 mos. 38 lbs. | Rajah Lincoln Park Zoo, Chicago | gorilla M 29 mos. 1 1 47 lbs. Irvin Young | Lincoln Park Zoo, Chicago | gorilla M 33 mos. 52 lbs. | Lotus Lincoln Park Zoo, Chicago | gorilla F 43 mos. 1 75 lbs. Miss Congof | Chicago Z. P. gorilla F 16 yrs. 1 325 lbs. Phil St. Louis Z. P. gorilla M 10 yrs. 1 320 lbs. Bobo St. Louis Z. P. gorilla M 2 yrs. 1 1 44 lbs. Big Boy | Cincinnati Zoo gorilla M 3 yrs. 35% lbs. | Albert | San Diego Zoo gorilla M 6 mos. 9% lbs. | Bouba | San Diego Zoo gorilla F 10 mos. 12% lbs. | Bata | San Diego Zoo gorilla F 8 mos. 10% lbs. | Phil | Colorado Springs, Col. | gorilla M 4 yrs. | | 40 lbs. Gargantua | Ringling Bros. Circus | gorilla M 17 yrs. | 550 lbst | Mtoto Ringling Bros. Circus gorilla F 1 6 yrs. | 438 lbs.f 1 * On October 31. 1949, three young specimens of Gorilla g. gorilla arrived in New York. They were a male and a female weighing 12-14 pounds, and a female weighing about 40 pounds. At the time this paper went to press they were still in the hands of Henry Trefflich, an animal dealer, t Died Sept. 22, 1949. t 1947 weight. or six times a week and is the only occasion on which the animals are in unrestricted contact with each other. From the time of separation until the main feeding of the day at approximately 2:30 P.M., the gorillas are unattended except insofar as the keepers play with them while passing in front of the cages. The typical afternoon meal consists of carrots, celery, oranges, apples, grapes, bananas, beets, beans, sweet-potatoes, cab- bage, onions, cherries and other seasonable fruits and vegetables. All food is fed raw and supplied through the front bars of the cages. Water, ad lib., is available to each animal from a continuously running spigot which empties into a trough running along the back wall of each cage. At 5:30 P.M. the lights are turned out and the gorillas remain unattended until the next morning. V. Observational Procedures. Preliminary study of the gorillas was started at the beginning of May, 1948. One of the writers (B.F.R.) visited the Primate House at the Zoo on two days of each week and observed the gorillas for one hour. The times selected were in the early morning when the keepers placed the two gorillas in one cage, and in the late afternoon, at about 3:30 P.M. following the afternoon meal. At the later time, the animals were in separate cages. As a result of the work during May and early June, a list was evolved on which both quantitative and qualitative indices of be- havior could be noted. The inventory con- sisted of 42 items divided into five catego- ries : posture and locomotion ; eating, drink- ing and elimination ; self-oriented activity (play?); inter-individual behavior; and observer-oriented behavior. The sheets of the check list were divided into 10 columns, each of which was used for a three-minute period of observation during the total 30 minute duration of the observational session. Thus it was possible to arrive at the total amount of each of the 42 types of behavior during the 30-minute period and also to determine the sequence of behavior during the period. Ad- ditional space was provided for running com- ments on the activity of the gorillas and for additional notes. The observers worked in pairs and rotated the pairing so that a measure of control over reliability of observation was possible. There was joint discussion of the meaning of each term on the check-list so that the observers would agree on how to label the activity ob- served. The problem of the animal’s reaction to the observer was considered and it was decided to standardize the position of the observers. Since the gorillas are a very popu- lar exhibit and attract large crowds of spec- tators, the public exhibition space was used as the location for the observers who sat in pairs in front of the glass partition opposite 114 Zoologica: New York Zoological Society [34: 13 the communicating grill between the two cages. Because the activity of the gorillas varied considerably with the time of day, the num- ber of visitors, the daily routine and other variables, it was decided to distribute the 30-minute observation periods over the whole working day of the animal. However, since the gorillas were allowed to be in the same cage with one another only in the early morning, there was a greater concentration of sessions between 9 :00 and 10 :00 A.M. The total number of observations made during these hours was 10 one-half hour periods. Sixty-eight sessions were devoted to taking notes on behavior of the animals in isolation. During these 68 periods, it was possible to get data on both Oka and Makoko so that each animal was studied an equal number of times. Every hour between 10:00 A.M. and 6:00 P.M. was covered. In addition to these systematic observations which started in July and lasted through the middle of August, 1948, each experimenter observed the gorillas several hundred times while passing through the Primate House to and from other areas in the Park. Any deviant behavior or peculiar activities were noted and added to the record. VI. Results and Discussion. Two factors limit the analysis of the data. In the first place, quantitative analysis af- fords very little insight into the problem at hand. The purpose of this study was to obtain information which would serve as a starting point for further investigation of the gorillas at the Bronx Zoo, particularly as they become physiologically more mature and show active sex and social behavior. In the second place, it was thought desirable to point to both the similarities and differences between our gorillas and those at the San Diego Zoo described by Carpenter. In this comparison, there are many diffi- culties and dangers arising from two sources. In the first place, the gorillas belong to dif- ferent sub-groups, ours being lowland and Carpenter’s mountain specimens. Other va- riables in this category include age and sex differences between the two sets of observed gorillas. In the second place, it is necessary to stress the differences in the environmental setting in which the San Diego and Bronx gorillas carried out their daily activity. At San Diego, the two male gorillas were housed in outside cages equipped with tree trunks for climbing and various devices which could be manipulated by the animals, for instance logs, swings, ropes, tires, etc. In addition, the experimenter could insert objects into the gorillas’ surroundings and study the effect of such introductions. At the Bronx Zoo, the separation of the gorillas from the public was much more rigorous and the cages much more bare. In interpreting the comparative findings, the obvious individual and environ- mental differences must be kept in mind. However, despite these limitations, the com- parison of the two groups should be of value, if only to emphasize the danger of generaliz- ing from any set of observations. With the restrictions specified above, the data in Table II represent the basic observa- tions made upon the two gorillas at the Bronx Zoo. Where information comparable to that obtained by us was derivable from Car- penter’s San Diego observations, it has been included in the Table. The discussion of the data in Table II will follow the general cate- gories outlined above. A. Posture and. Locomotion. Posture and general locomotion seem fair- ly well established as invariant gorilla pat- terns. Both Carpenter’s mountain and our lowland gorillas exhibited the same type and frequency of gross motor activity. Walking, running, standing and sitting were charac- teristically alike for Oka, Makoko and the San Diego pair. Differences were noted in the frequency of observed sleep and in swinging by the hands. Both of these differences may be the resultant of variable environmental and observational procedures. It was not feasible for us to observe night behavior, and swinging was made difficult for Oka and Makoko by the absence of a place suitable for that kind of activity. The complete ab- sence of nest-building in the Bronx pair is also related to the lack of adequate materials. Both Carpenter and Yerkes report that it was a fairly common behavioral pattern in their subjects. The observations on handedness in the Bronx gorillas are not comparable to other studies since this item was not listed by Car- penter. Oka was observed to make differen- tial use of her hands on 313 occasions and Makoko 184 times. In both animals, the right hand was more frequently employed regard- less of the nature of the activity. The fre- quency of use of this hand was 54% as com- pared with the report by Yerkes (6) who found that Congo used her right hand some 66% of the time. In Congo’s case the left foot was preferred to the hand, whereas in our gorillas there was relatively little pedal manipulation. B. Eating, Drinking and Elimination. In the presence of an abundance of food, Oka and Makoko both showed a form of be- havior somewhat akin to the hoarding of rats and lower mammals. The gorillas would sweep the food into a heap with either the hands or feet. The heaped food was then ex- amined, tossed around or eaten. This behav- ior is not mentioned in any other report on the gorilla and may well be the unique result of the absence of manipulatible material in the cages in the Bronx Zoo. Placing of the longer-stalked fruits and vegetables on the heads of the gorillas was a frequent after- math of the in-gathering of the material. 1949] 115 Riess, Ross, Lyerly & Birch: Behavior of Captive Lowland Gorillas An interesting aspect of the behavior of the isolated animals was a relatively infre- quent passage of food from Oka to Makoko through the bai-s of the intercommunicating grill between the two cages. The initiation of this activity usually came from Oka and was noted particularly on the several occa- sions when Makoko had been deprived of his usual rations because of diarrhea or other health considerations. In such circumstances, Oka was observed forcing potatoes and cab- bage through the double grill. Makoko did not seem to be particularly interested in the inserted foods. This behavior was seen on three occasions. The drinking of water is a form of be- havior common both to our animals and to those studied by Carpenter. According to Yerkes, Congo drank but little water. The manner of drinking in the Bronx animals was to bend over the fountain and suck the water into the mouth. Although there was no opportunity to study comparative satisfac- tion from milk and water, it is the opinion of the authors that the milk-egg liquid was preferred. Regurgitation of the milk-egg mixture was almost invariable. Following the feeding the animals would typically squat on their haunches, lean forward, and regurgitate some if not all of the milk. The gorillas then would examine the liquid manually and eventually bend all the way and lick up the regurgitated material. The time interval after ingestion varied somewhat but was in the neighbor- hood of 45 seconds. Some chemical changes took place during the brief digestive stay for the milk was usually curdled. This pat- tern is seen not only in the gorillas at the Bronx Zoo but also in the chimpanzees. Whether or not this is a primate characteris- tic and analagous to the similar behavior of the pre-socialized human infant or whether it is unique to the subjects in captivity re- mains a matter for further research. Voided fecal material was commonly handled both by Oka and Makoko and fre- quently was used to throw at the keepers, the observers or the public. Handling was casual and seemingly tactually motivated, for the feces were not examined nor used except for throwing. Since the keepers made heroic efforts to keep the cages clean, the oppor- tunities for greater concentration on feces were limited. Carpenter makes no mention of this type of behavior. Urination was a casual affair and no localization of territory for this or for defecation was noted. C. Self-oriented Activity. Self-manipulation of parts of the body was a common form of activity in both Oka and Makoko. The parts of the body selected for handling or fingering were not consistent, with the exception of the lack of attention to or focus on the external genitalia. The major phase of activity during which handling was observed was while the gorillas were lying on their backs or stomachs, when parts of the body such as the lips, ears, eyebrows and nose would be held. The absence of genital ma- nipulation may well be the result of the small size of the external genitalia of the gorilla, as described by Carpenter and Goss. Manipulation of objects in the environ- ment is a frequent finding wherever gorillas have been studied. Carpenter and Yerkes mention this behavior pattern and it was noticeable in the animals at the Bronx Zoo. The female, Oka, showed some tendency to manipulate and examine with greater fre- quency than her cage mate. In the absence of a variety of objects to examine, the fre- quency of occurrence of this activity is all the more remarkable. The attitude of the gorillas toward the water fountain has already been described. An observable difference was noted in the behavior toward drinking water and that emanating from the pressure hoses used in cage cleaning. As indicated above, in the dis- cussion of daily routine, water was used to separate the gorillas and to urge them toward the shift cages. The initial reaction to the stream from the hoses was retreat and ex- citement. However, once wet, the animals would face into the water and jump up and down. The keepers reported that the animals would on occasion approach more closely to the nozzle of the hose when thoroughly wet. No shaking of the body after the bath was seen during the periods of observation. Self-grooming does not seem to be a domi- nant activity in the lives of either the San Diego or New York gorillas. To what extent the absence of this form of self-manipulation is a function of the cleanliness of the en- vironment and the animal is not established by our observations but, as will be noted later, grooming as a pattern of behavior is markedly less present in the gorillas under study than in other primates at the Zoo. D. Inter-individual Behavior. The data in this section were obtained during those periods when the animals were together in the same cage. To the extent that the opportunity for such interaction was limited, the enhancement of activity during the periods of joint occupancy of the cage may be a function of the limitation of time during which the two animals could interact. Both in Carpenter’s study and in ours, the major forms of inter-individual activity were running, chasing and wrestling. These be- havior patterns were well marked and almost stereotypical in appearance. Chasing was especially vigorous when the cage floor was wet and the gorillas spent much time sliding in a pronograde posture from one end of the cage to the other. Initiation of this activity was fairly evenly divided between Oka and Makoko. Wrestling, too, was not started con- sistently by either male or female. A domi- nance pattern was not apparent. The sequence of individual motor acts in 116 Zoologica: Neiv York Zoological Society [34: 13 TABLE II. Comparative Behavior of Oka, Makoko and San Diego Gorillas * Behavioral Classification A. Posture and Locomotion Walking, pronograde Walking, upright Running, pronograde Running, upright Standing, pronograde Standing, upright Sitting Climbing Swinging by hands Hanging Sliding Left-handedness Right-handedness Lying down Sleeping Nest Building B. Eating, Drinking and Elimination Gathering food in heaps Sharing food with cage mate Drinking water Drinking milk Regurgitation of milk Handling of feces Attention to urination C. Self-oriented activity (Play?) Self-manipulation (non-genital) Manipulation of genitalia Manipulation of objects Manipulation of food (non-eating) Attitude toward stream of water Self-grooming D. Inter-individual behavior Chasing Wrestling Grooming Inspection and manipulation of genitalia Presenting Mounting Pelvic thrusts Chest thumping Vocalizing Dominance E. Observer-oriented behavior Throwing of feces Throwing of non-fecal material Attentional responses Vocalization * Key to symbols : O— Never observed or reported H — Little in frequency or amount H — | — Some or fairly frequent Frequency of Occurrence in San Diego Oka Makoko + + + + + + + + + 4" + + + + + + + + + + + + ND 4" 4" + + + + + + + + + + 4- 4" + + + +++ + + + + + + + + + + + + 4~ 4- + + + + + + + ND 46% 46% ND 54% 54% + + + + + + + + + + + + 4" + NP NP ND + + + 0 + 4* + + + + + +++ + + + + + + + ND + + ++ ND + + ND + + + + + + + + 0 0 0 + + + + + + ND + + + + positive + + + + + + + + + + + + + + + + + + + + + + + + + + + + ■+■ + 0 0 4- 0 0 0 0 0 0 0 0 + + + + + + + + ND ND + + + ND + +.+ ND 4- + positive ++ negative -f+ negative +- + + 1 ND ND + -j — | — Great deal, very frequent ND— No data reported NP— Not possible in the environment the wrestling behavior was free of pattern- ing and seemed to consist of random grap- pling at the anatomical point nearest to the initiator of the behavior. When the actions of either animal seemed to approach the point at which roughness would ensue, the animal at the moment on the receiving end would detach itself and a period of resting would follow. Of all the behavior noted in this study, wrestling was the most dramatic and illustrative of the great strength of the gorillas. As a rule, there were few vocaliza- tions during the bouts. Social grooming, certainly, does not seem to be as predominant in the behavior of the gorillas as in the case of other representa- tives of the great apes or other primates. Carpenter likewise observed little of this sup- 1949] Riess, Ross, Lyerly & Birch: Behavior of Captive Lowland Gorillas 117 posedly socially oriented activity. It is pos- sible that in the four animals for which data are available, the age and sex differences were not sufficiently well established to facili- tate the appearance of this type of social interaction. It may also be possible that there is a real species difference in such be- havior. Genital manipulation and exploration in the paired situation was not frequently ob- served. It was seen only three times in Oka and Makoko and was not reported for the San Diego pair. In the Bronx specimens, the initiator was always the male. This may be a reflection of the relatively greater maturity of Makoko. So, too, in our gorillas, mounting and pelvic thrusts as precursors of mating behavior were almost completely absent dur- ing the periods of observation in New York. Carpenter saw none of this at San Diego, but his animals were both males. The one in- stance of pelvic thrusts by Makoko occurred during a wrestling bout and was not repeated nor invited by Oka. Chest-thumping was much more prevalent during the periods of paired activity than when the animals were in their own cages. The causal sequences leading up to the thumping could not be determined for Oka and Makoko although the observers were all of the opinion that the behavior was socially oriented and significant. Some writers have suggested that thumping of the chest is a sign of well-being and general euphoria. Yerkes states that the behavior is a sign of “impatience or other mild dissatisfaction.” There was clear evidence of neither causal sequence in our observations. The range of situations during which the thumping was observed varied so widely that no specific factor can be assigned as the reason for its existence. The only statement that can be made from our data is that the male, Makoko, engaged in chest-thumping more frequently than did Oka and accompanied the beating with vocalizations more frequently than his companion. In the absence of sound-recording devices, description of vocalization is difficult. Fur- thermore, the public space was somewhat soundshielded by the glass partition sepa- rating the animals from the observers. In our experience the occurrence of this activity was less than that mentioned by Carpenter. E. Observer-oriented Behavior. The types of audience-attentive behavior observed in the Bronx Zoo consisted mainly of the throwing of feces or food at the glass plate between the cage and the visitors. The same aggressive behavior was noted during an attempt to get photographs of the ani- mals. Visitors invariably tried to attract attention from the gorillas by tapping on the glass partition and by yelling. The effect of such devices was negligible. This difference between our data and Carpenter’s may well be the result of the more restrictive condi- tions of the gorillas’ environment in the New York Zoo. VII. Summary. This report describes the behavior of two pre-adolescent lowland gorillas in the New York Zoological Park during the summer of 1948. Oka, then an eight-year-old female, and Makoko, a ten-year-old male, were observed in their regular living cages when alone and when placed together. A check-list was prepared and regular half-hour periods of observation were sys- tematically made. Significant behavioral items were compared for the sessions when the animals were alone and when they were together. Comparisons were also made with the data collected by Carpenter from two male mountain gorillas in the San Diego Zoo. The descriptive material obtained during the period of observation should serve as a base line from which to note variations aris- ing from the maturation of the two gorillas in the years to come. References. 1. Bingham, H. C. 1932. Gorillas in a native habitat. Carnegie Inst. Wash. Publ., No. 426. Pp. 66. 2. Carpenter, C. R. 1937. An observational study of two captive mountain gorillas (Gorilla beringei). Human Biol., 9, 175-196. 3. Goss, Leonard J. 1947. The external geni- talia of the gorilla, Gorilla gorilla gorilla (Savage & Wyman). Zoologica, 32, 97-99. 4. Leister, Claude W. Gorillas. New York Zoological Society, Popular Series No. 4. 5. Valker, A. 1931. La vie du gorille au Gabon. Bull. Mus. Nat. d’Histoire Nat., 3. 6. Yerkes, R. M. The mind of a gorilla, (a) Genet. Psychol. Monogr. 1927, 2. 1-193. (b) Genet. Psychol. Monogr., 1927, 2, 375-551. (c) Comp. Psychol. Monogr., 1928, 5, 1-92. 118 Zoologica: New York Zoological Society [34: 13: 1949] EXPLANATION OF THE PLATES. Plate I. Fig. 1. Makoko, the male lowland gorilla in the New York Zoological Park. Esti- mated age, 11 years; weight, 408 pounds. Plate II. Fig. 2. Oka, the female lowland gorilla in the New York Zoological Park, is still friendly and gentle with her keeper at the estimated age of 9 years. Her weight is 282 pounds. Fig. 3. Oka playing with her keeper. ilESS. ROSS. LYERLY & BIRCH PLATE I FIG. 1. THE BEHAVIOR OF TWO CAPTIVE SPECIMENS OF THE LOWLAND GORILLA, GORILLA GORILLA GORILLA (SAVAGE & WYMAN). | ESS. ROSS. LYERLY & BIRCH. PLATE II. FIG. 2. FIG. 3. THE BEHAVIOR OF TWO CAPTIVE SPECIMENS OF THE LOWLAND GORILLA. GORILLA GORILLA GORILLA (SAVAGE & WYMAN). Beebe : Migration of Papilionidae at Rancho Grande 119 14. Migration of Papilionidae at Rancho Grande, North-central Venezuela.1 William Beebe. Director, Department of Tropical Research, New York Zoological Society. (Plate I; Text-figure 1). [This is one of a series of papers resulting from the 45th, 46th and 47th Expeditions of the Department of Tropical Research of the New York Zoological Society, made during 1945, 1946 and 1948, under the direction of Dr. William Beebe, with headquarters at Rancho Grande in the National Park of Aragua, Venezuela. The expeditions were made possible through the generous cooperation of the National Govern- ment of Venezuela and of the Creole Petroleum Corporation. [The characteristics of the research area are in brief as follows; Rancho Grande is located in north-central Venezuela (10° 21' N. Lat., 67° 41' W. Long.), 80 kilometers west of Caracas, at an elevation of 1,100 meters in the undis- turbed montane rain forest which covers this part of the Caribbean range of the Andes. The migration flyway of Portachuelo Pass, which is also the water-shed between the Caribbean and Lake Valencia, is 200 meters from Rancho Grande. Adjacent ecological zones include sea- sonal forest, savanna, thorn woodland, cactus scrub, the fresh-water lake of Valencia and various marine littoral zones. The Rancho Grande area is generally subtropical, being uni- formly cool and damp throughout the year be- cause of the prevalence of the mountain cloud cap. The dry season extends from January into April. The average humidity during the expe- ditions, including parts of both wet and dry seasons, was 92.4%; the average temperature during the same period was 18° C; the average annual rainfall over a five-year period was 174 cm. The flora is marked by an abundance of mosses, ferns and epiphytes of many kinds, as well as a few gigantic trees. For further de- tails see Beebe and Crane, Zoologica, Vol. 32, No. 5, 1947. Unless otherwise stated, the speci- mens discussed in the present paper were taken in the montane cloud forest zone, within a radius of one kilometer of Rancho Grande.] For an account of Portachuelo Pass, to- gether with a general introduction to the groups of migrating insects and migrating factors see “Insect Migration at Rancho Grande,” by William Beebe, Zoologica, 1949, Vol. 34, No. 12, pp. 107-110. In Volume 26 of Novitates Zoologicae, W. J. Kaye has a paper entitled “A Geogra- phical Table to show the Distribution of the American Papilios.” Under the heading “Venezuela, North,” (pp. 352-355), the au- thor lists thirty-one species. In a letter Dr. Rene Lichy of Caracas sends me a list of thirty-one species of this family which he 1 Contribution No. 852, Department of Tropical Re- search, New York Zoological Society. has collected in northern Venezuela. A con- tinuation of this coincidence is that each list contains seven species not found in the other list. Both lists contain all the species which we took migrating through Portachuelo Pass, with the single exception of crassus which Lichy does not mention. In the limited width of twenty meters of Portachuelo Pass, and allowing a height of net reach of a maximum of five meters, we captured seventeen species of Papilio. This area may be considered, not unrealistically nor unconservatively, as, at the most, a mil- lionth of the extent of north Venezuela. Yet within this relatively microscopic bit of Andean air, we secured almost half the pa- pilios so far recorded from the entire north- ern part of the country. Thus, in the con- sideration of this family of butterflies, we are made to realize the wide-spread, impel- ling, migrational force affecting this group of insect life. So much of this migration — its causes and extent — is at present unknown, that every verifiable fact is of value. Reviewing the known distribution of the seventeen species of Papilio migrants, we find that most of them extend from Mexico to Paraguay, south Brazil or Argentina. The distribution of the subspecies, however, presents a very differ- ent picture, and a very significant one in its over-all pattern. In twelve out of the seven- teen, the subspecific range is confined to Colombia and Venezuela, with a few exten- sions to adjacent territory. Thus we may expect to find the northern point of origin of the movement of these forms a relatively short distance away. The twelve subspecies of Papilio with lim- ited distribution are as follows : anchises osyris anchisiades anchisiades agesilaus agesilaus areas areas cleotas coroebus erithalion zeuxis lycophron hippomedon paeon thrason polyxenes americus protesilaus archesilaus sesostris tarquinius torquatus orchamus 120 Zoologica: New York Zoological Society [34: 14 The remaining five forms of Papilio with wider distribution are : belus varus crassus phaon polydamus polydamus thoas neacles Let us take as an example Papilio agesilaus agesilaus or, as I called it before identifica- tion, Small Zebra Swallowtail. I recorded seven individuals captured and fifty-eight seen with certainty. Added to this number were the many papilios glimpsed too briefly, flying too high or too fast to be recognized as to species, and also those which must have passed during the hours of our absence from the pass. It was heartbreaking to realize what a minute fraction we could honestly record by sight identification, yet there is no other way, at present known, to glean definite, general knowledge of this phase of the lives of these splendid insects. Without exception, all sight-named species were sub- sequently confirmed or discarded by compari- son with captured, definitely classified speci- mens. In this as in some other species, the small number of records from the year 1946 and their entire absence from 1945 by no means indicates the absence of the species on migra- tion, but only reflects our non-recognition of the importance of the Portachuelo Pass migration during the early years, and con- sequent slight attention paid to this phenom- enon. An ultimate summary of the relatively few, disconnected observations made during 1945 and 1946 reveals a general movement on a scale equal in magnitude and as all-em- bracing of insect orders as we recorded dur- ing 1948. The mere recording of the capture of seventeen species of Papilio on migration is a worth-while fact, and when more and more individuals are taken on suceeding days and weeks the phenomenon is enhanced in in- terest. In few or in large numbers the insects continue to fly past, slowly or circling or alighting out of reach. As in the case of many other organisms, the time has come when sight records must be used to supple- ment specimens in net, envelope and cabinet. Ornithologists in general and British ento- mologists in particular have gone far in sight identifications, while at the same time main- taining as perfect accuracy as possible with man’s fallible eyesight and only too human brain. At the end of our many months of collect- ing and observation at the pass, I found, in my Journal, a significant assemblage of shorthand names of papilios. They were, of course, essentially personal, stimulated by mental comparisons with swallowtails fami- liar to me elsewhere, or in the case of strange tropical forms, by outstanding wing shapes, size, patterns and colors. Lepidopterists recognize three “natural groups” into which papilios may be divided. Text-fig. 1. Map showing location of Rancho Grande, Portachuelo Pass and surrounding ter- ritory. , These are based on various factors such as larvae, pupae, microscopical imaginal dis- tinctions and/or food-plants. Our migrant species fit into these three groups as follows : Aristolochia : sesostris, erithalion, an- chises, areas, polydamus , belus and crassus. Fluted: polyxenes, thoas, paeon, lycophron, anchisiades, torquatus and cleotas. Kite: phaon, agesilaus and protesilaus. Consideration of this arrangement shows no logical, technical or scientific agreement. This is only to be expected in sight identifi- cation, which can take no account of sexual relationships, or parallelisms, or the superfi- cial resemblance brought about by mimicry. After final identification of the seventeen species of Papilio migrants, I arranged mounted specimens of all in a large insect drawer, placed this upright on a chair in good light and studied them from a distance of ten meters. From this distance I made the following key : A — White and Black a — Small: agesilaus agesilaus b — Large : protesilaus archesilaus B — Yellow and Black a — Yellow-banded Broad-band-plus-spot, small: tor- quatus, orchamus, male Broad-band, medium : lycophron hippomedon, male Narrow-band : paeon thrason thoas neacles 1949] Beebe: Migration of Papilionidae at Rancho Grande 121 b — Y ellow-spotted Small: polyxenes americus Medium: polydamus poly damus Large: cleotas coroebus C — Red and Black (Hindwing) a — Cream-spot-forewing sesostris tarquinius, female erithalion zeuxis, female anchises osyris, female areas areas, female torquatus orchamus, female b — Green-spot-forewing erithalion zeuxis, male anchises osyris, male areas areas, male c — Black-forewing anchisiades anchisiades D — Green and Black a — Green-forewing sesostris tarquinius, male b — Green-hindwing phaon E — Black (Dominantly) belus varus crassus lycophron hippomedon, female I compared this key with the names made up on the spot in the field, and found a grati- fying agreement in species recognition. The differences were chiefly substitution for pat- terns and colors of the names of northern species suggesting resemblances, species with which I had long been familiar in the eastern United States. For example, age- silaus agesilaus was “small ajax or zebra,” polyxenes americus was “small asterias,” and thoas neacles was “cresphontes-like,” etc. I cite all this as in no way directly possible or in the same detail applicable for use by another observer, but merely to show a framework upon which can successfully be erected an observer’s sight key. Further comments, in greater or less detail, will be found under the treatment of many of the species. My special thanks go to Mr. Henry Flem- ing, entomologist of our Department of Trop- ical Research, for many direct observations and for frequent corroboration of my own. In addition I am beholden to him for looking up distribution data and for painstaking identification of all the species. Papilio anchises osyris Felder. Species Range: Colombia to Brazil, Bolivia and Paraguay. Subspecies Range : Venezuela. Field Characters: Both male and female indistinguishable in the field from erithalion zeuxis. Therefore all specimens observed and not taken are combined under the two species. Compared with areas areas the male lacks bright green forewing spot, and the female has a decidedly larger, 4-celled fore- wing cream spot. Number and Sex: Both sexes taken; eight males, five females. Date: April 13 to July 29. Condition: All taken were fresh. Record of Captures: 1945 — July 15 (male), 18 (female). 1946 — April 13 (male, km. 20). 1948 — April 29 (male); May 1 (female), 31 (female); June 6 (male and female) ; July 17 (male, km. 30), 23 (male, km. 15), 29 (female at Pass, 2 males, km. 35). Combined Sight Records: anchises and erithalion : (Total 62). 1946 — May 27 (4 females) ; June 29 (2 females). 1948 — May 26 (4 females) ; June 4 (11 females passed in 10 minutes), 15 (4 females resting on shrubs), 22 (5 seen) ; July 2 (12 females), 9 (14 males, 3 females), 29 (3 females). Papilio anchisiades anchisiades Esper. Field Name: Red-spot Black. Species Range: Mexico to southern Brazil. Subspecies Range: Colombia to Bolivia and Para (Brazil). Field Characters : Black with red on hind- wing. Closest in field appearance to wholly black lycophron hippomedon. Number: 2 specimens taken, a male and a female. Notes: This black-forewing-red-hindwing papilio came through the Pass with what were taken and identified as female areas areas and anchises osyris, all captured to- gether. Record of Captures : A male in frightfully worn condition collected on April 13, 1945, No. 45456. A second individual, a female, taken at the Pass May 1, 1948, No. 48474. Papilio agesilaus agesilaus Guerin. Field Name: Small Black-and-white Zebra Swallowtail. Species Range: Mexico to Paraguay. Subspecies Range: Magdalena Valley, Co- lombia, to northern Venezuela. Field Characters: Unmistakable resem- blance to our northern ajax or zebra. The only optically related species is protesilaus arehesilaus, but the present species is much smaller (forewing 43 mm. as compared with 53 mm.). The difference easily recognizable when either species is close at hand or near other butterflies. Number: Total recorded 58. Seven taken (48543, 48731, 481344, 481494). Sex : Both sexes taken. Date: From April 29 to July 26. Frequency: Recorded on seventeen days. Moderately but markedly gregarious. Two- thirds of agesilaus were recorded in groups of 4 to 8, while one-third appeared singly or in twos. Flight: Except when alarmed or fighting against a strong head wind, the flight was unhurried and wavering, always steadily south. Condition: With one exception all that I saw hovering or resting at the Pass were in 122 Zoologica : New York Zoological Society [34: 14 good condition, with both long slender tails visible to the naked eye or through three- power binoculars. No. 481344, taken July 26, 1948, at kilometer 16, was badly muti- lated, with one tail missing. Dissection re- vealed that this male had already mated. Record: 1946 — May 26 (2 seen), 27 (2 at Pass, 3 at km. 20), 30 (6 singly). 1948 — April 29 (2) ; May 14 (1 taken), 21 (4 seen, 1 taken), 23 (1 taken), 24 (4), 26 (4), 31 (8) ; June 6 (1), 17 (2 taken), 19 (2) ; July 13 (4), 16 (2), 23 (1 taken km. 16), 24 (6), 26 (1 at Pass, 1 at km. 16) . Papilio areas areas Cramer. Field Name: Green-spot (male). Two- celled-cream-spot (female). Species Range: Mexico to Colombia, Vene- zuela and the Guianas. Subspecies Range: Venezuela and the Guianas. Field Characters : Closest to anchises and erithalion, from which it differs in the bright green forewing spot in the male, and the smallex*, two-cell forewing spot in the female. It is also close to the exceedingly rare tor- quatus orchamus. From the male sesostris tarquinius this species differs in the red on the hindwing. Number: Total recorded 277. Twenty-one taken. Sex: Both sexes taken. In 1946 only fe- males were seen or taken. In 1948 females were dominant from May 1 to June 17, and males from July 6 to July 22. Date: May 1 to July 22. Frequency: Recorded on 20 days: 1 (15 times), 2 (4 times), 4 (once), 6 (twice), 7, 13, 14, 16, 18, 29 and 85. Decidedly gregari- ous, occurring singly and in twos nineteen times, comprising one-fourteenth of the total number passing in larger numbers — from 4 to 85. As mentioned above, the sexes showed a decided segregation. The flocks were us- ually compact waves. Flight: Rather low and fluttering. Condition: Most of areas observed were in fresh condition, decidedly unworn. Extent of Migration: On four separate days specimens of this species were taken both at the Pass and at kilometer 20, well to the south, and at kilometer 35, half way to the coast to the north. Record: 1945 — July 3 (3 taken), 16 (2 seen, 1 taken, Limon). 1946 — May 28 (27 seen, 2 taken), 29 (7 at Pass, 6 at km. 20), June 4 (female), 22 (female) ; September 7 (84 in half an hour, 1 taken). 1948 — May 1 (female), 4 (6 females), 5 (male), 28 (fe- male) 29 (female) ; June 6 (2 males, flock of 18 females), 17 (16 females flying low), 22 (male and female), 28 (28 females) ; July 6 (2 males), 9 (male caught and eaten by bat falcon, 21 males seen), 14 (7 males), 15 (11 males, 3 taken), 16 (13 seen, four fighting in midair), 19 (4 males, 1 taken km. 15) , 22 (2 males, 1 taken km. 35) . Papilio belus varus Kollar. Species Range: Mexico to Bolivia and Para (Brazil). Subspecies Range : Guatemala to northern Venezuela and Ecuador. Dichromatic Female, form latinus Felder. Field Name: Greenish-hindwing-band Black. Field Characters: Black, with a curved band of large, yellow-green spots on hind wing. Number: Total recorded 19. Eight taken. Sex: Females only taken. Date : May 29 to August 4. Frequency: Taken singly. Five once seen together. Condition: Fresh. Record: 1946 — July 7 (1 seen km. 20). 1948 — May 29 (1 seen) ; July 4 (4 seen at 8:30 A.M.), 14 (1 taken), 16 (3 taken), 21 (2 taken, 3 seen), 26 (1 taken, 2 seen km. 16) ; August 4 (1 taken). Dichromatic Female, form varus Kollar. Field Name: Cream-spot-forewing Black. Field Characters : Irregular splash of yel- low in forewing; hindwing blue-black. Number: Total recorded 18. Three taken. Sex: Females only taken. Date: May 30 to July 26. Frequency: On two occasions, five were seen together. Condition : Fresh. Record: 1948 — April 29 (1 taken); May 30 (5 seen); July 8 (3 seen), 9 (2 seen alighted, 2 taken, 5 seen km. 31). Papilio eleotas coroebus Felder, form dione Rothschild and Jordan. Field Name: Large Asterias Swallowtail. Species Range: Costa Rica to Brazil. Subspecies Range: North Colombia and Venezuela. Field Characters : Rather like a very large Asterias, or polyxenes americus, with fore- wing 67 mm. as compared with 40 mm. A very distinct species. Number: Total recorded 19. One taken. Sex: The single specimen taken was a female. Date: May 26 to July 17. Record: 1948 — May 10 (4 seen), 26 (2 seen) ; June 6 (female taken, 2 others flying in company with three of the small polyxenes americus ) ; July 10 (2 at Pass, 2 at km.15), 17 (3 at Pass, 3 at km. 30). Papilio crassus Cramer, male foi’m lepidus Felder. Field Name: Black Philenor. Species Range : Costa Rica south to Brazil. Field Characters : Wholly black except for concealed bluish-white anterior border of hindwing. Number: Total seen 23. One taken. Record: A single male specimen of the form lepidus taken on July 21, 1948, No. 481538. Twenty more, distinctly seen, passed 1949] Beebe : M igration of Papilionidae at Rancho Grande 123 at the same time, all out of reach. On the fol- lowing day, July 22, two more of these black papilios were seen. No other record. Papilio erithalion zeuxis Lucas. Species Range: Costa Rica to Colombia and northern Venezuela. Subspecies Range: North Venezuela and Colombia. Field Characters: Both sexes indistin- guishable from anchises osyris. Differs from areas areas in male lacking forewing green spot, and female with larger, 4-celled fore- wing cream spot. Number: Total number taken 14. Sex: Both sexes taken. Date: June 29 to July 24. Condition : All freshly emerged. Record of Captures: 1945 — July 3 (female taken, Limon). 1946 — June 29 (female, km. 20) ; July 7, 8 and 10 (Each day 1 female taken, km. 20). 1948 — July 9 (3 males, km. 31), 14 (male), 15 (male), 17 (2 females), 23 (female, km. 15), 24 (female). For joint sight identification records with anchises osyris, see under latter species. Papilio lycophron hippomedon Felder. Field Name: Male, Broad-band Medium Turnus. Female, Black Troilus-like Swallow- tail. Species Range : Mexico south to Argentina and Uruguay. Subspecies Range : Colombia and northern Venezuela. Field Characters: Male can be confused only with the very rare, smaller, yellow-spot torquatns orchamus; female recalling a mel- anistic troilus or phaon with black hind- wings. Number: Total recorded 20. Eight taken. Sex: Both sexes taken. Date: May 10 to July 20. Record: 1946 — May 27 (2 at Pass, 2 km. 20, all males) . 1948 — May 10 (3 males seen) , 21 (female taken), 23 (male taken), 24 (male taken), June 6 (2 females taken), 17 (2 males taken), 29 (4 males seen); July 10 (male seen), 20 (male taken). Papilio paeon thrason Felder. Field Name: Rare Cresphontes-like Swal- lowtail. Species Range: Mexico south to Argen- tina and Uruguay. Subspecies Range: North Colombia and Venezuela. Field Characters: Indistinguishable in the field from thoas neacles, but as only a single specimen of paeon thrason was taken, com- pared with more than one hundred of thoas, I am assuming that all Cresphontes-like papilios observed were of the more abundant species. Record of Capture: A single male taken on May 23, 1948, No. 481539, in extremely torn and worn condition. It was captured at the Pass at 12:30 P.M., the day being warm and sunny with a Force 4 wind from the south. Papilio phaon Boisduval, aberration metaphaon Butler. Field Name: Philenor-like Swallowtail. Species Range: Mexico to Ecuador and Venezuela. Field Characters: This is the only black papilio with green on the hind wings. Number: Total recorded 254. Seven taken. Sex: Both sexes taken. Date: April 13 to July 21. Record: 1945 — May 24 (1 taken). 1946 — (September 9, numbers of these black pa- pilios with large green spot on the hind wings were flying too high to catch. Several alighted and allowed detailed study with Number three glasses. Counted 228 and missed many more.) 1948 — April 13 (male taken km. 20), 16 (male taken), 27 (male taken, km. 20), 29 (female taken at Pass) ; July 21 (2 taken, 19 seen). Papilio polydamus polydamus Linnaeus. Field Name: Medium Asterias Swallow- tail. Species Range : South Atlantic states, West Indies and south to Argentina. Subspecies Range: Georgia south to Buenos Aires. Field Characters: Differs to the eye from polyxenes americus in the field by the single instead of double line of yellow spots across all wings. Another distinction is the larger size. Number: Total recorded 177. Nine taken. Sex : Both sexes taken. Date: May 15 to July 26. Frequency: Decidedly gregarious. One- eighth passed singly or in a scattering up to five individuals. Seven-eighths were ob- served in flocks of ten to forty-eight. Record: 1946 — May 27 (2 seen). 1948 — May 15 (5 seen), 29 (female taken) ; June 6 (female taken), 22 (11), 30 (female taken, 48 passing, 2 seen at km. 21) . July 9 (female taken), 10 (2 males taken, 16 seen. 2 taken km. 31), 11 (14 flurry, 1 single), 14 (1 taken, 23 seen), 17 (10 seen km. 30), 18 (3), 19 (3), 22 (3 km. 35), 26 (27 seen). Papilio polyxenes americus Kollar, form melasina Rothschild and Jordan. Field Name: Small Asterias Swallowtail. Species Range: Canada south to Cuba and Peru. Subspecies Range: Colombia, Venezuela and northern Peru. Field Characters : Under a new name this proved to be the same species as our northern Asterias. The only other resembling migrant butterfly was the markedly larger polydamus polydamus. Number: Total observed 34. Although ob- served on ten occasions during two seasons, only three specimens were taken. 124 Zoologica: New York Zoological Society [34: 14 Date: March 25 to July 20. Frequency: 1, 1, 1, 2, 2, 2, 3, 3, 3, 6, 6. Note: Three perfect specimens on June 5, after rain, clung to the extreme ends of large leaves. The wings were flat and expanded, with the fore edge straight across so that the transverse band and spots were continuous. Record: 1946 — March 25 (1 taken, km. 21) ; April 19 (1 taken, km. 21) ; May 28 (2 seen at Pass). 1948 — April 29 (3 seen) ; May 4 (6 flying together), 28 (2 seen); June 5 (3 seen), 6 (3 seen) ; July 19 (6 seen), 20 (1 taken at Pass) , 22 (6 seen at km. 35) . Papilio protesilaus archesilaus Felder. Field Name: Large Zebra Swallowtail. Species Range: Mexico to Paraguay. Subspecies Range : Colombia, northern Venezuela and western Ecuador. Field Characters: Larger than (forewing 53 mm. as compared with 43 mm.) but in general similar to agesilaus. Size difference quite apparent when near, but not when flying high, away from other known butter- flies. At least fifteen individuals were not counted because of uncertain sight identifi- cation. Number: Total recorded 42. Three taken (48543). Sex: Males only were taken. The female seems to be quite unknown. Date: Recorded on migration from April 29 to July 19. Frequency: The relative gregariousness corresponds to that in agesilaus. More than five-sixths were in 4 to 12 groups, and six only seen as solitary or dual migrants. Condition : All observed in detail appeared fresh and perfect. Additional Notes: The flurry of 12 large zebras on May 30, were in a compact body, and at a time when neblina and rain, while light, were continuous enough to discourage all other migrants. Yet these great swallow- tails flew steadily at a height of about 12 feet, up to and through the Pass and down into the fog on the south slope. The actions of six which passed on May 26 were typical. All flew slowly and with slightly wavering flight at 10 feet, until I swooped futilely at them with the net when all swerved sharply out and down, two pene- trating the underbrush and working their way separately through the Pass before ris- ing into the free air again. Three were fol- lowed with the glasses far down the south slope. Record: 1946 — May 27 (1 seen, 1 at km. 20), 29 (4 seen, 1 taken). 1948 — April 29 (4) ; May 1 (3), 4 (1 taken), 10 (1), 21 (5 seen, 1 taken), 26 (6), 30 (12) ; July 19 (1 at Pass, 1 at km. 15). Papilio sesostrh tarqulnius Boisduval. Field Name: Male, Green-spot Black. Species Range : Mexico to Bolivia and cen- tral Brazil. Subspecies Range: Panama, Ecuador, northern Venezuela. Field Characters: Male to be confused only with male areas areas, but wholly lacks the hindwing red. Number: Total recorded 39. Three taken. Sex: Males only taken. Date: April 30 to July 20. Frequency: 1, 4, 5, 6, 23. Condition: All freshly emerged. Record: 1948 — April 30 (1 taken) ; June 10 (4 seen), 17 (6 seen), 22 (22 seen, 1 taken) ; July 20 (4 seen, 1 taken). Papilio thoas neacles Rothschild and Jordan. Field Name: Common Cresphontes-like Swallowtail. Species Range: Texas to Buenos Aires. Subspecies Range: Nicaragua to Ecuador, Venezuela, Trinidad and the lower Orinoco. Field Characters: Cresphontes-like. Un- identifiable, even at close range, from paeon thrason, but only a single specimen of the latter was taken in two years of collecting. Number: Total recorded 105. Nine taken. Sex: Both sexes taken. Date : May 4 to September 8. Frequency: Usually seen passing in small groups, five to eight, maximum sixteen. Strong flyers, difficult to capture, but occa- sionally alighting, affording opportunity for a good look. Record: 1946 — May 4 (male taken, km. 20), 27 (16 seen); July 7 (female taken, km. 20) ; September 8 (3 seen) . 1948 — April 29 (3 seen) , 30 (3, km. 26, headed for Pass) ; May 1 (4), 10 (2), 11 (2), 15 (4 seen, 1 taken), 23 (8 seen, male taken), 26 (6), 31 (1 seen) ; June 6 (3 taken, male, 2 females. Eggs protruding from females), 10 (5), 18 (8), 22 (6), 29 (4), 30 (5); July 2 (5 seen), 9 (2 at Pass, 6 km. 31), 13 (male taken), 19 (3 seen, 1 taken km. 18). Papilio torquatus or chamus Boisduval. Field Name: Male, Small Yellow-Band- and-Spot. Female, mimic of areas. Species Range: Mexico to Bolivia, Brazil and Paraguay. Subspecies Range : Colombia and northern Venezuela. Field Characters : Male somewhat similar to but smaller than male lycophron hippo- medon; female very close to female of areas areas. The large, separate, anterior yellow spot on forewing of the male distinguishes it from the solid band of lycophron. Number: Two males were seen, and two females taken. Sex: Both sexes seen, female only taken. Dates: May 1 to July 2. Record: On May 1, 1948, I watched two new papilios fighting in the Pass. One flew down and alighted just out of reach, and the other soon followed. I made a detailed de- scription of them, recording them as yellow- 19491 Beebe: Migration of Papilionidae at Rancho Grande 125 banded-with-spot, tailless papilio. Not until our return north were we able to identify the insects by comparison with a male tor- quatus taken at Caripito. On May 26 and again on July 2, 1948, a female was taken. These were badly rubbed and torn, whereas the males I saw were freshly emerged. 126 Zoologica: New York Zoological Society [34: 14: 1949] EXPLANATION OF THE PLATE. Plate I. Seventeen species of butterflies of the genus Papilio taken as migrants at Porta- chuelo Pass, Rancho Grande, north-central Venezuela. Fig. 1. sesostris tarquinius ( male). Fig. 2. sesostris tarquinius (female) Fig. 3. erithalion zeuxis (male). Fig. 4. erithalion zeuxis (female). Fig. 5. anchises osyris (male). Fig. 6. anchises osyris (female). Fig. 7. areas areas (male). Fig. 8. areas areas (female). Fig. 9. polydamus polydamus. Fig. 10. belus varus form latinus. Fig. 11. belus varus form varus. Fig. 12. crassus form lepidus. Fig. 13. polyxenes americus form melasma (male) . Fig. 14. polyxenes americus form melasina (female). Fig. 15. thoas neacles. Fig. 16. paeon thrason. Fig. 17. lycophron hippomedon (male). Fig. 18. lycophron hippomedon (female). Fig. 19. anchisiades anchisiades. Fig. 20. torquatus orchamus (male). Fig. 21. torquatus orchamus (female). Fig. 22. cleotas coroebus form dione. Fig. 23. phaon aberration metaphaon. Fig. 24. agesilaus agesilaus. Fig. 25. protesilaus archesilaus. :ebe. PLATE I. MIGRATION OF PAPILIONIDAE AT RANCHO GRANDE, NORTH-CENTRAL VENEZUELA. Smith: Notes on Ergasilus Parasites 127 15. Notes on Ergasilus Parasites from the New Brunswick, New Jersey, Area, with a Check List of All Species and Hosts East of the Mississippi River. Roland F. Smith1. Dept, of Zoology, Rutgers University, New Brunswick, N. J. Introduction. The members of the genus Ergasilus rep- resent the most undifferentiated of all the copepod parasites and clearly show a pos- sible line of evolution from the free-living forms to the very specialized parasites that are to be found in other families and genera. Even in the genus itself one finds the tran- sition taking place, and E. chautauquaensis (which has never been found as a parasite) to E. elegans, which is parasitic only after the eggs begin to develop, and on to the other forms where the females are parasitic after they have become sexually mature. The main characteristics of the genus are its cyclops-like appearance; 2nd pair of an- tennae enlarged and prehensile, 1st antennae six-jointed; first thoracic segment and head fused to form a large carapace; five pairs of swimming legs; the first four biramose, the fifth pair very degenerate and uniramose. Egg sacs are similar to those in Cyclops. The genus is typically fresh-water, though some ergasilids are to be found in brackish and marine waters. The type species of the genus is a very common European species, E. sie- boldi, established by Nordmann in 1832. These parasites are generally found cling- ing to the gill filaments, but one species, E. megaceros, has been found in the nasal fos- sae of the Fulton cat, Ictalurus anguilla, and another, E. elongatus, has been found at- tached to the bony gill rakers of the spoon- bill cat, Polyodon spathula. At the present time they have been found to infest all of the major groups of fresh-water fishes and it is likely that no species of fresh-water fish is entirely free from the possibility of be- coming a host. As far as is known, fishes are the only hosts on which this genus has ever been found. (See Mueller, 1936; Tidd & Bangham, 1945; Wilson, 1911, 1916, 1925, 1932; Wright, 1844.) The males of Ergasilus are free-swimming i 1 The author wishes to express his appreciation to Mr. Herbert Groat, who first called attention to the epidemic ; Dr. R. F. Nierelli for assistance in making this paper ready for publication ; and to Herbert Treuting and others who gave assistance when it was needed. throughout their lives. They are small and easily overlooked in plankton samples and consequently few of them have ever been de- scribed. The chief distinguishing character- istics in the males are the powerful maxilli- peds, which are lacking in the females, and the small, weak second antennae. The females are free-swimming during their early developmental stages and only at- tach themselves after mating. It is generally concluded (Wilson, 1911) that mating takes place only once while the female is still free- swimming. The sperms are stored in the semen receptacle and fertilize the eggs as they pass out into the ovi-sacs. The breeding season apparently extends throughout the summer months. The length will vary from season to season, or from one region to another, depending on favorable water temperatures. In this area females were observed with fully extended egg strings on March 31. In the laboratory the in- cubation period was found to be around eight weeks, which is about the period of time observed by Wilson (“eight to nine weeks”). Henderson (1926) in her paper on E. luciovercarium from Canada, stated that it is likelv that the females carrv their egg strings throughout the winter. This is not tha case in Westons Mills Reservoir, but it may be that in Canada, where the summer season is considerablv shorter, the last batch of eggs does not get the chance to develop before cold water temperatures come and consequently must be carried over until the warmer temneratures of spring. In New Brunswick the breeding season is over by the middle of November. Observation on Epidemic of Ergasilus on Fishes in the Westons Mills Reservoir, New Brunswick, N. J. During the latter part of November, 1947, fishermen began to notice large numbers of fish dying in Westons Mills, a reservoir from which New Brunswick obtains its water supply. Local residents estimated that the fish were dying by the “thousands,” and in- deed the number of dying fish was so great 128 Zoologica: New York Zoological Society [34: 15 as to attract large flocks of sea gulls from the nearby Raritan River. The fish affected were apparently only two species, calico bass, Po- moxis sparoides, and bluegill, Lepomis ma- crochirus. Some of these fish were brought to the Rutgers University Zoological Laboratory where they were examined for possible para- sitic infestations. No intestinal parasites were found in excessive numbers nor did there appear to be any injury of body tissues, either external or internal. The gills ap- peared to be covered with an unusually heavy coating of mucus and microscopic examina- tion revealed great numbers of copepod par- asites of the genus Ergasilus. The blue color of these organisms, the three knobs on the inner edge of the second antennae towards the distal end, and the fact that they were all found in between the gill filaments seemed to indicate conclusively that these were of the species caeruleus. The hairs and spines on the appendages, along with other general morphological characteristics, were not quite in accord with Wilson’s caeruleus, but there seemed to be a considerable amount of varia- tion in this genus, depending on geographic location (Mueller, 1936). Wilson (1911) considers caeruleus a parasite of the vege- tative Centrarchidae, and gives the explana- tion that the copepods on these fishes must locate themselves between the gill filaments to escape the discomfort and irritation to their gills from bits of vegetation. A number of trips were made to the reser- voir to collect fish during this period. Oxygen determinations were made at all depths and at no time was the O2 concentration less than 10 ppm. Most of the fish were alive when cap- tured. They could easily be spotted as they swam feebly on their sides on the surface of the pond. It was possible to come up along- side these fish in a boat and pick them up by hand. Occasionally a fish would sound on ap- proach, but only to rise slowly to the surface after a short interval. On the first of such trips about a dozen fish were collected and taken back to the laboratory. These were placed in a well-aerated aquarium. At first they seemed almost dead — all lying motion- less on their sides, with only an occasional movement of a fin. The next day, however, they seemed fully recovered and all were swimming about apparently quite normal. A few of these examined at this time were found to have a very heavy infestation. In brief, the meager facts obtained on the epidemic from Nov. 27 to Dec. 10, 1947, can be summarized as follows : 1. Many calico bass and bluegills died during this period. No official estimates could be obtained, but laymen who observed the phenomenon estimated the deaths to be in the thousands. 2. One calico bass died for every ten blue- gills. These were apparently the only fish affected. 3. All fish that were infested by these para- sites appeared to have a heavy coating of mucus over the gills. 4. Estimated number of parasites on each fish was 250-300. 5. Age group (as determined by scale read- ings) was 1-2 years. During the late fall of 1948 these observa- tions were continued. A large fish trap was constructed and set in deep water as soon as ice covered the reservoir. All of the species of the Centrarchidae as well as one yellow perch were examined. Later in the year the fish were obtained by seining and from fish- ermen. In this survey the large-mouth bass, Mi- cropterus dolomieu, was found to be para- sitized by a second species, E. centrarchida- rum. This species is generally larger, broad- er, and found on the outside of the gill fila- ments, in contrast to caeruleus which is found between the gill filaments. In addition, there are no knobs on the inside surface of the distal end of the second pair of antennae. Again there were morphological differ- ences from Wilson’s description of centrar- chidarum and this phase of the work will be discussed in a later paper. Both species had the blue pigment but centrarchidarum was never as deeply pigmented as caeruleus. Strangely enough, both species had the same hair and spine formula on their appendages. This is: 1st exopod, 1-0; 0-1; II-5, endopod 0-0 ; 0-1 ; II-3 : 2nd exopod, 0-0 ; 0-1 ; 0-6, end- opod, 0-1 ; 0-2 ; 1-4 : 3rd exopod, 0-0 ; 0-1 ; 0-6, endopod, 0-1; 0-2; 1-4: 4th exopod, 0-0; 0-5, endopod, 0-1; 0-2; 1-3. In addition to the sampling of fish from Westons Mills, two other bodies of water in the New Brunswick area were samoled. These included the lower section of the Dela- ware-Raritan Canal and the small pond in Johnson’s Park. The bluegills and calico bass in Johnson’s Park were found to be free of ergasilids but the canal proved as fruitful as the reservoir. A summary of the fish ex- amined is given in Table I. Some interesting facts are foreshadowed in Table I. Although the amount of sampling from both bodies of water was not extensive enough to reveal fully the actual condition, nevertheless the methods by which these fish were obtained and the period of time over which the sampling was made certainly in- dicate the trend that one might expect to find if a more extensive sampling were to be taken. For example, only five large-mouth bass were obtained — but under totally different conditions and at various times. The fact that all five had infestations does not indi- cate that all the bass are nai'asitized. yet one would expect to find a high percentage of the bass serving as hosts to these parasites. On the other hand, not once during this entire period of research has there ever been found a pumnkinseed infested with these copepods. Certainly one can sav that for this species the incidence of parasitism is very low. Again, viewing the information in the same light, one should expect the bluegills to be parasitized about 50% of the time in f Cali Bto Pui Lai Yel W ca bi sf h' ii P t i ( 1949] Smith: Notes on Ergasilus Parasites 129 TABLE I. Result of the Samplings from Westons Mills and Delaware-Ra>’>tan Canal from Nov. 30, 1948- July 30, 1949. Name Total No. Percent of Average Length Caught parasitism Length Range Calico bass* 5 100% 5.79" 4"— 7" Bluegills 15 60% 5" 2"-7%" Pumpkinseed 6 0% 4" 3"— 5" Large-mouth bass 5 100% 6.4" 2%"-12' Yellow perch 1 0% 7" 7" * The highest infestation noted was on one calico bass 5'”K" in length : 259 caeruleua were counted on this fish. Westons Mills and the Canal. No data on calico bass could be obtained from the Canal but from the available information it would seem that these fish have become the most heavily infested in Westons Mills, not only in number of individuals but in number of parasites per fish. This is significant, for al- though the calico bass was found to be the most heavily infested as well as the most commonly parasitized fish, the epidemic of 1947 killed only one calico bass to ten blue- gills. Unfortunately, here also, too little in- formation is at hand. On the basis of the present data, however, there seems to be two logical explanations for this: 1. The calico bass are more resistant to the infestation of these gill parasites than are the bluegills and have been able to build up a resistance. 2. All the bluegills left are those that have built up a resistance to these parasites, or those that were only slightly parasi- tized. Both fish are to be found in the same gen- eral type of environment and both have simi- lar food habits, so apparently these two fac- tors may be discounted. Pathogenesis. There has been a great deal of controver- sy over the extent of damage done by the ergasilids, as well as exactly what kind is done. Wilson (1911) states that “Living as they do upon the fish’s gills, there can be but little doubt that they feed upon blood.” Halisch (1939) in his observations on E. sie- boldi and E. minor on the gills of tench states that extra-intestinal digestion is im- portant and that much more tissue is de- stroyed than is ingested by the parasitic co- pepod. Blood may be taken in. Fungi may grow in the lesions. In contrast to these two observations, Hen- derson (1926), in her work on E. lucioper- carium, observed that “The gills may be heavily attacked without affecting the health of the fish. It is a harmless parasite, which, while it only benefits the unbidden guests, causes no lesions and consequently does no injury to the host.” (If this is the case it should not even be called a parasite, but a commensal). She goes on to explain how, in her opinion, it is impossible for the parasites to injure the gill tissue, since the mouth parts of these organisms are too weak to pierce the gill tissue. One might suspect that the claw-like, pre- hensile, second pair of antennae of the para- site is capable of at least some damage to the gill lamellae, although the mouth parts may be too small to cause any injury. However, the observations in the present studies showed no evidence of mechanical injury; neither was there any evidence of digested gill tissue or blood in the intestine of the numerous copepods examined, even in fe- males with ovaries full of developing eggs. We agree with Henderson (1926) that these parasites in all probability feed on the excessive mucus produced under the condi- tion of the parasitism, or the many minute mucus or bits of organic debris and bacteria in this viscid material. However, it is alto- gether possible that the parasites may feed on sera, straining out the corpuscles. A close examination of the mouth parts of these ergasilids will reveal the mandibles and second maxillae heavily fringed with setae. These would seem ideal for feeding on mucus or bits of organic debris and bacteria but hardly suitable for feeding on blood. Any pathological conditions of the gills may cause the mucus glands to secrete an excessive amount of mucus (Nigrelli, 1949). This has been especially apparent during infestations of trematode gill parasites. This condition has also been observed on many of the fish that were infested with these copepods2. It may be that an excess mucus secretion over the surface of the gills may lower the efficien- cy of the gills to absorb the dissolved oxygen in the water. Under normal water conditions when there is sufficient oxygen and all the chemical and physiological factors are in proper balance, this may not have any dele- terious effects. However, if, for example, a factor such as the CO2 concentration in the water should be increased, it might be suffi- cient to reduce the efficiency of the gills in absorbing the dissolved oxygen and so cause the fish to die of suffocation. The highest death rate has been found among the younger fishes and it is the - It must be kept in mind that the heavy coating of mucus which seems to accompany fish that are heavily infested with this parasite does not mean too much in itself. Fish that have been placed in preservative or that have died from other causes may show the same condition. Moreover, some fish that were heavily infested did not have an excess of mucus covering the gills. In such cases these fish exhibited no symptoms whatsoever and appeared per- fectly normal. Apparently, there is a physiological balance here that is very delicately adjusted and which can be thrown out of balance only under certain specific conditions. 130 Zoologica: New York Zoological Society [34: 15 TABLE II. Check List of the Ergasilus Found East of the Mississippi River. Names and Synonyms Hosts Localities Studied Remarks E. caeruleus Wilson Synonyms: Yellow perch Trout perch Wall-eyed pike Gray pike, Blue pike Rock bass, Calico bass Lake Mendota Trout Lake region Lakes Erie & Michigan Oneida Lake Mississippi River, Iowa E. confuscus Bere White bass Warmouth bass E. skryabini Mueller Crappie Green sunfish Blue-spotted sunfish Pumpkinseed Bluegill Lake trout Cisco, White fish Sucker, Long-nosed gar E. centrarchidarum W right Rock bass. Calico bass, Large-mouth bass, Small-mouth bass, White bass, Warmouth bass, Green sunfish, Bluegill, Crappie, Pumpkinseed, Sunfish (?) Wall-eyed pike, Gray pike, Sauger, Silversides, Smelt, Microgadus tomcod Lake Erie Lake Michigan St. Lawrence River Watershed Lake Champlain Black Lake, N. Y. Lakes St. John & Chibogamo, Quebec St. Andrew’s Bay, N. B. Lake Maxinkuckee, Ind. Mississippi River, Iowa Clewiston, Fla. Woods Hole E. chautauquaensis Fellows Lake Champlain Lake Mendota Fairport, Iowa Has never been found as a parasite, but may be like elegans E. cotti Kellicot Rainbow darter Sculpin — Cottus bairdii Lake Erie Westerville, Ohio E. elegans Wilson Ameiurus sp. Northern black bullhead Channel catfish Short-nosed gar Long-nosed gar Lake Okeechobee Myakka River & Canals Peace River, Fla. Mississippi River, la., Parasitic only after eggs begin develop- ing E. elongatus Wilson Spoonbill cat Mississippi River, Iowa and Illinois E. funduli Kr0yer Fundulus ocellaris New Orleans Salt and brackish water E. labracis Kr0yer Striped bass Woods Hole Baltimore Marine E. lanceolatus Wilson Gizzard shad Cumberland River, Ky. E. lizae Kr0yer Common killifish Gulf killifish Broad killifish Striped mullet White mullet Englewood, Fla. New Orleans Salt and brackish waters E. luciopercarum Henderson Pike perches Lake St. John Lake Chibogamo, Que. E. manicatus Wilson Silversides Smelt Two-spined stickleback Top minnow ( Gambusia holbrooki ) Jordanella floridae Englewood St. Andrew’s Bay, N. B. Woods Hole Along Atlantic Coast Marine 1949] Smith: Notes on Ergasilus Parasites 131 Names and Synonyms Hosts Localities Sttidied Remarks E. megaceros W ilson Synonym : Fulton cat Fall fish Oneida Lake, N. Y. Mississippi River, Iowa Found in the nasal fossae and spiracles E. fragilis Mueller E. mugilis Vogt Striped mullet Beaufort, N. C. Marine E. nigratus Wilson Large-mouth bass Mississippi River, Iowa E. osburni Tidd & Bangham Burbot North Central States E. versicolor Wilson Synonym : E. celestis Mueller Fulton cat Red-mouthed buffalo fish Channel cat Common brown bullhead Mudcat Skip jack \ ellow cat Eel cat Mississippi River, Iowa Florida Lake Erie Lake Maxinkuckee younger fishes that would require the great- est amount of oxygen (in proportion to the gill area), due to their greater metabolic activity. At the same time it is the younger fisnes that are more susceptible to attacks from disease and parasites since they are using all their available energy toward growth. Wilson states that it is the young nsh that are most heavily parasitized and this is probably true — especially during an epidemic. However, larger fish may also be quite heavily parasitized. (See Table I). 'the ergasihds on the large-mouth bass were never very abundant — never more than thirty on any individual. Although Wilson (1916) mentions fish fatalities from the ergasilids he does not mention which species of copepod causes death, nor which species of fisn are killed. In checking all the litera- ture on centrarchidarum, never were their numbers found to be as great as for caeru- leus. Therefore, one wonders if caendeus is not the only one that may appear in such numbers as to bring about the death of a fish. This leads us to speculate on how many ergasilids must be present on a fish to cause death. It would seem that numbers that lead to the death of a fish at one time, appear to have no effect at another. However, during the epidemic of 1947 none of the dead fish had less than an estimated 250 copepods. Whether fewer parasites can bring about the death of a fish remains to be determined by further study. It is apparent from the literature that E. centrarchidarum is the most widespread par- asite and probably the best known. It has been found in all the main regions studied, including the marine habitat, but it has not been found on as many hosts (16) as has caeruleus (19), nor on as great a variety. Caeruleus has not been found in all the areas that have been studied and so far has proved to be an exclusively fresh-water par- asite. Additional research may also reveal that caeruleus is more widespread than cen- trarchidarum. It is unfortunate that most of the workers in this field have failed to make clear whether caeruleus has always been found between the gill filaments of its host. If it is found between the gill filaments of such clear-water fishes as the lake trout, cisco and white fish, certainly Wilson’s ex- planation for their being found between the gill filaments is not substantiated. E. versicolor is apparently third in abun- dance and seems to be more specific in its par- asitism. Many of the Ergasilus have only been found on one or a few hosts, but it is still too early to state definitely that they are spe- cific for only those hosts. Conclusions. 1. It seems quite likely that Ergasilus may in- directly cause death to fish, although it probably seldom causes extensive damage to any given fish population. 2. Young fish (1-2 years) are apparently the only ones on which these parasites may prove fatal. Although infestations have been found to be equally high on all age groups, more young fish appear to be heavily infested than older ones. 3. Ponds where the fish are overcrowded or that are small in area seem to be the only places where infestations become high enough to cause death. 4. Ergasilids do not feed on blood, but prob- ably on mucus or bits of organic matter found in the viscid material. 5. Death of fish is probably due to a delicate physiological imbalance affecting the diffu- sion of oxygen through the gill tissues. 132 Zoologica: New York Zoological Society [34: 15: 1949] 6. E. caeruleus can always be distinguished from E. ceritrarchidarum by the three knobs on the second pair of antennae, and by the fact that it is always found in between the gill filaments. Variations among these cope- pods are exceedingly great. 7. Although Wilson considers E. caeruleus a parasite of the plant-eating Centrarchidae, it is also found on species that are quite carnivorous and not generally found among aquatic vegetation. It undoubtedly is found most commonly on fishes that typically in- habit an environment of aquatic vegetation, such as the bluegill and calico bass. 8. Though ceritrarchidarum and caeruleus have often been found on the same fish (Wilson, 1911, 16), they were never observed together in the New Brunswick area nor was cen- trarchidarum ever observed on either blue- gills or calico bass in this area. Neither was observed on any but the above species of fish. 9. It may be that certain factors, chemical, physical, physiological and environmental, or a combination of these, determine which hosts may be parasitized. 10. No reason could be found to explain why the pumpkinseed, Lepomis gibbosus, is free from these copepod parasites in this region. Summary. An epizootic in Westons Mills Reservoir which caused the death of many bluegills and calico bass was investigated. A copepod parasite, Ergasilus caeruleus, was believed to be the indirect cause of the death of these fish. Apparently, metabolic wastes from the copepod or irritation to the gill tissues causes an excessive secretion of mucus. This mucus may lower the efficiency of the gills, so that when certain other fac- tors are not in proper balance the fish will suffocate. Large-mouth bass were found to be in- fested with another species E. centrarchida- rum. A brief survey of two other bodies of water in the New Brunswick area revealed that the incidence of parasitism from these ergasilids is quite high for bluegills, calico bass and large-mouth bass, but does not seem to be present on the pumpkinseed. References. Halisch, J. 1939. Anatomie und Biologie von Ergasilus minor. Zeitschr. Parasitenk., 11 (2/3) : 284-330. Henderson, Jean T. 1926. Description of a Copepod Gill Parasite of Pike Perches in Lakes of Northern Quebec, Including an Account of the Free-Swimming Male and Some De- velopmental Stages. Cou.tr. Canadian Biol. & Fish., N. S. 3 (7) : 235-246. Mueller, J. F. 1936. Notes on Some Parasitic Copepods and a Mite, Chiefly from Florida Fresh Water Fishes. The American Midland Naturalist, 17 (5) : 807-815. Nigrelli, Ross F. 1949. Notes from Lectures on Fish Diseases. Tidd, W. M. & Bangham, R. V. 1945. A New Species of Parasitic Copepod, Ergasilus osburni, from the Burbot. Trans. Amer. Micr. Soc., 44 (3) : 225- 227. Wilson, C. B. 1911. North American Copepods Belonging to the Family Ergasilidae. Proc. U. S. Nat. Mus., 39: 263-400. 1916. Copepod Parasites of Fresh- Water Fishes, and Their Economic Relations to Mussel Glochidia. Bull. U. S. Fish- eries, 34: 331-374. 1924. New North American Parasitic Cope- pods, New Hosts, and Notes on Cope- pod Nomenclature. Proc. U. S. Nat. Mus., 64 : 1-22. 1932. Copepods of the Woods Hole Region, Massachusetts. Bull. U. S. Nat. Mus., No. 158: 1-623. Wright, R. R. 1884. Notes on American Parasitic Cope- poda. Proc. Canad. Inst., N. S. 1:243- 254. Aronson: Reproductive Behavior in Tilapia macrocephala 133 16. An Analysis of Reproductive Behavior in the Mouthbreeding Cichlid Fish, Tilapia macrocephala (Bleeker).1,2 Lester r. Aronson. Department of Animal Behavior, The American Museum of Natural History. (Plates I-III; Text-figures 1-10). Introduction. Teleosts of the family Cichlidae are noted for their elaborate patterns of courtship, mating and parental care, and for the read- iness with which they breed in the restricted confines of the small aquarium. It is largely because of these attributes that cichlids have become the subjects of several extensive in- vestigations of fish behavior. Outstanding among these studies are those of Breder (1934) on the blue acara, Aequidens lati- frons ; Noble and Curtis (1939), Peters (1941) and Seitz (1942) on the jewel fish, Hemichromis bimaculatus ; Peters (1937) on the small Egyptian mouthbreeder, Haplo- chromis multicolor; and Seitz (1940) on a closely related mouthbreeder, Astatotilapia strigigena. These students have investigated topics such as schooling, sex recognition, courtship, territory, social dominance, spawning, pa- rental care, the stimuli causing the release of various innate responses and many other related items of behavior. In these studies, mating behavior has been described qualita- tively and in varying degrees of detail. Al- though the reports in most cases have been based on a number of observed spawnings, the results are given in a generalized or “av- eraged” form and the only suggestion of va- riability in behavior is found in such broad phrases as “this usually happens,” or “the typical mode of behavior is.” Moreover, the “averaging” is often achieved by means of subjective impressions rather than in terms of a calibrated or objectively weighted eval- uation of behavioral characteristics. Variability is a fundamental characteristic of biological phenomena, a characteristic 1 The experiments herein reported were supported by a grant from the Committee for Research in Problems of Sex. National Research Council. 2 Mrs. Magda Schonwetter assisted in many of the ob- servations. Drs. Frank A. Beach and Charles M. Breder made numerous helpful suggestions on the conduct of the observations and experiments. Dr. Myron Gordon helped solve the feeding problem. Dr. T. C. Schneirla, Mr. James W. Atz, Mrs. Marie Holz-Tucker and Mr. Christopher W. Coates read the manuscript and made innumerable con- structive criticisms. The author gratefully acknowledges his indebtedness to these people and those past and present associates of the Department of Animal Behavior who in ways too numerous to mention made possible the comple- tion of this study. which always merits careful consideration in studies of animal behavior. It is the writ- er’s belief that the study of teleost behavior cannot extend very far beyond the present descriptive stages unless and until methods of a more quantitative nature are employed. Students of mammalian psychology, and par- ticularly of rodent behavior, have made ex- cellent progress by utilizing quantitative procedures. The present study afforded an opportunity for testing the applicability of comparable techniques in the study of the reproductive behavior of fish. The present report is concerned with the average behavior and the range of varia- bility under constant aquarium conditions of a type which can be readily duplicated. In other investigations now in progress, the mating activities of brain-operated and hor- mone-treated animals will be compared to the norms obtained in the present report. Nothing appears to be known concerning the mating behavior of Tilapia macrocephala in the wild state, and the present report is hardly intended as a substitute for such an investigation. Nevertheless, wherever the be- havior of fishes has been studied both under field conditions and in captivity, agreement has been fairly good, as for example in the Centrarchidae. It is anticipated that the over-all picture obtained in this study should prove to be essentially similar to conditions prevailing in the natural state, and that dif- ferences if any would be expected only in some of the lesser details. Literature. Information concerning the breeding hab- its of Tilapia macrocephala and of related species belonging to the same genus has been furnished for the most part by aquarium hobbyists and through cursory observations by field naturalists. It is realized that because aquarists’ reports often fall below generally recognized standards of scientific accuracy, as might be expected considerable confusion exists in the literature concerning certain aspects of the breeding patterns of Tilapia. Some of these difficulties may no doubt be attributed to an improper identification of 134 Zoologica: New York Zoological Society [34: 16 the species in question, since aquarists some- times trust the knowledge and dependability of fish dealers for the identity of their sub- jects. Nevertheless it is possible to obtain from this literature a rough picture of the reproductive habits of the genus Tilapia. For these reasons the inclusion of numerous aquarists’ accounts is considered expedient. What may be offeied herein is by no means intended to be a comprehensive review of the extensive popular literature. Brief descriptions of the breeding habits of Tilapia macrocephala can be found in the aquarium texts of Stoye (1935), Arnold and Ahl (1936) and Innes (1944). In addit.on, reports on the spawning of Tilapia lieude- loti (which according to Boulenger (1915) may be a variety of Tilapia macrocephala) have been presented by b>reaer and Schoenfeld (1934). These accounts tell us briefly that (1) a nest is built by the mating pair; (2; the female deposits the eggs in the nest; (o) tne male fertilizes the eggs as soon as tney are deposited; (4) shortly thereafter the male taxes the eggs into his mouth; and to; tne eggs natcn anu develop in tne mouth of the maie. now long the eggs are retained in the maie’s buccai pouch is not indicated, but Stoye (1935; reports two cases where eggs were carried 24 to 29 days respectively. Stoye considers these periods abnormally long as a result of excessive disturbances. With four probable exceptions, all the species of Tilapia whose spawning habits have been reported are mouthbreeders. The four exceptional species remove their larvae to sand pits in typical cichlid fashion. These non-mouthbreeding species are (1) Tilapia guinasana (Rolon, 1938, 1939), (2) Tilapia spaarmami (ney, 1945, 1941; Anon., 194S)J, (3) Tilapia melanopleura (Svenssor, 1933; Bertram, Borley and Trewavas, 1942; and (4) Tilapia zillii (Bade, 1923; Stoye, 1935; Arnold and Ahl, 1936; Bertram, Borley and Trewavas, 1942). However, Liebman (1933) describes Tilapia zillii as a mouthbreeder. Incubation of the eggs is accomplished by the females of Tilapia flavomarginata (Pella- grin, 1906) i, Tilapia gaMlaea (Pellegrin, 1903, 1905), Tilapia martini (Boulenger, 1906), Tilapia microcephala 5 (Junghans, 1918) and Tilapia mossambica* * 4 5 6 * 8 (Bade, 1923; Dietz, 1926; Roloff, 1937; Peters, 1937a, 1939; Seleuthner, 1941; Hey, 1947). The same appears to be true for Tilapia squami- pinnis, Tilapia lidole, and Tilapia shirana (Bertram, Borley and Trewavas, 1942). The male is credited with the care of the eggs in Tilapia dolloi (Asch, 1939), Tilapia heudeloti (Breder, 1934; Schoenfeld, 1934), 8 Also recorded in "Report No. 1 (1944) Inland Fish- eries Dept., Union of South Africa”— 1946. 4 Designated by Boulenger (1911) as Tilapia andersonii. 5 Tilapia microcephala — Tilapia heudeloti. According to Boulenger (1916, p. 178) Tilapia macrocephala, and Tilapia multifaeciata “may ultimately have to be regarded as varieties of T. heudeloti. I am unable to find characters by which to separate them sharply.’’ 8 Tilapia mossambica = Tilapia nataleneis. Tilapia macrocephala (Stoye, 1935; Innes 1944), Tilapia microcephala (Schreitmiiller, 1920) and Tilapia simonis 7 (Lortet, 1875; 1883). However, there is some disagreement on this point since both the male and female are believed to incubate the eggs in Tilapia simonis (Pellegrin, 1903; Liebman, 1933), Tilapia galilaea (Liebman, 1933), Tilapia microcephala (Locke, 1932), Tilapia nilotica (Boulenger, 1901) and Tilapia zillii (Lieb- man, 1933). Bodenheimer (1927) claims that females alone incubate the eggs of Tilapia simonis, and Arnold and Ahl (1936) claim the same for Tilapia dolloi. Irvine (1947) states that the male or pos- sibly both sexes of Tilapia discolor and Tila- pia heudeloti incubate the eggs, but con- trary to the findings of Boulenger, Irvine relegates this function to the female in Tila- pia nilotica. Liebman (1933) believes that it is quite general in Palestine cichlids for both parents to incubate the eggs, but the number of females performing this function is higher than the number of males so doing. The length of the incubatory period has been reported for only a few species. Arnold and Ahl (1936) say about 14 days for Tilapia dolloi; Schreitmiiller (1920) gives 4 to 6 days for Tilapia microcephala; Roloff (1937) reports 21 days for Tilapia mossam- bica; Bade (1923) offers a value of 15 days while Dietz (1926) and Seleuthner (1941) both give 13 days as the incubatory period of this species. The retrieving into the female’s mouth of newly released young has been reported for Tilapia dolloi (Arnold and Ahl, 1936), Tila- pia macrocephala (Stoye, 1935), and Tilapia mossambica (Roloff, 1937; Seleuthner, 1941) while in Tilapia microcephala the male is credited with that activity (Schreitmiiller, 1920). Nest making by these mouthbreeders has received some general attention. On a num- ber of occasions, Lortet (1883) witnessed the female Tilapia simonis lay approximately 200 eggs m a small excavation which she had hollowed out and cleaned in the silt among the reeds. Nest building by both the male and the female Tilapia nilotica was observed in the field by C. L. Boulenger (1908 ) . Roloff (1937) describes the nest of Tilapia mossam- bica as being 20 cm. in diameter. Seleuthner (1941) reports a nest for this species which was 25 cm. in diameter and reached a depth of 4 cm. in the middle, while Hey (1947) pictures it as a “small saucer-shaped depres- sion.” Bertram, Borley and Trewavas (1942) describe the nest of Tilapia squamipinnis as a circular depression. Other mouthbreeding cichlids are listed by Peters (1937) as belonging to the genera Astatotilapia, Ectodus, Geophogus, Hap- l Placed in the genus Tilapia by Boulenger (1899) but now referred to a new genus Tristramella by Trewavas (1942). Lortet (1876) called this fish Chromis pater- familias. 1949] Aronson: Reproductive Behavior in Tilapia macrocephala 135 locliromis, Pelniatochromis and TropheusA It is to be noted that at least some of these genera also contain non-mouthbreeding spe- cies, suggesting a multiple origin of this habit even within the cichlid family. This problem has been considered in some detail by Breder (1933) and Myers (1939). The small Egyptian mouthbreeding cich- lid, Haplochromis multicolor, and a closely related form, Astatotilapia strigigena, have been the most popular and intensively studied of all the mouthbreeding fish. In addition to the scientific investigations previously men- tioned, more than 30 accounts of the spawn- ings of these two fish have appeared in the last three decades, the majority of them in the Wochenschrift fur Aquarien-und Terra- rienkunde. These accounts, which are rela- tively consistent in their general implica- tions, demonstrate that the spawning be- havior of these species differs considerably from that of the various species of Tilapia described above. For this reason what is known about the Haplochromis and Astato- tilapia mating patterns is summarized brief- ly for the purposes of comparison. In these species the male does practically all of the nest building. Upon the completion of the nest, the female starts the oviposition by depositing between four and ten eggs in tne nest. Tne maie immediately fertilizes the eggs after which they are picked up by the female. This cycle is then repeated as the female lays a second batch of eggs. Be- tween five and ten such cycles have been re- ported by various authors as comprising a spawning. The eggs are carried for 9 to 20 days, after which the young are released. However, the young are taken back into the female’s mouth at night and at other times when disturbed. Such a retrieving of the young has been the subject of a special in- vestigation by Peters (1937). Material and Methods. Tilapia macrocephala (Bleeker) is native to West Africa, particularly in the region of the Gold Coast. Boulenger (1915) de- scribes the species as coming from the Gold Coast, Ashantee and Lagos. Many of his specimens were taken from the Ancobra river and Secconda lagoon in the Gold Coast, and from the Lagos lagoon. According to Arnold and Ahl (1936) the fish is found in the brackish lagoons of the coast and the swampy deltas of rivers. The individuals utilized in this study were selected from a laboratory-bred stock which had been maintained for a number of years prior to the start of the present research.8 9 Males were chosen for the brightness of their yellow operculum which is a secondary sex 8 An older listing of mouthbreeding cichlids given by Pellegrin (1903) includes the genera Geophaaua. Aeara, Chaetobranch.ua, Tilapia, Paratilapia, Pelmatochromis, Ectodu8 and Tropheus. 9 I am greatly indebted to Miss Ethelwyn Trewavas of the British Museum for kindly checking and verifying the taxonomic identity of the fish as Tilapia macrocephala. character (PI. I, Fig. 1). The females (PI. I, Fig. 2) were selected on the basis of the complementary sex character, namely a deep red spot in the center of the gill cover.10 * * These dimorphic color patterns appear at sexual maturity and disappear after castra- tion (Aronson, in manuscript) . Pairs were established by random selec- tion and were placed in 54-liter aquaria, 60 cm. X 30 cm. X 30 cm. each containing roughly 36 liters of water. The side and rear walls of these tanks were painted pale blue to minimize any possible disturbing influ- ences from neighboring tanks, and also to facilitate the ability of the investigator to follow the activities of the fish. The tanks were located in a greenhouse the temperature of which was maintained throughout the year at 26° C. with a positive and negative variation of approximately 3° C. To furnish hide-outs for tne fish which at the same time would not obstruct the observer’s view, a mat of floating plants was placed in every tank. Cabomba was extensively used for this purpose, but Sagittana subulata was found to be somewhat more suitable and was used whenever available.11 The fish were fed mostly a dehydrated preparation consisting of dried shrimp, oatmeal, beef liver, lettuce ana spinacn. At times tms was supplemented by live tubifex worms. Occasionally the fish nibbled at the stonewort Niteila, and this was piaced in the tanks when available. The tanks were aerated continuously, and the water was changed whenever it became ex- cessively murky. This was approximately once a month. Tap water brought to the proper temperature was used in washing the tanks and for replacement. In order to avoid injury to the fish due to excessive nipping which often occurred after spawning, a transparent glass partition was placed in the aquarium, separating the male from the female as soon as observations of oviposition were completed. As the individ- ual carrying the eggs (generally the male) eats little or nothing during the incubatory period, brooding fish were not fed during this interval. By the time that the young were released from the male’s mouth, the females often were prepared to spawn again. However preliminary observations indicated that when such spawnings occurred males sometimes behaved abnormally, due appar- ently to the protracted period of inanition. To avoid this difficulty an arbitrary rule was established to the effect that males were separated by a transparent glass partition from females for one week after they had released their young or had swallowed their eggs. This interval allowed the males to feed and regain their strength. While thus iso- 10 Examination of this spot by Aronson and Holz-Tucker (in manuscript) has revealed that it is in actuality a semi- transparent window through which the underlying red gill can be seen. 11 The author wishes to express his appreciation to Dr. Myron Gordon of the New York Zoological Society for furnishing all of the sagittaria used. 136 [34: 16 Zoologica: New York Zoological Society lated, females often spawned alone but ob- servations showed that a considerable amount of courtship took place through the glass partition. To prevent the parents from eating their young after they were released, large masses of the stonewort Nitella were placed in all tanks where young were being incubated. The stonewort was distributed equally on both sides of the partition since the newly released fry could easily swim through the cracks at the intersection of the partition and the glass walls of the tank. Disturbances caused the young to swim into the fine inter- stices of the Nitella where they would not be followed by their cannibalistic parents.12 The criteria employed to indicate the ap- proach of oviposition were (1) persistent nest building, mostly by the female, (2) heightened courtship activity, and (3) pro- truding genital tubes. When these signs were observed, continuous records were taken of the courtship and mating activities of the pair up to the time of spawning and for one-half hour thereafter. Attempts were made to secure continuous pre-spawning records for three hours. However, this goal was attained in only a small number of cases with the result that the records vary from just a few minutes before spawning up to the full three-hour span. A serious difficulty was encountered here in that the activities of many promising pairs were observed continuously for many hours up to a whole day without the fish ever ovipositing. After the present experiment had been terminated, behavior during the interspawn- ing interval was studied, using different pairs of Tilapia. The experimental condi- tions were the same as before with the fol- lowing minor exceptions : (1) no plants were used but instead inside aquarium filters served as hide-outs; (2) the water in the tanks was never changed; (3) the males 12 This was the author’s first experience with the main- tenance of tropical fish. Since then, several innovations have been developed. Inside aquarium charcoal filters are now placed in every tank. These keep the water clean and eliminate the need for chancing it. Plants are not used since they grow poorly in Tilapia aquaria. The food formulae have been modified as follows: (1) Wet mash: 2V> lbs. liver ; % lb. chopped lettuce and spinach : Va lb. dried ground shrimp (mostly shell) : % lb. dried and ground refined shrimp (mostly muscle) : Pablum (or other pre- cooked infant cereal) —enough to make thick paste (ap- proximately 3 lbs.) ; 1 pinch salt. The liver is chopped, about 1 cup of water added, and the mixture is then liquefied in a blending machine. All ingredients are mixed together with sufficient Pablum to make a paste. The food is further solidified and preserved by packing into jars and immersing them in boiling water for about 10 minutes. (2) Dry food: 12 lbs. dried shrimp (mostly shell) : 12 lbs. dried refined shrimp (mostly muscle) ; 10 lbs. liver ; 6 lbs. chopped lettuce ; 6 lbs. chopped spinach ; 28 lbs. Pablum ; 2 level teaspoons salt. The ground spinach and lettuce are mixed with the Pablum and cooked for 15 minutes. The liver is cut into slices and boiled for 15 minutes in a minimum amount of water and then chopped. All ingredients are mixed together and the resulting paste spread about % inch thick on trays. When almost com- pletely dry, the food mixture is ground and sifted through screens of several coarsenesses. The sexes are no longer separated after the spawning. If the fry are to be saved they are forcibly removed from the parents’ mouths on the tenth day post-oviposition, and are placed in small aquaria. At this age the young do very well without further parental care, and thus, losses through cannibalism are easily avoided. were never separated from the females. The actual spawnings of these pairs were not witnessed, all ovipositions being recorded as having occurred at the time the eggs were discovered in the male’s mouth. All pairs were checked twice daily for eggs. A 15- minute record of the behavior of a given pair was taken 5 or 6 days after the spawning and again on the 15th or 16th day. The in- terval between successive spawnings varies from 8 days up to almost a year with a mode of 15 days (Aronson, 1945). Approximately two-thirds of the intervals are less than 29 days. Thus the 5- or 6-day score serves as an intermediate record for the shorter inter- spawning intervals, while the 15- or 16-day score serves in the same capacity for the longer intervals. Obviously some of the 15- day records could not be taken because of intervening ovipositions. Many of these ob- servations served, moreover, as behavior records for varying days before spawning. Qualitative Description of Reproductive Behavior. In order to furnish the reader with the proper background for the quantitative in- vestigation, it is appropriate to present first a general description of mating activities. This account does not take into consideration the question of the range of variability and any exceptional items of behavior. Details concerning many of the generalizations made here will be considered in the next section. Certain of the behavioral patterns which increase in their frequency of occurrence prior to spawning and which lead up to the acts of oviposition and fertilization gener- ally are classified as courtship activity. Such behavior appears to express the level of sex- ual excitability of the given individual. In accordance with the views of Huxley (1914, 1938), Howard (1929) and Marshall (1936), it is assumed that courtship tends to hold the pair together, and through mutual stim- ulation may lead to a well synchronized spawning. In the terms of Schneirla’s (1946) discussion, such relationship may be thought of as involving trophallactic processes, and the temporal aspects of these interactivities are of significance from the standpoint of adaptive function. Tilapia eggs (PI. I, Fig. 2; PI. II, Figs. 3, 4), as well as those of other oviparous teleosts, are shed in a flaccid state, but rapidly become hard and turgid upon entering the water. That is, they “water harden” (Breder, 1943). Hence, to insure fertilization the male must deposit his sperm over the eggs within a very short time after they are laid. An adequate synchronization of the pair’s reproductive processes thus appears to be critically important for effec- tive species survival. There follows a description of the early courtship behavior of Tilapia macrocephala : (1) The male and female approach each other and suddenly dip their heads; or one member of the pair lowers its head. This 1949] Aronson: Reproductive Behavior in Tilapia macrocephala 137 behavior has been termed “head-nodding.” (2) When one member of the pair approaches the other, spreads its opercula and expands its buccal pouch, we have called this act a “throat-puff.” (3) The male or female ceases swimming movements and the trunk muscu- lature appears to quiver for a fraction of a second. We have named this a “body- quiver.” (4) When one member of the pair slaps the other with its tail, this has been called a “tail-slap.” Included in this category were the frequent cases where tail-slapping motions were quite distinct, but where ac- tual contact with the partner was not made. Closely associated with the courtship acts, but displayed as well in many pairs through- out the interspawning interval, is a mode of behavior which we have termed “nipping.” This occurs when a fish swims after its part- ner, and •‘-hen with a sudden dart nips or bites the body of its mate. Nipping also occurred at times without a previous chase. Sometimes the male and female may nip each other simultaneously and occasionally they may even lock jaws. Frequently observed cases in which the pursuing fish darts ahead but misses the lleeing opponent also have been included under the general heading of nip- ping. in addition to its association with court- ship ana spawning, nipping behavior appears to be related to the establishment of social hierarchies and the formation of territories. These further relations of nipping have not as yet been investigated. The above-mentioned patterns of behavior usually appear as quite distinct, but occa- sionally they tend to merge into one another, so that discretion on the part of the observer is often called for in assigning a particular courtship act to its proper category. Often- times two or more courtship acts may be displayed in rapid succession, a frequent combination being a head-nod, throat-puff and body-quiver. Another commonly occur- ring combination is the throat-puff and tail- slap. i l Headers acquainted with the courtship be- havior of other cichlid fishes will readily recognize the resemblance of the Tilapia pat- tern with those of other cichlids. Reactions such as the body-quiver, the throat-puff and the tail-slap in some form seem to be preva- lent throughout the family. Nest-building is first observed after in- tensive courtship has been in progress for several hours or days. Most of this activity is conducted by the female who begins scoop- ing up mouthfuls of gravel from scattered locations in the bottom of the tank. Soon the excavating is confined to one location, and the construction of a nest begins. Often two or more nests are constructed prior to the spawning, and sometimes nests are built and then destroyed during the construction of an adjoining nest. The nests are most often round or slightly oval. If the gravel sub- stratum of the aquarium is not too thick, the fish dig down to the slate bottom of the tank. If, however, the depth of the gravel is more than 2 or 3 cm., the nests do not reach the slate. In our study the length of time taken to complete a nest varied considerably from as little as one-half hour up to what appeared to be several days. In the latter case, the nest- building activity occurred in spurts, followed by periods of quiescence. The rapid builders generally worked continuously until the nest was completed. A small amount of nest-build- ing was accomplished by sweeping move- ments of the tail and pectoral fins. However, this has been interpreted as incidental to swimming and balancing movements and not directly related to nest building. After the nest is more or less completed, nest-building decreases considerably and is supplanted to some extent by nest-cleaning (PI. I, Fig. 1) in which the female, and occa- sionally the male, pick continuously at the bottom of the nest. Nest-building and nest- cleaning are always interspersed among var- ious phases of courtship responses. As the nest takes form, the genital tubes of the male and female become more promi- nent. At this time, the male begins to swim slowly over the nest, rubbing his genital tube over the bottom. We have called this “passing-nest.” Later when the female com- pletes the nest, she likewise “passes-nest.” Thus the pair circle around and around, rub- bing their genital tubes over the nest. This behavior is often interrupted by periods of courtship, nest-building and nest-cleaning activity. As soon as a fish starts passing-nest, the genital tubes become fully erected, but if this activity ceases for a time, the tubes generally recede somewhat. This suggests that mechanical stimulation is one factor causing the erection of the genital tube. Since fish not on the verge of spawning are sometimes seen with partially extended geni- tal tubes, other stimuli seem to be involved in the partial erection of the genital tubes. Courtship activity and hormones are sug- gested as possible factors. After the passing-nest behavior of the male and female has been in progress for some time, the female stops in the nest dur- ing a “pass-nest,” and her body musculature quivers for a second or two. This has been called a “spawning-quiver.” Males also ex- hibit spawning-quivers, but in the male these responses generally are less distinct and are seen less frequently. Spawning-quivers were the final pre- spawning acts and indicated the imminence of the oviposition. During one of these quivers a batch of approximately 10 to 20 eggs is extruded by the female in what we have termed an “oviposition movement” (PI. I, Fig. 2). The female then swims just past the nest. She is followed by the male who passes-nest, usually rubbing his genital tube over the newly laid eggs and sometimes exhibiting a spawning-quiver. This com- plementary act has been termed a “fertili- zation movement” (PI. II, Fig. 3). Sperm 138 Zoologica: New York Zoological Society [34: 16 apparently are emitted at this time, although no male products were apparent in our ob- servations. The female then repeats her ovi- position movement which is followed closely by a second fertilization movement of the male. After two to four such egg-laying cy- cles, the female swims rapidly from the nest for a distance of 15 to 30 cm., then faces the nest. Meanwhile the male swims in the vicin- ity of the nest for a minute or so, then rapidly picks up the eggs with his mouth (PI. II, Fig. 4; PI. Ill, Fig. 5). In our investigations there were a few exceptional cases where the male did not pick up the eggs. Then the female nipped and tail-slapped the male violently, exhibited some courtship behavior, and finally after 10 to 20 minutes of this activity she picked up the eggs and carried them in her mouth (PI. Ill, Fig. 6). We found that at times, because of the unequal sizes of the male and female, all the eggs could not fit into the male’s mouth, the female would pick up the remaining eggs, but not until 10 to 20 min- utes had elapsed. Post-spawning activity consists for the most part in poking around the nest, first by the male, and later by the female as well. If any of the eggs are missed when the orig- inal spawn was picked up, they are almost always recovered during this poking activity. After several minutes, this poking be- havior sometimes gives way to extensive nipping and mouthing in which one member of the pair, generally the female, soon dom- inates and the other retreats into hiding. The eggs hatch in 5 days and are carried from 2 to 15 days further, during which time the embryos continue to develop. The young are released abruptly and most of them are sufficiently developed at this time to suggest that further parental care would not be advantageous to them. Parental care appears to end suddenly with the release of the young. In fact, parents sometimes eat their newly liberated offspring. Never did we see the young swim back into the parental mouth as has been described by some authors for this and other Tilapia species, and which is such a striking characteristic of the small Egyptian mouthbreeders, Haplochromis multicolor (Peters, 1937). With this brief description of the mating pattern we turn now to an analysis of the actual counts made of the frequency of occur- rence in relation to the time of spawning of many of the behavioral acts described above. Analysis of the Mating Pattern. For the purpose of analyzing the data, records were organized in the following manner. For each observed spawning, the time of appearance of the first batch of eggs was designated as the zero minute. The 15- minute period just prior to the zero minute was called the first pre-spawning interval. The period 15 minutes to 30 minutes prior to the zero minute was named the second pre-spawning interval. Twelve pre-spawning intervals were similarly measured. Again starting from the zero minute, the 15-min- ute interval which followed was called the first post-spawming interval, and a second post-spawning interval was likewise meas- ured. The number of times that the various courtship and mating activities (tail-slaps, passing-nest, etc.) were recorded during each 15-minute pre- and post-spawning in- terval was determined for both the males and the females for all observed spawnings. With data assembled in this manner, a series of distributions was obtained (one of each behavior pattern of both the male and female for each interval) . Almost all of these were strongly skewed to the right. The arithme- tical mean obviously is a poor representation of the central tendency of a markedly skewed distribution. Medians are generally more suitable, but a better method of treating such data is to employ a transformation. In many cases by use of the transformation X = vi, binomial distributions were ootained which could be treated as normal curves.13 These were checked by plotting cumulative distributions on arithmetic probability pa- per. However some of the distributions were not normalized following the above trans- formation, but approximated closely the Poisson series. This was particularly true with infrequently occurring items, where the highest frequency was zero and where the mean was considerably smaller than one. Theoretical Poisson distributions were cal- culated from Pearson’s (1914) tables and the goodness of fit of the actual distributions was tested by the chi-square method. Still other distributions did not approxi- mate either the normal or Poisson series. As will be noted later, these were not subjected to further statistical analysis. For the normal distributions, the means, range, theoretical range (M ± 3<0, standard deviation and standard error of the mean were calculated. These were plotted graphi- cally in a time sequence, using the method of comparing ranges and means developed by Dice and Leraas (1936) as modified by Simpson and Roe (1939). Since the length of the pre- and post-spawning records varied inadvertently for each spawning, the calcu- lated means for each interval are based upon a varying number of spawnings. Simpson and Roe (1939) point out that the method of Dice and Leraas is less reliable when the frequencies and standard errors of the mean vary greatly. Therefore, in critical cases where the graphic method was suspected of being inaccurate, P values were calculated. The solid lines in Text-figs. 1-6, 8 and 9, in- dicate females; the broken lines males. The heavy vertical lines designate actual ranges of the distributions. The adjacent light verti- cal lines indicate theoretical ranges (M ± 3<0. The large dots represent the means, while the short horizontal lines above and IS The writer wishes to acknowledge the aid given by Dr. Charles P. Winsor in suggesting the use of this transformation. 1949] Aronson: Reproductive Behavior in Tilapia macrocephala 139 below the means indicate the range of M ± 2<*m. When these ranges overlap, it may generally be assumed that the differences between the means are not significant. Con- versely, if M ± do not overlap, the dif- ferences between the means are significant. The limitation of this method has already been noted. For the Poisson series, the theoretical ranges were considered to run from zero to that value of the variate having a relative frequency of .003 or less. Means were com- pared by the method described by Snedecor (1946). Because of the asymmetrical nature of the Poisson distribution the graphical method of comparing means described above cannot be used. Hence, in the following graphs, the range of M ± 2.10). On the 5th or 6th post-spawning day, and on the 15th or 16th post-spawning day, the body-quiver frequency of the males was very close to that of the females. Of the 25 spawnings in which continuous records were available for one hour before the spawning, 100% of the females and 80.0% of the males exhibited body-quivering at least once. In the seven ovipositions in which continuous two-hour pre-spawning records were taken, 100% of the males gave body-quivers at least once. Thus while this courtship pattern is exhibited more frequent- ly by the females, practically all males show some body-quivering activity prior to the egg laying. Tail-Slaps. With the transformation pre- viously described, the data of both the male and female were found to be distributed in a Poisson fashion, excepting the first two post-spawning intervals of the female which were normally distributed (Text-fig. 3). On the several days prior to the spawning occa- sional tail-slapping by the female was ob- served, but at three hours before the egg laying, the occurrence of this behavior had increased considerably. This level was main- tained until the spawning, when there was another significant rise (P<.01) during the first post-spawning period. No tail-slapping by the male was observed on the several days prior to the spawning, and during the three-hour pre-spawning in- terval, the frequency of tail-slaps remained low with relatively little fluctuation. Ap- proximately this same frequency was ob- served during all the post-spawning obser- vation periods. Of the 25 pairs for which continuous rec- ords for the first hour were obtained, 100% of the females were recorded as tail-slapping at least once, but only 48% of the males. In the seven spawnings with two-hour continu- ous pre-spawning records, 71.4% of the males tail-slapped at least once. Head-Nods. Following the transforma- tion, head-nodding data for the female was characterized by a large number of zero fre- quencies and a small number of rather high frequencies. These did not fit Poisson dis- tributions. At times, head-nodding was not clear cut and easy to recognize, and it is possible that a considerable amount of head- nodding passed unrecognized. Before the spawning, head-nodding activity was quite high (Text-fig. 4), at least for some of the 142 Zoologica: New York Zoological Society [34: 16 DAYS BEFORE SPAWNING (ONE 15 MIN. INTERVAL ON EACH OF ABOVE DAYS) 15 MIN INTERVALS BEFORE SPAWNING 15 MIN INTERVALS AFTER SPAWNING 35 36 DAYS AFTER SPAWNING (ONE 15 MIN. INTERVAL ON EACH OF ABOVE DAYS) Text-fig.5. Fluctuation in male and female nipping behavior before and after spawning. females, and there was a still further rise after the egg laying. The data for male behavior fit Poisson dis- tributions quite closely. The low mean values indicate that this behavior occurred rather infrequently, and little fluctuation was no- ticeable befoi'e or after the spawning. Of the 25 ovipositions from which con- tinuous records were taken for the first pre- spawning hour, 84.0% of the females and only 20.0% of the males exhibited head- nodding at least once. Similarly, of the seven pairs where two-hour records were available, 100% of the females and 28.6% of the males head-nodded at least once. We may conclude that head-nodding is a typical female activity and that a small fraction of the males head- nod occasionally. Nips. Nipping data of both the male and female were treated as Poisson distributions. Both sexes displayed some nipping behavior on the several days they were observed be- fore the spawning (Text-fig. 5). During the three-hour pre-spawning observation period, approximately the same amount of nipping was shown by both the males and females. After the egg laying there was a significant rise (P<.01) in the nipping frequency of the females. The rise in female nipping dur- ing the ninth pre-spawning interval may be significant (P = .05), but it was mostly due to a marked spurt of activity of a single female. 1 1 \ Analysis of the 25 spawnings where com- plete records for the first hour before spawn- ing were taken showed that 68.0% of the females and only 28.0% of the males exhib- ited nipping behavior at least once. Where two-hour continuous records were available, 100% of the females, and 57.1% of the males engaged in some nipping activity. This sug- gests that practically all of the females and at least half of the males do some nipping before spawning. Nest-building Acts. On the several days before spawning, nest-building by either the male or female was not observed (Text-fig. 6), but by the third hour before oviposition, female nest-building activity had reached a rather high frequency. Since the presence of a nest and the occurrence of nest-building behavior was one of the more important criteria used to determine the imminence of spawning, and hence to ascertain the ap- propriateness of starting the observation, these data are likely to be somewhat biased in favor of early nest-builders. Actually at three hours before spawning, the average nest-building activity of the female may not be as high as that indicated by the data. The drop indicated in the seventh pre- spawning interval appears not to be signifi- cant if the nest-building values of the fifth and seventh intervals are compared (P = .13). On the other hand, there is a noticeable downward slope between the fifth and first pre-spawning intervals, and when these two intervals are compared, the difference was found to be highly significant (P<.01). It is clear that female nest-building behavior drops off as the time for the laying of the eggs approaches, and it is gradually super- seded first by nest-cleaning behavior (which is clearly distinguishable from nest-build- ing), and secondly by nest-passing activity, which, as we shall see in the next section, is increasing as the nest-building frequency is declining. Following the oviposition episode, nest- building activity dropped to a very low fig- 1949] Aronson: Reproductive Behavior in Tilapia macrocephala 143 NUMBER OF SPAWNINGS OBSERVED (TOTAL FREQUENCY) (ONE 15 MIN. INTERVAL AFTER SPAWNING (ONE 15 MIN. INTERVAL ON EACH OF ABOVE DAYS) ON EACH OF ABOVE DAYS) Text-fig. 6. Fluctuation in male and female nest-building behavior before and after spawning. ure. On the 5th or 6th day nest-building by females was not observed, and on the 15th or 16th day only one nest-building act was observed during the 15-minute observation interval by one female out of thirty-six. The data for the male was characterized by high frequencies of zero values and low frequencies of high values which neverthe- less did not fit Poisson curves even after the aforementioned transformation. In contrast to the extensive nest-building activity of the female, that of the male was quite limited. Similar to the female, there is a downward slope in activity between the fourth and first pre-spawning intervals. However, the de- cline is not very pronounced and its statisti- cal validity could not be readilv ascertained. After the spawning the males no longer engaged in nest-building except for a single male which on the 15th post-spawning day nest-built four times during the observation interval. This male was paired with the one female, which was also observed to build a nest during the 15-day post-spawning in- terval. Two nests were present in the tank and it is apparent that this pair was ap- proaching another spawning cycle. Turning again to the 25 soawnings with continuous records for the first pre-spawning hour, it was found that 100% of the females and 72% of the malps engaged in nest-build- ing at least once. Of the seven pairs w’th continuous two-hour nre-spawning records, 71.4% of the males did some nest-building. It is probable that only a small percentage of males do not engage in any nest-building prior to the spawning. Fifteen nests built by ten pairs were measured shortly after the spawnings. In each case the fish were first carefully re- moved without damaging the nests. Since in many cases the nests were oval, two diam- eters were taken, namely the short diameter, and at right angles to this the long diameter. The points used in these measurements are indicated in Text-fig. 7. The average short diameter was 11.8 cm., the average long diameter 13.2 cm., and the average depth 2.6 cm. The female fish (which as shown above are primarily responsible for the con- struction of the nests) varied from 10.7 gr. to 19.2 gr. with an average of 15.1 gr. The males were slightly heavier, weighing on the average 18.0 gr. There was no indication from these limited data of a correlation be- tween size of fish and size of nest. Passing-nest. The earliest nest-passing by the fema’e was recorded for the eleventh pre-spawning interval, two and one-half hours before the egg laying (Text-fig. 8). Following the previously mentioned trans- formation, the data for this interval fit a Poisson distribution. The same is true for the records of the 6th, 8th. 9th and 10th intervals. The data for the 2nd to 5th and the 7th intervals consisted of a series of zero or verv low frQouencies and a smaller group of relatively high values, vaguely suggesting 144 Zoologica: New York Zoological Society f34: 16 Text-fig. 7. Diagrammatic section through typical Tilapia nest showing points used for nest measurements. bimodal curves. The data for nest-passing for the first interval were normally distrib- uted. These data indicate a gradually rising frequency of nest-passing as the spawning approached, with a sudden spurt of activity during the second and first intervals. After the egg laying, nest-passing activity of the female dropped to almost zero and none was recorded on the 5th or 6th and 15th or 16th days. The nest-passing data of the male were normally distributed for the first interval. The records for the remaining pre-spawning intervals were highly skewed to the right with highest frequencies zero, which, how- ever, did not fit Poisson series. While no nest-passing was recorded for the males on the several days prior to the spawning, a substantial amount of nest-passing was ob- served by three hours before the egg laying. This level of activity remained fairly con- stant until the second interval when it started to rise precipitously. However, during the first pre-spawning interval, the nest-passing activity of the female surpassed that of the male for the first time (P<.01). After the spawning, the frequency dropped to almost zero and nest-passing was not observed on the two post-spawning observation days. Observations indicated that male and fe- male nest-passing were not entirely inde- pendent of each other, and calculation of the (ONE 15 MIN. INTERVAL AFTER SPAWNING (ONE 15 MIN. INTERVAL ON EACH OF ABOVE DAYS) ON EACH OF ABOVE DAYS) Text-fig. 8. Fluctuation in male and female nest-passing behavior before and after spawning. 1949] Aronson: Reproductive Behavior in Tilapia macrocephala 145 NUMBER OF SPAWNINGS OBSERVED (TOTAL FREQUENCY) 15 IS 14 15 13 16 18 21 31 39 52 66 60 51 ON EACH OF ABOVE DAYS) Text-fig. 9. Fluctuation in male and female spawning-quiver behavior before and after spawning. coefficient of correlation for the first pre- spawning interval yielded an r of + .63. This was transformed to Z = + .74 which is a highly significant correlation (P<.01). The nest-passing data for the remaining pre- spawning intervals appear to be comparably correlated, but the data do not readily lend themselves to this type of statistical treat- ment. Prior to the spawning, all of the males and females exhibited some nest-passing activity. Spawning-quivers. This behavior was not observed during the observation periods on the several days before the spawning (Text- fig. 9). Female spawning-quivers were first seen during the 8th pre-spawning interval, 1 % to 2 hours before the egg laying. Their frequency gradually increased and reached a peak during the first pre-spawning inter- val. There was a marked drop to almost zero after the egg laying, and on the 5th or 6th days and 15th or 16th days none were seen. A few male spawning-quivers were in evidence during the 12th pre-spawning in- terval, and a low level of this behavior was maintained until the second interval, 20 to 15 minutes before the egg laying, when there occurred an abrupt rise in frequency which terminated during the fii'st interval. During the first post-spawning interval, a very few spawning-quivers were recorded, and none were seen thereafter. While a few of the males exhibited spawning-quivers long be- fore the females, the peak of spawning- quiver activity of the females during the first pre-spawning interval was considerably higher than that of the males. However, the data did not permit further statistical anal- ysis. Selecting the 25 spawnings for which con- tinuous records for the first hour before oviposition were available, it was observed that 96.0% of the females and 72.0% of the males showed at least one spawning-quiver during this hour. Similarly, in the seven spawnings for which two-hour continuous records were available, 100% of the females and 71.4% of the males were recorded as performing at least one spawning-quiver during these two hours. It thus appears that while all the females showed this behavior, in about a fourth of the males spawning- quivers could not readily be distinguished from nest-passing behavior. Since all of the females exhibited spawning-quivers and be- cause of the sharp peak in the frequency of occurrence of this activity just before the spawning, this behavior can also be employed as an indicator of the approaching oviposi- tion. Oviposition and Fertilization. The be- havioral patterns considered thus far were recorded in terms of the number of times that the acts occurred during a short interval of time (i.e., 15 minutes), and the relative infrequency of some of this behavior ac- counts in part for the marked skewness of the distribution curves. On the other hand, the oviposition data which follow, and the data concerning the reactions of the male and female to the eggs and young, are based upon the total frequency of the behavior dur- ing a given spawning, and as might be an- ticipated, these data approximated more closely binomial distributions which could be treated as normal curves. A nest-passing act by the female during which eggs were oviposited was counted as a single oviposition movement. A nest-pass- ing act by the male when eggs were present in the nest was recorded as a fertilization movement. Actual contact with the eggs was not considered essential as a criterion for a fertilization movement, although in most 146 Zoologica: New York Zoological Society [34: 16 instances the male rubbed his genital tube over some of the freshly laid eggs. In 76 observed spawnings, the mean num- ber of oviposition movements by the female was 3.41 ± .13 with a standard deviation of 1.1 movements. The mean number of male fertilization movements was 3.46 ± .15 with a standard deviation of 1.3. The difference between the means is .05 ± .2 which indicates clearly that the number of oviposition move- ments of the female does not differ signifi- cantly from the number of fertilization movements of the male. Finally, there is a significant positive correlation (r = + .48, z = + .52, P calculated from <.01) be- o z tween these two activities, indicating that the number of times the males fertilize the eggs is partly related to, and probably de- pendent upon the number of oviposition acts of the female. Parental Behavior. In a total of 76 ob- served spawnings, the male alone picked up the eggs in 62 cases (81.8%), the female picked up the eggs in 6 cases (7.9%) while both male and female participated in this activity in 8 cases (10.5%). The time after the beginning of oviposi- tion for the male to start picking up eggs varied from 20" to 2'10" with a mean of 1'3" ± 3" and a standard deviation of 23". For the female this interval varied from 3'18" to 11T4" with a mean of 7'59" ± 1'22" and a standard deviation of 3'17". The difference between the means of these two distribu- tions is obviously significant, and from these data we may conclude that the male starts to pick up the eggs as soon as the ovinosi- tion has terminated, while the female allows several minutes to elapse before she will collect any of the eggs still available. Here then is an apportioning mechanism which results in the observed fact that the male usually incubates the eggs, and the female does so on infrequent occasions. Eggs remain in the nest available to the female under two circumstances. First, if the male’s mouth is of insufficient size to con- tain all of the eggs, a few may be left over in the nest. This was the situation in case 1 (Table I) where a small male was paired with a lar^e female. It was quite clear to the observer that in this instance not all of the eggs could fit into the male’s mouth. Sec- ondly, eggs would be available to the female when the male behaved atypically and did not touch the eggs. In three of these cases males had released broods seven to twelve days previous to the spawnings, and this may be a contributing factor causing the lack of response of the males to the eggs.* 1 2 3 * * * * * * * * * * 14 In most instances where the eggs remained in the nest for any length of time, the females would chase, nip and court the males. In a few cases, the latter retaliated and violent 14 On the other hand, recent observations by Aronson and Holz-Tncker (unpublished) reveal that males in the process of incubatine young may on occasion fertilize and pick up a new batch of eggs. TABLE I. Time from the Beginning of Oviposition for Eggs to Be Picked Up. Eight Cases Where Both Male and Female Engaged in This Activity. Male Case No. Start Finish Female Start Finish 1 25"— 50' 2 ll'OO"— 11'30' 3 3'00" — 4'00' 4 4'05" — 4'30' 5 2'05" — 5'00' 6 4'10" — 5'10' 7 6'24" — 10'30' 8 3'50" — 4'00' 7'15" — 6'50" — lO'lO" 2'30" — 4'00" 4'15" — 4 '30" 4'30" — 5'00" 4'1 0" — 4'55" 5'50" — 6'36" 1'25" — 3'30" fighting ensued; as a result the nests were destroyed and the eggs scattered. In cases 3, 6, and 7 (Table I), as soon as the female began to pick up the eggs, the males fol- lowed suit and both gathered up the eggs simultaneously. The typical pattern when eggs are left in the nest may be summarized as follows: (1) Immediately after the eggs are ovi- posited and inseminated, there is often a period of extreme quiescence lasting a minute or two. (2) This is followed by a period in which the female appears to be inhibited from approaching or touching the eggs, but at the same time she seems to be excited by the eggs, resulting in active nipping, chasing and courting of the male who sometimes responds similarly. (3) After several minutes the inhibitory action of the eggs begins to diminish; the female now approaches the nest, pokes around the eggs, and eventually picks them up. It was at this time that several of the recalcitrant males listed in Table I also approached the nest and in a few cases started to pick up eggs ahead of the female. It is suggested that in cases 2 to 8 (Table I), chasing, nipping and courting by the female, and also her poking around the nest, sufficiently stimulated the male to pick up the eggs, thereby completing the pat- tern. Once started, the length of time it took for males to gather up the spawn varied from 2" to 1'45" with a mean of 13" ± 2" and a standard deviation of 16". The high varia- bility noted here is a result of two excep- tional cases, one where the male took 1'7" and in the other 1'45". In the remaining 60 spawnings, the time was less than 46". On the other hand, six females took from 35" to 3’ with a mean of 2'6" ± 22" and a stand- ard deviation of 51.7". Thus we see that not only does the female wait longer before starting to pick up the eggs, but once started she performs this task at a significantly 1949] Aronson: Reproductive Behavior in Tilapia macrocephala 147 slower rate. In most cases, the males gath- ered up the eggs rapidly and then kept pok- ing around the nest for some time. Thus any scattered eggs were quickly recovered. Some of the females, on the other hand, would pick up part of the eggs, swim away from the nest, return and pick up more eggs, swim away again, and so forth. The egg-gathering records for the female were necessarily limited by the behavior of the males as noted above. It was therefore considered appropriate to use for compar- ison data from other experiments. Aronson and Holz-Tucker (unpublished data) ob- served the spawning of an isolated female that could see another female in an adja- cent tank. The ovipositing female took 24' to start gathering up the eggs and the process itself took 1'25" to complete. Similarly, we observed the spawning of a completely iso- lated female. This female did not start to pick up the eggs for 13'5". She took 1'15" to gather up most of the spawn, but left six eggs which she did not pick up for another eight minutes. A large number of normal females were paired with males suffering various types of brain lesions (Aronson, in manuscript). In 27 spawnings, these females took on the average 12'2" to start picking up eggs and an average of 1'15" to complete the job. Hence these data support our orig- inal conclusions. However, it is likely that in our first observations, the mean time for the six females to start picking up the eggs is somewhat low, while the time it took to complete the pi’ocess may be a little too high. It is of interest to note that in a few spawn- ings the females seemed unable to carry all of the eggs that they themselves had laid. Both the male and female are capable of successfully incubating the eggs. The per- centage of spawnings in which young were recovered at the termination of the incuba- tory period is shown in Table II. Where the spawnings were not witnessed, the slightly higher score made by the males may be ac- counted for by a possible failure to record a few cases where the spawn was swallowed immediately after the oviposition, and before it was observed. The data for the third set of observations are taken from a second ex- periment, (Aronson, 1945) . These spawnings were also not witnessed. In this experiment, aquarium conditions were considerably im- proved by the use of aquarium filters, thus avoiding any changes of water. The young were forcibly ejected from the parental mouth on or about the tenth day after spawn- ing and were counted immediately, thus largely eliminating the possibility of losses through cannibalism. Even with these improved techniques, only 40% of the males released viable fry. Two factors account for this low yield of young by the males. First is the failure of the eggs to be properly fertilized, or death of the em- bryos, with subsequent disintegration of the eggs. A second factor is swallowing the spawn. The relative importance of these two factors will now be considered. If freshly laid unfertilized eggs are placed in a jar of Tiiapia-conditioned water which is kept at approximately 26° C., very few of the eggs will show any gross signs of degen- eration before 24 hours. Starting with the second day, however, some of the eggs will have decomposed, and in all cases few if any intact eggs remain after the tenth day. As to the variation in the length of the incuba- tion period, it will be seen in columns 6 and 7 of Table III that females may carry unfer- tilized eggs for as long as ten days, after which time it may be assumed that all have decomposed. Note particularly that in almost 80% 15 of these cases, the dead eggs were retained longer than one day, and it is highly probable that in many of these cases the eggs were carried until they were quite de- generate. It was not unusual to examine the contents of a male’s or female’s mouth and is Since the presence of incubating eggs was checked only twice daily, spawn swallowed shortly after oviposition mi-ht have been overlooked. Hence this figure may be a little too high. TABLE II. Per Cent, of Spawnings in which Young Were Recovered. No. of spawn- ings in which males incubated eggs. No. of males releasing young. % of males releasing young. No. of spawn- ings in which female incu- bated eggs. No. of females releasing young. % of females releasing young. Spawnings witnessed. 68 22 32.4 14 3 21.4 Spawnings not witnessed. 86 33 38.8 2 0 0.0 Spawnings not witnessed, 2nd experi- ment. 70 28 40.0 4 4 100.0 Variation in Length of Incubatory Period. 148 Zoologica: New York Zoological Society [34: 16 >5 -O T3 > rO a £ w • >& r> N I ^ ; a? ! c$ a> "1 £w aJ £ . w ■H> 0) § * ° £ S-( ^ Ph CO CO CO* CO co co U-i w o £ • «J II Ah o O 3 S5 £ (MNNWH^NH 5C 05 W N W C ^ ci 03 t— t rH rH 03 O 03 rH W M Tf CO H IB H O >- I *■; N © i-H ioeide5®oddci M O* . cS I S 4) _a o3 X3 0 e « o "^3 « 2 Si S ®«w Ah O H N C rH N (N CO 05 N N N 03 O © Tf 03 O O 03 o a £ £ sh c o § JCi’-§ h~> c3 bCJS a s a> u -3 C !-HiH«OC-COOrH(Mr-iOO(MtH s< s- s< -3 -e JC H 00 ^