ZOOLOGICA SCIENTIFIC CONTRIBUTIONS OF THE NEW YORK ZOOLOGICAL SOCIETY VOLUME 54 • ISSUE 2 • SUMMER, 1969 PUBLISHED BY THE SOCIETY The ZOOLOGICAL PARK, New York Contents PAGE 4. Some Mexican and Central American Land Snails of the Family Cyclo- phoridae. By Fred G. Thompson. Plates I- VII; Text-figures 1-14 35 5. The Underwater Song of Erignathus (Bearded Seal). By Carleton Ray, William A. Watkins, and John J. Burns. Plates I-III; Text-figure 1; Phonograph Disk 79 Manuscripts must conform with Style Manual for Biological Journals (American Institute of Biological Sciences). All material must be typewritten, double-spaced. Erasable bond paper or mimeograph bond paper should not be used. Please submit an original and one copy of the manuscript. Zoologica is published quarterly by the New York Zoological Society at the New York Zoological Park. Bronx Park. Bronx. N. Y. 10460. and manuscripts, subscriptions, orders for back issues and changes of address should be sent to that address. Subscription rates: $6.00 per year; single numbers. $1.50. unless otherwise stated in the Society's catalog of publications. Second-class postage paid at Bronx. N. Y. Published November 17, 1969 © 1969 New York Zoological Society. All rights reserved. 4 Some Mexican and Central American Land Snails of the Family Cyclophoridae Fred G. Thompson1 (Plates I-VII; Text-figures 1-14) The systematics of some Middle American cyclophorid snails is revised based on studies of the soft anatomy. Two subfamilies are recognized ( Aperostominae and Neocyclotinae) as occurring in Mexico and Central America, not including the Diplommatinid genus Adelopoma. The subfamilies are separated by characteristics of the male and female reproductive systems. Genera placed in the Aperostominae include Aperostoma, Megalo- mastoma , Farcimen, and Tomocyclus. Genera placed in the Neocyclotinae include Dicrista new genus (type species— D. liobasis n. sp. ), Xenocyclus new genus (type species— A" patulus n. sp.), Amphicyclotus, and Neocyclotus. The following new species and subspecies are also described: Dicrista flavescens n. sp., D. indentata n. sp., D. rugosa n. sp., Amphi- cyclotus texturatus spiralis n. sp., A. paulsonoriun n. sp., Neocyclotus simplicostus n. sp., N. capscelius n. sp., N. ( Incidostoma ) impressus n. sp. The following nomenclatorial changes are made: Mexcyclotus petersi petersi Solem change to Dicrista petersi (Solem); M. p. damianensis Solem changed to D. damianensis (Solem); Cyclotus cooperi Tryon changed to D. cooperi (Tryon). Cyclostoma lutescens Pfeiffer is considered a nomina dubia. The generic name Mexcyclotus based on lutescens as genotype become unavailable for use among Mexican taxa. THE purpose of this study is to present new information bearing on the systematics and distributions of some Mexican and Central American cyclophorid snails based on material collected during recent years. The ma- terial includes numerous specimens preserved for anatomical investigations, and alters previ- ous systematic concepts. The study is not com- prehensive, and includes only critical synony- mies for particular species and genera. For a complete review of the taxa within this region the reader is referred to papers by Bartsch and Morrison (1942), Morrison (1955), and Solem (1956). Introduction The Cyclophoridae are typical of many groups of neotropical land snails. Although many spe- cies are known, more await to be discovered. Most of those that are known are represented by only a few specimens each. Distribution data are sparse or absent for most species. Large 1 Florida State Museum, University of Florida, Gainesville, Florida 32601. geographic areas from which material may be expected remain unexplored for mollusks. Very little is known about individual or geographic variation, and much less is known about the anatomies of the various species. The classification of the cyclophorid mollusks is as confused as that of any other comparable division of land mollusks. There is no general agreement whether the cyclophorids comprise a single very large family, or whether several distinct families are involved. Recent authors tend to recognize the group as a superfamily with several families, but no one has demon- stated the presence of characters that consist- ently separate and characterize the families. All of the genera discussed in this paper are placed in a single family and assigned to two sub- families. This does not imply that I favor the recognition of a single large family, the Cyclo- phoridae, but only that I am hesitant to recog- nize additional families until more is learned about the soft anatomy of the various divisions. Families or subfamilies in this group currently are based, in part, on the geographic distribu- tions of these categories, and this leaves much 35 36 Zoological New York Zoological Society [54: 2 to be desired in explaining their origins, evolu- tion, and relationships. Early systems of classification divided the helicoid cyclophorids into two subfamilies. Those with a calcified operculum were placed in the Cyclotinae, and those with a chitinous operculum were placed in the Cyclophorinae. The cylindroid genera were placed in several different subfamilies on even lesser basis. This system was modified by Kobelt (1902) and Thiele (1929) who used geographic distribu- tion as an important factor. All of the neo- tropical genera were placed in a single subfamily with the exception of Adelopoma , which belongs in the subfamily Diplommatininae. Bartsch (1942) recognized three subfamilies endemic to the American tropics, but made no attempt to differentiate them from Old World groups. These subfamilies were based exclusively on shell and opercular characters. The helicoid genera were placed in the Aperostominae with a calcified operculum, and the Amphicyclotinae with a chitinous operculum. The cylindroid genera were placed in the Megalomastominae. Tielecke (1940) divided the cyclophorid mollusks into five families, based primarily on the anatomy of the reproductive systems. His contribution is outstanding in that it was the first and only comprehensive attempt to study the soft anatomy of the cyclophorid mollusks. The neotropical genera were placed in a single family, the Poteriidae, characterized by possess- ing a common duct for the copulatory bursa, oviduct, and the seminal receptacle, and by hav- ing the verge in males located on the center of the nape or behind the right tentacle. Recently Morrison (1955) studied the ex- ternal male reproductive structures of the neo- tropical members and placed them in two fami- lies, the Amphicyclotidae and the Neocyclo- tidae. The Amphicyclotidae was characterized by having a seminal duct enclosed within the verge. The Neocyclotidae was characterized by having an open seminal groove extending to the tip of the verge, and included two subfamilies, the helicoid Neocyclotinae (= Aperostominae and Poteriidae of other authors) and the cylin- droid Neopupinae (= Megalomastominae) . He related the Neocyclotidae to the Cyclophoridae but derived the Amphicyclotidae from the marine family Lacunidae. Solem (1957) dem- onstrated that Amphicyclotus possesses a verge that places it in the Neocyclotidae, and indicated the nomenclatural changes that were necessary if the West Indian “amphicyclotids” were rec- ognized as a separate subfamily. Thompson (1967) showed that this West Indian group is related to the neocyclotids by the structure of the female reproductive system, but differed conspicuously in the tubular structure of the verge, and recognized the group as the sub- family Crocidopominae. Recent field work in Mexico and Central America has produced much new and interest- ing material of the Cyclophoridae. This material represents several new taxa and adds distribu- tional data to some species that previously were known from indefinite localities. Even more important, most of the species collected are rep- resented by soft parts, which provide new evi- dences on the phylogeny of the group in Middle America. New descriptions of four poorly known species are also included. Acknowledgment Many people have assisted me in this study, to all of whom I am grateful. For assistance in field work in Mexico and Central America I wish to thank Dennis R. Paulson and his wife Mary Lynn, Washington State University; F. Wayne King, New York Zoological Society; Roy McDiarmid, Norman Scott, and Jay Savage, University of Southern California; Andrew Star- red, Los Angeles County Museum of Natural History; Colin Little, University of Bristol; S. David Webb, Florida State Museum, University of Florida (UF). Albert Schwartz and Richard Thomas, Miami, Florida, have provided me with much valuable West Indian material. I wish to thank the following people for the loan of speci- mens in their charges. Tucker Abbott, Academy of Natural Sciences, Philadelphia (ANSP); Henry van der Schalie, Museum of Zoology, University of Michigan (UMMZ); Harold A. Rehder, United States National Museum (USNM). Photographs of shells used in this paper were taken by Ernest M. Collins, Jr., staff photographer of the Florida Division of Plant Industries. Field work in Mexico and Central America was supported by the National Institutes of Health Research Grant GM 12300. Family Cyclophoridae The material that I have studied alters pre- vious classifications of the neotropical cyclo- phorids. All apparently lack a copulatory bursa such as occurs in Cyclophorus (Tielecke, 1940: 321-327). I found no structure that appears homologous to that organ of the Old World cyclophorids. The absence of this organ indi- cates a degree of unity among the American genera that may justify recognition of them as a single family, the Neocyclotidae. However, the anatomical diversity of these genera oversha- dows the common relationships suggested by the absence of the bursa. The Megalomastominae differs from all other cyclophorid groups by 1969] Thompson: Some Mexican and Central American Land Snails of the Family Cyclophoridae 37 having a long vaginal slit extending nearly the length of the uterus. The subfamily is also pe- culiar among New World groups in having a multipapillaform seminal receptacle and a verge located on the side of the head behind the right tentacle. These last two characters suggest a relationship with the Asiatic Pupininae (-idae). Characters of the shell also suggest this rela- tionship, but the Asiatic pupinids differ by hav- ing a well developed copulatory bursa. Until more is learned about the anatomy or the vari- ous Asiatic pupinid genera, the relationship be- tween the Pupininae and the Megalomastominae will remain in question. The Neocyclotinae and the Crocidopominae are more alike to each other than to any other comparable group of cyclophorids by having a verge that is located on the center of the nape. The Neocyclotinae has a stout verge that bears an open seminal groove extending from the end of the prostate to the tip of the penis. The Crocidopominae has a long slender verge that possesses a tubular seminal duct. Apparently the Crocidopominae evolved from the Neocy- clotinae. The emphasis placed on the distinction between these two subfamilies may be ques- tionable, but their relationships and their differ- ences are clear in contrast to most other cyclo- phorid subfamilies. Subfamily Megalomastominae Kobelt and Mollendorff The external male reproductive structures of the following species have been described: Aperostoma m. mexicana (Menke) (Tielecke, 1940: 339) Aperostoma l walker i Baker (Morrison, 1955: 152) Tomocyclus simulacrum (Morelet) (Bartsch and Morrison, 1942: 142) Farcimen vindlense scopulorum T. and B. (Torre and Bartsch, 1942: 34) Farcimen superbum itinerium T. and B. (Morrison, 1955: 152) Farcimen ( Neopupina ) croteum (Gmelin) (Bartsch, 1942: 44) Megalomastoma pepiti Bartsch (Bartsch, 1942: 48) In addition I have examined both sexes of the following species: Aperostoma mexicana salleana (Martens) Aperostoma palmeri (Bartsch and Morrison) Farcimen ( Neopupina ) croteum (Gmelin) This subfamily is characterized by having the penis located on the side of the head behind the right tentacle. The penis is broad and flattened basally, and becomes cylindrical and attenuate distally. An open seminal groove extends from the end of the prostate and along the side of the nape to the penis where it passes along the outer and lower margin to the tip. The prostate has a corresponding slit along its columellar margin. The slit opens into the prostatic lumen and ex- tends from the seminal duct to the anterior end of the prostate where it continues with the seminal groove on the nape. The female has an open vaginal slit that ex- tends almost the length of the uterus and is con- tinuous with the uterine lumen. The uterus has a single continuous lumen and is nuiltilobate along its outer and distal margins. The copulatory bursa is rudimentary or absent. The seminal re- ceptacle consists of a series of digitiform or grape-like glandular lobes that discharge into a common chamber. The albumen gland is formed by a simple loop in the oviduct and is attached at its base by connective tissue to the side of the bursa. The albumen gland enters into the cham- ber of the seminal receptacle, which in turn opens into the distal end of the vaginal slit. Both the albumen gland and the seminal receptacle are imbedded in muscle and connective tissue and appressed to the side of the uterus. In addition the subfamily has an unusually enlarged hypobranchial gland, that occupies a major portion of the surface of the pseudolung (Text-fig. 1 ) . The megalomastomids have in common a multispiral, cornified operculum in which the outer edges of the whorls overlap the succeed- ing whorls and project obliquelly outward as thin chitinous lamella. The innermost whorls may be thickened but do not have calcareous deposits. Although these opercular characteris- tics do not distinguish this group, they show a degree of uniformity among the member genera. This subfamily consists of a homogeneous group of genera that is confined to eastern Mexico, Guatemala, and the Antilles. Aside from the helicoid genus Aperostoma all of the species are cylindroid in shape. This divergence in shape is more apparent than significant, for other features of the shell, such as the siphonal notch and the reflected peristome, indicate a close affinity between Aperostoma and Tomo- cyclus, which have allopatric but contigous ranges in eastern Mexico and Guatemala. Tomocyclus This genus has recently been reviewed (Thomson, 1963). No additional material has been examined. Aperostoma mexicanum salleanum (Martens) Cyclophorus salleanum von Martens, 1865; Malak. Blatt., 12: 15 1 . — Strebel, 1873; Beit. Mex. L. — Susswasser Couchy., I: 9; pi. 1, 38 Zoological New York Zoological Society [54: 2 fig. 2; pi. 1A, figs. 2-2a. — Martens, 1890; Biol. Cent. Amer. : 7-8. (Type locality: Cordoba, Veracruz). Cyrtotoma mexicanum salleanum (Martens), H. B. Baker, 1922; Occ. Pap. Mus. Zool. Univ. Mich., ( 106) :42-43; pi. 16, figs. 8-12. Aperostoma mexicanum salleanum (Martens), H. B. Baker, 1928; Occ. Pap. Mus. Zool. Univ. Mich., (193) :51. Cyrtotoma salleanum (Martens), Bartsch and Morrison, 1942; Bull. U.S. Nat. Mus., 181; 170-171; pi. 22, figs. 19-21. Aperostoma m. mexicanum (Menke), Solem, 1956; Proc. Acad. Nat. Sci. Phila., 108:48-50 {in part ) . Veracruz: 2.9 mi. E. Cordoba, 2500' (UF 20211. 2); 1.9 mi. S.W., 0.8 mi. N. Fortin, 2900' (UF 20212 . 13); 4.3 mi. E. Cordoba, 2300' (UF. 19). Anatomical material examined from the last two localities. Male. The penis is located on the side of head behind the right tentacle. The base of the penis is bulbous; the distal end is about twice the length of the base and is slender and flagellate. An open seminal groove extends from the pros- tate and along the side of the nape to the tip of the penis. Female. (Text-fig. 2, C) The single female examined illustrates the general features of the reproductive system, although the uterus does not appear to be completely mature. The vaginal slit extends the length of the uterus and is con- tinuous with the uterine chamber. The uterine chamber is uninterrupted through its length. The outer margin of the uterus is convoluted into Text-fig. 1. Aperostoma palmeri (Bartsch and Morrison). Dorsal and lateral view of the pseudolung and associated organs showing the relatively large hypobranchial gland. Scale equals 5 mm. 1969] Thompson: Some Mexican and Central American Land Snails of the Family Cyclophoridae 39 many small lobes that decrease in size distally. The distal end of the uterus is very thin and spatulate, with weakly indicated lobation. The seminal receptacle consists of about four digiti- form glandular lobes that discharge into the oviduct. The albumen gland consists of a simple convoluted loop in the oviduct. The loop is at- tached to the base of the seminal receptacle by fine fibers of connective tissue. Aperostoma palmeri (Bartsch and Morrison) Cyrtotoma palmeri Bartsch and Morrison, 1942, Bull. U.S. Nat. Mus., 181:172-173; pi. 22, figs. 1-3. (Type locality: Gomez Farias, Tamaulipas). Aperostoma mexicanum palmeri (Bartsch and Morrison), Solem, 1956; Proc. Acad. Nat. Sci. Phila., 108:50-51; pi. 5, figs. 1-4, pi. 6, figs. 1-5. San Luis Potosi: 11.4 mi. E. Xilitla, 1100' (UF 20209. 31), (UF 20208. 11): 7.5 mi. E. Xilitla (UF 20210 .1). One female and one male examined from first locality. The material before me does not indicate intergradation with A. mexicana as Solem (1956: 51 ) reported for material from the same region. A. palmeri is readily distinguished from A. mexicana by its deep parietal notch that is open to the dorsal surface of the whorl as a siphonal notch, and by its thickened, weakly flaring peristome. The parietal notch is partially obscured by a wing-like extension of the upper lip which leaves a narrow slit connecting the notch with the margin. A. mexicana has a more broadly flaring peristome, and the parietal notch consists of a simple U-shaped indentation that lies along the parietal wall of the aperture, but does not open through the dorsal wall of the whorl as in a siphonal notch. The nature Of the anatomical material is not satisfactory for detailed studies, but does show the gross morphological features of the repro- ductive systems. Male. Reproductive system typical for genus. The penis is located on the side of the head be- hind the right tentacle. The seminal groove ex- tends from the prostate and along the side of the nape to the tip of the penis. The groove is uninterrupted along its course. The prostate lies along the right margin of the pseudolung, and appears to have an open seminal channel that is continuous with the prostatic chamber and ex- tends from the spermatoduct to the anterior end of the prostate. The channel lies along the right margin of the pseudolung, and terminates over the seminal groove on the nape. Female. The reproductive system is similar to that described for A. mexicana salleana, ex- cept that the uterus is more developed, and has the same general appearance as does Farcimen croteum. Farcimen (Neopupima) croteum (Gmelin) Puerto Rico: 2.5 mi. S.W. Yabucoa, 800' alt. (3 females and 1 male); 1 km. S. Pueblito de Ponce (4 females, 1 male). Male. The verge is located on the right side of the head behind the tentacle. It has an open seminal groove that is continuous from the prostate. Female. The uterus (Text-fig. 2, A-B) is large, robust, and glandular with numerous large folds along its intestinal margin. The distal end is conspicuously flattened and palmate. The vagina consists of a continuous slit that extends the length of the uterus, and has a very thin membranous fold that overlaps the outer wall of the uterus. The vaginal slit is continuous with the uterine cavity, which continues distally into the seminal receptacle. The receptacle consists of a large mass of closely attached glandular lobes that discharge individually or in small groups into a chamber. The lumen of the re- ceptacle is lined with numerous small folds that ramify into the individual glandular lobes. The albumen gland consists of a long sigmoid loop in the oviduct that is closely attached by con- nective tissue to the side of the seminal recep- tacle in such a way that it is not distinguishable from the receptacle lobes upon superficial exami- nation. The ovary consists of five glands. The first gland is a small, single digitiform structure. Subsequent glands are multilobate. The mantle is similar to that of Aperostoma in that the hypobranchial gland is thick and padlike, but is confined to the left side of the pseudolung, is smaller, and is kidney-shaped. Subfamily Neocyclotinae Kobelt and Mollendorff, 1898 This subfamily includes the neotropical gen- era that Morrison (1955) placed here, plus the genera that he included in the Amphicyclotidae, exclusive of those that are now placed in the Crocidopominae (Thompson, 1967: 15). The subfamily also includes the seven Pacific Island genera discussed by Clench (1949: 4-13) and Solem (1959: 180-184). The subfamily Neocyclotinae is characterized by possessing a verge that is located on the center of the nape. An open seminal groove extends from the end of the prostate, across the nape, and to the tip of the verge. In some species of one genus ( Neocyclotus ) the groove may be secondarily coalesced to form a sperm duct that is connected to the surface of the verge by a keratinized raphe. This condition is different 40 Zoologica: New York Zoological Society [54: 2 uterus seminal rec. oviduct albumen gland vaginal slit Text-fig. 2. Female reproductive systems of two species of Aperostominae. A. Farcimen croteum (Gmelin); 2.5 mi. s.w. Yabucoa, Puerto Rico. B. Farcimen croteum (Gmelin); enlarged view of seminal receptacle with the receptacle chamber partially opened. The albumen gland has been partially freed from the side of the receptacle. C. Aperostoma mexicana salleana (Martens); 4.3 mi. e. Cordoba, Veracruz. Scale for A equals 5 mm; scales for B and C equal 2 mm. from the sperm duct that occurs in the Croci- dopominae, which has a tubular verge without any connecting raphe between the sperm duct and the outer surface. The female system is dis- tinct in possessing a hollow saculate seminal receptacle, an albumen gland that consists of an enlarged segment in a loop of the oviduct, and in having the uterine lumen divided longitudi- nally into two chambers by an involution. The division is generally incomplete so that a horse- shoe shaped cavity is formed in cross-section, but in one genus ( Neocyclotus ) the involution completely divides the lumen except at the distal end of the uterus. A copulatory bursa is absent or is rudimentary and imbedded in the uterine wall. 1969] Thompson: Some Mexican and Central American Land Snails of the Family Cyclophoridae 41 seminal Text-fig. 3. Diagramatic illustrations of three types of female reproductive systems in the evolution of the Neocyclotinae. A, C, and E are longitudinal sections through the uterus. B, D, and F are cross-sections through the above utera at the points indicated by the arrows, and show the respective degrees of involu- tion of the uterine wall. A. A primitive condition in which the seminal receptacle and the oviduct are separate and discharge on the outer surface of the uterus, as in Dicrista new genus. C. A more advanced stage in which the seminal receptacle enters the oviduct, and the genital duct terminates at the end of the vagina, as in Amphicyclotus. E. A highly modified stage in which the genital duct enters the uterus prior to the vagina. The uterine lumen is completely divided by an invagination so that a single sigmoid passage occurs through the uterus from the genital duct to the vagina. Evolution within the subfamily involved vari- ous elaborations of the terminal structure of the verge and modification of the female reproduc- tive system. Features of the shell and the oper- culum have no relationship to underlying ana- tomical changes. Suprageneric categories based upon conchological characters are without basis, for several prominent external features occur independently throughout the subfamily (color patterns, siphonal notches, calcified opercula, etc.) . In the primitive condition the seminal recep- tacle has a long slender duct that is independent of the oviduct throughout its length and both ducts discharge on the outer surface of the uterus above the vagina (Text-fig. 3, A). In more advanced stages the receptacle duct dis- charges into the oviduct, forming a common genital duct below their union, and the genital duct terminates at the vaginal opening (Text-fig. 3, C) or into the terminal segment of the uterine lumen (Text-fig. 3, E). Among the Middle American genera Dicrista new genus possesses the most primitive condi- tion, (as in Text-fig. 3, A-B) which is only slightly modified in Xenocyclus new genus. Amphicyclotus (as in Text-fig. 3, C-D) is in- termediate in structure between Dicrista and Neocyclotus (as in Text-fig. 3, E-F). Presum- ably other American genera are similar to Neo- cyclotus. Ostodes Gould and Gonatoraphe Mollendorff are the only members among the Pacific island genera that have been investigated sufficiently to indicate the structure of their female reproductive systems (Solem, 1959: ISO- 184). They too are similar to Neocyclotus. Mexcyclotus lutescems (Pfeiffer) Cyclostoma ( Cyclophorus ) lutescens Pfeiffer, 1851; Proc. Zool. Soc. Lond.: 250. Cyclophorus lutescens (Pfeiffer), Pfeiffer, 1852; Monogr. Pneum. Viv., 1:82. Mexcyclotus lutescens (Pfeiffer), Bartsch and Morrison, 1942; Bull. U.S. National Mus., (181): 179 (not 181). Apparently this is a South American species that has been identified with a form occurring in Mexico. A brief review of its history is given. Pfeiffer (1851: 250) described lutescens from Brazil. He subsequently (1865: 69) recorded 42 Zoologica: New York Zoological Society [54: 2 lutescens from Panistlahuaca, Oaxaca, on the basis of material collected by Boucard. Fischer and Crosse (1886: 139) included lutescens in the Mexican fauna on the basis of Pfeiffer’s identification of Boucard’s specimens. Angas (1879: 483) recorded the species from the Nicoya Peninsula of Costa Rica. Martens (1890: 7) hesitantly accepted the Costa Rican record and suggested the identity of Cyclotus cooperi Tryon from Mexico with Cyclostoma lutescens Pfeiffer. Bartsch and Morrison (1942: 181) recognized the two species as distinct, but they included both in the Mexican fauna, and desig- nated a new genus, Mexcyclotus, with lutescens as the type species. Solem (1956: 55) synonym- ized cooperi with lutescens, and discussed its distribution in Mexico and Central America. Apparently the sole basis for including lutes- cens in the Mexican fauna is Pfeiffer’s citation of specimens from Panistlahuaca, Oaxaca. Later authors followed Pfeiffer and credited the identity of the Mexican form to lutescens, but at no time did Pfeiffer state that the original locality of Brazil was in error, though implied by von Martens (1890: 7). The identity of Angas’s Costa Rican record is also doubtful. Von Martens accepted the locality as probable because it was bracketed by records from Brazil and Mexico, but he did not accept some of Angas’s other Costa Rican cyclophorid records (Martens, 1890: 4). An important distinction between lutescens and Mexican forms, or their identity, depends upon the opercular structure of the type of lutescens. All authors on Mexico subsequent to Pfeiffer ( 1865) assumed that Mexican popula- tions called lutescens have an operculum similar to that described for the Brazilian specimens. The types of lutescens currently lack opercula (J. F. Peake, personal communication), but some important points are stated in the original description (Pfeiffer, 1851: 250). A translation is given below. “Shell umbilicate, depressed conical, solid, obliquely threadstriate, silky (shiney), yellow- ish white; spire low and conical, acute; suture deep, simple; whorls 4.5, convex, rapidly in- creasing, last whorl not descending; umbilicus moderate, deep; aperture slightly oblique, broadly ovate; peristome simple, sharp, con- tinuous, narrowly adnate, upper corner weakly angulate. — Operculum membranous, pale horn colored, coarse and spiral [lamella], outer sur- face deeply concave. - Major diameter 20 mm, minor diameter 15.5 mm, height 12 mm, aper- ture height 10 mm.” “Habitat in Brazil.” Several critical points are apparent. The shell is more depressed than any Mexican specimens identified as lutescens by recent authors. The shell is unusually large, has unusually few whorls, and the whorls rapidly increase in size in contrast to Mexican specimens identified as lutescens. The operculum is described as being rough and membranous, not calcified. Similar observations on the operculum were made by von Martens ( 1 890: 2) and Kobelt (1902: 254) who had both examined Pfeiffer’s types. Kobelt merely stated that the operculum was mem- branous and tightly coiled. All authors subsequent to Pfeiffer assumed that Mexican specimens identified as lutescens have a membranous operculum as described by Pfeiffer. Actually they do not. The Mexican forms all have a well developed oblique cal- careous lamella in addition to a chitinous la- mella. The differences in opercular structure between lutescens and cooperi precludes their specific or generic identity. Because of these differences the name Mexcyclotus must be re- stricted to the “Brazilian” species (lutescens) , and the Mexican species (cooperi and others) must be placed in a separate genus described below. Even if we assume that the original locality given for lutescens was in error, but that the name applies to a Mexican species and that Pfeiffer omitted important points in his descrip- tion of the operculum, the name cannot be clearly identified with any Mexican form, for Pfeiffer’s description fails to mention charac- teristics of the embryonic sculpture which would be necessary to restrict the name among known taxa. Dicrista new genus Type Species: Dicrista liobasis new species. A neotropical genus of cyclophorid snails superficially characterized by the combination of characters of its shell and operculum, and more fundamentally by the characters of the male and female reproductive systems. The shell is variable in size and shape. It is usually heli- coid or depressed helicoid; occasionally very depressed. The shell has about five whorls at maturity. The whorls slowly increase in size. The umbilicus is variable in size, but always less than 0.30 times the major diameter. The sculp- ture is simple, consisting of axial threads or riblets. In some instances the axial sculpture may be quite heavy and rugose, but it is always simple, and follows the line of growth, not anas- tomosing or oblique or otherwise modified. The axial sculpture may begin on the second embry- onic whorl in one group of species or it may 1969] Thompson: Some Mexican and Central American Land Snails of the Family Cyclophoridae 43 not appear until after the third whorl in another group. The aperture is generally simple, hut two species have a small siphonal notch in the upper corner. The operculum is highly variable within the genus (PI. I and Text-fig. 4), but always con- sists of a chondroid basal plate with a calcareous and a chitinous lamella on the outer surface. The basal plate may he simple or it may be laminated with calcareous deposits. The lamella may be simple, oblique, and overlapping, or they may be nearly vertical and highly modified for particular species. The operculum is usually concave in cross section and consists of about nine to ten closely coiled whorls. The male possesses a stocky, moderately flat- tened verge that ends in a simple rounded or bluntly pointed tip, and has a small triangular penis on the dorsal surface (Text-fig. 5). The penis lies near the right margin and is reflected posteriorly. A low sigmoid lateral fold lies along the outer margin of the penis. An open seminal groove originates on the side of the nape at the anterior end of the prostate, and continues along the ventral surface of the verge, and back along the inside surface of the lateral fold. The inner wall of the prostate, as in the uterus of the female, is lined with a thick layer of glandular tissue that is involuted mesially to produce a large pendulous glandular fold that divides the prostatic lumen into two cavities that are con- nected mid-ventrally by a narrow gap. The uterus is elongate, cylindrical, and of nearly uniform width throughout its length (Text-fig. 6). The interior of the uterus is lined with a thick glandular layer that is involuted mesially to form a large pendulent fold which divides the uterine lumen into two lateral chambers that are interconnected mid-ventrally by a narrow gap. The vagina consists of a small slit located on the columellar side of the anterior end of the uterus, and is continuous with the uterine lumen. The seminal receptacle is a small saculate pouch appressed against the distal end of the uterus. It has a thick brown glandular wall, and has a simple muscular duct that is split along one side near its base. The albumen gland consists of a large sigmoid loop in the oviduct. The base of the loop is closely bound by connective tissue to the distal sixth of the uterus. Both the oviduct and the receptacle duct are partially imbedded in the uterine wall, and terminate just above the vagina beneath a thick fleshy flap on the columellar side of the uterus. The two ducts discharge into separate narrow grooves in the wall of the uterus. The grooves are parallel to the vaginal slit and are not inter- connected. The genus is known to occur in southwestern Mexico. It questionably has been recorded from Costa Rica. Dicrista contains seven known spe- cies representing two species groups that are distinguished by the sculpture on the early whorls of the spire. One group, containing the type species, has strong axial ribs on the second embryonic and following whorls. The other group does not develop axial sculpture until after the third whorl. The groups parallel each other in modifications of other characters of the shell and operculum. I consider the typical group with sculptured embryonic whorls to be the most specialized, and derived from the more gen- eralized smooth-whorled stock. Species of the latter group have also undergone a high degree of specialization, but along individual lines and not as a group. Dicrista and the following genus form an isolated stock that is not closely related to other genera of the subfamily Neocyclotinae, because of the structure of the reproductive systems. The generic name Dicrista is derived from the Latin and refers to the two opercular lamella, one calcareous and one chitinous. The name is of the feminine gender. Dicrista cooperi (Tryon) Cyclotus cooperi Tryon, 1863; Proc. Acad. Nat. Sci. Phila., 1863: 281; pi. 2, fig. 2. (Type locality: nr. Mazatlan, Sinaloa). Mexcyclotus cooperi (Tryon), Bartsch and Morrison, 1942; Bull. U. S. Nat. Mus., 181: 180; pi. 24, figs. 10-12. Mexcyclotus lutescens (Pfeiffer), Bartsch and Morrison, 1942; Bull. U. S. Nat. Mus., 191: 181; pi. 24, figs. 13-15. Mexcyclotus lutescens (Pfeiffer), Solem, 1956; Proc. Acad. Nat. Sci. Phila., 108: 55-56; pi. 5, figs. 12-13, 16. Diagnosis. A species of Dicrista allied to liobasis, damianensis, and flavescens, and dis- tinguished from others by its rugosely sculp- tured embryonic whorls. It differs from the first three species by its moderately small size, conical shape, closer sculpture on the early whorls, and narrower umbilicus. It approaches damianensis in these characters, and may be subspecifically related. The primary differences between the two forms are in the smaller shell, lack of a sub- sutural chord, and flattened operculum of cooperi. Shell (PI. II, A-C). Small or medium sized. Solid, but not thick. Helicoid; height 0.67-0.95 times major diameter. (The holotype is unusu- ally depressed. All other specimens examined have a height/width ratio greater than 0.70). 44 Zoologica: New York Zoological Society [54: 2 ^ CL O o ^ o -C p . ^ c < e c oo^ 5 £ T3 O £ ^ 3 0< r- w ? l- ^ C3 W o Cu 2 13 E 52 £ * W 0> o c g O 2 ro * <5-2 £ O ?Vi — ■ o S' •G 00 ^4 o E " 4y iu o D C/3 w o U U U fee - o- cO 43 o — — ... C/5 c y 3 « 5 O N «-s g S O O 1) 0) '5 5 q c •pp o ^ Q . « ’ (Jh O D, ^ ^ < o3 S3K o. . M © O 'V THOMPSON PLATE VII SOME MEXICAN AND CENTRAL AMERICAN LAND SNAILS OF THE FAMILY CYCLOPHORIDAE 5 The Underwater Song of Erignathus (Bearded Seal) Carleton Ray,1 William A. Watkins,2 and John J. Burns3 (Plates I-III; Text-figure 1; Phonograph Disk) Vocalization by mature males during breeding season. The call consists of a long oscil- lating frequency-modulated warble that may be more than a minute in duration, followed by a short unmodulated low-frequency moan. It typically starts at about 2000 cps with many frequency variations and ends as low as 200 cps. This call has been identified with the species by our own observations and those of others. Examination of seals heard calling and which were killed revealed them to be males in breeding condition. Thus, this “song” is apparently used solely by mature males in spring courtship season. It is suggested that its purpose is a proclamation of territory or of breeding condition or both. Introduction Erignathus harbatus (Erxleben, 1777), the bearded seal, produces a long combina- tion of complicated underwater sounds during the spring courtship season. This we have termed the “song” because of its complex musi- cal quality and apparent behavioral significance. The characteristic frequency-modulated warble and low moan, though produced entirely under- water, may occasionally be heard in air, radiat- ing through the water surface or ice-cover or through the hull of the boat. These musical underwater sounds have been identified with the bearded seal by the Eskimos who habitually hunt the species as a major source of food and skins (Burns, 1967). The association of singing sounds with the seals is reflected in the names used for these animals by the Alaskan Eskimos; when the seals are sing- ing, the term that is used (“aveloouk” — Upik dialect; “ayuktuk” — Inupik dialect) is trans- 1 Department of Pathobiology, School of Hygiene and Public Health, The Johns Hopkins University, Balti- more, Maryland 21205. 2 Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543 (Contribution No. 2028 from WHOI). s Division of Game, Alaska Department of Fish and Game, P.O. Box 862, Nome, Alaska 99762. lated as “the one that sings” or “the singer.” The reference is to the bearded seal (in-air known as “mukluk” or “oogruk,” respectively). The song of Erignathus was described by Peter Freuchen (1921-24: pp. 224-225) as a “shrill, siren-like note which is made in the water and becomes deeper and deeper till it ends in a long-drawn-out sigh.” The sigh he described as a “strange, dull, deep-toned sound,” remark- ing that some “sighs” were accompanied by bubbles which he thought indicated that the seal would shortly surface. The description fits the underwater sound very well and has been quoted by other more recent authors. Poulter (1966) has described the underwater signals of Erigna- thus, but the spectrographic analyses he pre- sents are hard to compare with ours since no scale is indicated and a non-standard logarithmic frequency portrayal is used. It seems likely that these are sounds of Erignathus, but we do not find either the high (6 kcps) frequency starts or the consistently distinct pulses he describes. Over the past 10 years two of us (C. R. and J. B.) have observed Erignathus in the Bering Sea. These observations have included a num- ber of instances when the “singing” could defi- nitely be attributed to a seal in the water. Usually only a part of the song was heard in-air as a seal was closely approached, but through a series 79 80 Zoologica: New York Zoological Society [54: 2 of such exposures a general impression of the song has been obtained. These impressions have been confirmed by underwater recordings made from shore, from pack ice, and from small boats on 7, 8, 9, 10, 14, 20, and 21 May 1966 near Gambell, St. Lawrence Island, Alaska (C. R.) and analyzed at Woods Hole Oceanographic Institution (W. W.). The dates of recording corresponded to the height of the courtship sea- son. No other mammals were seen in the imme- diate vicinity during most of the listening pe- riods, except for a few walrus, Odobenus, whose underwater sounds were noted to be similar to some of those reported from a captive (Schevill, Watkins, and Ray 1966). The recordings were made with an LC-50 (Atlantic Research) hydrophone, a preamplifier (Watkins, 1963), and a Nagra III tape recorder. Analysis playback was by means of a Crown (800 series) recorder. The system was essen- tially flat from 50 to 10,000 cps. Spectrographic analyses were made on a Kay Electric Vibra- lyzer. Acoustical Results The song of Erignathus is both complicated and highly variable, yet by listening for an ex- tended period an overall pattern for the song may be noted. The entire pattern occasionally is heard in one song; more often the song is frag- mented with only parts of the song given and the variations predominating. At times a short rising trill may signal the repetition of a large part of the song, usually with additional varia- tions. Our listening sample appears to have been large enough and from a sufficiently varied locale so as to offset limiting oceanographic factors, such as selective frequency attenuation and temperature/pressure effects. The fact that the song may be heard through the water-air interface indicates a relatively intense sound. The song of Erignathus is a highly frequency- modulated call, much more so than that of any other marine animal we have herad. It may be continuous for more than a minute and is com- posed of short oscillations of the carrier- frequency (frequency-modulation) superim- posed on longer-duration variations of the carrier-frequency. The song gives the aural im- pression of a long, siren-like oscillating warble. The song starts at 2000 to 3000 cps and ends at 200 to 300 cps; there is an overall downward sweep in frequency throughout the song, though there may be several short-term upward excur- sions. The song appears to end with a separate unmodulated low-frequency moan of two to three seconds duration and usually slightly de- creasing frequency. In our recordings, singing seals were numer- ous and never sufficiently isolated from their fellows to be certain of the attribution of the entire song to one individual. However, the pat- tern appeared generally to be the same. The variability noted in the detail of the pattern may be the result of individual preference. Our representative song (Text-fig. 1 ) was as follows: 1 ) An introductory short warble which rose in frequency from 2500 to 3000 cps and lasted 2.5 seconds. 2) A second phrase 20 seconds in duration in which the carrier-frequency dropped from 3000 to 1000 cps. 3) A third phrase which lasted 3.5 seconds and rose to 2000 cps in carrier-frequency. 4) A repetition of phrase two with additional variations, 40 seconds in duration, which dropped to just below 1000 cps and was fol- lowed by a pause of 15 seconds. 5) An unmodulated moan which lasted 3 seconds and fell in frequency from 400 to 300 cps. Detailed description of the song as heard from many individuals is as follows. The be- ginning phrase of the song (PI. I) is a rising warble characterized by short bursts (100 to 250 msec, long) of frequency-modulation with oscillations of up to 1000-cps variation, separated by periods (100-150 msec.) of either less variation in frequency or unmodulated tone. This first phrase of the song may last 2.5 to 8 seconds. The introductory phrase is followed by a long second phrase (PI. II), an oscillating warble that may last a minute or more and that is composed of regular frequency oscillations produced at a relatively rapid rate (12-30 per second) superimposed on slower oscillations of the carrier frequency which vary at the rate of 3 per second to 1 per several seconds. The overall frequency drops during this phrase of the song and both the rapid frequency-modula- tion and the slower oscillation of the carrier- frequency occur progressively more slowly. Thus during the later portions of the second phrase, the song becomes more like a simple frequency-modulated warble in which the long carrier-frequency oscillations are of 5 to 10 seconds in duration. The sound may vary during frequency modulation by as much as 1000 cps, but the largest is usually about 500 cps. A third phrase consisting of a short warble sometimes interrupts the second phrase, usually to be followed by a repetition of phrase two as another “verse” of varying length. This ascend- 1969] Ray, Watkins & Burns: The Underwater Song of Erignathus (Bearded Seal) 81 ing warble is relatively short (3 to 7 seconds) and has regular frequency-modulation as well as a steadily rising carrier-frequency (PI. III). A typical insertion of the ascending trill into the song raises the carrier-frequency from about 1000 to 2000 cps allowing a repetition or a varia- tion of the last portions of the song. The ascend- ing warble is never used following a moan and is always inserted when the carrier-frequency of the song has dropped to 1000 cps or below. The moan (lower portion, PI. II) appears at the end of many of the Erignathus songs and so we include it as a part of the total song. Actually, there usually is a silence of up to 30 seconds (the interval is difficult to determine) between the last part of the slowly oscillating warble and the beginning of the moan. The moan is always lower in frequency than the earlier parts of the song. It is an unmodulated tone usually of descending frequency between 500 to 200 cps and of 2 to 3 seconds duration, contrasting sharply with the long frequency- modulated warble that precedes it. At close range, it is the moan that is associated with the appearance of bubbles and the subsequent sur- facing of the seal. Behavioral Observations In spite of its superficial variability, the gen- eral pattern of the song of Erignathus is stereo- typed and repetitive. It is complex and musical; it is seasonally produced and apparently sexu- ally distinctive. It may also be territorial in func- tion. In these respects the song of Erignathus fits the traditional usage of the term “song” as it is applied to the complex sound combinations of passerine birds. In an effort to determine the role of the song a number of seals were collected (J. B.) It is the habit of Erignathus to swim in leads or openings in loose pack ice and individuals are not gregarious. Of 17 identified as singing and collected, all were males; of another 19 collected in the immediate vicinity of singing, 15 were males. All of these males proved to be sexually mature. Contrastingly, in July, though many seals of both sexes have been observed, none has been heard to sing. This strongly implies that the male Erignathus is proclaiming either its breeding territory in the pack ice or simply its availability or perhaps both. This is not in agree- ment with Freuchen’s (1921-24) often quoted assertion that the song is used for communica- — 1 — > 1 1 1 — 1 — * 1 1 > 1 ■ — —i ' 1 ' r- 0 10 20 30 40 50 60 70 80 TIME - SECONDS Text-fig. 1. The lower portion of the drawing is a spectrographic portrayal of the characteristic (irregular sine-wave-like) variations in the carrier-frequency of an Erignathus song. Additional modulation is superimposed on these carrier-frequency variations and consists of relatively rapid frequency-modulation whose swing about the carrier-frequency may be as large as 1000 cps; this is portrayed in the upper portion of the drawing. During phrase one the frequency-modulation typically is in short bursts; phrase two has continuous but varying frequency modulation; phrase three has slower and more regular modu- lation; and the moan at the end is separated from the rest of the song by a silent interval and is unmodulated. 82 Zoologica: New York Zoological Society [54: 2 tion between mother and pup, nor with Poulter’s ( 1966) evident belief that the song has a sonar function. Our suggestions are reinforced by the ob- servations of the Eskimos. Those that were inter- viewed agreed that the song is heard only from March through June and is associated with rela- tively short dives of about three minutes dura- tion. They emphasized that bubbles always ap- pear at the surface shortly after the moan is heard and that these bubbles are used as a con- venient marker for the seal’s appearance where it may be killed by the hunter. The observation that in-air sound is used by phocid seals in territorial or courtship activities on land has been made for Mirounga, the ele- phant seal, by Bartholomew ( 1952) . That sound is important in underwater courtship in its world of shore ice has also been suggested for Leptonychotes, the Weddell seal, by Ray (1967). Territory in the case of Erignathus would imply a lead or opening in the pack ice. Summary and Conclusions The underwater song of Erignathus usually consists of a long oscillating frequency-modu- lated warble that may be more than a minute in duration, followed by a short unmodulated low- frequency moan. The song typically starts at about 2000 cps with many frequency variations and ends as low as 200 cps. The song apparently is used only by mature males during the spring courtship season. It is suggested that its purpose is a proclamation of territory or of breeding condition or both. Literature Cited Bartholomew, G. A., Jr. 1952. Reproductive and social behavior of the northern elephant seal. Univ. Calif. Publ. Zool., 47:369-472. Burns, J. J. 1967. The Pacific bearded seal. Alas. Dept, of Fish and Game. Juneau. 66pp. Freuchen, P. 1921-4. Mammals. Part II, Rept. Fifth Thule Exped., 2(4 & 5) :68-278. POULTER, T. C. 1966. Systems of echolocation. In Les Systemes Sonars Animaux, Ed R.-G. Busnel. Lab. Physiol. Acous., Jouy-en-Josas, France, 1:157-185. Ray, Carleton 1967. Social behavior and acoustics of the Wed- dell seal. Antarctic Jour., 2(4 ): 105-106. Schevill, W. E., W. A. Watkins, and Carleton Ray 1966. Analyses of underwater Odobenus calls with remarks on the development and function of the Pharyngeal pouches. Zoo- logica, 51(3): 103-105. Watkins, W. A. 1963. Portable underwater recording system. Undersea Tech., 4(9):23-24. Acknowledgments We wish to thank Dr. F. H. Fay of the Arctic Health Research Laboratory, U. S. Public Health Service, College, Alaska, for aid in the identification of sounds. Winfred James and numerous eskimos of Gambell, Alaska, aided us in the field. We also wish to thank W. E. Schevill and Dr. R. H. Backus, both of Woods Hole Oceano- graphic Institution, for many helpful criticisms during preparation of the manuscript. This work was sponsored by the following grants: the New York Zoological Society (C. Ray); the Arctic Institute of North America to The Johns Hop- kins University under contractual agreements with the Office of Naval Research (C. Ray); the Office of Naval Research contract Nonr 4446 ( W. A. Watkins) ; Federal Aid to Wildlife funds, Pittman-Robertson Project W-6-R and W-14-R (J, J. Burns). 1969] Ray, Watkins & Burns: The Underwater Song of Erignathus (Bearded Seal) 83 EXPLANATION OF THE PLATES Plate I The song of Erignathus begins with an introduc- tory warble (phrase one) that alternates short bursts of frequency-modulation separated by periods of relatively unmodulated tone. The sounds below 2 kcps are from more distant seals. The analyzing filter bandwidth is 200 cps. Plate II A later portion during the Erignathus song (phrase two) is portrayed in the upper part of the spectro- gram and shows rapid frequency-modulation super- imposed upon carrier-frequency oscillations of longer duration. The lower part illustrates a moan at the end of another seal’s song. The effective filter bandwidth in this analysis is 120 cps. Plate III Sometimes interrupting phrase two is a short rising warble (phrase three) in which short-term modula- tions are regular and there are no longer oscillations as in phrase two. The analyzing filter bandwidth is 400 cps. Inserted Phonograph disk of the underwater song of Erignathus (Bearded Seal). RAY, WATKINS & BURNS PLATE I THE UNDERWATER SONG OF ERIGNATHUS (BEARDED SEAL) RAY, WATKINS & BURNS PLATE II in co O m o in cvJ cvi ^ O THE UNDERWATER SONG OF ERIGNATHUS (BEARDED SEAL) RAY, WATKINS & BURNS PLATE III Co NEW YORK ZOOLOGICAL SOCIETY The Zoological Park, Bronx, N. Y. 10460 Laurance S. Rockefeller President John Pierrepont Treasurer OFFICERS Robert G. Goelet Executi ve V ice-President Chairman of the Executive Committee Howard Phipps, Jr. Secretary Henry Clay Frick, II Vice-President Eben W. Pyne Assistant Treasurer Edward R. Ricciuti Editor & Curator, Publications & Public Relations EDITORIAL COMMITTEE Robert G. Goelet Chairman William G. Conway Hugh B. House Donald R. Griffin F. Wayne King William G. Conway General Director Joan Van Haasteren Associate Editor Peter R. Marler Ross F. Nigrelli ZOOLOGICAL PARK William G. Conway . . . Director & Curator, Ornithology Hugh B. House .... Curator, Mammalogy Grace Davall . . Assistant Curator, Mammals & Birds Walter Auffenberg . . . Research Associate in Herpetology Joseph Bell . . Associate Curator, Ornithology F. Wayne King .... Curator, Herpetology William Bridges . Curator of Publications Emeritus John M. Budinger . . . Consultant, Pathology Ben Sheffy Consultant, Nutrition James G. Doherty . Assistant Curator, Mammalogy Donald F. Bruning Ornithologist Joseph A. Davis, Jr Scientific Assistant to the Director AQUARIUM Ross F. Nigrelli Director Robert A. Morris Curator Christopher W. Coates . . . Director Emeritus U. Erich Friese Assistant Curator Nixon Griffis .... Administrative Assistant Louis Mowbray . Research Associate in Field Biology OSBORN LABORATORIES OF MARINE SCIENCES Ross F. Nigrelli . . . Director and Pathologist Harry A. Charipper . . Research Associate in Martin F. Stempien, Jr. ... Assistant to the Histology Director & Bio-Organic Chemist Kenneth Gold Marine Ecologist George D. Ruggieri, S.J. . . . Coordinator of Myron Jacobs Neuroanatomist Research & Experimental Embryologist Klaus Kallman Fish Geneticist William Antopol . . . Research Associate in Vincent R. Liguori Microbiologist Comparative Pathology John J. A. McLaughlin . . Research Associate in C. M. Breder, Jr. ... Research Associate in Planktonology Ichthyology Martin P. Schreibman . . Research Associate in Jack T. Cecil Virologist Fish Endocrinology Jay Hyman Research Associate in Comparative Pathology INSTITUTE FOR RESEACH IN ANIMAL BEHAVIOR [Jointly operated by the Society and The Rockefeller University, and including the Society’s William Beebe Tropical Research Station, Trinidad, West Indies] Donald R. Griffin .... Director & Senior Fernando Nottebohm . . . Research Zoologist Research Zoologist George Schaller Research Zoologist Peter R. Marler . . . Senior Research Zoologist Thomas T. Struhsaker . . . Research Zoologist Jocelyn Crane . . . Senior Research Ethologist C. Alan Lill Research Associate Roger S. Payne Research Zoologist O. Marcus Buchanan . . , Resident Director, Richard L. Penney .... Research Zoologist William Beebe Tropical Research Station Paul Mundinger Research Associate SMITHSONIAN INSTITUTION LIBRARIES