H X CO ^ o X* „ _ . „ 2 S v > xs^0j.ii5^/r ^ ^ 3> ^sss?*^ *s: >, CO * ^ W *■ 2 • M 2 DN NOlinillSNI NVIN0SH1IWS S3!HV*JSI1 LIBRARIES SMITHSONIAN INSTT CO — CO — m n LIBRARIES SMITHSONIAN INSTITUTION NOIlfUllSNI NVINOSH1MS S 3 1 Jd 2 r- 2 m \^vos»^ ^ rn >N%0!lfUIJL$NI~NViNOSHllWS S3 I B V 8 8 1 1~ L I B R A R I ES ^SMITHSONIAN INST! CO 2 W . 2 v CO 2 __ 5 jg 11 LIBRARIES SMITHSONIAN INSTITUTION^ NOlifUIJLSNI NVINOSHiUMS^SB I a’ CO = CO X \ CO O \^OSHiX — O 2 _J 2 _ )N NOimiliSNI NVINOSH1IWS S3 I HVH8n LIBRARIES SMITHSONIAN INSTil 2 ___ r* 2 r- 2 o /»\ — o *“* ~ co X x co x co M__L I B RAR I ES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1MS S3 I Hj 2 ^ CO 2 __ CO < 2 .< z .-w\>: -h /Mkt»WL> - 5 S > )N NOIXflXIXSNI NVINOSHXIINS^SB I HVH 8 ll"2 LI B RAR I ES^SMITHSONIAN INST!' CO ~ CO — tn O " ' ” Q \oN-D,c/ — ■’ ^ O 1 L! B RAR I ESZSMITHS0NIANJ|NSTiTUTI0N2N01inillSNI JNVIN0SH1!WSZS3 I 2 r- _ 2 03 / m x^yosv^>r ^ rn ^ {TJ >N NOIlOillSNi NVIN0SH1IWS S3 I HVH 3 II ~LI B R AR I ES ^SMITHSONIAN INSTH CO 2 CO 2 • CO 2 V 2 < V 2 X^Tri^Sv < & | VA z _ > 2 DiiniiiSNi^NivmosHiiHs^sa I ava a nf li b rar i es smithsoniaOnstitutiok C/3 ~r 07 2 ^77Tr\ O B RAR I ES^SMITHS0NIAN^INSTITUTI0NzN0linilXSNI(_NVIN0SHilWS2S3 I a VB a I JIXfUllSNI NVIN0SH1IWS“S3 I HVH 3 I1~LI B R AR I ES^SMITHSONIAN^INSTITUTIOt w¥% o I s '^$'rosS?/ >’ S ‘'.§P § y, B R AR I ESU,SMITHSONIANZ|NSTITUTIONt/5NOiinillSNI_NV]NOSHllWSwS3 I BV3 8 I 0 X^VOSV^X _ ^ ^UnilXSNI^NVINOSHilWS2 S3 IHV3Sn'J LIBRARIES SMITHSONIAN ^INSTITUTIO Z . o _ w m p.c>^ ^ rn B RAR I ES ^SMITHSONIAN ""INSTITUTION N01inilI$N!^NVM0SHlMS^S3 I 8 ! < „ S < S 2 f£\ O Is <£ •V x i ^ ^ s > 3UniIXSN|ZNVINOSHillNS^S3 I BVB 8 LI B RAR I ES^SMITHSONIAN^INSTITUTIC **$$0 5 ^ 2 | N35°y>' = w- o B RAR I ESZSMITHS0mAN'J|NSTITUTI0NZN0IXnXIXSNI^NVlN0SHXmSzS3 I HV3 8 ^ I A 5 1 S 1 5 ^Ifl t ~ CO ^ m x^osv^x ^ £} ' . w z DUnXIXSNI ~ NVIN0SHXUNS^S3 I 3 VH 8 11 ^L I B R AR I ES ^SMITHSONIAN^INSTITUTIC yy. = 1 1 S x^>Tir^x < h 1 LOS , * t ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE .'JJ 1 2- Il8Q8 | December 31, 1968 TABLE OF CONTENTS and AUTHOR INDEX 1967-1968 Nos. 122-154 Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 2 Contributions in Science TABLE OF CONTENTS No. 122. A new apterous genus and species of Aradinae from Mexico (Hemip- tera-Heteroptera, Aradidae), by Nicholas A. Kormilev. 4 pp., 2 figs. January 30, 1967. No. 123. New Species of North American Ammophila, Part III. (Hymenop- tera, Sphecidae), by A. S. Menke. 8 pp., 2 figs. January 30, 1967. No. 124. Studies on California ants. 3. The taxonomic status of Proceratium calif ornicum Cook ( Hymenoptera : Formicidae), by Roy R. Snelling. 10 pp., 1 fig. May 31, 1967. No. 125. The American atherinid fishes of the genus Coleotropis, by Carter R. Gilbert and David K. Caldwell. 16 pp., 5 figs. May 31, 1967. No. 126. A new species of Late Oligocene dog, Brachyrhynchocyon sesnoni, from South Dakota, by J. R. Macdonald. 5 pp., 2 figs. May 31, 1967. No. 127. A new species of Late Oligocene dog, Sunkahetanka sheffteri, from South Dakota, by J. R. Macdonald. 5 pp., 2 figs. May 31, 1967. No. 128. The marine fish fauna, based primarily on otoliths, of a Lower Pleis- tocene deposit at San Pedro, California (LACMIP 332, San Pedro Sand), by John E. Fitch. 23 pp., 31 figs. May 31, 1967. No. 129. Aspects of the biology of the lizards of the White Sands, New Mex- ico, by James R. Dixon. 22 pp., 12 figs. July 28, 1967. No. 130. Results of the 1966 Cheney Expedition to the Samburu District, Ken- ya. Ornithology, by Herbert Friedmann and Kenneth E. Stager. 34 pp.,5 figs. July 28, 1967. No. 131. On the age distribution of Smilodon calif ornicus Bovard from Ran- cho La Brea, by George J. Miller. 17 pp., 1 1 figs. February 20, 1968. No. 132. Taxonomic notes on some Mexican cephalotine ants (Hymenoptera: Formicidae), by Roy R. Snelling. 10 pp., 2 figs. February 20, 1968. No. 133. A mandible of Didelphodon vorax (Marsupialis, Mammalia), by William A. Clemens, Jr. 1 1 pp., 5 figs. February 20, 1968. No. 134. Populational variation in the frog genus Phrynohyas Fitzinger in Middle America, by Roy W. McDiarmid. 25 pp., 7 figs. February 20, 1968. No. 135. New genera and species of wasps of the tribe Trypoxylonini from the neotropical region (Hymenoptera, Sphecidae, Larrinae), by Arnold S. Menke. 9 pp., 1 fig. February 20, 1968. No. 136. A new porcupine from the Middle Pleistocene of the Anza-Borrego Desert of California, with notes on mastication in Coendou and Ere- thizon, by John A. White. 15 pp., 8 figs. February 20, 1968. No. 137. Swollen dorsal fin elements in living and fossil Caranx (Teleostei: Carangidae), by Harry L. Fierstine. 12 pp., 5 figs. May 7, 1968. Table of Contents 3 No. 138. Review of the lanternfish genus Lampadena with a description of a new species, by Basil G. Nafpaktitis and John R. Paxton. 32 pp., 10 figs. May 7, 1968. No. 139. A new genus of mustelid from the Ellensburg Formation, Washing- ton, by Laurie J. Bryant. 8 pp., 2 figs. May 7, 1968. No. 140. A new plethodontid salamander from Sonora, Mexico, by Charles H. Lowe, Clyde J. Jones, and John W. Wright. 11 pp., 3 figs. June 14, 1968. No. 141. A new subfamily of blind beetle from Idaho ice caves with notes on its bionomics and evolution (Coleoptera: Leiodidae), by Richard L. Westcott. 14 pp., 6 figs. June 14, 1968. No. 142. Tertiary birds from Laguna Hills, Orange County, California, by Hildegarde Howard. 21 pp., 3 figs. No. 143. A new piculet from Amazonian Bolivia, by Kenneth E. Stager. 2 pp. June 14, 1968. No. 144. Studies on North American bees of the genus Hylaeus. 4. The sub- genera Cephalylaeus, Metziella and Hylaeana (Hymenoptera: Col- letidae) , by Roy R. Snelling. 6 pp., 3 figs. June 14, 1968. No. 145. Three new epitonid gastropods from the Panamic Province, by Helen DuShane and James H. McLean. 6 pp., 6 figs. June 14, 1968. No. 146. Otoliths and other fish remains from the Timms Point Silt (Early Ple- istocene) at San Pedro, California, by John E. Fitch. 29 pp., 4 figs. June 14, 1968. No. 147. Occurrence of a giant bison, Bison latifrons, and a slender-limbed camel, Tanupolama, at Rancho La Brea, by Wade E. Miller. 9 pp., 6 figs. June 14, 1968. No. 148. Records of bats from Costa Rica, by Andrew Starrett and Richard S. Casebeer. 21 pp. June 28, 1968. No. 149. Photonectes munificus, a new species of melanostomiatid fish from the South Pacific Subtropical Convergence, with remarks on the con- vergence fauna, by Robert H. Gibbs, Jr. 6 pp., 1 fig. June 30, 1968. No. 150. Chiggers of the genus Pseudoschoengastia (Acarina: Trombiculidae) from Costa Rica, by Julius C. Geest and Richard B. Loomis. 49 pp., 22 figs. July 11, 1968. No. 151. A new Early Tertiary Perissodactyl, Hyracotherium seekinsi, from Baja California, by William J. Morris. 11 pp., 5 figs. December 13, 1968. No. 152. A new Central American sand fly breeding in crab holes (Diptera, Ceratopogonidae), by Charles L. Hogue and Willis W. Wirth. 7 pp., 13 figs. December 31, 1968. No. 153. A new piculet from southeastern Peru, by Kenneth E. Stager. 4 pp., 2 figs. December 31, 1968. No. 154. A new species of Eurhopalothrix from El Salvador (Hymenoptera: Formicidae), by Roy R. Snelling. 4 pp., 1 fig. December 31, 1968. 4 Contributions in Science AUTHOR INDEX Bryant, Laurie J. Caldwell, David K. Casebeer, Richard S. Clemens, William A., Jr. Dixon, James R. DuShane, Helen Fierstine, Harry L. Fitch, John E. Friedmann, Herbert Geest, Julius C. Gibbs, Robert H., Jr. Gilbert, Carter R. Hogue, Charles L. Howard, Hildegarde Jones, Clyde J. Kormilev, Nicholas A. Loomis, Richard B. Lowe, Charles H. Macdonald, J. R. McDiarmid, Roy W. McLean, James H. Menke, A. S. Miller, George J. Miller, Wade E. Morris, William J. Nafpaktitis, Basil G. Paxton, John R. Snelling, Roy R. Stager, Kenneth E. Starrett, Andrew Westcott, Richard L. White, John A. Wirth, Willis W. Wright, John W. No. 139 No. 125 No. 148 No. 133 No. 129 No. 145 No. 137 Nos. 128, 146 No. 130 No. 150 No. 149 No. 125 No. 152 No. 142 No. 140 No. 122 No. 150 No. 140 Nos. 126, 127 No. 134 No. 145 Nos. 123, 135 No. 131 No. 147 No. 151 No. 138 No. 138 Nos. 124, 132, 144, 154 Nos. 130, 143, 153 No. 148 No. 141 No. 136 No. 152 No. 140 LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE MBER 122 January 30, 1967 A NEW APTEROUS GENUS AND SPECIES OF ARADINAE FROM MEXICO (HEMIPTERA-HETEROPTERA, ARADIDAE) ii By Nicholas A. Kormilev I Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 A NEW APTEROUS GENUS AND SPECIES OF ARADINAE FROM MEXICO (HEMIPTERA-HETEROPTERA, ARADIDAE) By Nicholas A. Kormilev1 Abstract: A new genus ( Aradiohis ) and species (A. para- doxus) of flat bug belonging to the Aradinae is described from Mexico. The species represents the only apterous form known in the subfamily. By the kind offices of Dr. Charles L. Hogue, Curator of Entomology at the Los Angeles County Museum of Natural History, I have had an oppor- tunity to study a very interesting new apterous aradid from Mexico, for which I wish to express to him my sincere thanks. This specimen represents a new genus belonging to the subfamily Ara- dinae, in which were previously known, besides macropterous, also brachyp- terous and stenopterous, but not apterous forms. In the new genus the head, antennae, tip of the abdomen, and the whole ventral surface of the body, are those of an Aradus, but the thorax is quite different, somewhat resmbling that of Tretocoris Usinger and Matsuda (1959), an apterous genus from New Zealand belonging to the subfamily Chinamyersinae. Aradiolus, new genus Description : Head longer than width across the eyes. Anterior process of the head, and antennal segments II and III, covered with dense, short, inclined, incrustated bristles, giving them a look of roughness. Similar, but shorter, bristles on an- tenniferous tubercles and vertex. Anterior process long and strong, slightly tapering toward the tip, reaching over the base of antennal segment II. Anten- niferous tubercles acute, divergent, with a minute lateral tooth. Eyes globose, and slightly stalked, placed near the hind border of the head. Preocular teeth distinct, postocular rounded. Vertex raised in the middle anteriorly, elevation surrounded posteriorly by a semicircular depression. Antennae strong, less than twice as long as the head, and as thick as fore femora. Antennal segments I and IV the shortest and equal in length, II more than three times as long, and III more than twice as long as I. Rostrum placed far from the tip of anterior process; rostral atrium open; bucculae short, rounded, not contiguous anterior- ly; rostral groove moderately deep, open posteriorly; rostrum thin, reaches to hind border of prosternum. Notum partially, and dorsal surface of abdomen almost entirely, covered with dense, round scales. Pronotum very short and wide. Collar indistinctly separated from the disc. Anterior angles produced into small, rectangular lobes. Lateral borders ex- x365 Lincoln Place, Brooklyn, New York 11238 2 1967 New Flat Bug 3 Figures 1-3. Aradiolus paradoxus, new genus and species. Fig. 1. dorsal aspect. Fig. 2. hypopygium, dorsal aspect. Fig. 3. sternum VII and genital cap, ventral aspect. panded into two hammer-shaped lobes. Disc with two parallel carinae along median sulcus, and with two semicircular, black, opaque spots. Laterad of the latter are two fine longitudinal carinae. Mesonotum separated from pronotum by a thin, but distinct sulcus. Mid- dle of meso- and metanotum together occupied by a spear-shaped plate, de- pressed in the middle. Four small, black, opaque spots along fore border of mesonotum; two small, round, opaque spots behind outer of these. Lateral borders expanded into two “T”-shaped lobes. Metanotum laterad of median plate separated from mesonotum by thin sulci; its oblique hind borders, laterad of median plate, semifused with tergum I. Lateral borders expanded into two “T”-shaped lobes. Disc with similar spots as mesonotum. Wings completely absent. Abdomen ovate, longer than maximal width across segment V. Tergum I separated from tergum II by a fine, transverse sulcus. Tergum II separated from central dorsal plate by a more distinct sulcus. Formula of round callous spots on terga, 2:1:1. Scars of dorsal scent gland openings situated between terga III, IV, V, and VI respectively. Connexival segments clearly separated from each other, but terga separated from each other only laterally, boundaries 4 Contributions in Science No. 122 absent mesally. Disc raised along median line, forming a wide, gently sloping ridge. Segment VIII in the male forming two large genital lobes, similar to Aradus. Hypopygium covered from ventral side by a semiglobose genital cap (segment IX), but open on dorsal side, as in Aradus. Exterior borders of connexiva II to IV straight, but angularly rounded on connexiva V to VII. Spiracles small, ventral, placed far from lateral, and nearer to fore border, on sterna II to VII; lateral and visible from above on genital lobes (VIII). Ventral surface of the body similar to that of Aradus. Pro- and mesoster- num deeply sulcate on median line for the reception of rostrum; metanotum and venter finely sulcate on median line. Legs unarmed; trochanters completely fused with femora, the latter cylindrical; tibiae also cylindrical; fore tibiae near the apex interiorly with a minute tooth, or spine. Claws without arolia. Type species'. Aradiolus paradoxus, new species In addition to the above description, further characters pertaining to the type specimen are as follows: Aradiolus paradoxus, new species Figures 1 through 3 Male'. Ovate, flat. Piceous; central meso-metanotal plate, bases of meso- metanotal lateral expansions, the whole abdomen from above, and the ventral surface of the body reddish brown, with exception of genital cap and hypo- pygium which are black; round callous spots on terga II to VII reddish. Tibiae with three black rings (basal, mesal, and apical). Measurements : Head proportions 45:40 (= length:width of measured part); proportions of antennal segments, I to IV, are 11:37:24:11. Pronotum 17: 62.5; mesonotum, with exception of meso metanotal median plate, which is 30 units long, 20:81; metanotum 10:84. Abdomen 120:104, at segment V. Total length, 8.4 mm.; width of pronotum, 2.5 mm.; width of abdomen, 4. 16 mm. Holotype: Male, MEXICO, Oaxaca, 12 mi. so. of Chivela, A. S. Menke and L. A. Stange coll., VIII. 18, 1959. Deposited in the Los Angeles County Mu- seum of Natural History. LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE 1TBER 123 January 30, 1967 \D1> 73 NEW SPECIES OF NORTH AMERICAN AMMOPHILA, PART III. (HYMENOPTERA, SPHECIDAE) By A. S. Menke It i; ■ ! il jt Los Angeles County Museum of Natural EJistory Los Angeles, California 90007 Exposition Park NEW SPECIES OF NORTH AMERICAN AMMOPHILA, PART III.1 (HYMENOPTERA, SPHECIDAE)2 By A. S. Menke3 Abstract: Ammophila moenkopi and shoshone are de- scribed from Arizona, and Wyoming and Utah, respectively. Ammophila azteca Cameron, new subspecies clemente, is de- scribed from San Clemente Island off the coast of southern Cali- fornia. These species belong in the azteca species group which is defined, and the included species are listed. New synonymy is given as follows: azteca Cameron, 1888 (= pilosa Fernald, 1934, aculeata Fernald, 1934), peckhami (Fernald), 1934 ( = willi- stoni Fernald, 1934), strenua Cresson, 1865 (— denningi Mur- ray, 1951), varipes Cresson, 1865 (= comanche Cameron, 1888). Among material recently sent to me for identification by Mr. and Mrs. J. Davidson, Arizona State University, was a new species collected in north- east Arizona. This species, and another undescribed and related species from Wyoming and Utah are being described now so the names will be available to the Davidsons who are engaged in a study of Ammophila distribution in Arizona. An undescribed subspecies of Ammophila azteca from one of the Chan- nel Islands off the coast of southern California has been found in material sent to me by Dr. Fred S. Truxal of the Los Angeles County Museum of Natural History. This material was collected by the museum’s Channel Island Biologi- cal Survey team in 1939. Additional specimens of this new subspecies were subsequently located in material on loan from Cornell University and the United States National Museum. I am describing this subspecies here because azteca is a close relative of the two new species. A more comprehensive treat- ment of these new taxa will be given when my revision is published. Abbreviations used in citing type depositories are as follows: Arizona State University (ASU), Cornell University (CU), Los Angeles County Mu- seum of Natural History (LACM), United States National Museum (USNM), University of California, Davis (UCD), California Academy of Sciences (CAS), and the Museum of Comparative Zoology, Harvard (MCZ). The species described here belong to an assemblage I call the azteca group. All of the species in this group have a long preepisternal sulcus; e.g., it xParts I and II appeared in Acta Hymenopterologica 2:5-27 (1964) and Proc. Biol. Soc. Wash. 79:25-40 (1966). 2 A biproduct of research supported by a Sigma Xi-R.E.S.A. Grant-in-Aid. 3Research Associate, Los Angeles County Museum of Natural History; Department of Entomology, University of California, Davis. 1967 New Species of Genus Ammophila 3 extends well below the pronotal lobe and ends in the sternal region. In those species of the group in which the sculpture of the mesopleuron is obscured by appressed hair one must either scrape away the hair to see the sulcus, or study the hair carefully to see if a vertical line is discernible. With proper lighting and body angle this is easy to do after a little practice. The male genitalia are diagnostic in only a few species of this group. The aedeagi of the two species described here are similar to figures 12 and 14 in Menke (1964). The species assigned to this group are: acuta (Fernald), azteca Cameron (=z pilosa Fernald,4 aculeata Fernald4) breviceps Smith, calif arnica Menke, 4New synonymy based on study of types. evansi Menke, hard (Fernald), karenae Menke, mediata Cresson, peckhami (Fernald) (= willistoni Fernald4), pruinosa Cresson, regina Menke, strenua Cresson ( = denningi Murray4) and varipes Cresson ( = comanche Camer- on4). Ammophila moenkopi Menke, new species Figures 1 and 3 Holotype male : length 14 mm. Color : Black; tegula and metapleural flange red; petiole tergite red later- ally; gastral segments I-III red, tergite IV red laterobasally, sternite IV red; legs red except coxae, mid- and hindtrochanters and basal half of hind femur; wings clear, veins yellowish-red basally, brown apically. Vestiture : Head and mesosoma (except propodeal enclosure) covered with long dense erect silver hair; frons, clypeus, and gena with dense appressed silver hair; pronotal collar and scutum sparsely covered with appressed silver hair (sculpture not obscured); pronotal lobe, thoracic pleura and propodeal side densely covered with appressed silver hair (sculpture obscured). Structure : Labrum truncate but corners broadly rounded; clypeal surface broadly, evenly swollen, free margin arcuate in outline but with a broad, shal- low median emargination; collar outline in lateral view as in Figure 3a; collar and scutum moderately macropunctate, interspaces shining although with microsculpture; scutellum sparsely punctate anteriorly, longitudinally ridged posteriorly; propodeal enclosure transversely ridged, interspaces not punctate; metapleural flange broadly lamellate, outer margin angulate, angle with a deep, narrow notch (Fig. 1); midtibia with one spur. Female : Average length 15 mm, range: 11 to 18 mm. Color : As in male except mandible red basally, clypeal margin frequently red, petiole sternite red, petiole tergite frequently all red, gaster red except for black spot on tergite V, legs red except for black coxae and sometimes black hind trochanter dorsally. Vestiture : Same as male. Structure : Labrum rounded, clypeal disk moderately bulging, very sparse- ly macropunctate, punctures confined mainly to lateral areas, median free 4 Contributions in Science No. 123 Figures 1-2. Metapleural flange — 1, Ammophila harti and moenkopi; 2 a-c, A. azteca, showing variation of flange in this species. Figures 3-5. Lateral profile of the pronotum, a = male, b = female — 3, A. moenkopi ; 4, A. harti’, 5, A. shoshone. margin usually poorly defined because of weakly formed lateral tooth; inner orbits moderately converging below; collar outline as in Figure 3; midtibia usually with two spurs; mesosomal sculpture and other details as in male. Types'. Holotype male: 4.5 mi. E. Moenkopi, on Poliomintha incana, Coconino Co., Arizona, 14 June, 1966, J. M. Davidson and M. A. Cazier (ASU). Twenty-four male and sixty-nine female paratypes from the following Arizona localities: Coconino Co.: Moenkopi, 2 to 4.5 mi. E., 13-27 June, 1966, on Poliomintha incana, J. M. Davidson and M. A. Cazier (ASU, UCD, USNM, LACM, CAS, MCZ). Navajo Co.: Hotevilla, 7000', 28 June, 1966, on Poliomintha incana, J. M. Davidson and M. A. Cazier (ASU, UCD); Jadito Trading Post, 28 June, 1966, on Poliomintha incana, J. M. Davidson, and M. A. Cazier (ASU) . Discussion: In some males the petiole sternite shows some reddening ventrally and gastral sternite IV is sometimes black. The apex of the hind tibia is occasionally blackish in both sexes. Although the base of the hind femur is black, the extreme base is reddish in nearly all the specimens studied. Out of 1967 New Species of Genus Ammophila 5 twenty-four males only three had two midtibial spurs, and in two of these cases one leg had only one spur. Occasional females have one spur on one leg and one specimen had only one spur on both midtibiae. Except for the red legs and extensively red abdomen this species is similar to harti (Fernald) which has black legs, and which incidentally, was collected with moenkopi at its type locality on the same date. The form of the collar is also different in the two species. In harti it is somewhat longer in lateral pro- file, and more arcuate. In moenkopi the lateral profile is in the form of a grad- ual slope from the transverse line to the top of the collar. Thus, the top of the collar is narrower (from front to back) in moenkopi than in harti (compare Figs. 3 and 4). The clypeus of female moenkopi differs from harti in having much sparser punctation and a slightly duller surface. In harti the clypeus is rather evenly, moderately macropunctate and the surface is strongly shining. Besides moenkopi and harti the only other species in the azteca group that have a lamellate metapleural flange are shoshone n.sp. and azteca Cameron. The red legs and yellowish red color of the veins at the wing base easily separate moen- kopi from both of these however. Ammophila shoshone Menke, new species Figure 5 Holotype male : Length 15.5 mm. Color : Black; tegula pale posteriorly; media and cubitus of forewing red- dish brown basally, veins elsewhere brown; petiole tergite and gastral tergite I red laterally below spiracle, gastral sternite I red. Vestiture : Head and mesosoma (except propodeal enclosure) with dense erect silver hair; frons, clypeus, and pronotal lobe with dense appressed silver hair; mesopleuron sparsely covered with appressed silver hair (sculpture visi- ble) but denser along mesopleural suture forming a narrow band from mid- coxa to bottom of hypoepimeral area; metapleuron and propodeal side sparsely covered with appressed silver hair (sculpture visible) but somewhat denser on inferior metapleural area near hindcoxa. Structure : Labrum truncate but corner broadly rounded; clypeal surface broadly evenly swollen, free margin arcuate in outline but with a broad, shal- low median emargination; collar outline in lateral view as in Figure 5a; collar and scutum moderately macropunctate, interspaces shining although with microsculpture; scutellum moderately punctate anteriorly, longitudinally ridged posteriorly; propodeal enclosure transversely ridged, interspaces not punctate; pleura and propodeal side moderately macropunctate, the punctures somewhat larger than on scutum; metapleural flange lamellate, angulate, the angle notched (similar to Fig. 1) ; midtibia with one spur. Female : Average length 18 mm. Color : Same as male except petiole tergite and gastral segment I red, II 6 Contributions in Science No. 123 red but tergite has an apical black spot (entirely red in Pilgrim Creek Camp female). Vestiture : Essentially as in male except appressed silver hair of face sparser and sometimes lacking. Structure : Clypeal disk moderately and evenly bulging, moderately macro- punctate except anteromedially where punctures are sparser, median free mar- gin scarcely defined because lateral tooth is very obtuse or absent; inner orbits moderately converging below; collar outline as in Figure 5b; midtibia with two spurs; other details as in male. Types'. Holotype male: Horse Creek Camp, Shoshone National Forest, Wyoming 21 July, 1957, A. and H. Dietrich (CU) . Nine male and four female paratypes with same data as type (CU, UCD, LACM) and one female from Pilgrim Creek Camp, Grand Teton National Park, Wyoming, 27 July, 1957, A. and H. Dietrich (CU), and one male from Utah Lake, Utah, 1 August, 1933, G. P. Engelhardt (CU). Discussion : The only appreciable variation among the males is the almost complete lack of red on the abdomen in one male, the more extensive redden- ing of gastral tergite I (red extends above spiracle) in three others, the presence of a trace of appressed silver hair on the gena in some others, and the presence of two midtibial spurs in six of the males (only on one leg in three of these). This species is very similar to A . harti in every respect except color, scar- city of appressed silver hair, and the form of the pronotal collar. In lateral profile the collar is shorter in shoshone than in harti (compare Figs. 4 and 5). This seemingly slight difference is supported by the almost entirely black color of male shoshone. Color is very variable in harti, as would be expected in a species which has a broad geographical distribution (Arizona to Alberta and east to Quebec and North Carolina), but I have not seen a single male of harti in which melanism has reached the state found in shoshone males. Even the blackest males of harti have gastral tergite I completely red. The identification of females of shoshone and harti will be difficult since the only reliable means of separation is by the form of the collar and/or association with males. I have not seen any harti from the areas in which shoshone occurs, and this fact sug- gests that the latter may eventually prove to be a synonym. The red color and collar shape differences may only be the result of environmental influences, but a decision on this depends on the collection and study of more material from various Rocky Mountain area localities. However, I have seen one male of harti from 10 mi. N. Flowell, Millard Co., Utah (UCD) which is south of Utah Lake, a shoshone locality. This specimen is typical harti. Ammophila azteca and moenkopi are the only other members of the azteca group with a lamellate metapleural flange. Ammophila moenkopi is easily separated from shoshone by its red legs. The separation from azteca may prove more difficult. Ammophila azteca is one of the most wide ranging species of the genus in North America being found from coast to coast al- though only in the northern states east of the Rockies. Its range extends north- 1967 New Species of Genus Ammophila 7 ward beyond the Arctic Circle in the Yukon Territory, and into Labrador in the east. Over this range it displays a perplexing array of variation in both color and sculpture. Some of these varieties mimic shoshone fairly well and they occur in the same area. This situation makes it difficult to separate the two species unless one has a thorough knowledge of azteca. The wing veins of azteca are uniformly brownish black and the scutal and pleural sculpture is rougher. The scutal punctation in azteca is usually denser than in shoshone and the interspaces tend to form wrinkles. The mesosoma of azteca is shorter in comparison with shoshone because in the latter the propodeum is more elongate. The metapleural flange is variable in azteca and it is often only nar- rowly lamellate or not lamellate (Fig. 2a-c). A. azteca males usually have two midtibial spurs. The median lobe of the clypeus is well defined in azteca females because of a strong lateral tooth. Ammophila azteca clemente Menke, new subspecies This insular population is distinguished from typical azteca by color and vestiture. There are no structural differences. Diagnosis'. Black except for faint reddish tints on the petiole tergite later- ally and on gastral segments ITI ventrally; wings lightly infumate; erect hair of head and mesosoma dirty white tending towards brownish dorsally; ap- pressed silver hair found only on clypeus and frons of male, pronotal lobe, mesopleuron at midcoxa, and at side of petiole socket; face of female with sparse appressed brown hair, but sometimes with a trace of silver. Types : Holotype male, San Clemente Island, Los Angeles Co., California, 3 April, 1939, collected by Los Angeles County Museum Channel Islands Biological Survey team (LACM). Seven male and seven female paratypes, all from San Clemente Island: 3-4, April, 1939, L.A. Co. Mus. Channel Island Biol. Surv. (LACM, UCD); 8-12 April, 1923, collector unknown (CU, UCD) ; May, 1939, J. T. Scott (USNM). Discussion : The high degree of melanism in clemente is quite striking and easily sets this subspecies apart from the normally bicolored mainland azteca. In typical azteca males the petiole tergite and gastral segment I are red except for a narrow dorsal black stripe on each (sometimes the gastral tergite is completely red). In western azteca males the second, and sometimes the third, gastral segment may be largely red also. Females of typical azteca have the same amount of abdominal red but usually lack dorsal black markings. The erect hair of mainland azteca is silvery rather than dirty white or brown and often the appressed silver hair is more extensive than in clemente. Females, for example, nearly always have much appressed silver hair on the frons and clypeus. The metapleural flange in clemente is moderately lamellate in all speci- mens studied (Fig. 2b). In typical azteca it varies from broadly lamellate to not lamellate (Figs. 2a-c). 8 Contributions in Science No. 123 Ammophila azteca occurs on at least one of the other Channel Islands. I have seen one male (LACM) from Santa Rosa Island and interestingly it has the typical mainland color pattern and the erect hair is silver. Literature Cited Menke, A. S. 1964. New species of North American Ammophila. (Hymenoptera, Sphe- cidae). Acta Hymenopterologica, 2(1) : 5-27. (for 1963) LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE ■ IfMBER 124 May 31, 1967 STUDIES ON CALIFORNIA ANTS. 3. THE TAXONOMIC STATUS OF PROCERAT1UM CALIFORNICUM COOK ( Hymenoptera : Formicidae) By Roy R. Snelling Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) The 1960 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8V2 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) Method of literature citation must conform to CONTRIBUTIONS style — see number 90 and later issues. Spell out in full the title of non-English serials and places of publication. (7) A factual summary is recommended for longer papers. (8) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. David K. Caldwell Editor STUDIES ON CALIFORNIA ANTS. 3. THE TAXONOMIC STATUS OF PROCERATIUM CALIFORN1CUM COOK (Hymenoptera: Formicidae) By Roy R. Snelling1 Abstract: Based upon the type male, Proceratium cali- fornicum Cook is redescribed. The presumed females are de- scribed for the first time and the relationships to the species of the Old World and New World are discussed. A key to the workers and females of described New World Proceratium is provided. In his book The Ants of California, Cook (1953) described several taxa as new. All of these, with the exception of Proceratium calif ornicum, have since been synonymized with common, well-known species (Wilson, 1955; Cole, 1967). The identity of the Proceratium presented difficulties which, while not yet solved, can now be somewhat clarified. Historical Resume As its name implies, this species was originally based on a specimen from California. Brown (1958) indicated caution in acceptance of this record, since there were no prior records of Proceratium in the United States from west of the Great Plains. The members of this genus are all cryptobiotic in their habitats and show a decided preference for areas which maintain high summer humidity. There are few areas in California which can satisfy this requirement; the type locality of P. calif ornicum (the Santa Cruz Mountains) is one such area. On this basis, at least, it was not too unreasonable to expect a species of Proceratium to occur here. The species was described from a single male specimen. Males of Proceratium are extremely rare in collections, and Cook’s inadequate and inaccurate original description indicated no characters by which his species could be separated from the males of the eastern forms. The figures given by Cook are poorly executed, but at least offered clues lacking in the verbal description. During recent years several alate females of a distinctive Proceratium have been taken in California, but it has not been possible to secure any workers or additional males. Nonetheless, for reasons which I hope to justify below, I believe these to be conspecific with P. californicum. Before going further into a discussion of these females it is pertinent to review the status of the male which Cook originally described. Section of Entomology, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 124 At the time Cook’s publication appeared two similar and closely allied genera were recognized in the eastern United States, Proceratium and Sys- phincta. For American workers, at least, the generic distinctions were obvious and clear-cut in the workers and females; M. R. Smith (1943) indicated equally sound characters for the males. In the latter caste, the main character used was wing venation. Not mentioned by Smith, but equally distinctive, is the clypeal configuration. In all castes of Sysphincta the clypeus possesses a distinct anterior median projection and the petiole is more or less nodiform (although this was recognized to be somewhat variable). A cursory examination of Cook’s figures show a distinctly projecting clypeal margin, a somewhat nodiform petiole and wing venation typically that of Sysphincta. Had any competent myrmecologist examined this specimen it would have been placed in that genus without hesitation. Cook was aware of Smith’s work on male ants, having referred to it several times and taken a number of the illustrations directly from his paper, so it seems strange that he could have missed such obvious differences as were used to separate the two genera. Nowhere in his discussion of Proceratium does Cook make any men- tion of Sysphincta ; neither did he attempt to compare his ant with any then placed in that genus. Following a critical examination of nearly all the described species of Proceratium and Sysphincta, Brown (1958) found that the supposed generic differences would not hold up, as all the characters show graduation from the Proceratium extreme to the Sysphincta extreme. Accordingly Brown synony- mized Sysphincta under Proceratium. Attempts to recognize P. calif ornicum from either Cook’s description or figures prove futile. The textual comments are inaccurate and misleading and the figures bear little resemblance to the type specimen. The type, now the property of the Snow Museum, Oakland, California, has been made available to me, and with this specimen at hand, it is now possible to unravel some of the confusion and present a more detailed account of the species. Although I find it distasteful to review and criticize the original description in such detail as follows, I feel that if this is not done that there will remain the possibility of further confusion in the future. Critique of Cook’s Original Description and a Redescription of PROCERATIUM CALIFORNICUM Cook gave the length of the type specimen as 3.5 mm. I have carefully measured this individual, and arrive at a length of 4.25 mm; the distance from the anterior ocellus to the thoracic-petiolar articulation is 2.0 mm, the petiole measures 0.4 mm, and the abdomen (with apical segments reflected down- ward) is 1.85 mm. The statement that the head length (HL) is about equal to head width (HW) is correct; the HL is 0.82 mm, while HW is 0.84 mm, so that HW slightly exceeds HL. In specimens of P. croceum (Roger) HL is 1967 California Ants 3 0.60 mm and HW is 0.65 mm; according to Cook, the head of P. calif ornicum is “broader than in croceum while the above figures show the reverse to be true. Cook’s next statement that the antennae are 12-segment is refuted by his figure of a 13-segmented antenna. Male ponerines typically possess 13-seg- mented antennae and there is no reason to suppose that P. californicum is an exception, particularly since the males of other Proceratium, so far as known, follow the rule. Unfortunately the type lacks the right funiculus and the entire left antenna. However, in the figure the funicular segments appear much too long (twice as long as broad) , while typically they are hardly longer than broad in males of this genus. Since the facial view included only the right scape, the left antenna of the lateral view may be an illustrative addition to make the specimen complete. In the figure of the head the scape is shown distinctly longer than is actually the case. The statement that the scape is “equal in length to the last three segments of the funicle” can be neither proven nor disproven, although the figure shows the scape to be slightly longer; the scape is 0.42 mm long. The remaining cephalic features mentioned are correctly described, i. e., the posterior margin of the head is rounded, three ocelli are present, the eyes are large and prominent, the well-developed mandibles are edentate with pointed apex. These characters are common to all Proceratium males. The eye length is 0.37 times HL and the distance from the lower eye margin to the mandibular insertion is about equal to one-third of the eye length (6:19); the upper eye margin is slightly below the midpoint of the HL. The thorax is said to be short and massive. The thoracic length is 1.5 mm, maximum height is 1.2 mm and maximum width is 0.90 mm. I would not consider the thorax to be “short and massive” since the length exceeds both its height and width. Cook’s statement that the pronotum has distinct humeral angles is baffling, for I cannot discern anything resembling humeral angles. The statement that the anteromedian part of the mesoscutum is distinctly truncated is also confusing. I assume that he had reference to the dorsal portion adjoining the promesonotal suture; this however is evenly convex. In spite of the claim that there is a large, rounded tubercule terminating centrally on the mesonotum, no such tubercle exists. Presumably this was in error for the metanotal tubercle, which is not large, and is rounded only in lateral aspect. From above it is pointed behind with a distinct median longitudinal carina. The remaining gross characters are more or less correctly described, though without offering any distinctive features. In discussing the integument, his remark that most of the body is subopaque is not correct. Everywhere, except on the frons and epinotum, the surface is distinctly shining between the sculpturation. The cephalic sculpture is said to be fine, but the punctures are about equal to those of the thorax, where they are stated to be heavy and well-defined. On the frons and middle portions of the vertex the punctures are a little finer and much denser than elsewhere on the head; here the surface texture is roughened, but still there are sufficient shining raised interspaces that the aspect, on the whole, is that of a somewhat shining surface. The 4 Contributions in Science No. 124 punctures on the lower part of the face are irregularly linearly arranged, generally convergent toward the apex. The cheeks are crossed by a few very fine transverse striolae. The mandibles are shining, with numerous punctures a little larger than those of the adjacent portions of the face. The apical margin of the median clypeal lobe is acutely produced, with a pair of posteriorly convergent fine carinulae on its dorsal face. The pronotum is reticulopunctate, with the surface distinctly shining and the punctures equal in size to those of the lower part of the frons. The meso- pleurae are shining, with distinct punctures which are much denser below, especially anteriorly and posteriorly. The mesoscutum is densely punctate with distinct shining interspaces. The mesoscutellum is more strongly convex in lateral aspect than Cook’s figure shows; the punctation is much like that of the mesoscutum. The epinotum is distinctly duller than the remainder of the thorax and is densely reticulopunctate. The figure of the petiolar node is inaccurate as it shows the anterior face more steeply sloping than is actually the case, and the node is too thick from front to back. The postpetiole is densely punctate, and the remaining gastric segments are a little shinier and more sparsely punctate. The pubescence, both appressed and erect, is everywhere yellowish, not dark reddish-brown. The integument is dark reddish-brown. The wings are hyaline, very slightly whitish, with very pale yellowish veins and amber stigma. The hind wing has nine hamuli. The Type Specimen The above commentary of conflicting statements was based upon the type specimen, the only known male of this species. Cook cited the data for this specimen as follows: “Glenwood in the Santa Cruz Mountains (T. W. Cook, 1950),” thus suggesting that he collected the specimen. In fact, the male bears the following data: “Glenwood Cal./ 27 May 1908.” Dr. W. L. Brown has suggested (in litt.) that this specimen was “probably a Bradley-collected specimen from the MCZ.” However, I have seen other insects with an identical label in the Stanford University collection, to which Cook had access. This male bears, in addition, two other penciled labels in Cook’s handwriting. These read: “PROCERATIUM/ sp. /DRAWN” and “Proceratium/ calif ornicum / T. W. Cook/ Described,” indicating beyond any doubt that this is the type of P. calif ornicum. Additional Material I have before me four female Proceratium from three widely separated California localities. One female is from Yuba City, Sutter Co., 27 April 1965, collected by W. Wiard while sweeping mixed Rumex and Avena on a ditch levee. The second specimen was taken at Valle Vista, Oakland, Alameda Co., 21 April 1918, collector not indicated but probably J. C. Bradley. Two females are from the Santa Monica Mts., Los Angeles Co., 19 April 1959, collected by 1967 California Ants 5 Figure 1. Proceratium calif ornicum Cook. Frontal aspect of head and lateral aspect of body, respectively, of male Holotype (above) and female (middle); forewing (below). 6 Contributions in Science No. 124 an entomology student from the University of California at Los Angeles; according to Dr. J. N. Belkin the specimens were most likely taken at Tapia Park in Malibu Canyon. This park is a favored site for school field trips. Unfortunately, recent changes in the Tapia Park area have completely de- molished the most likely sites. All these females are conspecific; 15 years ago they would have been placed in the former genus Sysphincta with no difficulty. The clypeus is angularly produced in the middle, the petiole is somewhat nodiform in profile (though less so than is usual), the gastric configuration is more nearly that of Sysphincta than Proceratium, and the wing venation is typical of Sysphincta. As pointed out above, Cook erred in assigning his species to Proceratium; it should have, at that time, been placed in Sysphincta, with which it agreed in all essential characters. It was this realization, especially, which led me to assume that the females were also conspecific with Cook’s species; both the male type and the alate females are typically Sysphincta in both habitus and structure and therefore readily separable from most of the Nearctic forms. In addition, there was the fact that no other members of this group had been previously discovered in California. Comparative Notes on PROCERATIUM Species Dr. Brown has very generously sent me specimens of a number of species of Proceratium for comparison with California material. The three eastern United States species, P. croceum (Roger), P. pergandei (Emery), and P. silaceum Roger, are represented in this material. From P. croceum and P. silaceum, the California females may be separated immediately by the produced clypeus and more nodiform scale. Both of the species lack any indication of an angular projection on the clypeal margin and the petiolar scale is fully erect and compressed from front to back so that it is much higher than long. The resemblance to P. pergandei is much closer, but fundamental differences are abundant. The paired carinae of the middle clypeal lobe of P. pergandei form a broad-based inverted “V” near the apex of the lobe which coalesce well below the level of the antennal sockets, the occipital margin is distinctly convex in full face view, the head, in full face view is not so markedly narrowed above the level of the eyes, the petiolar node is more depressed, the ventral petiolar projection is spine-like and directed caudad and the gastric configuration is quite distinct. In the females here associated with P. calif orni- cum the clypeal carinae form a much elongated inverted “V” and coalesce above the level of the lower margin of the antennal sockets, the head is mark- edly narrowed above the eyes, and the ventral petiolar process appears as a blunt, somewhat triangular lamella directed cephalad. When compared with the Central American species, P. micrommatum (Roger), there are no obvious close relationships. In P. micrommatum, total length is much less, the antennal scapes are shorter and the petiolar and gastric configurations are different. Three of the Neotropical species, P. convexiceps 1967 California Ants 7 (Borgmeier), P. mancum Mann and P. brasiliense Borgmeier, are unknown to me except from their descriptions, which indicate, however, they are also quite distinct from the California form. In short, P. calif ornicum differs significantly from all its North American and Neotropical congeners and does not show any noteworthy affinity with any of these species. In its general habitus it is obviously much more closely related to the Eurasian species, P. melinum (Roger), a member of the melinum group of Brown. Tins group includes also the Japanese species, P. itoi (Forel). Al- though the latter is not available, Dr. Brown sent me a female of P. melinum and comparison with that species is possible. When P. calif ornicum and P. melinum are placed side by side, the similarity is striking; in size, color, punc- tation and configuration these ants are obviously very close. In P. melinum the occipital margin is evenly convex whereas in P. californicum the border is distinctly concave in the middle; the antennal scape is noticeably shorter in the Eurasian species, in full face view barely attaining the level of the hind margin of the posterior ocelli; in P. californicum the scape extends slightly above the level of the posterior ocelli. The frons is densely punctate, appearing granulose, and dull in P. melinum; in the case of P. californicum the punctures are finer, less distinct and the surface is slightly shining. Thoracic punctation offers an excellent character for separating the two species. Although it is consistently coarser and denser everywhere in P. melinum, the disdnction is most marked on the mesoscutum. Here, in P. melinum, the punctures are very crowded, with the surface appearing subgranulose; the individual punc- tures are difficult to distinguish. The mesoscutal punctures of P. californicum, while abundant, are discretely separated by shining interstices. The second gastric segment of P. melinum is abundantly marked by distinct punctures which are only slightly finer than those of the mesoscutum, obviously much larger than the diameter of the hairs arising from them. The Nearctic species has a very finely punctate second gastric segment, the punctures only slightly larger in diameter than the hairs arising from them, much finer than the mesoscutal punctures. The petiolar process of P. melinum, in lateral view, has the surface between the anterior and posterior angles emarginate, so that two spines are formed, one directed obliquely cephalad and the other obliquely caudad. In P. californicum the process is not emarginate and the entire, some- what triangular process, is obliquely directly cephalad. The above comparative comments apply solely to the females. The rarity of males makes it difficult to relate the type specimen to the males of other species. The clypeal configuration, wing venation and shape of the petiolar node adequately separate P. californicum from P. croceum and P. silaceum. I have seen no males of P. per gander, that caste was described by M. R. Smith (1928), but not in sufficient detail for adequate comparative remarks. The following comments are therefore tentative. The length of the male described by Smith is given as 3.6 mm so that it may be seen that the western species is 8 Contributions in Science No. 124 about 0.5 mm larger. In P. pergandei the head, “when measured from side to side thru the eyes,” is said to be slightly broader than long; when measured in this manner, the head of the male of P. calif ornicum is about 1.3 times as broad as long, which I would be inclined to call distinctly broader than long. Smith’s statement that the pronotum is “somewhat concealed by the mesonotum” would seem to imply that the latter overhangs the former; this condition is not true of P. californicutn. In this species, the posterior face of the pronotum is vertical and is not overhung by the mesoscutum. The gaster of P. pergandei is described as smooth and shining; since Smith noted the presence of punctures elsewhere on the body and made no mention of gastric punctures, I assume he meant there were no evident punctures, although very tine, piligerous punctures must surely be present. The first gastric tergum of Cook’s species is conspicuously and closely punctate; the second segment is likewise conspicuously punctate, but the punctures are notably sparser and somewhat finer than on the first segment. Males of the European species, P. melinum, have not been available to me. Although this caste was described by Emery (1895), the description is not sufficiently detailed to be of much assistance here. The size of Emery’s male and the mention of abdominal punctures may corroborate the relationship to P. californicum suggested by the females of these species. The evidence examined above indicates that P. californicum, based on a male specimen, is distinct from previously described species of New World Proceratium, while there is some indication that it may be more nearly related to the European species, P. melinum. This accords with what is known of the presumed females of P. californicum which are clearly distinct from those of the other known species of the New World. It is on the basis of the relationships expressed above that I have associated these females with P. californicum. The following key has been prepared to facilitate the identification of females and workers of the described species of Proceratium occurring in the New World. The following three species are placed in the key on the basis of comments in the literature as they have not been available to me: P. brasiliense Borgmeier, P. convexiceps (Borgmeier) and P. mancum Mann. Key to New World PROCERATIUM, Workers and Females 1. Petiole erect, compressed from front to back; middle lobe of clypeus not produced forward as a triangular process; females with distinct thin, blade-like process on middle of metanotum and with strong longitudinal carina on apical one-half or more of scutellum 2 Petiole nodiform, anterior face convex or strongly inclined, not compressed (except in P. californicum ); clypeus produced medially as a narrow triangular lobe (except in P. convexiceps ); female without blade-like process on metanotum; longitudinal carina usually absent from scutel- lum, when present, very faint, limited to posterior one-fourth, or less, of the segment 4 1967 California Ants 9 2. Larger species, 3.75 to 4 mm; petiolar node, in profile, thick, blunt above, base little thicker than crest; frons with longitudinal carina extending forward between frontal lobes to clypeal base; sides of thorax coarsely rugose (eastern U.S.) croceum (Roger) Smaller species, 2.75 mm or less; petiolar node, in profile, slender, base distinctly thicker than crest; frons with longitudinal carina ending at midpoint, or less, of distance toward clypeal base; sides of thorax with a few irregular rugulae, but mostly smooth 3 3. Epinotal spines distinct; genal area punctate, distinctly shining; dorsum of thorax without pronounced transverse rugulae behind (C. Amer., s. Mex) mancum Mann Epinotum without distinct spines, but angulate laterally at juncture of basal and declivious faces; genal area strongly rugulose; dorsum of thorax with prominent transverse rugulae behind (eastern U.S .) ....silaceum (Roger) 4. Small species, 3.5 mm or less; head broadest above, slightly but definitely narrowed toward mandibular insertions; frontal carinae approximate, subparallel (except in P. brasiliense) (Central and South America) 5 Larger species, 3.75 mm or more; head little, if any, narrowed toward mandibular insertions, broadest below level of eyes; frontal carinae well- separated, convergent above (U.S.) 7 5. Clypeal margin not medially produced convexiceps (Borgmeier) Clypeal margin slightly to strongly produced in middle 6 6. Eyes very small; frontal carinae convergent anteriorly; ventral petiolar process bispinose brasiliense Borgmeier Eyes larger, well-developed; frontal carinae parallel: ventral petiolar process with a single spine micrommatum (Roger) 7. In full face view occipital margin convex, sides of head sub-parallel to top of head; clypeal carinae forming inverted broad-based “V” near apex; ventral petiolar process forming a narrow spine obliquely directed caudad; reflected dorsum of second gastric segment strongly projected to rear so that reflected tip of gaster appears to arise from mid-ventral surface pergandei (Emery) In full face view occipital margin with median concavity, head distinctly narrowed above; clypeal carinae forming elongated inverted “V” before coalescing between frontal lobes; reflected dorsum of second gastric segment not strongly projecting to rear, forming an even curve with reflected tip californicum Cook During the preparation of this paper I have been supplied with material from several institutions. Especially helpful has been Dr. W. L. Brown, Jr.. Cornell University, who sent specimens which would not have been otherwise available; Dr. Brown has also read and criticized the manuscript. The holo- type of P. californicum was made available through the courtesy of Dr. C. 10 Contributions in Science No. 124 Don MacNeill, Snow Museum, Oakland, California. Dr. Marius Wasbauer sent the Yuba City specimen from the collection of the Bureau of Entomology, California Department of Agriculture. To these gentlemen, my very sincere thanks. To my wife, Ruth Ann, special thanks for the care and diligence with which she executed the habitus illustrations. Literature Cited Brown, W. L. 1958. Contributions toward a reclassification of the Formicidae. II. Tribe Ectatommini. Bull. Mus. Comp. Zool., 118:175-362. Cole, A. C. 1967. A monographic revision of the genus Pogonomyrmex Mayr in North America. Univ. Tenn. Press, in press. Cook, T. W. 1953. The ants of California. Palo Alto: Pacific Books, xiii— |— 462 pp. Emery, C. 1895. Beitrage zur Kenntnis der Nordamerikanischen Ameisenfauna. Zoolo- gische Jahrbiicher fiir Systematik, 8:257-360. !Smith, M. R. 1928. An additional annotated list of the ants of Mississippi. Ent. News, 39:242-246. 1943. A generic and subgeneric synopsis of the male ants of the United States. Amer. Midi. Nat., 30:273-321. Wilson, E. O. 1955. A monographic revision of the ant genus Lasius. Bull. Mus. Comp. Zool., 113:1-202. UCL LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE jllBER 125 May 31, 1967 THE AMERICAN ATHERINID FISHES OF THE GENUS COLEOTROPIS i By Carter R. Gilbert and David K. Caldwell Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) The 1960 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8V2 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) Method of literature citation must conform to CONTRIBUTIONS style — see number 90 and later issues. Spell out in full the title of non-English serials and places of publication. (7) A factual summary is recommended for longer papers. (8) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. David K. Caldwell Editor THE AMERICAN ATHERINID FISHES OF THE GENUS COLEOTROPIS By Carter R. Gilbert1 and David K. Caldwell2 Abstract: The atherinid fishes of the genus Coleotropis comprise two species, C. starksi (Meek and Hildebrand) and C. blackburni Schultz, which are found in shallow inshore waters of the eastern Pacific and western Atlantic oceans, respectively. A third species, C. colecanos Caldwell, recently described from Caribbean Costa Rica, is here synonymized with C. blackburni. The genus was included in the subfamily Menidiinae by Schultz (1948), who indicated that its closest relatives are the eastern Pacific genera Eurystole and Nectarges. The present paper describes the genus Coleotropis and its included forms, lists the morphological differences by which the species may be separ- ated, and extends both the geographic ranges and ranges of morphometric variation of the two species. Introduction The genus Coleotropis was erected by Myers and Wade (1942: 136-138) to include a single species, C. starksi, which had been described by Meek and Hildebrand (1923: 267, pi. 20, fig. 2) from Panama Bay. Later, Schultz (1949: 108-109, fig. 15) and Caldwell (1962) described, respectively, C. blackburni and C. colecanos, cognate forms from the Caribbean. The last species was described from a single specimen and was said to differ from C. blackburni in having a more slender body and a longer and more slender caudal peduncle. Although the morphometric data given in the original de- scription of C. colecanos appeared convincing, the absence of any meristic features different from those found in C. blackburni, together with the scarcity of specimens, cast some doubt as to the validity of the species. During the summer of 1963, intensive field work by the senior author in the vicinity of Tortuguero, Costa Rica (the type locality of C. colecanos) , resulted in the collection of over a hundred specimens of Coleotropis. This large series has permitted a reassessment of the two Atlantic species, as well as a more detailed comparison of the Atlantic and Pacific forms. Materials and Acknowledgments The specimens examined during this study are from the following museum collections: Field Museum of Natural History (formerly Chicago Natural History Museum) (FMNH); Florida State Museum, University of Florida (UF); Los Angeles County Museum of Natural History (LACM); Museum of Comparative Zoology, Harvard University (MCZ); University of Assistant Curator in Ichthyology, Florida State Museum. 2Curator of Ichthyology, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 125 California, Los Angeles (UCLA); University of Puerto Rico, Institute of Marine Biology, Mayaguez (UPR); and the United States National Museum (USNM). We wish to thank the individuals in charge of these collections, other than our own, for making the specimens available. We also want to express our appreciation to Dr. Archie F. Carr, of the University of Florida, whose generous help and financial support has resulted in many valuable fish collections from the Tortuguero area; to Dr. C. Richard Robins, Institute of Marine Science, University of Miami, and Mr. Alejandro Ciardelli, Univer- sidad de Cartagena, Cartagena, Colombia (formerly of the Institute of Marine Science, University of Miami), for information regarding the hydrography of the Rio Atrato and Gulf of Uraba (Darien) ; to Mrs. Mildred Eaddy, Florida State Plant Board, who took the photographs; to Mr. Paul Laessle, University of Florida, for the illustrations; and to the authorities of the United States Fish and Wildlife Service Biological Laboratory, Brunswick, Georgia, and par- ticularly to Mr. Herbert Gordy, who is responsible for the radiographs used in this study. Methods All counts and measurements included in this paper were taken by the senior author. These were made, using the standard methods described by Hubbs and Lagler (1958: 19-26), on the holotype and 21 topotypes of Coleo- tropis colecanos (ranging from 41.5 to 92 mm standard length) from Tortu- guero, Costa Rica; on three paratypes and one non-type specimen of C. black- burni from Venezuela and Brazil, respectively; on six paratypes of C. starksi from Panama; and on 14 specimens of C. starksi from Costa Rica. Additional counts were made on a fourth paratype of C. blackburni, eleven other topotypes of C. colecanos, and ten other specimens of C. starksi from Costa Rica. Measurements are expressed in thousandths of standard length (shortened hereafter to SL) ; they were taken with precision dividers and were read to the nearest tenth of a millimeter. Scale counts for the circumferential and caudal- peduncle series are written to indicate the relative disposition of scales above and below the lateral line. Thus, a count expressed as 7-2-7=16 signifies seven scales above and seven scales below and between the lateral lines on each side of the body. The sum of these two counts, together with the lateral-line rows, represents the total count. Since the lateral line is incomplete and irregular, a lateral-scale count was obtained by counting the series from above the oper- cular opening to the caudal base. A frequency distribution expressed as (21) 23 to 25 (26) indicates that 90 percent of the counts fall between 23 and 25, with the extremes 21 and 26. The caudal fin-ray count is written so as to indicate the total number of caudal elements, as well as the number of main (long) caudal rays emanating from the superior and inferior hypurals. Thus, a count of 9+8 = 17 indicates nine dorsal and eight ventral main caudal rays. Vertebral counts are written so as to indicate the numbers of precaudal and caudal vertebrae, as well as the total number. 1967 Fishes of Genus Coleotropis 3 In this paper a complete description is presented under the account of the genus Coleotropis; no description appears in either of the species accounts. Characters distinguishing the two species are compared in the species diag- noses, and the rough breakdown of meristic counts also appears in these sections. The illustrations in Figures 2 through 5 were made from the speci- mens appearing in Figure 1. Genus Coleotropis Myers and Wade Menidia Bonaparte, 1836: 91 (type species, Atherina menidia Linnaeus, by absolute tautonomy). Coleotropis Myers and Wade, 1942: 136-138 (type species, Menidia starksi, by monotypy). Description: Compressed, rather elongate atherinid fishes with the air bladder not tapering to a point posteriorly and not extending into five or more haemal arches; air bladder and posterior end of body cavity falling well short of anal fin origin; anal fin elongate, with a single spine; a sheath of scales present at base of anal fin, this sheath consisting of either one or two rows of scales anteriorly; body scales cycloid, the posterior edge somewhat irregular; origin of anal fin anterior to origin of first dorsal fin, midway between caudal base and posterior margin of orbit; dorsal fin widely separated, the height of the spinous portion less than distance between the origins of the spinous and soft-rayed segments; first dorsal fin with III or IV flexible spines, the origin over base of fourth or fifth anal ray; caudal fin forked; pectoral fin high on body, falcate, as long as or slightly longer than head, its tip extending beyond ventral base; pelvic fins posterior in position, close together and inserted equidistant between upper angle of pectoral base and anal origin; pelvic fins completely joined by a membrane; several enlarged scales present between bases of pelvic fins, the largest sharply pointed and extending posteriorly more than halfway along lengths of fins; anus normal in position, situated a short distance in front of anal fin. Premaxillaries protractile, the dermal covering separated by a deep fold from skin on head; premaxillaries broadly dilated posteriorly, the anterior part not separated by a notch from the posterior part; gape of mouth strongly curved, restricted at corners by a membrane between the jaws; rami of mandibles scarcely elevated; teeth well developed in both jaws, sharply pointed, and slightly curved, in two rows, the first row in upper jaw enlarged; lower jaw slightly included at tip of mouth when mouth is tightly closed. Abdomen more or less strongly compressed; peritoneum brownish-black; lateral line present, but irregular and incomplete; silvery lateral band present, very well marked, and sharply delimited above and below, somewhat con- stricted at caudal peduncle, and bordered above with a dark line. Sides of head and body scaled; scales large, 38 to 50 in lateral series; predorsal scales 18 to 28; body circumferential scales 9-2-11 or 10-2-10 to 11-2-11=22 to 24; caudal-peduncle circumferential scales usually 5-2-5=12 4 Contributions in Science No. 125 or 7-2-7=16 (depending on species), with intermediate counts often present; anal rays 1,19 to 1,32; dorsal rays II-I to IV-I,7 to 10; pectoral rays 12 to 14, usually 13; pelvic rays 1,5; gill rakers 4 to 6+1 + 14 to 17 (range of total counts=20 to 23); vertebrae 14 or 15+25 to 27 (range of total counts=40 to 42) ; total caudal rays 34, the main rays 9+8 = 17. Relationships: Although the genotype of Coleotropis, C. starksi, was originally described in the genus Menidia, the relationships of the two genera actually are not particularly close. Myers and Wade (1942: 136-138), in the description of Coleotropis, made the following comments regarding the affini- ties of their new genus: “In Jordan and Hubbs’ (1919) key, Coleotropis starksi keys down to the Thyrina group ( Thyrina , Thyrinops, and Atherinella) . It differs from all of these in its much larger size, the more curved gape, the considerably larger mouth (the maxillary reaching to below the front part of the eye), and very sharply in the deep anal sheath. None of the Melaniris { — Thyrina, preoccupied) group appears to possess any anal sheath at all. It may be remarked that Melaniris brasiliensis possesses neither the pinched belly nor the posteriorly produced air-bladder of the other species of Melaniris, 3 and it is a much larger fish. In these three characters brasiliensis resembles starksi, and it is possible that there is a close relationship. M. brasili- ensis, however, has no anal sheath and possesses the small mouth of the other species of Melaniris. We therefore do not at this time disturb its generic assignment.” Schultz (1948) regarded the degree to which the air bladder extends into the haemal arches of the caudal vertebrae as particularly significant in atheri- nid classification. Using this criterion, he placed Coleotropis in the subfamily Menidiinae, the members of which have the air bladder not tapering to a point posteriorly and not extending beyond the fifth haemal arch. Inasmuch as Schultz made no direct statements regarding the inter-relationships of the various genera (the characters of the different groups appearing only in the form of a key), one must assume that he intended the proximity to each other of the couplets in the key to be a direct indication of phylogenetic relation- ships. According to this, Coleotropis is most closely related to the eastern Pacific genera Eurystole and Nectarges, and somewhat more distantly allied to Adenops, Membras, and Hubbesia. All of these taxons are characterized by having the posterior tip of the air bladder falling well short of the anal-fin origin. Ecology: All specimens of Coleotropis black burni from Tortuguero, Costa Rica, were collected in the surge zone area, in water from one to 36 inches deep. None was ever taken in brackish or fresh water, and thus they were never found sympatrically with Thyrinops chagresi (Meek and Hilde- brand), the other atherinid fish in the Tortuguero area. 3According to Schultz (1948), the genus Melaniris does not have a pronounced posterior extension of the air bladder. He also erected a new genus, Xenomelaniris, for the sole reception of “M.” brasiliensis. 1967 Fishes of Genus Coleotropis 5 Data for the most recent collections of C. starksi (from Costa Rica) indicate that this species occurs in ecological situations very similar or identical to those in which C. blackburni is found. Range: (See under species accounts). Coleotropis starksi (Meek and Hildebrand) Figures 1A, 2 A, 3 A, 4A, 5 A Menidia starksi Meek and Hildebrand, 1923: 267-268, pi. XX, fig. 2 (original description; type locality, Taboga Island, Panama; holotype, 109 mm SL [incorrectly listed in original description as 235 mm SL], USNM 79732). Coleotropis starksi, Myers and Wade, 1942: 136-138 (new generic name). Specimens Examined (numbers in parentheses refer to number of speci- mens examined and size range in mm, respectively) : FMNH 8308 (2 para- types of Menidia starksi, 74-93.5), USNM 79733 (3 paratypes, 89-111), USNM 81747 (1 paratype, 111), all from Taboga Island, Panama; UCLA W 53-283 (26, 15-95), Isla de San Jose, Ensenada Playa Grande, Islas Perlas, Panama Bay, Panama; UCLA, W 54-168 (1, 43), Bahia Ballena, Golfo de Figure 1. A: Coleotropis starksi (Meek and Hildebrand), LACM 6893-1, 90.5 mm SL, from Playa del Coco, near Sardinal, Costa Rica. B: Coleotropis blackburni Schultz, UF 1 1205, 90.5 mm SL, from Tortuguero, Costa Rica. 6 Contributions in Science No. 125 Figure 2. Comparison of mouth size in the two species of Coleotropis. A: C. starksi. B: C. blackburni. Nicoya, Costa Rica; UCLA, W 54-172 (34, 31-99), UCLA, W 54-177 (24, 44-59), Isla Tortuga, Golfo de Nicoya, Costa Rica; LACM 6894-1 (2, 19-55), l/i mile south of Playa del Coco, near Sardinal, Costa Rica; LACM 6893-1 (22, 53-102), Playa del Coco, near Sardinal, Costa Rica. Diagnosis: Characters mentioned in the generic description are not re- peated here, except where greater clarification is required. Additional meristic data appear in Tables 1 through 5. A species of Coleotropis that differs from its Atlantic cognate, C. black- burni, in having fewer predorsal scales ( 18 or 19 vs. 23 to 28) ; fewer lateral scales (37 to 39 vs. 43 to 50); fewer caudal-peduncle scales (usually 12 vs. 1967 Fishes of Genus Coleotropis 7 usually 16); fewer scales in circumferential series above and between lateral lines (9 vs. 10 or 1 1) ; a higher average anal fin-ray count (25 to 33 vs. 20 to 26); a larger mouth, the maxillary extending posteriorly to beneath anterior part of orbit (instead of extending just anterior to orbit) (Fig. 2A); an anal sheath with two scale rows anteriorly (instead of one) (Fig. 3A); a longer pectoral fin, which is slightly longer than head and extends to above tip of pelvic fin (instead of being shorter than head and extending about halfway along length of pelvic fin); the upper margin of silvery lateral stripe on third (rather than fourth) scale row below mid-dorsal scale row (Fig. 4A); pig- mentation less extensive on lower jaw, covering less than half (instead of all) of jaw (Fig. 5 A) ; teeth in both jaws larger and more widely spaced; a probable greater maximum body length (largest specimen [of 1 15] examined, 111 mm SL vs. 92 mm SL [124 specimens]) . Lateral scales 37 or 38 (39) (a count of 40 listed in original description) ; dorsal-fin rays (II) III (IV), (7) 8 (9); total anal elements (25) 26 to 29 (30 and 33); circumferential scales (9-2-9 or 8-2-11=20 or 21) 9-2-10 or 9-2-11=21 or 22 (9-2-12=23); caudal-peduncle scales 5-2-5=12; vertebrae (14+25=39) 15+25=40 (16+25=41). Range: Recorded from the Gulf of Panama to northern Costa Rica. Further collecting undoubtedly will extend the range of this species. Figure 3. Comparison of anal sheaths in the two species of Coleotropis. A: C. starksi. B: C. blackburni. Coleotropis blackburni Schultz Figures IB, 2B, 3B, 4B, 5B Coleotropis blackburni Schultz, 1949: 108-109, fig. 15 (original description; 8 Contributions in Science No. 125 Figure 4. Comparison of scale size and position of lateral stripe on upper part of back in the two species of Coleotropis. A: C. starksi. B: C. blackburni. compared with C. starksi; type locality, Jacuque Point, Gulf of Venezuela, Venezuela; holotype, 82 mm SL, USNM 123205). Coleotropis sp., Caldwell, Ogren, and Giovannoli, 1959: 19-22, fig. 2 (one specimen from Tortuguero, Costa Rica; compared with C. blackburni ). Coleotropis colecanos Caldwell, 1962: 1-8, fig. 1-2 (original description; differs from C. blackburni in certain body proportions; type locality, Tortuguero, Costa Rica; holotype, 76 mm SL, UF 5652). Specimens Examined (numbers in parentheses refer to number of speci- mens examined and size range in mm, respectively) : USNM 123207 (3 para- types of C. blackburni, 45-64), MCZ 37293 (1 paratype of C. blackburni , 48), all from Point Macolla, Gulf of Venezuela, Venezuela; UPR 2490 (3, 35-65), Manzanilla, Isla Margarita, Venezuela; USNM 100830 (1, 72), Porto Inhauma, Brazil; UF 5652 (holotype of C. colecanos, 76), beach near Tor- tuguero, Limon Province, Costa Rica; UF 11205 (84 topotypes of C. cole- canos, 19-90.5), LACM 8335 (10 topotypes of C. colecanos, 37-61.5), shore in front of turtle camp, ca. 2 mi. S of mouth of Tortuguero lagoon, Limon 1967 9 Fishes of Genus Coleotropis a b Figure 5. Comparison of pigmentation on underside of head in the two species of Coleotropis. A: C. starksi. B: C. blackburni. Province, Costa Rica; UF 11254 (1 topotype of C. colecanos, 68.5), mouth of Tortuguero lagoon, Limon Province, Costa Rica; UF 11261 (23 topotypes of C. colecanos, 31-65), seaward side of sand spit, near mouth of Tortuguero lagoon, Limon Province, Costa Rica. Diagnosis: Characters mentioned in the generic description are not re- peated here, except where greater clarification is required. Additional meristic data appear in Tables 1 through 5. A species of Coleotropis that differs from its Pacific cognate, C. starksi, in having more predorsal scales (23 to 28 vs. 18 or 19); more lateral scales (43 to 50 vs. 37 to 39); more caudal-peduncle scales (usually 16 vs. usually 12); more scales in circumferential series above and between lateral lines (10 or 11 vs. 9); a lower average anal fin-ray count (20 to 26 vs. 25 to 33); a smaller mouth, the maxillary extending to just in front of orbit (instead of to beneath anterior part of orbit) (Fig. 2B); anal sheath with one scale row anteriorly (instead of two) (Fig. 3 B ) ; a shorter pectoral fin, which is shorter than head and extends about halfway along length of pelvic fin (instead of being slightly longer than head and extending to above tip of pelvic fin); the upper margin of silvery lateral stripe on fourth (rather than third) scale row below mid-dorsal scale row (Fig. 4B); pigmentation more extensive on lower jaw, covering nearly all (instead of less than half) of jaw (Fig. 5B); teeth in both jaws smaller and more narrowly spaced; a probable smaller maximum body length (largest specimen [of 124] examined, 92 mm SL vs. Ill mm SL [115 specimens]) . Lateral scales (43) 44 to 49 (50); dorsal-fin rays III (IV), (7) 8 (9 or 10); total anal elements (20) 21 to 24 (25 or 26); circumferential scales 10 Contributions in Science No. 125 (10-2-10 or 11-2-9=22) 11-2-10 or 11-2-11=23 or 24 (11-2-12=25); cau- dal-peduncle scales (6-2-6 or 7-2-6=14 or 15) 7-2-7 = 16; vertebrae (14+26= 40, 15+25=40, or 14+27=41) 15+26=41 (15+27=42). Range: Known from the Caribbean coasts of Costa Rica and Venezuela eastward to the Atlantic coast of central Brazil. Additional collecting probably will extend the range of this species. Status of COLEOTROPIS COLECANOS: In the preceding text only one species of Coleotropis (C. blackburni ) from the western Atlantic-Carib- bean area has been recognized, despite the fact that a second form (C. cole- canos) was recently described by Caldwell (1962) from the Caribbean coast of Costa Rica. The reasons for this are discussed herein. Coleotropis colecanos was described as differing from C. blackburni in having a longer and more shallow caudal peduncle and a lesser body depth. No meristic differences were noted. Measurements and counts were made (by Gilbert) on four paratypes of C. blackburni from Venezuela, a non-type specimen from Brazil, and 25 topotypes of C. colecanos. In addition, the holotype of C. colecanos was re- counted and remeasured. These data are summarized in Tables 1 through 5. The new value obtained for the caudal peduncle depth of the holotype of C. colecanos agrees closely with the original (88 versus 84). The new value for the caudal-peduncle length is somewhat less than that given in the original description (205 versus 236). This undoubtedly is attributable to a slight difference in measuring technique. Nevertheless, this new value is greater than for all but seven (out of 38) of the topotypes measured, and is also higher than any of the values obtained for the four specimens from Venezuela and Brazil (Table 5). As can be seen from Table 5, the overall average for caudal- peduncle length is slightly higher for the topotypes of C. colecanos than for the paratypes and non-type specimen of C. blackburni. However, the fact that values obtained for C. colecanos completely encompass those for C. black- burni indicates that these differences are not meaningful. Since there appar- ently are no striking morphological differences by which C. colecanos and C. blackburni can be distinguished, we conclude that C. colecanos should henceforth be regarded as a synonym of C. blackburni. Although the populations of Coleotropis blackburni from the extreme western Caribbean and from the southern Caribbean-western Atlantic appar- ently are indistinguishable, it should be noted that differences do exist between other elements of the inshore fish faunas occurring in these two areas. For example, in the southern and western Caribbean the sciaenid fish Umbrina broussonnetii is found only from western Colombia northward along the Central American coast, whereas the closely related U. coroides is confined almost exclusively to the South American coast. In the northern Caribbean these two species occur sympatrically (Gilbert, 1966). Another sciaenid, Ophioscion costaricensis, which is known from two specimens from the Tortuguero area, was said to differ from the closely related O. brasiliensis 1967 11 Fishes of Genus Coleotropis (from the southern Caribbean and southwestern Atlantic) in having a signifi- cantly smaller eye and wider interorbital space (Caldwell, 1958). Examina- tion (by Gilbert) of many additional specimens of O. brasiliensis confirms these differences. The above faunal break seems to center around the Gulf of Uraba (Darien), which marks the boundary between South and Central America. Recent exploration by the R/V Pills bury (of the Institute of Marine Science, University of Miami) in this area has resulted in ecological information pertinent to this problem. The following are preliminary observations, and thus may be subject to some modification when the data are analyzed in more detail (C. R. Robins, pers. comm.). The Gulf of Uraba is a long, relatively narrow arm of the sea, which appears to be of nearly uniform depth throughout (ca. 20 fathoms). The bottom is uniformly flat except near shore, where the sides slope very sharply upward. The large quantities of silt that are carried in by the Rio Atrato are mostly deposited along the eastern shore of the Gulf or along the adjacent Caribbean coast of South America. This easterly flow is thought to be due to the Corollis effect and/or perhaps an eastward flowing oceanic counter- current. The resultant ecological conditions favor such fishes as sciaenids and clupeids, most species of which characteristically inhabit shallow, turbid, inshore water where a silt bottom is present. In contrast, the western shore of the Gulf of Uraba, as well as much of the Caribbean coast of Panama, is largely silt-free, is characterized by extensive areas of small coral formations, and has a fish fauna that is basically an impoverished insular reef type. Such an area is poorly suited for most sciaenids and clupeids, and the species of these families collected to the east were not encountered here. Farther west one again finds ecological conditions similar to those on the coast of northern South America, and it is here that sciaenids and clupeids again appear. Thus, the distributional break noted for the sciaenid genera Ophioscion and Um- brina appears real, and might logically be expected in other groups of fishes. Table 1 Predorsal scale counts in the two species of Coleotropis 12 Contributions in Science No. 125 Vo VO 1—1 oo m (N cn •S -O «C> J5 ^ ^ j* O 'o q -2 -2 La «C> •~C) O O O O V3 W o £ % o 00 * t S-H o ■> o o *55 -C5 -S 4* p c o Oh o <3 o S3 > £ H * * ■55- s * * G G G * GOO *Non-topotypes of C. colecanos *Topotypes of C. colecanos Comparison of proportional measurements (expressed in thousandths of standard length) in the species of Coleotropis 14 Contributions in Science No. 125 ^.2 o Q O -S -A* o a CM Of- ^ On co co p CM nv ON o t"- r- © CO CO On ' CO ^7 On ^ 1 CO V *n "T On 7T cm’ cm «n . as (J <5 Cj J3 r-- O co cm • i co O O O CM cm w CO O co 7f cM cm o O i CM CO oo r- ^ CO /-S vo t- T1 cm’ CM »n rt *-h VO ^ r— i T1 oo n- o o ^ o CM CM >s . ~i*C ° a . T m O /-v CO NO VO t-h On CM o p r^ On . vd in in co *— l O ^ 00 In ’T vd oo VO in CO Tf r-H T1 vd Tf CM *5$ in Tf »n rv o o r- o in /-v t"- in Tf /-N l> CM co in •n VO 1 VD n V On T1 vd CM r- CO ON 1 OO »n vo CM vo OO CM CO vd «n OO 00 C, in ’—i NO 1-H t-H r-H M- w r-H W T"H r-H CM 1 § n .1 0X3 Oh C n TD O PQ Oh O JD 73 fl O T3 n V © m ON 6\ m 00 m 4 < ON (N n in w Tt m m w 00 in Tt W z c O co ro VO IVs Os ^ q OO vo /— S 00 vd «> C: r— 1 /— s in ^-s O cn VO ^ VO On ON ^ ■^t q r- 0 /—v Tt 00 00 q 4* in T d 4 On T1 ri ON 00 vo 00 O 00 m m o- m vd VO vo m «n in co 0 m T— 1 1-1 m co in w ro w in w r-l W co w VO IT on ^ 4 5 °° H 0 ^ ^ / s ^t ^ «n /-n in ON in On 0- OO in vo r- in 00 t''- ^ 4 CN O ^d «n r- IT 4 0 00 "T 4 <>s O ON 00 j ^ ^ J « 5 ; c H 1967 New Fossil Dog 3 Figure 2. Brachyrhynchocyon sesnoni, new species, LACM 17039. Jaws with canine- M3; occlusal and labial views. (To be viewed with a stereoscope. Ca. XI.) 4 Contributions in Science No. 126 Table 1 Measurements in millimeters of types of Brachyrhynchocyon sesnoni and B. intermedius Character B. sesnoni B. intermedius (after Loomis, 1931) Right Left l3 a-p 4.4 tr. 3.0 P1 a-p 3.3 tr. 2.8 P2 a-p 7.1 tr. 3.05 2.85 P3 a-p 9.6 7.1 tr. 4.3 P4 a-p (to parastyle) 13.8 13.9 tr. 8.3 M1 a-p ca. 10.00 tr. ca. 15.00 M2 a-p 5.6 ca. 5.9 tr. ca. 9.4 ca. 8.8 pi-M2 50.65 P2 a-p ca. 7.5 tr. 2.45 P3 a-p 8.4 8.7 tr. 3.8 3.45 P4 a-p 9.9 10.1 tr. 4.8 4.35 Mi a-p 15.2 15.6 tr. 6.8 ca. 6.8 M3 a-p 5.0 tr. 3.8 Px-M3 ca. 55.6 53.0 1967 New Fossil Dog 5 principal cusp above anterior end of posterior root; small well-defined pos- terior accessory cusp and posterior cingular cusp. P4 with small well-developed deuterocone extending antero-lingually, projecting slightly anterad of para- style; parastyle broadly rounded; protocone and metastyle worn, separated by open carnassial notch. M1 heavily worn; antero-labial corner less rounded, more angular than indicated by Loomis (1931: fig. 2). M2 worn; paracone ane metacone subequal. I4_2 missing. I3 root only; laterally compressed. Canine broken; large. P4 root only; separated from canine and P2 by diastema. P2 laterally compressed; low silhouette; apex of principal cusp worn away; small anterior and posterior accessory cusps; small posterior cingular cusp. P3 similar to P2; more robust; greatest transverse diameter between posterior cusps. P4 slightly larger than P3; greatest transverse diameter between princi- pal cusp and posterior accessory cusp. M4 heavily worn; paraconid and proto- conid widely separated; metaconid small, extending slightly posterad of protoconid; talonid heavily worn, showing presence of strong hypoconid and moderately developed entoconid. M2 worn to essentially featureless plain. M3 greatly reduced; paraconid, if present, broken away; protoconid and meta- conid form transverse crest; talonid slightly basined, opening labially. Sym- physis long; extending to posterior end of P2. Large mental foramen below center of P2, small one below anterior edge of posterior root of P3. Discussion: The lower P2_4 of B. sesnoni have anterior cingular cusps which are not found in B. intermedius (Loomis, 1931) but are found in an un- described Wounded Knee-Sharps species; there is more lateral compression of these teeth than in the older and younger species. The M3 has not been found in recorded specimens of the Brule species and is not as reduced as in the Sharps species. The canines in B. sesnoni do not dominate the symphysial region as they do in the Sharps form. The palate tapers from P4 to P1 while Loomis indicates an abrupt narrowing in B. intermedius at the P4 with the anterior premolars in parallel rows. The distal end of the humerus is somewhat catlike in aspect. It is flattened and laterally expanded with a wide internal condyle and a small entepicondylar foramen. This species is named for Mr. William T. Sesnon, Jr., Chairman of the Board of Governors, Los Angeles County Museum of Natural History, and patron of Vertebrate Paleontology. Literature Cited Loomis, F. B. 1931. A new Oligocene dog. Amer. J. Sci., 22(4) : 100-102. j c TlY^s LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE Timber 127 May 31, 1967 A NEW SPECIES OF LATE OLIGOCENE DOG, SUNKAHETANKA SHEFFLER1, FROM SOUTH DAKOTA By J. R. Macdonald Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) The 1960 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8V2 x 1 1 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) Method of literature citation must conform to CONTRIBUTIONS style — see number 90 and later issues. Spell out in full the title of non-English serials and places of publication. (7) A factual summary is recommended for longer papers. (8) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. David K. Caldwell Editor A NEW SPECIES OF LATE OLIGOCENE DOG, SUNKAHETANKA SHEFFLERI . FROM SOUTH DAKOTA By J. R. Macdonald1 Abstract: Simkahetanka sheffleri is described from the Poleslide member of the Brule formation in the White River Badlands of South Dakota. This species is older than the genotypic species which is found in the Gering formation ( sensu C. B. Schultz, et al.) of Nebraska and in the Sharps formation of South Dakota. It is less specialized than the later forms but does not give any greater indication of the origin of the genus. The Vertebrate Paleontology Section of the Los Angeles County Museum of Natural History spent part of the summer of 1966 continuing its prospect- ing in the late Whitneyan and early Arikareean on the south side of the White River in Shannon County, South Dakota. The purpose of this program was to expand the Museum’s reference collection and to look for ancestors of the forms which the writer has been collecting from the Arikareean Sharps forma- tion. The specimen described below was found by Mr. Douglas J. Macdonald. The 1966 field season was supported by Mrs. E. Hadley Stuart, member of the Museum Board of Governors. Sunkahetanka sheffleri, new species Type: LACM 17076, a partial skull with canine-M2 and P1 - M2. Paratype: LACM no. 17476, a cranium with P2 - M2. Locality and Horizon: LACM no. 2006, Medicine Root Creek, Shannon County, South Dakota. Poleslide member of Brule formation, from twenty five feet of gray clays whose top is fifty nine feet below the base of the Sharps formation. These clays are overlain by Protoceras channels and underlain by a red clay. Late Whitneyan. Diagnosis: Small, and teeth less massive than the genotypic species, Sunkahetanka geringensis (Barbour and Schultz); M1-2 unreduced; P2_3 with strong anterior cingular cusps. Description: Canine broken. P1 broken; small; single rooted; crowded between canine and P2. P2 set diagonally in tooth row; overlapping P1 and P3; no anterior cingular cusp; no posterior accessory cusp; posterior cingulum without cusp. P3 set diagonally in tooth row; no anterior cingular cusp; prin- cipal cusp worn; small posterior accessory cusp. P4 with broadly rounded parastyle; deuterocone prominent, lingual, but projecting slightly anterad of parastyle; protocone larger than metastyle; protocone-metastyle crest worn; closed carnassial notch below wear level. M1 large; elongated transversely; paracone and metacone subequal; protoconule and metaconule worn; proto- senior Curator of Vertebrate Paleontology, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 127 Figure 1. Sunkahetanka sheffleri, new species, type, LACM 17076. Cranium with canine-M2; lateral and palatal views. (To be viewed with a stereoscope. Ca. X-V2.) 1967 New Fossil Dog 3 cone large, worn; hypocone large, worn. M2 unreduced; paracone much larger than metacone, worn. Rostrum short; anterior edge of orbit above P4 meta- style; sagittal crest not as strongly developed as in S. geringensis. Pi small; single cusp; single root. P2 with tall principal cusp; large anterior cingular cusp; posterior cingulum. P3 similar to P2; with posterior accessory cusp. P4 similar to P3. with short worn paraconid; closed carnassial notch; tall worn protoconid; small metaconid, projecting slightly posterad of protoconid; large hypoconids, worn virtually flat; small entoconid; thin enamel rim closes talinid basin on lingual side. M2 broken; paraconid, if present, broken away; protoconid and metaconid worn flat, appear similar to Mx. M3 missing. All teeth massive. The back of the paratype cranium is well preserved. The bullae are large, proportionally comparable to those of S. geringensis; the sagittal and lambdoidal crests and the occiput are smaller and less massive than those of the younger species. Figure 2. Sunkahetanka sheffleri, new species, type, LACM 17076. Left mandible with P^M,,; occlusal and labial views. (To be viewed with a stereoscope. Ca. X-V2.) Discussion: Sunkahetanka geringensis (Barbour and Schultz) (1935) is based on a skeleton collected from “fifteen feet above the Brule, 400 feet west of the road in Redington Gap, . . . west of Bridgeport, Morrill County, Nebraska.” The beds are referred to the Gering formation by Barbour and Schultz although they are below the Gering formation as defined by Darton. Since 1953, eleven specimens (skulls, cranii, and mandibles) referable to this species have been found in the Sharps formation in the Wounded Knee area of Shannon County, South Dakota (Macdonald, 1963). The Whitneyan specimens were found 310 feet stratigraphically below and 14 miles geographically from the nearest record of S. geringensis. This occurrence is a unique example of a species to species evolution in a restricted 4 Contributions in Science No. 127 Table 1 Measurements in millimeters of types of Sunkahetanka. Character Sunkahetanka sheffleri Sunkahetanka geringensis LACM LACM NSM* ** LACM SDSM 17076 17476 Type Paratype 4-28-8-31 15910 5667 Right Left Right Left Type* Right Left P2 a-p 7.95 7.5 10.0 9.1 9.0 tr. 4.5 3.4 5.2 5.0 5.1 P3 a_P 9.7 8.4 10.5 10.8 11.1 tr. 5.5 5.0 6.2 6.0 6.0 P4 a-p (to parastyle) 15.1 13.1 16.0 17.2 17.0 tr. 10.0 8.8 10.0 11.5 11.6 M1 a-p 11.4 11.5 11.5 9.9 11.1 tr. ca. 18.3 17.0 15.5 16.7 M2 a-p 6.2 5.3 4.2 4.1 tr. 11.0 8.5 6.7 6.5 pi-M2 52.8 53.8 52.8 pi-4 38.8 40.0 42.5 40.9 M1-2 (parastyle- metastyle) 18.2 17.0 16.1 Pi a-p 3.0 4.1 tr. 3.2 3.5 P2 a-p 7.2 6.5 8.0 tr. 4.2 4.35 4.7 P3 a‘P 8.55 8.8 9.0 8.0 tr. 5.0 5.05 5.5 4.8 P4a-p 10.5 10.2 11.3 10.1 tr. 58.85 6.5 6.5 6.5 Mj a-p 17.1 18.0 17.6 tr. 7.8 8.2 7.6 * After Barbour & Schultze (1935:412) Nebraska State Museum. **South Dakota School of Mines Museum of Geology. 1967 New Fossil Dog 5 area, the younger species replacing the older one in place during a relatively short span of geologic time, perhaps on the order of 2 to 3,000,000 years. S. geringensis is slightly larger; the teeth are more massive; the M1 has become moderately reduced; the M2 greatly reduced; the jaw has become more massive and the sagittal crest, occiput, and lambdoidal crest much larger. These changes suggest a strengthening of the muscles for biting, chewing and manipulating the head. The general trend seems to be paralleling that seen in the Tomarctus - Osteoborus - Borophagus lineage of the Barstovian - Claren- donian - Hemphillian. This species is named for Mr. William J. Sheffler, a member of the Board of Governors, Los Angeles County Museum of Natural History, and patron of Vertebrate Paleontology. Literature Cited Barbour, E. H. and C. B. Schultz 1935. A new Miocene dog, Mesocyon geringensis, sp. nov. Bull. Nebraska State Mus., 1(6): 407-418. Macdonald, J. R. 1963. The Miocene faunas from the Wounded Knee area of western South Dakota. Bull. Amer. Mus. Nat. Hist., 125(3): 139-238. Ul fe'frSA 'f ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE Immber 128 May 31, 1967 J THE MARINE FISH FAUNA, BASED PRIMARILY ON OTOLITHS, OF A LOWER PLEISTOCENE DEPOSIT AT SAN PEDRO, CALIFORNIA (LACMIP 332, SAN PEDRO SAND) By John E. Fitch ;l Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) The 1960 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 814 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) Method of literature citation must conform to CONTRIBUTIONS style — see number 90 and later issues. Spell out in full the title of non-English serials and places of publication. (7) A factual summary is recommended for longer papers. (8) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. David K. Caldwell Editor THE MARINE FISH FAUNA, BASED PRIMARILY ON OTOLITHS, OF A LOWER PLEISTOCENE DEPOSIT AT SAN PEDRO, CALIFORNIA (LACMIP 332, SAN PEDRO SAND) By John E. Fitch1 Abstract: Several hundred pounds of fossiliferous “dirt” from a San Pedro sand deposit, yielded 2,746 otoliths, 584 vertebrae, 324 teeth, and 105 miscellaneous other fish remains representing at least 10 species of elasmobranchs and 30 kinds of teleosts. Since many of these fishes typically inhabit waters to the north of San Pedro, the deposit is believed to have been laid down during a period when ocean temperatures were colder than they are today at the same lati- tude. This deposit, one of the few remaining surface outcrops of San Pedro sand, will be destroyed by freeway construction prior to 1970. The Miraflores Street deposit at San Pedro, California (LACMIP 332), was first brought to the attention of the Los Angeles County Museum of Natural History in 1963 by Roger D. Reimer, a student of paleontology with an especially keen interest in the Pliocene and Pleistocene of southern Cali- fornia. Museum personnel, under the supervision of George P. Kanakoff, then Curator of Invertebrate Paleontology, sampled the site extensively during ensuing months, but it was Reimer who found the first otoliths and called them to my attention. Permission to sample the site was graciously granted by Jack Bell, owner of the property, and approximately 300 pounds of fossiliferous dirt were removed during two visits. This deposit is exposed at the base of a north-facing cliff, possibly 8 to 10 feet above present-day sea level, in the 600 block, Miraflores Street, San Pedro, California. The layer I sampled is composed primarily of friable molluscan remains in a fine, clean sandy matrix; it measures approximately 50 feet in a horizontal (east-west) direction by 18 to 24 inches vertically. Unfortunately, the Miraflores Street deposit is doomed by the impending southerly extension of the Harbor Freeway which is scheduled for completion prior to 1970. Until this occurs, it affords one of the last opportunities for the paleontologist to sample the San Pedro sand, and to learn of the various faunas represented by the contained assemblages. When I processed my 300-pound field sample, as explained in a previous publication (Fitch, 1966b), I found the identifiable remains of 10 species of elasmobranchs and 30 teleosts. In addition to 2,746 otoliths, there were 584 vertebrae (9 from elasmobranchs), 324 teeth (99 from elasmobranchs), and 105 miscellaneous items: bone fragments, fin spines, dermal denticles, etc. (Table 1). Research Associate, Los Angeles County Museum of Natural History; Marine Biologist, California Department of Fish and Game, Terminal Island, California 90731. 1 2 Contributions in Science No. 128 Table 1. Fish Remains Found in a San Pedro Pleistocene Deposit (Miraflores Street) Type and number of remains Scientific name Common name otoliths teeth vertebrae other ELASMOBRANCHS Carcharhinus sp. requiem shark 1 Galeorhinus zyopterus soupfin shark 9 I sums oxyrinchus mako 1 Myliobatis calif ornicus bat stingray 32 Notorynchus maculatus sevengill shark 6 Prionace glauca blue shark 3 Raja spp. skates 12 6* Squalus acanthias spiny dogfish 19 Squcitina californica Pacific angel shark 10 8** Triakis semifasciata leopard shark 6 unidentified elasmobranchs 9 TELEOSTS Atherinops affinis topsmelt 6 Brosmophycis marginata red brotula 1 Chitonotus pugetensis roughback sculpin 1 Citharichthys sordidus Pacific sanddab 22 Citharichthys stigmaeus speckled sanddab 191 Citharichthys spp. sanddabs 795 Clupea pallasi Pacific herring 29 Coryphopterus nicholsi bluespot goby 20 cottids sculpins 25 Cymatogaster aggregate shiner perch 1,129 Engraulis mordax northern anchovy 80 Enophrys taurina bull sculpin 1 Eopsetta jordani petrale sole 20 Genyonemus lineatus white croaker 2 Glyptocephalus zachirus rex sole 17 Icelinus tenuis spotfin sculpin 2 Icichthys lockingtoni medusafish 1 Leptocottus armatus staghorn sculpin 9 Lycodopsis pacifica blackbelly eelpout 12 Ly op sett a exilis slender sole 5 Microgadus proximus Pacific tomcod 1 Oxyjulis californica senorita 2 Porichthys notatus plainfin midshipman 148 Radulinus asprellus slim sculpin 3 Rhacochilus vacca pile perch 7 Scorpaenichthys marmoratus cabezon 1 Sebastodes carnatus gopher rockfish 1 Sebastodes goodei chilipepper 30 Sebastodes spp. rockfish 145 Seriphus politus queenfish 15 Spirinchus starksi night smelt 18 Stenobrachius leucopsarus northern lampfish 1 Trachurus symmetricus jack mackerel 3 unidentified teleosts 3 225 575 91 1 *“wing” spines ** dermal denticles fpharyngeals, fin spines, opercula, etc. 1967 Fossil Fish Otoliths 3 Woodring, Bramlette and Kew (1946), using molluscan data, postulated that the San Pedro sand represented a shallow-water habitat in some localities, but other deposits contained moderate-depth mollusks not found elsewhere. These localities were inferred to represent depths of 25 to 50 fathoms. Among the mollusks they identified from the San Pedro sand were six locally extinct northern species, but there were no locally extinct southern species. In addition, these deposits contain both northern and southern species that are now at or close to the limits of their range at the latitude of San Pedro. Based upon these and similar data, most geologists and paleontologists are prone to interpret the San Pedro sand as representing a portion of the Lower Pleistocene when local ocean temperatures were considerably colder than they are today. Deposits of the San Pedro sand generally lie unconformably below the Palos Verdes sand, and in terms of events that took place during the time interval represented by the unconformity, there was “deformation; almost complete submergence or probably complete submergence of the area now constituting the Palos Verdes Hills; and intermittent emergence during which the series of marine terraces were formed and the marine deposits now found on most of them were laid down, the Palos Verdes sand constituting the marine deposits on the lowest and most extensive terrace on the landward side of the hills.” (Woodring, Bramlette, and Kew, 1946). Systematic Account Hexanchidae — cow sharks Notorynchus maculatus Ayres — sevengill shark Although sevengill sharks range from northern British Columbia to San Carlos Point, Baja California, they seldom are seen or captured south of Point Conception. They reportedly attain lengths of 15 feet, but 7- to 9-foot speci- mens are quite large for California. A 5-foot 7-inch male caught at Santa Rosa Island in 1961 weighed 41 pounds. In southern California waters, individuals that size and smaller usually are caught at or near the bottom in depths exceeding 400 feet. Teeth of this species have been found in several Pliocene and Pleistocene deposits in southern California and in an Indian midden at Ventura, California (Fitch, 1964, In press, and unpublished data; Kanakoff, 1956). Material : 6 teeth. Isuridae — mako sharks Isurus oxyrinchus Rafinesque — mako This world-ranging species is fairly common in our coastal waters between about San Francisco and Magdalena Bay, Baja California. Applegate (1966) reported an 11 -footer from California that apparently represented a record 4 Contributions in Science No. 128 size for the species, but the usual mako in our waters is shorter than 8 feet. Mako remains, primarily teeth, have been found in numerous Pliocene and Pleistocene deposits in southern California (Fitch, 1964; and KanakofT, 1956, as I. glaueus; Fitch, unpublished data) and in several coastal Indian middens (Fitch, In press; Follett, 1932, 1963a, 1963b). Material : 1 tooth. 13 15 1967 Fossil Fish Otoliths 5 Carcharhinidae — requiem sharks Carcharhinus sp. — requiem shark, species undetermined At least four species of sharks belonging to the genus Carcharhinus have been captured off southern California at one time or another during the past four or five decades, but none of these tropical sharks occurs in any abundance north of about Magdalena Bay, Baja California. Although these sharks are rare visitors to our waters, all are common residents in tropical regions, especially around islands and offshore banks. A few members of this genus are pelagic forms, and some of these have worldwide distributions. Some species of requiem sharks may attain lengths of 15 feet, but others apparently never exceed 5. Because of confusion among taxonomists regarding speciation, it is difficult if not impossible to identify teeth beyond the generic level. Carcharhinus remains, mostly teeth, have been found in several Pliocene and Pleistocene deposits in southern California (Fitch, 1964, 1966b, and unpub- lished data; Fitch and Reimer, 1967; Kanakoff, 1956). Material : 1 tooth. Galeorhinus zyopterus Jordan and Gilbert — soupfin shark The soupfin shark ranges from northern British Columbia to about Magdalena Bay, but is not abundant at the more southerly latitudes. Females occur principally south of Point Conception, where they often are caught in depths as shallow as 100 feet. A 6 Vi -foot female may weigh as much as 100 pounds, but 50- to 70-pounders are the usual sizes caught. Soupfin shark remains, mostly teeth, have been found in a number of Pliocene and Pleisto- cene deposits in southern California (Fitch, 1964, and unpublished data; Fitch and Reimer, 1967; Kanakoff, 1956, as Triakis semifasciata ) and in several coastal Indian middens (Fitch, In press; Follett, 1963b). Material : 9 teeth. Figure 1 . Inner face, left sagitta of Clupea pallasi 3.7 mm long. Figure 2. Inner face, left sagitta of Atherinops affinis 2.5 mm long. Figure 3. Inner face, left sagitta of Coryphopterus nicholsi 3.8 mm long. Figure 4. Inner face, left sagitta of Engraulis mordax 4.4 mm long. Figure 5. Inner face, right sagitta of Oxyjulis calif ornicus 1 .7 mm long. Figure 6. Inner face, right sagitta of Brosmophycis marginata 3.7 mm long. Figure 7. Inner face, right sagitta of Lycodopsis paci fieri 3.9 mm long. Figure 8. Inner face, right sagitta of Spirinchus starksi 2.7 mm long. Figure 9. Inner face, left sagitta of Stenobrachius leucopsarus 1.7 mm long. Figure 10. Inner face, right sagitta of Enophrys taurina 4.3 mm long. Figure 1 1 . Inner face, left sagitta of Chitonotus pugetensis 2.8 mm long. Figure 12. Inner face, right sagitta (badly worn, rostrum missing) of Icichthys lock- ingtoni 3.0 mm long. Figure 13. Inner face, right sagitta of Rcidulinus asprellus 3.6 mm long. Figure 14. Inner face, righfisagitta (rostrum broken) of Scorpaenichthys marmoratus 3.7 mm long. Figure 15. Inner face, right^sagitta of Icelinus tenuis 3.5 mm long. Photographs by Jack W . Schott. 6 Contributions in Science No. 128 Prionace glauca (Linnaeus)- — blue shark The blue shark is a pelagic species occurring in tropical, subtropical, and warm temperate areas of all world seas. On our coast, individuals sometimes stray into shallow water just outside the surf zone, but most are seen swimming slowly at the surface considerable distances offshore. Small blue sharks are extremely abundant off the California coast during summer months, and almost invariably these small sharks are males, indicating a geographical 1967 Fossil Fish Otoliths 7 distribution by sex. Although blue sharks are reported to reach lengths of 20 feet, the largest verified record was 12 feet 7 inches (Bigelow and Schroeder, 1948). Twelve-footers have been caught off our coast, but the usual blue shark in our waters is shorter than 6 feet and weighs less than 50 pounds. Blue shark remains are not abundant in fossil deposits, being known from only one other southern California locality (Lomita Marl Pliocene; Fitch, unpublished data). Follett (1963b) reported blue shark remains from a coastal Indian midden in Los Angeles County. Material : 3 teeth. Triakidae — smoothhounds Triakis semifasciata Girard — leopard shark The leopard shark has been caught between Oregon and Mazatlan, Mexico. It usually frequents shallow areas where the bottom is sandy, but also abounds in shallow rocky areas around southern California’s offshore islands. The females grow larger than males and may reach lengths of 7 feet, but a 5-footer can be considered large. Leopard shark remains have been found in several Pliocene and Pleistocene deposits (Fitch, unpublished data; Fitch and Reimer, 1967) and in a few coastal Indian middens (Fitch, In press; Follett, 1963a, 1963b, 1964). Material : 6 teeth. Squalidae — dogfish sharks Squalus acanthias Linnaeus — spiny dogfish The spiny dogfish abounds in the north Pacific Ocean (eastern and western), ranging south on our coast to Sebastian Viscaino Bay, Baja Cali- fornia. Females attain larger sizes than males, and are reported to reach 5 feet, but a 4-footer can be considered large. Trawling off southern California yields best spiny dogfish catches in 100 to 250 feet of water, but they have been Figure 16. Inner face, right sagitta of Seriphus politus 7. 1 mm long. Figure 17. Inner face, right sagitta of Genyonemus lineatus 3.8 mm long. Figure 18. Inner face, left sagitta of Citharichthys sordidus 5.2 mm long. Figure 19. Inner face, right sagitta of Sebastodes carnatus 1 1.6 mm long. Figure 20. Inner face, right sagitta of Rhacochilus vacca 7.8 mm long. Figure 21. Inner face, left sagitta of Citharichthys stigmaeus 3.2 mm long. Figure 22. Inner face, right sagitta of Sebastodes goodei 12.1 mm long. Figure 23. Inner face, right sagitta of Cymatogaster aggregata 4.7 mm long. Figure 24. Inner face, right sagitta of Glyptocephalus zachirus 4.7 mm long. Figure 25. Inner face, right sagitta of Leptocottus armatus 6.6 mm long. Figure 26. Inner face, right sagitta of Porichthys notatus 5.4 mm long. Figure 27. Inner face, left sagitta of Eopsetta jordani 4.7 mm long. Figure 28. Inner face, left sagitta (rostrum missing) of Trachurus symmetricus 6.8 mm long. Figure 29. Inner face, left sagitta of Microgadus proximus 6.8 mm long. Figure 30. Inner face, left sagitta of Lyopsetta exilis 3.2 mm long. Photographs by Jack W . Schott. 8 Contributions in Science No. 128 caught both shallower and deeper. Remains of Squalus acanthias have been found in numerous Pliocene and Pleistocene deposits in southern California (Fitch, unpublished data; Fitch and Reimer, 1967) and in an Indian midden at Ventura, California (Fitch, In press). Material : 19 teeth. Squatinidae-— angel sharks Squatina calif ornica Ayres- — Pacific angel shark Although the Pacific angel shark ranges from southern Alaska into the Gulf of California, it apparently is not abundant north of Point Conception nor south of Magdalena Bay. Skin divers often observe them buried in sand or mud at depths of 8 to 150 feet or more, but greatest concentrations are noted in 50 to 70 feet. They are reported to attain lengths of about 5 feet, but the largest I have examined, a 44-inch female, weighed 31 pounds. Pacific angel shark remains have been found in numerous Pliocene and Pleistocene deposits in southern California (Fitch, 1964, and unpublished data) and in several coastal Indian middens (Fitch, In press; Follett, 1932, 1963a, 1963b). Material: 10 teeth and 8 dermal denticles. Rajidae — skates Raja spp. — skates, species undetermined Although six species of skates (at least) occur off California, I was unable to find foolproof characters for distinguishing the teeth of any one of these from those of the other five. By size alone, teeth of large R. binoculata can be distinguished if they are present, but no really large skate teeth were found in the Miraflores Street deposit. At various times, skates can be captured at all depths from the shallow subtidal to those exceeding several thousand feet. Skate remains have been found in numerous Pliocene and Pleistocene deposits in southern California (Fitch, 1964, and unpublished data) and in an Indian midden at Ventura, California (Fitch, In press). Material: 12 teeth and 6 “wing spines.” Myliobatidae — eagle rays Myliobatis calif ornicus Gill — bat stingray Bat stingrays range from Oregon to Magdalena Bay, occurring in shallow bays, along the mainland coast, and around offshore islands. A record speci- men weighed 209 pounds, but individuals exceeding 50 pounds are rare. Remains of M. calif ornicus, primarily teeth, have been found in many Pliocene and Pleistocene deposits in southern California (Fitch, 1964, and unpublished data; Fitch and Reimer, 1967; Kanakoff, 1956), and in several coastal Indian middens (Fitch, In press; Follett, 1963b; Harrington, 1928). Material: 32 teeth. 1967 Fossil Fish Otoliths 9 Unidentified elasmobranchs No effort was made to assign family or generic names to the nine small elasmobranch vertebrae recovered from this deposit. It is presumed that they came from some of the same species as the teeth that were found. Clupeidae — herrings Clupea pallasi Valenciennes — Pacific herring The Pacific herring ranges throughout the north Pacific Ocean (western and eastern), but because of pollution and bay development, is no longer important on our coast south of Point Conception, although it may still occur in small numbers as far south as Ensenada, Baja California. They are said to attain lengths of 18 inches, but a 12- to 14-inch individual can be considered large; a fish of this size will weigh about a pound. Pacific herring otoliths have been found in several Pliocene and Pleistocene deposits in southern California (Fitch, unpublished data). The sagittae of a large, adult Pacific herring will exceed 5.0 mm in length. Material : 29 otoliths 2.1 to 4.8 mm long (Fig. 1). Engraulidae — anchovies Engraulis mordax Girard — northern anchovy The northern anchovy is a schooling fish that ranges from British Columbia to Magdalena Bay and offshore for more than 100 miles. They are reported to attain lengths of 9 inches, but a 7-inch individual can be considered unusually large. Otoliths of E. mordax have been found in many Pliocene and Pleistocene deposits in southern California (Fitch, 1964, and unpublished data; Fitch and Reimer, 1967), and in Indian middens at Ventura and Corona del Mar (Fitch, In press). The sagittae of a large, adult northern anchovy will exceed 4.5 mm in length. Material : 80 otoliths 1.9 to 4.7 mm long (Fig. 4). Osmeridae — smelts Spirinchus starksi (Fisk) — night smelt The night smelt ranges from Shelikof Bay, southeastern Alaska to Point Arguello, California. It is reported to attain a length of 9 inches, but indi- viduals exceeding 7 inches are rarely encountered. McAllister (1963) in revising the smelt family (Osmeridae) mentions that only one genus ( Mallotus ) has a fossil record. I have found S. starksi remains only at one other site, a marine Pliocene deposit at Santa Barbara, California, but osmerid otoliths (apparently Spirinchus sp.) are abundant in a freshwater Pliocene deposit (Tulare formation) in Kettleman Hills, California (Fitch, unpublished data). The sagittae of a large, adult night smelt will exceed 4.5 mm in length. Material : 18 otoliths 2.1 to 3.3 mm long (Fig. 8). 10 Contributions in Science No. 128 Myctophidae— lanternfishes Stenobrachius leucopsarus (Eigenmann and Eigenmann) — northern lampfish S. leucopsarus is one of the most abundant bathypelagic fishes in the eastern north Pacific, where it ranges from the Bering Sea to about Cedros Island, Baja California. A large individual might be 5 inches long and weigh about one-half ounce. Otoliths of this species have been found in many Pliocene and Pleistocene deposits in southern California, more than 1,000 having been recovered from a San Pedro, California, (Lomita Marl) deposit (Fitch, 1966b, and unpublished data). The sagittae of a large, adult northern lampfish will exceed 1.8 mm in length. Material : 1 otolith 1.7 mm long (Fig. 9). Gadidae — cods Microgadus proximus (Girard) — Pacific tomcod The Pacific tomcod ranges from Alaska to about Morro Bay, California, usually at depths of 200 feet or more, but sometimes in shallow water just outside the surf zone. They are reported to reach lengths of 12 inches, but no weights are available for such a fish; a 1014 -inch female weighed just under 6 ounces. M. proximus otoliths are rare in southern California Pliocene and Pleistocene deposits, but are abundant in the Pleistocene of Oregon (Fitch, unpublished data). The sagittae of a large, adult Pacific tomcod will exceed 14.0 mm in length. Material : 1 otolith 6.8 mm long (Fig. 29). Bothidae — lefteyed flounders Citharichthys sordidus (Girard) — Pacific sanddab Pacific sanddabs are reported to range from southern Alaska to about Magdalena Bay, but the occurrence of the species in central and southern Baja California needs to be verified. The maximum length and weight attributed to C. sordidus (16 inches and 2 pounds) also are questionable. A 12-inch female, the largest of several thousand Pacific sanddabs recently examined, weighed less than 10 ounces. Citharichthys otoliths are abundant in southern California Pliocene and Pleistocene deposits, and many of these are from C. sordidus (Fitch, 1964, and unpublished data; Fitch and Reimer, 1967). The sagittae of a large, adult Pacific sanddab will exceed 8.0 mm in length. Material ; 22 otoliths 2.5 to 5.9 mm long (Fig. 18). Citharichthys stigmaeus Jordan and Gilbert— speckled sanddab Speckled sanddabs range along the coast from southeastern Alaska to Sebastian Viscaino Bay, Baja California, usually in depths shallower than 200 feet, and often just outside the surf zone. A large individual, about 5 inches long, would weigh less than an ounce. C. stigmaeus otoliths have been found in many southern California Pliocene and Pleistocene deposits (Fitch, 1964, 1967 Fossil Fish Otoliths 11 and unpublished data; Fitch and Reimer, 1967), and in an Indian midden at Ventura (Fitch, in press). The sagittae of a large, adult speckled sanddab will exceed 3.5 mm in length. Material : 191 otoliths 1.5 to 3.4 mm long (Fig. 21). Citharichthys spp. — sanddabs, species undetermined The otoliths of all three species of Citharichthys known to California are easily distinguished if they are in good condition, but most fossil sanddab otoliths are either worn, fragmented, or partially digested, making specific identification impossible. Some of the unidentified sanddab otoliths in this deposit could have been from C. xanthostigma (the third California species), but most of them probably were from Pacific and speckled sanddabs. Material : 795 otoliths identifiable to genus only. Pleuronectidae — righteyed flounders Glyptocephalus zachirus Lockington — rex sole The rex sole ranges from the Bering Sea to Ensenada, Baja California (at least) in depths from shallow water to 2,100 feet. They are reported to attain lengths of 22 inches, but no weights are available for fish that size; a 15-inch female weighed just under 1 Vi pounds. Rex sole otoliths are abundant in some southern California Pliocene and Pleistocene deposits, more than 700 having been recovered from the Lomita Marl (Pliocene) at San Pedro (Fitch, unpub- lished data). The sagittae of a large, adult rex sole will exceed 7.0 mm in length. Material : 17 otoliths 1.5 to 4.7 mm long (Fig. 24). Lyopsetta exilis (Jordan and Gilbert) — slender sole The slender sole ranges from southeastern Alaska to Cedros Island, usually in depths of 400 to 800 feet, but sometimes as shallow as 120 feet or as deep as 1,700. A large individual might exceed 12 inches in length, but probably would not weigh more than 14 pound. Otoliths of L. exilis have been found in numerous southern California Pliocene and Pleistocene deposits (Fitch, 1964; Fitch and Reimer, 1967), more than 2,000 having been recov- ered from the Lomita Marl at San Pedro (Fitch, unpublished data). The sagittae of a large, adult slender sole will exceed 5.5 mm in length. Material : 5 otoliths 2.6 to 4.2 mm long (Fig. 30). Eopsetta jordani (Lockington) — petrale sole The petrale sole ranges from northwestern Alaska to south of Ensenada, at least, migrating annually between deep and shallow water. They are reported to attain a length of 2514 inches and a weight of 8 pounds or more, but a 5-pounder is considered quite large by most commercial fishermen. The otoliths of E. jordani are difficult to distinguish from those of two other eastern Pacific flatfishes, Parophrys vetulus and Lepidopsetta bilineata, unless they 12 Contributions in Science No. 128 are in perfect or near perfect condition. Since several of the “Eopsetta” oto- liths from this deposit were in relatively poor condition, a few may not be properly assigned. The sagittae of a large, adult petrale sole will exceed 10.0 mm in length. Material : 20 otoliths 2.0 to 5.8 mm long (Fig. 27). Atherinidae — silversides Atherinops affinis (Ayres) — topsmelt Topsmelt range from the Straits of Juan de Fuca to and into the Gulf of California. Various subspecies inhabit bays, kelp beds, and offshore island areas where they live at or near the surface. A 141^ -inch female weighing slightly less than 1 2 ounces appears to be a record size. Topsmelt otoliths have been found in many southern California Pliocene and Pleistocene deposits (Fitch, 1964, and unpublished data), and in an Indian midden at Ventura (Fitch, In press). The sagittae of a large, adult topsmelt will exceed 5.0 mm in length. Material : 6 otoliths 1.7 to 4.0 mm long (Fig. 2). Carangidae — jacks Trachurus symmetricus (Ayres) — Pacific jackmackerel The Pacific jackmackerel, a schooling fish, is perhaps one of the three or four most abundant species off our coast, ranging from British Columbia to Cape San Lucas and offshore for several hundred miles. The commercial catch is comprised primarily of fish shorter than 15 inches, but jackmackerel are known to attain lengths of 30 inches and weights of 5 pounds or more. T. symmetricus otoliths have been found in several southern California Plio- cene and Pleistocene deposits (Fitch, 1966b, and unpublished data), and in an Indian midden at La Jolla (Fitch, unpublished data). The sagittae of a large, adult Pacific jackmackerel will exceed 10.0 mm in length. Material : 3 otoliths longer than 6.0 mm (Fig. 28). Centrolophidae- — medusaefishes Icichthys lockingtoni Jordan and Gilbert — medusafish The medusafish ranges throughout the offshore area between British Columbia and central Baja California. Very young individuals (to perhaps 3 or 4 inches) usually accompany jellyfishes, apparently enjoying some type of commensal relationship. Large individuals (10 to 12 inches long) do not seem to associate with medusae, solitary specimens occasionally being cap- tured in mid-water trawls and other nets. I have not encountered I. lockingtoni remains in any other fossil deposit. The sagittae of a large, adult medusafish will exceed 10.0 mm in length. Material : 1 otolith (Fig. 12) in poor condition, longer than 3.0 mm. 1967 Fossil Fish Otoliths 13 Sciaenidae — croakers Genyonemus lineatus (Ayres) — white croaker The white croaker ranges from Vancouver Island to Magdalena Bay, abounding in almost every type of habitat between the intertidal and depths of 600 feet. A near-record 141/2 -inch fish weighed 1.4 pounds. White croaker sagittae frequently are the most abundant fish remains in southern California Pliocene and Pleistocene deposits, comprising 6,409 of the more than 11,000 otoliths recovered from a San Diego Pliocene locality (San Diego Formation, unpublished data) . They are also a common constituent of many coastal Indian middens: Fitch (In press) reports 7,655 G. lineatus otoliths from an Indian midden at Ventura. In view of their abundance in other deposits, their scarcity in this deposit (Miraflores Street) is difficult to explain. The sagittae of a large, adult white croaker will exceed 12.5 mm in length. Material : 2 otoliths (Fig. 17) in poor condition, longer than 3.8 mm. Seriphus politus Ayres — queenfish Queenfish range from Yaquina Bay, Oregon, to San Juanico Bay, Baja California, living in much the same habitat as the white croaker. A near-record 12-inch female weighed just over 10 ounces. Queenfish otoliths have been found in many southern California Pliocene and Pleistocene deposits (Fitch, 1964, and unpublished data; Fitch and Reimer, 1967; Kanakoff, 1956), and in several coastal Indian middens (Fitch, In press). The sagittae of a large, adult queenfish will exceed 10.0 mm in length. Material : 15 otoliths 2.7 to 7.1 mm long (Fig. 16). Embiotocidae — surfperches Cymatogaster aggregata Gibbons — shiner perch The shiner perch ranges from Port Wrangel, Alaska, to Santo Tomas Point, Baja California, being restricted to the mainland coast, primarily in depths shallower than 50 feet, but occasionally trawled in 400 feet. A record- sized pregnant female 7 inches long, weighing just under 3 ounces, contained 16 young which were almost 2 inches long each. Although C. aggregata oto- liths have been found in many southern California Pliocene and Pleistocene deposits (Fitch, 1964, and unpublished data; Fitch and Reimer, 1967), they usually comprise less than 10 percent of the total sagittae recovered. In this deposit, they made up almost 50 percent of the more than 2,700 otoliths found. The sagittae of a large, adult shiner perch will exceed 6.5 mm in length. Material: 1,129 otoliths 2.0 to 6.4 mm long (Fig. 23) . Rhacochilus vacca (Girard) — pile perch The pile perch ranges from Port Wrangel, Alaska, to San Martin Island, Baja California, primarily in shallow water around kelp beds, pier and dock pilings, breakwaters, and similar habitat. A near-record 17-inch female weighed just under 4 pounds. Pile perch remains, mostly otoliths and pharyn- 14 Contributions in Science No. 128 geal teeth, have been found in several southern California Pliocene and Pleistocene deposits (Fitch, unpublished data), and in an Indian midden at Ventura (Fitch, In press). The sagittae of a large, adult pile perch will exceed 9.5 mm in length. Material : 7 otoliths 6.6 to 9.9 mm long (Fig. 20). Labridae — wrasses Oxyjulis calif ornica ( Gunther )— senorita The senorita ranges from Natural Bridges State Park to Cedros Island, inhabiting shallow areas of rocky substrate and similar relief where kelp beds and other vegetation grows. A near-record 9-inch male speared at Dana Point in January 1965 weighed 4 ounces. Senorita otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, unpub- lished data). The sagittae of a large, adult senorita will exceed 3.5 mm in length. Material : 2 otoliths 1.7 to 2.2 mm long (Fig. 5). Scorpaenidae — rockfishes Sebastodes carnatus (Jordan and Gilbert) — gopher rockfish The gopher rockfish ranges from Eureka, California, to San Roque, Baja California, living almost exclusively in areas of rocky substrate from the shallow subtidal to depths of 180 feet. The species is reported to attain a length of 15 inches, but the largest I have seen was a 12-incher that weighed an ounce over 1 pound. The otoliths of S. carnatus are not known from any other fossil deposit. The sagittae of a large, adult gopher rockfish will exceed 11.5 mm in length. Material : 1 otolith 11.6 mm long (Fig. 19). Sebastodes goodei Eigenmann and Eigenmann — chilipepper The chilipepper ranges from Vancouver Island to Magdalena Bay, usually at depths greater than 600 feet. A 201/^ -inch female (IVi inches short of the reported maximum) caught at Cortez Bank in December 1964 weighed just under 4 pounds. S. goodei otoliths have been found in several southern Cali- fornia Pliocene and Pleistocene deposits (Fitch, unpublished data). The sagittae of a large, adult chilipepper will exceed 17.0 mm in length. Material : 30 otoliths 6.0 to 12.1 mm long (Fig. 22). Sebastodes spp. — rockfishes, species undetermined The otoliths of most of the 52 members of the genus Sebastodes inhabit- ing the waters of California, can be distinguished one from the other if they are from adult fish, and if they are not worn or broken. Such characters as length and shape of rostrum, configuration of posterior end, presence or absence of marginal frills, angle of posterior taper, depth of sulcus, and number of growth zones (annuli) for otolith length are helpful for identify- 1967 Fossil Fish Otoliths 15 ing sagittae of the various species or species-complexes. Unfortunately, very few fossil otoliths are sufficiently well preserved to be identified to species. Material : 145 otoliths, probably representing four or five species, at least. Cottidae — sculpins Chitonotus pugetensis (Steindachner) — roughback sculpin The roughback sculpin ranges from northern British Columbia to Mag- dalena Bay, usually at depths of 100 to 350 feet, but sometimes shallower or deeper. A large individual might be 7 inches long and weigh about 2 ounces. C. pugetensis otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, 1964, and unpublished data). The sagittae of a large, adult roughback sculpin will exceed 6.0 mm in length. Material : 1 otolith 2.8 mm long (Fig. 11). Enophrys taurina Gilbert — bull sculpin The bull sculpin ranges from Monterey Bay to Anacapa Island, usually at depths of 100 to 250 feet, but sometimes as shallow as 30 feet or as deep as 800. A 6-inch specimen (perhaps near maximum size) will weigh about 2 ounces. E. taurina otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, unpublished data). The sagittae of an adult bull sculpin will exceed 5.5 mm in length. Material : 1 otolith 4.3 mm long (Fig. 10). Icelinus tenuis Gilbert — spotfin sculpin The spotfin sculpin ranges from Queen Charlotte to Guadalupe Island, Baja California, usually in depths of 100 to 400 feet, but some have been taken as shallow as 50 feet and as deep as 1,200. These small, slender fishes (4 inches might be about maximum length) probably never attain a weight of one ounce, even when full of eggs. Their otoliths are extremely difficult to distinguish from those of the yellowchin sculpin, I. quadriseriatus, but a critical comparison indicates /. tenuis sagittae are somewhat more pointed posteriorly, are more slender (height into length), and have a slightly less- pronounced rostrum. Since all of these characters are relative, it is necessary to compare each fossil otolith with an assortment of similar sized sagittae from living fishes before specific names can be assigned. Even with such critical evaluation, 100 percent accuracy is not assured. Otoliths of both these species of Icelinus are abundant in several southern California Pliocene and Pleisto- cene deposits (Fitch, unpublished data). The sagittae of a large, adult spotfin sculpin will exceed 4.5 mm in length. Material : 2 otoliths 3.0 to 3.5 mm long (Fig. 15). Leptocottus armatus Girard — Pacific staghorn sculpin The Pacific staghorn sculpin ranges from northwestern Alaska to San Quintin Bay, Baja California, being particularly common in shallow outer 16 Contributions in Science No. 128 coast waters, and in bays and lagoons. It is reported to reach a length of 12 inches, but the largest I have seen was a 10-inch female netted in 1958 that weighed 8 ounces. L. armatus otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, unpublished data), and in an Indian midden at Ventura (Fitch, In press). The sagittae of a large, adult staghorn sculpin will exceed 9.0 mm in length. Material : 9 otoliths 3.1 to 6.6 mm long (Fig. 25). Radulinus asprellus Gilbert — slim sculpin The slim sculpin ranges from Kodiak Island, Alaska, to about Ensenada, Baja California, usually at depths exceeding 150 feet. A 5!^ -inch fish (large for the species) will weigh about 1 ounce. Otoliths of R. asprellus have been found in a number of southern California Pliocene and Pleistocene deposits, over 1,300 having been recovered from a Pliocene site (Lomita Marl) at San Pedro (Fitch, unpublished data). The sagittae. of a large, adult slim sculpin will exceed 4.0 mm in length. Material : 3 otoliths 3.2 to 3.6 mm long (Fig. 13). Scorpaenichthys marmoratus (Ayres) — cabezon The cabezon ranges from northern British Columbia to Abreojos, Baja California, primarily in areas where the bottom is rocky, but often over sandy or sandy-mud substrate also. The cabezon is reported to reach a length of 39 inches (Miller, Gotshall, and Nitsos, 1965) and a weight of 20 to 25 pounds (Roedel, 1953), but individuals exceeding 25 inches and 10 pounds are rare. Cabezon otoliths have not been found in any other fossil deposit. The sagittae of a large, adult cabezon will exceed 6.5 mm in length. Material: 1 otolith (Fig. 14), in poor condition, longer than 3.7 mm. Unidentified sculpins A number of otoliths and otolith fragments were identifiable as sculpins (family Cottidae), but were not in good enough condition to determine genus or species. Material : 25 badly worn otoliths. Gobiidae-— gobies Coryphopterus nicholsi (Bean) — bluespot goby The bluespot goby ranges from British Columbia to San Martin Island. Although the adults are bottom dwellers in the shallow subtidal and to depths of about 200 feet, larvae and juveniles sometimes are taken in plankton nets 100 miles or more offshore (Ebert and Turner, 1962). C. nicholsi seldom exceeds a length of 5 inches or a weight of 1 ounce. Their otoliths have been found in many southern California Pliocene and Pleistocene deposits, over 1,700 having been recovered from a Pliocene site (Lomita Marl) at San Pedro 1967 Fossil Fish Otoliths 17 (Fitch, unpublished data). The sagittae of a large, adult bluespot goby will exceed 3.0 mm in length. Material : 20 otoliths 1.8 to 3.8 mm long (Fig. 3). Batrachoididae — toadfishes Porichthys notatus Girard — plainfin midshipman The plainfin midshipman ranges from southeastern Alaska to Cedros Island, being one of the half-dozen most abundant species at depths of 300 to 750 feet. During spawning and “nesting” they often move into intertidal areas, but during other periods they may travel into depths of 1,200 feet or more. P. notatus is reported to attain a length of 15 inches, but the largest I have seen was a 13 Vi -inch male weighing just over 14 ounces. Plainfin mid- shipman sagittae often are one of the most abundant otoliths in southern California Pliocene and Pleistocene deposits, nearly 700 having been recov- ered from a Pliocene site (Lomita Marl) at San Pedro (Fitch, 1964, and unpublished data; Fitch and Reimer, 1967). The sagittae of a large, adult plainfin midshipman will exceed 10.0 mm in length. Material : 148 otoliths 0.7 to 7.9 mm long (Fig. 26). Zoarcidae — eelpouts Lycodopsis pacifica (Collett) — blackbelly eelpout The blackbelly eelpout ranges from the Gulf of Alaska to Ensenada (at least), being very abundant in trawl catches made in depths of 100 to 800 feet or more. It is reported to attain a length of 18 inches and a weight of about 1/3 pound. L. pacifica otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, unpublished data). The sagittae of a large, adult blackbelly eelpout will exceed 5.0 mm in length. Material : 12 otoliths 2.2 to 5.1 mm long (Fig. 7). Brotulidae — brotulas Brosmophycis marginata (Ayres) — red brotula The red brotula ranges from southeastern Alaska to Ensenada, at least, mostly in depths of 60 to 400 feet in areas where a rocky bottom offers an opportunity for concealment. Although some individuals may reach lengths of 18 inches, few specimens are seen that exceed 16 inches or about 12 ounces. B. marginata otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, unpublished data). The sagittae of a large, adult red brotula will exceed 14.0 mm in length. Material : 1 otolith 3.7 mm long (Fig. 6). Unidentified teleosts Three badly deteriorated otoliths, 225 teeth, 575 vertebrae, and 91 assorted bone fragments recovered from this deposit could not be identified at 18 Contributions in Science No. 128 o CO o Figure 31. Present-day distributions of the 28 non-bathypelagic bony fishes identified from the Miraflores Street, San Pedro, California deposit (lat. 33° 44.8' N.) 1967 Fossil Fish Otoliths 19 this time except as “teleosts.” Most of these remains are presumed to have come from the same species that “left” their otoliths. Discussion Possibly 5 of the 40 fish species identified from the Miraflores Street deposit might be considered northern forms, 1 a “southerner,” and 3, offshore or pelagic species. Two of the northern species, Notorynchus maculatus and Clupea pallasi, occasionally have been noted south as far as the latitude of this deposit during modern times, but the known ranges of Spirinchus starksi, Microgadus proximus, and Enophrys taurina all terminate to the north of San Pedro (Fig. 31 ) . Several species of Carcharhinus are abundant in waters off southern Baja California, but only during warm summer months or years when local water temperatures are higher than normal do a few individuals stray into the southern California area. Two of the three offshore species, Prionace glauca and Icichthys locking- toni, sometimes are captured in or near the surf zone, but these may be sick individuals, because such behavior is not believed normal for either. Typically they are found in clear, blue offshore waters, generally in the upper 50 meters of the water column. Stenobrachius leucopsarus, the third “offshore” species, seldom is captured in water shallower than 1,000 feet, yet its otoliths were found in five of eight marine Pliocene and Pleistocene deposits that I have sampled extensively. In three of these five deposits, only one Stenobrachius otolith was recovered, but in the other two, Timms Point silt and Lomita marl, they were more abundant. In many respects, the Miraflores Street fish fauna is similar to what I found in both the Timms Point silt and the Lomita marl, yet in the proportion of surfperch otoliths (Embiotocidae) it is unique among all the deposits I have investigated. Three species, Clupea pallasi, Radulinus asprellus, and Bros- mophycis marginata, were present only in these three coldwater deposits. Four other species, Stenobrachius leucopsarus, Trachurus symmetricus, Cory- phopterus nicholsi, and Lycodopsis pacifica, were present only in these same three deposits plus one or at most two other deposits. (All but Trachurus were in the San Diego Pliocene, and Stenobrachius and Trachurus were both in the Playa del Rey Pleistocene.) A similarity also is seen in the insignificant role of Genyonemus lineatus and Seriphus politus otoliths in the Miraflores Street, Timms Point silt, and Lomita marl deposits. Genyonemus contributed 0, 8, and 2 otoliths respectively to the total otolith yield for these three “coldwater” deposits, while only 15, 8, and 3 Seriphus otoliths were recovered. In the five “non-coldwater” deposits that were sampled extensively (Table 2), Genyone- mus otoliths were very important, comprising from 13 (Playa del Rey) to 55 (Signal Hill and San Diego Pliocene) percent of the total recovered. Seriphus otoliths were also very important in these same five deposits. 20 Contributions in Science No. 128 Table 2. Otolith Yield from Eight Heavily-Sampled, Southern California Pliocene and Pleistocene Deposits Deposit Locality Designation Age Otoliths recovered Signal Hill (LACMIP 423 7 Pliocene 1,230* San Diego (LACMIP 305) San Diego fm. Pliocene 1 1,590* Miraleste Canyon (LACMIP 435) Lomita marl Pliocene 24,299t Timms Point (LACMIP 130) Timms Point silt Lower Pleistocene 2,370** Miraflores Street (LACMIP 332) San Pedro sand Lower Pleistocene 2,746t Playa del Rey (LACMIP 59) Palos Verdes sand Upper Pleistocene 2,591** 500 block N. Pacific (LACMIP 131) Palos Verdes sand Upper Pleistocene 662t 700 block N. Pacific Palos Verdes sand Upper Pleistocene 282t *recovered “by eye.” **about one-half recovered “by eye” and one-half with microscope, t recovered with microscope. In view of the similarities between otoliths from the Miraflores Street deposit and those recovered from the Timms Point silt and Lomita marl, I feel certain that additional sampling would yield many more species, particularly such northern forms as are unique to Timms Point and Lomita marl ( e.g ., Atheresthes stomias, Malacocottus zonurus, Icelinus burchami , Lyconectes aleutensis, Ammodytes hexapterus, etc.). No explanation is readily available as to why Cymatogaster aggregata otoliths comprised 41 percent of the total otolith yield of the Miraflores Street deposit. In no other deposit I have examined did surfperch otoliths exceed 10 percent of the total; however, a high yield of Cymatogaster otoliths was obtained in a 4-quart sediment sample from Ship Rock, on the leeward side of Santa Catalina Island, California. Department of Fish and Game biologist- divers had scooped up this sample (representing a sheltered environment) in 125 feet of water during routine diving operations, and saved it for me to examine. Sediment samples from exposed coastal areas have never yielded a disproportionate number of embiotocid otoliths, so such a phenomenon may exemplify deposition in a sheltered environment at moderate depth. 1967 Fossil Fish Otoliths 21 Interestingly, otoliths of Otophidium taylori were entirely lacking in the Miraflores Street deposit, yet they were important or very important in every other Pliocene and Pleistocene deposit sampled, and a few were found in the sediment sample from Ship Rock. This is not a reflection of inadequate sampling or poor recovery techniques at Miraflores Street — O. taylori otoliths simply were not present! Since most fossil assemblages represent death associ- ations, rather than life associations, the lack of O. taylori otoliths cannot be construed as positive proof the species was not an integral part of the fauna of the period, but it is a good indication that it was not. I have referred to the Miraflores Street site as a coldwater deposit because the otoliths represent a fish fauna such as one would find off central and northern California today (Fig. 31 ). A deposit such as this could reflect from one to 1 ,000 years or more of geological history, and within such a period many temperature anomalies could have occurred. If ocean temperatures had fluctuated widely, an admixture of “northern” and “southern” species should have resulted. Radovich (1961) showed that widespread changes in ocean temperature will result in a temporary displacement or redistribution of marine animals beyond their normal range. When such conditions exist for lengthy periods, breeding communities often become established, at least among the vertebrates. Fast-moving organisms (e.g., fishes) can also reflect short- duration temperature anomalies wherein both “northern” and “southern” species invade an area almost simultaneously (Fitch, 1966a). The lack of remains from southern fishes in this deposit indicates conditions were relatively stable during the period of deposition (i.e., there was no intrusion of warm water), or if oceanic warming did occur, very few fishes took advantage of it. Acknowledgments This study was made possible by a research grant (GB-1244) from the National Science Foundation. In addition, many individuals were helpful at one time or another: Roger D. Reimer not only found the site, but recovered the first otoliths; Jack Bell, owner of the property, granted permission to remove as much of the deposit as I needed; W. O. Addicott, U.S. Geological Survey, Menlo Park, provided information on pertinent references; Shelton P. Applegate, Los Angeles County Museum of Natural History, verified several of my shark identifications; Mrs. Gertrude Cutler helped with the numerical arrangement of the otolith photos; Richard A. Fitch assisted with obtaining field samples, and spent many hours with the microscope searching the washed residue for fish remains; Mrs. Loretta Morris, San Pedro, typed the several manuscript revisions; Jack W. Schott, San Pedro, took the excellent otolith photographs for me; Walter Thomsen, Long Beach, prepared the fish-distri- bution chart; and Charles H. Turner, Department of Fish and Game biologist- diver collected sediment samples for me from various environments and depths 22 Contributions in Science No. 128 during his diving activities. My deepest appreciation goes to my wife, Arline, whose constant encouragement and helpful advice are always uncomplain- ingly available, even after weeks-on-end when she is unable to use the dining room table for its intended purpose because it is piled high with microscope, dirt samples, vials of otoliths and other fish remains, references, manuscripts, and similar miscellany which are vital in this work. Literature Cited Applegate, Shelton P. 1966. A possible record-sized bonito shark, I sums oxyrinchus Rafinesque, from southern California. Calif. Fish and Game, 52(3) : 204-207. Best, E. A. 1957. Recent occurrences of the red brotula, Brosmophycis marginata (Ayres), in California waters. Calif. Fish and Game, 43(1): 97-98. Bigelow, Henry B., and William C. Schroeder 1948. Fishes of the Western North Atlantic. Part 1. Sharks. Mem., Sears Found. Mar. Res., ( 1 ) : 59-576. Bolin, Rolf L. 1944. A review of the marine cottid fishes of California. Stanford Ichthyol. Bull., 3(1): 1-135. Ebert, Earl E., and Charles H. Turner 1962. The nesting behavior, eggs and larvae of the bluespot goby. Calif. Fish and Game, 48(4) : 249-252. Fitch, John E. 1964. The fish fauna of the Playa del Rey locality, a southern California marine Pleistocene deposit. Los Angeles Co. Mus., Cont. in Sci., 82: 1-35. 1966a. A marine catfish, Bagre panamensis (Gill), added to the fauna of Cali- fornia, and other anomalous fish occurrences off southern California in 1965. Calif. Fish and Game, 52(2) : 214-215. 1966b. Additional fish remains, mostly otoliths, from a Pleistocene deposit at Playa del Rey, California. Los Angeles Co. Mus., Cont. in Sci., 119: 1-16. In press. Fish remains, mostly otoliths, from a Ventura, California, Chumash village site ( Ven-3 ) . In press. Fish remains recovered from a Corona del Mar, California, Indian midden (Ora-190). Calif. Fish and Game, 53. Fitch, John E., and Roger D. Reimer 1967. Otoliths and other fish remains from a Long Beach, California, Pliocene deposit. Bull. So. Calif. Acad. Sci., 66(2). Follett, W. I. 1932. Incomplete list of fishes from Point Magu [j/c] Shellmound. unpubl. ms, Los Angeles Co. Mus. Nat. Hist. 1963a. Fish remains from Arroyo Sequit Shellmound (LAn-52), Los Angeles County, California. Calif. Dept. Parks and Recr., Div. Beaches Parks, Archaeological Rept. 9, Appendix, p. 113-121. 1967 Fossil Fish Otoliths 23 1963b. Fish remains from the Century Ranch site (LAn-227), Los Angeles County, California. Ann. Rept. 1962-1963, Archaeol. Surv., Univ. Calif. Los Angeles, p. 299-313. 1964. Fish remains from a Sixteenth Century site on Drakes Bay, California. Ann. Rept. 1963-64, Archaeol. Surv., Univ. Calif. Los Angeles, p. 31-41. Garrick, J. A. F., and Leonard P. Schultz 1963. A guide to the kinds of potentially dangerous sharks (p. 3-60). In Sharks and survival, D. C. Heath & Co., Boston, 578, p. Harrington, John P. 1928. Exploration of the Burton Mound at Santa Barbara, California. 44th Ann. Rept., Bur. Amer. Ethnol., Washington, D.C., p. 21-168. McAllister, D. E. 1963. A revision of the smelt family, Osmeridae. Bull. Natl. Mus. Canada, (191): 1-53. Miller, Daniel J., Dan Gotshall, and Richard Nitsos 1965. A field guide to some common ocean sport fishes of California (second revision). Calif. Dept. Fish and Game, Sacramento, 87 p. Phillips, Julius B. 1957. A review of the rockfishes of California (family Scorpaenidae). Calif. Dept. Fish and Game, Fish Bull., 104: 1-158. Radovich, John 1961. Relationships of some marine organisms of the northeast Pacific to water temperatures particularly during 1957 through 1959. Calif. Dept. Fish and Game, Fish Bull., 1 12: 1-62. Roedel, Phil M. 1953. Common ocean fishes of the California coast. Calif. Dept. Fish and Game, Fish Bull., 91 : 1-184. Woodring, W. P., M. N. Bramlette, and W. S. W. Kew 1946. Geology and paleontology of Palos Verdes hills, California. U.S. Geol. Surv. Prof. Pap., 207: 1-145. LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE Timber 129 J July 28, 1967 b 7- 7 i 12 Li 67 ASPECTS OF THE BIOLOGY OF THE LIZARDS OF THE WHITE SANDS, NEW MEXICO By James R. Dixon Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM. — (1) The 1960 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8 Vi x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) Method of literature citation must conform to CONTRIBUTIONS style- — see number 90 and later issues. Spell out in full the title of non-English serials and places of publication. (7) A factual summary is recommended for longer papers. (8) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!5 All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings) ; photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF.— Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor ASPECTS OF THE BIOLOGY OF THE LIZARDS OF THE WHITE SANDS, NEW MEXICO By James R. Dixon1 Abstract: On the White Sands diurnal cycles of Uta, Scelo- porus and Holbrookia involved two daily activity periods, the first between 9 to 1 1 am, the second 3 to 5 pm. Cnemidophorus have one diurnal activity period, 9 to 12 am. Holbrookia and Sceloporus appear to have similar reproductive cycles and pro- duce one egg clutch each season, while Uta produces at least two egg clutches during the same period. Cnemidophorus seems to have a long incubation period, with only one egg clutch each sea- son. Uta and Sceloporus appear to maintain a body temperature between 34 to 36°C, Holbrookia between 35 to 39°C, and Cnemidophorus between 37 to 39°C. The Holbrookia population appears to have diverged most significantly from the parental stock, followed by Cnemidophorus, Sceloporus, and Uta. Introduction The White Sands of New Mexico are located in the south central part of the state, approximately 88 kilometers northeast of Las Cruces and 24 kilometers west-southwest of Alamogordo, on U.S. Highway 70-82. The sands represent a large, uniform environment, probably of pre-Pleistocene origin, and support several endemic plants, invertebrates, and vertebrates. Most of the populations on the dunes differ strikingly in color from allied populations from immediately adjacent situations and offer an interesting natural experi- ment in speciation. The organisms inhabiting the white sands have been studied by many people, notably Benson (1933), mammals; Bugbee (1942), invertebrates; Emerson (1935), plants; and Smith (1943), Lowe and Norris (1956), Ruth- ven (1907), reptiles. However, no one has attempted to study the lizards in their natural environment except Dr. Charles H. Lowe, Jr., and party in the late 1940’s and early 1950’s (Lowe, pers. comm.). The present study was initiated in June 1963, and was continued during the summer months of 1964 and 1965. The extreme heat, absence of shade, and frequent spring wind storms made field work difficult, and efforts to maintain a daytime vigil over certain study areas within the white sands was almost impossible, particularly in areas of adobe soil. Three species of lizards occur abundantly on the dunes, and the food habits of these forms were discussed by Dixon and Medica (1966), and no mention of food preferences is made in the present paper. Lizards were cap- curator of Herpetology, Los Angeles County Museum of Natural History, Los Angeles. 1 2 Contributions in Science No. 129 tured either by hand or killed by a 22 caliber dust shot. All temperature data were obtained with a Schultheis thermometer, calibrated from 0 to 50°C. A total of 440 lizard’s temperatures was taken. Geological History of Basin Herrick (1904) gives an accurate account of the geologic history of the basin and provides several interesting hypotheses involving its past. The essence of Herrick’s paper is that intensive faulting in late Cretaceous or early Tertiary reduced the Tularosa Basin to its present level, though some dislocations may have occurred at a later date. He believes that a river of considerable magnitude occupied the valley and flowed southwestward to enter the Rio Grande Valley (and river) in the vicinity of El Paso. This ancient river was later diverted, probably westward, by a great sheet of “mal pais” introduced by basaltic overflows in the Tertiary. This large lava mass essen- tially cut off the flow of water from the north end of the basin and blocked it completely. A long period of periodic rainfall and intermittent aridity assisted in silting up the outlet towards El Paso, forming a large ancient lake that is now recognized as Lake Lucero. Herrick estimated the size of the ancient lake as 1600 to 1800 square miles. The old shore lines are now buried under the talus from the mountains. Some of the talus deposits are 800 feet thick at the present time. Erosion has exposed the remnants of old lake benches along the northwest border of the basin and these are considered the early margins of the lake. The present lake bed has been examined to a depth of 200 feet, and con- sists of successional layers of gypsum and saline beds intercalated in gypsif- erous marls. The San Andres Mountains to the west and the Sacramento Mountains to the east are layered with Cretaceous sandstone and shale, and with Permian soft gypsiferous shales and sandstones. Weathering of the layers has probably produced the present sediments occupying the lake bed. The majority of the springs that feed the basin are only a few hundred yards long in most cases, and are high in dissolved salt content. Mal Pais Spring contains 15 per cent dissolved salt, Salt Creek 35 per cent, Lost River 7 per cent, and the Tularosa River 7.8 per cent. At the present time, as in the historic past, percolating water from summer rains and melting winter snow carries tons of dissolved gypsum and salt from the mountains into the lake bed. The warm spring and summer winds evaporate the water and deposit extensive beds of pure gypsum on the surface. The wind picks up the gypsum particles and deposits them to the northeast, the direction of the prevailing winds. These deposits have formed extensive dunes that are now under the jurisdiction of the U. S. Park Service, the White Sands National Monument. The white sand dunes are continually changing their shape and moving to the northeast at the rate of a few centimeters a year. The older dunes lying 1967 Biology of the Lizards of New Mexico 3 to the southwest have become stabilized with vegetation. These dunes are now encrusted with a mixture of adobe soils and gypsum. The present active dunes occupy some 444 square kilometers of the basin, and may rise as high as 10 meters from the basin floor. They are composed of almost pure gypsum (hydrous calcium sulphate) and are generally damp within a few centimeters of the surface. The water table is high most of the year, usually within one meter of the surface of the depressions that lie between the dunes. Vegetation and Climate The adobe soils surrounding the dunes support 13 species of grasses, the common ones being three awn ( Aristida spp.), drop seed ( Sporobolus sp.), muhly ( Muhlenbergia sp.), gramma ( Bouteloua sp.), alkali ( Distichlis sp.), burrograss ( Scleropogon sp.), and sand bunchgrass ( Oryzopsis sp.). Some of the larger trees and shrubs are cottonwood ( Populus wislizeni), mesquite (Prosopis glandulosa), creosote bush ( Larrea tridentata ), squawbush ( Rhus trilobata ), rabbitbush ( Chrysothamnus pulchellus ), saltbush ( Atriplex canes- cens ), and shrubby pennyroyal ( Poliomintha incana) . The old stabilized dunes that lie to the southwest are slowly being covered by the latter vegetation (Fig. 1 ) . A large alkali flat lies to the west of the major dune area and is essen- tially void of vegetation with the exception of two genera, burroweed ( Allen - rolfea occidental™) and sand verbena (Abronia augustifolia) . Figure 1. View of stabilized gypsum dunes along the southwest edge of the White Sands National Monument. Some of the reptiles found in this area are: Cnemidoph- orus inornatus, Cnemidophorus neomexicanus, Crotaphytus collaris, Sceloporus undulatus, Sceloporus magister, Phrynosoma modestum, Phrynosoma cornutum, Uta stansburiana, Crotalus atrox, Crotalus viridis, and Pituophis melanoleucus. 4 Contributions in Science No. 129 The southern edge of the dunes is slowly becoming stabilized and one may find several stages of plant succession taking place. Some of the more con- spicuous plants are Atriplex, Rhus, Chrysothamnus, and Populus. Figure 2. View of the central area of the White Sands looking northwest. The arrow points to a single cottonwood protruding from the top of a dune. The depression between the dunes on the left is almost void of vegetation while the depression in the center contains a fair amount of plant cover. The active dunes are lacking in vegetative cover with the exception of an occasional yucca ( Yucca elata ) and cottonwood (P. wislizeni ) rising from the top of a dune (Fig. 2). The depressions between the dunes frequently have a scattering cover of sand verbena ( Phyla incisa ), primrose ( Anogra gypso- phila), sand bunchgrass (O. hymenoides ), mormon tea ( Ephedra torreyana), drop seed ( Sporobolus sp.), four o’clock ( Abronia angustifolia) , yucca ( Y . elata), and rabbitbush (C. pulchellus) . The depressions closest to the edge of the dunes have dense vegetative cover, while those near the center of the dunes are sparse (Fig. 3). An excellent account of the plants of the white sands is given by Emerson (1935). The average summer (June, July, August) air temperature of the dunes (six feet above the ground) ranges from 16°C at night to 38 °C during the day. The average winter (December, January, February) air temperatures are 11°C during the day and — 6°C at night. The summer temperatures of the adobe and gypsum soils vary with the amount of wind, moisture, and cloud cover. However, the adobe soil is generally warmer than the gypsum soils (Fig. 4). 1967 Biology of the Lizards of New Mexico 5 Figure 3. View of a depression between the dunes in the center of the White Sands. The dark plant in the foreground is sand verbena, the scattered clumps of grass are single stands of sand bunchgrass, and the large plants in the background are Ephedra (mormon tea) and rabbitbush. Natural History of the Lizards Sceloporus undulatus: The western fence lizard is common in the depres- sions between the dunes wherever rabbitbush, yucca, mormon tea, and sand bunchgrass are present. The light color phase of this species found in the center of the active dune area is occasionally found along the edgs of the dunes in adobe soil. However, none was found farther than 10 meters away from a gypsum dune. The normal dark phase found on the adobe soil is present in the stabilized dunes where the dunes have become encrusted with adobe dust. They have also been found in adobe depressions within the gypsum dunes. The latter areas have recently been captured by the movement of the dunes to the northeast, but have not become completely covered by the sand. These lizards are never found more than a meter or two from plant cover. The majority of individuals were taken from the basal areas of rabbitbush, yucca, sand bunchgrass, or from the lower branches of rabbitbush and mormon tea. The lizards were reluctant to leave plant cover, even when pursued by hand. There are usually twice as many males as females present at all times during June, July, and August. Both sexes are active from about 8 am to 6 pm Ill I 1 I I I III I 7 8 9 10 11 12 1 2 3 4 5 6 HOURS (SUNLIGHT) • = ADOBE SOIL O = GYPSUM SOIL Figure 4. Typical daytime adobe and gypsum soil temperatures during mid-July, vicinity of White Sands National Monument headquarters, New Mexico. 1967 Biology of the Lizards of New Mexico 7 on a normal sunny day. The majority of individuals were observed between 9 and 1 1 am, the least number at 7 am and 2 pm. There appeared to be a late period of activity between 4 and 6 pm (Fig. 5). The latter activity appears to be correlated to the heat holding capacity of the sand beneath dense clumps of rabbitbush. The surrounding soil temperature was 28 °C while that below the shrubs was 35°C at 5:30 pm. TIME SCELOPORUS UNDULATUS Figure 5. Vertical axis represents the average number of individuals of Sceloporus undulatus observed each hour on the White Sands during the summer months of 1963-65, horizontal axis indicates hourly intervals during the same period. Gravid females were taken from June 12 through July 16. A few hatch- lings were observed on August 16, and became increasingly more abundant in the population during late August and early September. Gravid females dis- 8 Contributions in Science No. 129 played bright colors on the dorsum and tail. The mid-dorsal and lateral light stripes became progressively more yellow to yellowish-orange from late June to mid-July. Shortly after egg deposition, the colors faded back to their normal gray to light buff color. This color change may be attributed to the presence of fat deposits during pregnancy and/or to a hormonal mechanism that is induced while the females are carrying the eggs. These bright colors may also tend to warn the males that the female is gravid and copulation has already been achieved. Body temperature and associated soil and air temperatures were taken for 140 individuals during June, July, and August. The lizards were either captured by hand or shot. An analysis of cloacal temperatures resulting from the two capture methods indicated that there was no significant difference between the two methods. The body temperature was found to be higher than the ambient temperatures, except during the hottest part of the day (2 to 4 pm). The body temperature varied from 32.3 °C to 37.2°C between the hours of 8 am and 6 pm. The majority of individuals were observed sunning themselves between the hours of 9 and 1 1 am. During the latter hours, the lizards were reluctant to leave their basking site when disturbed. The basking site was usually beneath the branches of plants that allowed some sunlight to filter through to the soil. From about 12 noon to 4 pm, the lizards were somewhat inactive, remaining in the shade of plant cover (Fig. 6), at which time their body temperature varied from 33.0°C to 34.7 °C. The corresponding ambient temperatures indicated an appreciable drop in temperature from sunlight to shade (Fig. 6) between 12 noon and 1 pm, followed by a steadily increasing temperature towards the hottest part of the day. The lizards were again active in and out of sunlight between the hours of 4 to 5 pm. Three individuals were captured, marked with paint, and then removed from their home site for a distance of 600 meters. The place of release was three dunes and two depressions to the northeast of the homesite, a rabbitbush flagged with red cloth. All three lizards returned to their homesite within four days, one returned in three days, the other two in four. These lizards were observed to have a small home range within the dunes, seldom moving more than 10 meters from a particular plant. The experiment, although tried only once seems to indicate that these lizards have a decided tendency to remain within familiar cover, and have some powers of orientation. The Sceloporus undulatus cowlesi population occupying the center of the gypsum dunes is distinguishable from S. u. consobrinus populations off the dune area only by color. Lowe and Norris (1956) indicated that the two populations were distinct from one another on the basis of color, color pattern, and amount of prefrontal contact. An examination of 15 specimens from the adjacent adobe soils, 15 from the edge of the gypsum dunes, and 85 from the center of the gypsum dunes reveals evidence of gene flow between the three samples. A comparison of the number of femoral pore series, and the width 1967 Biology of the Lizards of New Mexico 9 • = cloacal 0= air ®= SOIL SCELOPORUS UNDULATUS Figure 6. Vertical axis represents the average cloacal and ambient temperatures of Sceloporus undulatus on the White Sands during the summer months of 1963 to 1965, horizontal axis indicates hourly intervals during the same period. The drop in cloacal temperature between the hours of 12 and 2 indicates a period of inactivity in the shade of vegetation. The rise in cloacal temperature between the hours of 4 and 5 indicates a period of activity and basking prior to burrowing in the sand for the night. of the mid-dorsal stripe are nearly identical in all three samples. The number of scales separating the lateral blue belly patches is ontogenetic, varying from 6 to 7 scales in specimens measuring 28 to 40 millimeters in snout-vent length, to 1 to 4 scales in specimens measuring 50 to 61 millimeters. The prefrontals are in contact with one another in 95 per cent of a sample of 23 individuals examined by Lowe and Norris (1956) from the White Sands. The amount of prefrontal contact is 37 per cent in the sample from the center of the dunes, 44 per cent in the edge of the dunes sample, and 36 per cent in the sample from the adjacent adobe soils. The latter two samples are small (15 each), but the larger series from the center of the dunes indicates a greater amount of non-contact of the prefrontals than found by Lowe and Norris (1956). The color and color pattern of the sample from the center of the gypsum dunes are more variable than indicated by Lowe and Norris (1956). The ground color of male S. undulatas cowlesi is dirty gray to pale yellowish-white, with a light blue mid-dorsal stripe from the occiput to the tail. The lateral dark stripe is buff, the ventrolateral dark stripe dark gray. The belly patches are bright blue, bordered mid-ventrally with black edges. The throat patch is similar to the belly patch in color. The whitish ground color of the belly is immaculate, lacking the gray and grayish-black flecks found in individuals from the adobe soils. The female retains the basic color pattern of individ- 10 Contributions in Science No. 129 uals examined from adobe soils, but the ground color is more buff, with some light orange color on the top of the head, and a mid-dorsal stripe of light bluish-gray. The ground color of the male S. undulatus occupying adobe soils is dark grayish-brown with a pair of dorsolateral gray stripes bordered above by a mid-dorsal grayish-brown stripe, whose intermargins are lined with a linear series of dark brown spots. The dorsolateral light stripe is bordered below by a dark brown stripe, that is in turn bordered below by light grayish-brown stripe that is bordered below by ground color. The belly is dirty white with scattered flecks of grayish or grayish-black. The lateral belly patches are bright blue, the inner margins edged with black. The throat patch is dark blue, out- lined with black and united across the throat. Females are similar in color but lack the brightness of the belly and throat patches. The latter color areas are usually smaller and lack the black borders. The striping is more pronounced in the females and in addition to the mid-dorsal stripe, a series of whitish linear spots is present along its outer margins. Individuals from the edge of the dunes are lighter in color than the adobe soil sample, but darker than those from the center of the dunes. The general ground color is light brown to buff with a lighter mid-dorsal stripe. Individuals examined from depressions between the dunes 1, 2, 3, 5, and 7 kilometers, respectively, towards the center of the dunes from the outer edge, reveal a progressively lighter ground color, especially within the first 3 kilometers. However, an occasional, relatively dark individual is found in the center of the dunes. There is apparently more gene exchange between normal and pallid color populations than was formerly believed. This may be due to the movement of the dunes towards the northeast, capturing fresh areas of adobe soil that include normal colored individuals. These captured areas are generally exten- sive and may include areas up to 0.4 square kilometers. The captured areas slowly fill in with gypsum sand, and slowly turn from dark brown soil to white soil. Light and medium dark specimens have been observed in various color stages in the captured areas. The behavior of this species may tend to offset the selective pressure favor- ing the light color. These lizards tend to stay in plant cover, and 99 per cent of the plant cover is found in the depressions between the dunes, usually upon an adobe base encrusted with a gypsum-adobe mixture. The soil is usually buff to dirty gray in these areas, offsetting the brilliant whiteness of the surrounding dunes. The majority of the specimens were found in or beneath rabbitbush, mormon tea, sand bunchgrass, or on the trunks of yucca. These plants have grayish to brown bark, suggesting a selective pressure acting on the lizard population to retain brown pigment as a means of camouflage while associated with the plant cover. This selection pressure would be working counter to selection for light colors effective for travel from plant to plant. 1967 Biology of the Lizards of New Mexico 11 Holbrookia maculata: This arenicolous lizard is common on the dunes and in depressions between the dunes. It does not seem restricted to the imme- diate vicinity of plant cover as is S. undulatus. Several individuals were found beneath rabbitbush, mormon tea, and sand bunchgrass during the heat of the day, but their foraging activity seems to be in more open areas. Individuals could not be found in adobe soils adjacent to the gypsum dunes. The nearest known locality to the white sands of normal colored individuals is 24 kilometers south of Alamogordo. The latter area is composed of reddish quartz sand and lies some 29 airline kilometers southeast of the Monument. The population inhabiting the dunes appears to be restricted to the central dune area. None was seen within 3.2 kilometers of the edge of the dunes. Both sexes appear to be active at the same time. However, there seemed to be a third more males present in the population during June and July, with females becoming slightly more numerous than males in August. Their greatest activity period was observed to be between 9 am and 12 noon, when the soil temperature ranged from 32 to 36°C. [On a day to day tally, greater numbers were always observed during the latter period (Fig. 7).] As in S. undulatus, there appears to be a period of inactivity between 2 and 4 pm, at which time the soil temperature reaches about 40°C. Individuals were observed burrowing in the surface sand of the dunes when the soil temperature reached 40°C. They reappeared about 4 pm when the soil tem- perature was approximately 38 °C but dropping. Individuals remained active beneath dense clumps of rabbitbush until about 6 pm, when the soil tempera- ture beneath the shrubs had dropped to 33 °C. Gravid females were observed in June and July, with the deposition of eggs occurring over an 8-day period, July 8 to the 16th (1963). Females revealed bright colors on the hind limbs, sides of the body, and top of the head during the last week of June. The colors began as light lemon yellow and turned progressively darker and brighter (yellow to orange) through the first week of July. The top of the head became bright yellow to orange in some females just prior to egg deposition. All females captured for obtaining body temperatures on July 16 had deposited their eggs and the bright colors of the head, body, and limbs were fading to a washed-out yellow. By July 27, the colors were no longer obvious. Hatchlings were seen for the first time on August 17, and became progressively more abundant through early September. Body temperatures and associated ambient temperatures were obtained on 176 individuals, including all sexes and ages, during June, July, and August. The body temperature was always above that for the air and ground (Fig. 8). The majority of the body temperatures were taken when the lizards were most abundant (9 to 11 am). Many were observed basking during the latter period. This species began seeking shade at about 1 1 am, when the soil temperature reached 36 to 38 °C. The lizards continued to be active in and around vegeta- tive cover until the soil temperature reached approximately 40 °C. 12 Contributions in Science No. 129 NUMBER OF INDIVIDUALS TIME HOLBROOKIA MACULATA Figure 7. Vertical axis represents the average number of individuals of Holbrookia maculata observed each hour on the White Sands during the summer months of 1963 to 1965, horizontal axis indicates hourly intervals during the same period. The white sands earless lizards, H. m. ruthveni, was described as a distinct form by Smith in 1943. His description of the form indicated that squamation was similar in all the samples examined, but color alone would separate the white sands form from all others (principally the closest race, H. maculata flavilenta). An examination of 160 specimens from the white sands and 50 normal colored individuals from southern New Mexico indicates a much closer rela- 1967 Biology of the Lizards of New Mexico 13 39- \ CENTIGRADE 8-9 9- 10- 11- 12- TIME 1- 2 - 3- 4- 5-6 • : CLOACAE Or AIR ® = SOIL H0LBR00KIA MACULATA Figure 8. Vertical axis represents the average cloacal and ambient temperatures of Holbrookia maculata on the White Sands during the summer months of 1963 to 1965, horizontal axis indicates the hourly intervals during the same period. The drop in cloacal temperature between the hours of 12 and 1 indicates a period of rest in the shade. The absence of cloacal temperatures between the hours of 2 and 4 indi- cates the period of rest beneath the surface of the sand. The temperature between the hours of 4 and 6 indicates a period of activity prior to burrowing beneath the sand for the night. tionship between the populations. In white sands males, the variation in ground color ranges from dark bluish-gray to light yellowish-white. A dorsal pattern of two dorsolateral linear series of small blackish spots from the rear of the head to the base of the tail is present in 90 per cent of the sample. In addition, a series of small white spots is scattered over the sides of the body, varying in intensity with the type of ground color of each specimen. The mid-dorsal area is pale bluish to bluish-white, lacking spots of any kind in most specimens. Occasionally, white spots may be present in some individuals. Normal colored males of H. m. fiavilenta have a ground color of light to dark brown with the dorsolateral series of black spots forming a series of chevrons that continue from the ear of the head on to the proximal one-third of the tail. The lateral white spots are present, but larger and less numerous than in the White Sands sample. The mid-dorsal area is bluish-gray to brown. The brown barring of the limbs is distinct, much less so in the white sands sample. White sands females exhibit dorsal colors and color pattern similar to 14 Contributions in Science No. 129 the males, but with less white spotting. One female is almost normal colored, with distinct chevrons on the dorsum and dark flanks. The opposite extreme is also obtained, with the near absence of any dorsal pattern of any kind in some individuals. It appears that the white sands population has retained some of the genes necessary to produce dark individuals. The majority of the samples examined were light colored, but enough darker individuals were present to cast doubt upon the validity of recognizing all white populations of lizards. Males were observed fighting with one another on several occasions in June. Biting, tail thrashing, pursuit, and normal bluff display (see Clarke, 1965, for details of combat and display in Holbrookia ) were used by the males. During the mid-morning period male Holbrookia often stationed themselves on an elevated area near the base of sand verbena or sand bunchgrass. From the latter vantage point they were observed in challenge displays. Males were seen in companionship with females on several occasions but copulation was not observed. During late morning hours males appeared to wander about more than females. The majority of females were found near vegetation during their activity periods, while males were found traversing large dunes some distance from plant cover. Usually when the tracks of Holbrookia were fol- lowed across the open dunes to the point where they buried themselves in the sand, the individual extracted from the sand was a male, except on one occasion. Cnemidophorus inornatus: The inornate whiptail lizard is abundant throughout the depressions between the dunes, along the edge of the dunes, and was seen as far as 50 meters from the dunes in adobe soil. C. neomexicanus was the only other whiptail seen beyond the fringes of the dunes. The latter species also occurs in the stabilized dunes along the southwest edge of the Monument. C. inornatus seems to be most abundant in the depressions between the dunes where there is a considerable amount of sand bunchgrass, yucca, and rabbitbush. Low dunes that were partly covered with vegetation were frequently utilized by the lizards. A few were seen traversing the large dunes that were void of vegetation. The sex ratio of the sample varied little over the three month summer period. Females represented 55 per cent of the sample in June, 38 per cent in July, and 45 per cent in August. Both sexes appeared to be equally active between 9 and 10 am, with the largest numbers observed at that time. This species seldom appeared upon the surface of the white sands before 9 am, while the population on the adjacent adobe soil was active about 8 am. There was a definite lag (about an hour) in soil temperature between the adobe and gypsum soils during the morning hours. The adobe soil reached 33 °C at approximately 8 am, while the gypsum soil did not reach this temperature until approximately 9 am. As the soil temperature of the dunes reaches 33°C, the lizards become 1967 Biology of the Lizards of New Mexico 15 active almost immediately. They appeared in large numbers (Fig. 9) during the first hour after their emergence from their burrows, but the numbers began to dwindle after an hour or two following their first appearance, and none was seen after 2 pm, unless the weather was extremely cool following a rain. Those individuals observed on the adobe soil disappeared at approximately 1 1 : 30 am when the soil temperature was between 49 and 51°C. NUMBER OF INDIVIDUALS 7-8 8- 9- 10- 11- 12- 1- 2- 3- 4- 5-6 TIME CNEMIDOPHORUS INORNATUS Figure 9. Vertical axis represents the average number of individuals of Cnemidoph- orus inornatus observed each hour on the White Sands during the summer months of 1963 to 1965, horizontal axis indicates hourly intervals during the same period. Gravid females were observed during late June and July, but hatchlings were not seen until the latter part of August (31st) . The females of this species apparently carry the eggs for a longer period of time, or the incubation period is longer. Medica (In press ) indicated an incubation period of 46 days for C. inornatus from the vicinity of Las Cruces, New Mexico. In any case, the young of this species appear later during the summer than the young of Hoi- 16 Contributions in Science No. 129 brookia and Sceloporus. Color change in females before and after egg deposi- tion was not observed in this species. Body and ambient temperatures were obtained on 84 individuals of this species. The body temperature always appeared to be 3 to 4°C above that recorded for the air and soil (Fig. 10). Most individuals were found foraging about plant cover when they were obtained for body temperatures. Most individuals were observed in direct sunlight during their foraging period, but prior to their subsequent disappearance into burrows, they were seen resting in the shade of vegetation. At the onset of the latter period, this species became very wary and sought burrows when approached. Several body temperatures were obtained from individuals while in their burrows in late afternoon. The lizard burrows were found approximately 10 to 15 centimeters below the surface at the base of sand bunchgrass, yucca, and rabbitbush. The body temperature of these lizards was approximately 2°C above that of the sur- rounding soil. • : cloacal 0= air (j = soil CNEMIDOPHORUS INORNATUS Figure 10. Vertical axis represents the average cloacal and ambient temperatures of Cnemidophorus inornatus on the White Sands during the summer months of 1963 to 1965, horizontal axis indicates hourly intervals during the same period. The absence of temperature data between the hours of 2 and 6 indicates a period of rest beneath the sand surface. The indicated burrow temperature represents 3 individuals taken from burrows on three different occasions. The color of C. inornatus from the central part of the dunes is strikingly different from typical normal colored samples from southern New Mexico. The mid-dorsal light stripe is absent in about 25 per cent of the sample. The ground color is pale yellowish-gray to pale bluish-gray with the 6 to 7 light stripes present but somewhat obscure. The limbs are pale blue without a 1967 Biology of the Lizards of New Mexico 17 suffusion of grayish bars on the dorsal surfaces in most specimens. The head is light brown to gray-blue in females, bright sky-blue to blue in males. The ground color of a sample taken 20 meters from the dunes is light grayish-brown to brown, with 7 white lines distinct in all specimens. Specimens taken from the edge of the dunes are much lighter in color than those taken 20 meters into the adobe soil, but appear to be intermediate between those off the dunes and those in the center of the dunes. However, those from the edge of the dunes lack the suffused grayish bars on the upper surfaces of the limbs that are present in those on the adobe soils. A sample of 200 indi- viduals from the vicinity of Las Cruces has a ground color of dark brown to chocolate brown with the 7 white lines greatly contrasting against the dark ground color. Two individuals of this color type were found within 50 meters of the dunes. However, an occasional individual is found in the central area of the dunes that is slightly darker than its lighter counterpart, indicating a possibility of some gene exchange with normal colored populations, or a retention of some genes for darker color. Males were observed fighting with other males throughout the summer. In some instances the fighting pairs were not collected and sex could not be determined. The fight consisted of a bluff display, followed by pursuit, biting, and tail thrashing. Males were observed following females in July, but the females often turned and pursued the males for a short distance. Copulation was not observed in this species. Both males and females seemed to wander aimlessly over a large area. One male was followed for approximately 30 linear meters. Fighting usually occurred when the paths of two lizards crossed one another at the same time. The majority of the burrows occupied by this species appeared to be those also occupied by the tenebrionid beetle, Eleodes sp. It is not known whether the beetle or lizard digs the burrow. In some instances, the burrows utilized were obviously those of Perognathus apache gypsi. One whiptail was found in a burrow of the pocket gopher, Geomys arenarius. Occasionally an individual of S. undulatus was found in a burrow with C. inornatus and tene- brionid beetles. Uta stansburiana: The side-blotched lizard is common on the adobe soils surrounding the white sands, but has never been observed for more than 30 meters into the dunes. It has been taken in association with Sceloporus undu- latus, S. magister, Crotaphytus collaris, Cnemidorphorus neomexicanus and C. inornatus on the adobe soils, and with S. undulatus and C. inornatus along the edge of the dunes. Smith (1943) indicated that Uta occur commonly on the dunes with S. undulatus and H. maculata, but we have been unable to find Uta anywhere in the dunes beyond 30 meters. Ruthven (1907) indicated that he did not find Uta anywhere in the white sands except on the outer edge of the dunes. Very few Uta were taken in open areas between vegetative cover on the 18 Contributions in Science No. 129 peripheral dunes. The majority of specimens were obtained beneath the foliage of squawbush ( Rhus trilobata ) and pennyroyal ( Poliomentha incana), both of which form dense ground cover. Uta were found associated with tar bush ( Flourensia sp.), creosote bush ( Larrea tridentata) , burrograss ( Scleropogon sp.), and gramma grass ( Bouteloua sp.) on the adobe soils. Both sexes were active from approximately 7 am to 12 noon. The lizards spent the remainder of the day in dense shade and in rodent burrows. The majority of individuals were observed between 8 and 1 1 am, with their num- bers decreasing gradually until the soil temperature reached 50°C+ in open sunlight on adobe soils, and 40°C+ on the edge of the dunes (Fig. 11). A few sporadic observations were made on Uta in the afternoon on the adobe soils. The area surrounding the dunes becomes unbearably hot from noon to about 4:30 pm, making continuous observations difficult. NUMBER OF INDIVIDUALS TIME UTA STANSBURIANA Figure 11. Vertical axis represents the average number of individuals of Uta stans- buriana observed each hour along the edge of the White Sands during the summer months of 1963 to 1965, horizontal axis indicates hourly intervals during the same period. The absence of individuals between the hours of 12 and 3 is probably due to inadequate observations during this period. 1967 Biology of the Lizards of New Mexico 19 Gravid females were observed in mid-June and mid-July, with hatchlings present in the population on July 23 and August 25. The latter observations lend support to Tinkle’s (1961) study that mature females may have as many as three clutches of eggs per year (reproductive season). The body temperature of 40 individuals of Uta from the adobe soils is higher than the air temperature, but much lower than the soil temperature (Fig. 12). The soil temperature increased at a rapid rate between 10 am and 12 noon, and the density of Uta was noticeably less after the soil temperature reached 39°C. Uta appeared to be active on the adobe soils from 7:30 am (soil temperature 36°C) to 10 am (soil temperature 43 °C and rising rapidly). 44 ' 43 - 42 - 41 - 40 - 39 - CENTIGRADE 38 - 37 - 36 35 34 8-9 9- 10- 11-12 TIME • = CLOACAL O = AIR ® = soil UTA STANSBURIANA Figure 12. Vertical axis represents the average cloacal and ambient temperatures of Uta stansburiana along the edge of the White Sands during the summer months of 1963 to 1965, horizontal axis indicates hourly intervals during the same period. The absence of temperature data for the afternoon period is due to inadequate sampling during this period. Note the high soil and low cloacal temperatures during the morn- ing hours. 20 Contributions in Science No. 129 They were observed feeding, basking, and presenting territorial displays during the latter period. The ground color of individuals on the gypsum dunes and on the adobe soils is similar. The small sample from the edge of the dunes is slightly lighter in color, but the basic color and color pattern are identical to the sample obtained from the adobe soils. Discussion Each of the four species of lizards found within the immediate vicinity of the White Sands of New Mexico shows a distinctive evolutionary mode. Each species has adapted to the white environment in a slightly different fashion as an interaction between the forces of genetic materials and natural selection. The dunes population of Sceloporus undulatus appears to have some gene flow with its parental population on the adobe soil. Individuals found along the fringes of the dunes are intermediate in color and color pattern between the dunes and adobe soil populations. Occasionally a dark individual is found well within the confines of the dunes and a light individual near the edge of the dunes on adobe soil. The presence of dark colored individuals is probably due to the behavior of the species. S. undulatus is associated with dense shrubby vegetation. Selec- tion for cryptic coloration probably favors the survival of dark individuals living in close association with vegetation. This is evident in the adobe soil population. The survival of individuals foraging for food on dark soils and traversing open areas in search of food and sexual companionship is enhanced by dark coloration. The white gypsum soil population is, in effect, being challenged by two opposing selective pressures : pressure favoring dark pigment associated with the microenvironment of conditions within the clumps of vegetation; pressure favoring light pigment, an adaption to the white soil macroenvironment. The latter selective pressure appears to be over-riding the selection for dark pigment on the dunes, but both pressures are still acting on the population. The few dark individuals present on the dunes are probably being selected against, but a behavior pattern of close association with the vegetation is apparently advantageous to the population, regardless of color. Eventually, the dunes population should reach an equilibrium, combining a minimum selection for dark color associated with the vegetation, and a maxi- mum selection for the white macroenvironment. Representatives of the dark soil counterpart to the gypsum dune popula- tion are seemingly able to wander into the dunes without being heavily preyed upon because of their association with the vegetation. Thus gene exchange between the dunes population and the dark soil counterpart does not seem to be greatly restricted at this time. The two populations probably represent ecological races in the sense of Grant (1963). 1967 Biology of the Lizards of New Mexico 21 The dunes population of Cnemidophorus inornatus appears to have re- stricted gene exchange with the population on the adobe soil. The individuals sampled from the edge of the dunes are intermediate in color and color pattern between the dunes and adobe soil populations. This zone of intermediacy is narrow and occupies a belt of mixed adobe and gypsum soils from 1 to 10 meters in width along the edge of the dunes. The boundaries of the latter soil belt are abruptly demarcated, somewhat tan in the middle, but rapidly grading into dark brown adobe soil and white gypsum soil on either side. The dunes population is very light colored with the dark stripes obscure in most specimens. Unlike S. undulatus this species requires open areas for foraging, between clumps of vegetation and on the bare slopes of the dunes. The absence of dark individuals on the gypsum soil indicates a strong selection for light pigment. The population occupying the intermediate zone may represent a deme or a narrow zone of restricted gene flow between the parental stock and the paternal offspring. The latter is more likely because of the narrow width of the intermediate zone and the close proximity of light and dark individuals on either side of the zone. The two totally different environments probably enforce the restriction of gene flow at the present time and allow only inter- mediates to survive out of each clutch of eggs from a cross between dark and light colored individuals. According to Grant’s (1963) classification of popu- lation systems, the two populations may represent contiguous allopatric races. The dunes population of Holbrookia maculata seems to be completely isolated from its dark soil counterpart. No evidence of gene exchange between the dunes population and the population found on the reddish quartz sands near the gypsum dunes has been found. One intermediate colored individual was found in the center of the dunes. The dunes population may have reached the final stages of speciation where almost all individuals are white with an occasional darker individual appearing in a clutch as a remnant of the original gene pool. If this is the case, the white sands Holbrookia population may represent an allopatric semi-species as defined by Grant (1963). However, without genetic compatability studies the status of the dunes population will remain in doubt. The evolutionary history of the Uta stansburiana populations from the vicinity of the White Sands is uncertain. Uta have not successfully invaded the dunes for an appreciable distance. Uta occur in the intermediate zone (adobe-gypsum soil mixture) and slightly beyond in more stabilized dunes. Three possible reasons for the absence of Uta in the major portion of the dunes are: (1) other species invaded first and filled the available niche, (2) Uta invaded the dunes along with the other species but could not maintain itself due to the absence of enough food and shelter (competition with the other species), (3) Uta is genetically unable to adapt to the white soil environment. 22 Contributions in Science No. 129 Acknowledgments I wish to thank the personnel of the White Sands National Monument for their cooperation while the author was a collaborator for the Monument from 1963 to 1965. My heartfelt thanks go to Philip A. Medica, Walter and Joan Conley, and Mike and Judy Rubick who spent many hours toiling in the dunes with me. I thank the Society of Sigma Xi Grants-in-Aid of Research committee for funds for summer field work in the white sands during 1963 and W. Ronald Heyer, Robert J. Lavenberg, Dr. Charles H. Lowe, Jr., and Dr. Jay M. Savage for critically reading the manuscript. Literature Cited Benson, S. B. 1933. Concealing coloration among some desert rodents of the southwestern United States. U. Calif. Publ. Zool., 40(1) : 1-70. Bugbee, R. E. 1942. Notes on animal occurrence and activity in the White Sands National Monument, New Mexico. Trans. Kansas Acad. Sci., 45(42) : 3 15-321 . Clarke, R. F. 1965. An ethnological study of the iguanid lizard genera Callisaurus, Copho- saurus, and Holbrookia. Emporia State Res. Stud., 13(4) : 1-66. Dixon, J. R., and P. A. Medica 1966. Summer food of four species of lizards from the vicinity of White Sands, New Mexico. Los Angeles County Mus., Cont. in Sci., 121:1-6. Emerson, F. W. 1935. Ecological reconnaissance in the White Sands, New Mexico. Ecology, 16(2) :226-233. Grant, V. 1963. The origin of adaptations. New York: Columbia Univ. Press, v+606 p. Herrick, C. L. 1904. Lake Otero, an ancient salt lake basin in southeastern New Mexico. Amer. Geol., 34:174-189. Lowe, C. H., and K. S. Norris 1956. A subspecies of the lizard Sceloporus undulatus from the White Sands of New Mexico. Herpetologica, 12(2) : 125-127. Medica, P. A. In press. Food habits, habitat preference, reproduction, and diurnal activity in four sympatric species of whiptail lizards. ( Cnemidophorus ) in South Central New Mexico. Bull. So. Calif. Acad. Sci., 66. Ruthven, A. G. 1907. A collection of reptiles and amphibians from southern New Mexico and Arizona. Bull. Amer. Mus. Nat. Hist., 23:483-604. Smith, H. M. 1943. The white sands earless lizard. Field Mus. Nat. Hist., Zool. Ser., 24(30) :339-344. Tinkle, D. W. 1961. Population structure and reproduction in the lizard Uta stansburiana stejnegeri. Amer. Midi. Nat., 66(1) :206-234. LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE NMBER 130 July 28. 1967 RESULTS OF THE 1966 CHENEY EXPEDITION TO fl-73 I n / the samburu district, kenya. ORNITHOLOGY By Herbert Friedmann and Kenneth E. Stager Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) The 1960 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8V2 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) Method of literature citation must conform to CONTRIBUTIONS style — see number 90 and later issues. Spell out in full the title of non-English serials and places of publication. (7) A factual summary is recommended for longer papers. (8) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor RESULTS OF THE 1966 CHENEY EXPEDITION TO THE SAMBURU DISTRICT, KENYA. ORNITHOLOGY By Herbert Friedmann1 and Kenneth E. Stager2 Abstract: The 1966 Cheney Kenyan Expedition was organ- ized and lead by Mr. William J. Cheney, whose primary goal was to collect a pair of leopards for an exhibition group in the Mu- seum. The choice of locality was determined by this considera- tion, but as the area visited is one that has been little studied, the birds collected help to fill gaps in the distributional data on many of the included species. Some 151 species and subspecies were obtained, and are reported on in this paper. The 1966 Cheney Expedition of the Los Angeles County Museum of Natural History worked in the Samburu District of the Northern Frontier Division of Kenya. The region was selected as a likely place in which to obtain a pair of leopards needed for an exhibition group, and of considerable interest ornithologically as well. The expedition suceeded in its chief objective, and not only assured the Museum a fine habitat group of leopards but also obtained a pair of the relatively little known aardwolf. This will give the Museum the opportunity to create a habitat display of this rare animal, a quite unexpected addition to our African exhibits. The expedition departed from Nairobi, Kenya, on June 1, 1966, and travelled north to the Samburu District by way of Fort Hall, Nyeri, Rumuruti and Suguta Marmor. Camp was established at the west base of the Karissia Hills, approximately 10 miles southeast of Maralal, on June 1, 1966, (Fig. 1). The camp was located at Bauwa in the Lorogi forest at an elevation of 6,500 feet, and the expedition collected in this area from June 2 through June 20. On June 21 the expedition moved camp to the west base of Mt. Nyiru (Fig. 1) and collecting continued in that area from June 21 through June 27. On June 28, 1966, the expedition returned to Maralal and then on to Nairobi, arriving there on June 30, 1966. Expedition personnel consisted of William J. Cheney, sponsor; Kenneth E. Stager, ornithologist; Peter Saw, professional hunter (Monty Brown Safaris) and his African staff of thirteen men. One trained specimen preparator was provided by Mr. John G. Williams of Nairobi. For kindly comparing one puzzling specimen with material in the large van Someren collections in the Field Museum of Natural History, grateful acknowledgment is made to M. A. Traylor of that institution. The main acknowledgment the Museum wishes to make is to the generous sponsor and director, Los Angeles County Museum of Natural History. 2Senior Curator of Ornithology, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 130 Figure 1 . Map of the Samburu District, Kenya, showing collecting localities. 1967 1966 Cheney Expedition to Kenya 3 leader of the expedition, Mr. William J. Cheney, a long time friend and bene- factor of the Museum. He made possible the trip, the results of which have enriched the Museum and have contributed to the advancement of our knowl- edge of the bird life of a seldom visited portion of eastern equatorial Africa. Mr. Cheney and his mother, Mrs. Eva Cheney, not only assumed the cost of the expedition but also provided the funds for the publication of this report on its work. Thanks are due the Chief Game Warden of Kenya, Mr. David W. J. Brown, for the granting of the official permits necessary for the accomplish- ment of our goals. Special thanks are due Major R. T. Elliott, Divisional Game Warden, Northern Frontier Division, Maralal, for his hospitality and assist- ance to the expedition while it was working in his administrative area. The junior author made the present collection, supplied the field notes and the photographs used in this paper. The senior author is responsible for the identifications, systematic notes, and the brief historical introduction. Each of the two authors read and approved the parts written by the other. The present collection is the most extensive yet reported from the area, and, as such, fills in locality records for many species. As might have been anticipated, these provide no surprising extensions of known ranges, but are briefly presented in this paper as supplemental to previously published data. The first explorers to traverse the area with which this report is concerned were the party of Count Samuel Teleki, who (1887-1888) proceeded north- wards from Kilimanjaro, past Mt. Kenya, and up along the eastern side of the Rift Valley, eventually discovering Lakes Rudolf and Stefanie. The count’s companion, Ludwig von Hohnel, published in 1894 a two volume narrative account of this trip, but it contains only a few casual references to the birds seen or collected. No specimen records of Teleki’s or Hohnel’s are extant so far as published evidence goes, if, indeed, any were collected. In 1912 the Childs Frick expedition, with Edgar A. Mearns as chief col- lector, skirted the eastern edge of the area on its way south from Ethiopia to central Kenya and to Nairobi. They collected assiduously in the areas south- east of Lake Rudolf, recording 17 species in one day on the southern slopes of Mt. Nyiru (called Nyero Mts. on its labels), others at South Horr, just to the east of Mt. Nyiru, and still others in the Indunumara and Endoto Moun- tains; all in about 10 days in July, 1912. Its material was included in the detailed, complete report on its entire traverse of Ethiopia and Kenya, by Friedmann (1930, 1937). One other collection, never published on and of unknown size and cover- age, was made in the Samburu district in 1911-1912 by A. B. Percival, then of the Kenya game department. In all probability, Percival collected only casually while passing through the area. Jackson (1938, pp. 880, 901, 1311, 1342) lists four species of birds taken by Percival on Mt. Nyiru: Stelgidocichla latirostris eugenia, Dioptrornis fischeri, Zosterops virens kaffensis, and Cin- 4 Contributions in Science No. 130 Figure 2. Forest cover in vicinity of expedition camp at the southwest foot of the Karissia Hills, N.F.D., Kenya, June 1966. Figure 3. Forest cover on the crest of the Karissia Hills, N.F.D., Kenya, June 1966, 1967 1966 Cheney Expedition to Kenya 5 Figure 4. View of the west face of Mt. Nyiru from the vicinity of the expedition camp at Turn. June 1966. Figure 5. View of plains area to the west from the base of Mt. Nyiru, in the vicinity of Turn. June 1966. 6 Contributions in Science No. 130 nyris mediocris mediocris (nomenclature given as in Jackson’s book). Of these the bulbul and the sunbird are not represented in the present collection. Dioptrornis fischeri is the same as the Malaenornis chocolafina fischeri of this paper, and Zoster ops virens kaffensis is probably our Zoster ops senegalensis jacksoni. Although much ornithological work has been done in the Mt. Kenya area and along the Northern Guaso Nyiro River, to the south of our present region, by van Someren, Lonnberg, and others, and at Marsabit to the north of it, no collections other than of casual, odd specimens appear to have been made in the Samburu District. In our treatment of the birds collected we have followed for the most part the most recent check list of C. M. N. White as far as nomenclature is concerned. Where we have thought it called for, we have recorded taxonomic comments and observations not necessarily in complete accord with White’s conclusions. In this way it is hoped that the present list may be used most easily by other students of East African birds. ANNOTATED LIST OF SPECIMENS Family Podicipedidae Podiceps ruficollis capensis Salvador! The African little grebe is widely distributed from Senegal to Ethiopia, south to South Africa, wherever there are lakes or permanent streams. One adult male, with somewhat enlarged gonads, in good breeding plumage, was collected in the Karissia Hills on June 19. The locality is an addition to pre- viously reported occurrences, but does not extend the known range of this bird. Several grebes of this species with downy young were observed on the reservoir formed by the dam across the stream at Bauwa in the Karissia Hills between June 5 and 20. Family Ardeidae Egret ta intermedia brachyrhyncha (Brehm) The African yellow-billed egret is represented by one specimen, an adult female, with slightly enlarged ovary, collected at Lake Kisima on June 17. This is a new, but not surprising, locality record for this bird. Family Anatidae Alopoehen aegypiiaca (Linnaeus) The Egyptian goose occurs throughout most of Africa. An adult male was obtained in the Karissia Hills on June 14. A common species at Bauwa in the Karissia Hills where one pair with five downy young were noted on June 15. Numerous on Lake Kisima on June 17. 1967 1966 Cheney Expedition to Kenya 7 Anas sparsa leucostigma Riippell The African black duck was met with in the Karissia Hills, where a male was collected on June 9. This bird is in an advanced stage of molt; it has lost all of its old remiges and the new ones are just sprouting, leaving the bird entirely incapable of flight. This duck is found chiefly in wooded streams in the uplands of Kenya, where it is apt to remain hidden in the vegetation along the banks. This causes it to seem less numerous than it probably is. Anas erythrorhynchus Gmelin The red-bill is one of the commonest of east African ducks. Three speci- mens in good plumage were collected as follows: Karissia Hills, June 9, one adult male; 15 miles south of Maralal, June 17, one adult of each sex. Family Accipitridae Aegypius tracheliotus (Forster) An adult female weighing 14 lbs. (6.36 kg.) was obtained on the open plains area, 10 miles south of Maralal, on June 9, 1966. On June 10, 1966, an adult male weighing 16 lbs. (7.26 kg.) was secured in the same general area, 15 miles south of Maralal. Neither bird showed any evidence of being in breeding condition. It is questionable whether Aegypius and Torgos are congeneric as held by White, but for the present we will go along with this arrangement. There are many basic differences between Torgos and Aegypius, including the size and conformation of the brain. The ratio of numbers of these large lappet-faced vultures to other vultures in the Maralal-Mt. Nyiru area was noted to be roughly one pair of the former to any large assemblage of vultures at a kill. Trigonoceps occipitalis (Burchell) Two specimens, one of each sex, of the white-headed vulture were ob- tained in the plains area at the south end of Mt. Nyiru, on June 23, 1966. The two are quite dissimilar, as the adult female has the large inner secondaries and broad edges on the median upper wing coverts white, while the subadult male has the inner secondaries dark silvery grayish and has largely lost the pale edges of the coverts through abrasion. Each of the above specimens weighed 10 lbs. (4.54 kg.). This vulture is generally stated to be relatively uncommon although widely distributed throughout the drier portions of Kenya (Jackson 1938: 139; Lonnberg, 1911: 54). Sharpe (1931: 10) found it on Marsabit Moun- tain; Mt. Nyiru seems to be a new, if not a surprising locality for this bird. White-headed vultures were frequently observed in the Karissia Hills- Maralal area, but no specimens were collected. The relative abundance of this species of vulture in the area was approximately one pair per gathering of 8 Contributions in Science No. 130 vultures at a given kill. Each time a zebra was killed for leopard bait in the Maralal area, a pair of Trigonoceps and a pair of Aegypius would join the throngs of Gyps. Gyps hengalensis africanus Salvador! Two specimens, representing two very different plumages, of this vulture were collected 10 miles south of Maralal, June 9, 1966, an adult female (weighing 10 lbs. or 4.54 kg.) and an immature of the same sex (7 lbs. or 3.18 kg.) . In treating the African white-backed vulture as a race of the Asiatic hengalensis we are following White (1965: 41 ) . This is a common vulture throughout Kenya and its occurrence near Maralal, though not previously reported, is not unexpected. White-backed vultures are exceedingly common in the Maralal area and, combined with Ruppeli’s griffon, make up the bulk of the vulture population of northern Kenya. It was interesting to note that these gregarious vultures have apparently become conditioned to the food opportunities afforded by hunting safaris, as large numbers of vultures will follow above and behind a hunting vehicle heading out across the plain. When the vehicle stops, the vul- tures will begin circling above in soaring flight. A hunter with rifle moving out from the vehicle on foot will cause many of the circling birds to begin their descent, apparently in anticipation of immediate food. Gyps ruppellii rupellii Brehm An adult male weighing 12 lbs. (5.45 kg.) was obtained on the open plains area, 10 miles south of Maralal, on June 9, 1966. Ruppell griffons are extremely common in the Maralal-Mt. Nyiru plains areas. Their numbers were roughly equal to those of the preceding species, Gyps hengalensis. Necrosyrtes monachus (Temminck) The hooded vulture is represented in the collection by an adult male taken in the Karissia Hills, June 17, 1966. It had small testes, and is in fairly abraded plumage; this is especially true of the retrices and inner greater upper wing coverts. In using a binomial for this bird we follow C. M. N. White (1965: 42), who pointed out that the change in size of these birds from the smallest West African examples to the largest East and South African ones is so gradual that it is difficult to know where, if at all, to draw a line. It is true that Mack- worth-Praed and Grant (1957: 137) consider the eastern and southern birds as forming the race pileatus, to which the present example would belong. The hooded vulture is one of the commonest species of its group in many areas of Kenya. Although White (1965) considers Necrosyrtes and Neophron to be con- generic, we hold that Necrosyrtes should be retained as a valid genus, as it differs very profoundly from the Egyptian vulture. Cranial casts of the brain 1967 1966 Cheney Expedition to Kenya 9 of the two genera show them to differ decidedly in shape. The external nares of Necrosyrtes are not nearly as large as those of Neophron. There are many differences in plumage texture and composition between the two forms. In Necrosyrtes the feathers of the occiput and nape are short and velvet-like while those of Neophron are narrow, elongate and pointed. The chest of Neophron is covered with normal contour feathers while in Necrosyrtes a pronounced crop patch is developed with a prominent border of white down feathers. The tail of Neophron has a sharply graduated tail of 14 feathers while that of Necrosyrtes is almost square with 12 feathers. Hooded vultures were well represented in all large vulture gatherings in the Maralal-Mt. Nyiru plains areas and several birds were always to be found about our camp areas, in the company of tawny eagles ( Aquila rapax ) . Perch- ing in the trees about the periphery of the camps, these scavengers were always on the alert for carcasses discarded during specimen preparation. Neophron percnopterus percnopterus (Linnaeus) One adult male Egyptian vulture was obtained in the plains at the west foot of the Karissia Hills, June 12, 1966. This species is chiefly a bird of the interior highlands along the Rift Valley in Kenya. Although a common bird in many parts of East Africa, the Egyptian vulture proved to be the least common of the six species of vultures encountered in the Maralal-Karissia Hills area. Circaetus cinereus Vieillot An adult female brown harrier eagle was collected on the west base of Mt. Nyiru, June 23, 1966. It was in non-breeding state, with small ovary, and was in fairly worn plumage. Jackson (1938: 190) gave a previous record from the Samburu district, but stated that the species is rather rare in the interior of Kenya. This individual was the only Circaetus observed during the course of the field work. In flight the bird appeared perfectly normal, but examination of the sternum showed evidence of a previous injury that had resulted in extensive fracturing and subsequent mending of the bone tissue. Accipiter badius sphenurus Riippell This small hawk is represented by one adult female, taken at the west base of Mt. Nyiru, June 23, 1966. The bird was in non-breeding condition, and in good plumage. Melierax poliopterus Cabanis The pale chanting goshawk was met with on June 24, on the west base of Mt. Nyiru, when one adult male and two adult females, all with small gonads, were collected. These three are definitely poliopterus, but the male has a very few dusky cross bars on the white upper tail coverts, suggesting a variation in the direction of the related and sympatric M. metabates. Similarly, one of the females shows some freckling on the inner webs of a few of the upper wing 10 Contributions in Science No. 130 coverts, but nothing like the degree to which this character is developed in M. metabates. Chanting goshawks were not met with in the Karissia Hills area, but observations were made on the harrier hawk, Polyboroides typus. A single individual of this species was observed in flight, vigorously pursued by large numbers of glossy starlings of several species. Disregarding the star- lings, this large hawk was repeatedly observed to raid the nesting colonies of the grey-headed social weaver ( Pseudonigrita arnaudis) . The hawk was ob- served to grasp the nest in its talons and pull it apart by hanging upside down and beating the air with its wings. Frequent movements of the hawk’s head towards the nest were made, but observations was too distant from the scene of predation to note whether or not any adult or nestling weavers were taken. Buteo rufofuscus augur (Riippell) The augur buzzard is represented by two adult specimens, a female taken in the Karissia Hills, June 9, and a male taken 5 miles west of Maralal, June 15. Both birds were in non-breeding state, with small gonads. This is a common, widely distributed bird in the interior of Kenya and numerous individuals were observed in the Maralal area. Aquila rapax belisarius (Levaillant) Two examples of the tawny eagle were obtained, as follows: 15 miles south of Maralal, June 12, one female in the “usual” grayish tawny plumage; Karissia Hills, June 18, one adult male in the dark plumage phase. The female is in very worn plumage. As was shown in a detailed account of the plumage stages and phases of this eagle by Friedmann (1930: 58-62), the very dark birds are the oldest ones, possibly 5 or 6 years old. The present male is very dark on the head, throat, and breast; less so on the wings and upper parts of the body, but still dark, and dark tawny brown on the abdomen. The tawny eagle is the commonest eagle in the interior of Kenya, where it is fairly well distributed. The present specimens add new, but not surprising, locality records for it. Tawny eagles proved to be great scavengers about our field camps, where they took up residence and fed on the discarded bird car- casses cast out each day. As many as eight individuals of this species were noted at one time at the Karissia Hills camp and always in the company of hooded vultures ( Necrosyrtes monachus) . Family Phasianidae Francolinus sephaena grantii Hartlaub Grant’s crested francolin is a common, widely distributed bird of the open country of eastern Africa from southern Sudan and Ethiopia to central Tan- zania. Two specimens were procured by the expedition, a non-breeding female in the plains area 15 miles south of Maralal, June 4, and a breeding male in the Karissia Hills, June 12. 1967 1966 Cheney Expedition to Kenya 11 Francolinus leucoscepus infuscatus (Cabanis) The yellow-throated francolin occurs over much of the open, drier parts of eastern Africa from Ethiopia to Tanzania. A female collected 15 miles south of Maralal, June 7, and a male in the same place June 15. Both birds were in full breeding condition; the hen was marked as “laying.” The male is larger and more deeply colored, with dark mahogany brown on the ventral feathers; the female has the white shaft streaks on the feathers of the upper part broader and the malar band of feathers less lined with dusky gray; other- wise the two are alike. Coturnix delegorguei delegorguei Delegorgue Two females of the wide-ranging harlequin quail were collected, one on June 7, 15 miles south of Maralal, and the other on June 11, in the Karissia Hills. Both birds had slightly enlarged ovaries; both are in fresh plumage. Family Otididae Eupodotis ruficrista gindiana Oustalet The bulf-breasted florican was met with at the west base of Mt. Nyiru, on June 25, when one female was collected. The bird had the ovary slightly enlarged. Mt. Nyiru is a new locality for this bird, and while not extending its known range, helps to fill in the still imperfect data on its occurrence. Family Burhinidae Burhinus capensis maculosus (Temminck) The South African dikkop ranges from the Cape Province north to Kenya, changing in northern Kenya to a tawnier race maculosus. Two male specimens obtained in the Karissia Hills, June 11 and 12, belong to the race maculosus, which had been recorded previously from Horr, Marsabit, and other localities not too far from the Karissia Hills. Both specimens are some- what abraded, and both were in non-breeding condition. Family Charadriidae Vanellus armatus (Burchell) One adult male blacksmith plover was collected 15 miles south of Maralal, June 6. It is in slightly abraded plumage, and shows signs of molt in the tail, the outermost rectrices being only an inch in length and still enclosed basally in their sheaths. The bird had the testes somewhat enlarged. Maralal must be about the northern limit of the range of this plover; Jackson (1938: 355) considered it to occur only in southern Kenya, south to Natal, and gave no locality record north of Naivasha, Elmenteita, and Nakuru. Mackworth- Praed and Grant (1957: 358) wrote that in Kenya it was most abundant in the Rift Valley, and noted its breeding there from April to August. The 12 Contributions in Science No. 130 present specimen, judging by its plumage and by its gonadal state, was probably through breeding, and was therefore one of the early breeders. We follow White (1965: 120) in considering the genus Hoplopterus, formerly used for this bird, as part of Vanellus. Vanellus coronatus coronatus (Boddaert) The crowned lapwing, a common bird of the fairly dry, short grass area of Kenya, was collected in two localities. On June 10 an adult male, with somewhat enlarged gonads, was taken 15 miles south of Maralal; on June 12 another male and a female with large ovarian eggs were collected in the open plains at the foot of the Karissia Hills. All three specimens are in somewhat abraded plumage. This species was formerly placed in the genus Stephanibyx, which White (1965: 122) later considered as part of the broader group Vanellus. Charadrius pecuarius pecuarius Temminck Kittlitz’s sand plover is a widely distributed bird over much of Africa, in dry open places. The expedition obtained one adult male, with enlarged testes, at Lake Kisima, on June 17. The specimen is in abraded plumage. Charadrius tricollaris tricollaris Vieillot The three-banded plover is a common bird of the mud flats, beaches, and open areas near water, throughout much of Africa. It is represented in the present collection by the following three specimens: June 4, Karissia Hills, one female with slightly enlarged ovary, one male with enlarged testes; June 6, plains area 15 miles south of Maralal, one female with enlarged ovary. Family Glareolidae Cursorius temminckii Swainson Temminck’s courser is a bird of the open, short grass country throughout Kenya. On June 4, two specimens, one of each sex, were obtained in the plains area 15 miles south of Maralal; both birds were in breeding condition. Family Columbidae Columba guinea guinea Linnaeus One adult male speckled pigeon, with considerably enlarged testes, was obtained at the west base of Mt. Nyiru on June 26. This is a bird of the open country where it goes about in small flocks. According to Peter Saw, the pro- fessional hunter with our field party, the speckled pigeon is known to form fairly large flocks in the Mt. Nyiru area where it is often sought as a game bird by hunting safaris. Streptopelia lugens lugens (Riippell) This dark pigeon is represented by two specimens in the collection; an 1967 1966 Cheney Expedition to Kenya 13 immature female taken in open plains country 15 miles south of Maralal, on June 10, and a non-breeding adult male collected in the Karissia Hills, June 11. White (1965: 156) considered this pigeon to be a bird of montane forests, but the present immature female, found in open country, suggests a broader ecological range of habitat for it. Streptopelia semitorquata (Rlippell ) In naming the specimens of the red-eyed dove binomially we are follow- ing White (1965: 156-157) who found the proposed races were not well enough differentiated for separate recognition. Three examples were collected by the expedition, all in the Karissia Hills, a female with small ovary on June 3, another with somewhat swollen ovary on June 14, and a male in breeding condition on June 4. Streptopelia senegalensis senegalensis (Linnaeus) The laughing dove, a widely distributed and common bird over much of Africa south of the Sahara, except in forested areas, was met with on the west base of Mt. Nyiru, where an adult male with greatly enlarged testes was collected on June 26. Oena capensis (Linnaeus) This long-tailed little dove is represented by one specimen, an adult female with slightly enlarged ovary, taken at Lake Kisima on June 17. Turtur tympanistria (Temminck) The tambourine dove occurs throughout Kenya in wooded areas. The expedition obtained an adult male in breeding condition in the Karissia Hills on June 10. We follow White (1965: 161) in treating this bird binomially. Turtur chalcospilos (Wagler) The emerald spotted dove is a woodland bird of very wide distribution in Africa. Three specimens were obtained, as follows: Karissia Hills, June 8, one adult male with enlarged testes; west base of Mt. Nyiru, June 22, one adult of each sex with somewhat enlarged gonads. Family Musophagidae Tauraco hartlaubi (Fischer and Reichenow) Hartlaub’s turaco is a bird of the highland forests of central and southern Kenya, eastern Uganda, and northern Tanzania. It has been found as far north as Marsabit Mountain (Sharpe, 1931: 99). Judging by the fact that the expedition obtained six specimens in the Karissia Hills, June 2 to 14, the bird must be as common there as it is on Marsabit and on Mt. Elgon. Five of the specimens are females, none in breeding condition and one of them a young bird, and the remaining one is an adult male with somewhat enlarged testes. 14 Contributions in Science No. 130 Turacos with a call similar to that of T. hartlaubi were heard on the steep, heavily-forested slopes of Mt. Nyiru, but no specimens were secured. Corythaixoides leucogaster (Riippell) The white-breasted go-away-bird is a denizen of the open, semi-arid plains and scrub country of Kenya. On the west base of Mt. Nyiru, June 21 to 25, a series of 3 adult males and 2 adult females was collected. Some of them showed gonadal enlargement while others did not. A common species in the area about our Mt. Nyiru camp. Family Cuculidae Cuculus solitarius solitarius Stephens The red-chested cuckoo was met with in the Karissia Hills, where a non- breeding female was taken on June 12 and a male, with slight gonadal enlarge- ment, on June 15. The female shows signs of active molt in the remiges, and has many worn feathers on the upperparts; the male shows no sign of ecdysis. Pitman (1964: 140) recorded an egg of this cuckoo taken at Baragoi, about 50 miles north of the Karissia Hills, on December 20. Cuculus clamosus clamosus Latham The name of the black cuckoo is once more clamosus since cafer Lich- tenstein has ruled a nomen rejectum by the International Commission on Zoological Nomenclature. The expedition collected an adult male, with some testicular enlargement, in the Karissia Hills, June 19. The fact that this bird, not a breeder but also not a migrant, judging by its testes, was collected at the same place and within a fortnight of the specimens of the next form, adds to, rather than clarifies the puzzling nature of the inter-relations of these cuckoos. This specimen has one very worn outer rectrix, apparently left over from an earlier plumage. Cuculus clamosus gabonensis Lafresnaye White (1965: 184) has combined jacksoni and mabirae with gabonen- sis, and we call our specimens by the last name to conform with his check list. However, it is not yet clear whether this is the correct solution of a puzzling series of bird plumages. The detailed discussion in earlier studies (Friedmann, 1930: 263-266; Chapin, 1939: 192) and the uncertainty expressed by Jackson (1938: 488-489) make it advisable to point out that if further study with additional specimen data should necessitate rearranging the birds presently grouped together as gabonensis, and should reinstate some of the races, for- merly recognized, our present examples would be jacksoni. They lack the pale ground color on the abdomen found in typical gabonensis and mabirae. Three examples, all males, and all like the type of jacksoni, were obtained on June 3 in the Karissia Hills. All had small testes and all are fairly similar, although 1967 1966 Cheney Expedition to Kenya 15 varying in the darkness of the ventral barrings and in the width of the white bars on the under tail coverts. Chrysococcyx cupreus cupreus (Shaw) The yellow-bellied emerald cuckoo occurs, chiefly in forest but also in scrub, country, throughout eastern Africa. Two adult males, both with some- what enlarged gonads, were collected in the Karissia Hills on June 15. The Karissia Hills are a new, but hardly surprising, locality record for this bird. We are following White (1965: 187) in considering sharpei a synonym of cupreus, although his action is subject to reconsideration, as the majority of previous authors considered the two distinct. Centropus superciliosus superciliosus Hemprich and Ehrenberg One adult female white-browed coucal was obtained at Pali Pali Hill, south of Karatina, June 1, when it was hit by the car. This terrestrial cuckoo is a common and widely distributed bird over much of Kenya. The above mentioned specimen was secured enroute to the Samburu district, but the species was not encountered in either of our collecting areas. Family Strigidae Bubo lacteus (Temminck) Verreaux’s eagle owl is represented in the collection by an adult female, taken in the Karissia Hills, on June 2. Its stomach contents showed it had been feeding on large coleoptera, a strange diet for so large an owl. Other observers have recorded snakes and lizards, but also caterpillars as parts of its usual food. The Karissia Hills adds a new “station” for this large owl. The above mentioned specimen was one bird of a pair that roosted in the large fever trees in the immediate vicinity of the Karissia camp. Family Caprimulgidae Caprimulgus frenatus Salvadori White (1965: 204) treated frenatus as a full species and not as a race of pectoralis, and we are here following his decision. Two specimens are referred to this nightjar, an adult “female” taken in the Karissia Hills, June 12, and another female in the same place on June 4. The first specimen may be wrongly sexed, as it has the large remigial spots pure white, untouched with buffy, and has very broad white terminal areas on the outer rectrices. The other specimen has the remigial spots tinged with rusty buff and has no white tips to the rec- trices, the terminal portions of which feathers are grayish sandy buff mottled and blurred with fuscous. The bird assumed to be a male agrees very closely with an undoubted male from “Mile 40” on the Nairobi-Magadi Road; the other fits very well with a female from Nairobi but has the terminal area of 16 Contributions in Science No. 130 the outermost pair of rectrices more clouded and mottled with fuscous, less plain sandy bull than the Nairobi bird. The descriptions of the female plumage given by Mackworth-Praed and Grant (1957: 667) and by Jackson (1938: 651) do not fit the present female, as they state that the terminal white tail areas are present in both sexes; but in this they seem to be in error. The speci- men with the large white tail marks, our probable male, is molting in the wings, as two of the remiges of the left wing are very short and basally are still enclosed in their sheaths from which they are expanding terminally. Family Apodidae A pus niansae niansae (Reichenow) Two specimens of this swift were collected at the west base of Mt. Nyiru, June 24, one of each sex. They agree with a series from Mt. Moroto, north- eastern Uganda, and from Naivasha, Kenya. The above specimens were shot on the wing in the late afternoon when large numbers of swifts, presumably of this species, were feeding over the scrub forest at the west base to Mt. Nyiru. Family Coliidae Colius striatus kikuyuensis van Someren The speckled mousebird is one of the commonest birds of most of Africa; the present subspecies ranges from northern Tanzania north at least to Marsa- bit. In the Karissia Hills, June 2 and 3, a series of 4 males and 2 females was collected, all but 2 of which were marked as in breeding condition. Mouse- birds of this species were exceedingly abundant in the Karissia Hills, but were absent in the Mt. Nyiru area, being replaced by the following species. Colius macrourus pulcher Neumann Two male blue-naped mousebirds were collected on the west base of Mt. Nyiru, June 26. Both had the gonads somewhat enlarged; one of the birds lacks the long rectrices, probably due to the shot. Mt. Nyiru appears to be an additional locality for this bird. Mousebirds were noted as relatively uncom - mon in the Mt. Nyiru area. Family Trogonidae A paloderma narina narina (Stephens) Narina’s trogon is a bird of the evergreen forest throughout much of eastern Africa, but its range is discontinuous because of the spotty distribution of suitable habitat. The expedition found it to be common in the Karissia Hills, where 5 male specimens were obtained between June 6 and 12. All the birds showed gonadal swelling. All of the above mentioned birds were located by their characteristic calls and no females were seen. 1967 1966 Cheney Expedition to Kenya 17 Family Alcedinidae Halcyon chelicuti chelicuti (Stanley) This is one of the wide-ranging, common kingfishers of the open country of Africa. One adult female was obtained on June 7, 15 miles south of Maralal; the bird was in non-breeding condition. Halcyon leucocephala leucocephala (Muller) The gray-headed kingfisher occurs over much of Africa except in treeless areas. The expedition met with it on the west base of Mt. Nyiru, where one male with somewhat enlarged gonads was collected on June 23, and one non- breeding female on June 26. Family Meropidae Merops pusillus meridionalis (Sharpe) This is one of the ubiquitous birds of the open country of eastern Africa. In the plains area 15 miles south of Maralal the expedition collected one adult female on June 4. It showed slight gonadal development. We follow White (1965: 232) in merging the genus Melittophagus with Merops. Merops lafresnayei oreobates (Sharpe) The cinnamon-chested bee-eater is a bird of the bush country of Kenya, usually at altitudes of from 6000 to 8000 feet. It was common in the Karissia Hills where the expedition collected 2 adult males and 2 immature birds of each sex, on June 14 and 15. In the Karissia Hills this bee-eater was encoun- tered only in heavy forest where it perched in numbers on the tops of large cedars along a stream course. Family Coraciidae Coracias caudata caudata Linnaeus The lilac-breasted roller was met with in the open plains 15 miles south of Maralal on June 7 and 10, on each of which days one male was obtained. With the deep lilac color extending well down over the breast, these specimens are clearly typical caudata and show no variation in the direction of lord of Ethiopia and northeastern Kenya, in which race that color is restricted to the chin and throat. Coracias naevia naevia Daudin The rufous-crowned roller is represented in the collection by an adult male taken on the west base of Mt. Nyiru, June 26. It had the testes slightly enlarged, and is in somewhat abraded plumage. This species is widely dis- tributed over much of eastern Africa; Mt. Nyiru is an additional, but not surprising, “station” for it. 18 Contributions in Science No. 130 Eurystomus glaucurus afer (Latham) The broad-billed roller is found over a good part of eastern Africa. One adult female specimen was collected in the Karissia Hills, on June 2. It was in non-breeding state. Family Upupidae Phoeniculus granti (Neumann) In calling our single, youngish, specimen P. granti, and not P. purpureus granti, we are conforming to the arrangement given by White (1965: 241). The present bird, a young male, taken in the plains area 15 miles south of Maralal, June 7, has the chin and throat dusky grayish fawn, and the black bill somewhat shorter (25 mm.) than birds in full adult plumage. Phoeniculus bollei jacksoni (Sharpe) This race of the white-headed kakelaar occurs in the highlands of Kenya west to the eastern Congo and north to the southern Sudan. The expedition obtained 5 male specimens in the Karissia Hills, June 2 to 12. Of these, 3 showed no signs of gonadal enlargement, one showed slight signs of it, and the last had no comment on its label. Phoeniculus minor cabanisi (Defilippi) This smallish kakelaar is a bird of the dry open country. The expedition collected one non-breeding male on the west base of Mt. Nyiru, on June 22. We follow White (1965: 243) in combining the genus Rhinopomastus with the broader Phoeniculus. Family Bucerotidae Tockus erythrorhynchus erythrorhynchus (Temminck) The red-billed hornbill has a wide range in the thorn bush country of Africa. The expedition collected 4 specimens on the west base of Mt. Nyiru, June 21 to 25; 2 adult males with somewhat enlarged testes, one female in non-breeding state, and 1 immature female. All the birds show signs of molt, especially in the tail, and all have some faded old feathers still present in the upperparts. Tockus deckeni (Cabanis) Von der Decken’s hornbill is represented in the collection by a single non-breeding female collected on the west base of Mt. Nyiru on June 25. The rectrices are not yet fully grown and are still enclosed in their sheaths basally. Tockus jacksoni O. Grant Jackson’s hornbill is not listed in White’s 1965 check list, and it is to be assumed that he considered it the same as deckeni. One immature female was collected on the west base of Mt. Nyiru on June 25, apparently together with 1967 1966 Cheney Expedition to Kenya 19 the specimen of deckeni listed above. However, this bird has numerous con- spicuous white spots on the upper wing coverts, which are absent in the other- wise very similar deckeni. The discussion of these two remarkably similar, sympatric species given in an earlier study (Friedmann, 1930, 420-425) still seems to be the most informative appraisal of them, and we prefer to follow the conclusions there reached, and since accepted by Mackworth-Praed and Grant (1957: 621). Tockus flavirostris flavirostris (Ruppell) The yellow-billed hornbill occurs widely in the acacia savannas of Kenya. Three examples were obtained by the expedition on the west base of Mt. Nyiru, on June 25, 2 males and 1 female, all with somewhat enlarged gonads. Both of the males show signs of active molt in the remiges. An adult male was observed actively feeding a nesting female in a dead snag in the center of our camp area. The bird would feed the sealed-in female several times a day and the food items were often noted as large green mantids. A young fledgling of this species was purchased alive from a Samburu warrior on June 26 at the Mt. Nyiru camp and kept as a pet. At the time the bird was acquired, it had its full complement of remiges, but no tail had developed. Family Capitonidae Lybius lacrymosus lacrymosus (Cabanis) The spotted-flanked barbet, a common bird of much of Kenya and adja- cent areas, was met with on the west base of Mt. Nyiru, on June 22 and 23, when 6 adults, 3 of each sex, were collected. We follow White (1965: 259) in considering the genus Tricholaema, formerly used for these birds, as part of Lybius. Lybius leucomelas diadematus (Heuglin) The red-fronted barbet was met with and collected at two localities: plains area 15 miles south of Maralal, June 4, one female with slight gonadal enlarge- ment; west base of Mt. Nyiru, June 24 and 26, two females, one with some ovarian enlargement, the other not. Pogoniulus leucomystax (Sharpe) This little tinker-bird ranges from central Kenya south to south-central Tanzania. One female, in non-breeding state, was obtained in the Karissia Hills on June 6. It agrees with several from Tanzania in size and coloration. The Karissia Hills must be near the northern limit of the range of this bird. Pogoniulus pusillus affinis (Reichenow) The red-fronted tinker-bird is represented in the present collection by 1 male and 5 females, all taken in the Karissia Hills. June 2, 3 and 4. The 20 Contributions in Science No. 130 locality is an additional, but not a surprising, “station” for this tiny barbet, which has an extensive range in the dry bush and acacia country of Kenya. These small barbets fed extensively on a red-berried mistletoe in the Karissia Hills area. Pogoniulus bilineatus jacksoni (Sharpe) This race of the golden-rumped tinker-bird occurs in the highlands of Kenya, chiefly west of the Rift Valley. Two adult specimens, one of each sex, were collected in the Karissia Hills on June 15. These examples agree with jacksoni, not with alius, the form found in the areas east of the Rift Valley in Kenya, although the difference between the two is slight. Trachyphonus darnaudii bohmi Fischer and Reichenow This race of Darnaud’s barbet, characterized by having the crown plain black with no terminal orange or red spots on the feathers, ranges farther to the west in northern Kenya than the statements of either Mackworth-Praed and Grant (1957: 735-736) or White (1965: 273) indicate. The present series of 4 specimens from the west base of Mt. Nyiru, June 25 and 26, are clearly bohmi and not typical darnaudii. The birds collected were in non- breeding condition, and are in somewhat abraded plumage. Trachyphonus erythrocephalus versicolor Hartlaub This subspecies of the red and yellow barbet, characterized by the yellow forehead and superciliary stripes, is represented in the collection by one adult male, taken on the west base of Mt. Nyiru on June 23. It was in non-breeding condition and is in a molting stage, especially noticeable on the chin and throat. Family Indicatoridae Indicator variegatus Lesson The scaly-throated honey-guide is treated binomially to conform with White’s list (1965: 274). It is represented in the present collection by one adult male taken in the Karissia Hills on June 3. The bird had the testes con- siderably enlarged, and is in fairly worn plumage. White is correct in saying that the dimensional characters by which juhaensis is characterized form a cline, but this in itself is not impossible of subdivision as the smaller form is, as far as known, a coastal and subcoastal population. Further studies may yet reestablish the usefulness of this racial recognition. Indicator indicator (Sparrman) The greater honey-guide was met with in the Karissia Hills, June 3, when a non-breeding female adult in good plumage was collected. The species is widely distributed over most of sub-Saharan Africa. Greater honey-guides 1967 1966 Cheney Expedition to Kenya 21 were noted as common in the Mt. Nyiru area and although no specimens were saved as skins, two specimens were liquid preserved for anatomical studies. Indicator minor minor Stephens The lesser honey-guide is widely distributed over much of Africa. One adult, non-breeding female was collected at the west base of Mt. Nyiru on June 22. The specimen is without a head, as that part was preserved in spirits for anatomical study of the olfactory tract. Family Picidae Campethera nubica nubica (Boddaert) The Nubian woodpecker, a common bird in most parts of Kenya, was collected on the west base of Mt. Nyiru, on June 24; one adult male, with enlarged testes, and two adult females with slight gonadal development. Dendropicos fuscescens hemprichii (Ehrenberg) Two specimens of the cardinal woodpecker are referred to the subspecies hemprichii. They are a non-breeding adult female taken 15 miles south of Maralal on June 4, and an adult male (by plumage, although marked female on the label) collected on the west base of Mt. Nyiru, June 26. The reason for calling them hemprichii and not lepidus is because they have the mantle very distinctly barred, and have little (male) or no (female) greenish or yellowish on the mantle and the underparts. Judging by the maps of the ranges of these two races in Mackworth-Praed and Grant: 1957: 762,763) Mt. Nyiru must be about where the two come together. In their dimensions (wing 80 in the male, 84 mm. in the female), they are large for hemprichii and less than the average for lepidus. Dendrocopos obsoletus ingens (Hartert) The little brown-backed woodpecker is represented by one adult female taken in the Karissia Hills, on June 2. It is in very worn plumage but has some new feathers coming in on the upperparts. This is one of the less frequently observed and collected of the east African woodpeckers, although it is widely distributed in the acacia and bush country. Karissia Hills is a new locality for it. Thripias namaquus schoensis (Ruppell) This race of the bearded woodpecker, characterized by having the black marking of the auriculars extending postero-ventrally so that it separates the white malar stripes from the white post-ocular area, ranges from Ethiopia to northern Kenya and west to the Ubangi-Shari area. Two specimens were obtained, as follows: Karissia Hills, June 3, one male adult; west base of Mt. Nyiru, June 24, one male. Both of these constitute additional, but not surprising locality records. 22 Contributions in Science No. 130 Family Alaudidae Mirafra africanoides intercedens Reichenow One specimen of this widely distributed lark was obtained in the Karissia Hills on June 12, a male with fairly large testes. The bird is darker, deeper chestnut on the auriculars, top of head, and the upperparts generally, and with more extensive blackish fuscous centers on the dorsal feathers, than in most of the comparative material examined, but is approached fairly closely by another male from Kajiado, Rift Valley, Kenya. The Karissia Hills bird is lacking some of its tail feathers, probably as a result of the shot as it shows no sign of molting. Eremopteryx leucopareia (Fischer and Reichenow) Fischer’s sparrow-lark ranges from northern Kenya south across Tan- zania, in fairly dry open country. One female, ovary slightly enlarged, was collected in the Karissia Hills, on June 10. Family Motacillidae Motacilla alba vidua Sundevall One example of the African pied wagtail, an adult male with fairly large testes, was collected in the Karissia Hills, on June 19. This is a common, wide- ranging bird over much of Kenya. Anthus leucophrys subsp. One male example of this pipit was obtained in the Karissia Hills, on June 12. The single specimen obtained is immature and is not fully grown. It is, therefore, not identifiable to subspecies. According to Hall (1961: 262), the races zenkeri and goodsoni intergrade just west of the Karissia Hills. It is possible that the present specimen should be considered as goodsoni, but to label it so would give it a definiteness that might prove misleading. We are indebted to M. A. Traylor of the Field Museum of Natural History for his opinion on our bird and for comparing it with material in the van Someren collections under his care. Family Timaliidae Turdoides plebejus cinereus (Heuglin) Four specimens of the brown babbler were taken in the Karissia Hills on June 12, one adult male with fairly large testes, one female in breeding condi- tion, and two young females. Argya mbiginosa rubiginosa (Ruppell) The rufous chatterer is represented in the collection by two males, in non-breeding state, taken on the west base of Mt. Nyiru on June 24. 1967 1966 Cheney Expedition to Kenya 23 Family Pycnonotidae Pycnonotus barbatus tricolor (Hartlaub) The common, ubiquitous yellow-vented bulbul was collected in the Karis- sia Hills, on June 3 and 4, three adult males and two adult females, all in non- breeding condition. In calling these birds tricolor, and not fayi, we are follow- ing White (1962: 74), who considers the latter name a synonym of the former. The Samburu district is in the area where tricolor and dodsoni intergrade. Phyllastrephus fischeri placidus (Shelley) This race of Fischer’s greenbull inhabits highland forests from Marsabit south across central Kenya and Tanzania to northern Mozambique. It was found to be common in the Karissia Hills, where a series of six adults, three of each sex, were collected June 8 to 14. The birds showed some to very great gonadal enlargement. Nicator chloris gularis Hartlaub and Finsch The east African race of the nicator was met with in the Karissia Hills, where an adult male with greatly enlarged gonads was collected on June 6. The specimen shows signs of active molting in the tail; the plumage of the anterior underparts is much abraded. Family Muscicapidae Muscicapa adusta interposita (van Someren) Three examples of this race of the pygmy flycatcher were obtained in the Karissia Hills, June 18 and 20, one non-breeding adult of each sex and one young bird. Melaenornis chocolatina fischeri (Reichenow) The white-eyed slaty flycatcher, here treated as a race of Melaenornis chocolatinus to conform with the arrangement of White (1963: 15), was common in the Karissia Hills, where one adult male with somewhat swollen testes, one female with slightly enlarged ovary, and three non-breeding females were collected June 2 to 12. The male is in very worn plumage; the females vary somewhat in the degree of feather abrasion, but are in better plumage than the male. Melaenornis pammelaina (Stanley) The shining black flycatcher is represented in the collection by six speci- mens from two localities, as follows: west base Mt. Nyiru, June 24 to 26, one non-breeding adult male and two adult females; Karissia Hills, June 8 and 12, one adult male and one adult female both with somewhat swollen gonads, and one immature female still in the spotted plumage. 24 Contributions in Science No. 130 Bradornis microrhynchus Reichenow One of these small grayish flycatchers was obtained on the west base of Mt. Nyiru, on June 26. It was sexed as a female, but with a query. It is in fairly fresh plumage. Bradornis pallidus subalaris Sharpe The pale flycatcher was met with on June 16 when one adult female and one unsexed young bird were collected 5 miles south of Maralal. The female was laying, and had a large ovum in it. Bat is molitor (Hahn and Kuster) The Kenya chin-spot flycatcher was collected in three localities, as fol- lows: Karissia Hills, June 3 and 7, one of each sex with slightly enlarged gonads; 5 miles west of Maralal, June 16, one adult male, two adult females, ail in non-breeding state; west base of Mt. Nyiru, June 26, one adult male with slightly swollen testes. Terpsiphone viridis ferreti (Guerin) This paradise flycatcher was obtained in two localities: on the west base of Mt. Nyiru, June 22, one non-breeding female; Karissia Hills, June 8 and 1 1, three non-breeding females. Family Turdidae Cercotrichas leucophrys leucoptera (Riippell) The white-browed scrub robin was met with on the west base of Mt. Nyiru where two males were collected on June 24 and 26. One had small gonads, the other showed some testicular swelling. White (1962) has combined Erythro - pygia with Cercotrichas, and we place the present species under the latter generic name to conform to his list. Cichladusa guttata rufipennis Sharpe The single specimen of the spotted morning warbler, obtained on the west base of Mt. Nyiru, June 22, agrees in coloration and in size with two from the lower Tana Valley, and is therefore placed with the race rufipennis. No exam- ples of typical guttata have been available for comparison. Mt. Nyiru must be close to the meeting ground of the two races. The present example shows signs of molt, especially in the tail; it had the ovary slightly enlarged. Myrmecocichla aethiops cryptoleuca Sharpe The Kenya anteater-chat was collected in the Karissia Hills; one adult male with somewhat enlarged testes, June 19. Monticola rufocinerea rufocinerea (Riippell) The lesser rock thrush was met with on Mt. Nyiru, when an immature 1967 1966 Cheney Expedition to Kenya 25 male was collected on June 23 and an adult of the same sex on June 27. We here follow White (1962: 120) in considering tenuis the same as typical rufocinerea, and find the present birds agree in tone of coloration with others from Mt. Moroto. Cossypha heuglini heuglini Hartlaub The white-browed robin chat is widely distributed over much of Eastern Africa. Six adults were obtained in the Karissia Hills, June 4 to 8, four males and two females; the birds vary in their gonadal development from not at all enlarged to fairly much so. In the case of one of the males, the testes were noted as black. In addition to these adults, one immature female was collected, in the same place, on June 4. Cossypha caff r a iolaema Reichenow The Kenya robin-chat ranges from northern Mozambique to the high- lands of Kenya and northwest to the mountains of southern Sudan. One young, unsexed specimen was taken on the west base of Mt. Nyiru on June 21. Turdus abyssinicus abyssinicus Gmelin This thrush is represented by one immature female and one adult male, taken on the crest of the Karissia Hills on June 20. The latter bird had enlarged gonads, and is in abraded plumage, possibly a bird that was about through with breeding for the season. Turdus pelios centralis Reichenow Two specimens of the Kurrichane thrush were obtained in the Karissia Hills, an immature female on June 6, and an adult, non-breeding female on June 12. Family Sylviidae Apalis cinerea cinerea (Sharpe) Two adult males of the grey apalis were obtained in the Karissia Hills on June 20; one had the testes small, the other somewhat enlarged; both are in fairly worn plumage. Apalis flavida malensis Neumann This race of the poorly named black-breasted apalis is represented by one adult of each sex, taken in the Karissia Hills, June 6 and 7. Apalis porphyrolaema Reichenow and Neumann Two adults, one of each sex, of the chestnut-throated apalis were collected in the Karissia Hills, on June 20. Both showed only slight gonadal enlarge- ment; the male is in molt, especially in the tail. These two birds, together with a male from South Kinangop Plateau, Aberdare Mountains, are very slightly 26 Contributions in Science No. 130 grayer, less tinged with brownish, above than others from Kaptagat Forest, Eldoret District, from Ol Kalow, and from Naro Mam, Mt. Kenya. Apalis rufifrons smithii (Sharpe) The red-faced apalis was met with in the plains area at the south end of Mt. Nyiru, where one adult male was taken on June 23. It showed slight gonadal enlargement, and is in somewhat worn plumage. Sylvietta whytii jacksoni Sharpe Two adults, one male, one female, of this race of the red-faced crombec were collected on the west base of Mt. Nyiru on June 22. Both specimens are somewhat abraded and were in non-breeding condition. Cisticola chiniana humilis Madarasz The Kenya highlands race of the rattling cisticola was met with in the Karissia Hills where an adult female was taken on June 2 and a male on June 8. The female had the gonads slightly enlarged, the male more so. Camaroptera brachyura griseigula Sharpe The Taita gray-backed camaroptera was very common in the Karissia Hills, where a series of seven males and three females was collected June 2 to 25. One of the females is a young bird and has the entire upperparts tinged with greenish, and the breast and upper abdomen obscurely streaked or clouded with greenish gray. The adults varied in the state of their gonads from not at all enlarged to fairly large. Camaroptera simplex simplex (Cabanis) Three adult male gray wren warblers were collected on the west base of Mt. Nyiru, June 22, 25 and 26, all in somewhat abraded plumage, and all with the testes somewhat enlarged. Eremomela icteropygialis griseoflava Heuglin One unsexed but adult yellow-bellied eremomela was collected in the plains area 15 miles south of Maralal on June 7. It agrees with the characters of griseoflava, but also suggests those of abdominalis, which race White (1962: 721) records from as far north as Archer’s Post and Meru. It seems that the range given for abdominalis in Mackworth-Praed and Grant (1955: 432) should be extended northward a very considerable distance. Our present bird agrees with abdominalis in lacking the white supraorbital streak, and disagrees with the description of griseoflava in Mackworth-Praed and Grant, who write that from the chin to the upper abdomen the plumage is white or huffish white. Our present specimen has these parts pale purplish slate gray, and this appears to be the normal thing, not white as in their description. i 1967 1966 Cheney Expedition to Kenya 27 Family Hirundinidae Psalidoprocne pristoptera massaica Neumann The black rough-winged swallow was met with in the Karissia Hills, June 3 to 15, when a series of 7 specimens, 3 males, 2 females, 2 unsexed, was collected. Three of these birds are molting in the wings and tail. In two the long rectrices are only partly grown, causing them to appear short-tailed, the depth of the fork being 12 and 15 mm., respectively, as opposed to 35 to 45 mm. in the others. These two birds are also slightly brownish, less deep black than the others, and may be immature birds. Family Campephagidae Coracina caesia pur a (Sharpe) The gray cuckoo-shrike is a bird of the wooded highlands. Three male specimens were collected in the Karissia Hills on June 14; all had somewhat enlarged testes, black in color. Campephaga quiscalina martini Jackson One adult female of the purple-throated cuckoo-shrike was obtained in the Karissia Hills on June 19, a bird in non-breeding condition. Campephaga phoenicea flava Vieillot The black cuckoo-shrike was collected in two localities, as follows: Karis- sia Hills, June 14 and 18, one adult male with enlarged testes, and one imma- ture female; Mt. Nyiru, June 22 and 26, two adult females with small ovaries. Family Dicruridae Dicrurus adsimilis divaricatus (Lichtenstein) This drongo is one of the common birds of the east African bush country. One adult female, in non-breeding condition, was collected in the Karissia Hills on June 12. It is in molt in the wings (outer remiges), it appears to be fully adult but has narrow white tips on the under tail coverts, recalling the immature plumage to that extent, but not elsewhere. Family Laniidae Nilaus afer minor Sharpe The Somali brubru was met with on Mt. Nyiru, where two adults, one of each sex, were collected on June 24 and 26; both birds were in non-breeding condition. Dryoscopus gambensis malzacii (Heuglin) This puff -back, characterized by its grayish scapulars and relatively heavy mandible (as compared with D. cubla), is represented in the collection by 28 Contributions in Science No. 130 one adult male with somewhat swollen testes and one non-breeding adult female, both taken in the Karissia Hills, June 6 and 14. The female may have been wrongly sexed as it is just as dark blue black above as the male, not earth-brown as described by Mackworth-Praed and Grant (1955: 620). Tchagra jamesi jamesi (Shelley) The three-streaked tchagra occurs in northern Kenya north to Ethiopia and Somalia. One adult female, in non-breeding state, was collected on the west base of Mt. Nyiru, June 24. Tchagra australis emini (Reichenow) This race of the brown-headed bush-shrike, somewhat darker brown above than T. a. minor (specimens seen from Nawisha and Kapenguria), was collected in the Karissia Hills, June 3 and 6, where one breeding female and one non-breeding male were taken. The Karissia Hills must be close to the northeastern edge of the range of the race. White (1962: 21) does not even include Kenya in its range, although listing Uganda; Mackworth-Praed and Grant (1955: 626) do record it from western Kenya. Tchagra cruenta hilgerti (Neumann) The Somali rosy-patched shrike was fairly numerous at the west base of Mt. Nyiru, where two adults of each sex were collected June 23, 24 and 25. All the birds had the small, resting gonads of the non-breeding season. Laniarius ferrugineus ambiguus Madarasz The boubou shrike was very common in the Karissia Hills, where six adults, three of each sex, were obtained June 2 and 3. One of the males and one of the females were in full breeding condition; the others were not. Laniarius funebris (Hartlaub) The slate-colored boubou shrike was common on Mt. Nyiru, June 21-26, when seven examples were collected; one immature female molting into adult plumage, three adult females, one adult male, and two unsexed adults. One of the females had the ovary enlarged, the others and the male had small gonads. Malaconotus sulfureopectus similis Smith Two examples of the sulphur-breasted bush-shrike, one of each sex, were taken in the Karissia Hills, June 4 and 6. This is a widely distributed species in Kenya. Lanius dorsalis Cabanis One specimen of the Teita fiscal was obtained in the plains area 15 miles south of Maralal, on June 7; an adult male with slightly enlarged testes. The bird is in worn plumage but is molting, especially in the tail. This shrike is 1967 1966 Cheney Expedition to Kenya 29 widely distributed in Kenya but tends to be local and hence very uneven in its local status. Lanius collaris humeralis Stanley The long-tailed fiscal is represented in the collection by one adult male taken in the Karissia Hills, June 15. The bird had the testes slightly enlarged. Family Paridae Parus albiventris Shelley The white-breasted tit occurs in forested areas, scrub, and bush country. Three examples were obtained, as follows: Karissia Hills, June 6, one adult male with enlarged testes; 5 miles west of Maralal, June 16, one adult of each sex with slightly enlarged gonads. Family Corvidae Corvus rhipidurus Hartert One adult male of the fan-tailed raven was collected 25 miles south of Baragoi, June 21, a bird in non-breeding condition. Corvus albicollis Latham The white-necked raven is represented in the collection by one adult female, non-breeding state, taken on the west base of Mt. Nyiru on June 22. The bird has an abnormal bill, apparently due to a mishap sometime prior to its death, giving it almost a small, narrow casque on the base of the culmen. Family Oriolidae Oriolus larvatus rolled Salvadori The black-headed oriole was common in the woodlands of the Karissia Hills where a series of six specimens was collected, June 2 to 18; three adult males with enlarged gonads, one adult female with somewhat swollen ovary, and one immature bird of each sex. The last two have the breast streaked with black, the chin and throat streaked black and yellow. The adults have red bills, the immature ones black. One of the adults shows sign of molt in the wings. Several occupied nests were noted in the large fever trees at the Karissia Hills camp. Family Sturnidae Lamprotornis chalybeus cyaniventris (Blyth) The blue-eared glossy starling is a common bird in much of Kenya. Five specimens were collected, as follows: plains area 15 miles south of Maralal, one adult female in non-breeding state, two immature females, June 4; west base of Mt. Nyiru, one immature male, June 24, and one adult female, June 27, ovary not enlarged. 30 Contributions in Science No. 130 Cinnyricinclus leucogaster verreauxi (Bocage) The violet-backed starling is represented in the collection by two males taken in the Karissia Hills, June 11. One is in full adult plumage, the other is molting into the adult feathering. Spreo superbus (Riippell) The superb starling was very common in the areas visited by the expedi- tion. In the plains area 15 miles south of Maralal, June 4 to 7, three adult males, one immature male, and two adult females were obtained; in the Karis- sia Hills one immature male was collected on June 12. In addition, three adult males were collected at the west base of Mt. Nyiru on June 27. One adult female was also obtained with them. Creatophora cinerea (Menschen) The wattled starling, a well-known bird of much of Africa, was collected as follows: plains area 15 miles south of Maralal, June 7, one adult female with slightly enlarged gonads; June 12, one adult male with swollen testes; Karissia Hills, June 12, one adult male with large testes. The wattles are well developed in the Maralal male, not in the other. Buphagus erythrorhynchus (Stanley) The red-billed ox-pecker is represented in the collection by four exam- ples, as follows: Karissia Hills, June 12, one adult female; west base of Mt. Nyiru, June 24, one adult male with somewhat enlarged gonads; also two adult males were taken on the west base of Mt. Nyiru June 27, one with small and the other with enlarged testes. Family Nectariniidae Anthreptes collaris garguensis Mearns This race of the collared sunbird is represented by two adult males and one adult female, taken in the Karissia Hills, June 7 to 19. None of the birds was in breeding condition. Nectarinia amethystina kirkii (Shelley) One adult of each sex of the amethyst sunbird was collected in the Karis- sia Hills, June 11 and 13. Nectarinia venusta falkensteini (Fischer and Reichenow) The Kenya buff-breasted sunbird was taken at two places, as follows : Karissia Hills, June 12 and 18, one adult of each sex; west base of Mt. Nyiru, June 26, one adult male; all in non-breeding state. Nectarinia preussi kikuyuensis Mearns This race of the double-collared sunbird is represented by three sped- 1967 1966 Cheney Expedition to Kenya 31 mens, two adult males and one adult female, all taken in the Karissia Hills, June 15 to 20. Nectarinia mariquensis osiris Finsch The mariqua sunbird was extremely common on the west base of Mt. Nyiru, June 22 to 25, where a series of five adult males, one immature male, and three adult females was collected. The immature male is molting into adult plumage. Two of the females have the chin and middle of the throat solid blackish; the other has these areas, as well as the upper and lateral parts of the abdomen, streaked with blackish. Family Zosteropidae Zosterops senegalensis jacksoni Neumann Jackson’s white-eye was very numerous in the Karissia Hills, where a series of seven adult females was obtained June 6 to 14. The absence of males in this series is puzzling as usually the sexes are found together in the same loose flocks during the non-breeding season. All the hens collected had small gonads, and were non-breeding birds. Family Ploceidae Ploceus baglafecht reichenowi (Fischer) One example of this wide-ranging weaver was obtained in the Karissia Hills, June 8, an adult male with somewhat enlarged testes. Ploceus velatus uluensis (Neumann) The masked weaver was met with on the west base of Mt. Nyiru, where one male, almost, but not quite, in adult plumage, was collected on June 27. Ploceus melanogaster stephanophorus (Sharpe) One adult female of this black-billed weaver was collected in the Karissia Hills, on June 14; it had the ovary slightly enlarged. Malimbus rubriceps leuconotus (Muller) The red-headed weaver was collected in two places, as follows: Karissia Hills, June 12, one adult male with somewhat swollen gonads; west base of Mt. Nyiru, June 22, one immature male. Dinemellia dinemellii dinemellii (Riippell) This noisy, conspicuous weaver was met with 25 miles south of Baragoi, where two adult females were collected on June 21. Plocepasser mahali melanorhynchus Bonaparte The black-billed sparrow-weaver was common on the west base of Mt. 32 Contributions in Science No. 130 Nyiru, where four adult females and one immature female were collected on June 22 and 24. The specimen marked as immature is similar to the adults. Pseudonigrita arnaudi arnaudi (Bonaparte) The gray-headed social weaver is represented in the collection by three examples, as follows: plains area 15 miles south of Maralal, June 7, one adult, non-breeding female; June 17, one adult male with somewhat enlarged testes; Karissia Hills, June 11, one adult male with somewhat swollen gonads. Passer iagoensis rufocinctus Finsch and Richenow Two examples of the Kenya rufous sparrow were collected in the plains area 15 miles south of Maralal, a female on June 4, and a male on June 7; both birds showed slight gonadal enlargement. The male is in very abraded plumage. Petronia xanthocollis pyrgita (Heuglin) The yellow-spotted petronia is represented in the collection by one adult, non-breeding male, taken in the Karissia Hills on June 12. Estrilda paludicola paludicola Heuglin This rather local waxbill was found in the Karissia Hills on June 20 when two non-breeding females were collected. In Kenya this species is found only in the more western half of the country; the Karissia Hills may be close to the eastern limit of its range. Estrilda ianthinogaster (Reichenow) The Kenya grenadier was common on the west base of Mt. Nyiru, where four examples were taken June 22 and 24; two adult males in non-breeding state, one immature male, and one non-breeding adult female. Estrilda bengala bengala (Linnaeus) The red-cheeked cordon-bleu is represented in the collection by two adults, both with large gonads, taken in the Karissia Hills, June 8. The female had a large developing egg. Family Emberizidae Emberiza flaviventris kalaharica Roberts One adult male, taken in the Karissia Hills on June 4, represents this bunting. It is in fairly worn plumage. Emberiza tahapisi tahapisi Smith The cinnamon-breasted rock bunting was met with on the west base of Mt. Nyiru on June 25 when an adult male with slightly swollen gonads was collected. 1967 1966 Cheney Expedition to Kenya 33 Family Fringillidae Serinus sulphur atus sharpii Neumann One adult female brimstone canary was collected 5 miles west of Maralal on June 16. It had a slightly enlarged ovary. Serinus striolatus striolatus (Riippell) The streaky seed-eater is represented in the collection by one adult, non- breeding female, taken in the Karissia Hills, June 20. Serinus reichardi striatipectus (Sharpe) Three adult females of this seed-eater were obtained on the west base of Mt. Nyiru, June 25; one had the ovary enlarged, the other two did not. Literature Cited Chapin, J. P. 1939. The Birds of the Belgian Congo, Pt. II. Bull. Amer. Mus. Nat. Hist., 75: 1-632. Friedmann, H. 1930- Birds collected by the Childs Frick expedition to Ethiopia and Kenya 1937. Colony. Bull. U.S. Natl. Mus., 153, Pt. I, Non-passeres, 1930: 1-516; Pt. II, passeres, 1937: 1-506. Granvik, H. 1934. The ornithology of North Western Kenya Colony, with special regard to the Suk and Turkana districts. Revue de Zoologie et de Botanique Africaines, Brussels, 25: 1-190. Hall, B. P. 1961. The taxonomy and identification of pipits (genus Anthus), Bull. Brit. Mus. (Nat. Hist.), Zool., 7, no. 5: 243-289. Hohnel, L. von 1894. Discovery of Lakes Rudolf and Stefanie. A narrative of Count Samuel Teleki’s exploring & hunting expedition in eastern equatorial Africa in 1887 & 1888. London: Longmans, Green and Co., 2 vols.: 1-435; 1-397. Jackson, F. J., and Sclater, W. L. 1938. The birds of Kenya Colony and the Uganda Protectorate. London: Gurney and Jackson, 3 vols.: 1-1592. Mackworth-Praed, C. W., and Grant, C. H. B. 1955. Birds of eastern and northeastern Africa. London: Longmans, Green and Co., Vol. 2: 1-1099. 1957. Birds of eastern and northeastern Africa. London: Longmans, Green and Co., Vol. 1, ed. 2: 1-806. Pitman, C. R. S. 1964. A further note on the egg of the red-chested cuckoo Cuculus solitarius Stephens. Bull. Brit. Ornith. Club, 84: 140-141. 34 Contributions in Science No. 130 Sharpe, H. B. 1930- The birds of Marsabit Mountain, Kenya Colony. Parts II-V. Bateleur, 1931. 2, 1930: 102-107; 3, 1931: 9-14, 35-39, 97-103. van Someren, V. G. L. 1921. On a collection of birds from Turkanaland. J. East Africa and Uganda Nat. Hist. Soc., no. 16: 3-38. 1922. Notes on the birds of East Africa. Novitates Zoologicae, 29: 1-246. 1932. Birds of Kenya and Uganda, being addenda and corrigenda to my previous paper in “Novitates Zoologicae,” 29: 1922. Novitates Zoo- logicae, 37: 252-380. White, C. M. N. 1960. A check list of the Ethiopian Muscicapidae (Sylviinae). Pt. I. Occas. Papers Natl. Mus. of Southern Rhodesia, Bulawayo, Nat. Sci., 3, no. 24B: 399-430. 1961. A revised check list of African broadbills, pittas, larks, swallows, wag- tails and pipits. Lusaka: Government Printer, 84 pp. 1962. A revised check list of African shrikes, orioles, drongos, starlings, crows, waxwings, cuckoo-shrikes, bulbuls, accentors, thrushes and bab- blers. Lusaka: Government Printer, 176 pp. 1962. A check list of the Ethiopian Muscicapidae (Sylviinae). Pt. II. Occas. Papers Natl. Mus. of Southern Rhodesia, Bulawayo, 3, no. 26B: 653-738. 1963. A revised check list of African flycatchers, tits, tree creepers, sunbirds, white-eyes, honey-eaters, buntings, finches, weavers and waxbills. Lusaka: Government Printer, 218 pp. 1965. A revised checklist of African non-passerine birds. Lusaka: Govern- ment Printer, 299 pp. \73 LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE UMBER 131 February 20, 1968 ON THE AGE DISTRIBUTION OF SMILODON CALIFORNICUS BOVARD FROM RANCHO LA BREA By George J. Miller MAR 141968 Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 814 x 1 1 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor ON THE AGE DISTRIBUTION OF SMILODON CALIFORNICUS BOVARD FROM RANCHO LA BREA By George J. Miller1 Abstract: Skulls of over 2100 specimens of Smilodon cali- fornicus Bovard from the collection of the Los Angeles County Museum of Natural History are studied in an effort to determine the percentages of the various age groups that were trapped in the La Brea Pits. The specimens are arbitrarily classified as Juvenile, Young Adult, Adult, and Aged, using tooth wear and closure of the basioccipital-basisphenoid suture as criteria. The percentages of the four age groups are found to be as follows: 16.6% Juve- niles, 23.2% Young Adults, 17.2% Adults, and 8.5% Aged, with 34.5% undetermined. The large number of undetermined, con- sisting mostly of adults that could not be positively classified as Young Adults, Adults, or Aged, are then classified with the Aged category eliminated, giving the following breakdown: 16.6% Juveniles, 25.7% Young Adults, 56.5% Adults, and 1.2% unde- termined. These data show that a cross section of the Smilodon calif ornicus population, rather than a preponderance of any one age group, was trapped in the tar pits. Introduction For the past 45 years the Rancho La Brea Fauna has been recognized by paleontologists as the outstanding collection of Pleistocene life available for study (Stock, 1961). Of the many large mammals trapped and preserved in the tar pools, those found in the greatest abundance are Canis dims (the dire wolf), with from 1646 (Marcus, 1960) to 2000 specimens (Stock, 1929a); and Smilodon californicus Bovard, (the sabre-tooth cat), with from 1029 (Marcus, 1960) to 1500 specimens (Stock, 1929a). The excellent state of preservation of this material (Stock, 1929b and 1961), together with its abundance, suggests many possibilities for population, variation, extinction, and microevolution studies. In the hope that the information obtained will be of value in the furtherance of the studies mentioned above, the present study attempts to determine the age distribution of Smilodon californicus from an examination of the skulls of the animals that were trapped in the tar pits. The term “skull,” as used in this paper, denotes any cranial specimen of sufficient completeness to determine the age. S. californicus was chosen for this study over Canis dims for convenience only — the Smilodon material is more readily accessible. It is hoped that a similar study will be done in the near future on C. dims. 1Research Assistant in Vertebrate Paleontology, Los Angeles County Museum of Natural History (LACM). 1 2 Contributions in Science No. 131 Material Studied Skulls and parts of skulls of Smilodon calif ornicus from the collection of the Los Angeles County Museum of Natural History were used exclusively in this study. All available catalogued and uncatalogued specimens were exam- ined and taken into consideration (fifty of the 576 catalogued skulls were out on loan and therefore unavailable for study). Out of over 2100 specimens examined, 918 were, for reasons discussed below, found to be suitable for this study. Procedure Lack of knowledge of the life span and the rate of maturation of Smilodon made it impossible to determine exact age groups, so arbitrary age groups were used (it was not deemed advisable to extrapolate from knowl- edge of recent carnivores because of the different patterns of tooth wear brought about by Smilodon' s unique dentition). The age groups were desig- nated as Juvenile, Young Adult, Adult, and Aged. The Juvenile group includes all skulls with any deciduous dentition or alveoli for deciduous dentition and with no wear on any of the permanent dentition (Figures 1 and 2). This group is probably under-represented because of separation and fragmentation of the soft-boned young skulls with their incompletely sealed sutures (churning, or movement of the petrolifer- ous mass in the tar pits, had a tendency to cause breakage and separation of bones and skeletons as well as cause the phenomenon known as pit wear). Complete skulls of Juveniles are rare (out of 150 Juvenile specimens examined, there were nine complete skulls, and the preparator had wired these together at the sutures as shown in Figure 3); most of the specimens in this age group consist of fragments (Fig. 4). Many skulls and fragments of skulls in this and in other age groups were not used in this study because of the possibility of duplication. When the uncatalogued fragmentary speci- mens were examined, all pit data were checked, and when a left and right part of an animal of the same age with the same pit data was found only one part of the animal was listed. Although this procedure could cause the elimination of some specimens from the study, any possibility of counting the same animal twice is avoided. As the workers who originally catalogued the fossils from Rancho La Brea are believed to have been extremely accu- rate (Leslie F. Marcus, 1966, personal communication), this study has not questioned possible duplication of any of the catalogued material. Skulls with all permanent dentition or alveoli for permanent dentition and no deciduous dentition or alveoli were classified as Young Adults. Other criteria used for this group were: no wear on the incisors, canines, or P3; less than one millimeter of wear on the paracone of the P4 (Figures 5 and 6); and the open basioccipital-basisphenoid suture (Fig. 7). It was observed during the course of this investigation that closure of the basioccipital- 1968 Age Distribution of Smilodon 3 Figure 1. (Upper) Smilodon californicus Juvenile. Lingual view of deciduous right C and M3 and right permanent C and P4. Permanent dentition not completely erupted (LACM 2001-30). Figure 2. (Lower) Smilodon californicus Juvenile. Lingual view of deciduous right M3 and permanent right P4. Permanent dentition not completely erupted and show- ing no wear (LACM 2001-104). 4 Contributions in Science No. 131 Figure 3 (Upper) Smilodon californicus Juvenile. Ventral view of one of the few complete Juvenile skulls recovered from the La Brea Pits (LACM 2001-104). Figure 4. (Lower) Smilodon californicus Juvenile. Typical skull fragment showing upper right permanent and deciduous dentition (LACM 2001-30). 1968 Age Distribution of Smilodon 5 basisphenoid suture coincided with the completion of the emergence of the permanent dentition; therefore, this feature was also used to classify the Young Adults. Presence of all permanent dentition or alveoli; moderate wear on the canines, incisors, and P3; the paracone and metacone of the P4 worn from one millimeter to one centimeter from the alveolar border (Fig. 8); and the closed basioccipital-basisphenoid suture (Fig. 9) were used as criteria for the Adult grouping. Extreme wear on all dentition and the P4 worn from one centimeter above the alveolar border to the alveolar border (Fig. 10) were the char- acteristics used to classify the Aged. Some carnassials were found that were worn through to the pulp cavity as shown in Figure 1 1. A tooth worn to this extent would not be likely to last much longer so animals with this condi- tion were considered to be near the end of their life span and were therefore classified as Aged. When a division was made between the Adults and Aged groups, the amount of wear on the P4, unless caused by malocclusion, took precedence over other criteria. Extensive wear was sometimes found on one carnassial and not on the other and was thus considered to be caused by malocclusion rather than by age. Although there are two hypotheses to explain the mode of existence of Smilodon (some authorities consider Smilodon to have been a fierce predator, whereas others consider him to have been a scavenger or carrion feeder), the carnassial would seem to have been essential to the animal in either case. Another possible cause of extensive wear could be additional use of the carnassials brought about by loss of the canines (Stock, 1961). Use of closure of the basioccipital-basisphenoid suture as a means of age determination was justified, as discussed above, by the fact that closure was observed to coincide with the emergence of the permanent dentition. Furthermore, the location of the suture at a sturdily constructed area of the skull prevented its opening on Aged specimens that had been subjected to pit churning. Aged skulls were found that had undergone severe strain, with many of the other sutures opened, but still having the basioccipital- basisphenoid suture closed. Results Counts taken during this study show that a total of at least 2400 Smilodon skulls have been collected from Rancho La Brea, approximately 2100 of these being in the collection of the Los Angeles County Museum of Natural History. This count is somewhat more than that of Stock’s (1929) census with 1500 specimens and Marcus’ (1960) census with 1029 specimens; however, these censuses were taken using bones other than the skull. The discrepancies may possibly be accounted for by Marcus’ (1960) 6 Contributions in Science No. 131 Figure 5. (Upper) Smilodon californicus Young Adult. Lingual view of permanent left P3, P4, and Ml showing less than one millimeter of wear on the paracone of the P4 (LACM 2001-101). Figure 6 (Lower) Smilodon californicus. Young Adult. Lingual view of permanent left P3 and P4 showing one millimeter of wear on the paracone of the P4 (LACM 2001-290). 1968 Age Distribution of Smilodon 7 Figure 7. (Upper) Smilodon calif ornicus Young Adult. Ventral view of skull showing basioccipital-basisphenoid suture not closed (LACM 2001-101). Figure 8. (Lower) Smilodon calif ornicus Adult. Lingual view of permanent left P3 and P4 showing more than one millimeter of wear on the paracone and metacone of the P4 (LACM 2001-250). 8 Contributions in Science No. 131 Figure 9. (Upper) Smilodon calif ornicus Adult. Ventral view of skull showing basi- occipital-basisphenoid suture closed (LACM 2001-96). Figure 10. (Lower) Smilodon californicus Aged. Lingual view of permanent left P4 showing tooth worn down to within one centimeter of the alveolar border (LACM 2001-102). 1968 Age Distribution of Smilodon 9 Figure 11. Smilodon calif ornicus Aged. Lingual view of permanent right P4 showing tooth worn through to the pulp cavity (LACM 2001-297) . Selection Hypothesis, in which he suggests the possibility of selective preser- vation and of selective collection. The Rancho La Brea field notes show that during excavation of the pits a great many specimens were discarded as being too poorly preserved to merit retention (Wyman, 1913). Many collectors have a tendency to be more selective when collecting a rich deposit than they would be under more normal circumstances and it seems that the La Brea collectors were affected by this human weakness. There are 576 catalogued skulls in the Los Angeles County Museum of Natural History collection, 50 of which were out on loan and thus not studied; 342 uncata- logued but identifiable; and over 1350 uncatalogued fragmentary specimens which were not suitable for this study, making a total of at least 2100 speci- mens in the Los Angeles County Museum collection. A recent census of the collection at the University of California at Berkeley (John E. Mawby, 1966 personal communication) showed 111 skulls and fragments of skulls (an additional 62 fragmentary specimens in this collection are probable duplica- tions). Many skulls from the Berkeley collection have been loaned or traded; it is probable that the collection originally contained at least 200 skulls. Data were not available on the smaller collections made by Occi- dental College, Los Angeles High School, and other institutions. If these 10 Contributions in Science No. 131 smaller collections are disregarded, there is a total of at least 2100 skulls, 918 or 43.7% of which were suitable for or available for and used in this study. It must be emphasized that the 2100 skulls include many fragments and thus may give an exaggerated picture of the number of animals repre- sented. It is quite probable that many of these fragments are from the same animal. This probable variable, as mentioned above, was taken into account in this investigation. Table 1 shows a breakdown of the data obtained into four age groups: Juvenile, Young Adult, Adult, and Aged, with an undetermined group con- sisting of 11 Juveniles, 24 Juvenile or Young Adult fragments, and 283 Adults. Because the 11 Juveniles are possible duplications, they are not included in the classification. The 283 Adults could not be classified as Adults or Aged because of their poor condition — too many teeth missing in some, and difficulty in distinguishing between age and pit wear in others; however, they could all be positively identified as mature adults by closure Table 1 FREQUENCY IN NUMBER AND PERCENT OF INDIVIDUALS OF EACH AGE GROUP Frequency in Number Percent of Age Group of Individuals Individuals Juveniles 150 16.6 Young Adults 213 23.2 Adults 158 17.2 Aged 79 8.5 Undetermined 318 34.5 Totals 918 100.0 Table 2 FREQUENCY IN NUMBER AND PERCENT OF INDIVIDUALS OF EACH AGE GROUP (THREE AGE GROUPS) Frequency in Number Percent of Age Group of Individuals Individuals Juveniles 150 16.6 Young Adults 237 25.7 Adults 520 56.5 Undetermined 11 1.2 Totals 918 100.0 1968 Age Distribution of Smilodon 11 of the basioccipital-basisphenoid suture. Although Table 1 shows a good age distribution, 34.5% of the animals studied remain unclassified. It was there- fore decided for purposes of comparison to combine the Adults and the Aged and to include the 283 undetermined Adults under the heading of Adults, and to place the 24 fragmentary young specimens under the Young Adult heading to give the results shown in Table 2. The decision to place the 24 young animals in the Young Adult grouping was made by using either tooth wear or closure of the basioccipital-basisphenoid suture as criteria instead of using both as was done in Table 1. In Table 3 the distribution of the four age groups is shown for the most heavily populated pits. The percentage distribution of the same data is shown in Table 4. The data from Tables 3 and 4 were combined into three age Table 3 FREQUENCY OF SKULLS OF EACH AGE GROUP IN THE MOST HEAVILY POPULATED PITS (FOUR AGE GROUPS) Pit Number 3 61-67 4 77 13 60 2 Age Group Juveniles 61 23 24 7 9 4 4 Young Adults 56 44 19 7 4 0 3 Adults 65 62 26 4 1 3 2 Aged 27 33 15 1 1 0 1 Undetermined 136 95 26 44 43 11 6 Totals 345 257 110 63 58 18 nr Table 4 PERCENT OF SKULLS OF EACH AGE GROUP IN THE MOST HEAVILY POPULATED PITS (FOUR AGE GROUPS) Pit Number 3 61-67 4 77 13 60 2 Age Group Juveniles 17.6 9.0 21.8 11.1 15.5 22.2 25.0 Young Adults 16.2 17.1 17.2 11.1 6.9 0.0 18.8 Adults 18.9 24.1 23.7 6.3 1.7 16.7 12.5 Aged 7.9 12.8 13.6 1.6 1.7 0.0 6.2 Undetermined 39.4 37.0 23.7 69.9 74.2 61.1 37.5 12 Contributions in Science No. 131 groups (Juvenile, Young Adult, and Adult) in the same manner as described above (the specimens under the undetermined heading were all classified except for the 11 doubtful Juveniles) to give the distribution of the animals in the seven most heavily populated pits as seen in Tables 5 and 6. A total of 867 specimens was used in the pit breakdown, which made a smaller sampling than was used in the over-all distribution because the smaller pits were not included and because skulls without pit data were necessarily eliminated. The data for Pit 61 and Pit 67 were combined because these two pits were actually “one large pit” (Marcus, 1960). Table 5 FREQUENCY OF SKULLS OF EACH AGE GROUP IN THE MOST HEAVILY POPULATED PITS (THREE AGE GROUPS) Pit Number 3 61-67 4 77 13 60 2 Age Group Juveniles 68 27 25 7 9 5 4 Young Adults 91 69 22 17 12 0 5 Adults 111 161 63 39 37 13 5 Undetermined 9 0 0 0 0 0 2 Totals 345 257 110 63 58 18 16~ Table 6 PERCENT OF SKULLS OF EACH AGE GROUP IN THE MOST HEAVILY POPULATED PITS (THREE AGE GROUPS) Pit Number 3 61-67 4 77 13 60 2 Age Group Juveniles 19.3 10.5 22.7 11.1 15.5 27.8 25.0 Young Adults 26.5 26.8 20.0 27.0 20.7 0.0 31.3 Adults 51.6 62.7 57.3 61.9 63.8 72.2 31.3 Undetermined 2.6 0.0 0.0 0.0 0.0 0.0 12.4 Discussions and Conclusions It is believed that the use of the two different age distributions (Juve- nile, Young Adult, Adult, and Aged as opposed to Juvenile, Young Adult, 1968 Age Distribution of Smilodon 13 and Adult), gives a more thorough view of the data than does a single breakdown. Both breakdowns are considered to be of value in that the four- class grouping gives a more detailed picture of the distribution, whereas the three-class grouping includes a larger sampling. It may be noted that the three-class grouping, “youth, middle life, and old age,” was favored by Kurten (1954). The most likely place for deviation is in the Juvenile category, wherein poor preservation of young animals would result in a low count. The possi- bility of increased fecundity (Merriam and Stock, 1932) would, however, have a tendency to partially offset this factor. It has been hypothesized that animals on the verge of extinction, undergoing a population decrease, experi- ence an increase in fecundity as a natural compensatory mechanism. This phenomenon may be compared to a similar condition that may be observed on a smaller scale in living populations today during so-called “good years.” In a time of drought or other adverse climatic conditions litter size and populations decrease. When environmental conditions improve and the food supply becomes abundant an increase in fecundity may be observed. Favor- able conditions would have a tendency to occur when the population is reduced, as the environment would then be more able to meet the needs of the smaller population. There is also the possibility that younger animals with their greater vigor would be better able to avoid being trapped or to escape after being trapped. However, it would seem that the lack of experi- ence of such animals coupled with their youthful exuberance would tend to lead them into more trouble, thus increasing the probability of their blunder- ing into the tar pools. The net result would thus probably be a mutual cancellation of these two factors. Two possibilities of discrepancies in the division of the Adult and Aged exist because some animals show excessive tooth wear for their age (Kurten, 1954); and because of difficulty in distinguishing between pit wear and age. Combining these two groups into the Adult grouping, as was done in Tables 2, 5, and 6, eliminated this problem. Selective preservation and collection (Marcus, 1960) is a possible variable that could not be controlled at this late date. The age distribution in Table 1 clearly shows that a cross section of the Smilodon population was trapped and preserved in the La Brea Pits, rather than a majority of any one age group. Table 1 also shows that there are a sufficient number of specimens from each age group for population, variation, extinction, and microevolutionary studies (it should be recognized that until dating of this material has been accomplished, the collection rep- resents a population over an indefinite period of time). These data also show that at least 8.5% of the Smilodon population reached old age. The need for age dating of the La Brea collection has long been recog- nized (Howard 1960, Stock 1961); however, very little has been done. 14 Contributions in Science No. 131 Although work is now in progress on the dating of animal remains and some information has been obtained, all published material to date has been from analyses of wood samples. A “wood sample” from an unidentified pit was dated as 16,325 + or — 2000 years B. P. (Before Present) (Douglas 1952). Howard (1962) commented on the questionable value of this dating because of the lack of proper data on the specimen and because of the method used. Stock (1961) referred to the La Brea fauna as “late middle Pleistocene” and “late Pleistocene.” Howard (1960) reported dating of a tree from Pit 3 as 14,500 B. P. + or — 200 years. As Howard observed, “The dating of the tree from pit 3 undoubtedly could apply as well to the fossil bones of the extinct ani- mals ” It is also possible that the bones found stratigraphically below the tree roots are even older (Hildegarde Howard, 1967, personal communi- cation). However, this is an isolated case. The fortuitous circumstance of a tree rooted in amongst the bones occurred only in one pit. This was the state of things as of 1962. Recent radio-carbon dating by Berger and Libby (1966) has extended the time range for Rancho La Brea considerably as shown below: Pit 4, 5 foot depth 33,700 + or - 1600 years B. P. Pit 9, 8V2 foot depth 13,300 + or — 160 years B. P. Pit 9, 16 foot depth > 40,000 years B. P. Pit 16, 6V2 foot depth > 40,000 years B. P. Pit 16, 12 foot depth > 40,000 years B. P. Pit 77, no record of depth 37,000 -f- or — 2660 years B. P. Although this gives a much better perspective of the age of some of the pits, this work was also done on wood samples, and any correlation between the wood and the bones is open to question. Due to pit churning the cor- relation between bones found in any one pit at the same depth and in the same grid may also be questioned. Berger and Libby (1966) also report a date of 23,300 + or — 510 years B. P. for cypress wood and 32,350 + or — 1400 years B. P. for leaves of California live oak from the University of California at Berkeley collection of Rancho La Brea material. Axelrod (1966) used these data for an analysis of the Rancho La Brea flora in which he was able to show that “at least two widely different communities” were “entombed in the tar pits at different times, and under wholly different climatic conditions.” Recent (July, 1966) unpublished dating of Canis dims humeri by Geochron Labs., Inc. showed a specimen from Pit 3 at a depth of 22 to 25 feet to have an age of 9860 + or — 550 C-14 years B. P., and a specimen from Pit 16 at 8 to 12 feet to have an age of 10,710 + or — 320 C-14 years B. P. This dating was, however, on carbonate, as it was found to be impractical to date the collagen because of contamination by petroleum and other organic compounds such as paint. (Paint was used by the original curators to label the bone.) New collagen extraction methods 1968 Age Distribution of Smilodon 15 now being developed promise to overcome this difficulty. There is, therefore, still a need for an extensive dating of bone. As will be suggested below, two of the pits may be considered as representative of the over-all distribu- tion of Smilodon from La Brea. Possibly the most valuable procedure would thus be to date all of one element from one of the more prevalent animals such as Smilodon calif ornicus or Canis dirus for one of the more representa- tive pits. As Berger and Libby (1966) show Pit 4 to be one of the oldest pits and Pit 4 will be suggested below to be one of the representative pits, it might be most useful to do this work on Pit 4, although Pit 3 should also be considered as one of the representative pits. Although Tables 1 and 2 show a larger proportion of Adults than of Juveniles, if the Juvenile and Young Adults are combined and then compared with the adults, a ratio of approximately 56 adults to 42 young is obtained. Without taking into consideration the poor preservation chances of young animals, this ratio is reasonably close to the age distribution (58 adults to 40 young) found in recent living populations of Panthera (Felis) leo (Guggisberg, 1963), the only recent large cat for which age distribution data were available. A chi-square test for deviation on these data based on the methods of Simpson, Roe and Lewontin (1960) showed a P value of .7 or no significant deviation from the expected. It may be argued that any comparison between Smilodon and the African lion is not necessarily perti- nent but there are no data on populations of any size of large carnivores, either fossil or living, available. The frequency of skulls from the four age groups (Juvenile, Young Adult, Adult, and Aged) in the most heavily populated pits (Tables 3 and 4) shows Pit 3 and Pit 4 to have an age distribution very close to that of all the pits combined (Table 1). Pit 3 and Pit 4 may thus possibly be repre- sentative of the Rancho La Brea Smilodon population, so either of these two pits could be utilized in a population study instead of studying the entire collection. Although similar frequencies being found in similar or related populations may not necessarily mean that the populations are identical (Leslie F. Marcus, 1967, personal communication), the fact that the age distributions are so nearly alike would seem to justify taking full advantage of the similarities. The breakdown of the data into three age groups (Juvenile, Young Adult, and Adult) also shows the age distribution of Smilodon from Pit 3 and Pit 4 to be in close agreement with the distribution of the entire Smilodon collection (Tables 2, 5, and 6) . The conclusion is therefore made that the Rancho La Brea population of Smilodon calif ornicus, as represented by the collection of the Los Angeles County Museum of Natural History, is a valid cross section of a population in time and as such, is excellent material for studies of population, variation, extinction, microevolution, and other related paleontological problems. 16 Contributions in Science No. 131 Acknowledgments I wish to express my profound gratitude to J. R. Macdonald, Senior Curator of Vertebrate Paleontology at the Los Angeles County Museum of Natural History, who suggested this project, for making available to me for study the Smilodon collection and associated catalogs and for many valuable suggestions. My thanks are also extended to Chief Curator of Earth Sci- ences, Theodore Downs, for the much needed background material used in this study. A difficult task was made easy and pleasant by the cooperation of the many people on the staff of the Los Angeles County Museum of Natural History. I would like to extend my deepest thanks to Leonard Bessom, Donald B. (Joe) Cocke, Charles McLaughlin, and Stephen Wright for their generous help in solving my many problems. I am very grateful to Hildegarde Howard for reading the manuscript and offering many help- ful suggestions. My special thanks go to Leslie F. Marcus for consultation and advice, of which I took full advantage. I would also like to thank James Asher of the Genetics Department at California State College at Long Beach for his valuable help with mathematical analyses. I would like to extend my special thanks to Professor John A. White, Curator of Verte- brate Paleontology at Idaho State University, for his many valuable sug- gestions and for his constant encouragement. None of the above-mentioned should be in any way held responsible for the opinions or conclusions expressed in this paper — I am entirely responsible. The specimens in the Los Angeles County Museum of Natural History collection were photo- graphed by the author. This research was partially supported by National Science Foundation Grant GB 5119. 1968 Age Distribution of Smilodon 17 Literature Cited Axelrod, D. I. 1966. The Pleistocene Soboba flora of Southern California. Univ. Calif. Publ. Geol. Sci., 60: 79 p., 14 plates. Berger, R., and W. F. Libby 1966. U.C.L.A. Radiocarbon Dates V. Radiocarbon, 8: 1-59. Douglas, D. L. 1952. Measuring low-level radioactivity. Gen. Electric Rev., 55 : 16-20. Guggisberg, C. A. W. 1963. Simba, the life of the lion. Philadelphia, Chilton Books, 309 p., illus. Howard, H. 1960. Significance of Carbon- 14 dates for Rancho La Brea. Science, 131: 712- 714. 1962. A comparison of avian assemblages from individual pits at Rancho La Brea, California. Los Angeles Co. Mus. Contrib. Sci., 58: 1-24. Kurten, B. 1954. Population dynamics and evolution. Evolution, 8: 75-81. Marcus, L. F. 1960. A census of the abundant large Pleistocene mammals from Rancho La Brea. Los Angeles Co. Mus. Contrib. Sci., 38: 11 p., 2 text figs. Merriam, J. C., and C. Stock 1932. The Felidae of Rancho La Brea. Carnegie Institution of Washington, xvi, 231 p., 152 text figs., 42 plates. Simpson, G. G., Anne Roe, and R. C. Lewontin 1960. Quantitative Zoology; rev. ed. Harcourt, Brace and Co., 440 p., 64 text figs- Stock, C. 1929a. A census of the Pleistocene mammals of Rancho La Brea, based on the collections of the Los Angeles County Museum. J. Mammal., 10: 281- 289, 2 text figs. 1929b. Significance of abraded and weathered mammalian remains from Rancho La Brea. Bull. So. Calif. Acad. Sci., 28: 1-5, 2 text figs. 1963. Rancho La Brea — A record of Pleistocene life in California. Los An- geles County Museum, Science Ser., 20 (Paleo. No. 11). Wyman, E. 1913. Rancho La Brea field notes. Unpublished. From the files of the Los An- geles County Museum of Natural History. IK 73 '.LOS ANGELES CONTRIBUTIONS COUNTY MUSEUM IN SCIENCE JMBER 132 February 20, 1968 TAXONOMIC NOTES ON SOME MEXICAN CEPHALOTINE ANTS (HYMENOPTERA: FORMICIDAE) By Roy R. Snelling ^\THS0/V/4^ MAR I ^ 1968 LIB RARttS Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8V2 x 1 1 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor TAXONOMIC NOTES ON SOME MEXICAN CEPHALOTINE ANTS (HYMENOPTERA: FORMICIDAE) By Roy R. Snelling1 Abstract: The previously undescribed male of Procry pto- cerus scabriusculus Emery is characterized and figured. Crypto- cerus insularis Wheeler, previously known only from the lost holotype, is recharacterized and its relationship to C. rohweri Wheeler and C. wheeled Forel is discussed. Soldiers and workers of all three are figured and a key is given for their separation. Procryptocerus scahnuscuSus Emery, 1894 Figure 1 In his excellent treatment of the cephalotine ants Kempf (1951) divided the genus Procryptocerus into a number of species complexes. In his analyses of these complexes he was able to place some names into synonymy; others were elevated to species level, and in general the systematics of the genus was put on a sound basis. Unfortunately, relationships within a given complex were left unresolved due to a lack of adequate representation of the alate forms. One of Kempf’s complexes was composed of the forms previously ascribed to P. striatus (F. Smith), a starting total of thirteen “subspecies and varieties,” which he reduced to seven species. Of these species, males were known and described for only two, P. adlerzi (Mayr) and P. convergens (Mayr) . While collecting in the vicinity of Cordoba, Veracruz, Mexico, during July, 1965, I was fortunate enough to secure several complete colonies of P. scabriusculus Emery, a member of this complex, in two of which were found alate individuals of both sexes. Since the previously undescribed male of this species exhibits a number of interesting characters which readily permit its separation from those of P. adlerzi and P. convergens, it seems worthwhile to describe it at this time. The male differs from that of P. adlerzi most obviously in the presence of distinct apical spurs on the middle and hind tibiae, a trait which it shares with P. convergens. If I correctly understand Kempf’s description of the P. con- vergens male, that of P. scabriusculus may be separated by the presence of a distinctly shining, sculptureless area on the frons (sculptured throughout in P. convergens, according to Kempf), the shorter hairs on the underside of the head (said to be longer than the antennal scape in P. convergens) , and the decidedly less truncate subgenital plate. Entomology Section, Los Angeles County Museum of Natural History, Los Angeles, California. 1 2 Contributions in Science No. 132 Male. Length of body, 7.2 - 8.5 mm; of forewing, 4.8 - 5.7 mm. Distance from anterior ocellus to apical margin of clypeus 0.8 times interocular dis- tance; greatest ocular diameter 0.5 times maximum head length; maximum diameter of lateral ocelli slightly less than minimum distance between them; antennal scape longer than first two funicular segments combined. Mandibles rugosopunctate, with distinct, essentially longitudinal rugulae. Clypeal disc minutely tessellate, laterally with a few short rugulae; apical margin evenly concave medially. Longitudinal rugulae of face variable, rather coarse and close between eyes and antennal sockets; fine, close and parallel between antennal sockets; elsewhere moderately coarse, spacing variable, but absent from triangular area on frons in front of anterior ocellus; inter-rugal sculptur- ing consisting of very fine tessellation. Occiput rounded, without distinct corners. Thoracic dorsum in profile not evenly rounded, the scutellum slightly bulging. Shoulders angulate, but not dentate, pronotal sides divergent caudad; pronotal dorsum free of sculpturing, the sides with a few fine striae below. Mayrian furrows of scutum deeply impressed; median lobe with numerous very fine oblique striae anteriorly and posteriorly, medially free of striae. Mesopleurae largely smooth, with a few fine striae below. Base of epinotum with a short triangular tooth on each posterior corner, the discal area with strong, essentially transverse, rugae; declivous face with strong transverse rugae. Middle and hind tibiae each with a single distinct apical spur. Wings slightly infuscated, marginal cell of fore-wing closed; hind wing with 6-8 hamuli. Petiole about 1.6 times as long as wide, its sides slightly convex; an- teriorly above with a few well-separated rugulae, with very short longitudinal rugulae along hind border. Postpetiole about 1.2 times longer than wide, free of sculpturing except for very short rugulae anteriorly and posteriorly. Abdo- men subcylindrical; first tergite with fine short striae basally; remaining tergites smooth and shining, devoid of sculpturing. Genitalia and subgenital plate as illustrated. Pubescence as described for P. adlerzi by Kempf (1958). Head, thorax, petiolar segments, and most of the first gastric tergite black; antennal scape fuscous-brown; funicular segments brown; legs reddish-brown, the tibiae lighter than the other segments; gastric segments brownish medially, lighter apically. A male from colony No. 76513-2, collected at Cordoba, Veracruz, Mexico, July 13, 1965, by the author, has been selected as the andro-type and is deposited in the Los Angeles County Museum of Natural History. At the time of my visit, P. scabriusculus was the most commonly en- countered cephalotine; foraging (?) individuals were seen at dusk crawling up and down the trunks of the trees in which they nested. Of the six colonies taken, four appear to be complete; the two remaining were surely only frag- 1968 Mexican Cephalotine Ants 3 Procryptocerus scabriusculus Figure 1. Procryptocerus scabriusculus, male: a, head; b, lateral aspect; c, genital capsule (left half, dorsal aspect; right half, ventral aspect without aedeagus); d, vol- sella, inner aspect; e, aedeagus, lateral aspect; f, subgenital plate, lateral aspect; g, subgenital plate, ventral aspect; h, fore and hind wings. Figures by Ruth DeNicola Snelling. 4 Contributions in Science No. 132 ments of a polydomous colony. All colonies were taken from dead twigs or limb stubs in living trees; none were found in the epiphyte samples which I examined, although Cryptocerus (C.) multispinosus biguttatus Emery was commonly found in such surroundings. Foraging (?) individuals moved rather slowly, but when disturbed were capable of swift, agile movements for a short distance. Individuals of one colony (No. 76513-1) were seen on several occasions to move across the soil surface to another tree about seven feet from the nest tree. Although numerous presumed foragers were seen, none gave any clue as to the food habits. Twice, however, two different individuals accepted dead Nasutitermes workers; indi- viduals regularly came to a honey-water mixture set out as bait. On the evening of July 12, at 2045 hours, alate females from colony No. 76513-2 appeared and promptly took wing. Three males appeared at 2112 and remained on the stub for several minutes, showing no interest in the half dozen females nearby, which had emerged at 2105. The latter flew away at 2120, followed a few minutes later by the males. Two dealate females were taken the following night at 2203. A few workers were present on the stub at the same time as the sexual forms but did not pay them any attention. Shortly before the sexual phases appeared, a brief rain shower had thoroughly satu- rated the area; temperature was 72°F, humidity ca. 80-85%. On several preceding nights, males were attracted to 15 watt ultraviolet “black light” tubes set up about 150 feet from the tree in which colony 76512-2 was located; unfortunately, the times at which these individuals arrived at the light source were not noted. Cryptocerus (Cryptocerus) insularis Wheeler, 1934 Figure 2 Wheeler (1934) described C. pilosus insularis on the basis of a single worker from Maria Madre Island (Tres Marias group), Nayarit, Mexico. Kempf (1958) showed that this ant was not allied to C. pilosus Emery, but belonged in his rohweri- subgroup, which also included C. wheeled. Since the location of the unique type of C. insularis was unknown, he provisionally accorded it species status, but intimated it might prove to be either C. rohwed or C. wheeled. C. insularis supposedly differed in the more prominent lateral pronotal teeth and the slightly emarginate anterior gastral border. In 1964 I spent a day at the California Academy of Sciences, where the type evidently had been originally placed, in a futile effort to locate the missing type. In the Academy’s type drawers is a unit tray with the original name of this ant, but no specimen. The type record carries the notation “type lost.” A careful search through both the identified and unidentified formicid collections failed to produce the missing specimen. Since Kempf indicated that the type of C. pilosus insularis was not in the Wheeler Collection at the Museum of Comparative Zoology, it seems safe to assume that the specimen is no longer 1968 Mexican Cephalotine Ants 5 extant. Accordingly, I have designated a NEOHOLOTYPE specimen from material at hand, and have selected a worker specimen taken on Maria Magdalena Island (Tres Marias group), Nayarit, March 25, 1964, ex colony No. 36425-c (R. R. Snelling)2. This specimen is deposited in the collections of the Los Angeles County Museum of Natural History. Kempf (1958) separated the workers of C. rohweri and C. wheeled as follows: “Lower face of head longitudinally striato-rugose; frontal carinae testaceous and semitranslucid wheeled Forel “Lower face of head reticulate-rugose; frontal carinae partly infuscated and solid rohwed Wheeler” The workers of the insular form, when run through Kempf’s key, will go directly to C. wheeled with no difficulty. In his discussion of C. wheeled, Kempf indicates additional characteristics by which the two species may be separated; the C. insulads workers differ from C. rohweri in exactly the same characters. As pointed out above, Kempf allowed C. insulads to stand because of two apparently definitive characteristics which seemed to be at variance with the other species. The first of these is the more prominent lateral pronotal teeth. Wheeler, in his original description, stated that this form differs from “. . . typical pilosus in having the three lateral teeth of the pronotum longer and more acute.” Wheeler based his concept of C. pilosus, specimens of which he had not seen, on Emery’s description and figures of that species, and on a “related form” from Brazil. I must confess that I, too, have never seen C. pilosus ; however, the normally excellent figures by Kempf should provide a good idea of the thoracic configuration. I have compared my material of C. insulads closely with Kempf’s figures of C. pilosus, and fail to agree with Wheeler’s claim. The anterior and middle lateral teeth in C. insulads exhibit some variation, but no individuals have these teeth any longer or more acute than the corresponding teeth in C. pilosus ; if anything, they tend to be some- what stouter. In many of the C. insularis workers the middle tooth is much reduced. The posterior tooth in the series before me is highly variable. It is safe to say, however, that on the whole it is better developed in C. insularis than in C. pilosus, if Kempf’s figure may be considered to represent the average condition in that species. Of the basal tergite Wheeler says “anterior border of the gaster at the articulation of the postpetiole less concave than in C. pilosus and without angular projections. . . In C. pilosus the first gastric segment at the articula- tion of the postpetiole is very decidedly concave, and on each side there are 2For the opportunity to collect ants on the Tres Marias Islands I am very deeply in- debted to Mr. Richard F. Dwyer of Newport Beach, California. Mr. Dwyer provided transportation to the Islands aboard his vessel, then the “Gringa’’ now the “Sea Quest!’ To Mr. and Mrs. Dwyer and the crew of the “Gringa” my sincere thanks. 6 Contributions in Science No. 132 rohweri insularis wheeleri Figure 2. Cryptocerus species, soldier (including dorsal aspect of head) and workers, respectively of C. rohweri, C. insularis and C. wheeleri. Figures by Ruth DeNicola Snelling. 1968 Mexican Cephalotine Ants 7 strong angular projections. On the basis of this statement by Wheeler, Kempf evidently was led to expect at least some indication of such a concavity in C. insularis. It does exist, but certainly not to the degree that Wheeler’s state- ment implied; the concavity, and it can barely be called that, is no more pro- nounced in my C. insularis samples than it is in C. rohweri or C. wheeleri. The above discussion intends to show that the characters by which Kempf separated the three species in this complex are highly variable. This by no means, however, can be considered evidence that there are no good distinc- tions between them. If Wheeler’s original description of C. insularis had been less vague, much of the difficulty which Kempf encountered could have been avoided. Through the courtesy of M. R. Smith and D. L. Smith I have been able to examine the U. S. National Museum’s material of C. wheeleri; Professor Creighton has sent me samples of C. rohweri. With these specimens, plus my material of C. insularis, it is now possible to restudy this complex and to affirm the specific nature of all three forms. The majors of the three species present the most obvious and consistent distinction. That of C. rohweri differs from both C. insularis and C. wheeleri in the presence of a fringe of setae along the lower margins of the cephalic disc; no indication of this fringe is present in available material of the other two. Furthermore the major of the first species has coarser and more close-set foveolae on the cephalic disc; these foveolae, and the spaces between them, are distinctly granulose, and the foveolae lack flattened silvery hairs. Creighton {in litt.) has noted these distinctions between C. rohweri and C. insularis, and the cotypes of C. wheeleri available to me show they will separate the former from that species as well. The floor of the cephalic disc is similarly shaped in C. rohweri and C. insularis ; when viewed from directly above, it is strongly humped in the middle. In C. wheeleri, the disc is essentially flat all the way across when viewed in this manner. When the head of C. rohweri is viewed from the side, the rim of the cephalic disc appears much more pronounced above than in most majors of either of the other two species. One of our specimens of C. insularis, however, has the rim very nearly as well-developed as in C. rohweri. This same speci- men, and one other from the same colony, has the cephalic foveolae nearly as large and close as they are in C. rohweri; but these foveolae all possess the flattened, shining hairs which are absent in C. rohweri and the interspaces are not granulose and dull as in that species. In C. wheeleri the transverse pronotal carina is rather strongly humped on either side of the pronounced median excision. In C. insularis and C. rohweri the carina, when viewed from the front, is distinctly sinuate, but the median excision is barely detectable, and the carina is not humped on either side. In C. rohweri the lateral spines of the petiole and post-petiole are longer and sharper than in the other two species. 8 Contributions in Science No. 132 The integument of C. rohweri, including that of the cephalic and thoracic foveolae, is everywhere granulate, imparting to the entire insect a dull appear- ance which is not much offset by the few shining hairs in the thoracic foveolae. This is in rather strong contrast to the condition of both C. insularis and C. wheeleri, in which the integument, while tessellate between the foveolae, is not at all granulate and hence somewhat shining. In addition, each foveola possesses a shining, flattened hair at its bottom, imparting a further luster to the insect. The minors are by no means as readily separable; differences do exist, but they are more subtle and, apparently, subject to greater variation. As is the case with the majors, the minors of C. rohweri differ from those of the other species in the more distinctly granulose integument. In this character, however, the distinction is one of degree, and accordingly difficult to appre- ciate unless all three species are available for comparison. However, in both C. insularis and C. wheeleri, the sides of the thorax and the posterior surface of the epinotum, while conspicuously tessellate, are nonetheless moderately shining; in C. rohweri these areas are granulate and dull. The gaster of the latter species is conspicuously duller than is the case with the other two. This is especially obvious along the sides of the first gastric tergite. The underside of the head of C. rohweri is reticulate-rugose (as pointed out by Kempf), while in the other two species it is striato-rugose. Furthermore, in the latter two species the frontal carinae are testaceous and semitransluscent, while in C. rohweri they are somewhat thickened and partially infuscated. The minor of C. rohweri is relatively easily separated from C. insularis and C. wheeleri ; the latter two species, however, are less readily separated from one another. The two principal distinctions which I have noted, and which I use in the key below, are of questionable validity, since only two minors of C. wheeleri are available for study. In C. insularis the maximum head width, at the upper margin of the eyes, is slightly less than the maximum length, the mandibles excluded. The two cotypes of C. wheeleri both have the maximum head width slightly greater than the maximum head length. The longitudinal rugulae of the promesonotum of C. insularis are rather regularly spaced, and are basically parallel to one another, not noticeably convergent anteriorly. In C. wheeleri, on the other hand, these rugulae are irregularly spaced, not essentially parallel with one another, and are definitely convergent anteriorly. The three species are allopatric in distribution: C. rohweri is known from the mountain ranges of southern Arizona and northwestern Mexico3, C. wheeleri only from the types taken at Cuernavaca, Morelos, Mexico. C. insu- laris has been taken on the Tres Marias Islands, Nayarit, and on the coastal lowlands near Mazatlan, Sinaloa, Mexico. 3The presence of this species in Mexico is indicated by a single female taken 13.6 miles west of Alamos, Sonora, on July 17, 1963 by R. L. Westcott. 1968 Mexican Cephalotine Ants 9 I consider the two specimens taken in quarantine and recorded by Kempf (1958: 134) as C. wheeleri to be examples of C. insularis', both were from unknown localities in Mexico. These two specimens have been examined and compared with my material of C. insularis, with which they agree quite closely. In all the characters discussed above, they coincide with C. insularis rather than C. wheeleri . The removal of these specimens from the records of C. wheeleri leaves only the original type series to represent that poorly known species. The following key is intended to supplement that of Kempf (1958) in securing separations of the three species involved, since he was unable to include C. insularis. KEY TO MEMBERS OF CRYPTOCERUS WHEELERI COMPLEX 1. Majors 2 Minors 4 2. Lateral projecting lobe of mesonotum angulate or dentate; rim of cephalic disc without projecting setae 3 Lateral projecting lobe of mesonotum broadly rounded, not angulate or dentate; rim of cephalic disc with fringe of projecting setae C. rohweri Wheeler 3. Seen from above, floor of cephalic disc strongly humped in middle; transverse pronotal carina lacking distinct median excision C. insularis Wheeler Seen from above, floor of cephalic disc flat, not at all humped in middle; transverse pronotal carina sharp, strongly excised medially C. wheeleri Forel 4. Genal area longitudinally striato-rugose; frontal carinae testaceous and semitransluscent 5 Genal area reticulate-rugose; frontal carinae thickened, partly infus- cated C. rohweri Wheeler 5. Maximum head width, at upper margin of eyes, slightly less than maxi- mum length (mandibles excluded) ; rugulae of promesonotum regularly spaced, parallel, not convergent anteromedially C. insularis Wheeler Maximum head width slightly greater than greatest length (mandibles excluded); rugulae of promesonotum more irregular, not essentially parallel, distinctly convergent toward anterior middle C. wheeleri Forel 10 Contributions in Science No. 132 Literature Cited Creighton, W. S., and W. L. Nutting 1965. The habits and distribution of Cryptocerus rohweri Wheeler. Psyche, 72: 59-64. Kempf, W. W. 1951. A taxonomic study on the ant tribe Cephalotini. Revista de Entomologia, 22: 1-244. 1958. New studies of the ant tribe Cephalotini. Studia Entomologia, New ser., 1: 1-176. Wheeler, W. M. 1934. Ants from the islands off the west coast of lower California and Mexico. Pan-Pacific Ent., 10: 132-144. 7. 73 LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE JMBER 133 February 20, 1968 A MANDIBLE OF DIDELPHODON VORAX ( M ARSUPI ALI A, MAMMALIA) By William A. Clemens, Jr. ^THSOW/^ tuples J-/eRARVt£ I Los Angeles County Museum of Natural EIistory Los Angeles, California 90007 Exposition Park CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957, The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8V2 x 1 1 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor A MANDIBLE OF D1DELPHODON VORAX (MARSUPIAL! A, MAMMALIA) By William A. Clemens, Jr.1 Abstract: A fragment of a mandible of Didelphodon vorax found in the Hell Creek Formation, Montana, confirms the provi- sional identification of lower premolars and molars of this species, establishes the orientation of the lower premolars, and serves as the basis for further speculation on the orientation of the upper premolars. The functional significance of prominent lateral lobes on mammalian premolars is examined. Introduction Didelphodon is the largest of the Late Cretaceous marsupials and prob- ably the largest mammal known from any Mesozoic fauna. In addition to its size, the striking modifications of its dentition raise questions concerning the role of Didelphodon in the reptile-dominated terrestrial environments of the Cretaceous. When I prepared my study of the marsupials of the Late Cretaceous Lance local fauna (Clemens, 1966), approximately 200 isolated teeth but only seven mandibular fragments of D. vorax were available for study. None of these mandibular fragments is as complete as that included in the type of Didelphodon padanicus from Late Cretaceous deposits in South Dakota, which contains only two undamaged although heavily worn teeth, the posterior premolar and a questionably associated molar. Also available for comparison is a fragment of an edentulous mandible from the Hell Creek Formation that has been described by Simpson (1927) and can be provisionally identified as representing a species of this genus. The fossil found in the Hell Creek Formation and described in this paper is the most nearly complete mandible of Didelphodon yet discovered. It provides sig- nificant information correcting and extending previous restorations of the mandible and dentition of Didelphodon. Acknowledgments The mandible of Didelphodon described here was found by Mr. Harley J. Garbani, a member of a field party from the Los Angeles County Museum of Natural History that collected fossil vertebrates from the Hell Creek Formation of Montana in the summer of 1965. This work was made possible by a generous gift from Mr. and Mrs. William T. Sesnon, Jr. I am greatly indebted to Dr. J. R. Macdonald of the Los Angeles County Museum of Natural History (LACM) who made the fossil available to me for study. 1Museum of Natural History and Department of Zoology, University of Kansas, Lawrence. Current address: Museum of Paleontology, University of California, Berkeley. 1 2 Contributions in Science No. 133 My research is part of a study of North American Late Cretaceous mam- mals supported by a grant from the National Science Foundation (GB- 5121) and the Museum of Natural History of the University of Kansas. The illustrations were prepared by Mr. Merton Bowman, to whom I am par- ticularly indebted for help with the restoration of the mandible. I also express my thanks to the following individuals and institutions for the opportunity to study materials in their care: Donald E. Savage, Museum of Paleontology, University of California, Berkeley (UCMP); Seth B. Benson, Museum of Vertebrate Zoology, University of California, Berke- ley (MVZ); J. Knox Jones, Museum of Natural History, University of Kansas (KU); Donald E. Russell, Institut de Paleontologie, Paris; and to Jason A. Lillegraven for his constructive criticism of the manuscript. Material Locality. Dr. J. R. Macdonald provided the following information: Lower Hell Creek no. 1. Late Cretaceous. Hell Creek Formation. In purple to gray clay and sandy silt, crusty brownish-gray with yellow and brown spots, also some medium-size, round sand concretions and carbonized wood resembling charcoal. Maloney Hill Quadrangle, Garfield County, Montana. Additional locality data are on file at the Los Angeles County Museum of Natural History and available to qualified investigators. Description: The fossil (LACM 15433) is a fragment of a left man- dibular ramus lacking that part of the symphyseal region anterior to the canine. Posteriorly it is broken at the level of the anterior edge of the coronoid process. The root of the canine and alveoli of P4 are present. P2, its apex slightly blunted by wear, and P3, which shows less wear, are pre- served. Mj-2 are missing but M3, which has large wear facets on the trigonid and talonid, is preserved. Only the trigonid of M4, less worn than that of M3, remains. The symphysis extends posteriorly to a point below the an- terior edge of P3. Two mental foramina are present, one below P2-3, the other below M^. Measurements of the fossil are given in Table 1. Table 1 Measurements of LACM 15433 in millimeters Dentition P2 P3 m3 m4 Length 6.5 7.3 6.5 Width 5.4 4.8 4.8 (trigonid) 4.5 (talonid) Dentary 5.4 (trigonid) Below P3 Below M3 Height 14.7 ca. 19. Width 8.6 ca. 8. 1968 Mandible of Didelphodon 3 c Figure 1. Didelphodon vorax, LACM 15433: left mandibular ramus with P2.3, Ma, and trignoid of M* preserved; a, occlusal; b, lingual; and c, labial views; all X 2. 4 Contributions in Science No. 133 Comments In my review of the marsupials of the Lance Formation (Clemens, 1966), two species of Didelphodon, the type species D. padanicus and D. vorax, were recognized. Fossils from the Lance Creek area demonstrate D. vorax has three lower premolars of which P4 is the smallest and P3 the largest. The evidence afforded by the only known mandible of D. padanicus does not clearly document its premolar structure. What remains of the pre- molars indicates the presence of either four premolars or three premolars with the second being distinctly larger than the last. Differences between species were also observed in morphology of the presumed posterior upper premolars and, possibly, in the smaller size of the teeth of D. vorax. The premolar morphology of LACM 15433 is of the type character- istic of Didelphodon vorax. Only three premolars are present. P2 is smaller than P3, although the difference is not so large as would be expected from the isolated teeth in the sample of the Lance local fauna. In only one of its dental dimensions (width of P2) does the fossil exceed the observed ranges of variation of the Lance sample. The characters of the molars and dentary also support allocation of the fossil to the species D. vorax. This discovery extends the range of the species to include the Hell Creek Formation of eastern Montana. Restoration of the Dentition LACM 15433 permits confirmation of some points in my study (Clem- ens, 1966) of the lower dentition of Didelphodon vorax. These can be summarized as follows: Association of the lower molars and premolars pro- visionally allocated to D. vorax is now demonstrated. The lateral lobes of P2 and, no doubt, P4 are on the lingual sides of their crowns. The trigonid of Mx is labial to the posterior end of P3. Although the evidence is still not conclusive, the lower molars appear to have increased in size from Mx, the smallest, to M4. The observed range of the ratios of widths of trigonid/tal- onid derived from isolated teeth in the Lance sample thought to be M3’s, 1.07 to 1.13, includes the value of the ratio, 1.07, for the M3 in LACM 15433. Using LACM 15433 as a base and selecting teeth collected from the Lance Formation to replace those postcanine teeth missing from it, a partial restoration of the mandible (Fig. 2) has been prepared. The restoration of the canine is based on canines of Sarcophilus, and a large isolated canine, YPM 10667, from the Lance Formation described by Marsh (1892, PI. VIII, fig. 6) and tentatively referred to Didelphodon vorax. In my earlier paper an attempt was made to determine the orientation of the upper premolars of Didelphodon vorax, which are known only from isolated teeth. Because of the morphology and apparently regular pattern 1968 Mandible of Didelphodon 5 Figure 2. Didelphodon vorax; partial restoration of left mandible; a, occlusal, and b, lingual views; both X 1.8. and sequence in development of wear facets on these teeth, it was assumed the entire battery of lower premolars made contact with the upper pre- molars. Orientation of the lower premolars with their lobes on the lingual sides of the teeth was supported by some evidence. From study of the wear patterns it appeared to follow that the upper premolars were oriented with the lobes on their labial sides. The orientation was also supported by con- sideration of the general pattern of symmetry of mammalian dentitions. Although not fully disproving the proposed orientation, several points have come to my attention suggesting that the reverse orientation — the lobes of the upper premolars on the lingual sides of their crowns — is probably correct. The first restoration of the upper premolars of Didelphodon was pro- posed with the implicit assumption that development of prominent lateral lobes of Pjlfj was the result of selection for increase in the area of contact of the upper and lower premolars. Information obtained from a large series of skulls of Sarcophilus (in the Museum of Vertebrate Zoology) supports an alternative assumption. In gross morphology the lower premolars of Sarcophilus (Fig. 3) resemble those of Didelphodon and show the same pattern of wear. The 6 Contributions in Science No. 133 Figure 3. Sarcophilus harrisi, MVZ 127035: a, occlusal view, P1-2 and M1; b, lateral view of left upper and lower incisors, canines, premolars, M \ , and part of M2; c, oc- clusal view, Pi. 2 and Mr, all X 2.25. 1968 Mandible of Didelphodon 7 lower premolars of Sarcophilus do not contact the uppers when the molars are occluded. P2 lacks a lobe on the lingual side of its crown. This premolar is situated with the long axis of its crown forming an oblique angle with the midline of the palate and, when the molars are occluded, it lies close to the crown of Mx. In contrast, in those dentitions of Sarcophilus studied, a few P1^ have a small lingual lobe, but most Prs resemble the tooth illustrated here (Fig. 3), which has only a small lingual expansion bordered by a prominent cingulum. Because this cingulum is well removed from the apex of the crown of P2 when the molars are occluded, it cannot act to increase the area of contact of the upper and lower dentitions. The convex sides of brachyodont mammalian cheek teeth serve to protect the margin of the gingiva from damage through direct contact with hard or abrasive materials. Because the upper and lower premolars of Sarcophilus do not come into contact during occlusion of the molars, development of lobes and cingula on their crowns cannot be attributed to selection for increase of the occlusal area. The modifications probably were developed in response to selection for increase of the area protected by the premolar crowns or change in configuration of the crown to protect the margin of the gingiva. Similarity in premolar morphology and wear patterns of Sarcophilus and Didelphodon suggests the assumption that P4_2 of Didelphodon occluded with P12 could be erroneous. If these premolars of D. vorax did not occlude, the function of their lateral lobes could have been to provide pro- tection for the gingiva, and lobes on the upper premolars could have been on the lingual sides of their crowns. This function and orientation is favored by the presence of steeply inclined wear facets on the lobes of some pre- molars that might not have developed through contact of upper and lower premolars. Many extant mammals have premolars with bulbous crowns and, less frequently, lateral expansions or lobes. Restricting comparisons to mammals of approximately the same size — as far as can be determined from man- dibular dimensions — the closest resemblance in dental morphology to Didel- phodon is found in Sarcophilus and it is not close. Although the molars (Fig. 4) of the sea otter, Enhydra, and Didelphodon are distinctly different in function and morphology, their premolars show points of resemblance. The upper and lower first premolars of Enhydra usually are lost, but a small peglike P1 is present in either the upper right or left quadrant of some dentitions (2 of 12 individuals represented in the University of Kansas collections). P2 and P3, which do not occlude with upper premolars, have prominent lobes on the lingual sides of their crowns. The small lobe on the lingual side of P2 is dorsal to but does not occlude with P3. The lobe on the lingual side of P3 is larger than that of P2 and has a shallow basin on its surface that receives the apex of P4. 8 Contributions in Science No. 133 c Figure 4. Enhydra lutris, KU 44672: a, occlusal view, P--4 and M1; b, lateral view of entire dentition; c, occlusal view, P2.4, M4.2; all X 2.17. 1968 Mandible of Didelphodon 9 b Figure 5. Quercitherium tenebrosum, Collections de Paleontologie, Museum national d’Histoire naturelle, 1893-11: left maxillary fragment with P1-4 and M1-3; illustra- tions based on figures in Piveteau (1935) and cast of fossil; a, occlusal, and b, lingual views; both approximately X 2. Comparisons can be carried further among placental mammals. The recently established oxyaenoid family, Teratodontidae Savage, 1965, is com- posed of Teratodon and Quercitherium. Both have dentitions resembling the dentition of Didelphodon in the presence of massive premolars and some shearing molars. Apparently in one species, Teratodon enigmae, the first molars also have greatly modified, bulbous crowns. The bulbous teeth of Teratodon do not have the prominent lobes found in Didelphodon. Savage (1965: 252) suggests the upper and lower premolars of Teratodon occluded and were employed in a grinding action. With the exception of P1, which has a lingual lobe, the premolars of Quercitherium (Fig. 5) more closely resemble those of Teratodon than Didelphodon. The available casts and illustrations do not reveal whether the wear facets on the premolars of Quercitherium are the result of occlu- sion or apical, non-occlusal wear of the type described by Mac Intyre (1966: 123). However, its position relative to the crowns of other premolars sug- gests the lingual lobe of P1 functioned to protect the gingiva. Because of their position relative to the insertion of the masseter and 10 Contributions in Science No. 133 the condyle, the massive premolars of Teratodon would have obtained little mechanical advantage from the leverage of the mandibular ramus. Savage (1965) noted that functionally the dentition appears to be a compromise between shearing and crushing or grinding functions but one bringing out the worst attributes of both. In teratodontids and Didelphodon the lack of great mechanical advantage for the premolars could have been offset by the evolution of a massive masticatory musculature. This modification is found in the jaw musculature of Sarcophilus. As Macalister (1872: 18) noted, “The most expressive way of representing the enormous size of these muscles [of Sarcophilus ] is by stating that the weights of the muscles which elevate the lower jaw (masseters, pterygoids, and temporals) were equal to the sum of the weights of all the scapular and brachial muscles (deltoids, spinates, biceps, brachiales, triceps, & c.), or to the entire series of muscles which act on the shoulder-joint (pectorals, latissimus dorsi, spinati, deltoids, & c.).” Data obtained from mammals with premolars resembling those of Didelphodon suggest the following points concerning the function and orientation of this kind of premolar: Evolution of lateral lobes need not be a response to selection for increase of the occlusal area of the dentition, but may serve to change the configuration of the margin of the gingiva to give it greater protection or increase the area protected by the crowns of the premolars. Presence of these lobes does not necessarily indicate the opposing upper and lower premolars occluded. In all species in which the orientation of the teeth can be definitely established, the lobes are found only on the lingual sides of the upper and lower premolars. Summary A fragment of a mandible of Didelphodon vorax found in the Hell Creek Formation is described and used as the basis for a restoration of the lower dentition of the species. The morphology of the teeth preserved in this mandibular fragment and evidence from other sources suggests the pre- viously proposed orientation of the upper premolars was erroneous. Promi- nent lobes on some premolars of Didelphodon probably result from selection for increase of the area protected by the premolars or changes in configura- tion of the crown to protect the margin of the gingiva, not from selection for increase in the area of contact of the upper and lower premolars. Although not fully demonstrated, it now appears most likely that, like the lobes on the lower premolars, those on the upper premolars were also on the lingual sides of their crowns. 1968 Mandible of Didelphodon 11 Literature Cited Clemens, W. A., Jr. 1966. Fossil mammals of the type Lance Formation, Wyoming. Part II. Mar- supialia. Univ. California Publ. Geol. Sci., 62, 122 p., 77 figs. Mac Intyre, G. T. 1966. The Miacidae (Mammalia, Carnivora). Part I. The systematics of Icti- dopappus and Protictis. Bull. Amer. Mus. Nat. Hist., 131:115-210, 21 figs., 20 pis. Macalister, A. 1872. Further Observations on the Myology of Sarcophilus ursinus. Ann. Mag. Nat. Hist., ser. 4, 10: 17-20. Marsh, O. C. 1892. Discovery of Cretaceous Mammalia. Part III. Amer. J. Sci., Ser. 3, 43:249-262, pis. v-xi. Piveteau, J. 1935. Etudes sur quelques Creodontes des Phosphorites du Quercy. Annales de Paleontologie, 24:75-95, 12 figs., 2 pis. Savage, R. J. G. 1965. Fossil mammals of Africa: 19, The Miocene Carnivora of East Africa. Bull. Brit. Mus. (Nat. Hist.), Geol., 10: 241-316, 62 figs., 5 pis. Simpson, G. G. 1927. Mammalian fauna of the Hell Creek Formation of Montana. Amer. Mus. Novitates, 267:1-7, 6 figs. ,73 LOS t ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE '|JMBER 134 February 20, 1968 POPULATIONAL VARIATION IN THE FROG GENUS PHRYNOHYAS FITZINGER IN MIDDLE AMERICA By Roy W. McDiarmid :cpTOOft^ «rumi JUQFwmt Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8 V2 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS.— All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings) ; photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor POPULATIONAL VARIATION IN THE FROG GENUS PHRYNOHYAS FITZINGER IN MIDDLE AMERICA By Roy W. McDiarmid1 Abstract: A detailed populational analysis indicates that Phrynohyas latifasciata, inflata, and spilomma are all representa- tive of a single wide-ranging form, Phrynohyas venulosa. Impor- ant characteristics, including size, coloration, and skin texture, exhibit considerable variation and extensive overlap from popu- lation to population or within the same population. The largest specimens of Phrynohyas occur in the more arid portions of its range. Large size is considered an important mech- anism for increasing the frogs’ efficiency in unfavorable habitats and allowing them to cope more effectively with the problems of desiccation. The degree of development of the dermal glands cor- relates with the annual climatic cycle characteristic of the wet and dry seasons in Costa Rica. Glandular secretions of Phryno- hyas venulosa are considered a secondary adaptation against des- iccation, as well as being important in deterring predators. During the summer months of 1962, a field party from the University of Southern California conducted an extensive survey of the lowland amphibian and reptilian faunas of northwestern Mexico. On August 3 a series of nine large hylid frogs, genus Phrynohyas, was obtained from a rain-filled pond approximately 9 miles south of Escuinapa, Sinaloa. The frogs, eight males and one female, were located by their calls. Apparently this species was just initi- ating breeding activities, as only one pair was in amplexus. A second series of these frogs was collected on July 1 in Nayarit by a field party from California State College at Long Beach. This second series, 18 males and one female, was located by the loud chorus at a pond situated 0.1 mile west of the junction of Mexico Highway 15 and Highway 46, in Nayarit. For the past eight years field parties from the University of Southern California have collected amphibians and reptiles in Costa Rica. In 1964, 1966, and again in 1967, I collected specimens of Phrynohyas in Guanacaste and Puntarenas Provinces of Costa Rica. Several of these specimens show striking similarities, particularly in color pattern, to specimens from western Mexico. The results of a preliminary examination of the two series from Sinaloa and Nayarit indicate the need for a re-evaluation of the present understanding of the systematic arrangement of the genus in western Mexico. Also, with the availability of this additional material from Costa Rica, it seems appropriate to attempt to elucidate the relationships among the named populations of Phrynohyas in Middle America. department of Biological Sciences, University of Southern California, Los Angeles, California 90007, and Research Assistant in Herpetology, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 134 Historical Review As Duellman (1956), in a monograph on the genus Phrynohyas, pre- sented an historical account that included complete synonomies and literature references for each species, a review of material published prior to 1956 will not be repeated. Duellman (1956:8) proposed that Phrynohyas be used to refer to those frogs that possess “paired lateral vocal sacs behind the angles of the jaws and without the skin of the head co-ossified with the skull.” The generic name Phrynohyas was adopted and, following Opinion 520 of the International Commission on Zoological Nomenclature (Hemming, 1958), has been used by most authors (Gans, 1960; Porter, 1962; Stuart, 1963; Zweifel, 1964; Savage, 1966). Rivero (1961:128-131) discussed Duellman’s proposal at some length and presented an alternative for treatment of frogs referred to Phrynohyas. He suggested that the species be maintained in the genus Hyla (see Rivero, 1961, for additional discussion). On the basis of the unique characters of the larvae of Phrynohyas (Zweifel, 1964:204-205) and the characters used by Duellman to diagnose this group of frogs, it seems reasonable to maintain the generic name Phryno- hyas. For the sake of nomenclatural stability, in accordance with Opinion 520, those frogs referred to Phrynohyas zonata (Spix) by Duellman (1956), and to Hyla tibiatrix Laurenti by Rivero (1961), are assigned to Phrynohyas venulosa (Laurenti). The five species of Middle American Phrynohyas recognized by Duell- man (1956) are: P. zonata from Costa Rica, Panama, and tropical South America; P. spilomma from southern Tamaulipas, Mexico, southward along the eastern coast of Central America to Nicaragua and along the Pacific coast from the Isthmus of Tehuantepec southward to Guatemala; P. modesta from El Salvador north along the Pacific coast of Chiapas through the region of the Isthmus of Tehuantepec into the Atlantic drainage; P. inflata from scattered localities along the Pacific coast of Mexico from central Guerrero north to Colima; and P. latifasciata from Presidio, Sinaloa, Mexico. Shannon and Humphrey (1957) described Phrynohyas corasterias from a single young female collected 4.8 miles east of San Bias, Nayarit. This species was diagnosed on the basis of pattern and coloration. The only known speci- men of P. corasterias was collected from a locality approximately midway between the northernmost locality for P. inflata and the type locality of P. latifasciata. Duellman (1961:45-46) discussed the characters used to diagnose P. corasterias and placed it in synonomy with P. inflata. He suggested that additional material from northwestern Mexico might show that P. inflata and P. latifasciata are conspecific (Duellman, 1956:21; 1961:45). Examination of additional material from Guatemala convinced Duellman that the uniform brown or tan dorsum characteristic of P. modesta is only a color variant of P. spilomma, and accordingly he placed P. modesta in the synonomy of P. spilomma (Duellman, 1966:277). 1968 Variation in Phrynohyas 3 Methods and Materials As presently understood, the species of Phrynohyas are distinguished by differences in size, body proportions, coloration, and pattern. To supplement the data gathered by Duellman from more than 350 specimens, I examined an additional 153 specimens of Phrynohyas from localities in western Mexico, Guatemala, Honduras, Costa Rica, and Panama. Seven measurements, includ- ing body length, tibia length, foot length, head width, head length, interorbital distance, and internarial distance, were taken on each suitable specimen; five proportions were calculated from these measurements. These data are con- tained in Table 1 . In order to standardize the measurements, only specimens larger than 45 mm in body length and capable of breeding are included. This procedure reduces the possibility of allometric growth in subadult stages confusing the data. The measurements taken are the same as those used by Duellman (1956:6-7). All measurements are in millimeters and were made with Helios dial calipers. Each available specimen was examined in detail with reference to skin texture, coloration, and pattern, because these characters, as presently understood, are the primary features used to distinguish between the species (see key to adult Phrynohyas in Duellman, 1956:43-44). Analysis of the Populations Western Mexico. — The northernmost representatives of the genus were collected at Presidio, Sinaloa (Boulenger, 1882). Examination of these two specimens led Duellman (1956:24-25) to describe P. latifasciata. The Sinaloan species was described as being most closely related to P. inflata, a species known from six males collected from localities in Colima, Michoacan, and Guerrero to the south. The characters he used to distinguish between P. latifasciata and P. inflata were differences in size, skin texture, coloration, and pattern. The utilization of size in diagnosing a species of frog known from only two specimens is open to criticism. Furthermore, in the case of P. latifasciata, there is no mention of the time of year the frogs were collected nor of the breeding condition of the specimens. The paratype of P. latifasciata is a young specimen. If it was collected at the same time as the holotype, as presumed by Duellman (1956:24), then I suspect that the specimens were taken before the breeding season. The holotype may be an adult, but if it is sexually mature, it probably is a young adult. This supposition is supported by the size range of males taken less than 50 miles to the south of the type locality (Table 1 ) . At this locale, breeding males measuring from 74.2 to 90.2 mm were collected. In addition, a male (UMMZ 108019) referred to P. inflata, measures 63.5 mm, approximately 5 mm less than the type of latifasciata. The size of the vocal sacs as depicted in the illustration of the holotype of P. latifasciata (Duellman, 1956 :pl. Ill, fig. 1), when compared to those of breeding males from Sinaloa 4 Contributions in Science No. 134 ! Table 1 Measurements and Proportion >: P. latifasciata1 if $ 9 Sinaloa 3 S 9 9 Nayarit $ S 9 9 Colima-Guerrero SS 9 9 Oaxaca-Veracru j S$ 9iji Number of Specimens 1 0 6 1 17 1 5 0 4 Body Length mm mean — — 80.9 — 84.9 — 80.3 — 65.6 5*1 range 68.0 — 74.2-90.2 89.0 81.1-92.6 95.8 63.5-92.0 - 55.9-69.0 47.4 Tibia Length mm mean — — 35.7 — 39.7 — 34.5 — 30.3 2'. range 30.0 - 33.5-39.3 39.5 37.9-42.8 42.9 32.0-42.0 - 26.0-33.0 24.1 j Foot Length mm mean — — 32.1 — 35.0 — 29.7 — 24.4 21 range 28.5 - 30.0-35.1 35.5 31.0-37.1 40.1 26.0-32.0 - 21.2-26.0 19.4} Head Length mm mean 23.8 25.0 23.6 19.5 T range 20.5 - 22.5-26.2 25.7 23.0-26.6 28.1 20.0-27.0 - 15.8-21.0 15.4: Head Width mm mean — — 25.8 — 27.0 — 26.0 — 20.5 n range 22.0 - 24.0-27.8 28.5 26.0-28.2 30.5 21.0-29.0 - 16.5-23.0 15.8’ Interorbital Distance mm mean — — 7.5 — 7.8 — 7.5 — 5.7 5 range 6.0 - 6.9-8. 2 9.5 7.0-9.4 8.8 5. 5-9.0 - 4.6-7.0 5.1 ( Intemarial Distance mm mean — 5.6 — 6.2 — 6.2 — 4.9 4 range 4.0 - 4.9-6. 1 6.2 5. 5-6.7 6.5 5.5-7.0 - 4. 1-6.0 35 % Tibia Length/ Body Length mean — — 44.2 — 46.8 — 47.6 — 46.2 4;i range 44.1 - 42.5-45.6 44.4 44.6-48.4 44.8 45.6-50.7 — 44.1-47.8 49.1 % Head Length/ Body Length mean — — 29.5 — 29.4 — 32.5 — 29.6 3| range 30.1 - 27.6-31.5 28.8 28.0-31.3 29.3 31.5-33.3 — 28.3-30.4 32.8; % Head Width/ Body Length mean — 32.0 — 31.8 — 29.6 — 31.1 31 range 32.3 - 30.3-33.5 32.0 29.4-35.0 31.8 27.7-31.5 — 29.2-33.3 31.(4 % Interorbital Distance/ Head Width mean — — 29.0 — 28.9 — 28.7 — 28.0 31 range 27.3 - 27.0-31.2 33.3 25.8-33.3 28.8 26.2-31.0 — 26.1-31.1 28.: % Internarial Distance/ Head Width mean - — 21.7 — 23.3 — 24.0 — 24.0 2 range 18.2 — 18.8-23.1 21.7 20.4-25.3 21.3 22.2-26.2 — 21.7-26.7 24. J 1 : Data from Duellman, 1956. 1968 Variation in Phrynohyas 5 Table 1 Viynohyas from Middle America . spilomma1 f 9 9 P. modesta1 S$ 99 Guatemala S $ 9 9 Honduras $$ 9 9 Costa Rica $ S 9 9 Panama SS 99 >9 85 13 14 53 20 1 4 2 9 4 * .7 67.7 63.7 63.5 63.6 64.6 66.2 81.5 77.5 88.9 54|87.0 56.0-86.5 54.0-69.5 52.0-80.5 54.5-71.1 60.5-79.3 67.0 62.7-68.5 74.3-88.6 54.5-94.1 84.0-92.0 82.0 .8 32.6 31.0 31.7 30.2 31.1 _ 32.9 38.1 37.2 41.8 2: 40.0 27.0-41.0 25.5-33.5 26.0-40.0 28.0-34.2 28.8-36.4 31.3 30.9-34.0 34.4-41.7 29.6-45.8 40.6-43.5 39.4 .9 27.8 25.5 26.5 24.8 25.8 27.3 32.6 31.7 35.9 _ 1! 35.0 22.5-34.5 20.5-29.0 21.5-32.5 22.5-30.2 23.5-30.8 26.5 24.5-28.4 29.6-35.5 26.0-35.7 32.5-38.3 34.5 ] 20.7 19.5 19.5 19.2 19.8 20.0 24.0 23.8 25.2 _ lil 26.0 16.5-27.0 16.5-22.5 17.0-23.0 17.3-21.9 18.8-21.7 20.8 19.1-20.9 22.4-25.5 18.9-27.7 23.5-26.0 23.8 1.2 22.3 20.8 20.9 19.3 20.2 20.7 25.1 25.4 27.0 _ 1(27.0 19.0-27.0 17.5-23.0 18.0-26.0 17.5-21.4 18.5-22.8 20.8 20.1-21.4 22.5-27.7 19.8-30.0 25.0-28.6 25.3 J 6.1 5.6 5.8 5.6 6.2 6.1 6.5 6.3 8.1 _ 8.0 5.0-7.5 4.5-6.5 5.0-7.0 4.5-6.6 5.4-7. 8 5.5 5.9-6.6 6.0-7 .0 4.0-7.0 7.S-9.2 7.5 9 4.9 4.7 4.8 4.8 5.1 4.8 6.0 5.5 6.6 _ 6.5 3. 5-6.0 4.0-5.0 4.0-5.0 4.2-5.8 4.5-5.8 4.7 4.7-5. 1 6.0-6.0 4.S-6.5 6.5-7.0 5.9 •8 47.7 48.7 49.9 48.5 48.2 49.7 46.7 48.2 47.0 4-51.6 42.9-52.5 46.8-51.2 46.8-55.9 43.9-53.0 45.9-50.4 46.7 49.3-50.3 46.3-47.1 45.2-52.1 45.1-48.9 48.0 ).l 30.4 30.6 30.8 29.8 30.8 _ 30.2 29.5 31.3 28.3 2!: 33.3 27.0-33.1 28.8-32.6 28.6-32.8 27.7-32.8 26.0-32.7 31.0 29.2-32.0 28.8-30.1 26.9-43.1 27.6-29.0 29.0 .4 32.6 32.6 32.9 30.0 31.3 _ 31.3 30.8 33.5 30.4 2-36.4 29.2-36.0 31.4-34.3 31.2-35.1 27.0-33.8 28.8-32.4 31.0 29.5-32.8 30.4-31.3 29.4-50.8 29.5-31.9 30.8 ’.5 27.3 27.1 28.0 28.7 30.3 29.6 26.0 25.4 30.0 2 -33.3 22.2-34.3 25.6-30.6 25.5-32.4 24.2-34.1 25.9-35.1 26.4 28.0-32.7 25.3-26.7 14.4-29.8 26.2-34.6 29.6 ’.0 21.9 22.8 22.9 25.0 25.1 23.4 24.2 22.1 24.6 1 ■26.1 18.2-26.3 19.2-23.9 19.0-26.5 22.2-28.1 22.9-27.1 22.6 22.3-25.2 21.7-26.7 16.2-25.3 22.7-26.0 23.3 Contributions in Science No. 134 1968 Variation in Phrynohyas Table 1 Table 1 Measurements and Proportions of phrynohyas from Middle America 6 Contributions in Science No. 134 Figure 1. Series of Phrynohyas venulosa collected 9.4 miles south of Escuinapa, Sinaloa, Mexico. These frogs exhibit both unicolor and blotched color patterns. and Nayarit, also suggests that the specimen is a young male. If the Sinaloan specimens referred to P. latifasciata represent a population distinct from P. inflata, then differences in proportions might be anticipated. It is apparent from Table 1 that the additional material from southern Sinaloa and Nayarit is intermediate between P. latifasciata and P. inflata in measurable characters. Although only two adult females of Phrynohyas are known from western Mexico, data suggest that females attain a larger size than males. The development of the dorsal pustules in the Phrynohyas from Sinaloa and Nayarit varies considerably. The series includes animals with a nearly smooth dorsum, animals with moderately developed pustules, and animals with well developed pustules. The range of variation in this character en- compasses the differences attributed to latifasciata and inflata. A specimen from Nayarit (CSCLB 641) that was preserved in alcohol has a relatively smooth skin. Additional specimens (CSCLB 625-640) collected from the same locality and on the same night were fixed in ten percent formalin and then preserved in alcohol. These specimens exhibit a wide range of variation in skin texture but in all cases are more rugose than the specimen initially preserved in alcohol. This suggests that the type of latifasciata was fixed in alcohol, apparently a common practice in the late 1800s. The differences in coloration and pattern between species of Phrynohyas, as pointed out by Duellman (1956), include the nature of the dorsal color 1968 Variation in Phrynohyas 7 pattern and the arrangement and size of the transverse leg bands. These differences also form the primary basis for the key to the adult Phrynohyas (Duellman, 1956:43-44). A series of seven specimens from Sinaloa (LACM 6314-19, 7245) exhibits a range of variation in dorsal color pattern encom- passing three of the nominal species of Phrynohyas from Mexico (Fig. 1 ) . Five of the specimens have a pattern similar to the unicolor form of P. spilomma , originally considered distinct and referred to P. modesta by Duellman (1956). Dorsally these specimens are mottled brown on tan, or dark brown on brown. Two of the specimens (D and E) show a slight indication of a darker mid- dorsal color, a condition considered intermediate between the unicolor phase and the blotched forms (F and G) . The upper surface of the thighs is unicolor or speckled dark brown (A through E). Duellman (1956:27) pointed out that the unicolor brown dorsum and the speckled condition of the hind legs is typical of specimens of P. modesta found in southern Mexico and Guatemala. A second dorsal color pattern (F) consists of a light brown ground color with a large chocolate brown patch extending posteriorly from behind the eyes to near mid-body. A broad band of ground color bounds a posterior brown patch that continues to the vent and onto the dorsal surface of the thighs. The brown bands on the legs are separated by tan interspaces. This color pattern is typical of the specimens from Colima and Guerrero referred to P. inflata and shows striking similarities to the specimen from Michoacan figured by Duellman ( 1956: pi. II, fig. 1 ) . Specimen G has a dorsal color pattern similar to specimen F but differing in having the anterior dorsal patch olive brown and broken into two parts rather than forming a continuous central patch. The middle spot is connected by a narrow band to the posterior blotch. The presence of two or more anterior dorsal spots is characteristic of the nominal species Phrynohyas latifasciata (Duellman, 1956:25, pi. Ill, fig. 1). The bands on the legs are not as wide as in the type of latifasciata (three times the width of the interspaces) but rather are very much like the condition of specimen F and the type of P. inflata. The vocal sacs of the Sinaloan frogs in life varied from chocolate brown to dark olive. With the differences in coloration of the vocal sacs noted in the Sinaloan series, it is not surprising to find that Boulenger (1882:327- 328) described the type of P. latifasciata as possessing black vocal bladders. In life, the ventral surfaces of the Phrynohyas from Sinaloa vary from a dirty white to brownish white. The throat is usually creamy white with or without brown vermiculations. In the series of specimens from Nayarit (Fig. 2), the anteriodorsal blotch is generally continuous in most specimens. However, one individual (A) has the anterior blotch broken into two parts. The nature of this break in the blotch is easily traced to the solid blotched pattern through intermediate specimens (B, C, G). In addition to the split of the anterior central patch, there is considerable variation in other specimens in the shape of the blotch. Certain individuals have a rectangular shaped blotch (E, H) ; some exhibit an 8 Contributions in Science No. 134 hourglass-shaped blotch (J, N). Some have posterior indentations of the blotch (K, L), anterior indentations of the blotch (B, G), or both (C, I, M). The color of the blotch varies from a uniform chocolate (E) to a mottled light brown blotch that is darker laterally (N). Some individuals have isolated circles of ground color within the blotch (F, G) . In two specimens (M and N) the anterior and posterior blotches connect, while in others (L and O) the two dorsal blotches are almost completely fused. In addition to the variation in dorsal pattern, the Nayarit series (Fig. 2) also exhibits a wide range of variation in color pattern of the legs. Some specimens have continuous bands on the femur and tibia of both hind legs (A, E, N). In some the bands are four times the interspace width (J, O); in some they are only twice the interspace width (B, C). There are progressive stages from specimens with nearly unicolor legs (L) through an individual with one tibia solid and one banded (H), or both tibiae solid (F, G), to specimens on which the interspaces do not meet across the face of the tibia on Figure 2. Selected individuals from a series of Phrynohyas venulosa collected 0.1 mile west of the junction of Mexico Highway 15 and Highway 46 (=22.9 miles east of San Bias), Nayarit, Mexico. The specimens show variation in dorsal color pat- tern and leg pattern. 1968 Variation in Phrynohyas 9 either one side (K) or both sides (J). The leg markings vary from nearly unicolor (D, O) to a banded condition with varying degrees of spotting or marbling with a darker color (I, K). Most, but not all, of the leg bands are edged in dark brown. The Nayarit specimens possess a brownish white to creamy white venter, generally with some brown marking on the chin. The vocal sacs vary from light brown and olive to dark brown. I examined the six specimens previously called P. infiata from western Mexico and found no differences between these specimens and the Nayarit or Sinaloa material, either in measurements or in proportions (Table 1) . In terms of coloration and pattern, the six known specimens of P. infiata are almost exactly duplicated in the series from Nayarit. A single specimen (KU 73879) taken approximately halfway between the localities of the Nayarit and Sinaloan series possesses characteristics typical of some of the Nayarit specimens as well as of one of the Sinaloan specimens. The description and illustration of P. corasterias (Shannon and Humph- rey, 1957:15-18) indicate that this nominal species shares characters of color and pattern with some of the Nayarit material. In addition, the condition of the subarticular tubercle on the penultimate joint of the fourth finger (toe) varies from a bifid condition to a typical round condition in both the Nayarit and Sinaloan material. The webbing characters appear to be constant in all specimens examined from western Mexico and, in general, are the same as the illustration of the hand and foot of P. corasterias (Shannon and Humphrey, 1957:17, fig. 2). The other characters supposedly diagnostic of P. corasterias are well within the range of variation found in the Nayarit and Sinaloan series. Duellman (1961:45-46) considered the holotype of P. corasterias as a repre- sentative from the population of frogs from Colima, Michoacan, and Guerrero. The new material from Sinaloa and Nayarit indicates that there is extensive overlap between the characters used to distinguish between P. latifasciata and P. infiata and argues for recognition of but a single species of Phrynohyas in western Mexico. Southern Mexico-Northern Central America. — The Phrynohyas from southern and eastern Mexico and northern Central America were considered by many workers to represent two species, P. spilomma and P. modes ta (Smith and Taylor, 1948; Duellman, 1956, 1960; Neill and Allen, 1959a, 1959b; Neill, 1965; Stuart, 1963). Recently, Duellman (1966:277) presented evi- dence to show that P. modesta was a color variant of P. spilomma. Examination of material from Mexico, Guatemala, and Honduras augments the conspeci- ficity of P. modesta and P. spilomma. In his monograph of the genus in 1956, Duellman examined 335 speci- mens of P. spilomma. I have examined an additional 85 specimens. The measurements and ratios for the adult specimens are listed in Table 1. It is immediately obvious from the measurements that presumed P. spilomma generally average smaller than specimens of Phrynohyas from western Mexico. 10 Contributions in Science No. 134 Figure 3. Part of a large series of Phrynohyas venulosa from Cuyuta, Guatemala. These specimens show a gradation from unicolor to blotched dorsal pattern. However, it should be pointed out that there is overlap in the ranges of all measurements of the frogs from southern and eastern Mexico and west Mexican Phrynohyas. In addition, the averages of the ratios are nearly the same from Sinaloa, Mexico to Honduras. While there is considerable variation in the measurements and proportions of this sample, there is no consistent geographic variation within the sample. Besides the overlap in measurements and proportions, there is a wide range of variation in skin texture. No con- sistent differences in skin texture or glandular development exist to aid in distinguishing the west Mexican individuals from those referred to P. spilomma. The primary characters, other than overall size, that have been used to distinguish between eastern Mexican and Central American Phrynohyas and the Phrynohyas from western Mexico are differences in coloration and pattern. Examination of a large series of Phrynohyas from Cuyuta, Guatemala ( AMNH 74377-90, +58) indicates that all the pattern types characteristic of Phry- nohyas from Mexico and northern Central America may be present in a single breeding population. Selected individuals in this series (Fig. 3) exhibit a gradation from unicolor to blotched pattern. Some possess a unicolor brown dorsum with various degrees of spotting on the back and legs (A and B); 1968 Variation in Phrynohyas 11 some have faint traces of a dorsal pattern and leg barring (C and D) ; three individuals (E, F, and G) are heavily mottled dorsally, tending towards a uniform dorsal blotch similar to those frogs referred to P. spilomma, as figured by Duellman ( 1 956 :pl. IV, figs. 1 and 2); others (H, I, and J) exhibit various modifications of a dorsal pattern and barred legs; and two specimens (K and L) are nearly identical in pattern with specimens from Nayarit, Mexico (Fig. 2, D, F, N). Ventral coloration in these 12 specimens is uniform white or dirty white with some brown mottling on the throat, or white with numerous brown spots. Others have mentioned the color patterns characteristic of populations in a single area (Neill, 1965:88). Honduranean specimens were described by Duellman (1956:32) as closely resembling those from La Libertad, Guate- mala (Duellman, 1956: pi. IV, fig. 2) and considered typical of the nominal P. spilomma. Meyer (1966:173) reported five specimens from Honduras, all of which were unicolor. Thus in Honduras, as in Guatemala, British Honduras, and Mexico, Phrynohyas exhibits a wide range of color pattern within a single population or between populations. Costa Rica and Panama. — Material from lower Central America, al- though scanty, provides a basis for further understanding of the relationships between the populations of this wide ranging hylid frog. As previously men- tioned, both unicolor and blotched forms are known from Honduras. The only Figure 4. Costa Rican specimens of Phrynohyas venulosa showing the variation in coloration and size. 12 Contributions in Science No. 134 known specimen from Nicaragua has a blotched pattern typical of some of the Guatemalan individuals. All previously known specimens from Costa Rica and Panama exhibited a blotched dorsal pattern and banded legs. These frogs were assigned to Phrynohyas venulosa by Taylor (1952:800) and Zweifel (1964:201), and to Phrynohyas zonata by Duellman (1956:37). Several additional specimens from Costa Rica closely resemble the individual illus- trated by Duellman (1956:pl. V, fig. 2) from Palmar, Puntarenas Province, Costa Rica, and the individual illustrated by Zweifel (1964:206, fig. 5) from Nueva Gorgona, Panama Province, Panama. Thus it appeared that lower Mid- dle American Phrynohyas were generally consistent in their color pattern and large size and possibly represented a species distinct from the Phrynohyas from Honduras and areas to the north. In late May, 1964, four unicolor specimens of Phrynohyas were collected in the Atlantic drainage at Los Chiles, Alajuela Province, Costa Rica. These specimens are smaller than other Costa Rican individuals but larger than most specimens from Honduras and Guatemala. The measurements and proportions of all the specimens examined from Costa Rica and Panama are listed in Table 1. The specimens from lower Central America are generally larger than individuals of Phrynohyas to the north, but, as was true with the forms already discussed, there is overlap with the northern Central American forms in most of the measurements and very little difference, if any, among the ratios. Ten of the Costa Rican specimens that were examined are pictured (Fig. 4). The four Los Chiles specimens, three females and one male, range from dark brown to brownish tan dorsally (A through D). Three specimens have unicolor legs, while the fourth possesses many dark spots on the hind legs. Ventrally they are yellowish white with a faint brown mottling on the throat and belly. The dorsal pattern of the remaining six adults from Costa Rica consists of a wide dark brown blotch beginning between the eyes and extend- ing along the lateral edge of the dorsal blotch to about mid-body. The width of the lateral band varies from a narrow line (E and F) to a wide band (I and J) and determines the shape of the dorsal blotch. The legs of all the specimens are variously crossed by solid bands or a series of broken bands which give the legs an overall mottled appearance. Ventrally, these frogs are yellowish white with a faint brown mottling on the throat or covered with numerous brown spots. The skin texture varies from nearly smooth (A, C, and I) to very rugose and pustulate (B, E, and G). The ventral surfaces also show some variation in pustulation. The Panamanian specimens examined exhibit the pattern (Fig. 5) char- acteristic of most of the Costa Rican frogs. The dorsal blotches are somewhat narrower than in some of the Costa Rican individuals and the lateral bands from the eyes are correspondingly wider. Ventrally, these frogs show varying degrees of faint brown mottling. The intermediate size of the Los Chiles specimens and the presence of 1968 Variation in Phrynohyas 13 Figure 5. Panamanian specimens of Phrynohyas venulosa illustrating the lateral restriction of the dorsal blotch. both color patterns in Costa Rica suggest that there is a very close relationship between the Panamanian and Costa Rican Phrynohyas and those to the north. Analyses of several characters of specimens from the entire range of the genus in Mexico and Central America indicate that there are no consistent differences among the populations studied. The characteristics utilized by Duellman (1956) to distinguish between species of Phrynohyas have been shown to exhibit considerable variation within each population and extensive overlap among several populations. Based on the preceding discussion and analysis, I consider all known specimens of frogs of the genus Phrynohyas from Mexico and Central America as representative of a single, wide-ranging, variable species. All Middle American specimens presently referred to Phrynohyas latifasciata, P. infiata, P. spilomma, P. modesta, and P. zonata are regarded as representatives of P. venulosa. Some comments concerning the South American species seem appropri- ate. Duellman discussed the relationship of P. hebes and P. ingens. Phrynohyas hebes is very close to P. zonata and Duellman (1956:42) suggested that they may be subspecifically related. Phrynohyas ingens differs from P. venulosa primarily in coloration and size, and Rivero (1961:131) considered ingens to 14 Contributions in Science No. 134 be a subspecies of P. venulosa. Based on a knowledge of the coloration and size variation found in the Middle American Phrynohyas venulosa, I suspect that detailed analysis of the South American forms will reveal that only a single, wide-ranging, and highly variable species may be involved. Growth and Ontogenetic Change Zweifel (1964:201) described the eggs and larvae of P. venulosa from Panama. He found that the larvae metamorphosed in the laboratory in approx- imately 37 days, and suggested that under natural conditions a faster rate of development might be expected. Several newly transformed frogs were col- lected in Costa Rica from June 18 to September 18, 1964. Two of these individuals (CRE 8178), measuring 26 and 27 mm in body length, were taken on the road about 12 miles NW of Liberia on the night of July 26. Field records indicate that three adult females (CRE 8102, 8105, 8121) were taken in the same area on the nights of June 29 and June 30. The retention of a few scat- tered eggs in the reproductive tracts of these frogs indicates that they had deposited their eggs shortly before being collected, probably the previous night. Both juveniles had attained the characteristic coloration typical of the adult frogs. If we assume that the females had finished depositing eggs just prior to their collection and that the young frogs began development at about that time, then the period from egg to frog encompassed thirty or more days. Zweifel (1964:205) pointed out that the characteristic adult pattern is not reached until several days after metamorphosis appears complete. The evidence, while circumstantial, suggests that under natural conditions the rate of development probably is faster than in the laboratory. In either case, the inferences con- cerning the length of time from egg laying to metamorphosis generally support Zweifel’s findings. Other recently metamorphosed frogs from Parrita and Rincon de Osa, Costa Rica, are between 13 and 17 mm body length and do not exhibit the adult pattern. These specimens are assignable to Phrynohyas venulosa by their possession of green bones and the leg stripe characteristic of the recently transformed frogs (Zweifel, 1964:205). Duellman (1956:33) mentioned the presence of ontogenetic change in the tibia/ body length ratios of Phrynohyas from central Veracruz, Mexico. He stated that “Only in the small adults and juveniles does the tibia exceed 50 per- cent of the body length”, and he refers to a graph of body length plotted against tibia length. All that can be determined from these data (Duellman, 1956:33, fig. 9) is that specimens above 60 mm body length generally have a shorter tibia and show a lower percent tibia/ body length than do specimens below 60 mm. While there is no way to determine at what size (body length) he con- sidered specimens to represent small adults and juveniles, it can be seen from his graph that there are specimens over 60 mm body length which have a tibia/ body length ratio greater than 50 percent. 1968 Variation in Phrynohyas 15 The tibia/ body length ratios for 72 specimens from Guatemala show that ratios greater than 50 percent are found in six specimens ranging between 54.5 and 66.3 mm. The average body length for these six frogs is 60.7 mm, a value lower than the average for the total sample (Table 1 ) . There are seven indi- viduals with body lengths less than the average of 60.7 mm that have ratios lower than 50 percent. These seven specimens range between 58.7 and 60.5 mm (5r= 59.7 mm) and have tibia/body length percents ranging between 47.9 and 49.3 (x = 48.7). It can be seen that there are about as many specimens with a tibia/ body length ratio greater than 50 percent as there are with a tibia/ body length ratio less than 50 percent at the same or larger body length. For the total sample this means that, proportionally, some of the smaller frogs have a slightly longer tibia than the larger frogs. These findings generally are supported by the few smaller individuals from other populations. In these instances the higher tibia/ body length ratios are found in the smaller specimens. As Duell- man and, to a degree, my data indicate, there is ontogenetic change in tibia length so that the smaller the individual, the relatively longer the tibia. The data presented by Duellman in support of this change are for animals greater than 40 mm in body length. Apparently, he did not have specimens smaller than 40 mm from central Veracruz, Mexico. A series of 20 specimens (TCWC 16782-91, 16800-09) from Zacapa, Santa Rosa Department, Guatemala, has body lengths between 24.4 and 30.1 mm (x = 26.3 mm); the tibia length ranges between 11.5 and 14.1 mm (x = 12.6 mm) ; and the percents of tibia/ body length range between 44.6 and 51.5 (x = 47.9) . Unfortunately, no adults are available from this locality. A graph plotting body length against tibia length for these specimens (Fig. 6) exhibits the same distribution in reference to the 50 percent line as Duellman found for the adults from central Veracruz, Mexico. The majority of the specimens lie below the 50 percent line. These results indicate that a parabolic curve rather than a sigmoid curve of tibia/ body length ratios would be found in a given population. Two juvenile frogs (26 and 27 mm) and five adults (78.5 to 94.1 mm) from the same general locality in Costa Rica exhibit similar patterns. The smallest and the largest specimens have comparable ratios which are slightly lower than the intermediate sized specimens. Again the sample size, degree of differences among the ratios, and the lack of specimens from 30 to 70 mm body length do not allow for a meaningful analysis and only suggest the para- bolic oscillation in ratios for a given population. There is also an indication of ontogenetic change in the number of vomerine teeth. All juvenile frogs smaller than 30 mm body length that were examined averaged 4 to 5 fewer teeth than the adults from the same areas. Again the lack of sufficient material from the same population makes it difficult to evaluate these differences adequately. Color pattern is the only other character which shows a well defined ontogenetic change. Recently metamorphosed frogs maintain the hind leg 16 Contributions in Science No. 134 15 14 12 25 26 27 28 29 30 BODY LENGTH Figure 6. Graph plotting body length against tibia length for 20 small specimens of Phrynohyas venulosa from Zacapa, Guatemala. The 50 percent index is indicated by the dashed line. stripe of the tadpoles for a few days after transformation. Zweifel (1964:205) described this characteristic coloration, and my material supports his findings. By the time the frogs reach 24 mm body length, they have attained the adult color pattern. Discussion During the course of this study, several evolutionary trends among the different populations of Phrynohyas became obvious. I have already shown that color patterns, once thought to be consistent for each population, exhibit a wide range of variation from sample to sample or within a single population. The arrangement of the dorsal coloration in specimens from Mexico and Guatemala exhibits a clinal trend. All of the lower Middle American patterned specimens examined exhibit a single dorsal blotch. Most of the Guatemalan individuals possess a single dorsal blotch that extends from between the eyes to the vent. A single specimen from western Guatemala has the two blotched dorsal pattern characteristic of the western Mexican specimens. Individuals from Colima and Guerrero to the north exhibit a pattern of two blotches. In some specimens from the northern portion of the range of the genus along the 1968 Variation in Phrynohyas 17 west coast of Mexico, the dorsal pattern consists of three blotches. The trend is towards a fragmentation of the dorsal blotch from northern Central America northward along the Pacific coast of Mexico. In addition to the fragmentation of the dorsal pattern into two or three blotches, the anterior dorsal blotch tends to split along the longitudinal axis of the body. The split appears first in the Nayarit series and becomes prevalent in the southern Mexican and Guatemalan populations. There is no indication of a longitudinal split in dorsal patterns of specimens from Costa Rica and Panama. The dorsal blotch is constricted in Panamanian and Costa Rican specimens, but most marked in the former. The lateral constriction narrows the dorsal blotch and decreases the amount of dark dorsal coloration, but in a different manner from the longitudinal splitting of the dorsal blotch found in the Guatemalan specimens. The blotched dorsal coloration is considered the primitive condition. In the southern Mexican and Guatemalan populations, completely gradating series from blotched to unicolor forms are known. Apparently, the unicolor condition is a local variant, perhaps the result of a single gene mutation, which is distributed throughout the range of the species. The unicolor pattern may occur in all known specimens from a single locality, as is the case for the popu- lations from near La Lima, Honduras, and Los Chiles, Costa Rica, or may appear with typical blotched forms without obvious intermediates, as in the population from near Escuinapa, Sinaloa. Of primary importance to a discussion of the relationships among the major populations of Phrynohyas in Middle America is an understanding of the differences in sizes of the individuals from different populations. The data (Table 1) indicate that Phrynohyas from two areas are similar in size. The largest frogs are found in the terminal portions of the range of the species in Middle America, along the northwest coast of Mexico and in Costa Rica and Panama. The central portion of the range is occupied by smaller frogs. In the past these differences have been interpreted as favoring the recognition of three species. As previously pointed out, there is complete overlap in the measurements from population to population and only slight differences be- tween the ratios. Further, an examination of material from Central America clearly indicates a clinal gradation in body length for frogs from Guatemala (about 64.0 mm average size), Honduras (about 66.4 mm average size), Los Chiles, Costa Rica (about 70.5 mm average size) , other Costa Rican specimens (about 83.0 mm average size), and Panamanian specimens (about 87.0 mm average size). This gradation generally is reflected in all other measurements as well (Table 1 ) . If a clinal trend in size does exist in western and southern Mexico, it is not immediately obvious for two reasons. First, there is insufficient material from Nayarit southward to Chiapas for adequate analysis. Secondly, the data for the southern and eastern Mexican populations, presented by Duellman (1956) 18 Contributions in Science No. 134 and included in Table 1, are lumped for all populations. It is obvious from these data that there is variation in size, sometimes marked and sometimes clinal, from population to population throughout the range of Phrynohyas venulosa in Middle America. The larger specimens of Phrynohyas generally are found in areas of low mean annual rainfall, characterized by definite wet and dry seasons. The largest specimens were collected in western Mexico and along the Pacific coasts of Costa Rica and Panama. Rainfall data from the major localities reveal that the largest individuals of the genus come from areas which have between 200 and 400 mm of precipitation in September and that the smaller individuals come from areas which have between 300 and 500 mm of precipi- tation in the same month (Vivo Escoto, 1964:200, fig. 11). Duel 1 man (1956: 33) presented evidence which supports the proposed correlation between large size and drier environments. He recorded the largest specimens of his Phryno- hyas spilomma from San Luis Potosi and Yucatan (mean monthly precipita- tion for September, between 100 and 300 mm). Frogs of intermediate size were taken from La Libertad, Guatemala (300 to 400 mm precipitation) , while the smallest specimens are from Veracruz, Mexico (300 to 500 mm precipita- tion) . This evidence indicates that there is a correlation between the size of the Phrynohyas and the amount of annual precipitation as evaluated by the mean monthly precipitation in September, the peak of the rainy season. Rainfall data for March (Vivo Escoto, 1964:202, fig. 12), the driest month of the year, also correlate with the sizes of the individuals from different populations. The largest specimens come from areas characterized by less than 25 mm precipi- tation in March, while the smallest specimens come from areas where the mean March precipitation exceeds 50 mm. There are several selective forces that may operate to regulate an animal’s size relative to the amount and distribution of the precipitation in a given habitat. The primary factor favoring large size in a dry habitat is water economy. The primary site of water loss in terrestrial and arboreal frogs is the skin. It has been well established that water loss via the skin is reduced in environments of high humidity and increased in environments of low humidity (Prosser and Brown, 1962:32). Thorson (1955:100-116) demonstrated that, within a species, small individuals lose water at a more rapid rate than large individuals. Schmidt-Nielsen (1964:23-32) discussed the advantage of a large body size in arid environments as an adaptation to cope with the continual problems of overheating and desiccation. Bogert and Cowles (1947:33) found that reptiles which live in moist environments lose moisture at a more rapid rate (in terms of the percentage of their original body weight) than those that are found in dry environments. They also point out that there is a definite correlation between habitat selection and ability to resist desiccation. Based on this evidence, it seems that the correlation between size and habitat in Phryno- hyas represents an adaptive response to the environment. The occurrence of 1968 Variation in Phrynohyas 19 the largest specimens in the driest areas suggests that the osmo-regulatory powers of Phrynohyas are increased in drier environments by large size. There are other factors which should be considered in a discussion of the correlation between the frog’s size and habitat. It is generally true that species reach their highest population density in areas of most favorable habitat. As a corollary, it may be assumed that the species probably is less successful in an unfavorable as opposed to a favorable habitat. Success in a less favorable habitat may involve some biological adaptation. In the long term this might lead to a gradual morphological modification through natural selection toward a closer adaptation to the specific habitat. An immediate measure advantageous to the species in an unfavorable habitat would be an increase in size. Salt (1962: 912) has pointed out that large size is a factor in certain avian species to in- crease their chance of success in unfavorable habitats as a temporary adaptation pending the eventual evolution of specific adaptations to the new habitat. Salt has shown that under some circumstances larger birds are more efficient than smaller individuals. For example, large birds can support a larger amount of tissue without a proportional increase in food requirement. If a large animal can feed at less energy output per gram than a small animal, then the large animal is more efficient, and greater efficiency has a selective value. A selective advantage could give the larger animal more latitude in moving into new habitats than the smaller animal. On this basis it is anticipated that the well adapted smaller form will occur in the favorable habitat, that is, the area occupied for the longest time. Conversely, the largest individuals are to be expected in unfavorable habitats, peripheral to the ecologic and distributional center of the species. An examination of the distributions of the various Phrynohyas popula- tions supports this suggestion. The most favorable habitat, as determined by species density based on collecting records, is in southeastern Mexico and northern Central America and is occupied by small individuals. That all the basic color patterns are represented in the populations from this area adds support to my contention that this area represents the possible center of dis- persal of the species in Middle America. The Middle American center of dispersal for Phrynohyas is not necessarily the center of evolutionary origin for the species. I suggest that the general habitat now found in southeastern Mexico and northern Central America is similar to the habitat where Phrynohyas originated. In an historical sense, one would expect the best ecological fit be- tween a species and its habitat to be in the area where adaptive evolution is most complete. In peripheral populations the degree of adaptation to the habitat is lower because there has been less time for evolutionary change. An interesting sideline to this argument is that peripheral, drier environ- ments, for several reasons, may not be able to support the same number of individuals as more mesic areas. This could explain the apparent greater density of smaller frogs in the more favorable humid habitats. If the populations of 20 Contributions in Science No. 134 larger frogs in dry regions are composed of fewer individuals than are the populations of smaller frogs, as is indicated by locality records, then another factor might favor a large size frog in drier areas. A large male potentially can call louder and attract more females from a wider area than can a small male at the same site. If the population density of Phrynohyas in dry environments is lower than in wet environments, then it would be advantageous to breeding males to be larger and have a louder call. Small males would not face the same problems in the more mesic environments where the population density is higher. Bogert and Cowles (1947:33) found a rough correlation between the ability of representatives of a species to resist desiccation and the vagility of the species. The populations that exhibit wide latitudes in terms of physiological or ecological parameters usually have the greatest distribution and make the best colonizers. If it is assumed that the frog population that originally moved into peripheral habitats evolved a larger size to counteract suboptimal condi- tions in the new habitat, and that this larger size was also secondarily advan- tageous as a water conserving mechanism, then, after a given period of time, it might be expected that further adaptive modifications would appear. In the Phrynohyas from Costa Rica such an adaptation exists in the development of the dermal glands. In specimens collected in late June and July, during the rainy season, the dermal glands in the neck region show relatively little development; the glands of specimens collected during the dry season, in February and March, show extensive development (Fig. 7) . In many of the dry season specimens, glandu- lar development is so extensive that parts of the tympanum are concealed ( A) . The tympanum is never concealed in Costa Rican specimens taken during the rainy season (B). Duellman (1956:31, 38) mentioned that many individuals of his Phrynohyas spilomma and P. zonata possess heavy glandular folds which obscure part or all of the tympanum. Because of the effects that the apparent Figure 7. Lateral view of the heads of two Costa Rican Phrynohyas venulosa from Guanacaste Province showing the development of the dermal glands in the region of the tympanum. Specimen A was collected in February during the dry season; specimen B was collected in June during the rainy season. 1968 Variation in Phrynohyas 21 seasonal variation in glandular development has on tympanum size, I did not consider tympanum size in the populational analysis. Small frogs have propor- tionally larger tympanums than larger frogs, and this difference is ontogenetic. Duellman (1956:39) mentioned a small form of his Phrynohyas zonata with a larger tympanum than in typical P. zonata. The secretions of the dermal glands apparently are volatile and poisonous. The effectiveness of this secretion in deterring collectors (Smith, 1941:38; Duellman, 1956:41; Shannon and Humphrey, 1957: 18; Neill and Allen, 1959: 26; Janzen, 1962:651) is well attested. It probably is very effective on natural predators, as well. Although differential predation on Phrynohyas in the dry season may account for the seasonal variation in glandular development, an- other explanation seems more plausible. Slime secretion may be a factor decreasing skin permeability to water (Prosser and Brown, 1962:32). The use of slime secretion as a mechanism to prevent desiccation is well documented in lungfish (Herald, 1961:290; Smith, 1961:77-78). Vellard (1948:143) regarded the thick skin of the frog Lepto- dactylus bufonius, which is very rich in cutaneous glands, as protection against desiccation. McClanahan (1967:88) mentioned the appearance of dark, keratinized skin on Scaphiopus while the frog is hibernating, a change in integument effective in reducing water loss through the skin. Robert Stebbins (pers. comm.) reported the appearance of a cellophanelike membrane in Pyxicephalus adspersus, an African frog which is especially successful in arid environments. Stebbins wrote that whenever he wished to show the formation of this membrane, in which the frog completely encases itself, he merely placed the animal in a dry container for a few days. Neill and Allen (1959a: 25) proposed that the slime of Phrynohyas might serve to prevent desiccation in addition to lessening predation. These authors point out that viscous organic liquids are comparatively resistant to evaporation. I have found that this secretion is water insoluble and is difficult to remove when dried. McConkey (in Duellman, 1956:41) mentioned that a collecting sac used to carry the frogs became stiff as a board a day or so after the secretion had dried. Vellard (1948:150) reports that in northern Argentina, these frogs use this cutaneous secretion to line the cavities of trees in which they seek refuge. If this is a response to aridity, then this peculiar behavior suggests that Phrynohyas has developed a modification to decrease water loss, which is similar to that mechanism utilized by lungfish. All Phrynohyas venulosa examined have these glands, whether the frogs are from areas with a definite wet-dry season or from areas where there is some precipitation throughout the year. Frogs from the dry forests of Costa Rica, where there is a marked wet-dry season, apparently exhibit a change in the glandular development from season to season. It is suggested that the greater development and subsequent secretion of the glands in the dry season is pro- duced as an adaptive response to arid environments. Additional investigation 22 Contributions in Science No. 134 with other populations of Phrynohyas is needed to corroborate the evidence. Similar adaptations may occur in other genera of frogs inhabiting arid environments. Acknowledgments I would like to express my appreciation to Jay M. Savage for his help and encouragement throughout the preparation of this report. Several other associ- ates, especially Norman J. Scott, have offered suggestions or aided in the completion of this project in other ways. To them I extend my thanks. I am indebted to the following people and institutions for the loan of comparative material or for permission to examine specimens in their care: Richard J. Baldauf, Texas Cooperative Wildlife Collection (TCWC) ; James L. Christian- sen, University of New Mexico ( JLC) ; James R. Dixon, Los Angeles County Museum of Natural History (LACM); William E. Duellman, Museum of Natural History, University of Kansas (KU); Robert F. Inger and Hymen Marx, Field Museum of Natural History, Chicago (FMNH); Richard B. Loomis, California State College, Long Beach (CSCLB); Charles F. Walker, Museum of Zoology, University of Michigan (UMMZ); and Richard G. Zweifel, American Museum of Natural History (AMNH). Thanks are also due my wife, Mercedes, for checking the text and typing the manuscript, and to Robert J. Lavenberg and Richard A. Anderson for helping with the photo- graphs. Part of the field work was done in 1962 with financial assistance from a Grant-in-Aid of Research from the Society of Sigma Xi. Specimens were collected in 1964 during the tenure of a National Science Foundation Summer Fellowship and in 1965 while participating in the program sponsored by the Organization for Tropical Studies. Specimens Examined Mexico Sinaloa: 9.4 mi S. Escuinapa (LACM 6314-19, 7245). Nayarit: 46.5 mi S. Escuinapa (KU 73879) ; 22.9 mi E. San Bias (KU 74339, CSCLB 625-31, 633-41). Colima: 1 mi N. Colima (UMMZ 80018); Paso del Rio (UMMZ 108019). Michoacan: Barranca de Bejuco (UMMZ 104814). Guerrero: nearLaVenta (FMNH 10046, 10835, 10836). Oaxaca: Temascal (LACM 28215-16, 36220-21). Veracruz: 44 km S. Tampico (JLC 903-04) ; Salinas (TCWC 19103) . Guatemala Escuintla: Cuyuta (AMNH 74377-90 +58). Santa Rosa: 45 km S. Guatemala City (LACM 8442); 23 km W. Zacapa (TCWC 16782-91, 16800-09). 1968 Variation in Phrynohyas 23 Honduras Cortes: 1 mi W. La Lima (TCWC 19180-84). Costa Rica (Specimens designated CRE are in the University of Southern California collections.) Alajuela: Los Chiles (CRE 7215 +1,7219 +1). Guanacaste: 20.6 mi N. Liberia (CRE 8157) ; 12.7 mi N. Liberia (CRE 8105, 8178 +1); 12.5 mi N. Liberia (CRE 8102); 12.4 mi N. Liberia (CRE 8121); 10.7 mi N. Liberia (CRE 8141); Rio Higueron, 0.5 mi E. Finca Jimenez (CRE 842); Finca Jimenez (CRE 892 +3). Puntarenas: Parrita (CRE 6158 +2; 6159-60); Parrita, La Julieta (CRE 8253-54,8258,8264); 1% mi S.W. Rincon de Osa (CRE 7237 + 1);5± mi W. Rincon de Osa (CRE 891 ) . Panama Panama: 3 mi S. Bejuco (AMNH 69803-04); Nueva Gorgona (AMNH 69805-06) ; 1 mi W. Nueva Gorgona (AMNH 69807) . Literature Cited Bogert, C. M., and R. B. Cowles 1947. Results of the Archbold Expeditions, No. 58. Moisture loss in relation to habitat selection in some Floridian reptiles. Am. Mus. Novitates, 1358:1-34. Boulenger, G. A. 1882. Description of a new genus and species of frogs of the family Hylidae. Ann. Mag. Nat. Hist., ser. 5, 10:326-328. Duellman, W. E. 1956. The frogs of the hylid genus Phrynohyas Fitzinger, 1843. Misc. Publ. Mus. Zool. Univ. Michigan, 96:1-47. 1960. A distributional study of the amphibians of the Isthmus of Tehuantepec, Mexico. Univ. Kansas Publ. Mus. Nat. Hist., 13(2): 19-72. 1961. The amphibians and reptiles of Michoacan, Mexico. Univ. Kansas Publ. Mus. Nat. Hist., 15(1): 1-148. 1966. Taxonomic notes on some Mexican and Central American hylid frogs. Univ. Kansas Publ. Mus. Nat. Hist., 17(6) :263-279. Gans, C. 1960. Notes on a herpetological collecting trip through the southeastern lowlands of Bolivia. Ann. Carnegie Mus., Pittsburgh, 35:283-314. Hemming, F. (ed.) 1958. Opinion 520. Suppression under the plenary powers of the specific name tibiatrix Laurenti, 1768, as published in the combination Hyla tibia- trix, and of the generic name Acrodytes Fitzinger, 1843, and interpre- tation under the same powers of the nominal species Rana venulosa Laurenti, 1768 (Class Amphibia). Opin. Declar. Inti. Comm. Zool. Nomen., 19:169-200. Herald, E. S. 1961. Living fishes of the world. Doubleday and Co., New York, 304 p. 24 Contributions in Science No. 134 Janzen, D. H. 1962. Injury caused by toxic secretions of Phrynohyas spilomma Cope. Copeia, 1962(3): 651. McClanahan, L., Jr. 1967. Adaptations of the spadefoot toad, Scaphiopus couchi, to desert envi- ronments. Comp. Biochem. Physiol., 20:73-99. Meyer, J. R. 1966. Records and observations on some amphibians and reptiles from Hon- duras. Herpetologica, 22:172-181. Neill, W. T. 1965. New and noteworthy amphibians and reptiles from British Honduras. Bull. Florida State Mus., 9:77-130. Neill, W. T. and R. Allen 1959a. Studies on the amphibians and reptiles of British Honduras. Publ. Res. Div., Ross Allen’s Reptile Inst., 2:1-76. 1959b. Additions to the British Honduras Herpetofaunal list. Herpetologica, 15:235-240. Porter, K. R. 1962. Mating calls and noteworthy collections of some Mexican amphibians. Herpetologica, 18:165-171. Prosser, C. L., and F. A. Brown, Jr. 1962. Comparative animal physiology. 2d ed. W. B. Saunders Co., Philadel- phia, 688 p. Rivero, J. A. 1961. Salientia of Venezuela. Bull. Mus. Comp. Zool., Harvard Coll., 126:1- 207. Salt, G. W. 1962. Avian body weight, adaptation, and evolution in western North Amer- ica. Proc. XIII Intern. Ornith. Congr., 2:905-917. Savage, J. M. 1966. The origins and history of the Central American herpetofauna. Copeia, 1966:719-766. Schmidt-Nielsen, K. 1964. Desert animals. Oxford Univ. Press, London, 277 p. Shannon, F. A., and F. L. Humphrey 1957. A new species of Phrynohyas from Nayarit. Herpetologica, 13:15-18. Smith, H. M. 1941. Snakes, frogs, and bromelias. Chicago Nat, 4:35-43. Smith, H. M., and E. H. Taylor 1948. An annotated checklist and key to the Amphibia of Mexico. Bull. United States Natl. Mus., 194:1-118. Smith, H. W. 1961. From fish to philosopher. Doubleday and Co., Garden City, N. Y., 293 p. Stuart, L. C. 1963. A checklist of the herpetofauna of Guatemala. Misc. Publ. Mus. Zool. Univ. Michigan, 122:1-150. 1968 Variation in Phrynohyas 25 Taylor, E. H. 1952. A review of the frogs and toads of Costa Rica. Univ. Kansas Sci. Bull., 35:577-942. Thorson, T. B. 1955. The relationship of water economy to terrestrialism in amphibians. Ecology, 36:100-116. Vellard, J. 1948. Batracios del Chaco Argentino. Acta Zoologica Lilloana, 5:137-174. Vivo Escoto, J. A. 1964. Weather and climate of Mexico and Central America. In Wauchope, R. (ed.), Handbook of Middle American Indians. Univ. of Texas Press, Austin, Vol. 1, Natural environment and early cultures, Chapt. 6:187- 215. Zweifel, R. G. 1964. Life history of Phrynohyas venulosa (Salientia: Hylidae) in Panama. Copeia, 1964:201-208. |7/TS LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE ' UMBER 135 February 20, 1968 NEW GENERA AND SPECIES OF WASPS OF THE TRIBE TRYPOXYLONINI FROM THE NEOTROPICAL REGION (HYMENOPTERA, SPHECIDAE, LARRINAE) By A. S. Menke p\THSO%^ MAR 1 4- 1968 Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 814 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor NEW GENERA AND SPECIES OF WASPS OF THE TRIBE TRYPOXYLON INI FROM THE NEOTROPICAL REGION (HYMENOPTERA, SPHECIDAE, LARRINAE)1 By A. S. Menke2 Abstract: Pisoxylon xanthosoma, new genus and new species, is described from Peru. Four subgenera are recognized in Pison : Pison, Pisonoides, Krombeiniellum, and the new subgenus Entomopison (type: Pison pilosus Smith, 1873). Pison (Pison) eremnon is described from Brazil. Pison areolatus Spinola, 1851, is transferred to the genus Pisonopsis. A lectotype is designated for Pison variicornis Reed, 1894, which is synonymized under areolatus Spinola. A key is provided to the genera of the Trypoxy- lonini and the subgenera of Pison. A checklist of the New World Pison is appended. The following new generic and specific entities have been discovered while studying the tribe Trypoxylonini in connection with a world revision of the genera of the Sphecidae. This revision is being conducted by R. M. Bohart and the author. In addition to the new taxa described here, I have provided a key to the genera of the Trypoxylonini. Notes on the subgenera of Pison, a check- list of the New World Pison, and some new synonymy in Pisonopsis are also included. The material on which this paper is based comes from several sources: the Los Angeles County Museum of Natural History; Department of Ento- mology, University of California, Davis; Carnegie Museum, Pittsburgh; and the U. S. National Museum, Washington, D. C. I would like to thank the curators of these institutions for the loan of this material. Deposition of types will be cited in the descriptions. Dr. I. H. H. Yarrow of the British Museum (Natural History) kindly sent me notes on the type of Pison pilosum Smith and related species. Pisoxylon new genus3 Diagnosis : Inner orbits deeply notched; antennal sockets not contiguous with frontoclypeal suture; flagellum with eleven articles; frontal carina nar- rowly forking above (Fig. 2); clypeus high, trapezoidal; mandible simple; labrum hidden, with two narrow apical fingerlike processes; mouthparts short, palpi 6-4; occipital carina meeting hypostomal carina near the latter’s apex; ’Research supported by the National Science Foundation, grant numbers GB-5839 and GB-3074. 2Research Associate, Los Angeles County Museum of Natural History; Department of Entomology, University of California, Davis. 3Based on male, female unknown. 1 2 Contributions in Science No. 135 pronotal collar narrow; form of thorax as in Trypoxylon, landmark sulci of mesopleuron consisting of episternal sulcus curving forward ventrally but disappearing before meeting anterior margin, and a sulcus running from subalar pit to depression behind pronotal lobe; intercoxal carina straight (carina between mid - and hindcoxae) ; propodeum without carinae, propodeal enclosure not defined; area surrounding dorsal margin of propodeal orifice simple, not lamellate nor rimlike; no propodeal sternite; gaster sessile, compact (Fig. 5); wings as in Trypoxylon : one submarginal cell, marginal cell acumi- nate and with anterior veinlet of cell (Ri) extending a short distance beyond apex of cell, hamuli of hindwing divided into two groups, outermost group short and separated from basal group by much more than length of outer group; legs simple, mid - and hindcoxae contiguous, pulvilli moderately large, equal on all legs; gonostyle of male genitalia simple (Fig. 6) . Type of genus: Pisoxylon xanthosoma Menke. Distribution : Known only from Peru in South America. Discussion: This genus is very similar to Trypoxylon, but differs in having a nonpetiolate compact gaster (Fig. 5). Trypoxylon is a huge genus (over 350 species), and accordingly contains many divergent groups. Therefore it could be argued that Pisoxylon simply represents an extreme group within Trypoxy- lon and as such should be considered as a subgenus. However, despite the morphological complexity of Trypoxylon, it is still an easily recognized taxon because of the long clavate (and often petiolate) gaster. To place Pisoxylon in Trypoxylon would, in my opinion, tend to weaken a popular and morphologi- cally sound generic concept. Pisoxylon xanthosoma new species Holotype male: Length 9 mm. Color: Straw yellow; frons, vertex and back of head black; antenna yellow basally but grading to black at tip; scutum black but with a large rectangular central yellow spot, and humeral area yellow; posterior half of scutellum black; metanotum with a small lateral black spot; dorsum of propodeum and median sulcus of posterior face black; subalar pit of mesopleuron black; metapleuron narrowly black beneath metapleural flange; gastral tergite I with a transverse subapical brownish band, remaining tergites slightly suffused with brown; tarsomere V of mid- and hindlegs and all pulvilli black; wings with a faint yellow tint, veins reddish brown. Vestiture: Appressed hair of clypeus, lower frons and gena golden; rest of body with short, sparse, pale hair. Structure: Flagellomeres IX and X shorter than preceding articles, flagel- lomere XI slightly longer than combined length of articles VIII-X, flagello- meres VI-VIII bearing narrow tyloides ventrally, flagellomere VIII swollen apicoventrally (Fig. 2); frontal carina Y-shaped, the stem bearing two short lateral arms just above antennal sockets (Fig. 2); frons minutely granulate 1968 Neotropical Wasps 3 with moderate shallow punctation, dull; least interocular distance at vertex slightly greater than at clypeus (24.5:21.0); ratio of ocellocular distance to diameter of lateral ocellus to distance between lateral ocelli: 2.0: 8.0: 5.0; clypeal outline as in Figure 2; surface of scutum and scutellum less shiny than propodeum and pleura, punctation of thorax fine and sparse; propodeal dor- sum finely diagonally ridged basally, ridges merging laterally with evanescent striatopunctation which rapidly becomes simple punctation; median sulcus of propodeal dorsum broad, and with fine transverse arcuate ridges; postero- lateral corner of propleuron with a roughly egg-shaped platelike area covered with short hair and delimited inwardly by a carina; sternites VII-VIII as in Figures 3 and 4, respectively; genitalia as in Figure 6. Holotype male : Pucallpa, Loreto, Peru, 200 m, April 10-19, 1965, J. Schunke. Type deposited in the Los Angeles County Museum of Natural History. Discussion : The extensively yellow body of this wasp is distinctive, al- though a few Neotropical Try poxy Ion species are similarly colored. The features of the head, especially the antenna, should separate xanthosoma from other species of Pisoxylon that may be discovered. The two narrow tufts of long setae at the apex of the last sternite are distinctive (Fig. 4). These setae project from the apex of the abdomen, giving the appearance of a sting. The subgenera of Ptson Three subgenera are currently recognized in Pison : Pison, Pisonoides and Krombeiniellum. Krombeiniellum* is a valid subgenus in my opinion. It differs from typical Pison in having short dense hair on the eyes. Most Krombeiniel- lum have only two submarginal cells in the forewing, but there are several undescribed species from the Neotropical Region with three submarginals. The subgenus includes browni Ashmead, differens Turner, and koreense Radoszkowski. These three are Oriental species, but koreense has been success- fully introduced to the United States (Krombein, 1958). The taxon Pisonoides includes all other Pison species that have only two submarginals. However, like Turner (1916) and Leclercq (1965), I do not believe that Pisonoides is a valid subgenus if based solely on wing venation. The size of the second submarginal cell of Pison species having three submarginals varies from large to very small. In species in which the cell is pinhole size, some specimens exist in which the cell is entirely obliterated ( Pison xanthopum Brulle, inaequale Turner and other African species, for example). The type species of Pisonoides, Pison obliteratum Smith, differs from most other Pison species in having a semipetiolate abdomen. This condition is found in at least two other two-celled Pison species: icarioides Turner and difficile Turner. It * Richards ( 1962, p. 118) proposed Krombeiniellum for the preoccupied name Paraceramius Radoszkowski. 4 Contributions in Science No. 135 Figures 1-6. 1. Head of Pison eremnon, holotype. 2-6. Pisoxylon xanthosoma, holo- type: 2. head, with three views of terminal flagellomeres, 3. sternite VII, 4. sternite VIII, 5. lateral view of gaster, 6. ventral view of genitalia (aedeagus is stippled). 1968 Neotropical Wasps 5 would appear that if Pisonoides is to be recognized as a subgenus, it must be redefined on the basis of the semipetiolate abdomen and presence of two submarginal cells. All two-celled Pis on with a nonsubpet iolate abdomen thus revert back to the typical subgenus Pison. A fourth group of Pison species has been found which is sufficiently dis- tinct to warrant subgeneric status. It is characterized as follows: Pison (Entomopison) new subgenus Mandible with a deep notch (or angle) on externoventral margin; fore- wing with three submarginal cells, the second petiolate; recurrent veins of forewing received by second submarginal, or first recurrent vein interstitial or received by first submarginal. Type of subgenus : Pison pilosum Smith, 1873, present designation. Except for the notched mandible, Entomopison is a typical Pison in every respect. Notched mandibles also occur in the trypoxylonine genus Pisonopsis, but this taxon differs from Pison in having oblique grooves on sternites II I- IV, and a shorter marginal cell, the apex of which is rounded or truncate. Further- more, in some species of Pisonopsis the female has a pygidium. Pisonopsis has a narrower (more transverse) clypeus than most Pison species. The clypeal outline of Pison usually approximates a high trapezoid. Entomopison is a Neotropical group and from the material on hand, it is clear that there are at least ten species in South America, all of which are undescribed save for pilosum, aurofaciale, and convexifrons. Pison (Pison) eremnon new species Holotype female : length 14 mm; a large black, rather coarsely and irregularly punctate wasp. Color : Black; inner face of hindtibia rust colored; anterior margin of forewing strongly infumate, hindwing faintly infumate anteriorly. Vestiture : Appressed hair of face sparse, brown; head and thorax with long erect brown hair, propodeum posferolateraily with some shorter pale erect hair in addition to the brown hair; fine appressed hair of gastral tergites brown, tergites without transverse apical bands of hair. Structure : Inner orbits converging above, ratio of interocular distance at clypeus and vertex: 49:34; lateral ocelli much closer to each other than to eyes, ratio of ocellocular distance to diameter of lateral ocellus to inter- ocellar distance: 10:5.5:3; frons with a median longitudinal carina the length of which is equal to about two ocellus diameters; frons and clypeal disk closely to confluently punctured, punctures of two sizes, the larger ones scattered, interspaces weakly shagreened, surface subshining; clypeus with an obtusely trunctate median lobe which is broadly impunctate and shining; mandible with a small tooth on inner margin (Fig. 1) ; antenna as in Figure 1; collar weakly tumid mesally; scutum shagreened, dull, scutal punctures of two sizes, closely 6 Contributions in Science No. 135 punctate anteriorly, punctation becoming irregularly less dense posteriorly, where punctures are separated by one to one and a half puncture diameters; scutellar punctation sparser than that of scutum; metanotum with very fine, dense, pinhole punctures and a few scattered large punctures; propodeal dorsum with a median longitudinal ridge which is contained in a narrow sulcus, sulcus obscured by transverse striatopunctation basally, remainder of dorsum irregularly densely punctate, the punctures of uniform size, interspaces shining; punctures of propodeal side and posterior face of two sizes, smaller punctures dense, becoming weakly striatopunctate posterolaterally, larger punctures (about size of those on dorsum) sparse; upper two-thirds of vertical posterior face of propodeum with a deep sulcus; propodeum without lateral carinae; mesopleural punctation same as that of anterior one-half of scutum but inter- spaces shining; sulcus dividing metapleuron from propodeal side foveolate between upper and lower pits; metapleural flange narrowly lamellate; gastral segment I set off from remaining segments by a constriction, tergite I sharply transversely depressed subapically forming a bandlike margin; punctation of tergite I similar to that of posterior one-half of scutum, interspaces shining, punctation of II denser except for some polished subapical prominences, punctation of III- VI very dense and punctures of uniform size; sternites strongly shining, sparsely and finely punctate although with a few scattered larger punctures; marginal cell of forewing rounded apically, the apex just barely surpassing outer veinlet of third submarginal cell; first recurrent vein received by second submarginal cell, second recurrent interstitial. Holotype female : Santarem, Brazil, H. H. Smith, deposited in the Carnegie Museum, Pittsburgh. The large size, black body with dark hair, and rounded marginal cell quickly distinguish this wasp from all other known Neotropical Pison. The size of eremnon is exceeded only by the Old World species regale Smith and its relatives. The rounded marginal cell is a rarity in Pison, occurring only in a few Old World species. Checklist of New World Pison4 subgenus Pison cameronii Kohl, 1893. Mexico or Peru. fasciatum Kohl, 1883. (preocc.) chilense Spinola, 1851. Chile. conforme Smith, 1869. Mexico. cressoni Rohwer, 1911. Nicaragua. eremnon Menke, 1967. Brazil. 4Pison argentinus Schrottky, 1909; P. flavopictus Smith, 1860; P. laetus Smith, 1860; and paraensis Spinola, 1853 are no longer assignable to Pison. See Menke, 1968, for details. 1968 Neotropical Wasps 7 flavolimbatum Turner, 1917. British Guiana. llaeve Smith, 1856. “Georgia.” Verification that this is a North American or a New World species is needed. maculipenne Smith, 1860. Brazil. subgenus Krombesnidlum koreense Radoszkowski, 1887. Eastern United States, eastern Asia. subgenus Enfomopison aureofaciale Strand, 1910. Paraguay. convexifrons Taschenberg, 1870. Brazil. pilosum Smith, 1873. Brazil. New synonymy in Pisonopsis Pisonopsis areoSesfus (Spinola) Pison areolatus Spinola, 1851. in Gay, Historia Fisica y Politica de Chile, Zook, 6:327. Holotype $ , Chile (type probably in Turin or Paris). Pison variicornis Reed, 1894. Anal. Univ. Chile, 85:634. Lectotype $ , Valparaiso, Chile (Museum of Comparative Zoology, Cambridge), present designation. New synonymy. ? Pisonopsis anomala Mantero, 1901. Bull. Soc. Ent. Ital., 33:202. Holo- $ type , Rio Santa Cruz, Argentina (Museo Civico de Storia Naturale, Genoa) . Tentative new synonymy. This species has been assigned to Pison until now. I have studied material of a small black Chilean Pisonopsis which agrees perfectly with Spinola’s description, and there is little doubt that the type of areolatus is this species. I have seen Reed’s syntypes of variicornis and they agree with Spinola’s descrip- tion of areolatus. Mantero’s description of anomalus seems to fit areolatus, but I have seen no Argentine material. Pisonopsis areolatus apparently is the only representative of the genus in Chile. Besides areolatus there is one other South American Pisonopsis : australis Fritz from Argentina. Pisonopsis argentina Schrottky is probably a species of a new genus belonging in the tribe Bothy- nostethini (see Menke, 1968). Key to genera of the TRYPOXYLONINI 1. Forewing with one submarginal cell; antennal sockets not contiguous with frontoclypeal suture 2 Forewing with two or three submarginal cells; antennal sockets con- tiguous with frontoclypeal suture 3 2. Gaster long, clavate, often petiolate, segment I usually slender, clublike, at least two times as long as wide; Old and New World Trypoxylon Latreille 8 Contributions in Science No. 135 Gaster compact, sessile, segment I not elongate (Fig. 5) ; South America Pisoxylon Menke 3. Gaster compact, sessile, segment I not petiolate or at most subpetiolate in dorsal view; mesopleuron without coarse horizontal ridges; Old and New World 4 Gaster petiolate, segment I rodlike (tergite nodose at apex) and nearly as long as remaining segments combined; mesopleuron with many coarse horizontal ridges; Neotropical Region Aulacophilus Smith 4. Marginal cell of forewing rounded or truncate apically, the apex not or only slightly extending beyond outer veinlet of third submarginal cell, and externoventral margin of mandible notched or strongly angulate, and sternites III-IV with a lateral oblique groove; female gastral tergite VI usually flattened or with a distinct pygidium bounded by carinae; North and South America Pisonopsis Fox Marginal cell of forewing acute apically, the apex extending well beyond outer veinlet of third submarginal cell, or if apex rounded and/or not extending much beyond third submarginal (exceptional Old World species), then outer margin of mandible not notched; outer margin of mandible entire (except in some South American forms but wing characteristics typical) ; gastral sternites without oblique grooves; female gastral tergite VI conical, sometimes weakly keeled along midline; cosmopolitan Pison Jurine 5 5. Outer margin of mandible entire 6 Outer margin of mandible notched or strongly angulate; South America subgenus Entomopison Menke 6. Eyes bare 7 Eyes densely covered with short hair; Oriental Region, North and South America subgenus Krombeiniellum Richards 7. Gaster sessile, not subpetiolate, apical width of segment I much more than one-half apical width of II and usually subequal to II (dorsal view) ; forewing with two or three submarginal cells; Old and New World subgenus Pison Jurine Gaster subpetiolate, apical width of segment I about equal to one-half apical width of segment II; forewing with two submarginl cells; Australasian Region subgenus Pisonoides Smith 1968 Neotropical Wasps 9 Literature Cited Krombein, K. V. 1958. Pison ( Paraceramius ) koreense (Rad.), A new adventive wasp in the Eastern United States. Entomol. News, 69:166-67. Leclercq, J. 1965. Sphecidae, subfamily Trypoxyloninae. Parc. Nat. Garamba, Mission H. de Saeger, fasc. 46(5) : 67- 153. Menke, A. S. 1968. New South American genera and species of the tribe Bothynostethini. Acta Zool. Lilloana (in press). Richards, O. W. 1962. A revisional study of the masarid wasps. London, William Clowes and Sons, 294 p. Turner, R. E. 1916. Notes on the wasps of the genus Pison, and some allied genera. Proc. Zool Soc. London, 1916:591-629. LOS L%y% COUNTY MUSEUM IN SCIENCE fUMBER 136 February 20, 1968 ANGELES CONTRIBUTIONS A NEW PORCUPINE FROM THE MIDDLE PLEISTOCENE OF THE ANZA-BORREGO DESERT OF CALIFORNIA With notes on mastication in Coendou and Erethizon By John A. White MAR HI Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 814 x 1 1 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor A NEW PORCUPINE FROM THE MIDDLE PLEISTOCENE OF THE ANZA-BORREGO DESERT OF CALIFORNIA With notes on mastication in Coendou and Erethizon By John A. White1 Abstract: A new species of Coendou is described from the Middle Pleistocene of the Anza-Borrego Desert. The genera Coendou and Erethizon are distinguished from one another by means of inferred differences in mastication, and it is postulated that Coendou is ancestral to Erethizon. Introduction A large and diversified collection of fossil vertebrates has been obtained from the highly fossiliferous badlands of the Vallecito Creek Valley and Fish Creek Wash areas of the Anza-Borrego Desert in the western Imperial Valley of Southern California. This collection has resulted from the extensive field operations in that area conducted by the Vertebrate Paleontology Section of the Los Angeles County Museum of Natural History under the direction of Theodore Downs. The sediments comprise the Palm Springs formation (Woodring 1931; Dibblee 1954) and consist of mudstones, siltstones, and sandstones which are occasionally cross-bedded. That part of the formation from which the above collection was made is exposed on the south limb of a west-plunging anticline (Woodard 1963), and dips 24 to 26 degrees to the south. It is in continuous sequence through more than 10,000 feet. The porcupine specimens described herein have been collected from within the upper 3,000 feet of this section, which is Irvingtonian (Middle Pleistocene) in age (Downs 1957; White and Downs 1961; Howard 1963; White 1964). Acknowledgements The author thanks the following individuals for permission to examine specimens in their care: Seth Benson, Museum of Vertebrate Zoology, Uni- versity of California, Berkeley (MVZ); W. A. Clemens and Theodore Eaton (Vertebrate Paleontology), and E. Raymond Hall and J. Knox Jones, Jr. (Mammalogy), Museum of Natural History, University of Kansas (KUM); C. L. Gazin and Clayton E. Ray (Vertebrate Paleontology), and David H. Johnson (Mammalogy), United States National Museum (USNM); Claude W. Hibbard, Museum of Paleontology, University of Michigan (UMM); J. R. Macdonald (Vertebrate Paleontology), and C. A. McLaughlin (Mam- malogy), Los Angeles County Museum of Natural History (LACM); and 1Curator in Vertebrate Paleontology, The Museum, Idaho State University, and Re- search Associate, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 136 Donald E. Savage, Museum of Paleontology, University of California, Berkeley. For valuable suggestions the writer is indebted to the late R. A. Stirton, Claude W. Hibbard, and Theodore Downs. Theodore Downs and David E. Fortsch critically read the manuscript. John E. Mawby collected the type specimen. Kent Wilkinson made the stereo-photograph, and Lisa A. Hansen made the other illustrations. The cooperation of the personnel and management of the Anza-Borrego Desert State Park is gratefully acknowledged. The National Science Foundation supported this research through grants GB-1333 and GB-5116. Specimens Examined An estimated 400 specimens of Erethizon and 25 Coendou were exam- ined. Only the specimens that were measured are listed here with their museum numbers. Coendou mexicanus : MVZ 116815, 116817, 124088; USNM (Mam- malogy) 296308, 298915, 303135, 303277, 306980, 324110, 335666; skele- tons: 240269, 240312, 257008, 257364; KUM (Mammalogy) 18190, 18191, 19888, 19889, 24489, 24490, 24491, 32180, 93790, 93791, 93793, 93794; UMM Coendou sp. V47106; ICoendou brachignathum USNM (Vertebrate Paleontology) 13684. Erethizon dorsatum : MVZ 4052, 4508, 4510, 4467, 4496, 51382, 40874, 42073, 42077, 42078, 42079, 46906, 53702, 53263, 58496, 67689, 68707, 68708, 74292, 79588, 84036, 84969, 84970, 87862, 89556, 106164; USNM (Mammalogy): skeletons: 49420, 90476, 192546, 241005; (Vertebrate Pal- eontology) : 7668-7673, 7996, 8128, 8130, 8134, 8174. A number of specimens of porcupines referable to the genus Coendou, to which belong the prehensile-tailed porcupines of the rain forests of Central and South America, have been collected in the above-mentioned geological section, and differ morphologically from known species of the genus to such a degree that they are assigned to a new species. Coendou stirtoni new species Type : LACM 17633, fragmentary palate with LP4- M1, RP4 (partial), M1 and M2 (Fig. 1). Type Locality : LACM 1428, Arroyo Tapiado, Badlands in Anza-Borrego Desert State Park, San Diego County, California. Horizon : Approximately 2900 feet stratigraphically below top of the Palm Spring formation in the Tapiado member of the formation (Woodard 1963). 1968 Anza-Borrego Fossil Porcupine 3 Figure 1. Coendou stirtoni n. sp., LACM 17633, type specimen, stereophotograph of the palate (A); ventral view of palate (B). 4 Contributions in Science No. 136 Referred Specimens : LACM 6136, a partial skeleton with palate and medial part of RM2, ventral portion of rostrum, fragmentary left mandible with broken teeth, sacrum and 1 2 caudal vertebrae, proximal end of left radius, broken proximal right ulna, partial ischium with part of acetabulum, proximal end of left femur; 6210 and 62 10 A, two fragmentary left mandibles with broken teeth; 4325, a partial skeleton lacking skull parts, but with distal part of right humerus, right radius and ulna with distal epiphyses missing, proximal portion of left ulna, right femur with distal epiphysis missing, left astragalus, right calcaneus, left cuboideum, and seven phalangeal elements. Diagnosis ; Erethizontid with width of palate between right and left P4’s at least 40 per cent of width of palate between M3’s in adults; width of P4 approximately the same as width of M1; adults as large as a large adult Erethizon, and far larger than any species of living Coendou. Description : The occlusal pattern of the cheek teeth in Coendou stirtoni is essentially the same as in other species of Coendou and in Erethizon. The large size of the teeth in the new species tends to cause the occlusal patterns of the cheek teeth to resemble those in Erethizon more than those in Coendou ; however the living species of the latter genus have teeth that are markedly smaller in size than in Erethizon and in the larger, extinct species of Coendou. This marked difference in size may account for the dissimilarity of tooth pat- tern. P4 is as wide as M1 in the new species and in all specimens of living Coendou examined, while in all specimens of Erethizon P4 is markedly larger than M1. In juvenile specimens of Erethizon, DM4 is as wide as M1, and DM4 and P4 resemble one another morphologically in Coendou and Erethizon. In the type specimen of C. stirtoni, the broken roots of DM4 are visible on the sides of the alveolus of P4. Although mandibular cheek teeth are unknown for C. stirtoni, P4 in Erethizon is, on the whole, larger than M1 while in Coendou the two teeth tend to be subequal in size. The relationship between P4 and M1 is more con- sistent than that between P4 and M4 (compare Figs. 4 and 5). The latter exhibits considerable variation and is of limited taxonomic use (Hibbard and Mooser 1963). Only in rare cases when P4 is markedly larger than can Erethizon be identified by this character. In mandibles LACM 6210A and 6136, the base of the ascending ramus indicates that it slanted posteriad, permitting the cheek teeth to be visible from the side (Hibbard and Mooser ibid.), as in other species of Coendou. In mandibles LACM 6210 and 6210A the proximal, approximately three- fourths of the inflected angular process is present and is as in Coendou, wherein it is subparallel to a midline separating the mandibles, and not in- flected to the degree it is in Erethizon. The greater inflection seems to reflect the marked divergence posteriad of the cheek teeth in the latter genus. Unfortunately, only four skeletons each of Coendou and Erethizon were 1968 Anza-Bgrrego Fossil Porcupine 5 Figure 2. Coendou stirtoni n. sp. Lateral (A) and ventral (B) views of left mandible, LACM 6210A; anterior (C) and distal (D) views of right humerus; dorsal view of left astragalus (E), LACM 4325; and dorsal view of right calcaneus (F), LACM 4325. available for comparison with two partial skeletons of C. stirtoni. C. stirtoni resembles both genera in most observable skeletal characters but differs in the following: The medial epicondyle of the humerus (Fig. 2 C and D) in Coendou and in C. stirtoni is oriented laterad, whereas in Erethizon it extends posteriad. Since the digital flexor muscles have origin mostly on the medial epicondyle, one is tempted to relate this difference to the known difference in habitus existing between living Coendou and Erethizon. More needs to be known about the functional anatomy of these forms before such conclusions can be made, however. The proximal articular surface of the radius in specimen LACM 6136 is ovoid, similar to the condition in Coendou, while in Erethizon it is rounded. The portion of the articular facet of the acetabulum present on the ischium in specimen LACM 6136 is elongated as it is in Coendou, while in Erethizon it is rounded. The femur (Fig. 3A) in specimens 6136 and 4325 has a well- developed third trochanter as in Erethizon. In Coendou this trochanter is poorly developed (Gupta 1966). The neck of the astragalus in specimen 4325 is distinct as in Coendou while in Erethizon it is less so. The caudal vertebrae in specimen 6136 were not found in articular posi- tion, but when they are lined up in a sand box it seems likely they did articulate. The last caudal to have a zygapophyses is the ninth. If this interpretation is 6 Contributions in Science No. 136 Figure 3. Coendou stirtoni n. sp., LACM 6136. Posterior view of left femur (A); ventral view of palate and rostrum (B); and right-lateral view of the sacrum (C). correct, then it would seem that C. stirtoni, like Erethizon, did not have a prehensile tail. In living Coendou the last caudal to have zygapophyses is the eighteenth (Gupta ibid.) . The sacrum (Fig. 3C) in specimen 6136 consists of four fused vertebrae, the neural spines of which are unfused as in adults of living Coendou (Gupta ibid.) . Named in honor of the late R. A. Stirton. 1968 Anza-Borrego Fossil Porcupine 7 4.5 l 4.0 4.5 5.0 5.5 6.0 wmA 6.5 7.0 7.5 8.0 Figure 4. Scatter diagram and calculated reduced major axes (diagonal lines) modi- fied from Imbrie (1956) for adults of Erethizon dorsatum and Coendou mexicanus. The wide separation of the two lines indicates a significant difference between the two samples. 8 Contributions in Science No. 136 8.0- 7.5- 7.0- 6.5- 5.5- 5.0- 4.5- © © ■3* : Coendou mexiconus, Recant 4.0- i i i i i i ' 4.0 4.5 5.0 5.5 6.0 6.5 7.0 WM| Figure 5. Scatter diagram and calculated reduced major axes (diagonal lines) for adults of Erethizon dorsatum and Coendou mexicanus. See Figure 4 for additional details. 1968 Anza-Borrego Fossil Porcupine 9 Discussion The living species of Coendou are adapted to an arboreal habitus, more so than the species of Erethizon. The tail in the former is prehensile, while in the latter it is not. It seems likely that C. stirtoni had a habitus more like Erethizon than the living Coendou. The latter inference is based upon the presence of a well-developed third trochanter on the femur, on the presence of zygapophyses only up to the ninth caudal vertebra, and on the large size of C. stirtoni. The degree of divergence posteriad of the upper and lower cheek teeth in adults of Erethizon versus the subparallel arrangement of the cheek teeth in Coendou (Figs. 6 and 7), seems to reflect a difference in mastication between the two genera. This functional difference can be established as follows: 1 . The scratches on the enamel of the occlusal surfaces in both the upper and lower cheek teeth are oriented anteromediad and form an angle greater than 35 degrees with the longitudinal axis of the tooth rows in Coendou (Landry 1959). In Erethizon this angle is less than 30 degrees (Fig. 8). 2. The angular process of the mandible in adults of Erethizon is inflected mediad and the posterolateral surface of the mandible is convex, while in Coendou and in juvenile Erethizon the angular processes are sub- parallel and the posterolateral surface of the mandible is flattened. 3. The fossa for the insertion of M. masseter medialis pars posterior on the side of the mandible is deeper in Coendou than in adults of Erethizon. This structure seems related to the degree of inflection of the angular process and to the degree of convexity seen in the postero- lateral part of the mandible. 4. The ascending ramus of the mandible in Coendou slants posteriad to a greater degree than in adults of Erethizon. 5. In adults of Erethizon the projection of a line superimposed upon the longitudinal axis of the lower tooth row and bisecting the occlusal surfaces of P4 and M3, passes mediad to the incisor, while in Coendou it extends to the posteromedial surface or even laterad to the incisor (Fig. 8). This measurement makes it possible to determine if the mandibular cheek teeth converge as in adults of Erethizon, or are subparallel as in Coendou and juvenile Erethizon. Juvenile specimens of Erethizon have subparallel cheek-tooth rows as well as all the characters mentioned above for the mandibles of Coendou. This sug- gests a paedomorphic relationship between the two genera, in that the adult condition in Coendou is found only in the juveniles of Erethizon. In his description of mastication in Coendou, Landry (1957: 15) stated, “The teeth occlude on one side at a time, the entire lower jaw pivoting around the condyle on the opposite side. Thus if the stroke begins on the right side the jaw pivots around the left condyle, grinding the right cheek teeth together as WIDTH OF PALATE BETWEEN 10 Contributions in Science No. 136 wi 6.0- Q = E.dorsatum. Juveniles 0=Mean of two coordinates 5.S- 2.5 3.0 3 Is 4.0 4^5 Slo S.S 6.0 6^5 WIDTH OF PALATE BETWEEN P-'s Figure 6. Scatter diagram and calculated reduced major axes (diagonal lines) for adults of Erethizon dorsatum and Coendou mexicanus. See Figure 4 for additional details. 1968 Anza-Borrego Fossil Porcupine 11 1 .2 5 ' ' ' ' .2'0 ’ ' ' ' .l'S ' ' ’ ' .10 ' ‘ ' ' .o‘s ‘ ' ' ' -6+ ' .02 ' Figure 7. Ratio diagrams modified from Simpson et al (1960), comparing several cranial dimensions of Coendou stirtoni n. sp. (A), with Erethizon dorsatum (B), and Coendou mexicanus (C). The logs of the means of the dimensions of C. stirtoni are assumed to be zero, while the differences between the log of the mean in the latter species (standard) and species being compared are plotted on the positive (--) or negative ( — ) sides of the zero line. the right condyle slides forward in the glenoid fossa.” In Erethizon mastication occurs in a similar manner, but since the teeth in adults of the latter genus converge markedly anteriad, each of the masticatory strokes more closely parallels the longitudinal axis of the tooth row than in Coendou, where the tooth rows are subparallel. From the latter it can be inferred that adult Erethizon have a more efficient chewing mechanism than either Coendou or juvenile Erethizon. Hibbard and Mooser (1963: 247-248) in referring a mandible from the Late Pleistocene (Hibbard in litt.) of Aguascalientes, Mexico, to Erethizon, noted that their specimen resembled the mandibles in Erethizon, except that: “The fossa on the lateral surface and posterior border of the coronoid process for the insertion of M. masseter medialis, pars posterior is slightly deeper than observed in the Recent specimens.” and “The base of the coronoid process of the fossil does not ascend as vertically as in Recent specimens and therefore makes it possible to see the crowns of the teeth in lateral view.” The scratches on the enamel of the occlusal surface of the cheek teeth in this specimen, and the projection of the longitudinal axis of the mandibular cheek teeth, are as in Coendou (Fig. 8). This indicates that the specimen (an adult) probably had a masticatory apparatus as in Coendou and juvenile Erethizon. The base of the ascending ramus in the mandible described by Wilson (1935) as Erethizon brachignathum indicates that it probably sloped posteriad, and the anterior end of the fossa below the coronoid process suggests a condi- 12 Contributions in Science No. 136 Figure 8. Dorsal views of the mandibles of Coendou from Aguascalientes, Mexico, UMM V47106 (A), and Erethizon dorsatum, USNM 90476 (B). Lines a-b bisect the cheek teeth longitudinally and project as shown by the arrows. The orientation of the scratches on the enamel of the occlusal surfaces of the cheek teeth is indicated by cross-hatching. See text for additional explanation. tion essentially as in Coendou. The orientation of the scratches on the enamel of the cheek teeth of this specimen has not been determined. In E. brachignathum Wilson, the Aguascalientes specimen of Hibbard and Mooser, and the maxillaries of C. stirtoni, the cheek teeth closely resemble those in living Erethizon. It is difficult to compare the structure of the cheek teeth of living Coendou with the above specimens or with those in living Erethizon, since the cheek teeth in living Coendou are markedly smaller in size, and there is great variability in the cheek-tooth structure of all these forms. The differences in masticatory function between Coendou and Erethizon are clearcut and seem, consequently, to be of greater importance than dentition in characterizing these two genera. On functional grounds, then, it seems reasonable to refer tentatively Erethizon brachignathum Wilson to ICoendou brachignathum (Wilson), and to refer the specimen from Aguascalientes described by Hibbard and Mooser (1963) to Coendou. 1968 Anza-Borrego Fossil Porcupine 13 The paedomorphic nature of the structures relating to mastication in Erethizon strongly suggests that the condition found in adults of Coendou and in juveniles of Erethizon are primitive and presumably functionally less effi- cient than the condition found in adult Erethizon. Thus, it would seem that Coendou is the structural ancestor of Erethizon, and that the living species of Coendou either are relicts adapted to a narrow niche in the rain forests, or that both genera descended from an unknown, common ancestor in South America. Unfortunately, a fossil record of prehensile-tailed Coendou is unknown. Literature Cited Dibblee, T. W.. Jr. 1954. Geology of the Imperial Valley Region, California. Bull., Div. of Mines, State of Calif., 170, Chap. II: 21-28, 3 figs., 1 map. Downs, T. 1957. Late Cenozoic vertebrates from the Imperial Valley region, California. (Abstract) Bull. Geol. Soc. Amer., 68: 1822-23. Gupta, B. B. 1966. Skeleton of Erethizon and Coendou. Mammalia, 30(3) : 495-497. Hibbard, C. W., and O. Mooser 1963. A porcupine from the Pleistocene of Aguascalientes, Mexico. Contrib. Mus. Paleo. Univ. Michigan, 18(16) : 245-250, 1 fig., 1 pi. Howard, H. 1963. Fossil birds from the Anza-Borrego Desert. L. A. Co. Mus. Contrib. Sci., no. 73: 1-33, 1 fig., 3 pis. Imbrie, J. 1956. Biometrical methods in the study of invertebrate fossils. Bull. Amer. Mus. Nat. Hist., 108: 215-252, 10 figs. Landry, S. O., Jr. 1957. The interrelationships of the New and Old World hystricomorph ro- dents. Univ. Calif. Publ. in Zool., 56: 1-118, 37 figs., 5 pis. Simpson, G. G., Anne Roe, and R. C. Lewontin 1960. Quantitative Zoology; rev. ed. Harcourt, Brace and Co., 440 p., 64 text figs. White, J. A. 1964. Kangaroo rats (Family Heteromyidae) of the Vallecito Creek Pleisto- cene of California. (Abstract) Geol. Soc. Amer. Special Paper, 82: 288-89. White, J. A., and T. Downs 1961. A new Geomys from the Vallecito Creek Pleistocene of California, with notes on variation in Recent and fossil species. L. A. Co. Mus. Contrib. Sci., 42: 1-34, 17 figs. Wilson, R. W. 1935. A new species of porcupine from the later Cenozoic of Idaho. Jour. Mamm, 16: 220-222. 14 Contributions in Science No. 136 Woodard, G. D. 1963. The Cenozoic Stratigraphy of the Western Colorado Desert, San Diego and Imperial Counties, Southern California. Ph.D. thesis, Univ. Calif., Berkeley. Woodring, W. P. 1931. Distribution and age of the Tertiary deposits of the Colorado Desert. Carnegie Inst. Washington Publ., 148: 1-25. Table 1 Measurements in millimeters of skull parts in Coendou stirtoni n. sp. lacm 17633 LACM6136 lacm 6210A lacm 6210 (Type specimen) Breadth of rostrum lateral to incisors 17.6 Alveolar length of upper cheek-tooth series 27.7 25.9 Length of upper diastema from posterior surface of 39.2 _ incisor to P4 Width of crown on P4 8.4 Width of crown on M1 8.4 — — — Width of crown on M2 7.8 — — — Alveolar length of lower cheek-tooth series 30.7 Depth of mandible below M2 17.7 15.8 16.7 Breadth of palate between alveoli of P4’s 6.6 6.0 Breadth of palate between alveoli of M3’s 11.5 10.5 Width of upper incisor — 4.8 — — Anteroposterior thickness of upper incisor 6.1 1968 Anza-Borrego Fossil Porcupine 15 Table 2 Measurements in millimeters of skeletal parts in Coendou stirtoni n. sp. Humerus: distance from medial to lateral epicondyle. Ulna : from proximal surface of articular facet of humerus to tip of olecranon process. Radius: proximal head: Anteroposterior width Transverse width LACM 6136 LACM 4325 — 28.2 — 11.5 12.8 8.2 Femur: from mediolateral base of third trochanter to proximal tip of greater trochanter. — 35.3 Anteroposterior width of head 21.5 — Astragalus: width across trochlea — 16.6 greatest anteroposterior length of lateral trochlea — 15.2 width of distal articular head — 8.6 Calcaneus: greatest anteroposterior length — 34.5 greatest width — 20.1 Cuboideum: from groove of M. peroneus longus to lateral margin of antiplantar surface — 7.2 LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE Number 137 May 7, 1968 \n 73 U Lftf SWOLLEN DORSAL FIN ELEMENTS IN LIVING AND FOSSIL CARANX (TELEOSTEI: CARANGIDAE) j! By Harry L. Fierstine Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8V2 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor SWOLLEN DORSAL FIN ELEMENTS IN LIVING AND FOSSIL CARANX (TELEOSTEI: CARANGIDAE) By Harry L. Fierstine1 Abstract: Thirty-two specimens of Caranx hippos (Lin- naeus) and two large specimens each of C. caballus Gunther, C. lugubris Poey, C. marginatus Gill, and C. melampygus Cuvier from the Gulf of California and the eastern Pacific Ocean were examined for an hypertrophied basal bone of the first dorsal spine. The hyperostoses were found only in specimens of C. hippos larger than 343 mm standard length, regardless of sex. The bone is egg-shaped with a deep posterior notch. It is here suggested that this structure may aid in erection of the dorsal fin by increasing the surface area and by changing the direction of pull of the inclinator muscles of the dorsal fin. The element is formed of a thin outer layer of acellular compact bone which sur- rounds an hypertrophied center of cancellous bone. The cancel- lous bone is arranged in honeycomblike chambers which are filled with adipose tissue. Bony elements found in the marine middle Miocene deposits of Sharktooth Hill, Kern County, California, are identified as be- longing to Caranx sp. and constitute the first record of the genus from the California Miocene. The presence of numerous branch- ing canals, laminae, and lack of honeycomblike chambers elimi- nate their positive identification as C. hippos. The presence of an incipient posterior notch and a more oval shape eliminate their positive identification as C. carangopsis Heckel from the Austrian Miocene. This record lends support for assuming a near-shore, tropical or sub-tropical environment for the Sharktooth Hill fauna. Introduction Hyperostosis or hypertrophied bone is often encountered in various parts of the skeleton of perciform fishes (Korschelt, 1940; Schlumberger and Lucke, 1948; Lucke and Schlumberger, 1949; Breder, 1952; Konnerth, 1966). When hyperostosis occurs, it is normally not restricted to one ele- ment but is found in more than one area of the skeleton. The structures most commonly affected are the supraoccipital crest and the basal elements of the dorsal and anal fins. In certain species, hypertrophied bony elements are so common that the condition cannot be considered abnormal even though closely related species lack swollen bones (Breder, 1952). The fam- ilies of fishes that commonly display hyperostoses are the Carangidae (Stein- research Associate in Vertebrate Paleontology, Los Angeles County Museum of Natural History, and Department of Biology, California State College, Long Beach, California. Current address: Biological Sciences Department, California State Poly- technic College, San Luis Obispo, California, 93401. 1 2 Contributions in Science No. 137 dachner, 1859; Starks, 1911; Korschelt, 1940; Barnard, 1948; Gopinath, 1952; Konnerth, 1966), Ephippidae (Gregory, 1933; Schlumberger and Lucke, 1948; Breder, 1952), Sciaenidae (Chabanaud, 1926), Sparidae (Kesteven, 1928; Takahashi, 1929; Ebina, 1936), and Trichiuridae (Gun- ther, 1860; James, 1960). The following study reveals the morphology of the swollen dorsal basal elements in the carangid fish, Caranx hippos (Linnaeus). This information is applied to the identification of hyperostoses in the marine middle Miocene deposits of Sharktooth Hill, Kern County, California. Materials and Methods Thirty-two specimens of Caranx hippos (Linnaeus) and two large specimens each of C. caballus Gunther, C. lugubris Poey, C. marginatus Gill, and C. melampygus Cuvier from the Gulf of California and the eastern Pacific Ocean were examined by dissection or by X-ray photographs. Except for fifteen small specimens of C. hippos examined by Mr. Edmund S. Hobson (Zoology Department, University of California, Los Angeles) and discarded in the field, all specimens are contained in the ichthyological collections of the University of California, Los Angeles, or the Los Angeles County Museum of Natural History. Fossil hyperostoses from the marine middle Miocene deposits of Shark- tooth Hill, Kern County, California, are housed in the Vertebrate Paleon- tology Section of the Los Angeles County Museum of Natural History. Ground thin sections of the fossil bones were prepared by the method described by Enlow and Brown (1956). Acknowledgements I wish to thank Dr. Boyd W. Walker and Mr. Wayne Baldwin for the use of the ichthyological collection and equipment at the University of Cali- fornia, Los Angeles. Dr. Shelton P. Applegate, Associate Curator of Verte- brate Paleontology, Los Angeles County Museum of Natural History, offered encouragement and advice as well as access to the paleontological collection. Mr. Frederick H. Berry, Mr. Robert K. Liu, and Drs. David K. Caldwell, Tilly Edinger, and Richard R. Rosenblatt referred me to important literature. Mr. Edward D. Mitchell called attention to examples of fossil hyperostosis, Mr. Dick Bishop gave the Los Angeles County Museum of Natural History fossil hyperostoses from his personal collection, and Mr. Edmund Hobson made observations on 18 Caranx hippos at Rancho Buena Vista, Baja California del Sur, Mexico. One recent hyperostosis was stained, sectioned, and photographed through the courtesy of Dr. Marshall R. Urist, M.D., Director of the Bone Research Laboratory, School of Medicine, Uni- versity of California, Los Angeles. Mr. Armando Solis of the Los Angeles 1968 CARANX Dorsal Fin Elements 3 County Museum of Natural History deserves the entire credit for Figures 2 and 4 and for the printing of Figures 1, 3, and 5. Figure 1. Caranx hippos (Linnaeus), tracings from X-ray photographs showing the enlarged basal bone of the first dorsal spine. Note that the hypertrophied bone sur- rounds the second and third basal elements and is not a fusion of more than one struc- ture. UCLA number W51-21, Mexico, Gulf of California. A. Specimen number 9, Standard length 344 mm, male. B. Specimen number 8, Standard length 425 mm, sex undetermined. 4 Contributions in Science No. 137 Results The basal bone of the first dorsal spine is swollen in all specimens of Caranx hippos larger than 343 mm standard length, regardless of sex (Fig. 1). The swollen fin element is absent in 22 smaller specimens (ranging from 170 to 225 mm standard length). The length of the enlarged basal tends to increase linearly with the length of the fish. The excised bone (Fig. 2) is egg-shaped with a deep posterior notch which surrounds the succeeding one or more basal elements of the dorsal fin. Histological examination reveals a thin outer layer of acellular compact bone which surrounds an hypertrophied center of cancellous bone (Fig. 3). The cancellous bone is formed into a regular network of honeycomblike chambers which are fat filled. In dried preparations, the thin compact layer often cracks and pulls away from the hypertrophied cancellous bone (Fig. 2). No unusual blood vessel or nerve supply was noted. The great lateral myomeric musculature does not attach to the hyper- Figure 2. Caranx hippos (Linnaeus), basal bone of first dorsal spine, field number 63040203, male, fork length 443 mm, Pacific coast of Costa Rica. Enlarged 1.5 times. A. Left lateral view. Note the thin first dorsal spine which articulates with the swollen basal and the stouter second dorsal spine which is attached to the hypertrophied basal only by ligaments. Additional comments in text. B. Ventral view. Note the deep pos- terior notch which is partly filled with the basal element of the second dorsal spine. 1968 CARA NX Dorsal Fin Elements 5 Figure 3. Caranx hippos (Linnaeus), longitudinal section of the enlarged basal bone of the first dorsal spine, UCLA number W51-21, specimen number 9, standard length 344 mm, Mexico, Gulf of California. A. Acellular outer compact layer. Enlarged 225 times. B. Acellular inner cancellous layer with adipose tissue enclosed in honeycomb- like chambers. Enlarged 225 times. 6 Contributions in Science No. 137 Figure 4. Caranx sp. Basal bone of first dorsal spine, LACM number 10702, Middle Miocene, Sharktooth Hill, Kern County, California. Enlarged 2.5 times. A. Left lat- eral view. B. Ventral view. Note the shallow posterior notch. C. Posterior view. Note the shallow posterior notch. ostosis, but curves around it. The supracarinales muscle forms a thin anter- ior layer which originates on the posteriormost free basal and inserts on the anterior midborder of the hyperostosis. The dorsal surface of the swollen basal serves as the point of origin for the inclinator muscles of the first and second spines of the dorsal fin. The inclinator muscles are unusual in two ways: (1) in most fishes they originate from the surface of the great lateral muscles rather than from a bony surface and (2) they normally take a lateral-medial orientation rather than an anterior-posterior direction. These two differences suggest that the enlarged basal may function to erect the dorsal fin by increasing the surface area for the origin of the inclinator muscles of the first and second dorsal spines. 1968 CARA NX Dorsal Fin Elements 7 Two large specimens each of C. caballus, C. lugubris, C. marginatus, and C. melampygus from the eastern Pacific Ocean lacked the enlarged basal bone. Fossil bones from the marine middle Miocene deposits of Sharktooth Hill, Kern County, California, are readily identified as hypertrophied teleost elements. The shallow posterior notch in the larger bones and the general shape suggest that they are dorsal basal elements of a fish similar to C. hippos (Fig. 4). Ground bone sections reveal a very different histology than that found in recent C. hippos (Fig. 5). The compact bone is hypertrophied, laminated, and filled with a lacework of fine canals. These smaller canals interconnect with larger spaces in the cancellous center. Discussion and Conclusion The shape of the hyperostosis, particularly the deep posterior notch, seems to be characteristic for the swollen first dorsal basal in Caranx hippos. Starks (1911) mentioned a swollen post-temporal in a 355-mm C. crysos (Mitchill) and readily distinguished it from the hypertrophied dorsal bone of a 736-mm C. hippos. Steindachner (1859) described the unique shape of the swollen dorsal basal in an 1220-mm C. carangus Bloch (= C. hippos, see Berry, 1959) and readily distinguished it from swollen ribs, post-temporals, and other swollen bones in the same fish. Gopinath (1952) described the swollen supra- occipital of C. sexfasciatus Quoy and Gaimard as peapod-shaped. None of the published figures of swollen elements in other teleost fishes appear similar to the swollen dorsal basal of C. hippos. The histology of the hypertrophied dorsal basal is similar to a figured cross-section of the normal (unswollen) lower jaw of C. bartholomaei Cuvier figured by Moss ( 1961 ) , except for the hypertrophy of the cancellous tissue in Caranx hippos. The function of the swollen basal element is enigmatic. Breder (1952) states that bony growths are never added to a fish as a response to equilibrium or for counter-balancing structures. Furthermore, he noted that they are fat- filled and would not alter the specific gravity. To me, the best functional explanation seems to be that it provides a better mechanism for erection of the dorsal fin. If the growth has a selective advantage (such as surface area for muscle origin), it could not increase linearly with the length of the fish unless it could increase its volume without increasing in weight. This problem could be solved by increasing the cancellous tissue and filling the spaces with buoyant fat, so that an increase in volume would not change the center of gravity for the fish. However, a large C. hippos may have other hypertrophied bones (Starks, 1911; Konnerth, 1966), and it is likely that a more complex explana- tion is necessary to explain the phenomena of hyperostoses. 8 Contributions in Science No. 137 Figure 5. Caranx sp. Cross-section of fossil hyperostosis, LACM number , Middle Miocene, Sharktooth Hill, Kern County, California. A. Outer and inner lay- ers showing small canals, lamellae, and large central spaces. Enlarged 21 times. B. Anastomosing small canals. Enlarged 225 times. C. Inner layer showing anasto- mosing of large central spaces and small canals. Enlarged 225 times. 1968 CARANX Dorsal Fin Elements 9 The branching tubular network in the fossil bones may have contained fat and may represent an early attempt to increase in volume without increas- ing the specific gravity. Since the largest fossil hyperostoses are only about one-half the size of those found in living C. hippos, it is possible that this method of increasing volume was unsuccessful. The shape of the fossil hyperostoses from Sharktooth Hill indicate that they belong to the genus Caranx. The histological differences detract from their identification as C. hippos. Steindachner (1859), in a redescription of C. carangopsis Heckel from the Miocene of the Vienna Basin, Austria, de- scribed a swollen basal bone. His material was fragmentary and neither his illustrations nor his description indicate a shallow posterior notch. The overall shape may be more spherical than that of the Sharktooth Hill specimens. The type description (Heckel, 1852) is too brief to be helpful. Considering these discrepancies, I choose to identify the hyperostoses as belonging to Caranx sp. If reexamination of the Austrian material shows a similar histology, then the Sharktooth Hill specimen should be considered as belonging to C. carangopsis. Since most living species of Caranx are restricted to near-shore tropical and subtropical waters, a similar environment for the Sharktooth Hill fauna is suggested. Other piscine evidence (Applegate and Fierstine, unpublished data) supports this supposition. Literature Cited Barnard, K. H. 1948. Further notes on South African marine fishes. Ann. So. African Mus., 36:341-406. Berry, F. H. 1959. Young Jack Crevalles ( Caranx species) off the southeastern Atlantic coast of the United States. U.S. Fish and Wildlife Serv., Fish. Bull., 59(152) : 4 17-535. Breder, C. M., Jr. 1952. The problem of directives to cellular proliferation as illustrated by onto- genetic processes in certain fishes. Growth, 16:189-198. Chabanaud, P. 1926. Frequence, Symetrie et Constance specifique d’Hyperostoses Externes chez divers Poissons de la Famille des Scienides. Comptes Rendus de 1’Academie des Sciences, Paris, 182: 1647-1649. Ebina, K. 1936. On the growth of Evynnis cardinalis (Lacepede). J. Imp. Fish. Inst., Tokyo, 31:69-78. Enlow, D. H., and S. O. Brown 1956. A comparative histological study of fossil and recent bone tissues. Part I. Texas J. Sci., 8:405-443. Gopinath, K. 1952. On a peculiar bone formation in the supra-occipital crest of some car- angid fishes. J. Zool. Soc. India, 3:267-276. 10 Contributions in Science No. 137 Gregory, W. K. 1933. Fish skulls: a study of the evolution of natural mechanisms. Trans. Amer. Philosophical Soc., 23:75-481. Gunther, A. 1860. Catalogue of the Acanthopterygian fishes in the collection of the British Museum, 2:1-548. Heckel, J. J. 1852. [Caranx carangopsis und fossile Delphin-Wirbel.] Jahrbuch der geo- logischen Reichsanstalt, Vienna, 3(2) : 160-161. James, P. S. B. R. 1960. Instances of excessive thickening of certain bones in the ribbon fish, Trichiurus lepturus Linnaeus. J. Mar. Biol. Assn. India, 2:253-258. Kesteven, H. L. 1928. Contributions to the cranial osteology of the fishes. No. VI. Rec. Aus- tralian Mus., 16:316-345. Konnerth, A. 1966. Tilly bones. Oceanus, 12(2) :6-9. Korschelt, E. 1940. fiber Besonderheiten im Aufbau des Knochenfischskelets. Zeitschrift fur wissenschaftliche Zoologie, 152:507-546. Lucke, B., and Schlumberger, H. G. 1949. Neoplasia in cold-blooded vertebrates. Physiol. Reviews, 29:91-126. Moss, M. L. 1961. Studies of the acellular bone of teleost fish. I. Morphological and syste- matic variations. Acta Anatomica, 46:343-362. Schlumberger, H. G., and B. Lucke 1948. Tumors of fishes, amphibians and reptiles. Cancer Research, 8:657-754. Starks, E. C. 1911. Osteology of certain scombroid fishes. Stanford IJniv. Publ., Univ. Series, 5:1-49. Steindachner, F. 1859. Beitrage zur Kenntniss der fossilen Fisch-Fauna Osterreichs. Sitzungs- berichte der Mathem.-Naturw. Classe der Kaiserlichen Akademie der Wissenschaften, Wien, 37:673-704. Takahashi, K. 1929. Studie fiber die Fischgeschwfilste. Zeitschrift fur Krebsforsch., 29:1-73. CONTRIBUTIONS IN SCIENCE LOS ANGELES COUNTY MUSEUM Number 138 May 7, 1968 b'z 7 3 loi LiU? £ REVIEW OF THE LANTERNFISH GENUS LAMP ADEN A WITH A DESCRIPTION OF A NEW SPECIES By Basil G. Nafpaktitis and John R. Paxton Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 814 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS.— All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF.— Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor REVIEW OF THE LANTERNFISH GENUS LAMPADENA WITH A DESCRIPTION OF A NEW SPECIES By Basil G. Nafpaktitis1 and John R. Paxton2 Abstract: The genus Lampadena, family Myctophidae, includes seven known species, one of which, occurring in the southern hemisphere, is here described as new. We consider L. nitida to be conspecific with L. luminosa and do not recognize the subgenus Lychnophora. The otoliths of the species in the genus show interspecific differences, and thus are of taxonomic value. The distribution of each species is plotted and tentative conclu- sions concerning relationships within the genus are presented. Introduction In spite of the considerable amount of work which has been done on the family Myctophidae, many groups within the family need to be reworked. There are three reasons for this: (a) inadequate original and subsequent descriptions, including, as Bolin (1959) has emphasized, extremely general statements concerning the distribution of individual species; (b) addition of new species; and (c) the potential for more critical comparative studies due to the increase in the number and size of collections. Prior to Bolin’s work (1959) the genus Lampadena comprised a total of nine described species assigned by Fraser-Brunner (1949) to two subgenera, Lampadena and Lychnophora. Two of these nine species were placed by Bolin in a separate genus, Taaningichthys. Further, the same author placed Lampa- dena braueri Zugmayer in the synonomy of L. speculigera Goode and Bean. Paxton (1963) described a new species, L. urophaos, from the eastern north Pacific, and placed it in the subgenus Lychnophora. At that point the composi- tion of the two subgenera was as follows: subgenus Lampadena — L. speculi- gera, L. dea, L. chavesi, L. anomala; subgenus Lychnophora — L. luminosa, L. nitida, L. urophaos. The two main purposes of the present work are (a) to bring “under one roof” detailed descriptions of all the species of the genus found scattered in the literature, and (b) to present a survey of those species which occur in the southern oceans. The genus as a whole is rather uncommon. Examination of many more collections around the world will probably result in an increase in the number of known species in the genus and further elucidate individual patterns of distribution. With representatives of all currently known species department of Biological Sciences and Allan Hancock Foundation, University of Southern California. 2Present address, The Australian Museum, Sydney, Australia. 1 2 Contributions in Science No. 138 available to us, we thought it advisable to redescribe them in as much detail as possible, thus hoping to minimize confusion in future investigations. Materials and Methods The ultimate rays of the dorsal and anal fins are double, but they have here been considered as single. Only the well-developed gill rakers of the first gill arch (one side) have been counted. In the method of taking measurements and presenting data we follow Bolin (1939). A few modifications and addi- tions were made, however. These are as follows: depth of body — the vertical through the upper end of the base of the pectoral fin; caudal peduncle length — the distance from the base of the last anal ray to the end of the hypural plate; caudal gland length — the greatest extent of the luminous tissue, not always a midline measurement. Most measurements and proportionality data have proven unsatisfactory for the elucidation of differences between species of the genus Lampadena. Rather than present what we consider meaningless data under each species description, those measurements which have proven useful are summarized in Table 1. Certain proportions, notably the lengths of the dorsal and anal fin bases, have been utilized in the past to characterize genera. To enable a future comparison of the genus Lampadena with other myctophid genera, those measurements, expressed as per cent of standard length (SL), which are not useful at the specific level, are summarized for all species (except L. anomala) as follows: length of head, 28.4-36.6; depth of head, 18.2-23.3; length of upper jaw, 19.1-26.9; depth of body, 18.3-23.8; predorsal distance, 44.3-50.0; prepectoral distance, 28.9-37.4; preventral distance, 40.9-50.5; preadipose distance, 69.2-78.8; length of base of dorsal fin, 12.6-20.7; length of base of anal fin, 1 1.8-16.7. A list of individual measurements and calculated percentages of standard lengths for all specimens is available on request. Vertebral counts and structure were determined from X-rays. Under the description of each species, locality data are presented for all specimens examined and ranges of standard lengths are given. Coordinates for the starting position only of each trawl are included. All time is in local time. All captures were made with a 10' Isaacs-Kidd Midwater Trawl (IKMT) , unless otherwise indicated. For open IKMT captures, the depth sampled is recorded from the surface to maximum depth reached by the gear. For captures with an IKMT equipped with a closing device, the discrete depth sampled is indicated. The terminology of body photophores is that of Bolin (1939) . Consistent within all specimens examined, and indeed in the entire family, is the presence of three Br photophores on the branchiostegal membranes under the ventral sheet of the dentaries. Since they are invariably present, their occurrence is omitted from the species descriptions. Otoliths were measured with an ocular micrometer, then lightly smeared 1968 Review of Lanternfish LAMP ADEN A 3 with graphite to bring out details in their sculpture and photographed. The entire series of otoliths is deposited in the extensive collection established by John Fitch of the California Fish and Game at Terminal Island and supported by a National Science Foundation Grant. The terminology on otoliths follows that of Frizzell and Dante (1965). Key to the Species of the Genus Lampadena la. None of the PO abruptly and highly elevated 2 lb. P04 abruptly and highly elevated, about over P03 L. luminosa (Garman, 1899) 2a. VO plus SAO 7-9; AOa 5-7; Pol immediately below or very close to lateral line 3 2b. VO plus SAO 5-6; AOa 3-4; distance between Pol and lateral line about half as great as that between Pol and ventral contour of caudal peduncle L. anomala Parr, 1928 3a. Prc1-Prc2 interspace equal to, or greater than, three times the diameter of a photophore of this series 4 3b. Prc1-Prc2 interspace much shorter than three times the diameter of a photophore of this series 5 4a. Last two (sometimes three) AOa entirely behind base of anal fin; two AOp; infracaudal luminous gland very long, at least 1.5 times as long as least depth of caudal peduncle, almost twice as long as diameter of eye; crescent-shaped patch of whitish tissue on iris above pupil; pterotic spine directed posteriorly L. chavesi Collett, 1905 4b. No AOa behind base of anal fin; four to five AOp (rarely three); infra- caudal luminous gland shorter than 1.5 times the least depth of caudal peduncle and about 1.5 times as long as diameter of eye; no crescent- shaped patch of whitish tissue on iris above pupil; pterotic spine directed downward and forward (in specimens longer than about 30 mm) L. dea Fraser-Brunner, 1949 5a. Gill rakers 6-8+1 + 12-17; supracaudal gland shorter than infracaudal gland; mesopterygoid teeth uniformly small 6 5b. Gill rakers 3-5+1+8-10; supracaudal gland equal in length to or some- what longer than infracaudal gland; posterior mesopterygoid teeth notice- ably enlarged L. urophaos Paxton, 1963 6a. Distance between posterior end of base of anal fin and anterior margin of infracaudal gland equal to, or slightly greater than, length of this gland; photophores small, in specimens smaller than 30 mm AOa about two organ diameters apart from each other; first and usually second AOp in front of infracaudal gland; AOa level; gill rakers 6-7+1 + 12-14, total 4 Contributions in Science No. 138 19—22 L. speculigera Goode & Bean, 1896 6b. Distance between posterior end of base of anal fin and anterior margin of infracaudal gland equal to about one-fourth of length of this gland; photo- phores large, in specimens smaller than 30 mm AOa less than one organ diameter apart from each other; all AOp well over infracaudal gland; last AOa usually distinctly raised above level of rest of organs of same series; gill rakers 7-8 + 1 + 16-17, total 24-26 L. notialis sp. n. Lampadena luminosa (Garman, 1899) Fig. 1 R/V ANTON BRUUN, cruise III, sta. 147, 06°54'N, 59°55'E, 16 August 1963, 1845-2213 hrs, depth sampled 0-750 m, bottom depth 1775-3015 m, one specimen, 31.0 mm; sta. 151, 04°52'S, 60°02'E, 23 August 1963, 0625- 1350 hrs, depth sampled 0-2030 m, bottom depth 3895 m, two specimens, 67.5-86.0 mm. R/V ANTON BRUUN, cruise VI, sta. 334A, 06°01'N, 64°59'E, 24 May 1964, 1912-2345 hrs, depth sampled 0-700 m, bottom depth 4663 m, one specimen, 53.0 mm; sta. 335B, 03°46'N, 65°05'E, 26 May 1964, 0100-0850 hrs, depth sampled 0-275 m, bottom depth 2926 m, two specimens, 39.5-49.0 mm; sta. 341B, 07°56'S, 65°14'E, 1-2 June 1964, 2200-0300 hrs, depth sampled 0-504 m, bottom depth 4200 m, one specimen, 22.0 mm; sta. 342A, 09°57'S, 64°55'E, 2 June 1964, 1755-2250 hrs, depth sampled 0-525 m, bottom depth 3200 m, 12 specimens, 16.5-29.0 mm; sta. 343 A, 12°10'S, 64°54'E, 4 June 1964, 0020-0510 hrs, depth sampled 0-798 m, bottom depth 3200 m, one specimen, 20.0 mm; sta. 345E, 17°58'S, 65°34'E, 7 June 1964, 2130-2400 hrs, depth sampled 0-120 m, bottom depth 3000 m, five specimens, 17.0-22.0 mm. Scripps Institute of Oceanography, SIO 60-239-25H, 04°56'N, 142°54'W, 6 July 1960, one specimen, 22.5 mm. R/V ATLANTIS II, cruise 13, sta. RHB 1005, 41°26'N, 59°01'W, 4 September 1964, 1045-1410 hrs, depth sampled 0-555 m, one specimen, 28.0 mm; sta. RHB 1013, 41°36'N, 52°21'W, 6 September 1964, 1940-2320 hrs, depth sampled 0-65 m, one specimen, 23.5 mm. R/V CHAIN, cruise 17, sta. RHB 801, 00°15'S, 18°40'W, 26 April 1961, 0250-0605 hrs, depth sampled 0-85 m, one specimen, 49.0 mm. R/V CHAIN, cruise 35, sta. RHB 960, 07°39'N, 45°14'W, 11 February 1963, 2005-2350 hrs, depth sampled 0-60 m, one specimen, 27.0 mm; sta. RHB 966, 01°13'S, 34°35'W, 17 February 1963, 0335-0605 hrs, depth sampled 0-102 m, four specimens, 23.0-33.0 mm. R/V CHAIN, cruise 60, sta. RHB 1252, 16°45'N, 64°18'W, 24 May 1966, 2015-2335 hrs, depth sampled 0-84 m, 12 specimens, 19.0-33.0 mm; sta. RHB 1253, 16°38'N, 64°27'W, 25 May 1966, 0046-0400 hrs, depth sampled 1968 Review of Lanternfish LAMP ADEN A 5 Figure 1. Lampadena luminosa , 67.4 mm in standard length; 04° 52' S, 60° 02' E. 0-133 m, two specimens, 20.0-27.0 mm; sta. RHB 1258, 13°32'N, 71°24'W, 27 May 1966, 2030-2300 hrs, depth sampled 0-210 m, one specimen, 19.0 mm; sta. RHB 1261, 13°04'N, 73°12'W, 28 May 1966, 1628-1955 hrs, depth sampled 0-300 m, four specimens, 29.0-66.0 mm; sta. RHB 1263, 12°58'N, 73°34'W, 29 May 1966, 0040-0450 hrs, depth sampled 0-120 m, nine speci- mens, 27.0-69.0 mm; sta. RHB 1286, 19°46'N, 83°07'W, 10 June 1966, 0020- 0505 hrs, depth sampled 0-86 m, nine specimens, 17.0-51.0 mm; sta. RHB 1289, 21°11'N, 85°12'W, 10 June 1966, 2010-2320 hrs, depth sampled 0-170 m, two specimens, 41.5-64.0 mm; sta. RHB 1290, 21°17'N, 85°22'W, 11 June 1966, 0030-0400 hrs, depth sampled 0-124 m, eight specimens, 17.0-71.0 mm (otolith of 71.0 mm specimen photographed, Fig. 10); sta. RHB 1307, 27°01' N, 90°02'W, 22 June 1966, 0007-0400 hrs, depth sampled 0-95 m, six speci- mens 39.0-67.5 mm; sta. RHB 1315, 25°46'N, 79°47'W, 25-26 June 1966, 0002-0400 hrs, depth sampled 0-71 m, 41 specimens, 18.0-26.0 mm. D. 15; A. 14(13-15) ; P. 16(15-17) ; V. 8; gill rakers 4+1+9(8-10), total 14(13-15); PO 5; VO 4-5; AOa 5-6 (rarely 7); AOp 2; Prc 2+1; lateral line scales 35-37; vert. 36(37), 8 x-rayed specimens. Of 16 specimens counted, only one had P. 17. A relatively large myctophid fish; snout obtuse and round; mouth large, terminal, its cleft somewhat oblique; eye of moderate size, its diameter 3.7 to 4.3 in length of head; opercular margin concave posterodorsally, convex posteroventrally, tapering into a rather sharp point located above PV02 and with a distinct indentation at about level of same organ; pterotic spine strong and directed posteriorly. Origin of dorsal fin distinctly in advance of vertical through base of outer- most ray of ventral fin; origin of anal fin well behind vertical through end of base of dorsal fin; base of adipose fin somewhat in advance of vertical through end of base of anal fin; pectoral fin long, extending to level of second or third VO; ventral fin reaching anus. 6 Contributions in Science No. 138 A small, roundish Vn posteroventrad to nasal apparatus, capped by crescent-shaped, darkly pigmented tissue; a narrow strand of black tissue extending along entire anterior margin of orbit. A very small, deeply embedded Opi close behind angle of mouth; Op2 larger than general body photophores, directly over or slightly posterior to Op! and about at level of ventral margin of orbit. PLO in advance of base of pectoral fin and 1.5 to 2 times its own diameter below lateral line. PVOi at about level of posterior end of maxillary and some- what in advance of vertical through PV02; PV02 somewhat behind straight line connecting PLO with PVO], distinctly above level of Op2 and in front of middle of base of pectoral fin. First PO interspace 1.2 to 1.4 times as wide as second, which is equal to or somewhat narrower than that between P03 and P05; P05 displaced laterally, situated about its own diameter in front of and anteromesad to base of outermost ray of ventral fin; P04 highly and abruptly elevated, over, slightly anterior or slightly posterior to vertical through P03 and on same level with, or somewhat higher than, PVOi. VLO distinctly behind vertical through base of outermost ray of ventral fin and somewhat nearer to lateral line than to base of ventral fin. VO series forming a gentle arc; VC^ posteromesad to base of inner ray of ventral fin. SAO slightly angular, with last VO, SA04 and SAOo equidistant and on straight, steeply ascending line; SA02-SA03 interspace about twice as wide as SAOj-SAOo interspace; SA03 on or slightly posterior to vertical through SA02 and immediately below lateral line. AOa level and usually evenly spaced. Pol behind vertical through last AOa and immediately below lateral line. AOpx slightly in front of, or directly over anterior end of infracaudal luminous gland and separated from AOp2 by a distance about 1.5 times the diameter of these organs. Prc4 and Prc2 on horizontal line, space between them varying from 1 to 2.5 times the diameter of these organs; Prc3 at base of middle rays of caudal fin, on level of lateral line and well behind vertical through Prc2. Supra- and infracaudal luminous glands of about equal length, somewhat shorter than diameter of eye; distance between anterior end of infracaudal gland and posterior end of base of anal fin about equal to or slightly less than length of gland. Several series of needlelike teeth on both jaws, those of inner series dis- tinctly larger and more widely spaced; four to five large, recurved teeth directed anteriorly at posterior end of dentary; one to two large teeth hooked forward in inner dentary row near symphysis (these peculiar teeth are undifferentiated in young specimens and often missing in large ones) ; anteriormost premaxillary teeth in outer row very slightly hooked. Palatines with a narrow band of small teeth. A small patch of similar teeth on distal end of each limb of vomer. Mesopterygoid teeth in large patch with medial series greatly enlarged in maturing fish (about 65 mm), undifferentiated in size in young individuals; last few teeth on posterolateral margin of patch also enlarged. Quite often, in 1968 Review of Lanternfish LAMP ADEN A 1 the fish examined, several specimens of comparable size from the same haul showed marked differences in the development of the medial series of mesop- terygoid teeth. Neural arches of anterior vertebrae not greatly expanded. A direct comparison of the type specimens of Lampadena luminosa and L. nitida (Taning, 1928) revealed no differences. Taning, it seems, established his subspecies L. luminosa nitida mainly on the basis of the division of the VO series into two groups (Bolin, pers. comm.). A grouping of the VO into 1 + 1+3 is clearly visible in Garman’s type. Furthermore, examination of several specimens, from both the Atlantic and Indian Oceans, shows that irregularities in the spacing of both the VO and the AOa series are not un- common within a given population. Some specimens have five VO evenly spaced, others, from the same station, have the same number of VO but the series is divided into two groups (2+3). Still others possess only four VO either evenly spaced or with the V02-V03 interspace distinctly wider than the other interspaces in the series. Finally, in some cases where there are five VO, the third is markedly reduced in size. With these differences appearing to be a simple expression of intra-specific variation and with no other differences sharp enough to justify the establishment of two species, or even subspecies, we place Taning’s L. luminosa nitida in the synonymy of L. luminosa Garman. It should also be added here that among the Indian Ocean specimens two trends involving the numbers and arrangement of the VO, AOa and Prc can be discerned. Specimens with four VO usually, but not invariably, have five AOa. The Prcx and Prc2 in these specimens are separated by a distance equal to 2 to 2.5 times the diameter of a body photophore. On the other hand, specimens with five VO tend to have six AOa. The PrCj-Prc2 interspace in the latter individuals is equal to 1 to 1.5 times the diameter of a body photophore. Inter- mediates between the two patterns are not uncommon and the differences in the three characters mentioned seem to intergrade. Lampadena luminosa appears to be a tropical species, found mainly in equatorial waters of all three oceans, between 20°N and 20°S (Fig. 8). Its occurrence in the Gulf of Mexico, Straits of Florida and high latitudes in the western north Atlantic is probably due to the influence of the Gulf Stream. Individuals of many tropical myctophid species have often been taken all along the path of this current as far as 42°N in the western half of the north Atlantic (O’Day and Nafpaktitis, 1967). It is reasonable to suspect a similar situation in the western north Pacific. For instance, Matsubara (1952) reports, under the name L. nitida, a specimen 150 mm (SL) taken by commercial fishermen off Owase, Mie Prefecture, Japan. Lampadena urophaos Paxton, 1963 Fig. 2 Most of the examined specimens are from the southern California off- 8 Contributions in Science No. 138 Figure 2. Lampadena urophaos, 56.0 mm in standard length; San Nicolas basin, California. shore waters, and only a few localities were plotted (Fig. 9). Station data for the Pacific localities are from Berry and Perkins (1966), Paxton (1963), and Pearcy (pers. comm.); coordinates for the latter are 34°49'N, 160°19'W, and 42°00'N, 126°59'W. The following Atlantic specimens were also examined: CAP’N BILL III, sta. RHB 910, 38°24'N, 71°08'W, 12 October 1962, 1330- 1830 hrs, 0-713 m, 64' midwater trawl, one specimen, 73.6 mm; R/V CHAIN, cruise 49, sta. RHB 1111, 21°33'N, 70°12'W, 17 June 1965, 1520- 1940 hrs, 612-668 m, two specimens, 19.0-22.7 mm. D. 15(14-16); A. 14(13); P. 16(15-17); V. 8; gill rakers 4(3-5) + l+8 -9(10), total 13-14(16); PO 5; VO 5(4-6); AOa 5(6); AOp 2(3); Prc 2+1; lateral line scale pockets 37; vert. 36(35), 6 x-rayed specimens. Of 22 specimens examined, the following counts were observed only once: D. 16; P. 17; gill rakers 3+1+9, 5+1 + 10; VO 6; AOa 6; AOp 3. Snout bluntly rounded; mouth large, terminal, its cleft slightly oblique; eye moderate in size, its diameter 3.5 to 4.1 in length of head; posterior opercular margin produced into a point situated above PV02; pterotic spine strong, directed posteriorly and slightly ventrally. Origin of dorsal fin distinctly anterior to origin of ventral fin; origin of anal fin behind vertical through end of base of dorsal fin; base of adipose fin somewhat in advance of vertical through end of base of anal fin; pectoral rays extending about to level of V02; ventral rays almost reaching origin of anal fin. Vn small, posterior and ventral to nasal organ; anterior margin of orbit, from below Vn to above nasal organ, lined with black tissue. Opx small and difficult to discern, situated about on same level as and close behind posterior end of maxillary; Op2 larger than body photophores, located about at level of ventral margin of orbit, slightly behind vertical through Op*. PLO one to three photophore diameters below lateral line and anterior to PVOo. PVOo slightly anterior to upper half of base of pectoral fin; PVOx about on level of posterior end of mouth, slightly anterior to or under PV02 in 1968 Review of Lanternfish LAMP ADEN A 9 Pacific specimens and under or slightly posterior to PV02 in Atlantic speci- mens. First PO interspace about twice as wide as others; P02 to P05 evenly spaced or P04-P05 interspace slightly larger; PC+ to P03 in a horizontal line, with P04 and POs diverging posteriorly; P04 somewhat closer to level of P05 than to that of P03; P05 slightly anteromesad to base of outermost ray of ventral fin. VLO distinctly behind vertical through base of outermost ray of ventral fin, closer to lateral line than to base of ventral fin. V04 slightly postero- mesad to base of innermost ray of ventral fin; rest of VO somewhat higher, the entire series forming a gentle arc, with the organs evenly spaced or with V02-V03 interspace largest, particularly if only four VO present. Last VO, SA04 and SAOo evenly spaced and in a straight line; SA03 over or slightly posterior to SAOo, about under base of last dorsal ray and immediately below lateral line; SA02-SA03 interspace about twice as wide as others in series. AOa in a straight line, evenly spaced or noticeably grouped (2+1+2); last AOa directly over or slightly anterior to base of last anal ray. Pol behind last AOa, immediately below lateral line. AOpT over anterior margin of infra- caudal gland, separated from AOp2 by 1 to 1.5 photophore diameters (in the one specimen with three AOp, the third is half the size of other body photo- phores). Prc4 over second or third ventral procurrent spine, one to two photo- phore diameters anterior to Prc2; Prc3 at base of middle rays of caudal fin and slightly above level of lateral line. In small specimens, supra- and infracaudal luminous glands are about equal in length; in all large Pacific specimens (over 50 mm), the supracaudal gland is somewhat longer, while they are of equal length in the large Atlantic specimen; distance from base of last anal ray to anterior margin of infracaudal gland two-thirds to one times the length of infracaudal gland. Premaxillary and dentary with wide band of conical, closely-set teeth; entire inner row of dentary composed of larger, more widely-spaced conical teeth; three to five large, recurved teeth directed anteriorly at posterior end of dentary; inner row of premaxillary teeth only slightly enlarged; dentigerous portion of premaxillary, near symphysis, somewhat widened to form an ap- proximately triangular area, with an upper (outer) row of three to five large, recurved teeth directed posteroventrally. A narrow band of small teeth along length of palatine. A small patch of teeth on distal ends of vomer. Very small mesopterygoid teeth in an elongated, oval-shaped patch; posterolateral margin of patch with row of six to ten moderate-size teeth. Neural arches of anterior vertebrae not greatly expanded. A number of errors were made in the original description (Paxton, 1963) , which are corrected in the above description. The antorbital organ, which is difficult to discern in adult specimens and is obscured by the black tissue at the anterior portion of the orbit, is definitely ventral to the nasal organ. Thus, L. urophaos agrees with all of its congeners in the presence of a small Vn. The ventral ray count for the species is eight, plus a small accessory splint on the 10 Contributions in Science No. 138 base of the first ray. With very few exceptions (notably species in the genera Gonichthys and Notolychnus) , eight ventral rays plus an accessory splint is constant for the family Myctophidae. The P04 and P05, while slightly raised out of the line of P04-P03, are not as highly elevated as is the P04 in L. lumi- nosa. L. urophaos can therefore be referred to the subgenus Lampadena, as defined by Fraser-Brunner (1949). The status of the subgenus Lychnophora will be discussed after the species descriptions. Lampadena urophaos occurs between 25°N and 42°N in the central and eastern Pacific and between 21 °N and 38°N in the western Atlantic (Fig. 9). The minor differences noted between the Pacific and Atlantic populations, namely the position of the PVOj and the relative sizes of the caudal glands, may be expected in widely allopatric populations. A more complete comparison of the populations must await the collection of additional large Atlantic specimens. Lampadena speculigera Goode and Bean, 1896 Fig. 3 USNS ELTANIN, cruise 15, sta. 1402, 39°15'S, 179°35'W, 30 November to 1 December 1964, 2353-0600 hrs, depth sampled 0-2489 m, bottom depth 2946-3514 m, one specimen, 20.0 mm. USNS ELTANIN, cruise 23, sta. 1710, 41°44'S, 178°18'W, 25 May 1966, 1658-1956 hrs, depth sampled 0-900 m, bottom depth 2641 m, one specimen, 66.0 mm (otolith photographed, Fig. 10). USNS ELTANIN, cruise 26, sta. 1820, 40°22'S, 168°25'E, 2 December 1966, 2000-2319 hrs, depth sampled 0-750 m, bottom depth 920-1028 m, one specimen 109.0 mm. R/V ANTON BRUUN, cruise III, sta. 157, 32°11'S, 59°30'E, 8 Sep- tember 1963, 1510-1835 hrs, depth sampled 150-750 m, bottom depth 4389 m, four specimens, 21.0-24.0 mm. R/V ANTON BRUUN, cruise VI, sta. 352B, 34°14'S, 64°58'E, 30 June 1964, 1420-2000 hrs, depth sampled 350-750 m, bottom depth 2700 m, two specimens, 20.0-20.5 mm. R/V ANTON BRUUN, cruise XIII, coll. 54, 33°42'S, 73°35'W, 2 Feb- ruary 1966, 0320-0920 hrs, depth sampled 0-500 m, one specimen, 70.5 mm. R/V CAP’N BILL III, sta. RHB 913, 39°26'N, 71°00'W, 13 October 1962, 1020-1505 hrs, depth sampled 0-715 m, 64' midwater trawl, one speci- men, 128.0 mm; sta. RHB 914, 39°32'N, 71°00'W, 13 October 1962, 1545- 2025 hrs, depth sampled 0-715 m, 64' midwater trawl, one specimen, 130.0 mm; sta. RHB 915, 39°36'N, 71°12'W, 13 October 1962, 2105-0025 hrs, depth sampled 0-550 m, 64' midwater trawl, one specimen, 126.0 mm (otolith photo- graphed, Fig. 4). R/V ATLANTIS II, cruise 13, sta. RHB 1023, 43°16'N, 45°03'W, 10 1968 Review of Lanternfish LAMP ADEN A 11 Figure 3. Lampadena speculigera, 66.0 mm in standard length; 41° 44' S, 178° 18' W. September 1964, 0855-1230 hrs, depth sampled 0-700 m, two specimens, 33.0 mm; sta. RHB 1006, 41°16'N, 57°37'W, 4 September 1964, 1947-2330 hrs, depth sampled 0-85 m, four specimens, 29.0-36.0 mm; sta. RHB 1003, 41° 36'N, 60°30'W, 3 September 1964, 2021-2345 hrs, depth sampled 0-60 m, one specimen, 28.0 mm. D. 14(13-15); A. 14(15); P. 14(15); V. 8; gill rakers 6-7+1 + 12-14, total 19-22; PO 5(6); VO 5(4-6); AOa 6-7; AOp 3-4(5); Prc 2+1; lateral line scales 39-41; vert. 38-39(40), 5 x-rayed specimens. Of 15 specimens ex- amined, the following counts were found only once: D. 13; PO 6; VO 4; AOp 5. Snout bluntly rounded; mouth large, barely subterminal, its cleft very slightly oblique; eye large, its diameter 2.7 to 3.2 in length of head; opercular margin not pointed, but with a shallow indentation about on level of upper end of pectoral base; pterotic spine strong, straight, directed posterolaterally. Origin of dorsal fin on or slightly behind vertical through base of outer- most ray of ventral fin; origin of anal fin behind vertical through end of base of dorsal fin; base of adipose fin over end of base of anal fin; pectoral fin extending to about level of first VO; ventral fin reaching origin of anal fin. A small Vn, immediately posteroventrad to nasal apparatus and directed ventrally; anterior third of orbital margin lined with black tissue. Opx small, immediately behind lower preopercular margin and opposite posterior end of mouth; Op2 markedly larger than general body photophores, at least one times its own diameter above level of ventral margin of orbit and about 2.5 times its own diameter above and slightly behind Op!. PLO distinctly in advance of vertical through base of uppermost ray of pectoral fin and about one photophore diameter below lateral line; PVOx somewhat above level of posterior end of maxillary, slightly behind vertical through PV02, which is situated in front of lower half of base of pectoral fin and on a straight line connecting PLO with PVOi. POj-POo interspace about 12 Contributions in Science No. 138 twice as wide as PO4-PO5 interspace, which is equal to or slightly wider than rest; POj-POs in a horizontal line; P05 situated about its own diameter in front of, and slightly mesad to, base of outermost ray of ventral fin; P04 much closer to horizontal level of P03 than to that of P05. VLO distinctly behind vertical through base of outermost ray of ventral fin and somewhat closer to lateral line than to base of ventral fin; VO more or less level and evenly spaced, sometimes last VO slightly raised; VOj posteromesad to base of innermost ray of ventral fin. SAO forming a wide angle; SAOx above and behind last VO; SA02 above and behind SA04; SA02-SA03 interspace about 2.5 times as wide as SA04-SA02 interspace, which is equal to that of VOs-SAOj; SA03 slightly in advance of vertical through SA02 and its own diameter, or less, below lateral line. AO small, especially in young specimens; AOa level, evenly spaced, about one photophore diameter (one to two diameters in fishes less than 30 mm) apart from each other. Pol behind last AOa and about its own diameter below lateral line; AOp evenly spaced, on a straight, gently ascending line; AOp4 and usually AOp2 in front of anterior end of infracaudal luminous gland. Prc4 and Prc2 horizontal or Prc2 slightly raised, the two photophores being only one organ diameter apart from each other; Prc3 at base of middle rays of caudal fin and very slightly above level of lateral line. Figure 4. Lampadena speculigera, medial view of left otolith, 4.9 mm long, 3.5 mm high, belonging to an 126 mm specimen from the north Atlantic. 1968 Review of Lanternfish LAMP ADEN A 13 Length of supracaudal gland one-half to two-thirds that of infracaudal; posterior margin of both glands at about same vertical; posterior margin of supracaudal gland distinctly emarginate; distance between posterior end of base of anal fin and anterior margin of infracaudal luminous gland 1 to 1.5 times as great as length of this gland. Premaxillary and especially dentary with bands of small, needlelike teeth; an inner series of 10 to 15 enlarged, broad-based, recurved teeth directed anteriorly on posterior dentary; two to five large, anteriorly directed, recurved teeth on inner margin of dentary close to symphysis (more prominent in large specimens, 80 mm or more) ; anterior end of premaxillary broadened, forming a roughly triangular dentigerous area outside the mouth; teeth of outer row of this area enlarged, recurved, directed posteroventrally. Palatine with long, narrow band of small teeth. Few minute teeth on each distal end of vomer. Mesopterygoid with large, roughly oval, patch of uniformly small teeth. Neural arches of anterior vertebrae greatly expanded to form an almost closed tube above vertebrae. Two trends can be distinguished between specimens from the two hemi- spheres. These involve AO and gill-raker counts. Individuals from the southern oceans tend to have 6+4 AO and 7+1 + 13-14 gill rakers, whereas in those from the north Atlantic, 7+3 AO and 6+1 + 12-13 gill rakers are the most frequent counts. In addition, John Fitch called our attention to the fact that the otoliths from a specimen 126 mm long taken in the north Atlantic (Fig. 4), were distinctly different from those of specimens from the southern hemisphere (Fig. 10 (4)). Some changes in size and shape, depending on the size of the specimens and the length of time of preservation, are to be expected. However, the differences in the present case are strong enough to warrant an examination of larger material, especially from the north Atlantic. Until this is done we prefer to consider the two populations as conspecific. Our material from the southern hemisphere indicates that Lampadena speculigera is a subtropical-temperate form, ranging, in the Indian and Pacific Oceans, between the latitudes of approximately 30°S and 45 °S. It is considered relatively common in the north Atlantic (Bolin, 1959). On the basis of material from the collections of the Woods Hole Oceanographic Institution, this species appears to range between the latitudes of approximately 35°N and 45 °N in the north Atlantic (Fig. 9). Further study of large collections is necessary before its apparent absence from the south Atlantic can be ascertained. Lampadena notialis, new species Fig. 5 Holotype : LACM 11321-1; immature specimen, 66.3 mm, USNS ELTANIN, cruise 26, sta. 1830, 42°00'S, 160°11'E to 42°08'S, 160°05'E, 8 14 Contributions in Science No. 138 December 1966, 0839-1140 hrs, depth sampled 0-800 m, bottom depth 4895 m (otolith photographed, Fig. 10). Paratypes: LACM 11331-1; a male specimen, 105.0 mm, USNS EL- TANIN, cruise 26, sta. 1841, 47°20'S, 161°54'E to 47°28'S, 161°52'E, 12-13 December 1966, 2334-0221 hrs, depth sampled 0-800 m, bottom depth 4648 m. MCZ 45892; one specimen, 25.2 mm; R/V ANTON BRUUN, cruise VI, sta. 354A, 40°48'S, 65°03'E, 4 July 1964, 0915-1510 hrs, depth sampled 0-1650 m, bottom depth 4600 m. D. 14; A. 14; P. 14; V. 8; gill rakers 7-8 + 1 + 16-17, total 24-26; PO 5; VO 5(6); AOa 6; AOp 3; Prc 2+1; lateral line scales 38-39; vert. 37-38, 2 x-rayed specimens. A relatively large myctophid fish; body rather robust, gently tapering from vertical through base of pectoral fin to a deep caudal peduncle; dorsal and ventral contours anterior to pectoral fin evenly curved; snout rather high and bluntly rounded; mouth large, slightly subterminal, its cleft moderately oblique; eye large, its diameter 2.8 to 3.0 in length of head; posterior opercular margin produced into two rounded lobes separated by a triangular indentation located at about level of PV02; pterotic spine directed posterolaterally in younger specimen (holotype) and posteroventrally in mature specimen (paratype). Origin of dorsal fin distinctly in advance of vertical through base of outer- most ray of ventral fin; origin of anal fin behind vertical through end of base of dorsal fin; base of adipose fin about over end of base of anal fin; rays of pectoral fin extending slightly beyond base of ventral fin; rays of ventral fin reaching first or second ray of anal fin. Vn well developed, immediately posteroventral to nasal apparatus, round- ish, directed ventrally and framed dorsally and laterally by darkly pigmented tissue; a narrow strand of black tissue along entire anterior margin of orbit, Figure 5. Lampadena notialis, new species, holotype, 66.3 mm in standard length; 42° 00' S, 160° 1 1' E; LACM 11321-1. 1968 Review of Lanternfish LAMPADENA 15 extending ventrally and posteriorly up to about vertical through center of pupil; suborbital region with numerous pores, each with a raised, darkly- pigmented ridge. Op2 small, immediately behind preopercular margin and directly opposite or slightly below level of posterior end of maxillary, often masked by whitish tissue of adjacent neuromasts; Op2 larger than general body photophores, at about level of ventral margin of orbit, slightly posterior to vertical through Op! and separated from latter by a distance about twice as large as its own diameter. PLO slightly in advance of vertical through upper end of base of pectoral fin and 1.5 to 2 times its own diameter below lateral line. PVOj at about level of posterior end of mouth, directly under or slightly posterior to PV02, which is about its own diameter in front of ventral half of base of pectoral fin. PO series forming gently diverging lines; first PO interspace 2 to 2.5 times as wide as spaces between rest of organs of same series; POs about its own diameter from, and slightly mesad to, base of outermost ray of ventral fin. VLO slightly behind vertical through base of outermost ray of ventral fin and somewhat closer to lateral line than to base of ventral fin. LACM paratype with five VO distinctly divided into two groups (2+3) on the left side and six evenly spaced VO on the right side; holotype and smaller paratype with five evenly spaced VO on both sides, the organs forming a gentle arch; VOx posteromesad to base of innermost ray of ventral fin. SAO slightly angular; SAOx about its own diameter behind and above last VO; SA02 (not developed on left side of larger paratype) about its own diameter above, but only slightly behind SAOj; SA03 directly over or slightly behind SA02, its own diameter or less below lateral line and separated from SA02 by a distance 2 to 3 times as wide as that between SAOx and SAOo. AOa evenly spaced, interspaces about 1.5 times the diameter of an organ of same series (less than the diameter of a photophore in fishes less than 30 mm) ; all but last AOa level, latter distinctly raised, its ventral margin touching tangent to dorsal border of preceding organ. Pol behind last AOa and less than its own diameter below lateral line; all three AOp on anterior, ascending dorsal contour of infracaudal luminous gland; AOp interspaces equal to one diameter of an organ of same series, or less; Prc1-Prc2 interspace equal to about half the diameter of an organ of same series; Prc2 slightly raised; Prc3 at base of middle rays of caudal fin and slightly above level of lateral line. Supra- and infracaudal luminous glands very well developed; length of supracaudal gland about three-fourths that of infracaudal gland; length of infracaudal gland equal to (in small specimens) or greater than (in large specimens) diameter of eye; distance between posterior end of base of anal fin and anterior end of infracaudal gland equal to about one-fourth of length of that gland. Premaxillary and dentary studded with small, villiform teeth; an inner series of eight to nine large, recurved forward, very broad-based teeth at 16 Contributions in Science No. 138 posterior end of dentary (particularly large in medium-sized fish — i.e., size of holotype) ; an inner series of widely spaced, large teeth hooked forward along anterior end of dentary (best developed in large individuals); anterior denti- gerous ends of premaxillaries broadened near symphysis, forming a triangular area with an outer row of enlarged teeth recurved anteroventrally. Palatine teeth very small, papilliform, arranged in a long and narrow band. A small patch of minute teeth on distal end of each limb of vomer. A large, roughly oval patch of somewhat larger teeth on each mesopterygoid. Neural arches of anterior vertebrae not greatly expanded. The specific name comes from the Greek notia, which means southern, in reference to the occurrence of the species in high southern latitudes (Fig. 8) . Lampadena dea Fraser-Brunner, 1949 Fig. 6 R/V ANTON BRUUN, cruise III, sta. 156, 28°54'S, 60°01'E, 6-7 Sep- tember 1963, 2155-0240 hrs, depth sampled 0-275 m, bottom depth 4114 m, one specimen, 23.0 mm; sta. 156, 29°13'S, 60°05'E, 7 September 1963, 0245- 0620 hrs, depth sampled 0-150 m, bottom depth 4114 m, two specimens, 21.0 mm. R/V ANTON BRUUN, cruise VI, sta. 352B, 34°14'S, 64°58'E, 30 June 1964, 1420-2000 hrs, depth sampled 350-750 m, bottom depth 2700 m, one specimen, 37.0 mm; sta. 353A, 37°59'S, 64°56'E, 2 July 1964, 1115-1925 hrs, depth sampled 350-2390 m, bottom depth 4400-4600 m, one specimen, 63.5 mm. R/V ANTON BRUUN, cruise XIII, coll. 22, 33°51'S, 87°49'W, 18 January 1966, 0013-0530 hrs, depth sampled 0-375 m, bottom depth 3550- 3900 m, one specimen, 45.5 mm; coll. 28, 30°45'S, 92°34'W, 22 January 1966, 0045-0650 hrs, depth sampled 0-320 m, bottom depth 3450-3710 m, two specimens, 35.0-39.5 mm; coll. 30, 31°07'S, 89°29'W, 24 January 1966, 0025- 0350 hrs, depth sampled 0-410 m, bottom depth 3550-3950, two specimens, 27.0-40.0 mm. USNS ELTANIN, cruise 24, sta. 1739, 40°15'S, 144°45'W, 26 July 1966, 0204-0530 hrs, depth sampled 0-1125 m, bottom depth 5325 m, one specimen, 59.0 mm (otolith photographed, Fig. 10). D. 14; A. 14(15); P. 14(15); V. 8; gill rakers 6(7) + l + 14(15), total 21(22-23); PO 5; VO 5(4-6); AOa 6-7(5); AOp 4-5(3); Prc 2+1; lateral line scales 37-38; vert. 38, 3 x-rayed specimens. Of ten specimens examined, the following counts occurred only once: A. 15; gill rakers 7+1 + 14, 7+1 + 15; VO 4, 6; AOp 3. The 11 specimens in the collection yielded 22 AO counts, as follows: 5+4(4), 6+3(1), 6+4(6), 6+5(6), 7+4(4), and 7+5(1). Snout bluntly rounded; mouth large, slightly subterminal, its cleft some- 1968 Review of Lanternfish LAMPADENA 17 what oblique; eye large, its diameter 2.9 to 3.3 in length of head; opercular margin with a very distinct triangular indentation at about level of upper half of base of pectoral fin; pterotic spine very strong, curved downward and for- ward in larger specimens, posterolaterally in smaller specimens. Origin of dorsal fin over or slightly in advance of vertical through base of outermost ray of ventral fin; origin of anal fin behind vertical through end of base of dorsal fin; base of adipose fin somewhat behind vertical through end of base of anal fin; pectoral fin extending to about level of second VO; ventral fin reaching base of second or third anal ray. Vn present, posteroventral to nasal apparatus, directed ventrally; anterior third of orbital margin lined with black tissue. A small Opl9 difficult to see, close to and directly behind posterior end of maxillary; Op2 distinctly larger than general body photophores, about on level of ventral margin of orbit and slightly behind vertical through Op4. PLO slightly in advance of vertical through upper end of base of pectoral fin and about its own diameter below lateral line. PVOx at about level of posterior end of mouth, directly under or slightly in advance of vertical through PV02, which is found in front of middle of base of pectoral fin and slightly behind straight line connecting PLO with PV04. PO series forming posteriorly diverging lines, with P05 situated about its own diameter in front of, and slightly mesad to, base of outermost ray of ventral fin; first PO interspace about 1.5 times as wide as others; P04 closer to horizontal level of P03 than to that of P05. VLO directly over or slightly in advance of vertical through base of outermost ray of ventral fin and markedly closer to lateral line than to base of ventral fin. VO equally spaced or distinctly grouped (2+3), the entire series on a gently arched line; VOx posteromesad to base of innermost ray of ventral fin. SAO forming a wide angle; SAOj above and behind last VO; SA02 above and behind SA04; SA02-SA03 interspace 2 to 2.5 times as wide as SA04 SAOo interspace, which is equal to that of V05-SA0i; SA03 directly over, 18 Contributions in Science No. 138 slightly anterior, or immediately posterior to SA02, and 0.5 to 1 time its own diameter below lateral line. AOa level and usually evenly spaced. Pol behind vertical through last AOa and less than one times its own diameter below lateral line. AOp evenly or irregularly spaced, but all organs over anterior two-thirds of infracaudal gland. Prcx anterior to first ventral procurrent spine; Prcx and Prc2 on same horizontal level or Prc2 slightly raised, the two organs separated by a distance equal to at least three photophore diameters; Prc3 at base of middle rays of caudal fin, and slightly above level of lateral line. Supra- and infracaudal luminous glands well developed; length of supra- caudal gland about two-thirds length of infracaudal gland; posterior margin of both glands on same vertical; posterior margin of supracaudal gland slightly or distinctly emarginate; distance between anterior end of infracaudal gland and posterior end of base of anal fin less than one-fourth the length of infracaudal gland. Premaxillary and dentary with a band of small, conical teeth; inner series of 6-8 large, broad-based, widely-spaced, recurved teeth directed anteriorly on posterior third of dentary; two or three larger inner teeth on anterior dentary near symphysis, but these are not recurved; anterior dentigerous portion of premaxillary near symphysis slightly widened, with curved teeth. Narrow band of small teeth on palatine. No teeth evident on vomer. Elongated oval patch of small, uniformly-sized teeth on mesopterygoid. Neural arches of anterior vertebrae greatly expanded to form an almost closed tube above centra. Lampadena dea occurs in the southern parts of all three oceans, between the latitudes of approximately 20°S and 50°S (Fig. 9). Lampadena chavesi Collett, 1905 Fig. 7 R/V ANTON BRUUN, cruise VI, sta. 35 ID, 31°45'S, 65°08'E, 29 June 1964, 0359-1507 hrs, depth sampled 350-1786 m, bottom depth 4480 m, one specimen, 54.5 mm. R/V ANTON BRUUN, cruise XIII, coll. 26, 31°16'S, 92°28'W, 20-21 January 1966, 2014-0138 hrs, depth sampled 0-380 m, bottom depth 3300- 3700 m, one specimen, 60.0 mm; coll. 30, 31°07'S, 89°29'W, 24 January 1966, 0025-0350 hrs, depth sampled 0-410 m, bottom depth 3550-3950 m, two specimens, 36.0-38.0 mm (otolith of 36.0 mm specimen photographed, Fig. 10). R/V ATLANTIS II, cruise 13, sta. RHB 1014, 41°34'N, 52°15'W, 7 September 1964, 0115-0500 hrs, depth sampled 0-85 m, one specimen, 19.0 mm; sta. RHB 1022, 42°35'N, 45°56'W, 10 September 1964, 0135-0505 hrs, depth sampled 0-50 m, two specimens, 22.0-24.0 mm; sta. RHB 1044, 39°37' N, 31°10'W, 26 September 1964, 0050-0535 hrs, depth sampled 0-475 m, one 1968 Review of Lanternfish LAMP ADEN A 19 Figure 7. Lcimpadena chavesi , 54.5 mm in standard length; 31° 45' S, 65° 08' E. specimen, 35.0 mm; sta. RHB 1047, 39°25'N, 36°56'W, 27 September 1964, 2030-2250 hrs, depth sampled 0-52 m, one specimen, 21.0 mm. R/V CHAIN, cruise 49, sta. RHB 1127, 31°28'N, 70°25'W, 22 June 1965, 0035-0330 hrs, depth sampled 0-83 m, two specimens, 22.0-25.0 mm; sta. RHB 1129, 34°12'N, 70°21'W, 22 June 1965, 2145-2355 hrs, depth sampled 0-80 m, one specimen, 27.0 mm. D. 14; A. 13-14(12); P. 16-17; V. 8; gill rakers 6-7+1 + 13, total 20-21; PO 5; VO 5(6); AOa 7(8); AOp 2; Prc 2+1; lateral line scales 38(39); vert. 37-38, 2 x-rayed specimens. Snout rounded; mouth large and terminal, its cleft very slightly oblique; maxillary somewhat expanded posteriorly; eye large, its diameter 2.6 to 3.3 times in length of head; opercular margin with slight indentation opposite level of middle of pectoral base; pterotic spine strong, directed posteriorly. Origin of dorsal fin over or slightly behind vertical through base of outer- most ray of ventral fin; origin of anal fin behind vertical through end of base of dorsal fin; base of adipose fin markedly behind vertical through end of base of anal fin; pectoral fin reaching base of ventral fin; ventral fin extending to origin of anal fin. A small, roundish Vn immediately posteroventrad to nasal apparatus; anterior third of orbital margin lined with black tissue; a conspicuous, crescent- shaped strip of whitish (luminous?) tissue on iris dorsal to pupil, which is somewhat elliptical; distinct aphakic space anteroventrad to lens. Op! small, immediately behind preopercular margin and about opposite posterior end of mouth; Op2 distinctly larger than general body photophores, at about level of ventral margin of pupil, separated from Opj by a distance about three times as wide as its own diameter. PLO slightly in advance of vertical through base of uppermost ray of pectoral fin and 1 to 1.5 times its own diameter below lateral line. PVOi at about level of posterior end of mouth, slightly posterior to vertical through 20 Contributions in Science No. 138 re 8. Chart showing stations which yielded material examined in the present study (the type iity of L. chavesi was added from the literature) . 1968 Review of Lanternfish LAMP ADEN A 21 Figure 9. Chart showing stations which yielded material examined in the present study (the type localities of L. dea and L. speculigera were added from the literature) . 22 Contributions in Science No. 138 PV02, which is situated in front of middle of base of pectoral fin and on a straight line connecting PLO with PVOi. P04 and P02 on a straight, hori- zontal line; P02 through POs in straight, diverging lines, with POs situated about 1.5 times its own diameter in front of and slightly mesad to base of outermost ray of ventral fin; first PO interspace 1.5 to 2 times as wide as last, which is somewhat wider than P02-P03 and P03-P04 interspaces; VLO over base of outermost ray of ventral fin and midway between latter and lateral line. VO evenly spaced or distinctly grouped (2+3) forming a markedly arched line; VOj directly mesad, or slightly anterior, to base of innermost ray of ventral fin. SAO series very slightly angular; SAOx above and behind V05; SA02 above and behind SA04; SA03 directly over, slightly anterior, or imme- diately posterior to vertical through center of SAOo and about its own diameter below lateral line; distance between SA03 and SA02 about twice as wide as distance between SA02 and SA01? the latter being equal to that between SA04 and VOs. First and second AOa somewhat depressed; space between AOa2 and AOa3 usually distinctly wider than other AOa interspaces; last two to three AOa entirely behind base of anal fin; at least one and usually two AOa over infracaudal luminous gland. Pol over or slightly posterior to last AOa and about its own diameter below lateral line. AOp widely spaced, about over middle of, and in contact with, infracaudal gland. Prc4 over posterior end of infracaudal gland; Prc4 and Prc2 horizontally arranged and very widely spaced, distance between them equal to, or somewhat smaller than, distance between Prc2 and Prc3, which is posterior to Prc2 and slightly above level of lateral line. Caudal luminous gland largest in genus, their posterior borders on same vertical; supracaudal gland bifurcating posteriorly; infracaudal gland flat in cross section, limited to ventral area of caudal peduncle and not conspicuous laterally, tapering posteriorly rather than anteriorly (in contrast to all other species of the genus); distance between end of base of anal fin and anterior margin of infracaudal gland very short, about twice the diameter of a photo- phore. Premaxillary and dentary with band of small, needlelike teeth; six to eight large, recurved, broad-based teeth directed anteriorly on posterior part of dentary; three to four inner, moderately large teeth hooked forward on anterior portion of dentary near symphysis; premaxillary slightly broadened near symphysis, with outer row of large, recurved teeth. Palatine with long, narrow band of minute teeth. A small patch of minute teeth on each limb of vomer. Large, oval patch of similar teeth on mesopterygoid. Neural arches of anterior vertebrae greatly expanded to form an almost closed tube above centra. Bolin (1959) states that “Lampadena chavesi appears to occur through- out the Atlantic between the latitudes of approximately 38 °N and 33 °S.” Our data suggest that this species occurs in the North Atlantic between the latitudes of about 30°N and 45 °N and in both the Indian and Pacific Oceans between 1968 Review of Lanternfish LAMP ADEN A 23 the latitudes of approximately 30°S and 40°S (Fig. 8). An antitropical distri- bution is indicated. Lampadena anomala Parr, 1928 R/V PAWNEE, sta. 58B, 32°24'N, 64°29'W, 20 March 1927, depth sampled 0-ca. 2000 m, 14-foot Ring Net, one specimen (holotype), ca. 48 mm, BOC 2272. R/V CHAIN, cruise 60, sta. RHB 1260, 13°12'N, 72°47'W, 28 May 1966, 1050-1400 hrs, depth sampled 0-890 m, Marinovich Trawl, one speci- men, about 48 mm (otolith photographed, Fig. 10). Parr (1928) described this species on the basis of a single 48 mm speci- men. Because of the poor condition of the fish, Parr was unable to take ac- curate measurements. Unfortunately, the only specimen from the Woods Hole collection is in equally poor, if not poorer, condition. However, direct com- parison with Parr’s type left us with no doubt that it belongs to L. anomala. The following description is based on the characters which we have been able to positively locate, identify, and count in the Woods Hole specimen. D. 16; A. 13; P. 16; V. 8; gill rakers 5+1 + 11; VO 3; SAO 3; AOa 3 on left side, 4 on right; AOp 2; Prc 2+1; vert. 36. Photophores small, more so than in any other species of the genus; two Op (Parr’s “. . . a series of six small organs . . .”, mistaken for photophores, are nothing but a series of neuromasts immediately behind the preopercular margin, found in all myctophids); PLO in advance of vertical through upper end of base of pectoral fin, about twice its own diameter below lateral line. PVOi slightly behind vertical through center of PV02, which is immediately in front of middle of base of pectoral fin. VLO slightly nearer lateral line than base of ventral fin; three VO widely spaced. SA03 about over origin of anal fin and about its own diameter below lateral line. AOa level, widely spaced; last AOa well in advance of end of base of anal fin. Pol behind end of base of anal fin, about twice its own diameter below lateral line. AOpt in advance of anterior margin of infracaudal luminous gland; AOp2 over same gland; AOpj- AOp2 interspace about twice the diameter of an organ; Prcx and Prc2 hori- zontal and about one organ diameter apart from each other; Prc3 near base of middle rays of caudal fin and slightly above level of lateral line. Supracaudal luminous gland slightly shorter than infracaudal, the length of the latter being equal to distance between end of base of anal fin and anterior margin of infracaudal gland. Premaxillary and dentary with band of small, needlelike teeth, the inner series distinctly enlarged; five broad-based, strongly recurved teeth directed anteriorly on posterior part of dentary; no evidence of enlarged teeth on anterior part of dentary or premaxillary near symphysis. Palatine with what appears to be a single, long row of relatively large, sharp teeth. One or two Figure 10. Medial views of left otoliths, anterior end to the right. (1) Lampadena luminosa, otolith 5.4 mm long, specimen 71 mm; (2) L. urophaos, otolith 7.6 mm long, specimen probably about 80 mm; (3) L. species A, otolith 9.2 mm long; (4) L. speculigera, otolith 3.8 mm long, specimen 66 mm; (5) L. notialis, otolith 3.9 mm long, specimen 66.3 mm; (6) L. dea, otolith 3.4 mm long, specimen 59 mm; (7) L. chavesi, otolith 1.7 mm long, specimen 36 mm; (8) L. anomala , otolith 1.8 mm long, specimen about 48 mm; (9) Taaningichthys bathyphilus, otolith 1.9 mm long, specimen probably about 50 mm. teeth at each distal end of vomer. A large, roughly oval-shaped patch of similar teeth on mesopterygoid, with posterior ones somewhat enlarged. Neural arches of anterior vertebrae not greatly expanded. Inadequate data preclude any statement concerning the horizontal dis- tribution of L. anomala (Fig. 9). However, this fish appears to occur at greater depths than the rest of its congeners. 1968 Review of Lanternfish LAMP ADEN A 25 Otoliths Sagittae of the species of the genus Lampadena display three main pat- terns (Fig, 10). L. luminosa and L. urophaos have large otoliths, considerably longer than high, with a length to height ratio of 1.7: 1 to 1.8:1. The postero- dorsal angle is conspicuously notched and the ventral margin is not smooth. A distinct rostrum and small antirostrum are present. The lateral face is slightly rounded and weakly sculptured. The posterior angle of the dorsal margin is pronounced (broken off in the photographed specimen of L. luminosa). The ventral margin of L. urophaos is scalloped or finely serrate, while that of L. luminosa has more distinct spinous processes. The collum divides the sulcus into two unequal sections in L. urophaos, whereas the collum is about in the middle of the sulcus in L. luminosa. L. speculigera, L. notialis, L. dea and L. chavesi have more oval-shaped otoliths, with length to height ratios varying between 1.3:1 and 1.5:1. The posterodorsal notch is slight or absent, the collum is about in the middle of the sulcus, the ventral margin has small but distinct spines, and the lateral face is slightly rounded and weakly sculptured. A distinct rostrum and antirostrum are present in L. dea and L. chavesi, while these structures are less conspicuous in L. speculigera and L. notialis. The posterodorsal angle is slightly indented in L. chavesi and essentially unnotched in the other three species. The anterior region of the dorsal margin is dis- tinctly scalloped in L. notialis, while that of L. speculigera is almost smooth. L. anomala has a small otolith with a length to height ratio of 1.2: 1, no postero- dorsal notch, a smooth ventral margin, a greatly developed rostrum, and a distinctly rounded lateral surface. The otolith of Taaningichthys bathyphilus is small and nearly round, with a length to height ratio of 1.0:1, no postero- dorsal notch, a smooth ventral margin, a slight to moderate rostrum, and a rounded lateral surface. Although otoliths from fishes of comparable sizes were not available, a difference in relative sizes is suggested. The ratios of standard length of specimen to otolith length are as follows: L. luminosa — -13.1:1; L. notialis— 16.9: 1; L. speculigera and L. dea- — 17.4:1; L. chavesi — 20.6:1; L. anomala - — 26.7:1. Standard lengths for the other species are not available; however, L. urophaos, like L. luminosa, has a large otolith, whereas Taaning- ichthys bathyphilus, like L. chavesi, has a small one. Discussion Fraser-Brunner (1949) erected the subgenus Lychnophora to include the forms L. luminosa and L. nitida. The distinguishing characters of the subgenus were given as follows : P03 (=P04) much elevated (versus PO all on same level) ; diameter of eye more than four times in length of head (versus eye diameter less than four times in head); and inner row of pterygoid teeth conspicuously enlarged (versus pterygoid teeth uniformly small). The present study has revealed that the latter two characters are not diagnostic for the Table 1 Measurements as per cent of standard length for the species of Lampadena. 26 Contributions in Science No. 138 puB$ -odns pUBl§ ’OJUI p2UB3ld ipdap ’P9cl -pnBD qjSuaj 'pad -pnBD (N CN CO 00 i-h on 4 © »n on ^ vd oo in r- x> ci H; q •"t rH GO vq as 00 in o cn 00 i-h rH rH i— H VO 1 q cA cA 1—H vd vd VO o o 1—1 d o 00 00 CN CN On CN cA in CN t-H CN CN CN O 1— H ON H- H- m L. chavesi 5 34.0- 1968 Review of Lanternfish LAMP ADEN A 27 subgenus. The eye diameter in head length for L. luminosa varies from 3.7 to 4.3 times, for L. urophaos from 3.5 to 4.1 times, and for the other species from 2.6 to 3.3 times. The most posterior mesopterygoid teeth are enlarged in L. urophaos and slightly enlarged in L. anomala. The subgenus can be defined on the basis of one character only, the elevated P04. Since L. nitida is herein considered conspecific with L. luminosa, the subgenus is restricted to one form; its retention suggests that L. luminosa stands apart from its congeners. However, L. urophaos shares a number of characters with L. luminosa. In addition to the two characters noted above, each of these species has a very low gill-raker count, total 13-16, versus 17-26 for the other species, and a distinct similarity in the shape and size of the otolith. Moreover, in L. urophaos, the P04, although not as highly elevated as that of L. luminosa, is more elevated than the P04 of the rest of the species. L. chavesi, with extremely long caudal glands, white tissue (possibly luminous) on the iris, and the V04 in advance of or directly mesad to the base of the innermost ray of the ventral fin, stands as far from, if not further from, the main Lampadena stock as L. luminosa. L. anomala, as indicated by its otolith size and shape, development and pattern of photophores, etc., appears to be evolving in a different line. At our present state of knowledge, the degree of relationship between the various species of Lampadena is not clear and the retention of Fraser-Brunner’s subgenera and/or the erection of new ones is, in our opinion, of little value in elucidating evolutionary trends within the genus. Therefore, we do not recognize the subgenus Lychnophora. A tentative interpretation of species relationships is nevertheless possible. L. luminosa and L. urophaos are clearly related. The relatively low gill-raker count, total 17, and the absence of expanded anterior neural arches in L. ano- mala may indicate derivation from a stock ancestral to these two species. L. speculigera has more characters in common with L. notialis than with any other species, while L. dea appears to be intermediate between the latter two species and L. chavesi. L. speculigera, L. dea, and L. chavesi share with the genus Taaningichthys the expanded neural arches of the anterior vertebrae. The presence of white tissue on the iris and the restriction of the infracaudal gland to the ventral portion of the caudal peduncle suggest a possible relation- ship between L. chavesi and Taaningichthys. Certain measurements (Table 1) will distinguish some of the species. Allometric growth is common for most structures, including the caudal glands. Knowledge of the biology of most lanternfishes in general, and Lam- padena in particular, is meager. Many species of the genus are relatively large; specimens of L. speculigera, L. notialis, and L. luminosa attain a standard length of over 100 mm. Most species appear to be among the deepest-dwelling of myctophids; the two known captures of L. anomala with open nets have been made below 750 meters. Shallow captures of large specimens during the night, indicative of extensive vertical migration, are known only for L. lumi- 28 Contributions in Science No. 138 nosa and L. urophaos. Small (20-35 mm) specimens of L. speculigera, L. dea, and L. chavesi have been taken during the night in the upper 200 meters. After the manuscript was completed, John Fitch brought to our attention a large series of otoliths recovered in good condition from the stomach con- tents of a pygmy sperm whale ( Kogia simus) caught off Taiji, Japan. These otoliths (Fig. 10(3)) belong to a species of the genus Lampadena. The new otoliths are large, with posterodorsal angle distinctly notched, ventral margin sculptured, a distinct rostrum and antirostrum, a posteriorly divided sulcus, and with a slightly rounded lateral face. Distinctive differences between these oto- liths and those most similar to them, i.e. L. luminosa and L. urophaos, include stronger development of the antirostrum, strong sculpturing of the lateral face, reduction of the area posterior to the sulcus, and a length to height ratio of 1.3:1 to 1.4:1. We have little doubt that the new otoliths belong to a form as yet un- described. Moreover, we are inclined to believe that this new form, when caught, will prove to be more closely related to L. urophaos and L. luminosa than to any other species. If we may venture a further prediction, the new species will probably have a small number of gill rakers, unexpanded anterior neural arches, and short caudal luminous glands. Acknowledgments We wish to thank our colleagues of the University of Southern California who, on board the USNS ELTANIN and the USC research vessel VELERO IV, helped collect material for the present study. We also wish to thank Giles W. Mead of the Museum of Comparative Zoology (MCZ), Harvard University, for allowing us to examine the type specimen of Lampadena luminosa and to study material collected by the R/V ANTON BRUUN during the International Indian Ocean Expedition. Erik Bertelsen, of the Danish Marine Biological Laboratory, Carlsberg Foundation, and Keith S. Thomson, of Peabody Museum, Yale University (BOC), kindly placed the types of L. nitida and L. anomala, respectively, at our disposal. Thanks to Richard H. Backus (RHB) and James E. Craddock of the Woods Hole Oceanographic Institution, Richard H. Rosenblatt and Robert L. Wisner of the Scripps Institution of Oceanography (SIO), and Robert J. Lavenberg of the Los Angeles County Museum of Natural History (LACM), we were able to examine and compare material from the north Atlantic and the north Pacific. We are indebted to Rolf L. Bolin for having made available to us his personal file containing meticulous notes and careful descriptions of type material examined by him in various museums and collections around the world. We thank John E. Fitch and Jack W. Schott of California Fish and Game for taking the otolith photographs. Robert L. Brownell kindly provided the identification and locality of the pigmy sperm whale. Professor Bolin and Robert Lavenberg have generously given of their time to read and criticize the present work. 1968 Review of Lanternfish LAMP ADEN A 29 Partial financial support by National Science Foundation Research Grants GA 448, under sponsorship of the United States Antarctic Research Program, and GB 6576 are here gratefully acknowledged. Literature Cited Berry, F. H., and H. C. Perkins. 1966. Survey of pelagic fishes of the California Current area. U. S. Fish Wildl. Serv., Fish. Bull., 65: 625-682. Bolin, R. L. 1939. A review of the myctophid fishes of the Pacific coast of the United States and of Lower California. Stan. Ichthyol. Bull., 1: 89-156. 1959. Iniomi. Myctophidae from the “Michael Sars” North Atlantic Deep- Sea Expedition 1910. In Rep. Sci. Res. “Michael Sars” N. Atlantic Deep-Sea Exped. 1910, Bergen, 4, pt. 2(7) : 1-45. Collett, R. 1905. On some fishes from the sea off the Azores. Zoologischer An- zeiger, 28: 723-730. Fraser-Brunner, A. 1949. A classification of the fishes of the family Myctophidae. Proc. Zool. Soc. Lond., 118: 1019-1106. Frizzell, D. L., and J. H. Dante. 1965. Otoliths of some early Cenozoic fishes of the Gulf Coast. J. Paleontol., 39: 687-718. Garman, S. 1899. The fishes. Mem. Mus. Comp. Zool., Harvard Univ., 24: 1-431, 97 pis. Goode, G. B., and T. H. Bean. 1896. Oceanic ichthyology. U. S. Nat. Mus., Spec. Bull., 553 p., 123 pis. Matsubara, K. 1952. The capture of the Atlantic lantern-fish, Lampadena nitida (Taaning) in Japan. Japan Jour. Ichthyol., 2:11 1-112. O’Day, W. T., and B. Nafpaktitis. 1967. A study of the effects of expatriation on the gonads of two myctophid fishes in the North Atlantic Ocean. Bull. Mus. Comp. Zool., Harvard Univ., 136: 77-90. Parr, A. E. 1928. Deepsea fishes of the order Iniomi from the waters around the Bahama and Bermuda Islands. Bull. Bingham Oceanogr. Coll., 3(3) : 1-193. Paxton, J. R. 1963. A new lanternfish (family Myctophidae) of the genus Lampa- dena from the eastern Pacific Ocean. Copeia, 1963: 29-33. Taaning, A. V. 1928. Synopsis of the scopelids of the North Atlantic. Vidensk. Medd. Dansk Naturh. Foren., 86: 49-69. Accepted for publication September 25, 1967. CONTRIBUTIONS IN SCIENCE LOS ANGELES COUNTY MUSEUM Iumber 139 May 7, 1968 'n I? c 2jJL $u A NEW GENUS OF MUSTELID FROM THE ELLENSBURG FORMATION, WASHINGTON By Laurie J. Bryant Los Angeles County Museum of Natural EIistory Los Angeles, California 90007 Exposition Park CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8P2 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor A NEW GENUS OF MUSTELID FROM THE ELLENSBURG FORMATION, WASHINGTON By Laurie J. Bryant1 Abstract: Beckia grangerensis, n. gen. and sp., is the sec- ond reported genus of mellivorine mustelids from the Pliocene of North America. Probably representative of an Asian emigrant, the type specimen was found in association with remains of other late Clarendonian animals, including horse and camel. The jaw is that of an animal about the size of a modern wolverine, showing similarities to both Mellivora and Eomellivora. Introduction American mellivorine mustelids have previously been known only from one specimen found in Pliocene deposits. A fragmentary palate of Eomellivora cf. wimani was recorded by Stock and Hall (1933) from the Hemphillian Kern River beds in Southern California. Beckia grangerensis, n. gen. and sp., de- scribed in this paper, is the second reported specimen of an American melli- vorine. Presumably, both Eomellivora and Beckia were emigrants from Asia, where mellivorines are well known, which may have crossed the Bering land bridge and moved south down the Pacific coast. The specimen described below was collected in Yakima County, Wash- ington, and donated to the Los Angeles County Museum of Natural History by Mr. George F. Beck of Yakima in 1963, as part of a larger collection from the Ellensburg Formation. The names of institutions to which specimens referred to in this paper belong, are abbreviated as follows: GSI, Geological Survey of India; LACM, Los Angeles County Museum of Natural History; SAMD, Science and Art Museum, Dublin, Ireland. Specimens belonging to the Geological Survey of India and to the Science and Art Museum, Dublin, have previously been reported and figured. Their descriptions and measurements are taken from the original citations. Acknowledgments Acknowledgment is made to Theodore Downs and J. R. Macdonald, who read and criticized the manuscript, and to Mrs. E. Templeton, who drew the figures. 1Student Professional, Vertebrate Paleontology, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 139 MUSTELIDAE Swainson, 1835 MELLIVORINAE Gill, 1872 Beckia, new genus Genotypic Species:Beckia grangerensis, new species Geologic Range : ? Clarendonian Diagnosis : Jaw shallow, symphysis gently sloped. Mt with protoconid one-third taller than paraconid; bladelike hypoconid, separated from proto- conid by deep, narrow, open notch. Suggestion of metaconid on posterior blade of protoconid. Premolars elongate oblongs. P3 lacking accessory cusps. Mental foramina below anterior and posterior roots of P3. Tooth row straight. Beckia grangerensis, new species Figs. 1 , 2 Type : LACM 10642, a right ramus with P3-M4, roots of canine and P2. Type Locality : LACM 6431, Granger Clay Pit, Yakima County, Wash- ington. Horizon : Ellensburg Formation, ? Clarendonian. Diagnosis : As for the genus. Description : Mi with hypoconid only remaining cusp of the talonid; sug- gestion of metaconid on posterior blade of protoconid; slight cingulum around hypoconid. M4 and P4 overlap. P4 trenchant, with slight cingulum; posterior accessory cusp small, forming part of posterior blade of principal cusp; anterior cusp very small, lying at base of principal cusp; minute anterior cingular cusp. P3 crowded against P4; trenchant, without anterior or posterior accessory cusps; faint anterior cingular cusp. Both premolars oblong in outline, tapered at the anterior end. P2 represented by roots and fragment of base; crowded against lingual side of P3. Comparison with Other Species of Mellivorinae : Mellivorines are repre- sented in the Pliocene of Asia and North America by five species in two genera; Mellivora punjabiensis Lydekker, M. sivalensis (Falconer), Eomellivora necrophila Pilgrim, and E. tenebrarum Pilgrim, all from the Siwalik Hills of India, and E. wimani Zdansky from Shansi and Honan provinces, China. The only previously reported mellivorine from North America is a fragmentary palate and isolated teeth from the Hemphillian Kern River beds in California which has been identified (Stock and Hall, 1933, p. 63) as Eomellivora cf. wimani. Unfortunately, E. necrophila and E. tenebrarum are known only from fragmentary specimens, and their status is questionable. The jaw of Beckia grangerensis is shallower than in Eomellivora, the angle of the symphysis is less steep, and the premolars are more elongate than tri- angular. The M4 is as compressed as in E. wimani. The premolars are taller and more trenchant. The strong cingula seen in E. wimani are lacking in Beckia grangerensis. 1968 New Genus of Mustelid 3 Figure 1. Beckia grangerensis, n. gen. and sp., LACM 10642. Right ramus with P3 to Mi; labial view, (x 1) Figure 2. Beckia grangerensis , n. gen. and sp., LACM 10642. Right ramus with P3 to Mi; occlusal view, (x 1) The tooth row forms a straight line, as in Eomellivora; in Mellivora sivalensis, the tooth row is convex outward. The P2 apparently was set at an angle to the other teeth, the anterior end pointing outward and the posterior end overlapping the antero-lingual corner of P3. This condition is also present, and more pronounced, in Eomellivora wimani and Mellivora sivalensis. Geology and Associated Fauna of the Locality : The Granger Clay Pit is presently a source of clay used in making brick. In late Clarendonian times, it may have been a pond bottom or lake area where animals were trapped in the mud or died of natural causes. The clay lens is contained in the upper or N aches Member of the Ellensburg Formation (Russell, 1900), above the Wenas Basalt (Beck, pers. comm.; Macdonald, field notes, 1965). Associated with Beckia grangerensis are a large tortoise, a gomphothere, Hipparion lanthonyi, IPliauchenia sp., a medium and a large sized camel, and an antilocaprid. It is difficult to determine the age of the clay pit deposit. Smiley, in his study of the flora of the Ellensburg Formation (1963, p. 206), states that “Vertebrate evidence indicates an Upper Miocene (Barstovian) to Lower Pliocene (Clarendonian) age for the entire formation.” From the study of the large collection made by Mr. Beck, it appears that the vertebrates indicate an age ranging from Clarendonian to Hemphillian. None seems to indicate a 4 Contributions in Science No. 139 Barstovian age. Even localities near the base of the formation yield assemblages including such Pliocene genera as Hipparion and Nannippus. Several Pliohip- pus teeth found near the top of the formation indicate a Hemphillian or younger age (Shotwell, 1961, p. 207). This genus is named for Mr. George F. Beck of Yakima, Washington, who donated the specimen to the museum. Literature Cited Colbert, E. H. 1935. Siwalik mammals in the American Museum of Natural His- tory. Trans. Amer. Phil. Soc., N.S., 26: 1-401, 198 figs. Lydekker, R. 1884. Indian Tertiary and Post-Tertiary vertebrata. Siwalik and Nar- bada carnivora. Palaeontologia Indica (ser. 10) 2: 178-355, 21 figs., 19 pis. Pilgrim, G. E. 1932. The fossil carnivora of India. Palaeontologia Indica, (N.S.) 18: 1-232, 2 figs., 10 pis. Russell, I. C. 1900. A preliminary paper on the geology of the Cascade Mountains in northern Washington. 20th ann. rpt., U. S. Geol. Surv., 2: 83-210, maps, plates. Shotwell, J. A. 1961. Late Tertiary biogeography of horses in the northern Great Basin. J. Paleo., 35: 203-217, 10 figs. Smiley, C. J. 1963. The Ellensburg flora of Washington. Univ. Calif. Publ. Geol. Sci., 35: 159-276, 8 figs., 17 pis. Stock, C., and E. R. Hall. 1933. The Asiatic genus Eomellivora in the Pliocene of California. J. Mamm., 14: 63-65, 1 pi. Zdansky, O. 1924. Jungtertiare carnivoren Chinas. Palaeontologia Sinica (ser. C) 2 (1): 1-155, 24 figs., 33 pis. Accepted for publication October 9, 1967. 1968 New Genus of Mustelid 5 Table 1 COMPARISON OF SELECTED CHARACTERS OF MELLIVORINES* Beckia grangerensis Eomellivora wimani Eomellivora necrophila Mellivora punjabiensis Mellivora sivalensis Depth of jaw shallow deep deep shallow not known Angle of symphysis low steep steep low not known Mental foramina below roots below post, of Po of Po, ant. of P4 not known below ant. ends of P3 and P4 not known Shape of tooth row straight straight straight nearly straight convex outward Shape of premolars elongate oblong broad triangle elongate oblong P4 oblong, P3 elongate broad oblong Relative size of P3 to P4 smaller much smaller slightly smaller smaller slightly smaller Position of P2 anterior end labiad anterior end labiad straight; P3 has anterior end linguad straight anterior end labiad Accessory cusps not prominent prominent prominent not prominent not known Relative height of protoconid 1/3 taller than paraconid 1/4 taller than paraconid 1/2 taller than paraconid not known not known Width: length of Mi 1 :2.5 1:2.6 1:3 not known ca. 1 :2 Cingula of M4 slight strong strong strong slight *So little material of Eomellivora tenebrarum is known that it is impossible to include it in this table. Measurements (in millimeters) of the jaw and lower teeth of Beckia grangerensis and related genera 6 Contributions in Science No. 139 +M. sivalensis ^ SAMD 45 Type 06 +M. punjabiensis ^ GSI D20 Type ^ +E. tenebrarum GSI D22 Type A E. necrophila * GSI D243 Type A E. necrophila : GSI D254 A E. necrophila * m GSI D255b 2 A E. necrophila : GSI D256b * *E. wimani Type B. grangerensis LACM 10642 Type c 3 c ’C « 8 H a) 35 rs <*> "u' U tH Ctf D 300,000 years (5 samples) and 325,000 ± 60,000 years (1 sample). This marine terrace is the oldest late Pleistocene of southern California, having been laid down after the San Pedro sand. Unfortu- nately, because Fanale and Schaeffer (1965) report that the Lomita marl is 4 Contributions in Science No. 146 intermediate in age between the 70- and 1,200-foot terraces, considerable doubt is cast upon the validity of their techniques for determining geo- chronology. Valentine (1961), using data for molluscan assemblages, speculated that the Timms Point silt was deposited at depths between “about 25 and 100 fathoms, and waters at those depths were cooler in some season [sic] at least than at present.” The fish remains that I identified (Table 1) lend support to these theories of depth and temperature. In all, 2,601 otoliths, 121 vertebrae, 68 teeth (56 from elasmobranchs and 12 from teleosts), 2 elasmobranch dermal denticles, and 1 fin spine were recovered from 1,000 pounds or more of fossiliferous matrix excavated from this site. These items represent a minimum of 53 species of bony fishes (teleosts) and 9 of sharks, skates, and rays. Previously, only Bagg (1912) had noted teleost remains from the Timms Point silt, but his report, comprising a single plate (PI. 28 a-e) illustrating five assorted sagittae, fails to identify them except as otoliths. Unfortunately, two of these (figures c and e) show outer faces, so they are identifiable only as cottids (Cottidae); however, his figures a, b, and d are of Tarletonbeania crenularis, Sebastodes sp., and Cymatogaster aggregata, respectively. Clark (1931) listed shark teeth (3 or fewer specimens at 1 of 15 stations) in one of his tables. He questionably identified these as “ Galeus zyopterus .” Nine of the 56 elasmobranch teeth that I identified (Table 1) were from this species ( Galeorhinus zyopterus) . Most of the Timms Point silt exposures mentioned by Arnold (1903), Clark (1931), and Woodring et al. ( 1 946) have long since disappeared, and at the present rate of human encroachment, another decade could see the last surface outcrop of this important horizon covered by a building, highway, retaining wall, or some similar “mark of progress.” Thus, unless some concerted effort is made to salvage additional fossiliferous Timms Point silt in the near future, the small handful of otoliths, teeth and other remains reported here may be our only record of the fish fauna during the million years or so of geologic history required to deposit this material. Except for a small amount of material in the Zinsmeister collection, the otoliths and other fish remains reported upon here are deposited in the Los Angeles County Museum of Natural History; the figured specimens have been curated separately. SYSTEMATIC ACCOUNT Heterodontidae — horn sharks Heterodontus francisci (Girard) — horn shark Horn sharks are abundant in rocky subtidal areas between about Morro Bay, California, and Magdalena Bay, Baja California. They seldom move about during daylight hours, but at night they can be found foraging the bottom for food, primarily crustaceans. During their nocturnal feeding, they occasion- ally stray into areas of sandy or sandy-mud bottoms. They are reported to 1968 Otoliths and Fish Remains From Timms Point Silt 5 attain a length of about 4 feet, but the largest I have seen was 3 feet 2 Vs inches long and weighed 22 pounds. Horn shark teeth have been found in several southern California Pliocene and Pleistocene deposits (Fitch, 1966, and unpublished data; Kanakoff, 1956). Material : 1 tooth. Isuridae — mako sharks Carcharodon carcharias (Linnaeus) — white shark The white shark, an inhabitant of temperate and subtropical waters throughout the world, ranges from Alaska to Mazatlan, Mexico, in the eastern north Pacific (Kato, 1965). A large individual for our coast would be 12 to 15 feet long, although white sharks are reported (unreliably) to attain 35 feet. This species probably is responsible for most of California’s unprovoked attacks on swimmers and bathers, usually occurring in relatively shallow water near shore. White shark teeth have been found in several southern California Pliocene and Pleistocene deposits (Fitch, 1964, and unpublished data; Fitch and Reimer, 1967; Kanakoff, 1956). Material : 2 teeth. Isurus oxyrinchus Rafinesque — mako This world-ranging species is fairly common in our coastal waters between about San Francisco and Magdalena Bay, Baja California. An 11-footer from Santa Catalina Island, California, is the largest authentic record for the species, but the usual mako in the eastern north Pacific is shorter than 8 feet. Mako remains, primarily teeth, have been found in numerous southern Cali- fornia Pliocene and Pleistocene deposits, sometimes reported as /. glaucus (Fitch, 1964, 1967, and unpublished data; Kanakoff, 1956). Material : 7 teeth. Triakidae — smoothhounds Triakis semifasciata Girard — leopard shark Leopard sharks have been caught at many localities between Oregon and Mazatlan, Mexico. They usually frequent shallow areas where the bottom is sandy, but they also abound in rocky areas, especially around southern California’s coastal islands. Females grow larger than males and may reach lengths of 7 feet, but a 5-foot, 35-pounder can be considered quite large. Leopard shark remains have been found in several southern California Plio- cene and Pleistocene deposits (Fitch, 1967 and unpublished data; Fitch and Reimer, 1967). Material : 1 tooth. Carcharhinidae — requiem sharks Galeorhinus zyopterus Jordan and Gilbert — soupfin shark The soupfin shark ranges from northern British Columbia to about 6 Contributions in Science No. 146 Magdalena Bay, but is not abundant at the more southerly latitudes. Females occur principally south of Point Conception, usually at depths of 100 to 600 feet or more. A 61/2-foot female may weigh as much as 100 pounds, but 50- to 70-pounders are most often encountered. Soupfin shark remains, mostly teeth, have been reported from the Timms Point silt (Clark, 1931), and have been found in numerous other southern California Pliocene and Pleistocene de- posits (Fitch, 1964, 1966, 1967, and unpublished data; Fitch and Reimer, 1967; Kanakoff, 1956). Material : 9 teeth. Prionace glauca (Linnaeus) — blue shark The blue shark is a pelagic species occurring in tropical, subtropical, and warm temperate areas of all world seas. Along the southern California coast they are extremely abundant during summer months, and they almost invari- ably are males. Twelve-footers have been caught off California (close to record size for the species), but our usual blue shark will not exceed 6 feet and 50 pounds. Blue shark remains are not abundant in fossil deposits, being known from only a few other localities (Fitch, unpublished data). Material : 1 tooth in the Zinsmeister collection. Squalidae — dogfish sharks Squalus acanthias Linnaeus — spiny dogfish The spiny dogfish abounds in the north Pacific Ocean (eastern and western), ranging south on our coast to Sebastian Viscaino Bay, Baja Califor- nia. Females attain larger sizes than males, and are reported to reach 5 feet, but a 4-footer is rare. Trawling off southern California yields the best spiny dogfish catches in 100 to 250 feet of water, but they have been caught as deep as 1,200 feet. Remains of Squalus acanthias have been found in several southern Cali- fornia Pliocene and Pleistocene deposits (Fitch, 1967 and unpublished data). Material ; 19 teeth. Squatinidae — angel sharks Squatina calif ornica Ayres — Pacific angel shark Although the Pacific angel shark ranges from southern Alaska into the Gulf of California, it is not abundant north of Point Conception nor south of Magdalena Bay. They seem to prefer sandy or sandy mud bottoms at depths of 50 to 70 feet, but many are taken or seen in 6 to 8 feet of water and others in 150 feet or deeper. A 44-inch female, the largest I have seen, weighed 31 pounds. Squatina remains have been found in many Pliocene and Pleistocene deposits in southern California (Fitch, 1964, 1967, and unpublished data). Material : 1 dermal denticle in the Zinsmeister collection. 1968 Otoliths and Fish Remains From Timms Point Silt 7 Rajidae — skates Raja spp. — skates, species undetermined There do not appear to be any foolproof tooth characters for distinguish- ing the six species of skates that occur off California. At various times, one or more of these six skates can be captured at any depth from the shallow subtidal to several thousand feet. Four of the species ( R . inornata, R. stellulata, R. trachura, and R. kincaidi) are “small” forms which probably never exceed lengths of 2 Vi feet (including tail) as adults. R. rhina is reported to attain 4 to 5 feet, and R. binoculata may reach 8 feet and exceed 150 pounds. Skate re- mains, mostly teeth and “wing” spines, have been found in numerous southern California Pliocene and Pleistocene deposits (Fitch, 1964, 1966, 1967, and unpublished data) . Material : 16 teeth and 1 wing spine. Clupeidae — herrings Clupea pallasi Valenciennes — Pacific herring The Pacific herring, a schooling fish, ranges throughout the north Pacific Ocean (eastern and western), but because of pollution and bay development, herring seldom are observed south of Point Conception at present, although they were abundant in San Diego Bay three decades ago. Pacific herring are said to attain lengths of 18 inches, but a 12- to 14-inch individual is considered quite large. Pacific herring otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, 1967 and unpublished data). The sagittae of a large, adult Pacific herring will exceed 5.0 mm in length. Material : 5 otoliths from 2.9 to longer than 3.3 mm (Fig. la). Engraulidae — anchovies Engraulis mordax Girard — northern anchovy The northern anchovy, a schooling fish, ranges from British Columbia to Magdalena Bay, and offshore for more than 100 miles. It is reported to attain a length of 9 inches, but the largest I have seen was an 8-inch fish that weighed 2 ounces. Otoliths of E. mordax have been found in many southern California Pliocene and Pleistocene deposits (Fitch, 1964, 1966, 1967, and unpublished data; Fitch and Reimer, 1967) . The sagittae of a large, adult northern anchovy will exceed 4.5 mm in length. Material : 4 otoliths from 2.8 to 4.1 mm long (Fig. lb). Myctophidae — lanternfishes Ceratoscopelus townsendi (Eigenmann and Eigenmann) — dogtooth lampfish This little fish (attaining perhaps 4 inches when fully grown) is widely 8 Contributions in Science No. 146 distributed in tropical and temperate waters of the Pacific, Indian, and Atlantic Oceans. Along our coast, C. townsendi ranges from north of San Francisco to Magdalena Bay, Baja California, at least. Except at night, they seldom approach within 600 feet of the surface. Their otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, unpublished data). The sagittae of a large, adult C. townsendi will exceed 3.0 mm in length. Material : 1 otolith 2.6 mm long (Fig. lc). Diaphus theta Eigenmann and Eigenmann — California headlightfish Diaphus theta ranges from the Gulf of Alaska to Cedros Island, Baja California, and offshore for several hundred miles, at least. They are reported to attain a length of 4 Vi inches (Clemens and Wilby, 1961), but in the central and southern parts of their range a 3 Vi -inch individual is a rarity. At night, D. theta, along with other creatures that make up deep scattering layers, migrates upward and has been captured within 30 feet of the surface. During daylight hours it usually is found at depths of 1,000 feet or more. Otoliths of D. theta have been found in several southern California Pliocene and Pleisto- cene deposits, more than 400 having been recovered from a San Pedro, Cali- fornia (Lomita marl) deposit (Fitch, unpublished data). The sagittae of a large, adult D. theta will exceed 3.0 mm in length. Material : 10 otoliths from 1.0 to 3.1 mm long (Fig. Id). Electrona rissoi (Cocco) — chubby flashlightfish The chubby flashlightfish apparently lives in the scattering layers 1,000 feet or more beneath the surface during daylight hours but nearer the surface at night. It is rarely caught off California today, but may be abundant in other areas. In the eastern Pacific, it ranges from about Point Conception to Mag- dalena Bay, at least. The validity of the name rissoi is questionable for the form in the eastern north Pacific, but the otoliths found in this deposit match perfectly, in morphometry but not necessarily in size, the otoliths removed from a IVi -inch-long individual caught off our coast (the species attains 23A inches, at least). Otoliths of E. rissoi have been found in small numbers in several southern California Pliocene and Pleistocene deposits (Fitch, unpub- lished data; Fitch and Reimer, 1967). The sagittae of a large, adult E. rissoi will exceed 3.5 mm in length. Material : 8 otoliths from 2.3 to 4.1 mm long (Fig. le). Lampadena urophaos Paxton— sunbeam lampfish L. urophaos ranges from about mid-Oregon to Magdalena Bay and off- shore for a considerable distance. At night, it sometimes migrates to within about 150 feet of the surface, but during daylight hours it seldom is captured shallower than about 800 feet. This species may attain 5 inches when full grown. The otoliths of L. urophaos are larger than those of any other mycto- phid I have examined. (My comparative collection contains sagittae from more 1968 Otoliths and Fish Remains From Timms Point Silt 9 than 100 myctophid species representing all but one of the recognized genera.) L. urophaos otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, unpublished data). The sagittae of a large, adult L. urophaos will exceed 7.5 mm in length. Material : 1 incomplete otolith 4.0 mm long (Fig. li). Protomyctophum crocked (Bolin) — California flashlightfish P. crocked is distributed throughout the western and eastern north Pacific Ocean (apparently discontinuously), ranging on our coast from off Puget Sound to Cape San Lucas and offshore for a considerable distance. A large individual might measure 3 inches in total length, but few exceeding 2Vi inches are caught. As with other myctophids, this species also approaches the surface at night, but not to the same extent, preferring to remain 500 feet deep or deeper. Otoliths of P. crocked have been found in several southern California Pliocene and Pleistocene deposits, nearly 200 having been recovered from the Lomita marl (Pliocene) at San Pedro, California (Fitch, unpublished data). The sagittae of a large, adult P. crocked will exceed 2.5 mm in length. Material : 8 otoliths from 1.2 to 2.4 mm long (Fig. If). Stenobrachius leucopsarus (Eigenmann and Eigenmann) — northern lampfish S. leucopsarus is one of the most abundantly captured myctophids off California. It ranges from the Bering Sea to about Cedros Island and offshore for a considerable distance. At night it sometimes approaches to within 150 feet of the surface, returning to 750 feet or deeper during daylight hours. A large individual might measure 5 inches and weigh one-third of an ounce. Otoliths of S. leucopsarus have been found in many southern California Plio- cene and Pleistocene deposits, more than 1,000 having been recovered from a San Pedro, California, (Lomita marl) deposit (Fitch, 1966, 1967, and unpub- lished data). The sagittae of a large, adult northern lampfish will exceed 1.8 mm in length. Material : 35 otoliths from 0.8 to 2.0 mm long (Fig. lg). Tarletonbeania crenularis (Jordan and Gilbert) — blue lanternfish The blue lanternfish ranges on our coast from northern Alaska to San Diego and offshore for a considerable distance. This species often migrates to the surface of the ocean at night, but during daylight hours it apparently prefers depths of 600 feet or more. It is reported to attain a length of 5 Vi inches, but off California a 4 -inch fish is unusual. Otoliths of T. crenularis have been found in several southern California Pliocene and Pleistocene de- posits, more than 300 having been recovered from the Lomita marl (Pliocene) at San Pedro (Fitch, unpublished data). Bagg (1912) illustrated a sagitta of T. crenularis, but did not identify it beyond calling it an otolith. The sagittae of a large, adult blue lanternfish will exceed 1.8 mm in length. Material : 10 otoliths from 1.2 to 2.0 mm long (Fig. lh). 10 Contributions in Science No. 146 Merlucciidae — hakes Merluccius productus (Ayres) — Pacific hake Pacific hake range from Alaska to the southern tip of Baja California, and offshore for 350 miles or more. Sometimes they are found in shallow water near shore, but mostly they travel in dense schools near the bottom where the water is deeper than 600 feet. They are said to reach lengths of 3 feet, but a 30-inch fish is rare; a 26-inch female weighed slightly more than 4 pounds. M. productus otoliths have been found in many southern California Pliocene and Pleistocene deposits (Fitch, 1964, 1966, and unpublished data; Fitch and Reimer, 1967) . The sagittae of a large, adult Pacific hake will exceed 30.0 mm in length. Material : 22 otoliths from about 7.5 to longer than 26.0 mm (Fig. lj). Gadidae — cods Theragra chalcogramma (Pallas) — walleye-pollack The walleye-pollack inhabits bottom waters of the north Pacific (western and eastern), ranging south to about Monterey Bay on our coast. It is reported to reach a length of 3 feet and a weight of 7 or 8 pounds, but maximum sizes need authentication. Sagittae of T. chalcogramma have not been recovered from any fossil deposit except the Timms Point Pleistocene (this report), and a late Pleistocene deposit at San Pedro. The sagittae of a large, adult walleye- pollack will exceed 20.0 mm in length. Material : 4 otoliths from 9.8 to 18.2 mm long (Fig. Ik). Microgadus proximus (Girard) — Pacific tomcod The Pacific tomcod ranges from Alaska to about Morro Bay, California, usually at depths of 200 feet or more, but sometimes in shallow water just out- side the surf zone. A large individual might be 12 inches long but no weights are available for such a fish; a 1014 -inch female weighed just under 6 ounces. M. proximus otoliths are abundant in some southern California Pliocene de- posits, and in the Pleistocene of northern California and Oregon (Fitch, un- published data). The sagittae of a large, adult Pacific tomcod will exceed 14.0 mm in length. Material : 11 otoliths 3.4 to larger than 6.0 mm (not figured). Melamphaidae — bigscales Melamphaes lugubris Gilbert — black bigscale M. lugubris ranges from the Bering Sea to Cedros Island, Baja California, and offshore for a considerable distance. Although young individuals usually inhabit shallower depths and more-northerly waters than adults, black big- scales are most often captured 150 to 1,500 feet beneath the surface in quite deep water. They are reported to attain lengths of about 4 Vi inches. Otoliths of 1968 Otoliths and Fish Remains From Timms Point Silt 11 M. lugubris have been found in only one other California deposit (Lomita marl Pliocene, San Pedro), but even there they are extremely rare (Fitch, unpublished data). The sagittae of a large, adult black bigscale will exceed 5.0 mm in length. Material : 1 otolith 1.3 mm long (Fig. 11). Scopelogadus bispinosus (Gilbert) — twospine bigscale S. bispinosus ranges from about latitude 40° N. to latitude 20° S., and off- shore across the Pacific. Young and half-grown individuals have been captured 400 feet beneath the surface and deeper, but adults seldom are captured within about 1,300 feet of the surface. The females attain larger sizes than males, reaching perhaps AVi inches. Otoliths of S. bispinosus have been found in only one other California deposit (Lomita marl Pliocene, San Pedro), but even there they are very rare (Fitch, unpublished data). The sagittae of a large, adult twospine bigscale will exceed 2.0 mm in length. Material : 1 otolith 2.0 mm long (Fig. lm). Bothidae — lefteyed flounders Citharichthys sordidus (Girard) — Pacific sanddab Pacific sanddabs are reported to range from southern Alaska to about Magdalena Bay, but their occurrence in central and southern Baja California needs to be verified. The maximum length and weight attributed to C. sordidus (16 inches and 2 pounds) also needs verifying. Citharichthys otoliths are abundant in most southern California Pliocene and Pleistocene deposits, and many of these are identifiable as C. sordidus (Fitch, 1964, 1966, 1967, and unpublished data; Fitch and Reimer, 1967). The sagittae of a large, adult Pacific sanddab will exceed 8.0 mm in length. Material : 16 otoliths from 4.2 to 6.7 mm long (Fig. In). Citharichthys stigmaeus Jordan and Gilbert — speckled sanddab Speckled sanddabs range along the coast from southeastern Alaska to Sebastian Viscaino Bay, Baja California, usually at depths shallower than 200 feet. A possible record-sized male just over 5 Vi inches long weighed just over 1 ounce. C. stigmaeus otoliths are abundant in most southern California Pliocene and Pleistocene deposits (Fitch, 1964, 1966, 1967, and unpublished data; Fitch and Reimer, 1967). The sagittae from a large, adult speckled sanddab will exceed 3.5 mm in length. Material : 25 otoliths from 1.7 to 2.9 mm long (Fig. lo). Citharichthys xanthostigma Gilbert — longfin sanddab The longfin sanddab ranges from Monterey Bay, California, to about Magdalena Bay, usually in water shallower than 500 feet. It probably attains a maximum size somewhat in excess of 10 inches and 8 ounces. C. xanthostigma 12 Contributions in Science No. 146 otoliths are fairly abundant in several southern California Pliocene and Pleisto- cene deposits (Fitch, unpublished data). The sagittae of a large, adult longfin sanddab will exceed 7.0 mm in length. Material : 22 otoliths from 3.0 to 7.3 mm long (Fig. Ip). Citharichthys spp. — sanddabs, species undetermined The otoliths of all three species of Citharichthys known to California are easily distinguished if they are in good condition, but most fossil sanddab otoliths are either worn or fragmented, making specific identification difficult if not impossible. Probably all three species are represented among the unidenti- fied sanddab otoliths found in this deposit Material : 267 otoliths. Pleuronectidae — righteyed flounders Atheresthes stomias (Jordan and Gilbert) — arrowtooth flounder The arrowtooth flounder ranges from the Bering Sea to Morro Bay, Cali- fornia, sometimes as shallow as 60 feet and other times as deep as 2,400. They are known to attain a length of 3214 inches and a weight of 10 pounds— a 26- inch female weighed 614 pounds. A. stomias otoliths have been found in only one other fossil deposit (Lomita marl Pliocene, San Pedro), but even in that deposit they are rare (Fitch, unpublished data). The sagittae of a large, adult arrowtooth flounder will exceed 13.0 mm in length. Material : 1 otolith 9.0 mm long (Fig. Iq). Glyptocephalus zachirus Lockington— rex sole The rex sole ranges from the Bering Sea to Ensenada, Baja California (at least), in from 25 to over 2,000 feet of water. They are reported to attain a maximum size of 22 inches and about 3 pounds. Rex sole otoliths are abundant in many southern California Pliocene and Pleistocene deposits (Fitch, 1967), more than 700 having been recovered from the Lomita marl Pliocene at San Pedro (Fitch, unpublished data). The sagittae of a large, adult rex sole will exceed 7.0 mm in length. Material : 45 otoliths from 1.2 to 5.3 mm long (Fig. lr). Lyopsetta exilis (Jordan and Gilbert) — slender sole The slender sole ranges from southeastern Alaska to Cedros Island, usu- ally in depths of 400 to 800 feet, but sometimes as shallow as 120 feet or as deep as 1,700. A large individual might exceed 12 inches in length, but would not weigh more than 14 pound. L. exilis otoliths have been found in many southern California Pliocene and Pleistocene deposits (Fitch, 1964, 1966, 1967, and unpublished data; Fitch and Reimer, 1967), more than 2,000 having been recovered from the Lomita marl Pliocene at San Pedro (Fitch, unpublished data). The sagittae of a large, adult slender sole will exceed 5.5 mm in length. Material : 219 otoliths from 1.0 to 5.0 mm long (Fig. Is). 1968 Otoliths and Fish Remains From Timms Point Silt 13 Microstomus pacificus (Lockington) — Dover sole The Dover sole ranges from northwestern Alaska to Ensenada (at least), usually at depths of 400 to 1,200 feet, but sometimes as shallow as 100 feet or as deep as 2,400. They are said to attain a maximum size of 30 inches and 10 pounds, but individuals exceeding 5 pounds are rarely seen. I have not found otoliths of M. pacificus in any other fossil deposit. The sagittae of a large, adult Dover sole will exceed 8.0 mm in length. Material : 1 otolith 2.8 mm long (Fig. lu). Parophrys vetulus Girard — English sole The English sole ranges from northwestern Alaska to Cedros Island, migrating inshore to spawn but living in deep water (to 1,000 feet) most of the remaining time. They attain a maximum size of 21 inches and about 3 pounds, but individuals exceeding 2 pounds are rarely seen. The otoliths of P. vetulus are difficult to distinguish from those of two other California flat- fishes, Eopsetta jordani and Lepidopsetta bilineata, unless they are in perfect or near-perfect condition. Most of the Timms Point otoliths that I identified as Parophrys were in relatively poor condition. The sagittae of a large, adult English sole will exceed 9.0 mm in length Material : 24 otoliths from 1.6 to longer than 6.5 mm (Fig. It). Atherinidae — silversides Atherinops affinis (Ayres) — topsmelt Topsmelt range from the Straits of Juan de Fuca to and into the Gulf of California. Various subspecies inhabit bays, kelp beds, and offshore island areas where they live at or near the surface. A 1414 -inch female weighing slightly less than 12 ounces appears to be a record size. A. affinis otoliths have been found in many southern California Pliocene and Pleistocene deposits (Fitch, 1964, 1966, 1967, and unpublished data). The sagittae of a large, adult top- smelt will exceed 5.0 mm in length. Material : 2 otoliths from 2.1 to 4.4 mm long (Fig. lv). Carangidae — jacks Trachurus symmetricus (Ayres) — Pacific jackmackerei The Pacific jackmackerei, a schooling fish, is perhaps one of the three or four most abundant species off our coast, ranging from British Columbia to Cape San Lucas and offshore for several hundred miles. Jackmackerei are known to attain a length of 30 inches and a weight of 5 pounds, but most individuals in the commercial catch are smaller than 15 inches and 1 pound. T. symmetricus otoliths have been found in several southern California Plio- cene and Pleistocene deposits, but they are not abundant in any of them 14 Contributions in Science No. 146 (Fitch, 1966, 1967, and unpublished data). The sagittae of a large, adult Pacific jackmackerel will exceed 10.0 mm in length. Material. 1 otolith 4.3 mm long (Fig. lw). Sciaenidae — croakers Genyonemus lineatus (Ayres) — white croaker The white croaker ranges from Vancouver Island to Magdalena Bay, abounding in almost every type of habitat from the intertidal into depths of 600 feet. A near-record 1414 -inch fish weighed 1.4 pounds. G. lineatus oto- liths often are the most abundant fish remains in southern California Pliocene and Pleistocene deposits (Fitch, 1964, 1966, 1967, and unpublished data; Fitch and Reimer, 1967; Kanakoff, 1956), comprising 6,409 of the more than 11,000 otoliths recovered from the San Diego formation (Pliocene) near Imperial Beach (Fitch, unpublished data). The sagittae of a large, adult white croaker will exceed 12.5 mm in length. Material : 8 otoliths from 4.5 to 7.0 mm long (Fig. lx). Seriphus politus Ayres — queenfish The queenfish ranges from Yaquina Bay, Oregon, to San Juanico Bay, Baja California, living in much the same habitat as the white croaker. A near- record 12-inch female weighed just over 10 ounces. S. politus otoliths have been found in many southern California Pliocene and Pleistocene deposits, often comprising a major portion of the total fish remains recovered (Fitch, 1964, 1966, 1967, and unpublished data; Fitch and Reimer, 1967; Kanakoff, 1956). The sagittae of a large, adult queenfish will exceed 10.0 mm in length. Material : 8 otoliths from 3.0 to 6.6 mm long (Fig. ly). Embiotocidae — surfperches Cymatogaster aggregata Gibbons — shiner perch The shiner perch ranges from Port Wrangel, Alaska, to Santo Tomas Point, Baja California, being restricted to the mainland coast, primarily in Figure 1. Teleost sagittae found in Timms Point silt at San Pedro, California. Lengths (in mm) are given for each otolith, and notations are made regarding its position in the skull (left or right), and condition if imperfect, a. Clupea pallasi, 3.3, /; b. En- graulis mordax, 4.1, r; c. Ceratoscopelus townsendi, 2.6, r, badly worn; d. Diaphus theta, 2.7, r; e. Electrona rissoi, 3.4, /; f. Protomyctophum crocked, 1.7, /; g. Steno- brachius leucopsarus, 1.6, /; h. Tarletonbeania crenularis, 2.0, r; i. Lampadena uro- phaos, 4.0, r, anterior end missing; j. Merluccius productus, 10.2, r, posterior tip miss- ing; k. Theragra chalcogramma, 9.8, r; 1. Melamphaes lugubris, 1.3, /, badly eroded; m. Scopelogadus bispinosus, 2.0, r; n. Citharichthys sordidus, 4.5, /; o. C. stigmaeus, 2.7, /; p. C. xanthostigma, 6.3, r; q. Atheresthes stomias, 9.0, /, badly eroded; r. Glyp- tocephalus zachirus, 3.3, r; s. Lyopsetta exilis, 3.6, r; t. Parophrys vetulus, 5.0, r; u. Microstomus pacificus, 2.8, /; v. Atherinops affinis, 4.4, r, anterior tip missing; w. Trachurus symmetricus, 4.3, r, badly worn; x. Genyonemus lineatus, 5.0, r, badly worn; y. Seriphus politus, 6.6, /, badly worn. Photographs by Jack W. Schott. 1968 Otoliths and Fish Remains From Timms Point Silt 15 16 Contributions in Science No. 146 depths shallower than 50 feet but sometimes caught as deep as 400. A record- sized pregnant female weighed just under 3 ounces. C. aggregata otoliths are abundant in most southern California Pliocene and Pleistocene deposits (Fitch, 1964, 1966, and unpublished data; Fitch and Reimer, 1967), comprising 1,129 of the more than 2,700 otoliths recovered from a San Pedro Pleistocene deposit (Miraflores Street: Fitch, 1967). Bagg (1912) illustrated a C. aggregata sagitta from Timms Point, but did not identify it except as an otolith. The sagittae of a large, adult shiner perch will exceed 6.5 mm in length. Material : 1 14 otoliths from 2.2 to 6.8 mm long (Fig. 2a). Embiotocidae — unidentified surfperches Seven otolith fragments in the Zinsmeister collection could be identified as belonging to family Embiotocidae, but it was impossible to assign them to a genus or species. Labridae — wrasses Oxyjulis calif ornica (Gunther) — senorita The senorita ranges from Natural Bridges State Park, California, to Cedros Island, inhabiting shallow areas of rocky substrate where kelp beds and other vegetation grow. A near-record 9-inch male speared at Dana Point in January 1965 weighed 4 ounces. Senorita otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, 1967, and unpublished data). The sagittae of a large, adult senorita will exceed 3.5 mm in length. Material : 2 otoliths 1.5 mm long (Fig. 2b). Pimelometopon pulchrum (Ayres) — California sheephead The California sheephead ranges from Monterey Bay to and into the Gulf of California. Throughout much of this area they abound in rocky areas, particularly where dense kelp beds grow. The male sheephead grows bigger and lives longer than the female. A record weight seems to be 3614 pounds, but this fish was not measured. A 29-pound male was 32 inches long. P. pulchrum remains, primarily jaw and pharyngeal teeth, have been found in many south- ern California Pliocene and Pleistocene deposits (Fitch, 1964, and unpub- lished data; Kanakoff, 1956). Material : 3 jaw teeth, plus 1 pharyngeal clump in the Zinsmeister collection. Scorpaenidae — rockfishes Sebastodes goodei Eigenmann and Eigenmann — chilipepper The chilipepper ranges from Vancouver Island to Magdalena Bay, in- habiting depths of 350 to 1,080 feet when adult, but living in shallower water as a juvenile. A 201/2-inch female (they are reported to attain 22 inches) 1968 Otoliths and Fish Remains From Timms Point Silt 17 weighed 3 pounds 5 ounces. S. goodei otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, unpublished data). The sagittae of a large, adult chilipepper will exceed 17.0 mm in length. Material : 1 otolith 7.3 mm long (Fig. 2d). Sebastodes hopkinsi Cramer — squarespot rockfish The squarespot rockfish ranges from Halfmoon Bay, California, to south of Ensenada, Baja California, usually in depths of 60 to 600 feet. They are reported to attain a size of 1 1 inches and about V2 pound. An 8 Vi -inch fish weighed just over 4 ounces and was found to be 5 years old. S. hopkinsi remains have not been identified from any other fossil deposit. The sagittae of a large, adult squarespot rockfish will exceed 10.0 mm in length. Material : 1 otolith 4.5 mm long (Fig. 2c). Sebastodes rosaceus Jordan and Gilbert — rosy rockfish The rosy rockfish ranges from Queen Charlotte Sound to Turtle Bay, Baja California, usually in 60 to 450 feet of water. A 10-inch female (the species attains a maximum of \23A inches) weighed slightly less than Vi pound. I have not identified S. rosaceus remains from any other fossil deposit. The sagittae of a large, adult rosy rockfish will exceed 13.0 mm in length. Material : 1 otolith 8.7 mm long (Fig. 2e). Sebastodes spp. — rockfishes, species undetermined The otoliths of most of the 52 species of Sebastodes inhabiting the waters of California can be distinguished one from the other if they are from adult fish, and if they are not badly worn or broken. Such characters as length and shape of rostrum (if present), configuration of posterior end, presence or absence of marginal frills, angle of posterior taper, depth of sulcus, and num- ber of growth zones (annuli) for otolith length are helpful for identifying sagittae of the various species or species-complexes. Unfortunately, very few fossil rockfish otoliths are well enough preserved to be identified to species. Bagg (1912) illustrated a Sebastodes sagitta from this deposit but did not identify it except as an otolith. Kanakoff (1956) reported Sebastodes otoliths from a Newport Beach Pleistocene deposit. Material : 464 otoliths, apparently representing more than five species, mostly from juveniles. Sebastolobus sp. — channel rockfish Two species of Sebastolobus inhabit the waters off California: S. alas- canus, which attains nearly 30 inches, lives in shallower depths (to 2,000 feet) than the smaller (15 inches maximum) S. altivelis, which has been caught as deep as 2,500 feet. Sebastolobus otoliths are readily distinguished from those of Sebastodes, but I have not found any characters (except size) to distinguish sagittae of S. alascanus from those of S. altivelis. Otoliths from a large (12- 18 Contributions in Science No. 146 inch) S. altivelis are about 10.0 mm long, whereas those from a large (21- inch, and 414 -pound) S. alascanus are about 16.0 mm — both fish appeared to be roughly the same age. Material : 3 otoliths from 7.0 to 9.0 mm long, the largest (Fig. 2f) probably from the shortspine channel rockfish, S. alascanus Bean. Cottidae — sculpins Artedius notospilotus Girard — bonehead sculpin The bonehead sculpin ranges from Puget Sound, Washington, to Cape Colnett, Baja California, being taken in the intertidal and in 180 feet at the other extreme. A record-sized female trawled off Anaheim Bay, California, in February 1957, was 814 inches long and weighed just over 6 ounces. A. noto- spilotous otoliths have been found in several southern California Pliocene and Pleistocene deposits, but they are never abundant (Fitch, unpublished data; Fitch and Reimer, 1967). The sagittae of a large, adult bonehead sculpin will exceed 6.0 mm in length. Material : 6 otoliths from 3.5 to longer than 5.3 mm, the length of a fragmented one (Fig. 2g). Chitonotus pugetensis (Steindachner) — roughback sculpin The roughback sculpin ranges from northern British Columbia to Mag- dalena Bay, usually at depths of 100 to 300 feet. A 7-inch fish, about maximum for the species, weighed just under 3 ounces. C. pugetensis otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, 1964, 1967, and unpublished data). The sagittae of a large, adult roughback sculpin will exceed 6.0 mm in length. Material : 7 otoliths from 2.6 to 4.4 mm long (Fig. 2h). Enophrys taurina Gilbert — bull sculpin The bull sculpin ranges from Monterey Bay to Santa Catalina Island, California, in depths of 30 to 800 feet (usually 100 to 250). A 6-inch fish, about maximum for the species, will weigh about 3 ounces. E. taurina otoliths have been found in several southern California Pliocene and Pleistocene de- posits (Fitch, 1967 and unpublished data). The sagittae of an adult bull sculpin will exceed 5.5 mm in length. Material : 6 otoliths from 4.4 to 5.2 mm long (Fig. 2i). Icelinus burchami Evermann and Goldsborough — dusky sculpin The dusky sculpin ranges from Alaska (55° N) to Santa Catalina Island, California (at least). It has been captured at depths of 600 to 1,500 feet, but appears to be rare throughout its range. A 5*4 -inch fish, about maximum for the species, weighs just over 2 ounces. /. burchami sagittae bear no resemblance to those of any other Icelinus otolith in my collection (six species), in fact 1968 Otoliths and Fish Remains From Timms Point Silt 19 they are so different, they suggest a generic affinity other than with Icelinus. The dusky sculpin apparently was much more abundant in the Pliocene than it now is — nearly 600 of its otoliths were recovered from the Lomita marl at San Pedro (Fitch, unpublished data). The sagittae of an adult dusky sculpin will exceed 5.0 mm in length. Material : 1 otolith 3.9 mm long (Fig. 2j). Icelinus filamentosus Gilbert — threadfin sculpin The threadfin sculpin ranges from northern British Columbia to the vicinity of Ensenada, at least. It is abundant on a muddy or sandy mud bottom in 60 to about 1,200 feet of water. A near-record lOV^-inch female weighed just under 7 ounces. Otoliths of I. filamentosus have been found in several southern California Pliocene and Pleistocene deposits (Fitch, unpublished data). The sagittae of a large, adult threadfin sculpin will exceed 10.0 mm in length. Material : 7 otoliths from 3.6 to 5.7 mm long (Fig. 2k). Icelinus fimbriatus Gilbert — fringed sculpin The fringed sculpin ranges from about San Francisco to San Diego, having been taken in 150 to nearly 1,000 feet of water. An 8-inch female (maximum for the species is reported as 9 inches) weighed exactly 3 ounces. /. fimbriatus otoliths have been found in several southern California Pliocene and Pleisto- cene deposits (Fitch, unpublished data). The sagittae of a large, adult fringed sculpin will exceed 7.5 mm in length. Material : 7 otoliths from 3.5 to 5.2 mm long (Fig. 2n). Icelinus quadriseriatus (Lockington) — yellowchin sculpin The yellowchin sculpin is one of the most abundant members of the family in moderate depths (50 to 250 feet) between about Point Reyes, Cali- fornia, and Cape San Lucas, Baja California. A 3 Vi -inch fish, the largest of nearly 10,000 examined, weighed slightly less than 1 ounce. /. quadriseriatus otoliths have been found in several southern California Pliocene and Pleisto- cene deposits (Fitch, 1966, and unpublished data). The sagittae of a large, adult yellowchin sculpin will exceed 3.8 mm in length. Material : 43 otoliths from 2.4 to 4.1 mm long (Fig. 2 1). Icelinus tenuis Gilbert — spotfin sculpin The spotfin sculpin ranges from Queen Charlotte Sound to Guadalupe Island, Baja California, usually in depths of 100 to 400 feet. At maximum length, perhaps AVi inches, a female I. tenuis probably weighs less than 1 ounce, even when full of roe. Spotfin sculpin otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, 1967, and un- 20 Contributions in Science No. 146 published data). The sagittae of a large, adult spotfin sculpin will exceed 4.5 mm in length. Material : 1 1 otoliths from 2.5 to 4.0 mm long (Fig. 2m). Leptocottus armatus Girard — Pacific staghorn sculpin The Pacific staghorn sculpin ranges from northwestern Alaska to San Quintin Bay, Baja California, being especially abundant in shallow outer coast waters, and in bays and lagoons. A 10-inch female (2 inches short of the reported maximum) netted in 1958, weighed V2 pound. L. armatus otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, 1967, and unpublished data; Kanakoff, 1956). The sagittae of an adult staghorn sculpin will exceed 9.0 mm in length. Material : 3 otoliths from 4.2 to 5.8 mm long (2o). Malacocottus zonurus Bean — blackfin sculpin The blackfin sculpin ranges from the Gulf of Alaska to southern Wash- ington, being relatively abundant in 90 to 500 feet. It is reported to attain 8 inches and 1/3 pound, but no data are available for a record-sized fish. M. zonurus otoliths have been found in only one other southern California fossil deposit (Lomita marl Pliocene, San Pedro) — -more than 650 having been recovered from a relatively small field sample (Fitch, unpublished data). The sagittae of an adult blackfin sculpin will exceed 9.0 mm in length. Material : 4 otoliths from 6.5 to 8.0 mm long (Fig. 2p). Radulinus asprellus Gilbert — slim sculpin The slim sculpin ranges from Kodiak Island, Alaska, to about Ensenada, being fairly abundant in 150 to 600 feet of water. A 51/i-inch fish (large for the species) weighed about 1 ounce. Otoliths of R. asprellus have been found in several southern California Pliocene and Pleistocene deposits, over 1,300 having been recovered from the Lomita marl Pliocene at San Pedro (Fitch, 1967, and unpublished data). The sagittae of a large, adult slim sculpin will exceed 4.0 mm in length. Material : 81 otoliths from 2.5 to 4.3 mm long (Fig. 2q). Cottidae — genera and species undetermined Cottid otoliths are easy to identify to family, but if they are worn or broken it is nearly impossible to assign a generic (except for Icelinus ) or specific name. Most of the cottid otoliths in this deposit were badly eroded (perhaps from digestive action by some predator), and even though the major- ity of these appeared to be Icelinus, I preferred to list them simply as “unde- termined cottids” for ease in accounting. Bagg (1912) illustrated two cottid sagittae from this deposit but did not identify them except as otoliths. Material : 419 otoliths 1968 Otoliths and Fish Remains From Timms Point Silt 21 Agonidae — poachers Xeneretmus cf. latifrons (Gilbert) — blackedge poacher The sagittae of the four species of Xeneretmus known to California are impossible to distinguish one from the other unless they are in perfect or near-perfect condition, and even then the task is difficult. Two of the species are rare, deep-water forms (X. ritteri and X. leiops ), but the other two, X. triacanthus and X. latifrons are relatively common in moderate depths. X. lati- frons is by far the most abundant of the four, and on this basis alone, I have questionably assigned the poacher otoliths from this deposit. The blackedge poacher ranges from southern British Columbia to about Ensenada, usually in 300 to 900 feet of water. A 7Vi-inch individual, which is apparently maxi- mum length for X. latifrons, weighed slightly less than 1 ounce. Xeneretmus otoliths have been found in only one other southern California fossil deposit: Lomita marl Pliocene, San Pedro (Fitch, unpublished data). The sagittae of a large, adult blackedge poacher will exceed 5.0 mm in length. Material: 2 otoliths, only one measurable (3.4 mm long, Fig. 2r). Gobiidae — gobies Coryphopterus nicholsi (Bean) — bluespot goby The bluespot goby ranges from British Columbia to San Martin Island, Baja California, typically from the shallow subtidal into 200 feet or more. The maximum size attained by the species is about 5 inches and 1 ounce. C. nicholsi otoliths have been found in many southern California Pliocene and Pleistocene deposits, over 1,700 having been recovered from the Lomita marl Pliocene at San Pedro (Fitch, 1967, and unpublished data). The sagittae of a large, adult bluespot goby will exceed 3.0 mm in length. Material : 276 otoliths from 1.1 to 3.5 mm long (Fig. 2s). Lethops connectens Hubbs — halfblind goby The halfblind goby ranges from Monterey Bay to about Cape Colnett, Baja California, but it is not very common north of Point Conception. Juveniles often have been observed in the water column in and around kelp beds, but adults are more secretive in their habits and usually remain hidden under rocks, in kelp holdfasts, and similar places. A large adult, slightly longer than 2 inches, will have sagittae longer than 1 .4 mm. Otoliths of halfblind gobies have been found in a number of Pleistocene and Pliocene deposits in southern California (Fitch, unpublished data). Material : 78 otoliths from 0.7 to 1.3 mm long (Fig. 2t). Batrachoididae — toadfishes Porichthys notatus Girard— -plainfin midshipman The plainfin midshipman ranges from southeastern Alaska to Cedros 22 Contributions in Science No. 146 1968 Otoliths and Fish Remains From Timms Point Silt 23 Island, Baja California, being one of the half-dozen most abundant species trawled in 300 to 750 feet of water. A 13 145 -inch male (2 inches short of the reported maximum) weighed just over 14 ounces. P. notatus otoliths usually are abundant in southern California Pliocene and Pleistocene deposits (Fitch, 1964, 1966, 1967, and unpublished data; Fitch and Reimer, 1967), nearly 700 having been recovered from the Lomita marl Pliocene at San Pedro (Fitch, unpublished data). The sagittae of a large, adult plainfin midshipman will exceed 10.0 in length. Material : 101 otoliths from 1.5 to 6.0 mm long (Fig. 2u). Stichaeidae— pricklebacks Lyconectes aleutensis Gilbert — dwarf wrymouth The dwarf wrymouth ranges from the Bering Sea to Eureka, California, usually at depths of 1 50 to 1 ,200 feet or more. The species apparently attains a length of 12 inches and a weight of about 2 ounces. Otoliths of L. aleutensis have been found in several other southern California Pliocene and Pleistocene deposits, but they are not abundant in any of these (Fitch, unpublished data). The sagittae of a large, adult dwarf wrymouth will exceed 5.0 mm in length. Material : 25 otoliths from 2.8 to 5.2 mm long (Fig. 3a). Zoarcidae — eelpouts Lycodopsis pacifica (Collett) — blackbelly eelpout The blackbelly eelpout ranges from the Gulf of Alaska to Ensenada (at least), being very abundant in trawl catches made in depths of 100 to 800 feet or more. The species is known to attain a size of 14 inches and nearly 1/3 pound. L. pacifica otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, 1967, and unpublished data). The sagittae of an adult blackbelly eelpout will exceed 5.0 mm in length. Material : 34 otoliths from 2.7 to 4.8 mm long (Fig. 3b). Figure 2. Teleost sagittae found in Timms Point silt at San Pedro, California. Lengths (in mm) are given for each otolith, and notations are made regarding its position in the skull, and condition if imperfect, a. Cymatogaster aggregate, 5.2, r; b. Oxyjulis californica, 1.5, r; c. Sebastodes hopkinsi, 4.5, r; d. S. goodei, 7.3, /; e. S. rosaceus, 8.7, /; f. Sebastolobus sp., 9.0, l, badly worn; g. Artedius notospilotus, 5.3, r; h. Chitono- tus pugetensis, 3.2, /; i. Enophrys taurina, 5.0, /; j. Icelinus burchami, 3.9, r, badly eroded; k. I. flamentosus, 5.4, r, badly worn; 1. 1. quadriseriatus, 4.1,/; m. /. tenuis, 2.9, /; n. /. fimbriatus, 5.2, r\ o. Leptocottus armatus, 5.8, /; p. Malacocottus zonurus, 8.0, /, badly worn; q. Radulinus asprellus, 2.8, l ; r. Xeneretmus sp., 3.4, /, rostrum missing and badly eroded; s. Coryphopterus nicholsi, 2.8, r; t. Lethops connectens, 1.2, r, u. Porichthys notatus, 5.5, r. Photographs by Jack W. Schott. 24 Contributions in Science No. 146 Brotulidae — brotulas Brosmophycis marginata (Ayres) — red brotula The red brotula ranges from southeastern Alaska to Ensenada (at least), mostly in depths of 60 to 400 feet. Although reported to attain 18 inches, few individuals are seen that exceed 16 inches and 12 ounces. B. marginata otoliths have been found in several southern California Pliocene and Pleistocene deposits, but they are never abundant (Fitch, 1967, and unpublished data). The sagittae of a large, adult red brotula will exceed 14.0 mm in length. Material : 15 otoliths from 1.4 to 5.8 mm long (Fig. 3c). Ophidiidae— cusk-eels Otophidium taylori (Girard) — spotted cusk-eel The spotted cusk-eel ranges from about northern Oregon to San Cristobal Bay, Baja California, primarily in 60 to 800 feet of water. A near-record 1414 -inch specimen caught off San Pedro in 1960 weighed about 10 ounces. O. taylori otoliths are abundant in many southern California Pliocene and Pleistocene deposits (Fitch, 1964, 1966, and unpublished data; Fitch and Reimer, 1967; Kanakoff, 1956). The sagittae of a large, adult spotted cusk-eel will exceed 11.0 mm in length. Material : 113 otoliths from 2.9 to 8.0 mm long (Fig. 3d). Ammodytidae — sandlances Ammodytes hexapterus Pallas — Pacific sandlance *■ The Pacific sandlance ranges from the Bering Sea to San Simeon, Cali- fornia, often occurring in schools from the surf zone to a considerable distance Figure 3. Teleost sagittae found in Timms Point silt at San Pedro, California. Lengths (in mm) are given for each otolith, and a notation is made as to whether it is left or right, a. Lyconectes aleutensis, 4.9, /; b. Lycodopsis pacifica, 4.4, r; c. Bros- mophycis marginata, 5.2, /; d. Otophidium taylori, 5.6, r; e. Ammodytes hexapterus, 2.4, /. Photographs by Jack W. Schott. 1968 Otoliths and Fish Remains From Timms Point Silt 25 offshore. The species is reported to attain a size of 8 inches and about Vi ounce. A. hexapterus otoliths have been found in several southern California Pliocene and Pleistocene deposits (Fitch, unpublished data). The sagittae of a large, adult Pacific sandlance will exceed 3.0 mm in length. Material : 8 otoliths from 1.4 to 3.2 mm long (Fig. 3e). Unidentified teleosts Two otoliths (very badly worn), 8 teeth, 121 vertebrae, and 1 fin spine recovered from this deposit could not be identified except as teleost remains. Most of these are presumed to have come from the same species that “left” their otoliths. DISCUSSION Nine of the more than 53 species of teleosts identified from the Timms Point silt belonged to two families of mesopelagic fishes (Myctophidae and Melamphaidae) that seldom are captured in depths shallower than about 1,000 feet. All of these are found in waters offshore from San Pedro as well as at latitudes to the north and south of this fossil deposit, and all but one, Electrona rissoi, are taken abundantly in midwater trawls. It is commonplace to find otoliths of one to three species of myctophids in southern California Pliocene and Pleistocene deposits, including those presumably laid down at relatively shallow depths (±60 feet), but few deposits that I have investigated have contained melamphaid otoliths, or sagittae from as many myctophid species (7) as I found at this locality. Otoliths of deep-sea fishes can arrive in moderately-deep or shallow deposits in any of several ways: (i) fishes can be eaten by a predator in deeper waters and their otoliths transported to a shallow area in the predator’s diges- tive tract, which they can pass through in a relatively unaltered condition; (ii) when mass mortalities occur, the dead fishes can be transported to shallower areas by currents or wind action, or in the stomachs of scavengers; (iii) mesopelagics which undertake diurnal migrations are known to approach close to shore at night near the heads of submarine canyons or where the bottom gradient rises abruptly near shore — predation and mortality in these shallow waters will result in otoliths being deposited there; (iv) spawning migrations take some species into shallower waters than they inhabit the rest of the year; and (v) deep areas which contain otoliths of offshore species can be folded or faulted upward so that subsequent otolith deposition will be from shallow species (downfaulting of shallow bottom areas could also result in an admixture of otoliths from nearshore and offshore species). Six of the 53 teleost species found in this deposit ( Ammodytes , Ather- esthes, Lyconectes, Malacocottus, Microgadus, and Theragra ) have not been captured or noted south of about Monterey, California, during modern times (Figure 4). Only two of these ( Ammodytes and Microgadus ) are found in 26 Contributions in Science No. 146 CD sO O ro Figure 4. Present-day distributions of 51 of the 53 species of teleosts identified from Timms Point silt at San Pedro, California (lat. 33° 43.8' N.). Otoliths of Lethops connectens (Monterey Bay to Cape Colnett) and Microgadus proximus (Alaska to Morro Bay) were identified after this chart was completed. 1968 Otoliths and Fish Remains From Timms Point Silt 27 relatively shallow water, but even these species are more abundant in offshore waters where depths exceed 100 feet. Only 10 of the more than 53 species identified from this deposit were captured during 3 years of beach seining (451 hauls) along the southern Cali- fornia coast to depths of about 10 or 12 feet (Carlisle, Schott, and Abramson, 1960), but 7 of these 10 were also taken during otter-trawling operations in Santa Monica Bay using fine-mesh nets in 60 to 600 feet. In all, these otter- trawling operations captured 27 of the 53+ Timms Point species (Calif. Dept. Fish and Game unpublished data). Obviously the Timms Point fish fauna admirably supports the statement of Valentine (1961) and others that deposi- tion occurred at depths between 25 and 100 fathoms at a time when water temperatures were considerably cooler than at present. In view of the numbers of northern fishes (Figure 4, Table 1 ) , it would be interesting to know how many individuals of a given species are needed in a square mile of ocean before an otolith of that species finds its way into a given area of bottom. If such data were available, it would be possible to speculate as to the density of Lyconectes aleutensis that resulted in 25 of their otoliths being found in the 1,000 pounds or so of fossiliferous matrix that was examined. It would also be interesting to know if the single arrowtooth flounder otolith would represent a comparable number of Atheresthes stomias. It seems reason- able to assume that otoliths of a prey species or potential prey species (e.g., Ammodytes and Lyconectes ) will be present in greater numbers than those of a large predator (e.g., Atheresthes ) which would have few natural enemies. Unfortunately, as has been pointed out by a multitude of investigators, fossil deposits represent only death associations, so conclusions as to life associations must be relegated to the realm of pure speculation unless accompanied by irrefutable proof. Such proof is lacking in the present case. ACKNOWLEDGMENTS This study was made possible by research grants (GB-1244 and GB-6490) from the National Science Foundation. In addition, many individuals were helpful in an assortment of ways. Richard A. Fitch washed, screened, and sorted the bulk of about 1 ,000 pounds of fossiliferous Timms Point silt that had been collected by numerous students and volunteer workers during Museum field trips supervised by George P. Kanakoff, then Curator of Inverte- brate Paleontology, Los Angeles County Museum of Natural History. Dr. Charles R. Wright, Sanger, California, sent many otoliths that he had found in material from this site. Dan Gotshall, California Department of Fish and Game, Eureka; Richard B. Grinols, U.S. Bureau of Commercial Fisheries, Seattle; Robert J. Lavenberg, Los Angeles County Museum of Natural History; and S. J. Westrheim, Fisheries Research Board of Canada, Nanaimo, B.C., sent me otoliths from living species that were critical in identifying some com- 28 Contributions in Science No. 146 ponents of this fauna. Jack W. Schott, San Pedro, took the excellent otolith photographs; Walter Thomsen, Carmichael, produced the fish distribution map; and Mrs. Loretta Morris, San Pedro, typed the manuscript and its revisions. The several student volunteers who excavated the deposit were: Nick Furjanick, Pat LaFollett, Louis Marinkovich, Peter Oringer, Robert Rashkin, Roger Reimer, and William Warner. My wife, Arline, listened to my reading of the manuscript to see if it sounded satisfactory, and proofread various typed and printed versions. To all these I offer my sincere thanks. If I have forgotten to mention anyone who helped bring this study to a successful conclusion, it has not been intentional. LITERATURE CITED Arnold, Ralph. 1903. The paleontology and stratigraphy of the marine Pliocene and Pleistocene of San Pedro, California. Mem. Calif. Acad. Sci., 3: 1-420. Bagg, R. M. 1912. Pliocene and Pleistocene Foraminifera from southern California. Bull. U.S.Geol.Surv., 513: 1-153. Carlisle, J. G., Jr., J. W. Schott, and N. J. Abramson. 1960. The barred surfperch (Amphistichus argenteus Agassiz) in southern California. Calif. Dept. Fish and Game, Fish Bull., 109: 1-79. Clark, Alex. 1931. The cool-water Timms Point Pleistocene horizon at San Pedro, California. San Diego Soc. Nat. Hist. Trans., 7: 25-42. Clemens, W. A., and G. V. Wilby. 1961. Fishes of the Pacific coast of Canada. Fish. Res. Bd. Canada, Bull., 68 (2nd edit.) : 1-443. Fanale, F. P., and O. A. Schaeffer. 1965. Helium-uranium ratios for Pleistocene and Tertiary fossil aragonites. Science, 149(3681): 312-317. Fitch, John E. 1964. The fish fauna of the Playa del Rey locality, a southern Califor- nia marine Pleistocene deposit. Los Angeles County Mus. Nat. Hist., Contr. in Sci., 82: 1-35. . 1966. Additional fish remains, mostly otoliths, from a Pleistocene deposit at Playa del Rey, California. Los Angeles County Mus. Nat. Hist., Contr. in Sci., 119: 1-16. — . 1967. The marine fish fauna, based primarly on otoliths, of a lower Pleisto- cene deposit at San Pedro, California (LACMIP 332, San Pedro Sand). Los Angeles County Mus. Nat. Hist., Contr. in Sci., 128: 1-23. Fitch, J. E., and R. D. Reimer. 1967. Otoliths and other fish remains from a Long Beach, California, Pliocene deposit. Bull. So. Calif. Acad. Sci., 66: 77-91. Kanakoff, G. P. 1956. Fish records from the Pleistocene of southern California. Bull. So. Calif. Acad. Sci., 55: 47-49. Kanakoff, G. P., and J. H. McLean. 1966. Recognition of the cancellariid genus Neadmete Habe, 1961, in the west American fauna, with description of a new species from the Lomita marl of Los Angeles County, California. Los Angeles County Mus. Nat. Hist., Contr. in Sci., 116: 1-6. Kato, Susumu. 1965. White shark Carcharodon carcharias from the Gulf of Califor- nia with a list of sharks seen in Mazatlan, Mexico, 1964. Copeia, 1965(3) : 384. 1968 Otoliths and Fish Remains From Timms Point Silt 29 Obradovich, J. D. 1965. Isotopic ages related to Pleistocene events. INQUA VII Int. Congr. Abstr., p. 364. Valentine, J. W. 1961. Paleoecologic molluscan geography of the Californian Pleistocene. Bull. Univ. Calif., Publ. Geol. Sci., 34: 309-442. Woodring, W. P., M. N. Bramlette, and W. S. W. Kew. 1946. Geology and paleon- tology of Palos Verdes Hills, California. U.S. Geol. Surv. Prof. Pap., 207: 1-145. Accepted for publication March 28, 1968. LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE 1 JMBER 147 June 14, 1968 5 07. 15 OCCURRENCE OF A GIANT BISON, BISON LAT1FRONS , AND A SLENDER-LIMBED CAMEL, TANUPOLAMA , AT RANCHO LA BREA By Wade E. Miller 3 ;l Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8 Vi x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor OCCURRENCE OF A GIANT BISON, BISON LATIFRONS, AND A SLENDER-LIMBED CAMEL, TANUPOLAMA , AT RANCHO LA BREA By Wade E. Miller1 Abstract: Tanupolama is recognized as a new constituent in the Rancho La Brea fauna. A previously overlooked record of Bison latifrons is confirmed by additional material and its associ- ation with Bison antiquus is noted for the first time. The geo- graphic range of the former species of Bison is extended to south- ern California and its chronologic range into later Wisconsin time. Introduction Despite the fact that the Rancho La Brea fauna is widely known, the richness of its contents has not yet been fully evaluated. In using this collection for comparative purposes, it was found that a specimen belonging to a camel other than the only heretofore identified species, Camelops hesternus, is present (Webb, 1965; Stock, 1963, p. 44). Although Stock (1963, p. 47-48) indicated that the only bison in the fauna is a moderate-sized species, Bison antiquus, a few elements of an extremely large bison are also in evidence. Additional species, even of large vertebrates, may yet be discovered in this extensive accumulation of Pleistocene life. Acknowledgments Study of the Rancho La Brea collection at the Los Angeles County Museum of Natural History was made possible by the kind permission of Drs. Theodore Downs, Chief Curator of Earth Sciences, and J. R. Macdonald, Senior Curator of Vertebrate Paleontology. Thanks are due to Dr. Hildegarde Howard, Research Associate at the Los Angeles County Museum of Natural History, for encouraging the writing of this article. Discussion of Taxa Tanupolama Stock, 1928 The recognition of Tanupolama, a slender-limbed camel, in the Rancho La Brea collection is based on a complete adult astragalus, Los Angeles County Museum of Natural History (LACM) No. Z 2717 (Fig. 1). Maturity is indi- cated by the clear definition of processes and ridges and the noncancellous exterior. This specimen is 59.2 mm in greatest length, 35.7 mm in greatest width across the proximal ginglymi and 39.5 mm in greatest width across the distal ginglymi. These dimensions are significantly smaller than the smallest 1Research Associate, Vertebrate Paleontology, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 147 Figure 1. Comparison of camel astragali from Rancho La Brea. A, Tanupolama, an- terior view. B, smallest adult Camelops, anterior view. C, Tanupolama, lateral view. D, smallest adult Camelops, lateral view. Scale IV2 x. 1968 Rancho La Brea BISON LATIFRONS and TANUPOLAMA 3 Figure 2. Scatter diagram of camel astragali, x = Camelops from Rancho La Brea; o Tanupolama from McKittrick; t = Tanupolama from Rancho La Brea. adult Camelops astragalus in the Rancho La Brea collection (Figs. 1 and 2), and even much smaller than the smallest juvenile observed. However, they are well within the range of those noted for Tanupolama from McKittrick, Cali- fornia (Fig. 2), the locality of the genotypic species. The only other genus of 4 Contributions in Science No. 147 camel recognized in post-Blancan deposits of North America is Titanotylopus (Webb, 1965, p. 44-45). This camelid is larger than Camelops and greatly exceeds Tanupolama in size (Barbour and Schultz, 1934, p. 291). In the field notes taken by Wyman ( 1914, p. 50) during the excavations of the Rancho La Brea deposits, he states that the skull of a new camel was found, llama-like in structure. No further mention of the specimen was made, nor can any trace of it be found now in the collection. With the possible exception of proboscidean specimens, almost all the material from the pit of discovery (Pit 9) was unrecoverable due to ground water saturation. Possibly the speci- men was never collected. The astragalus of the camel here identified as Tanupolama was recovered from this same pit near the location given for the skull. It occurred below a wood sample which was dated in excess of 40,000 years B.P. (Berger and Libby, 1966, p. 492). Slaughter (1966, p. 86), in a table indicates the presence of Tanupolama at Rancho La Brea. He (1967, pers. comm.) has informed me, however, that this location was inadvertently listed and that the referred material came from the McKittrick deposits. Bison latifrons (Harlan, 1825) Wyman (1926) listed the fossils that had been recovered at Rancho La Brea and discussed the presence of a skull (p. 32) of Bison latifrons whose horn-cores measured a “full six feet from tip to tip.” According to Skinner and Kaisen (1947, p. 205-206), B. latifrons is the only species of Bison with a horn-core spread this great. Unfortunately, the skull reported by Wyman dis- integrated upon removal. Whether any of the fragments were saved is not known. In the Rancho La Brea field notes (p. 66), Wyman states that two metapodials of a very large bison were collected. These elements, as well as the skull, were taken from Pit 9. The only bison element that I have seen from this pit is the distal one-third of a metatarsal, LACM No. Y 2557. Its proportions exceed those of any others seen in the Rancho La Brea collection and the specimen probably is assignable to the species B. latifrons (greatest medio- lateral width at distal end, 87.0 mm; greatest anteroposterior thickness at distal end, 51.0 mm) . A few Bison latifrons bones are apparently also present in the material recovered from Pit 4. Its occurrence here marks the first time this species of Bison has been reported associated with Bison antiquus (no positive identifi- cation has been made of B. antiquus in Pit 9). Of further interest are the radiometric dates extracted from bone samples taken with these specimens from Pit 4, of 26,700 years B.P. at an eight foot depth and 28,000 years B.P. at a 15.5 foot depth (unpublished data currently being compiled at the UCLA Radiocarbon laboratory). This extends the chronologic range of B. latifrons into very late Wisconsin time. Material representing this species was recovered from a depth of ten to fifteen feet in Pit 4; Bison antiquus specimens have been 1968 Rancho La Brea BISON LATIFRONS and TANUPOLAMA 5 l A B Figure 3. Comparison of bison atlases from Rancho La Brea, dorsal view. A, Bison latif roiis. B, Bison antiquus (large specimen). About Vt. x. GREATEST DORSOVENTRAL WIDTH ( in mm ) Figure 4. Scatter diagram of bison atlases from Rancho La Brea, o = Bison antiquus', x = Bison latif rons. 6 Contributions in Science No. 147 TABLE 1 Skeletal measurements of Bison latifrons from Rancho La Brea Atlas (Y 6737) mm Greatest width across wings 260 Greatest dorsoventral thickness (across tubercles) 131 Greatest transverse distance (across proximal articulating surface) 152 Axis (Y 6734) Greatest length of centrum (including dens) 148 Greatest width (across anterior articulating surface) 138 Greatest width of dens 059 Height of neural spine (superior border of neural arch to tip of spine) 124 Scapula (Y 6738) Greatest anteroposterior distance (across articular surface) 090 Greatest transverse distance (across articular surface) 075 Ulna (Y 6736) Greatest anteroposterior distance (through anconeus process) 126 Height of olecranon process (above sigmoid notch) 160 Mandible (Y 6710) Length of jaw (lateral border of C/1 alveolus to angle) 468 Depth of jaw (at anterointernal border of P/2 alveolus) 060 Depth of jaw (at posterointernal border of M/3 alveolus) 095 Length of premolar series (from alveolar borders) 068 Length of molar series (from alveolar borders) 115 A B Figure 5. Comparison of bison axes from Rancho La Brea, lateral view. A, Bison lati- frons. B, Bison antiquus (large specimen). About Vz x. 1968 Rancho La Brea BISON LAT1FRONS and T AN U POL AM A 7 E E c Ui 0£ O 150 X 145 - 140 - 135 130 125 o o o o 120 o o o ° o o - ° ° ° o o o o o o o 115 o O 0 0 ooo 0 l i ° _J 1 _l 105 110 115 120 125 130 135 140 GREATEST WIDTH ( in mm ) Figure 6. Scatter diagram of bison axes from Rancho La Brea, o = Bison antiquus', x — Bison latifrons. taken from above and below this level. This record of the giant species of Bison now extends its geographic range to southern California. The first California record was reported by VanderHoof (1942) who described a com- plete skull from the northern part of the state. Savage (1951, p. 284) men- tioned the presence of B. latifrons in Contra Costa County, California. Recovered specimens from Pit 4 include an atlas, LACM No. Y 6737; an axis, LACM No. Y 6734; the proximal end of an ulna, LACM No. Y 6736; and a complete mandible, LACM No. Y 6710. All these specimens greatly exceed corresponding ones of B. antiquus in size and two show differences in proportions, the atlas and the axis (Figs. 3 and 5 and Table 1 ) . 8 Contributions in Science No. 147 Atlas. This element far surpasses the equivalent one of Bison antiquus in size and further differs by being more rugose, possessing more massive ventral and dorsal tubercles and having a tendency toward a trapezoidal rather than a rectangular form (dorsal and ventral view). The present atlas was found to be identical to the corresponding element in B. latifrons specimens in the Uni- versity of California Museum of Paleontology and the Los Angeles County Museum of Natural History collections. It is a little larger than the one of Bison latifrons and distinctly larger than that of B. crassicornis reported by Allen (1876, p. 14). Axis. Like the atlas, this cervical vertebra is much larger than any similar element of B. antiquus in the Rancho La Brea fauna. It further differs by possessing a greater rugosity and a larger neural spine whose anterior edge is more steeply inclined (Fig. 5). The axis is indistinguishable from axes of B. latifrons used in comparison. Scapula. The portion of this element available indicates a much larger bone but one whose proportions are similar to B. antiquus. Ulna. Like the scapula, the proximal portion of the ulna compares favor- ably in shape to Bison antiquus but is much larger. Mandible. The greater size of this element distinguishes it from lower jaws of B. antiquus. However, the teeth are relatively smaller, being about equal in size to many observed specimens of the smaller species. This observa- tion was also made by VanderHoof (1942, p. 11); the complete jaw described by him is almost identical in size to the Rancho La Brea specimen. Based on their large size, several postcranial juvenile specimens from Pit 3 of the Rancho La Brea deposits could possibly represent Bison latifrons. Literature Cited Allen, J. A. 1876. The American bison, living and extinct. Mem. Harvard Mus. Comp. Zool., 4(10): 1-246. Barbour, E. H., and C. B. Schultz. 1934. A new giant camel, Titanotylopus nebras- kensis. Bull. Nebr. State Mus., 1 : 291-294. Berger, R., and W. F. Libby. 1966. UCLA radiocarbon dates. Radiocarbon, 8: 467- 497. Savage, D. E. 1951. Late Cenozoic vertebrates of the San Francisco Bay region. Bull. Univ. Calif. Publ. Dept. Geol. Sci., 28: 215-314. Skinner, M. F., and O. C. Kaisen. 1947. The fossil Bison of Alaska and preliminary revision of the genus. Bull. Amer. Mus. Nat. Hist., 89(3) : 1-256. Slaughter, B. H. 1966. The Moore Pit local fauna, Pleistocene of Texas. J. Paleont., 40: 78-91. Stock, C. 1963. Rancho La Brea, a record of Pleistocene life in California. Los An- geles County Mus., Sci. Series, 20: 1-83. VanderHoof, V. L. 1942. A skull of Bison latifrons from the Pleistocene of north- ern California. Bull. Univ. Calif. Publ. Dept. Geol. Sci., 27: 1-24. 1968 Rancho La Brea BISON LATIFRONS and TANUPOLAMA 9 Webb, S. D. 1965. The osteology of Camelops. Los Angeles County Mus., Bull. 1: 1-54. Wyman, L. E. 1913-1915. Unpublished field notes of the excavations at Rancho La Brea for the Los Angeles County Museum. 1 : 1-134. 1926. Notes on the Pleistocene fossils obtained from Rancho La Brea Asphalt Pits. Los Angeles Mus., Misc. Publ., 2(rev. ed.) : 1-39. Accepted for publication March 30, 1968. angeles CONTRIBUTIONS uZH IN SCIENCE JMBER 148 June 28, 1968 f 01-73 t 2. L&kfe RECORDS OF BATS FROM COSTA RICA By Andrew Starrett and Richard S. Casebeer Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8V2 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor RECORDS OF BATS FROM COSTA RICA By Andrew Starrett1 and Richard S. Casebeer2 Abstract: Twenty-five species of bats from Costa Rica are reported, ten species and one subspecies of which are new records for the country. Information on distribution patterns and ecologi- cal observations are included. New to the chiropteran fauna of Costa Rica are : Saccopteryx leptura, Peropteryx macrotis, Ptero- notus davyi, Chilonycteris psilotis, Micronycteris schmidtorum, Chrotopterus auritus, Mesophylla macconnelli, Natalus strami- neas, Myotis simus, Myotis chiloensis and Eptesicus furinalis gaumeri. Balantiopteryx plicata is reported from Costa Rica for the first time since the type description, and the separate generic status of Mesophylla macconnelli is reaffirmed. Introduction In 1962, the Los Angeles County Museum of Natural History (LACM) was awarded a four-year grant from the U.S. Army Medical Research and Development Command to support a survey of the mammals and mammalian ectoparasites of Costa Rica. The mammals collected during the three years of the field phase of this study, and specimens collected in 1957 (Starrett and De la Torre, 1964), 1959, and 1961 (Casebeer, Linsky and Nelson, 1963, and Tamsitt and Valdivieso, 1961 ) by various field parties from the University of Southern California and the Los Angeles County Museum of Natural History, have provided this museum with a good representation of the chirop- teran fauna of Costa Rica. Using this material as a nucleus, the LACM has been attempting to put together a picture of the distributional patterns of the bats of Costa Rica, with the aid of published records and specimens available in other collections, as well as continued collecting in that country. There are now known from Costa Rica at least 78 species of bats, based on authenti- cated published records and specimens reported in this paper, and the total number may be expected reasonably to reach 100 as the various collections continually being made by this museum and other institutions are worked up. An increase in interest in Neotropical bat faunas is represented by a number of recent publications which have included records of species from Costa Rica (in addition to those already cited): Carter, Pine and Davis, 1966; Davis, 1965 and 1966; Davis and Carter, 1962; Davis, Carter and Pine, 1964; Nelson, 1965; Walton, 1963. This paper reports on 226 specimens of 25 species of bats, ten species and one subspecies of which are new records for Costa Rica. 1Research Associate, Los Angeles County Museum of Natural History, and Depart- ment of Biology, San Fernando Valley State College, Northridge, California 91324. 2Research Associate, Los Angeles County Museum of Natural History, and Bio- logical Sciences Curriculum Study, University of Colorado, Boulder, Colorado. 1 2 Contributions in Science No. 148 Localities are listed after capitalized province names which are reported in alphabetical order. Elevations have been converted to meters for the sake of uniformity and ease of comparison; (when shown in italics, they represent reasonable estimates, usually based on topographic maps). Measurements are in millimeters and are for adult specimens unless otherwise indicated. Capital- ized Forest Formations are those of L. R. Holdridge, from Slud, 1964. Species of Bats Rhynchonyctens naso (Wied-Neuwied) Vespertilio naso Wied-Neuwied, 1820, Reise nach Brasilien in den Jahren 1815-1817, 1: 251, footnote. Rhynchonycteris naso, Peters, 1867, Monatsb. preuss. Akad. Wiss., Berlin, p. 478. Specimens : GUANACASTE: Rio Colorado at Hacienda La Norma (Finca Coyolar) 5 km N, 4 km W Liberia, 50 m, 3 $ $ , 11 $ $ ; 0.8 miles by road SW Tilaran, (Santa Rosa drainage) , 500 m, 1 $ ; Rio Diria at Santa Cruz, 60 m, 1 $ , 2 $ $ ; Samara, sea level, 1 $ . HEREDIA: 4 miles W Puerto Viejo de Sarapiqui, 92 m, 1 $ (TCWC). LIMON: R(o Madre de Dios, near Finca La Lola, 50 m, 1 $ . PUNTARENAS: 6.5 miles N, 2 miles W Puntarenas, 30 m, 1 $ (TCWC) ; Rio La Vieja on Interamerican Highway, 1 10 m, 1 $ ; Boca de Barranca, sea level, 4 $ $ ; 2.8 miles E Golfito, 60 m, 1 $ ; 8 miles ENE Golfito, 30 m, 1 $ ; Rincon de Osa, headquarters area Osa Productos Fore- stales, 10 m, 1 $ . Measurements : Forearm, averages and ranges for 7 males and 12 females, respectively, 35.8 (34.5-37.3), 37.7 (36.3-38.8). Remarks : Although this bat is known from every Middle American country and most tropical countries in South America (Sanborn, 1937), all previous Costa Rican records are based on 10 specimens, without locality, which were collected by G. K. Cherrie and cited first (as 11 specimens) by J. A. Allen (1897). The 30 specimens recorded here were all collected at low elevations (up to 500 m) and from localities which include both the driest and wettest regions of the country as well as both Pacific and Caribbean lowlands. All of the collecting localities, with the exception of that at Rincon de Osa, are in areas which have been modified ecologically by one type of agricultural use or another; the virgin tropical wet forest of the Rincon area is undergoing change due to investigatory lumber operations. The only immediately ap- parent ecological preference shown by Rhynchonycteris in Costa Rica is for watercourses. The series from Hacienda La Norma was collected during the day from two adjacent colonies hanging from the face of cliffs along the Rio Colorado, the larger of the two colonies having an estimated 45 individuals, some of which were females carrying young. All the other LACM specimens were netted or shot over or next to streams or rivers along which they were hunting. They were taken variously in association with Saccopteryx bilineata, 1968 Records of Bats From Costa Rica 3 Noctilio leporinus, Noctilio labialis, Chilonycteris parnellii, Molossus sp., and various stenodermines which used the watercourses as flyways. Saccopteryx Septum (Schreber) Vespertilio lepturus Schreber, 1774, Die Saugethiere, Teil 1, Heft 8, pi. 57 (name); 1775, ibid., Teil 1, Heft 9, pp. 173-174 (description). Saccopteryx lepturus, llliger, 1811, Prodromus systematis mammalium et avium, p. 121. Specimens : GUANACASTE: Playas del Coco, sea level, 1 2 ; Mojal de Santa Cruz, Arenal (Cartajena), 55 m, 6 8 8,2 $ 2. LIMON: “El Tigre,” approximately 14 km S Siquirres, 825 m, 1 2. PUNTARENAS: 15 km SE Jaco, Batalla’s Finca, Esterillos Oeste, sea level, 1 2 ; 3.8 miles NW Villa Neilly along Interamerican Highway, 55 m, 1 2 ; Rincon de Osa, headquarters area Osa Productos Forestales, 10 m, 1 8. Measurements'. Averages and ranges: forearm, 7 males, 5 females, re- spectively, 38.4 (37.0-39.6), 40.9 (39.8-42.5); (cranial, males and females together) greatest length of skull (6 specimens), 13.5 (13.2-13.8); condylo- canine length (7), 11.8 (11.5-12.3); length of maxillary toothrow, C-M3 (8), 5.2 (4. 9-5. 6); postorbital constriction (8), 2.2 (2. 0-2. 6); zygomatic breadth (7), 8.6 (8.5-9. 1 ) ; mastoid breadth (7) , 7.3 (7. 1-7.5) . Remarks : This species was known previously in Middle America from Panama (Sanborn, 1937; Handley, 1966), Nicaragua (Davis et al., 1964), El Salvador (Burt and Stirton, 1961) and Chiapas, Mexico (Carter et al, 1966). The 13 specimens included here represent the first records of S. leptura from Costa Rica. All were collected at low elevations: the one from “El Tigre” represents the only record from the Caribbean slope and is from the highest elevation (825 m) ; the others are all from elevations below 60 m on the Pacific lowlands. These specimens were shot at night over roads or clearings or taken during the day from roosting sites in caves (Playas del Coco) or on sheltered areas of tree trunks (Mojal de Santa Cruz). Peropteryx macrotis macrotis (Wagner) Emballonura macrotis Wagner, 1843, Arch. Naturgesch., Jahrg. 9, 1; 367. Peropteryx macrotis macrotis, G. M. Allen, 1935, Jour. Mamm., 16:227. Specimens'. ALAJUELA: 4 km NE La Palmira, Finca Las Vegas, 400 m, 1 2 ; 8.2 miles W bridge at Atenas on road to San Mateo, 67 1 m, 1 2 . LIMON : 1 mile NW Limon, Les Cuevas, sea level, 1 8,2 2 2. PUNTARENAS: Boca de Barranca, sea level, 1 2 ; 35 miles SE San Isidro del General along Inter- american Highway, at bridge over Rio Ceibo, 320 m, 1 2. SAN JOSE: Rio Canas, Las Cuevas, 200 m, 2 8 8. Measurements : Averages and ranges: forearm (3 males, 6 females, re- spectively), 42.6 (41.9-43.0), 45.4 (43.8-46.9); (cranial, males and females together) greatest length of skull (7 specimens), 14.9 (14.5-15.3); length of 4 Contributions in Science No. 148 maxillary toothrow, C-M3 (9), 5.8 (5. 5-6.0); postorbital constriction (9), 2.9 (2. 7-3. 2) ; zygomatic breadth (9), 8.5 (8. 3-8. 8). Remarks : These nine specimens mark the first records of P. macrotis for Costa Rica, although the species is known from southern Mexico to and throughout tropical South America (Hall and Kelson, 1959; Sanborn, 1937). The Costa Rican specimens are all from lower elevations and represent both Pacific and Caribbean slopes. They were shot at night over roads in areas where trees (but not necessarily forest) occurred, or were taken from diurnal roosts in caves (Limon, Rio Canas, Finca Las Vegas). Although P. macrotis was collected in association with P. kappleri in all three of the roosts, the two species were never taken together while feeding over the roads at night. The roosting Peropteryx were found in rather open situations at or very near the mouths of the caves. Centronycteris maximiliani centralis Thomas Centronycteris centralis Thomas, 1912, Ann. Mag. Nat. Hist., ser. 8, 10: 638. Centronycteris maximiliani centralis, Sanborn, 1937, Zool. Ser. Field Mus. Nat. Hist., 20: 337. Specimens'. PUNTARENAS: Rincon de Osa, Osa Productos Forestales, Camp Seattle, 35 m, 2 $ $ , and near headquarters area, 10 m, 1 $ . Measurements'. Forearm (3 females), 42.7, 44.2, 44.4; (cranial, third female) greatest length of skull, 14.4; condylobasal length, 13.2; length of maxillary toothrow, C-M3, 5.9; postorbital constriction, 3.0; zygomatic breadth, 9.1; breadth across M3-M3, 6.1. Remarks : This species is known from relatively few localities from southern Mexico to southern Brazil (Hall and Kelson, 1959; Handley, 1966; Sanborn, 1937) and was first recorded from Costa Rica, without definite locality, by Allen and Barbour (1923). The same specimen was later listed by Sanborn (1937) as coming from Viragua, Prov. Alajuela, and then by Goodwin (1946) from Vijagua, Prov. Alajuela. Sanborn’s spelling was in error: the label on the specimen (MCZ 7092) reads “Vijagua, Costa Rica,” Vijagua being an alternative spelling for Bijagua, or La Bijagua, a small town on the Caribbean slope of Volcan Miravalles, Prov. Alajuela (some 320 km NW of Rincon de Osa) . The rarity of specimens of C. maximiliani, in the light of our limited experience with this bat, suggests that it requires relatively heavy forest (possibly virgin stands) in which to live. The very slow and highly maneuver- able, floppy flight of this species makes it well adapted to hunting insects among trees and along natural (and now man-made) pathways and clearings. The three specimens reported here were all collected in just such situations. Balantiopteryx plicata plicafa Peters Balantiopteryx plicata Peters, 1867, Monatsb. preuss. Akad. Wiss., Berlin, p. 476. 1968 Records of Bats From Costa Rica 5 B [ alantiopteryx ] p [ licata ] plicata , Burt, 1948, Occ. Pap. Mus. Zool., Univ. Mich., No. 515, p. 2. Specimens : GUANACASTE: Playas del Coco, sea level, 16 8 8, 17 $ $ ; Mico de Matapalo, 50 m, 13 8 8,1 2 . Measurements : Averages and ranges: forearm (22 males, 10 females, respectively) , 40.0 (39.0- 42.3 ) , 42.0 (40.2- 43.2) ; (cranial, 2 males, 2 females together) greatest length of skull, 14.4 (14.2-14.6); condylobasal length, 13.0 (12.9-13.1); length of maxillary toothrow, C-M3, 5.4 (5.4); interorbital breadth, 5.5 (5. 4-5. 5); postorbital constriction, 3.1 (2. 9-3. 3); zygomatic breadth, 9.1 (8. 9-9. 2); breadth of braincase, (3 specimens), 6.9 (6. 9-7.0); mastoid breadth, 8.1 (7.9-8.1); alveolar breadth across upper canines, 3.6 (3. 5-3. 7) ; breadth across M3-M3, 6.6 (6.5-6.6) . Remarks'. Since the description of B. plicata in 1867, based on one speci- men from “Puntarenas in Costa Rica,” this bat has been collected from a number of localities, and often in rather large numbers, from tropical Mexico, Guatemala and El Salvador (Hall and Kelson, 1959; Felten, 1955; Burt and Stirton, 1961), but only recently has it been reported from localities further to the south. Jones (1964) includes a single specimen from each of two localities in Nicaragua, and this paper records the first specimens to be taken in Costa Rica since the type description. Hall and Kelson (1959) assume Peters’ type locality to be Puntarenas, Prov. Puntarenas, but since the place name is used to refer to both the town and the province and Peters did not specify more than “Puntarenas,” the exact type locality must remain open to question. Be- cause most (if not all) of the known specimens of this bat apparently come from localities in the relatively drier portions of Middle America, usually on the Pacific side of the cordillera (all localities in Central America), it is most likely, on ecogeographic grounds, that the type of B. plicata came from the lowland Tropical Dry Forest region in the small portion of Puntarenas Province north and northwest of Punta Leona (some 10 km south of the Rio Grande de Tarcoles), if not actually from the town of Puntarenas itself. Both series of B alantiopteryx recorded here came from colonies in diurnal roosts. The specimens from Playas del Coco were roosting in a series of fissures in the face of a sea cliff which was accessible only at low tide; those from Mico de Matapalo were hanging in the shadow of a huge rafter near the mouth of a short, partially collapsed, abandoned manganese mine. The bats taken from the latter colony were all adults (January 26, 1963), while the former series included 18 adults, 14 subadults and one large juvenile (August 27, 1962). Didedurus virgo Thomas Diclidurus virgo Thomas, 1903, Ann. Mag. Nat. Hist., ser. 7, 11: 337. Specimen : PUNTARENAS: 2.7 miles NW Villa Neilly along Interameri- can Highway, at bridge over Rio Nuevo, 55 m, 1 2 . Measurements'. Forearm, 65.0; greatest length of skull (to front of canines), 18.4; condylobasal length, 17.6; length of maxillary toothrow, C-M3, 6 Contributions in Science No. 148 7.9; interorbital breadth, 8.7; postorbital constriction, 5.6; zygomatic breadth, 12.1; breadth of braincase, 9.2; mastoid breadth, 10.2; alveolar breadth across upper canines, 4.4; breadth across M3-M3, 8.4. Remarks : This specimen is apparently the first record from the Costa Rican lowlands for this species, which is known by relatively few records from Oaxaca, Mexico, to western Panama (Hall and Kelson, 1959; Handley, 1966). It has been reported previously from Costa Rica from three localities, all of which are on the Meseta Central (central plateau)— Escazu (type locality), San Jose and La Palma (all Prov. San Jose)- — at elevations ranging approxi- mately from 1 100 to 1500 m (with some leeway, depending on the exact locali- ties represented by the locality names). The Rio Nuevo individual was shot at night as it hawked insects over the road, flying relatively high and apparently straight courses, out of the range of the lights of the field truck, and uttering a unique musical twittering not heard in other Costa Rican bats. On the night this bat was collected (September 9, 1963), we had parked the truck by the road and a phenomenal number of moths were swarming in the headlights. Numerous bats were dodging in and out of the headlight beams, snapping up the moths and other insects, and within about an hour (starting at dusk or approximately 6:00 PM local time) we collected a number of Saccopteryx bilineata and Pteronotus suapurensis as they chased insects in our lights, as well as the Diclidurus, which was making passes above the truck. The peculiar musical call of the white bat was heard above us in the dark during most of the time until the specimen was shot, after which the sound was noticeably absent from the night air. White bats, which were almost certainly Diclidurus and not the smaller and undoubtedly less conspicuous Ectophylla alba, were mentioned by local people at Boca de Barranca, Prov. Puntarenas (sea level, Pacific lowlands) and Finca La Lola, Prov. Limon (50 m, Caribbean lowlands). In both places they were said to have been seen hanging from beneath the leaves of coco palms during the day. One was shot by one of the men working at Pension Chanita, Boca de Barranca, in 1959 (and thrown away). Cyttcirops aSecto Thomas Cyttarops alecto Thomas, 1913, Ann. Mag. Nat. Hist., ser. 8, 11: 134. Specimens : HEREDIA: Puerto Viejo de Sarapiqui, 100 m, 3 $ $ , 5 2$. Measurements'. Forearm (male, 3 females, respectively), 45.8, 47.1, 46.7, 46.6; (cranial, same male, first two females) greatest length of skull, 13.6, , 14.3; condylobasal length, 12.6, , 13.1; length of maxillary tooth- row, C-M3, 5.4, 5.6, 5.5; interorbital breadth, 5.1, , ; postorbital constriction, 4.2, 3.7, 3.9; zygomatic breadth, 8.7, — — , ; breadth of braincase, 7.1, 7.1, 7.1; mastoid breadth, 7.7, , 7.6; alveolar breadth across upper canines, 3.0, — , ; breadth across M3-M3, 6.3, , ; length of mandibular toothrow, C-M3, 6.0, 6.0, 5.9. Remarks'. This rare bat was represented only by the type specimen from 1968 Records of Bats From Costa Rica 7 Para, Brazil, and a second specimen from British Guiana (also used in the original description) until 1964, when three specimens were recorded from Los Diamantes, Prov. Limon (300 m), with a fuller description of its charac- ters (Starrett and De la Torre, 1964). The eight specimens reported here were collected at a town some 35 km northwest of Los Diamantes and also on the Caribbean lowlands, and constitute the fourth locality record for this mono- typic genus. These bats were shot during the day as they hung, in small groups, from beneath the fronds of coco palms in the town of Puerto Viejo, on three successive days (September 15-17, 1964). The series includes one adult male, four adult females, two subadult males and one subadult female (LACM 26614-26617, 26625-26627, 26633). The specimens are in a better state of preservation than those from Los Diamantes (LACM 14572-14574), which were collected from a coco palm by some workers and preserved in local alcohol. The new specimens show color variation in that some of the adults have a somewhat paler gray area in the region of the shoulders and dorsal portion of the base of the neck. In other respects they agree with those from Los Diamantes and with the type description. Noctileo ieporinus mexicanus Goldman Noctilio Ieporinus mexicanus Goldman, 1915, Proc. Biol. Soc. Wash., 28: 136. Specimens : LIMON: Tortuguero, sea level, 1 $ (UF 4647). PUN- TARENAS: Boca de Barranca, sea level, 1 8,1 $; Rio Coronado, 20 km NW Puerto Cortes, sea level, 1 $ ; Dominical, sea level, 1 $ . Measurements : Forearm (1 male, 3 females, respectively), 82.9, 85.0, 82.0, 83.9; (cranial, male and first female) greatest length of skull, 27.5, 26.1; condylobasal length, 23.9, 23.3; length of maxillary toothrow, C-M3, 10.4, 10.0; postorbital constriction, 7.2, 7.0; zygomatic breadth, 19.2, 18.2; mastoid breadth, 17.8, 16.4; breadth across M3-M3, 12.5, 12.3. Remarks : Although this fish-eating bat ranges throughout most of the Neotropics (Cabrera, 1957; Hall and Kelson, 1959) and has been known from Panama since 1923 (Allen and Barbour, 1923), it was first recorded from Nicaragua in 1964 (Davis et al., 1964) and from Costa Rica in 1966 (Carter et al., 1966). The additional specimens here reported were caught in mist nets set either across or parallel to streams, rivers or creeks near the ocean (with estuarine tidal influence at Boca de Barranca). At Hacienda Bonilla, near Boca de Barranca, we watched fish-eating bats, along with Rhynchonycteris, coursing over a long rain “puddle” (roughly five by 30 yards in extent) contain- ing minnow-sized fish. Saccopteryx bilineata, Molossus sp. and seven species of phyllostomatids were taken in various combinations in the same nets with N. Ieporinus at the three localities. Pteronotus davyi Gray Pteronotus davyi Gray, 1838, Mag. Zool. Bot., 2: 500. Chilonycteris gymnotus Wagner, 1843, Arch. Naturgeschichte, 9 (Bd. 1) : 367. 8 Contributions in Science No. 148 Chilonycteris davyi fulvus Thomas, 1892, Ann. Mag. Nat. Hist., ser. 6, 10: 410. Dermonotus davyi, Miller, 1902, Proc. Biol. Soc. Wash., 15: 155. Pteronotus davyi davyi, Miller, 1912, Bull. U.S. Nat. Mus., 79: 33. Pteronotus davyi fulvus, Miller, 1912, Bull. U.S. Nat. Mus., 79: 33. Specimens : ALAJUELA: 4.6 miles (1067 m) and 8.4 miles (671 m) W bridge at Atenas on road to San Mateo, 1 $ , 2 $ $ . GUANACASTE: Playas del Coco, sea level, 1 $ ; 5 miles N Liberia along Interamerican Highway, 150 m, 1 $ . Measurements : Averages and ranges (4 females) : forearm, 45.7 (44.7- 47.6); greatest length of skull, 16.1 (15.8-16.5); condylobasal length, 14.9 (14.7-15.2); length of maxillary toothrow, C-M3, 6.7 (6.5-6.8); postorbital constriction, 3.7 (3.5-3.8); zygomatic breadth (3 females), 8.9 (8. 9-9.0); breadth of braincase, 7.9 (7. 7-8.0); mastoid breadth, 8.5 (8. 0-8. 8); alveolar breadth across upper canines, 4.8 (4. 8-4.9) ; breadth across M3-M3, 6.0 (6.0) ; length of mandibular toothrow, C-M3, 7.1 (7.0-7.1); length of mandible, 12.0 (11.8-12.1). Remarks : Previous to the collection of the specimens reported here, there has been a gap, which extended from El Salvador (Hall and Kelson, 1959) to Venezuela (Cabrera, 1957), in the distribution of this tropical American species. The northern and Antillean-southern populations were distinguished as subspecies when Thomas (1892) described P. d. fulvus from Mexico. Apparently, however, the species essentially is distributed contin- uously. A comparison of measurements of the specimens from Costa Rica with those of P. d. fulvus (including the type) from Mexico (Goodwin, 1946; LACM specimens) and El Salvador (Burt and Stirton, 1961; Felten, 1956a) and P. d. davyi from Trinidad (Dobson, 1878; Goodwin, 1946; Goodwin and Greenhall, 1961) indicates that, although some clinal variation may exist, these forms do not warrant taxonomic distinction (as has already been sug- gested by J. A. Allen, 1894) . The status of the P. davyi mentioned by Goodwin (1946), from the Island of Dominica, may be open to question. The specimen of P. davyi from Playas del Coco was caught, along with seven species of phyllostomatids and Desmodus, in a mist net set over a stream. The other four were all shot over roads at night, the three from the road to San Mateo while they were flying low hunting courses along the roadside vegetation. Also shot with the latter were Peropteryx macrotis and Eptesicus furinalis gaumeri. Pteronotus sucspurensis suctpurensis (J. A. Allen) Dermonotus suapurensis J. A. Allen, 1904, Bull. Amer. Mus. Nat. Hist., 20: 229. Pteronotus suapurensis, J. A. Allen, 1911, Bull. Amer. Mus. Nat. Hist., 30: 265. Pteronotus davyi suapurensis, G. M. Allen, 1935, Jour. Mamm., 16: 227. Pteronotus suapurensis centralis Goodwin, 1942, Jour. Mamm., 23: 88. 1968 Records of Bats From Costa Rica 9 Pteronotus suapurensis suapurensis, Davis, Carter and Pine, 1964, Jour. Mamm., 45: 377. Specimens-. PUNTARENAS: 35 miles SE San Isidro del General along Interamerican Highway, near bridge over Rio Ceibo, 320 m, 1 $ ; 2.7 miles NW Villa Neilly, at bridge over Rio Nuevo, 55 m, 3 $ $ , 3 $ $ . Measurements : Forearm (averages and ranges, 4 males, 3 females, re- spectively), 52.7 (51.9-53.6), 50.9 (50.5-52.3); (cranial, first two males, second and third females) greatest length of skull, 18.1, 17.7, 17.9, 16.9; condylobasal length, 16.9, 16.7, 16.7, 16.3; length of maxillary toothrow, C-M3, 7.8, 7.5, 7.5, 7.2; postorbital constriction, 4.4, 4.3, 4.1, 4.4; zygomatic breadth, 10.6, 10.2, 10.1, ; breadth of braincase, 9.0, 8.5, 8.6, 8.5; mastoid breadth, 9.4, 9.4, , 9.8; alveolar breadth across upper canines, 6.0, 5.7, 5.4, 5.8; breadth across M3-M3, 7.3, 7.0, 6.9, 7.0; length of mandibu- lar toothrow, C-M3, 8.3, 7.9, 7.9, 7.7; length of mandible, 14.1, 13.6, 13.5, 13.1. Remarks : This species, which is known from a limited number of localities from Veracruz, Mexico (Davis et al., 1964), through Central America into northern South America (Cabrera, 1957; Hall and Kelson, 1959), has been recorded previously from Costa Rica from Boruca (562 m), Prov. Puntarenas (Carnegie Museum 1613, Sanborn, 1932). The nearest neighboring records are from one locality in Nicaragua (Goodwin, 1942) and six in Panama (Handley, 1966). Our specimens come from two localities which are both within 60 km of Boruca: all were shot over roads at night, the specimen from Rio Ceibo with Peropteryx macrotis and those from Rio Nuevo with Saccop- teryx bilineata and Diclidurus virgo. Chilonycteris psifotis Dobson Chilonycteris psilotis Dobson, 1878, Cat. Chiroptera British Mus., p. 451. Chilonycteris torrei continentis Sanborn, 1938, Occ. Pap. Mus. Zool., Univ. Mich., No. 373, p. 1. Specimens : GUANACASTE: 4.4 miles S Liberia, 125 m, 1 $ . PUNTA- RENAS: Rincon de Osa, Osa Productos Forestales, Camp Seattle, 35 m, 1 $ . Measurements : Forearm 45.4, 47.0; greatest length of skull, , 15.6; condylobasal length, , 14.2; length of maxillary toothrow, C-M3, 6.2, 6.2; postorbital constriction, 3.6, 3.5; zygomatic breadth, 8.5, 8.5; breadth of braincase, 7.2, 7.4; mastoid breadth, 8.3, 8.7; alveolar breadth across upper canines, 4.4, 4.4; breadth across M3-M3, 5.6, 5.8. Remarks : This species, which apparently occurs from tropical Mexico at least to Venezuela (De la Torre, 1955), has been recorded previously in Central America south of El Salvador (Burt and Stirton, 1961) only from localities in Panama (Handley, 1966). Goodwin and Greenhall (1961) con- sider this small Chilonycteris a subspecies of C. personata Wagner. However, at this point, without the opportunity to make a fuller study of the necessary comparative material, we are following De la Torre (1955) in considering 10 Contributions in Science No. 148 C. psilotis Dobson (type locality: Oaxaca, Mexico) specifically distinct from C. personata Wagner (type locality: Matto Grosso, Brazil) and assigning the Costa Rican specimens to the former. We prefer to retain the generic distinction between the “naked-backed” species (Pteronotus) and those with the wing membranes attached farther down on the sides of the body (Chilonycteris). The distinction is clear and definitive, in spite of the fact that there is some variation in this character among the species of Chilonycteris, as pointed out by Burt and Stirton (1961 ) in their discussion which argued in favor of considering all the species of the two genera as congeneric, under the name Pteronotus. Both LACM specimens were shot over the road at night, the one from Rincon de Osa while it was hunting relatively low over the ground and vegetation next to a small swamp. Micronycteris schmidtorum Sanborn Micronycteris schmidtorum Sanborn, 1935, Zool. Ser., Field Mus. Nat. Hist., 20: 81. Specimens : GUANACASTE: 3 km S Playas del Coco, 70 m, 1 $ , 5 $ $ ; 9 km N Liberia along Interamerican Highway and 4 km E Interamerican High- way, 150 m, 1 $ . Measurements : Forearm (2 males, average and range of 5 females, re- spectively), 33.1, 34.0, 35.1 (34.6-35.9); (cranial, 2 males, third and fifth females) greatest length of skull, 20.4, 19.8, 19.8, 20.0; condylobasal length, 17.1, 17.0, 17.7, 17.5; length of maxillary toothrow, C-M3, 7.7, 7.8, 7.6, 7.8; postorbital constriction, 4.2, 4.3, 4.3, 4.4; zygomatic breadth, 9.3, 9.5, 9.1, 9.4; breadth of braincase, 8.2, 7.9, 7.9, 8.1; mastoid breadth, 9.2, 9.0, 9.0, 8.8; alveolar breadth across upper canines, 3.3, 3.3, 3.4, 3.2; breadth across M3-M3, 6.4, 6.1, 6.2, 6.1; length of mandibular toothrow, C-M3, 8.2, 8.4, 8.3, 8.3; greatest length of lower premolar series, 3.0, 3.0, 3.1, 3.0. Remarks’. These specimens mark the first Costa Rican records for this bat, which was previously known from four localities from Yucatan, Mexico, to southeastern Nicaragua (Hall and Kelson, 1959: Davis et al., 1964), and one locality in Panama (Handley, 1966). The bats from near Playas del Coco were all taken from one hollow tree (on two successive days) ; that from near Liberia was captured, along with five other genera of phyllostomatids, in a mist net set across a small stream. Macrophyllum macrophyllum (Schinz) Phyllost [ oma ] macrophyllum Schinz, 1821, Das Thierreich , 1 : 163. Macrophyllum macrophyllum, Nelson, 1912, Proc. Biol. Soc. Wash., 25: 93. Specimens’. GUANACASTE: Rio Colorado, 9 km N Liberia along Inter- american Highway and 4 km E Interamerican Highway, 150 m, 5 $ $ (1 imm. ) . LIMON : Finca La Lola, Rio Madre de Dios, 50 m, 1 $ . 1968 Records of Bats From Costa Rica 11 Measurements : Female, Rio Madre de Dios (LACM 15147), male, Rio Colorado (LACM 23365), respectively: forearm 35.3, 34.4; greatest length of skull, 16.9, 17.2; condylobasal length, 13.8, 14.2; length of maxillary toothrow, C-M3, 5.3, 5.6; postorbital constriction, 3.3, 3.2; zygomatic breadth, 9.4, 9.6; mastoid breadth, 8.7, 9.0; alveolar breadth across upper canines, 3.5, 3.4; breadth across M3-M3, 6.4, 6.3; length of mandibular toothrow, 5.9, 6.0. Remarks : These specimens appear to be the first records of Macrophyllum from Costa Rica. Previous published records for Central America, including localities in Nicaragua and Panama, are discussed by Davis et al. (1964) , Jones (1964) , and Handley ( 1966) . The female from La Lola was caught early in the evening in a mist net set across the Rio Madre de Dios. The five males from the Rio Colorado were taken from a diurnal roost in an irrigation tunnel, along with numerous Carollia sp. and Chilonycteris parnellii. Trachops cirrhosus (Spix) Vampyrus cirrhosus Spix, 1823, Simiarum et vespertilionum Braziliensum. . . ., p. 64. Trachops fuliginosus Gray, 1865, Proc. Zool. Soc. Lond., p. 14. Specimens : PUNTARENAS: Boca de Barranca, 5 m, 9 $$, 8 9$; Dominical, 3 m, 5 $ $ , 5 99;5 km N Villa Neilly on Interamerican Highway, 150 m, 1 $ ; Monteverde, Rio Guacimal, 1380 m, 1 $ . Measurements : Two females, averages and ranges for 4 males, respec- tively: forearm, 59.7, 61.1, 59.1 (57.3-61.7); greatest length of skull, 28.1, 27.7, 28.5 (28.3-29.1); condylobasal length, , 24.3, 24.7 (24.6-24.9); length of maxillary toothrow, C-M3, 9.9, 9.8, 10.1 (10.1-10.2); postorbital constriction, 5.3, 5.2, 5.4 (5. 1-5. 3); zygomatic breadth, 13.5, 13.6, 13.7 (13.6-13.9); mastoid breadth, 13.0, 13.2, 13.3 (13.1-13.4); alveolar breadth across upper canines, 11.1, 11.1, 11.4 (11.1-11.9); breadth across M3-M3, 9.5, 9.7, 9.8 ( 9. 7-9.9 ); length of mandibular toothrow, 10.8, 10.8, 11.0 (10.8- 11.2). Remarks : This species has been reported previously from only one locality in Costa Rica: Rio Tenorio, 3 miles S and 10 miles W Las Canas, 3 m, Prov. Guanacaste (Davis and Carter, 1962). The next closest localities are in Hon- duras (Goodwin, 1942) and Panama (Handley, 1966). Felten (1956b) con- siders Trachops coffini Goldman to be a subspecies of T. cirrhosus and uses the name T. cirrhosus coffini for specimens from El Salvador, as do also Burt and Stirton (1961). Handley (1966) lists the Trachops from Panama as T. c. cirrhosus. Examination of further comparative material will be necessary in order to clarify the subspecies question in this genus and determine which name should be used for specimens from Costa Rica. The specimens from Boca de Barranca were taken during the day from an abandoned railroad tunnel, on four different dates from July, 1962, to late June, 1963. Those from Dominical and near Villa Neilly were caught in 12 Contributions in Science No. 148 culverts in daytime. The Monteverde specimen was captured in a mist net set across a narrow stream gorge. Chrotopterus auritus Peters Vampyrus auritus Peters, 1856, Abhandl. preuss. Akad. Wiss., Berlin, p. 305. Chrotopterus auritus, Peters, 1865, Monatsb. preuss. Akad. Wiss., Berlin, p. 505. Specimens : LIMON: Finca La Lola, 50 m, 7 $ $ (1 imm.), 1 9. Measurements : Female, 3 males, respectively: forearm, 83.0, 79.3, 79.7, 79.8; greatest length of skull, 38.5, 36.9, 37.4, 37.9; condylobasal length, 32.9, 31.4, 32.0, 32.4; length of maxillary toothrow, C-M3, 13.9, 13.5, 13.5, 13.8; postorbital constriction, 15.9, 15.7, 16.1, 16.0; zygomatic breadth, 19.5, 18.4, 19.3, 18.9; mastoid breadth, 17.1, 16.6, 16.9, 16.7; alveolar breadth across upper canines, 17.8, 17.2, 17.8, 17.4; breadth across M3-M3, 12.3, 11.5, 12.2, 12.0; length of mandibular toothrow, 15.9, 15.1, 15.6, 15.5. Remarks'. These specimens represent the first Costa Rican records for this species. Chrotopterus has been reported previously from Mexico (Hall and Kelson, 1959; Davis et al., 1964), El Salvador (Burt and Stirton, 1961), Guatemala (Carter et al, 1966), Panama (Handley, 1966), and South America (Cabrera, 1957). The first seven specimens were taken from a diurnal roost in a large hollow tree (July 30, 1961). When the tree was revisited in September, 1962, a large colony of vampires ( Desmodus rotundus ) were occupying the tree and only one male Chrotopterus was present. Vampyrodes major G. M. Allen Vampyrodes major G. M. Allen, 1908, Bull. Mus. Comp. Zool., 52: 38. Vampyrodes ornatus Thomas, 1924, Ann. Mag. Nat. Hist., ser. 9, 13: 532. Vampyrodes caraccioloi ornatus, Cabrera, 1957, Rev. Mus. Argent. Cien. Nat. “Bernardino Rivadavia,” Cien. Zool., 4: 82. Vampyrodes carracioloi major, Carter, Pine and Davis, 1966, Southwest. Nat., 1 1 : 494. Vampyrodes caraccioloi major, Handley, 1966, p. 766, in Wenzel and Tipton, Ectoparasites of Panama. Specimens : HEREDIA: Puerto Viejo de Sarapiqui, 100 m, 1 $ (KU 88153). LIMON: Finca La Lola, 50 m, 1 9 ; Los Diamantes, 100 m, 3 $ $ , 7 9 9. PUNTARENAS: Rincon de Osa, Osa Productos Forestales, near Rio Rincon, 5 m, 1 9 . Measurements'. (Averages and ranges): forearm (3 males, 9 females), 53.1 (52.2-54.4); (cranial, 3 males, 2 females) greatest length of skull, 27.9 (27.4-28.4); condylobasal length, 24.5 (24.4-24.7); length of maxillary toothrow, C-M2, 9.9 (9.8-10.1); postorbital constriction, 6.8 (6. 6-7.0); zygo- matic breadth, 17.2 (16.7-17.8); breadth of braincase, 12.0 (11.7-12.2); 1968 Records of Bats From Costa Rica 13 mastoid breadth, 13.9 (13.7-14.3); alveolar breadth across upper canines, 6.9 (6. 8-7. 3) ; breadth across M2-M2, 12.4 (12.0-12.6). Remarks : V ampyrodes major has been recorded previously in Costa Rica only from Cerro Santa Maria, Prov. Alajuela (Goldman, 1920). This bat is apparently not uncommon at lower elevations on both coasts, although it seems to prefer the wetter portions of the country. Our specimens were all captured in mist nets. From a comparison of specimens, published data and the original de- scriptions of Vampyrodes major, V. ornatus Thomas, and V. caracciolae Thomas, it seems that there is no reason for retaining V. ornatus as a species distinct from V. major nor, in fact, for even distinguishing the two subspecifi- cally (in agreement with Handley, 1966). The measurements of V. ornatus, as well as its salient characters, fit in with those of V. major (including the type and LACM series of the latter from Costa Rica and Mexico) and there seems to be no break in these features anywhere in the ranges of localities represented by the two names. V. caracciolae, on the other hand, appears to be a distinct, smaller, and less brightly colored species which deserves specific status (con- trary to the opinions of Cabrera, 1957, Goodwin and Greenhall, 1961, and Handley, 1966). Two obvious dental characters separate V. major from V. caracciolae : the former has distinctly bifid and rounded outer upper incisors while in V. caracciolae these teeth are entire with a more flattened profile (a character illustrated by Thomas, 1889, and by Goodwin and Greenhall, 1961, but otherwise apparently overlooked); the M1 of V. major has a distinct cusplet arising from the anterolingual portion of the cingulum, whereas this structure is absent in V. caracciolae. The small size of the male V ampyrodes from Utinga, Para, Brazil, mentioned by Thomas (1920), suggests that it also belongs to the latter species rather than to V. major. Since the specimen from Para apparently was returned to Brazil, Thomas’ inclusion of it in V. carac- ciolae will have to remain as the standing opinion on this matter until the dental characters can be checked. V. major is currently known from Peru (San Lorenzo, Depto. Loreta: type locality of V. ornatus) and most of the countries in Middle American from Panama to Mexico (summarized by Davis et al., 1964, and Handley, 1966), and has a distribution quite separate from that of V. caracciolae. Further collecting in the intermediate areas in South America may bring the two ranges closer together. EcfophyHa a!ba H. Allen Ectophylla alba H. Allen, 1892, Proc. U.S. Nat. Mus., 15: 441. Specimens : CARTAGO: Turrialba, Instituto Interamericano de Ciencias Agricolas, 602 m, 1 $ (ISU 230), 1 (?) (IICA). HEREDIA: Finca “La Selva,” near Puerto Viejo de Sarapiqui, 100 m, 2 $ $ . Measurements'. Forearm (male, 2 females), 28.1 (dry), 28.6, 29.5; (cranial, male) greatest length of skull, 17.0; condylobasal length, 15.8; length of maxillary toothrow, C-M2, 6.1; zygomatic breadth, 10.6; breadth of brain- 14 Contributions in Science No. 148 case, 6.2; mastoid breadth, 8.8; alveolar breadth across upper canines, 4.3; breadth across M2-M2, 7.6; length of mandibular toothrow, C-M2, 6.6. Remarks : The Turrialba specimens represent the third locality for the little white bat in Costa Rica, extending its range some 65 km to the South- southeast from Finca “La Selva” from which three specimens have previously been reported (Casebeer et al., 1963: see also for a summary and discussion of earlier records). This species is known only from the Caribbean lowlands of Nicaragua, Costa Rica and Panama (Handley, 1966), ranging from an un- known elevation (undoubtedly well below 300 m) at the type locality on the Rio Coco (Rio Segovia), Nicaragua, to approximately 602 m at Turrialba and 732 m in Panama (Handley, 1966). All specimens of this bat were caught in mist nets. Mesophylla macconnelii Thomas Mesophylla macconnelii Thomas, 1901, Ann. Mag. Nat. Hist., ser. 7, 8: 143. Ectophylla macconnelii, Laurie, 1955, Ann. Mag. Nat. Hist., ser. 12, 8: 269. E[ctophylla] m[acconnelli\ macconnelii, Goodwin and Greenhall, 1962, Amer. Mus. Nov., no. 2080, p. 3. Ectophylla macconnelii flavescens Goodwin and Greenhall, 1962, Amer. Mus. Nov., no. 2080, p. 2. Ectophylla macconnelii, Handley, 1966, p. 768, in Wenzel and Tipton, Ecto- parasites of Panama. Specimens : CARTAGO: Moravia de Chirripo, 1116 m, 1 $ (UMMZ 111724). Measurements : Forearm, 33.0; greatest length of skull, 18.5; condy- lobasal length, 16.7; length of maxillary toothrow, C-M2, 6.3; postorbital con- striction, 4.5; zygomatic breadth, 10.4; breadth of braincase, 8.1; mastoid breadth, 9.2; alveolar breadth across upper canines, 4.4; breadth across M2-M2, 7.7; length of mandibular toothrow, 7.1. Remarks : This bat has been reported previously from several localities in Panama, widespread localities in tropical South America, and from Trinidad (Handley, 1966; Cabrera, 1957; Goodwin and Greenhall, 1962). The speci- men from Costa Rica is much grayer in color than LACM specimens of M. macconnelii from Peru and Brazil, and is apparently even grayer than M. m. flavescens (Goodwin and Greenhall) from Trinidad. Its measurements are generally larger than those of M. m. macconnelii listed by Goodwin and Greenhall (1962) and approach those of M. m. flavescens. Handley (1966) considers the species to be monotypic, on the basis of variability in Pan- amanian specimens, and we follow Handley’s judgment in this matter at this time. We disagree with Laurie (1955) and Goodwin and Greenhall (1962), and consider Mesophylla Thomas to be a genus distinct from Ectophylla H. Allen in several prominent cranial, as well as dental, features. The genera Ectophylla, Mesophylla and V ampyressa Thomas form a well-defined group 1968 Records of Bats From Costa Rica 15 within the Stenoderminae, sharing cranial, dental and external characters, and showing varying degrees of modification of a similar dentitional pattern. The skull of Vampyressa pusilla thyone shows more similarity to that of Meso- phylla than does the skull of Ectophylla, and exhibits dental features seen in both the other genera. Ectophylla, with its shortened rostrum, raised nasals, and exaggerated circular and flattened last lower molar (M2), appears to be the most highly modified of the three genera, and we see no justification for making it congeneric with Mesophylla. Natalus stramineus mexicanus Miller Natalus mexicanus Miller, Proc. Acad. Nat. Sci. Philadelphia, 54: 399. Natalus mexicanus mexicanus, Dalquest and Hall, 1949, Proc. Biol. Soc. Wash., 62: 153. Natalus mexicanus saturatus Dalquest and Hall, 1949, Proc. Biol. Soc. Wash., 62: 153. Natalus stramineus saturatus, Goodwin, 1959, Amer. Mus. Nov., no. 1977, p. 7. Natalus stramineus mexicanus, Handley, 1966, p. 770, in Wenzel and Tipton, Ectoparasites of Panama. Specimens : GUANACASTE: Curiol de Santa Rosa, 25 m, 3 $ $ . Measurements : Forearm, 36.9, 36.4, 36.4; (cranial, second specimen) greatest length of skull, 16.4; condylobasal length, 15.0; length of maxillary toothrow, C-M3, 6.9; postorbital constriction, 3.1; zygomatic breadth, 8.4; breadth of braincase, 7.8; alveolar breadth across upper canines, 3.6; breadth across M3-M3, 5.5. Remarks : These specimens mark the first records for this genus in Costa Rica. The species (following Cabrera, 1957, and Goodwin, 1959) ranges from tropical Mexico into eastern South America (to Brazil), but until now it has not been recorded from any country between El Salvador and Panama (Hall and Kelson, 1959; Handley, 1966). The Natalus from Costa Rica were caught during the day, in a mist net stretched across the entrance to an abandoned tungsten mine, as they attempted to escape capture in the mine tunnel. Also taken in the same tunnel were Saccopteryx bilineata (near the entrance to a small side opening), Chilonyc- teris parnellii, Carollia subrufa, Glossophaga soricina, and Desmodus. Myotis simus riparius Handley Myotis simus riparius Handley, 1960, Proc. U.S. Nat. Mus., 112: 466. Specimen : CARTAGO: 4 km SW Moravia de Chirripo, 1200 m, 1 $ (KU 88207). Measurements : Forearm, 40.4; greatest length of skull, 14.8; condy- lobasal length, 13.8; length of maxillary toothrow, C-M3, 5.7; postorbital con- striction, 3.5; zygomatic breadth, 8.9; breadth of braincase, 6.7; alveolar breadth across upper canines, 3.7; breadth across M3-M3, 5.8. 16 Contributions in Science No. 148 Remarks'. This specimen marks the first record of Myotis simus for Costa Rica and extends the range of the subspecies into Central America from eastern Panama (Handley, 1966). This same specimen was recently reported as Myotis albescens by Nelson (1965). Myotis chiloensis (Waterhouse) Vespertilio chiloensis Waterhouse, 1838, Zool. Voy. Beagle, p. 5, pi. 3. Myotis chiloensis, Cabrera, 1903, Rev. Chilena Hist. Nat., 7: 295. Specimen : SAN JOSE: Cerro de la Muerte, on Interamerican Highway, 4.3 miles S restaurant “La Georgina,” 2590 m, 1 $ (LACM 25445) . Measurements’. Forearm, 39.5; greatest length of skull, 14.4; condylobasal length, 13.4; length of maxillary toothrow, C-M3, 5.5; postorbital constriction, 3.9; zygomatic breadth, 8.7; breadth of braincase, 6.9; alveolar breadth across upper canines, 3.6; breadth across M3-M3, 5.7. Remarks'. This species is here recorded for the first time north of Panama (Handley, 1966). The bat was shot at night while it was hawking insects over the highway. This specimen is listed without subspecific designation due to the lack of a sufficiently large series for proper comparison with South American forms (see further Handley, 1966). Eptesicus furincdis qaumeri (J. A. Allen) Adelonycteris gaumeri J. A. Allen, 1897, Bull. Amer. Mus. Nat. Hist., 9: 231. Eptesicus albigularis, Goodwin, 1942, Amer. Mus. Novitates, No. 1199, p. 1 (not Vesperus ( Marsipolaemus ) albigularis Peters, 1872). Eptesicus brasiliensis propinquus, Dalquest, 1953, Louisiana State Univ., Bio. Sci. Ser., 1: 55 (not Vesperus propinquus Peters, 1872). Eptesicus gaumeri gaumeri, Davis, 1965, Jour. Mamni., 46: 233. Eptesicus furinalis gaumeri, Davis, 1966, Southwest. Nat., 11: 268. Specimens: ALAJUELA: 8.4 miles from bridge at Atenas on road to San Mateo, 671 m, 1 $ . LIMON : Finca La Lola, 50 m, 1 $ ; Tortuguero, sea level, 1 $ (UF 8025). PUNTARENAS: Boca de Barranca, sea level, 2 $ $ , 3 $ $ . Measurements: (Averages and ranges): forearm (5 males, 3 females), 38.8 (37.8-39.6); (cranial, 3 males, 3 females) greatest length of skull, 15.7 (15.6-15.9); condylobasal length, 14.4 (14.1-14.8); length of maxillary toothrow, C-M3, 5.6 (5. 3-5. 9); postorbital constriction, 3.8 (3. 6-3. 9); zygo- matic breadth (2 males, one female, this measurement only), 10.5 (10.4- 10.6); breadth of braincase, 7.3 (7. 1-7. 5); mastoid breadth, 8.2 (8. 0-8. 4); alveolar breadth across upper canines, 4.6 (4. 4-4. 9); breadth across M3-M3, 6.5 (6. 3-6. 6) . Remarks: Davis (1966) gives the range of this subspecies essentially as lowland Middle America and Caribbean South America, from Mexico to Surinam, and lists specimens (Davis, 1965) from every Middle American 1968 Records of Bats From Costa Rica 17 country except Belize and Costa Rica. The records reported here serve to fill in the distribution of this bat on both lowlands of Costa Rica. The specimens from Boca de Barranca were caught in mist nets set along the edge of a creek, along with Rhynchonycteris, Rhogeessa tumida, Molossus sp. and several species of phyllostomatids; the one from the road to San Mateo was shot over the road at night not far from the locality at which Peropteryx macrotis and Pteronotus davyi were collected. The flight of E. f. gaumeri is quite similar to that of Myotis nigricans. Eptesicus andinus J. A. Allen Eptesicus andinus J . A. Allen, 1914, Bull. Amer. Mus. Nat. Hist., 33: 382. Eptesicus chiriquinus Thomas, 1920, Ann. Mag. Nat. Hist., ser. 9, 5: 362. Eptesicus brasiliensis andinus, Hershkovitz, 1949, Proc. U.S. Nat. Mus., 99: 451. Eptesicus andinus andinus, Davis, 1965, Jour. Mamm., 46: 239. Eptesicus andinus, Davis, 1966, Southwest. Nat., 11: 252. Specimens : SAN JOSE: San Jose, campus Universidad de Costa Rica, 1200 m, 1 $ ; Candelaria, elevation uncertain, 1 $ (MCZ 20445) ; Cerro de la Muerte, S restaurant “La Georgina” along Interamerican Highway, 4.2 miles, 2743 m, 2 $ $, 6.0 miles, 2590 m, 1 $, 6.8 miles, 2500 m, 2 $ $, and 17.5 miles, 1525 m, 1 $ ; San Isidro del General, 705 m, 1 $ . Measurements : (Averages and ranges) forearm (5 males, 3 females), 44.2 (42.8-46.4); greatest length of skull (6 specimens), 17.0 (16.7-17.3); condylobasal length (8), 16.0 (15.5-16.4); length of maxillary toothrow, C-M3 (8), 6.2 (5. 9-6.5); postorbital constriction (8), 4.1 (3. 8-4.5); zygo- matic breadth (7), 11.2 ( 10.9-1 1.6) ; breadth of braincase (7), 7.8 (7.7-8.1); mastoid breadth (7), 8.9 (8.4-9. 2) ; alveolar breadth across upper canines (8), 5.0 (4.8-5. 3); breadth across M3-M3 (8), 7.9 (6.9-7. 3). Remarks : The LACM specimens have been cited by Davis in his review of South American Eptesicus (1966); the one from Candelaria (MCZ) rep- resents a new record. Only one locality north of Costa Rica is reported, based on one specimen from Chiapas, Mexico (Davis, 1966). The animals from Cerro de la Muerte were all shot over the highway at night; that from San Jose (previously reported as E. chiriquinus by Tamsitt and Valdivieso, 1961) was taken in a mist net set across a small stream. Also captured in the same net with the latter specimen were Eptesicus fuscus mira- dorensis, Myotis nigricans, and several species of phyllostomatids. The E. andi- nus from Candelaria was part of a series of bats collected by C. F. Underwood, presumably all from the same locality, which included also an adult male E. fuscus miradorensis and a number of subadults and juveniles of the latter species. Rhogeesa tumida tumida H. Allen R [ hogeesa ] tumida H, Allen, 1866, Proc. Acad. Nat. Sci. Phila, 18: 285. 18 Contributions in Science No. 148 Rhogeessa parvula tumida, Hall, 1952, Univ. Kans. Publ., Mus. Nat. Hist., 5: 231. Rhogeessa tumida tumida, Goodwin, 1958, Amer. Mus. Nov., No. 1923: 3. Specimens'. GUANACASTE: Samara, sea level, 2 9 $ ;PUNTARENAS: Boca de Barranca, sea level, 1 $ ; Hacienda Bonilla (near Boca de Barranca), sea level, 1 9 . Measurements'. Forearm (male, 3 females, respectively), 29.2, 29.2, 29.3, 29.4; (cranial, male, first 2 females) greatest length of skull, 12.7, 12.8, 12.5; condylobasal length, 11.6, 11.4, 11.6; length of maxillary toothrow, C-M3, 4.4, 4.4, 4.4; postorbital constriction, 3.2, 3.4, 3.1; zygomatic breadth, , 8.1, ; breadth of braincase, 5.9, 6.0, 5.8; mastoid breadth, 6.7, 6.8, 6.8; alveolar breadth across upper canines, 3.7, 3.6, 3.6; breadth across M3-M3, 5.2, 5.4, 5.3. Remarks'. This tiny bat is known from Costa Rica from Miravalles, Prov. Guanacaste, (Sanborn, 1932), and “Pacific Coast” (Goodwin, 1946), in addi- tion to the two localities recorded here. It has been taken in most countries of Central America (Goodwin, 1958; Handley, 1966). Our specimens are in agreement with the description and measurements of R. t. tumida published by Goodwin (1958). All four were caught in mist nets in association with several species of phyllostomatids. Lasiurus ega panamensis (Thomas) Dasypterus ega panamensis Thomas, 1901, Ann. Mag. Nat. Hist., ser. 7, 8: 246. Lasiurus ega panamensis, Handley, 1960, Proc. U.S. Nat. Mus., 112: 474. Specimens'. PUNTARENAS: Interamerican Highway, near bridge over Rio Puerto Nuevo, 90 m, 1 9 , and near bridge over Rio Catarata, 180 m, 1 $ . Measurements : (Male, female, respectively) forearm, 44.5, 45.9; great- est length of skull, 15.8, 16.2; condylobasal length, 14.5, 15.4; length of maxillary toothrow, C-M3, 5.5, 5.6; postorbital constriction, 4.6, 4.4; zygo- matic breadth, 10.3, 10.8; breadth of braincase, 8.4, 8.2; alveolar breadth across upper canines, 5.9, 6.0; breadth across M3-M3, 7.1, 7.4. Remarks'. The yellow bat is poorly represented from Central America. Only two specimens, both males, have been reported from Costa Rica, one from San Jose, Prov. San Jose, and the second from “Lajas Villa Quesada,” Prov. Alajuela (Goodwin, 1946). The two individuals recorded here were both shot at night over the road, where they passed as they flew hunting pat- terns along the edge of the gorge of the Rio Terraba. These bats were extremely rapid and maneuverable and they made considerable effort to stay out of our lights. In two nights of shooting at a fair number of yellow bats we were able to secure only these two specimens. Acknowledgments The specimens recorded in this paper were collected mainly by various field parties working with the support of grant funds (field years and principal 1968 Records of Bats From Costa Rica 19 investigators) : National Academy of Sciences, Bache Fund, no. 465 (1961 — Starrett); National Science Foundation, no. G-18002 (1961 — Starrett) and no. G-6089 (1959 — J. M. Savage and Starrett); U.S. Army Medical Research and Development Command, nos. DA-MD-49-183-62-G54 and DA-MD-49- 193-63-G94 (1962-1964 — F. S. Truxal and C. A. McLaughlin, Jr.). These field parties included, in addition to the authors, respectively, Ronald B. Linsky; Arden H. Brame, Arnold G. Kluge, Robert J. Lavenberg; the late Jorge Perez; Harry N. Coulombe, Charles L. Hogue, Anthony G. Hollister, Julius E. Geest, David Marqua, Charles A. McLaughlin, Jr., Allan A. Schoenherr, Robert Stephens and Fred S. Truxal. Others who have contributed field time and/ or specimens include Keith Arnold, Edgar Arrieta, William A. Bussing, John V. Dennis, Jorge Hernandez-Camacho, Roy W. McDiarmid, Jay M. Savage, Norman J. Scott, James R. Tamsitt, Fred G. Thompson, Dario Valdivieso, James L. Vial and Alvaro Wille. Leslie R. Holdridge and Robert J. Hunter generously allowed members of various parties to use facilities and collect specimens on their tineas “La Selva,” near Puerto Viejo de Sarapiqui. Similar courtesies and assistance have also been extended by Albert Wright, director, Walt Strain, foreman, and Gus Zambrano, at Osa Productos Forestales, Rincon de Osa; Don Alfredo Paredes, Finca La Lola; Gerardo Budowski, Instituto Interamericano de Ciencias Agricolas, Turrialba (IICA) ; Rex Benson, Gromaco Foundation; William and Daryl Cole, Quaker Colony, Monteverde; the various members of the staff at Los Diamantes experimental station; and Milton Lopez, Ministerio de Salubri- dad Publica. John de Abate and Rafael L. Rodriguez, Universidad de Costa Rica, on many occasions gave of their time to provide advice and help. Loans of specimens, permission to publish records, and courtesies during visits to various museums were extended by Dale E. Birkenholz, Illinois State University, Normal (ISU) ; John H. Kaufmann, University of Florida (UF); William B. Davis and Dilford C. Carter, Texas A. and M. University (TCWC) ; Barbara Lawrence and Charles W. Mack, Museum of Comparative Zoology, Harvard (MCZ) ; William H. Burt and Emmet T. Hooper, Museum of Zoology, University of Michigan (UMMZ); E. Raymond Hall and J. Knox Jones, Jr., Museum of Natural History, University of Kansas (KU); Richard G. Van Gelder, American Museum of Natural History; Joseph C. Moore, Field Museum of Natural History, Chicago; Charles O. Handley, Jr., U.S. National Museum; J. Kenneth Doutt, Carnegie Museum. It is a pleasure to acknowledge the time and assistance provided by all of these people and institutions, as well as others whose names may have been omitted inadvertently. So many people have contributed in one way or an- other, particularly in Costa Rica, that it is not possible to provide a complete list. Nonetheless, we do here express our appreciation to all. 20 Contributions in Science No. 148 Literature Cited Allen, G. M., and Thomas Barbour. 1923. Mammals from Darien. Mus. Comp. Zool., Harvard, Bull., 65 : 259-274. Allen, J. A. 1894. Remarks on specimens of Chilonycteris rubiginosus from west- ern Mexico, and on the color phases of Pteronotus davyi Gray. Amer. Mus. Nat. Hist., Bull., 6: 247-248. . 1897. Additional notes on Costa Rican mammals, with descriptions of new species. Amer. Mus. Nat. Hist., Bull., 9: 31-44. Burt, W. H., and R. A. Stirton. 1961. The mammals of El Salvador. Mus. Zool., Univ. Mich., Misc. Pub., No. 117: 1-69. Cabrera, Angel. 1957. Catalogo de los mamiferos de America del Sur. I. Rev. Mus. Argent. Cienc. Nat. “Bernardino Rivadavia” Cienc. Zool., 4: 1-307. Carter, D. C., R. H. Pine, and W. B. Davis. 1966. Notes on Middle American bats. Southwest. Nat., 11: 488-499. Casebeer, R. S., R. B. Linsky, and C. E. Nelson. 1963. The phyllostomid bats, Ectophylla alba and Vampyrum spectrum, in Costa Rica. J. Mammal., 44: 186- 189. Davis, W. B. 1965. Review of the Eptesicus brasiliensis complex in Middle America with the description of a new subspecies from Costa Rica. J. Mammal., 46: 229- 240. . 1966. Review of South American bats of the genus Eptesicus. Southwest. Nat., 11: 245-274. , and D. C. Carter. 1962. Notes on Central American bats with description of a new subspecies of Mormoops. Southwest. Nat., 7 : 64-74. , D. C. Carter, and R. H. Pine. 1964. Noteworthy records of Mexican and Central American Bats. J. Mammal., 45: 375-387. Dobson, G. E. 1878. Catalogue of the Chiroptera in the collection of the British Museum. British Mus. (Nat. Hist.), London, 576 p. Felten, Heinz. 1955. Fledermause (Mammalia, Chiroptera) aus El Salvador. Teil 1. Senckenb. Biol., 36: 271-285. . 1956a. Fledermause (Mammalia, Chiroptera) aus El Salvador. Teil 2. Senckenb. Biol., 37: 69-86. . 1956b. Fledermause (Mammalia, Chiroptera) aus El Salvador. Teil 3. Senckenb. Biol., 37: 179-212. Goldman, E. A. 1920. Mammals of Panama. Smithson. Misc. Coll., 69(5): 1-309. Goodwin, G. G. 1942. New Pteronotus from Nicaragua. J. Mammal., 23: 88. . 1946. Mammals of Costa Rica. Amer. Mus. Nat. Hist., Bull., 87: 275-473. . 1958. Bats of the genus Rhogeessa. Amer. Mus. Nov., No. 1923: 1-17. • . 1959. Bats of the subgenus Natalus. Amer. Mus. Nov., No. 1977: 1-22. , and A. M. Greenhall. 1961. A review of the bats of Trinidad and Tobago. Amer. Mus. Nat. Hist., Bull., 122: 187-302. , . 1962. Two new bats from Trinidad, with comments on the status of the genus Mesophylla. Amer. Mus. Nov., No. 2080: 1-18. 1968 Records of Bats From Costa Rica 21 Hall, E. R., and K. R. Kelson. 1959. The mammals of North America. Vol. I. Ronald Press, N. Y. 546 p. Handley, C. O., Jr. 1966. Checklist of the mammals of Panama, pp. 753-795. In Wenzel, R. L., and V. J. Tipton, Ectoparasites of Panama. Field Mus. Nat. Hist., Chicago. 861 p. Jones, J. K., Jr. 1964. Bats new to the fauna of Nicaragua. Kans. Acad. Sci., Trans., 67: 506-508. Laurie, E. M. O. 1955. Notes on some mammals from Ecuador. Ann. Mag. Nat. Hist., ser. 12, 8: 268-276. Nelson, C. E. 1965. Lonchorhina aurita and other bats from Costa Rica. Texas J. Sci., 17: 303-306. Sanborn, C. C. 1932. Neotropical bats in the Carnegie Museum. Ann. Carnegie Mus., 21: 171-183. . 1937. American bats of the subfamily Emballonurinae. Field Mus. Nat. Hist., Zool. Ser., 20: 321-354. Slud, P. 1964. The birds of Costa Rica. Amer. Mus. Nat. Hist., Bull., 128: 1-430. Starrett, Andrew and Luis de la Torre. 1964. Notes on a collection of bats from Central America, with the third record for Cyttarops alecto Thomas. Zoologica, 49: 53-63. Tamsitt, J. R., and D. Valdivieso. 1961. Notas sobre actividades nocturnas y estados de reproduction de algunos Quiropteros de Costa Rica. Rev. Biol. Trop., 9: 219-225. Thomas, Oldfield. 1889. Description of a new stenodermatous bat from Trinidad. Ann. Mag. Nat. Hist., ser. 6, 4: 167-170. . 1892. Note on Mexican examples of Chilonycteris davyi, Gray. Ann. Mag. Nat. Hist., ser. 6, 10: 410. . 1920. On mammals from the lower Amazons in the Goeldi Museum, Para. Ann. Mag. Nat. Hist., ser. 9, 6: 266-283. Torre, Luis de la. 1955. Bats from Guerrero, Jalisco and Oaxaca, Mexico. Field- iana: Zool., 37: 695-704. Walton, D. W. 1963. A collection of the bat Lonchophylla robusta Miller from Costa Rica. Tulane Stud. Zool., 10: 87-90. Accepted for publication April 3, 1968 LOS ANGELES CONTRIBUTIONS COUNTY ■ MUSEUM IN SCIENCE UMBER 149 June 30, 1968 5^-7, 73 C2Z, ecf PHOTONECTES MUNIF1CUS, A NEW SPECIES OF MELANOSTOMIATID FISH FROM THE SOUTH PACIFIC SUBTROPICAL CONVERGENCE, WITH REMARKS ON THE CONVERGENCE FAUNA By Robert H. Gibbs, Jr. Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8V2 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M, Halmos Editor PHOTONECTES MUNIFICUS, A NEW SPECIES OF MELANOSTOMIATID FISH FROM THE SOUTH PACIFIC SUBTROPICAL CONVERGENCE, WITH REMARKS ON THE CONVERGENCE FAUNA By Robert H. Gibbs, Jr.1 Abstract: Photonectes munificus, a new species in the sto- miatoid family Melanostomiatidae, is described from a single specimen from the Subtropical Convergence of the South Pacific. The holotype, 371 mm standard length, is the largest ever re- corded for the genus and is distinguished from other species by the highest IV and OV photophore counts (49-50 and 38), the lowest VAL count (2-4), and the greatest number of vertebrae (67). It most closely resembles species in the subgenus Trachi- nostomias, which have the dorsal and anal fins covered with black skin, and especially P. ( T .) margarita, which, like P. munificus, often lacks the pectoral fin. A distinct circumglobal Subtropical Convergence fauna is characterized by nine restricted species of stomiatoids, and P. munificus is attributed to it. A specimen of the genus Photonectes, collected in the Subtropical Con- vergence during cruise 24 of the USNS Eltanin, is the largest ever recorded for the genus and represents a previously undescribed species. Although there is confusion regarding the species of Photonectes, I have completed enough of a revisionary study to assure the validity of the new species. Photonectes munificus new species Fig. 1 Holotype — Los Angeles County Museum of Natural History (LACM) 11282-1; female, 371 mm standard length. USNS ELTANIN, USC Station 1777; 41 °50'S, 130°12'W to 41°46'S, 130°16'W; 14 August 1966. Ten-foot (3m) Isaacs-Kidd midwater trawl, surface to maximum depth of ca. 440 m. Local time 2050-2317 hours, two hours at depth. Diagnosis — Dorsal and anal fins covered with black skin; pectoral fin absent; IV photophores 49-50; OV 38; VAL 2-4; vertebrae 67. Description (most characters in Diagnosis not repeated) — Dorsal rays 19, the first small, the last split to the base; anal rays 21, the first tiny, the last split to the base (dorsal and anal rays counted from radiograph); pelvic rays 7. Serial photophores: IV 49 (left side), 50 (right side), those on isthmus not forming a separate group, the last three before the pelvics close together; VAV 12, anterior three and posterior four forming close groups; AC 12, the first four close together, next four more widely spaced, last four gradually becoming Curator, Division of Fishes, U.S. National Museum, Washington, D.C. 20560 1 Contributions in Science No. 149 1968 A New Species of Melanostomiatid Fish 3 smaller and closer together; OV 38; VAL 4 (left side), 2 (right side), widely spaced, the last well anterior to anal-fin origin. Branchiostegal photophores 9. Vomerine teeth, one on each side. Palatine teeth, 3 on each side. Gill teeth 9, single, on lower limb only. Measurements (in mm; percent of standard length in parentheses) — standard length 371; snout to dorsal origin 300.1 (80.7); snout to anal origin 313.1 (84.2) ; snout to pelvic insertion 258.3 (69.5) ; head length 49.1 (13.2); snout to anterior bony margin of orbit 8.9 (2.4) ; fleshy orbit length 7.0 ( 1.9) ; postocular organ, pale part only 4.0 (1.1), entire black surrounding area 6.9 (1.9); upper jaw length 43.3 (11.6); body depth behind head 48.8 (13.1); greatest body depth 77.4 (20.8) ; least caudal peduncle depth 8.2 (2.2) ; pelvic fin length 91.4 (24.6); length of dorsal fin base 51.4 (13.8); length of anal base 51.5 (13.9). Barbel with a small bulb, but broken distally; length to end of bulb 19.0 (5.1), length to broken end 22.3 (6.0). No luminous tissue in mouth, on jaws, or on body. Orangish eggs present, about 0.35 mm in diameter. Relationships — The new species, P. munificus, is obviously closely related only to those species included in the subgenus Trachinostomias Parr by Morrow (in Morrow and Gibbs, 1964), who revised the subgeneric classification of Regan and Trewavas (1930). These species — P. bifilifer Beebe, P. fimbria Regan and Trewavas, P. parvimana Regan and Trewavas, and P. margarita (Goode and Bean) — are characterized by dorsal and anal fins that are more or less completely covered with thick black skin, by the highest IV and OV photophore counts within the genus (total range 41-48 and 30-36 respectively) , and by the highest vertebral number (the highest so far counted, either in the literature or on my numerous radiographs, is 64). The previous maximum photophore and vertebral counts in the genus Photonectes are, thus, exceeded by several in P. munificus, while the VAL count of 2-4 is much lower than the previously recorded minimum of 10 for the genus. Because pectoral fins are absent in P. munificus, a relationship with species of the subgenera Photonectes Gunther and Dolichostomias Parr, which also lack pectorals, might be con- sidered. Pectoral fin development in the genus Photonectes, however, is weak in all species, and P. margarita (subgenus Trachinostomias ) may have one ray, ranging from long to very short, or may lack the ray on one or both sides. The sole known representative of the subgenus Dolichostomias, P. gracilis, is well- differentiated. It is generally more slender than other Photonectes, has the pelvic fins inserted before the mid-length, and has long dorsal and anal fin bases. I believe that P. munificus is closest to P. margarita. Distribution — The capture of P. munificus at a depth of 0-440 m in the Subtropical Convergence suggests that this new species is a member of the fauna of the Convergence, a fauna that I believe constitutes a distinct zoogeo- graphic region encircling the Southern Ocean. The same trawl collected Batho- philus ater (Brauer), another stomiatoid species of similar zoogeographic pro- 4 Contributions in Science No. 149 pensities. Brauer (1902) described B. ater from 26°49'S, 5°54'E, and M.A. Barnett and I have examined other specimens from between 27°10'S and 32°29'S, all in the southeastern Atlantic. Specimens of B. ater from cruises of the Eltanin show this species to occur all across the south Pacific from New Zealand to Chile at latitudes between 33° and 45° S. Very few species of stomiatoid fishes occur south of the limits of the great central water masses or the waters underlying these water masses and fewer still are restricted to the far southern waters. I am aware of 17 species that have the Subtropical Convergence as the southern boundary of a wide- spread distribution (Table 1). Stomias gracilis is the only stomiatoid that, to my knowledge, is found only in subantarctic and antarctic waters; it is replaced by S. boa boa in the Subtropical Convergence. Borostomias antarcticus, an antitropical species, and Idiacanthus atlanticus occur in the Convergence and in more southern waters as well. Nine species of stomiatoids appear restricted to the Subtropical Convergence (Table 1), ten if Photonectes munificus is included. But for a recent capture in the central Atlantic, Trigonolampa miriceps might also be considered an antitropical member of the Convergence fauna. Further collecting and study will doubtless confirm the reality of a Sub- tropical Convergence fauna. With regard to Photonectes munificus, I predict that future collecting will show this species to occur around the world in the Convergence and northward into the southern reaches of the Benguela and Humboldt currents. Etymology — The specific name munificus, is a Latin adjective meaning bountiful, used in reference to the high meristic counts of the species and the large body size of the holotype. Acknowledgments— -l thank Robert J. Lavenberg for making this speci- men available. The work of the USNS Eltanin is conducted under the auspices of the National Science Foundation. The present study is supported by NSF grant GB-3906 and by Smithsonian Research Award 3362. Literature Cited Barnard, K. H. 1948. Further notes on South African marine fishes. Ann. S. Afr. Mus. 36: 341-406. Brauer, A. 1906. Die Tiefseefische. 1. Systematischer Teil. In Wiss. Ergeb. Deutschen Tiefsee-Exped. Valdivia, Jena, 15 (1): 1-432. Bussing, W. A. 1965. Studies of the midwater fishes of the Peru-Chile Trench. Biol. Antarctic Seas II, Am. Geophys. Union, Antarctic Res. Ser. 5: 185-227. Ege, V. 1934. The genus Stomias Cuv., taxonomy and biogeography. In Carlsberg Foundation’s oceanogr. exp., 1928-30, Dana-Rpt., Copenhagen, 5: 1-58. 1948. Chauliodus Schn., bathypelagic genus of fishes. In Carlsberg Founda- tion’s oceanogr. exp., 1928-30, Dana-Rpt., Copenhagen, 31 : 1-148. Gibbs, R. H., Jr., and B. A. Hurwitz. 1967. Systematics and zoogeography of the stomiatoid fishes, Chauliodus pammelas and C. sloani, of the Indian Ocean. Copeia 1967 (4) : 798-805. 1968 A New Species of Melanostomiatid Fish 5 Gunther, A. 1887. Report on the deep-sea fishes collected by H.M.S. ‘Challenger’ during the years 1873-1876. In Great Britain, Rpt. Sci. Res. Challenger, 1873- 76, Zool., Edinburgh, 22 (57) : 1-268, 331-335. Morrow, J. E., Jr. 1964. Family Stomiatidae. In Fishes of the western North At- lantic. Yale Univ., Sears Found. Mar. Res., Mem. 1 (4) : 290-310. Morrow, J. E., Jr., and Robert H. Gibbs, Jr. 1964. Family Melanostomiatidae. Ibid : 351-511. Norman, J. R. 1930. Oceanic fishes and flatfishes collected in 1925-1927. In Discov- ery Rpts., Cambridge, Eng., 2: 261-370. Regan, C. T., and E. Trewavas. 1930. The fishes of the families Stomiatidae and Malacosteidae. In The Danish “Dana”-Expeds., 1920-22, Copenhagen, Rpt. No. 6: 1-143. Table 1 Stomiatoid genera and species represented in the Subtropical Convergence and/ or in subantarctic waters. Approximate total number of species in each genus in par- entheses. SA — species occurring in subantarctic waters. CONVERGENCE — species occurring only in the Convergence (x) or also in subantarctic waters (o) . S. LIMIT — species known from central waters that occur in the Convergence area. SOURCES — 1. R. H. Gibbs et al., unpublished; 2. W. H. Krueger, unpublished; 3. this paper; 4. Barnard, 1948; 5. Brauer, 1906; 6. Bussing, 1965; 7. Ege, 1934; 8. Ege, 1948; 9. Gibbs and Hurwitz, 1967; 10. Gunther, 1887; 11. Morrow, 1964; 12. Norman, 1930. A — Atlantic Ocean. I- —Indian Ocean. P — Pacific Ocean. Genus and Species SA Convergence S. A I P A Limit I P Source Chauliodus (6) sloani X 9 danae X X 6,8 Stomias (9) *boa boa XXX 1,7 gracilis X 1.7 Macrostomias ( 1 ) longibarba X X 1,11 Astronesthes (23) boulengeri XXX 1 sp. nov. X X 1 indicus X 1 Boro stomias (5) antarcticus X o o o 1 Neonesthes (2) capensis X X X 1 microcephalus X X 1 Eustomias (35) trewavasae XXX 1 enbarbatus X 1 Flagellostomias ( 1 ) boureei X 12 *S. boa boa also has a disjunct population in the Mediterranean Sea and the adjacent eastern Atlantic. 6 Contributions in Science No. 149 Leptostomias (11) macropogon gladiator Bathophilus (18) ater irregularis pawneei nigerrimus Echiostoma (1) barbatum Melanostomias (7) tentaculatus Thysanactis (1) dent ex Opostomias (2) micripnus Photonectes (15) munifcus Trigonolampa ( 1 ) miriceps l diacanthus (3) atlanticus Malacosteus ( 1 ) niger Total Species 142 Table 1 (continued) x 3 x x XXX X X X X X X X X X X X X o o o X X 9 17 12 1 1,5 1,12 1,12 1 1 1,12 1 1,4,10 3 1 2 1 LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE UMBER 150 July 11, 1968 5*7,75 Czjjcb CHIGGERS OF THE GENUS PSEUDOSCHOENGASTIA (ACARINA: TROMBICULIDAE ) FROM COSTA RICA By Julius C. Geest and Richard B. Loomis Los Angeles County Museum of Natural History Los Angeles, California 90007 Exposition Park CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8 Vi x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor CHIGGERS OF THE GENUS PSEUDOSCHOENGASTIA (ACARINA: TROMBICULIDAE) FROM COSTA RICA By Julius C. Geest1 and Richard B. Loomis1 Abstract: Chigger larvae of thirteen species of the genus Pseudoschoengastia are described and reported from mammals of Costa Rica. Nine species are new and the other four are recorded from this country for the first time. The subgenus Pseudoschoen- gastia Lipovsky contains ten species: P. bulbifera Brennan, P. in- termedia sp. n., P. montana sp. n., and P. peromysci sp. n. belong- ing to the Bulbifera Complex of the Hungerfordi Group; P. zona Brennan, P. hoguei sp. n., P. hooperi sp. n., and P. rheomys sp. n. of the Farneri Group; P. abditiva Brennan and P. fnitima Bren- nan and Yunker in the Anomala and Aeci Groups respectively. The second subgenus, Walchioides Vercammen-Grandjean, has three new species: P. costaricensis, P. guanacastensis and P. verdensis. These chiggers were found on fifteen species of rodents: Het- eromys desmarestianus, Liomys salvini, Oryzomys (five species), Ototylomys phyllotis, Peromyscus nudipes, Rheomys hartmanni, Rheomys underwoodi, Scotinomys teguina, Sigmodon hispidus, Zygodontomys microtinus and Proechimys semispinosus ; and on one opossum Philander opossum. Species of Pseudoschoengastia are recorded from 24 locali- ties. The life zones, biotic provinces and biotic districts of Costa Rica are discussed and mapped. Three species are limited to the Guanacaste Biotic District of the Pacific Mexico-Nicaragua Biotic Province and two species are known from the Golfo Dulce Biotic District of the Pacific Costa Rica-Panama Biotic Province, both along the Pacific versant. Two species occur in the Caribbean Costa Rica Biotic District of the Caribbean Costa Rica-Panama Biotic Province, also of the Tropical Lowlands. Nine species occur in the Tropical Highlands or Costa Rica-Panama Highlands Bi- otic Province, which is divided into the Costa Rica Highlands Bi- otic District in the north with six species and the southern Panama Highlands Biotic District, also with six species of Pseudoschoen- gastia. The larval stage of each species is described and illustrated, and notes are given on the ecology, including a complete host list. A key is included to the larvae of the twenty species of Pseudo- schoengastia known from Central and South America. Acknowledgments We are grateful to Drs. Fred S. Truxal and Charles A. McLaughlin, Los Angeles County Museum of Natural History, for the opportunity to study these chiggers from Costa Rica and for making it possible for the senior author to department of Biology, California State College at Long Beach, 90801, and Re- search Assistant and Research Associate respectively, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 150 collect mammals and their ectoparasites and to observe the various habitats in Costa Rica for a period of six months. We gratefully acknowledge the aid provided by the Government of the Republic of Costa Rica and the University of Costa Rica. We wish to express our appreciation to the following persons who worked as field assistants or assisted the senior author during his stay in Costa Rica: Dr. Charles L. Hogue, Dr. Jay M. Savage, Dr. Andrew Starrett, Dr. Keith A. Arnold, Ricardo Batalla, Rex Benson, Dr. Richard Casebeer, Walter Fiala, David G. Marqua, Roy W. McDiarmid, Norman J. Scott, Robert C. Stephens, and Dr. James L. Vial. Also we thank Dr. Emmet T. Hooper of the University of Michigan Museum of Zoology, who loaned us specimens of Rheomys, and Dr. James M. Brennan of the Rocky Mountain Laboratory for chiggers of several species from Panama, and for a copy of the manuscript by Brennan and Yunker ( 1966) prior to publication. The paper is, in part, a result of research supported by U.S. Army Medical Research and Development Command Grant DA-MD-49-193-63-G94, F. S. Truxal and C. A. McLaughlin, Principal Investigators. Additional support was provided by Research Grant AI-3407 from the National Institute of Allergy and Infectious Diseases, to Richard B. Loomis. This paper is a modification of a thesis submitted by the senior author to the Department of Biology, California State College, Long Beach, in partial fulfillment of the requirements for a Master of Arts Degree. Introduction Chiggers of the genus Pseudoschoengastia Lipovsky belong to the sub- famliy Trombiculinae, family Trombiculidae. As larvae, they possess six legs and parasitize small mammals, especially rodents. Both the taxonomy and classification have been based on the larvae, the only stage easily obtained. The nymphs and adults are free-living and have not been taken in Costa Rica. From January, 1962, to December, 1964, more than 500 rodents from Costa Rica were examined and chiggers were recovered from approximately 291 specimens. Of those with chiggers, 112 hosts (nearly 39 percent) had one or more species of Pseudoschoengastia. These larvae were recovered from the ears of 16 host species, including one opossum and 15 kinds of rodents belonging to three families: Heteromyidae, Cricetidae, and Echimyidae. Examination of 548 larvae revealed 13 species of Pseudoschoengastia , nine of which are new; ten species in the subgenus Pseudoschoengastia Lipov- sky, and three species in the subgenus Walchioides Vercammen-Grandjean. Description of Costa Rica Costa Rica is a small Central American Republic with numerous vol- canoes, high mountains, plateaus and coastal plains providing diverse habitats for many kinds of plants and animals. Physiographically Costa Rica consists 1968 Chiggers of the genus PSEUDOSCHOENGASTIA 3 of Pacific and Atlantic lowlands almost completely separated by central high- lands, which follow roughly a northwest to southeast orientation. The Pacific lowland is widest at the Nicoyan Peninsula in Guanacaste Province, and after narrowing near mid-country, widens again in the region of Golfo Dulce and the Osa Peninsula. The Atlantic lowland extends almost entirely across the northern border but narrows to less than 50 kilometers at the Panamanian border. The elevation near the Nicaraguan border is no greater than 100 meters; however, just to the south is the Cordillera de Guanacaste with elevations over 1000 meters. This range consists of a series of volcanoes called the Cordillera Volcanica, and at the southern end is the east-west range, the Cordillera Central. Here the lowest elevation is 1500 meters, but between Volcan Irazu and Volcan Turrialba the lowest elevation is 2150 meters. To the south, the Cordillera Central forms the central plateau, or Meseta Central, which sep- arates the northern and southern highlands. South of this plateau is the Dota Region, the lower northern portion of the Cordillera de Talamanca, which for one hundred and sixty kilometers forms the major northwest-southeast axis with elevations above 2150 meters. The following numbered localities for Pseudoschoengastia are plotted in Figs. 1 and 2 and are listed consecutively from northwest to southeast: GUANACASTE PROVINCE; 1. 5-8.3 km N Liberia; 2. Liberia; 3. 1.4 km S and 7.3 km S Liberia; 4. 3 km S Playa del Coco; 5. 5 km NW Tilaran. PUNTAREN AS PROVINCE: 6. Monteverde, 1380 m. HEREDIA PROVINCE: 7. 2.9 km S Puerto Viejo; 8. El Angel Falls. ALAJUELA PROVINCE: 9. Volcan Poas, 2493 m, and Rio Poasito, Poasitos, 2000 m. CARTAGO PROVINCE: 10. Rio Claro, 600 m, 27.5 km N San Jose on San Jeronimo Rd. LIMON PROVINCE: 11. Finca “La Lola.” CARTAGO PROVINCE: 12. Turrialba, IIC A, 600 m; 13. Tapanti, Rio Quiri, 1220 m. SAN JOSE PROVINCE: 14. Rio Damitos, 14 km N Quepos; 15. 11.3 km S La Georgina, 2500 m; 16. 15-20.8 km N San Isidro del General, 1495-1600 m. PUNTAREN AS PROVINCE: 17. Finca “Los Helechales,” 12-15 km E Potrero Grande, 1040 m; 18. Finca de Senor Treno, 8 km E Potrero Grande, 660 m; 19. 1.7 km W Palmar Norte; 20. Rio Coronado, 20 km N Puerto Cortez; 21. 7.3 km S Palmar Sur; 22. Rincon de Osa, 50 m; 23. 8.3 km W Rincon, Camp Seattle, 100 m; 24. vicinity of Villa Neily. Life Zones The five life zones and the forest formations in the Tropical Life Zone, as discussed and mapped by Slud (1964), are listed below. I. Tropical Life Zone. A. Tropical Dry Forest Formation B. Moist Forest Formation C. Tropical Wet Forest Formation II. Subtropical Life Zone. 4 Contributions in Science No. 150 III. Submontane Life Zone. IV. Montane Life Zone. V. Subalpine Life Zone. The map (Fig. 2) follows Slud except that the Tropical Wet and Moist Forest Formations are combined and the Subalpine Life Zone is not shown. This latter Life Zone is surrounded by the Montane Life Zone and occurs on two mountain tops where no Pseudoschoengastia are found. Biotic Provinces and Biotic Districts Many authors have stated that Costa Rica can be divided into areas based on the distribution of plants and animals and on the environmental factors. Goodwin (1946), in discussing mammalian distributions, did not delineate distinct provinces but mentioned five regions. Taylor (1951) suggested essen- tially the same patterns of distribution based upon the herpetofauna. Slud (1964) proposed, but did not map, four distinct avifaunal zones based on the geography, climate, geologic history, and the distribution of birds. Although he stated that there were differences in the compositions of bird species between the two major mountain areas, he found it convenient to keep them together as a single avifaunal zone. Slud utilized plant data adapted from Holdridge (1959) to give more definite boundaries to his avifaunal zones. West (1964) listed four “natural regions” that seem to correspond to the avifaunal zones of Slud. Ryan ( 1963) utilized the distribution of mammals to define the biotic provinces of Central America. He divided Costa Rica into two provinces, the Puntarenas-Chiriqui Biotic Province of the Pacific slope and the Guatuso-Talamancan Biotic Province of the Caribbean slope, separated by the central highlands. Ryan did not believe that the highlands should be a separate biotic province, and both the northern dry and southern wet forests of the Pacific slope were placed into one province. This arrangement does not seem as appropriate for Costa Rica as that of Stuart (1964), who mapped and named the biotic provinces of Central America, including four biotic provinces within Costa Rica, based upon the flora and the known vertebrate and invertebrate faunas. These biotic provinces are comparable to the avifaunal Figure 1. Map of Costa Rica showing Biotic Provinces and Districts, political prov- inces and all localities for Pseudoschoengastia. Localities 1-24 are listed in the text. A. Tropical lowlands Al. Guanacaste Biotic District of the Pacific Mexico-Nicaragua Biotic Province. A2. Golfo Dulce Biotic District of the Pacific Costa Rica-Panama Biotic Prov- ince. A3. Caribbean Costa Rica Biotic District of the Caribbean Costa Rica-Panama Biotic Province. B. Tropical highlands (Costa Rica-Panama Highlands Biotic Province) Bl. Costa Rica Highlands Biotic District. B2. Panama Highlands Biotic District. Figure 2. Life Zones in Costa Rica and localities for Pseudoschoengastia. 1968 Chiggers of the genus PSEUDOSCHOENGASTIA 5 6 Contributions in Science No. 150 zones of Slud. For zoogeographic patterns the biotic provinces, as originally conceived by Dice (1943), represent a more realistic picture than do life zones. In a recent paper, Savage (1966:723) listed and mapped the Recent Central American Herpetofaunas. Four separate herpetofaunas were shown for Costa Rica. His Nicaraguan herpetofauna corresponds to our Guanacaste Biotic District, the Golfo Dulcean is the same as the Golfo Dulce Biotic District and the Isthmian has the same distribution as the Costa Rica Caribbean. Savage considered the tropical highlands as a single herpetofauna, the Talamancan. The entire Central American Region is called Mesoamerica, which extends northward to the Nearctic in Mexico and southward to the Neotropical of eastern Panama and South America. The following biotic provinces and districts are used to aid in the under- standing of the distribution of Pseudoschoengastia and their mammalian hosts. The names assigned to the four biotic provinces presented below, and shown in Fig. 1, follow Stuart (1964). In addition, we have subdivided these biotic provinces of Costa Rica into biotic districts, and each is given an appropriate name. The Tropical Lowlands (A) consists of the Pacific Mexico-Nicaragua Biotic Province and its Guanacaste Biotic District ( Al) ; the Pacific Costa Rica- Panama Biotic Province and its Golfo Dulce Biotic District ( A2) ; the Carib- bean Costa Rica-Panama Biotic Province with its Caribbean Costa Rica Biotic District (A3). The Tropical Highlands (B) is composed of the Costa Rica-Panama High- lands Biotic Province subdivided into the Costa Rica Highlands Biotic District (Bl) and the Panama Highlands Biotic District (B2). Rainfall data presented under each biotic province and district were taken from the official records of the Ministerio de Agricultura y Ganaderia, Seccion Climatologia, for 1961. Pacific Mextco-Nicaragua Biotic Province Guanacaste Biotic District The Pacific Mexico-Nicaragua Biotic Province, as represented in Costa Rica, is called the Guanacaste Biotic District. It covers the northern half of the Pacific slopes of Costa Rica and includes the Tropical and Subtropical Life Zones. It is the southern part of a relatively arid Pacific Coast forest which extends intermittently from the state of Sinaloa, Mexico, southward to the mouth of the Gulf of Nicoya (Slud, 1964). The Guanacaste Biotic District is named for the large political province which is the major part of the district. The average annual rainfall in 1961 for ten stations was less than 2000 mm and most of this was concentrated during the months of June through November with over 50 mm of rainfall recorded for each of these months. Taylor (1951) stated that the rainfall is confined for the most part to the summer months (May to August), while the remaining months have little or 1968 Chiggers of the genus PSEUDOSCHOENGASTIA 7 no rainfall and the winds during this dry season contribute to further moisture loss for this already dry area. Although many areas with better soil have been cleared and burned, it originally supported subhumid to moist forest associations with tall stands of deciduous forest composed of relatively few species (Slud, 1964). The mammals (Goodwin, 1946) and birds (Slud, 1964) of this region are primarily northern in origin. Three new chigger species, Pseudoschoen- gastia hoguei, P. costaricensis and P. guanacasiensis, occur only in this biotic district. Pacific Costa Rica- Panama Biotic Province Golfo Dulce Biotic District That part of the Pacific Costa Rica-Panama Biotic Province in Costa Rica includes the Pacific versant from the southern tip of the Nicoyan Peninsula into Panama. The Tropical and parts of the Subtropical Life Zones are included in this district. In 1961, the average rainfall of ten stations was in excess of 3500 mm, with one station recording more than 7200 mm. All months except January (85 mm) and February (36 mm) had rainfall in excess of 100 mm. The forest association is composed predominantly of tall evergreen species typical of the tropical humid to wet rainforest. On good undisturbed soils, approximately 100 species of trees make up this three or four story tropical rain forest. The canopy is almost closed and the crowns of the scattered emergents project above it. The forest floor has only a small amount of litter and the easy passage through this forest reveals buttressed trunks of the larger trees, stilt roots of the smaller trees, and the presence of many lianas, as mentioned by Allen (1956). Slud (1964) stated that many birds of this region can tolerate the humid conditions on both the eastern and western slopes, although the many endemics found here and the absence of Caribbean species attest to a distinct avifauna. Two species of Pseudoschoengastia occur in this district. Pseudoschoen- gastia bulbifera has been taken at seven of the eight localities, and P. zona is known from one locality. Caribbean Costa Rica-Panama Biotic Province Caribbean Costa Rica Biotic District This district extends along the entire length of the eastern slope, and includes the Tropical and Subtropical Life Zones. The average annual rainfall in 1961 at ten stations was in excess of 3400 mm. Yearly rainfall in excess of 7200 mm has been recorded in the extreme northeastern corner. The average recorded rainfall for each month in 1961 was in excess of 100 mm. There was less rainfall in January (102 mm) and February (170 mm) and a second relatively “dry spell” occurred from Septem- 8 Contributions in Science No. 150 ber to November, as each month had less than 300 mm of rainfall. The evergreen floral composition of this biotic district has extremely moist to wet forest associations similar to those found in the Golfo Dulce Biotic District. The lower slopes near sea level are covered by extensive tracts of palm swamps that are almost impassable because of numerous water courses and a year-round high water table. Large tracts of this land in Costa Rica and Panama have been cleared to grow bananas, cacao and sugar cane. Many typical species of trees in the Golfo Dulce Biotic District are also found here. Slud (1960, 1964) listed the species of trees. Goodwin (1946) stated that the mammals of the Caribbean are closely related to forms in adjacent Panama, and Slud (1964) noted that affinities of the Caribbean species of birds are with species of Panama and South America. Of the two species of Pseudoschoengastia taken in this district, P. bulbifera were taken from all localities and P. finitima from only one site. Costa Rica-Panama Highlands Biotic Province This biotic province of the Tropical Highlands in Costa Rica consists of two distinct areas which are separated into the northern Costa Rica Highlands and the southern Panama Highlands Biotic District. Seven of the nine species of Pseudoschoengastia from this province have not been taken elsewhere, but only P. montana sp. n. is known from both districts. Costa Rica Highlands Biotic District This biotic district, with many endemic species of mammals (Goodwin, 1946), consists of the northern mountains and Cordillera Central, including the Montane, Submontane and parts of the Subtropical Life Zones. The average annual rainfall in 1961 for ten stations was in excess of 2600 mm. The rainfall pattern was similar to that of the adjacent Guanacaste Biotic District, although the rainfall was heavier and the pattern showed sharper peaks. Slud (1964) stated that the evergreen trees of the subtropical forest often form a dense canopy but are not of value as timber and are poorly known. The montane moist to very humid forest formations are dominated by many species of oaks that are often bathed in clouds. Many of these temperate mountain areas have been cleared to raise fruits, vegetables, and cattle. Goodwin (1946) believes that many mammals of this district have afffin- ities with species of the Mexican Central Plateau. Six species of Pseudoschoengastia are reported; one new species (P. montana) is known only from these two highland districts, and three new species (P. hooperi, P. peromysci, and P. verdensis) were taken in this district. Panama Highlands Biotic District This district is south of the central plateau and includes the mountains of 1968 Chiggers of the genus PSEUDOSCHOENGASTIA 9 the north-central Dota Region of Costa Rica and the high Talamanca Cor- dillera. These mountains terminate with Cerro Pittier in the Chiriqui region of Panama bordered by the arid savannas of northwestern Panama. Subtropical to Subalpine Life Zones occur in this biotic district. The average rainfall for eight stations was in excess of 2300 mm and had a pattern similar to that of the Costa Rica Highlands. The Panama Highlands has fewer human inhabitants, and rainfall records probably are from the drier areas. The moist to wet forest formations of this region are dominated by species of oaks and a subalpine wet paramo above the tree line on Cerro de la Muerte and Cerro Chirripo, which according to Slud (1964) has an affinity with the paramos of the Andes. Most species of mammals (Goodwin, 1946) and birds (Slud, 1964) are related to forms of Panama and South America. Six species of Pseudoschoengastia have been taken here, of which three (P. abditiva, P. intermedia sp. n., and P. rheomys sp. n.) were not obtained elsewhere in Costa Rica. Table 1 The distribution of Pseudoschoengastia in Costa Rica, based on the occurrence of the species in each biotic district. Numbers in each column refer to the number of localities for each species. Species A1 BIOTIC DISTRICTS A2 A3 B 1 B2 Total localities with Pseudoschoengastia 5 8 6 2 3 1. P. bulbiferci Brennan 7 6 1 2 2. P. intermedia sp. n. — — — — 1 3. P. montana sp. n. — — — 1 2 4. P. peromysci sp. n. — — — 1 — 5. P. zona Brennan — 1 — 1 2 6. P. hoguei sp. n. 3 — — — — 7. P. rheomys sp. n. — — — — 1 8. P. hooperi sp. n. — — — 1 — 9. P. abditiva Brennan — — — — 1 10. P. finitima Brennan & Yunker — — 1 — — 1 1. P. guanacastensis sp. n. 3 — — — — 12. P. costaricensis sp. n. 3 — — — — 13. P. verdensis sp. n. — — — 1 — Number of species in each 3 2 2 6 6 district (or districts) 3 9 Number limited to biotic district (in Costa Rica) 3 0 1 3 3 1 10 Contributions in Science No. 150 Materials and Methods From 1962 through 1964 approximately 3,500 mammals from Costa Rica were examined for ectoparasites. Most of these were bats, but 505 were rodents. They were taken primarily in Sherman live traps, rat and museum special kill traps using baits of rolled oats, bananas and peanut butter. Each specimen was sealed in a separate cloth or plastic bag until it was examined for ectoparasites with the aid of a stereoscopic microscope. The estoparasites were preserved in 75 per cent ethyl alcohol and each kind of parasite and its location on the host usually were listed in the field notes. The mammals were prepared as wet specimens in formalin or as dry study skins and have been deposited in the Los Angeles County Museum of Natural History (LACM) . Representatives of the larval chiggers were mounted in polyvinyl alcohol- lactophenol (PVA-LP). These larvae were studied with the aid of a phase- contract microscope, and all drawings, made with the aid of a drawing tube by the senior author, were based on the type series and other specimens from Costa Rica. Specimens designated with an O- were taken by members of the LACM field parties and those specimens with the designation “RML” belong to the Rocky Mountain Laboratory, Hamilton, Montana. Unless otherwise noted, the chiggers have been deposited in the Acarina collection, LACM. The holotype and at least one paratype of each new species has been retained in the LACM collection. When available, paratypes will be deposited in the following collections: Chigger Research Collection, California State College, Long Beach; Rocky Mountain Laboratory, Hamilton, Montana; The University of Kansas; The George Williams Hooper Foundation, University of California Medical Center, San Francisco; United States National Museum; Institute of Acarology, Ohio State University; Bishop Museum, Honolulu, Hawaii; Dr. Anita Hoffmann, Mexico, D.F., and other appropriate institutions and individuals. Accounts of the Species The terminology for the larva usually follows that of Wharton et al. (1951), supplemented by the terms from Newell (1957). All measurements are in microns. Unless otherwise noted, the description and illustrations of each new species are based upon the holotype. Paratypes from Panama and referred specimens from Costa Rica were utilized to characterize those species already named. The term “group” is utilized to bring together similar species below the subgeneric level. Any subgroup of several more nearly similar and presumably related species is called a complex. Genus Pseudoschoengastia Lipovsky, 1951 Type-species : Pseudoschoengastia hungerfordi Lipovsky, 1951. Vanidicus Brennan and Jones, 1961 (type species, Vanidicus tricosus ), new synonymy . 1968 Chiggers of the genus PSEUDOSCHOENGASTIA 11 Pseudoschoengastia was proposed by Lipovsky ( 1951 ) to include two new species, P. hungerfordi and P. farneri, and Ascoschoengastia diazi (Hoffmann 1948). Two additional species, Ascoschoengastia anomala and A. pedregalen- sis, were described from Mexico by Hoffmann (1951), and Brennan (1952) named P. guatemalensis from Guatemala and P. occidentalis from California. Seventeen additional species of Pseudoschoengastia have been described from southwestern United States and from Mexico southward to Panama (Brennan and Jones, 1959; Brennan, 1960; Hoffmann, 1960; Brennan, 1965; Brennan and Yunker, 1966). Fauran (1960) described Pseudoschoengastia myoproc- tae from French Guiana, the only species of this genus known from South America. Brennan and Jones ( 1961 ) described a new genus and species, Vanidicus tricosus, from Panama, and indicated a close relationship to Pseudoschoen- gastia. The absence of differentiation in characters and the close similarity to P. abditiva Brennan has prompted us to synonymize Vanidicus with the genus Pseudoschoengastia. Walchioides Vercammen-Grandjean (1960) was proposed originally as a subgenus of the Asiatic genus Susa Audy. Although the type-species, Walchia gouldi Hoffmann from Mexico, is unique in lacking the AM scutal seta, it closely resembles several species of Pseudoschoengastia, and therefore we consider Walchioides a subgenus of Pseudoschoengastia. We are placing all species of Pseudoschoengastia with the PL setae on the scutal plate in the sub- genus Walchioides. Twenty- seven species of Pseudoschoengastia have been described: four from the United States; 13 from Mexico and Guatemala; nine from Panama; one from northern South America. The nine additional species from Costa Rica described below brings the total number to 36. Larval Pseudoschoengastia are parasitic on small rodents, lagomorphs, insectivores and other small terrestrial mammals. Referred species: Thirty-six species, arranged below in two subgenera: Pseudoschoengastia (26 species) and Walchioides (10 species). Diagnosis: Pseudoschoengastia differs from all other genera in the sub- family Trombiculinae by having the following combination of larval charac- ters: scutum small; sensilla clavate to capitate; palpotibial claw trifurcate; palpal tarsus with five branched setae; body with many setae, usually with two or more pairs of humeral setae and one or more pairs of lateral humeral and lateral sternal setae between coxae II and III; leg I with stout subterminala (dorsal eupathid) and parasubterminala (companion seta); basifemur and telofemur III completely or partially fused (showing line of previous articula- tion) , with two or more internal bars; tibiala III present; leg III without masti- setae. Closely resembling the genera Euschoengastoides Loomis and Cordiseta Hoffmann; but differing from Euschoengastoides in having subterminala and parasubterminala I (usually absent in Euschoengastoides)-, lateral humeral and lateral sternal setae between coxae II and III (absent in Euschoengas- 12 Contributions in Science No. 150 toides ); long slender vestigiala on leg I (short dagger-shaped vestigiala in Euschoengastoides) ; femora II and III fused (not fused in Euschoengas- toides ) ; differing from Cordiseta in lacking large foliate dorsal body setae. Description : Larval characters as in diagnosis, plus: scutum with postero- lateral setae on or off the scutal plate; expanded sensilla with setules large to extremely small; cheliceral blade with a terminal tricuspid cap (some with a serrated medial margin), and with or without a dorsomedial tooth; palpal tarsus without subterminala (eupathid); body small, ellipsoidal and somewhat constricted when engorged; eyes 2/2 (rarely absent), usually on a plate; anus at level of 4th and 5th rows of ventral body setae; postanal setae resembling dorsal setae; all legs terminating in two lateral claws and median clawlike empodium, without tenent hairs (onychotriches) ; cheliceral bases, capitular sternum, scutum and all leg segments punctate. Galeala usually nude, rarely with one to four fine branches. Comments : There is some confusion in the names of the lateral body setae which are present between and near coxae II and III of Pseudoschoen- gastia. Lipovsky (1951) and Brennan (1952) followed Hoffmann (1948) and referred to these setae as dorsal and ventral humerals. However, these lateral setae are of two distinct morphological types: 1) a humeral seta, and 2) a sternal seta. Because of the position between coxae II and III, the ventralmost humerals will be called lateral humeral setae, and the sternals will be con- sidered lateral sternal setae. The leg segmentation of this genus should be considered as 7-6-6, with the fusion of the basifemur and telofemur on legs II and III. Most species of Pseudoschoengastia were examined, and in some species a definite line of fusion was seen. Loomis (1956:1419) reported that a femoral suture could be discerned in P. hungerfordi but seemed to be entirely absent or indistinct in P. farneri. The degree of fusion was determined by the thickness and number of heavily sclerotized rings within the femur. When completely fused, there were fewer, less distinct bars. The condition of fusion seemed to be consistent within each species, but varied among different species. The leg index is the sum of one measured length for each of the legs I, II and III. Taxonomic remarks : Cordiseta is another New World genus that shares many characteristics with Pseudoschoengastia. These similarities include five branched setae on the palpal tarsus, expanded sensilla, multiple lateral humerals and sternals, and reduced (7-6-6) leg segmentation. The many characteristics shared by Cordiseta and Pseudoschoengastia indicate a close relationship. They are not closely similar to Ascoschoengastia, Laurentella, or other Old World genera. Subgenus Pseudoschoengastia Lipovsky, 1951 Referred Species : Twenty-six species placed into four groups: Hunger- 1968 Chiggers of the genus PSEUDOSCHOENGASTIA 13 SAMPLE TUfc L- 1. GOLFO DULCE 1 “1 N= 33 2. LA LOLA i N = 10 3. PUERTO VIEJO r izn — wtacn — 1 N = 6 4. RIO CLARO i i ^ii i N =1 3 AW l i 1 j J 30 40 50 N = 12 2 r . l.Ai l N = 10 3 T I N = 6 SD 4 L_ 30 N = 1 3 1 40 50 1 i rdrr — i. nos 2 I ZZE^CLZ: I n = io 4 I I di I I N = 1 1 AL I L I I 1 50 60 70 80 90 1 I . ~ ~ 1^1 ! N = 1 3 2 I | ^ | | N = 1 0 LEG INDEX 3 I nfa I n = 6 (total of 3 legs) 4 I 1 M 1 I N = 1 3 L J I 500 600 700 800 Figure 3. Modified Dice-Leraas diagrams of selected measurements of Pseudoschoen- gastia bulbifera from four different populations: 1. Golfo Dulce Biotic District; 2. Finca “La Lola”; 3. Puerto Viejo de Serapiqui; 4. Rio Claro. Measurements, top to bottom, are: AW, anterior width of scutum; SD, scutal depth; AL, length of antero- lateral scutal seta; leg index, total length of three legs. N = size of sample. In each diagram the lower line indicates the total variation of the sample (N); the shaded area one standard error, the adjacent bar twice the standard error on each side of the mean; and the upper line is two standard deviations on each side of the mean. 14 Contributions in Science No. 150 fordi Group (8 species), Farneri Group (7 species), Anomala Group (8 species), and Aeci Group (3 species). Diagnosis’. Larva with posterolateral setae off the scutum; AM < AL> PL, AM=ALALALAM and PL scutal setae (Hungerfordi Group) 3 2.1 Anterolateral seta =or = < femur / genu / tibia * D-L-V scutum iceral Tooth III 1. P. bulbifera B / B / BBB N 3 1 1 off + 2 > 2. P. intermedia B / B / NNB N 3 1 1 off + 2.5 > 3. P. montana B / B / NBB N 3 1 1 off + 2.5 > 4. P. peromysci B / B / NNB N 3 1 1 off + 2.5 > 5. P. zona B / N / NNN N 3 1 1 off + 2 < 6. P. hoguei B / N / NNN N 2 1 1 off + 2.5 < 7. P. rheomys B / B / BNB N 3 1 1 off + 2 = 8. P. hooperi B / B / NNN N 3 1 1 off + 2 < 9. P. abditiva B / N / NNN N 3 1 1 off 0 2 < 10. P. finitima B / N / BNB N 2 1 1 off 0 2 < 1 1 . P. guanacastensis B / B / BNB i N (IB) 2 1 1 on 0 3 < 12. P. costaricensis B / B / BNB B 2 0 0 on 0 3 < 13. P. verdensis *B=branched N=nude B / B / NNB N 1 0 0 on 0 3 < 1968 Chiggers of the genus PSEUDOSCHOENGASTIA 47 Table 3 List of Mammalian Hosts for Species of Pseudoschoengastia in Costa Rica Species of Mammals* Chiggers (number of hosts with Pseudoschoengastia ) ORDER MARSUPIALA Family Didelphidae Philander opossum (Linnaeus) P. bulbifera (Four-eyed Opossum) (1) P. zona ORDER RODENTIA Family Heteromyidae Heteromys desmarestianus Gray P. bulbifera (Desmarest’s Spiny Pocket Mouse) (2) P.finitima Liomys salvini (Thomas) P. hoguei (Salvin’s Spiny Pocket Mouse) (30) P. guanacastensis P. costaricensis Family Cricetidae Oryzomys alfaroi (Allen) P. abditiva (Alfaro’s Rice Rat) (1) P. bulbifera P. montana Oryzomys bombycinus Goldman P. abditiva (Silky Rice Rat) (2) P. montana Oryzomys caliginosus (Tomes) P. bulbifera (Costa Rican Dusky Rice Rat) (24) P. zona Oryzomys albigularis (Tomes) (=0. devius ) P. bulbifera (Chiriqui Rice Rat) (7) P. intermedia P. montana P. zona Oryzomys fulvescens (Saussure) P. bulbifera (Costa Rican Pygmy Rice Rat) ( 1 ) Ototylomys phyllotis Merriam P. guanacastensis (Nicaraguan Climbing Rat) (2) P. costaricensis Peromyscus nudipes (Allen) P. bulbifera (Naked-footed Deer Mouse) (12) P. montana P. peromyscus P. zona P. verdensis Rheomys hartmanni Enders P. rheomys (Panamanian Water Mouse) (6) Rheomys underwoodi Thomas P. hooperi (Costa Rican Water Mouse) (3) Scotinomys teguina (Alston) P. bulbifera (Alston’s Brown Mouse) (1) Sigmodon hispidus Say and Ord P. bulbifera (Hispid Cotton Rat) (5) P. zona Zygodontomys microtinus Thomas ( —Z . cherriei) P. bulbifera Family Echimyidae Proechimys semispinosus (Tomes) P. bulbifera (Tomes’ Spiny Rat) (5) P. zona *The specific names in this list follow Handley (1966), with previously used names in parentheses. 48 Contributions in Science No. 150 Literature Cited Allen, P. H. 1956. The rain forests of Golfo Dulce. Univ. Florida Press, Gaines- ville, xi + 417 p. Brennan, J. M. 1952. The genus Pseudoschongastia Lipovsky, 1951 with the descrip- tion of two new species and a key to the world species, also Neoschongastia paen- itens, new name for Neoschongastia kohlsi Brennan, 1951, preoccupied (Aca- rina, Trombiculidae). Proc. Ent. Soc. Wash. 54:133-137. 1960. Eight new species of Pseudoschongastia from Mexico and Panama with a revised key to species (Acarina: Trombiculidae). Acarologia 2:480-492. 1965. Five new chiggers from southwestern United States (Acarina: Trom- biculidae ) ; J . Parasit. 51:108-113. Brennan, J'. M., and E. K. Jones. 1959. Pseudoschongastia and four new neotropical species of the genus (Acarina: Trombiculidae). J. Parasit. 45:421-429. 1961. New genera and species of chiggers from Panama (Acarina: Trombicu- lidae). J. Parasit. 47:105-124. Brennan, J. M., and C. E. Yunker. 1966. Chiggers of Panama (Acarina: Trombicu- lidae). In Wenzel, R. L., and V. J. Tipton, eds. Ectoparasites of Panama. Field Mus. Nat. Hist., Chicago, pp. 221-266. Dice, L. R. 1943. The biotic provinces of North America. Univ. Michigan Press, Ann Arbor, 78 p. Fauran, P. 1960. Description de quatre nouvelles especes et d’une nouvelle sous- espece de trombiculides de Guyane Frangaise. Archives, Institut Pasteur Guy. Frang. No. 459:1-20. Goodwin, G. G. 1946. Mammals of Costa Rica. Bull. Amer. Mus. Nat. Hist. 87: 275-473. Hall, E. R., and K. R. Kelson. 1959. The mammals of North America. Ronald Press, New York. 2 vols. v. 1 : xxx -f- 546 p.; v. 2: viii + 547-1083 p. Handley, C. O., Jr. 1966. Checklist of the mammals of Panama. In Wenzel, R. L., and V. J. Tipton, eds. Ectoparasites of Panama. Field Mus. Nat. Hist., Chicago, pp. 753-795. Hoffmann, A. 1948. Dos especies nuevas de trombiculidos mexicanos. Rev. Inst. Salubr. Enf. Trop., Mexico, 9:177-189. 1951. Contribuciones al conocimiento de los trombiculidos mexicanos, 4a Parte. Ciencia. 11:97-103. 1954. Contribuciones al conocimiento de los trombiculidos mexicanos (Aca- rina, Trombiculidae) (6a parte). An. Esc. Nac. Ciencias Biol., Mexico, 8:17-30. 1960. Contribuciones al conocimiento de los trombiculidos mexicanos (Aca- rina: Trombicul.) (8a parte) . Ciencia 20:99-105. Holdridge, L. R. 1959. Mapa ecologico de Costa Rica, A. C. con la clave de clasifi- cacion de vegetales del mundo. San Jose, Instituto Interamericano de Ciencias Agricolas de la O. E. A., Proy. 39, Prog. Coop. Tec. Lipovsky, L. J. 195 1 . A new genus of Walchiinae (Acarina, Trombiculidae) . J. Kans. Ent. Soc. 24:95-102. Loomis, R. B. 1956. The chigger mites of Kansas (Acarina, Trombiculidae). Univ. Kans. Sci. Bull. 37:1195-1443. Newell, I. M. 1957. Studies on the Johnstonianidae (Acarina, Parasitengona), Pa- cific Sci. 11:396-466. Ryan, R. M. 1963. The biotic provinces of Central America. Acta Zool. Mexicana 6(2-3) : 1-54, map. 1968 Chiggers of the genus PSEUDOSCHOENGASTIA 49 Savage, J. M. 1966. The origins and history of the Central American herpetofauna. Copeia 1966:719-766. Slud, Paul. 1960. The birds of tinea “La Selva” Costa Rica: a tropical wet forest lo- cality. Bull. Amer. Mus. Nat. Hist. 121:49-148. 1964. The birds of Costa Rica. Bull. Amer. Mus. Nat. Hist. 128:1-430. Stuart, L. C. 1964. Fauna of Middle America. In Handbook of Middle American Indians, Vol. I. Natural environment and early cultures, ed. by R. C. West. Univ. Texas Press, Austin, pp. 316-362. Taylor, E. H. 195 1. A brief review of the snakes of Costa Rica. Univ. Kans. Sci. Bull. 34:1-188. Vercammen-Grandjean, P. H. 1960. Introduction a un essai de classification ra- tionelle des larves de Trombiculinae Ewing 1944 (Acarina, Trombiculidae). Acarologia 2:469-471, 1 table. West, R. C. 1964. The natural regions of Middle America. In Handbook of Middle American Indians, Vol. I. Natural environment and early cultures, ed. by R. C. West. Univ. Texas Press, Austin, pp. 363-383. Wharton, G. W., D. W. Jenkins, J. M. Brennan, H. S. Fuller, G. M. Kohls, and C. B. Philip. 1951. The terminology and classification of trombiculid mites (Acarina: Trombiculidae) J. Parasit. 37:13-31. Accepted for publication March 20, 1968. Z™. CONTRIBUTIONS z:z: IN SCIENCE jMBER 151 December 13, 1968 p7. 7/ C-z/tc? A NEW EARLY TERTIARY PERISSODACTYL, HYRACOTHERIUM SEEKINSI , FROM BAJA CALIFORNIA By William J. Morris ,i ii 2* Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8 Vi x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor A NEW EARLY TERTIARY PERISSODACTYL, HY RACOTHER1UM SEEKINSI, FROM BAJA CALIFORNIA By William J. Morris1 Abstract: Hyracotherium seekinsi, a new Paleocene spe- cies from Baja California, is described as being closely related to H. angustidens. Rectangularity of the upper molars suggests a morphological difference between H. angustidens, H. seekinsi, and Desmatoclaenus. On the basis of rectangularity, Desmato- claenus appears to be ancestral to H. seekinsi, but not directly ancestral to H. angustidens. A sequence of red mudstone and siltstone capped by coarse conglomerate forms two small but topographically prominent buttes between Punta Prieta and Rancho Rosarito approximately 325 miles south of Ensenada in Baja California, Mexico. Five poorly preserved but significant vertebrate fossils were found at this locality during 1965 and 1966 by field parties from the Los Angeles County Museum of Natural History (LACM). In 1965 four upper molars were collected and referred to Hyracotherium (Morris, 1966). These molars were associated with barylambdid pantodonts and Esthonyx sp. The pantodonts were found approximately 20 feet above and the specimen of Esthonyx 30 feet below the horizon containing the molars. Since Hyracotheri- um has not been reported before as occurring with or stratigraphically below barylambdid pantodonts, it seems pertinent to stress the vertical continuity of the section. Structural complications are lacking and the strata are not over- turned or faulted. Terrestrial vertebrate fossils are rare at the Punta Prieta-Rancho Rosarito locality and, aside from the Hyracotherium molars, the collection includes only a partial skeleton of a barylambdid pantodont, a barylambdid pantodont scapula, various teeth of Esthonyx (?), and a partial skeleton of an immature creodont. A few fresh water gastropods were collected and hackberry seeds are common. Acknowledgments The field work in Baja California is supported by the National Geo- graphic Society. Cooperation of the University of Baja California and the University of Mexico is acknowledged. Dr. Theodore Downs, Los Angeles County Museum of Natural History, reviewed the manuscript, which is more lucid as a result of his suggestions. The drawings (Fig. 1) were done by Mr. Donald Cocke, LACM. 1Research Associate, Vertebrate Paleontology, Los Angeles County Museum of Natural History, and Professor of Geology, Occidental College, Los Angeles, Cali- fornia 90041. 1 Contributions in Science No. 151 My thanks to Dr. Donald Savage, Museum of Vertebrate Paleontology, University of California, Berkeley, for allowing me to study specimens from the Four Mile Creek locality. Abbreviations used in this contribution include LACM, Los Angeles County Museum of Natural History, and AMNH, American Museum of Natural History. Geologic Age of the Punta Prieta-Rancho Rosarito Deposits Heim ( 1922) named the sequence of conglomerates, siltstones, and mud- stones, outcropping near the now abandoned Rancho Tepetate, the Tepetate Formation. Later Beal (1948) extended the geographic and stratigraphic limits to include the oldest Tertiary rocks in Baja California belonging to the Paleocene and Eocene series and extending along the western side of the pen- insula from San Ysidro to the latitude of La Paz. The unit is not continuous. The Punta Prieta-Rancho Rosarito outcrop is approximately 10 miles long and separated from adjacent outcrops by Late Tertiary volcanics or deep arroyos. Invertebrates collected from a sequence lithologically similar to Punta Prieta-Rancho Rosarito, but occurring about 12 miles south of the verte- brate locality, include the gastropod, Turrit ella pachecoensis, a species char- acteristic of the Lower Martinez Formation considered Paleocene in the west coast invertebrate chronology. Outcrops 60 to 70 miles north and south of the Punta Prieta-Rancho Rosarito locality have yielded Early to Middle Eocene invertebrates (Darton, 1921; Vaughn, 1929; and Beal, 1948). Most published information, including Beal (1948), is of a preliminary nature and only gross stratigraphic units were defined. The Baja California pantodont will be described in a later paper. Though poorly preserved, it certainly belongs in the family Barylambdidae as defined by Simons (1960). Details of the skeleton and scapula are remarkably similar to Barylambda faberi (Patterson, 1937) from Tiffanian Debeque Formation of Colorado. It does not resemble the Asiatic Early Eocene barylambdid, Haplo- lambda planicanum. In a preliminary report (Morris, 1966), the Punta Prieta-Rancho Rosa- rito vertebrates were considered probably of Clarkforkian age. Recently Wood (1967) reviewed the evidence for a Clarkforkian fauna and found it incon- clusive. He stated (p. 28), “Such evidence scarcely warrants recognition of the Clark Fork as a provincial age, faunal zone, or member of the Polecat Bench Formation.” If the Clarkforkian is not valid, then the Punta Prieta- Rancho Rosarito vertebrates could be either Wasatchian or Tiffanian. The primitive nature of the Hyracotherium and the presence of barylambdids favor a Tiffanian Age but future discoveries may alter this conclusion. 1968 A New Tertiary Perissodactyl 3 Order Perissodactyla Owen, 1848 Family Equidae Gray, 1821 Hyracofherium Owen, 1840 Hyracofherium seekinsi, new species Holotype : LACM 15349. Left M2, M3, and right M2, M3 from single maxillary. Collected by W. J. Morris field party, 1965. Locality and horizon : LACM Locality No. 65155, approximately 25 kilometers south of Punta Prieta, Baja California. Locality characterized by two prominent buttes. Specimen found approximately one hundred and fifty feet below conglomerate capping most northern butte. Diagnosis’. M1 and M2 considerably smaller than those of Hyracotherium angustidens. Length and width of M3 is within the range of H. angustidens, but length and width of M1, M2 outside the range (Fig. 3). M1 and M2 more rectangular than M3, the greatest dimension being from the labial to lingual margins. M1 and M2 more rectangular than in H. angustidens. The principal cusps are closely appressed toward the midline, forming deep valleys between the protocone-paracone and hypocone-metacone. This feature is most con- spicuous on M1 and M2 when compared with M3. The angle formed between the paracone and metacone margin and the posterior border is more obtuse than in specimens of H. angustidens, although this characteristic is notably variable among hyracotheres. M1 lacks about one-third of the labial side but a well developed protoloph and metaloph are present on the lingual side. These lophs are noticeably parallel and point towards the anterior margin of the tooth. Neither proto- cone nor hypocone are prominent as these two cusps form an integral part of the lophs. Lophs are not well developed on M2. The ectoloph is barely discernible between the closely appressed paracone and metacone. Both the paraloph and metaloph are sharp ridges diminishing rapidly from the paracone and meta- cone to the labial margin. The four major cusps are sharply pointed and stand well above the lophs. The protocone and hypocone are conical and separated by a deep depression. The labial side of both metacone and protocone slopes less steeply than the lingual side. The two cusps, geometrically, form a pierc- ing, almost shearing blade. The parastyle of M2 is compressed and crowded against the paracone. The sharp definition of the principal cusps and the lack of loph develop- ment on M2 and M3 is very different from the semi-parallel strong loph devel- opment on P4. M2 and M3 resemble H. angustidens and condylarths much more than does M1. It is difficult to assess the taxonomic significance of the cusps on M3, as in Hyracotherium these are variable in geometry and position. The number and arrangement of cusps on the posterior lingual margin of M3 of H. seekinsi, 4 Contributions in Science No. 151 Figure 1. Hyracotherium seekinsi, n. sp.; A, upper left M2-M3 labial side; B, occa- sional view upper left M2-M3 and upper right IVF-M2; C, upper left M3 posterior view. 1968 A New Tertiary Perissodactyl 5 Figure 2. Hyracotherium seekinsi, n. sp.; A, upper left M2-M3 labial aspect; B, oc- casional view upper left M2-M3; C, occlusional view upper right IVF-M2; D, upper right M2 labial aspect; E, upper left M3 posterior view. 6 Contributions in Science No. 151 Figure 3. Scatter diagram comparing individual upper molar length-width ratios of Hyracotherium seekinsi, n. sp., with H. angustidens (Four Mile Creek). however, do seem significant. Posterior from the protocone are two cusps lo- cated on the platform of the tooth. A third cusp, arising from the lingual cingulum, is also well developed. It is difficult to decide which of these repre- sents the hypocone, but regardless of the cusp selected the hypocone is appre- ciably smaller than in observed specimens of H. angustidens. This new species is named in honor of its discoverer, Mr. William Seekins. Discussion : It is becoming increasingly apparent that Early Tertiary peris- sodactyls and their progenitors among the Paleocene condylarths are morpho- logical intergrades, at least as regards dentition. It is impossible to ascertain the degree and kind of morphological overlap until many more geographically and temporally isolated samples have been examined. As a result it becomes extremely difficult to formulate precise parameters of diagnostic taxonomic value at the species level. This is particularly valid for Hyracotherium seek- insi, which may well be representative of a morphological grade between cer- tain Paleocene condylarths and H. angustidens. The condylarths, T etraclaenodon and Desmatoclaenus, have each been proposed as progenitors of Hyracotherium. Apparently neither geometry nor number of cusps precludes either from this ancestral possibility (Kitts, 1956; Radinsky, 1966). Both condylarth genera have the necessary primary cusps to provide a morphological ancestor for Hyracotherium. The geometry of the teeth, aside from rectangularity (to be discussed later), appears quite differ- ent in Hyracotherium, T etraclaenodon, and Desmatoclaenus. The upper molars of T etraclaenodon are broadly ovate, while those of Desmatoclaenus are triangular. The difference is partly due to the development of at least an incipient hypocone on M1 and M2 of some samples of Tetraclaenodon, as well as the strong posterior cingulum of M3. In addition, the protocone has 1968 A New Tertiary Perissodactyl 7 developed directly labial to the paracone so that lines drawn through para- cone-protocone crests and metacone-hypocone crests are parallel. This is not the case in the more triangular molars of Desmatoclaenus, particularly in D. hermaeus, as the protocone is located at the V formed by the three prin- cipal cusps. The geometry of Tetraclaenodon is in this respect more like Hyracotherium than is that of Desmatoclaenus . Hyracotherium seekinsi shows the characteristic development of parallelism of the principal cusps on M1 and M2, but the teeth are much more rectangular than in Tetraclaenodon. Aside from the geometry of the principal cusps, H. seekinsi has many morphological features suggestive of desmatoclaenid affinities. Desmatoclae- nus hermaeus has a respectable parastyle on M2 that is closely appressed against the paracone and connected to it by an anterior spur of the ectoloph. This is almost a duplicate of the morphological condition in M2 of Hyraco- therium seekinsi. In addition, M2 of both Desmatoclaenus hermaeus and Hyracotherium seekinsi have protoconules and metaconules less developed than in primary cusps, paracone and metacone appressed towards the midline, and slight development of the ectoloph between paracone and metacone. Another feature common to the two genera is the similarity of the protoloph, which trends in an anterior direction towards the parastyle, gradually dimin- ishing in height and interrupted midway by a modestly developed protoconule. M3 of Hyracotherium seekinsi has three weakly developed cusps or cuspules on the hypocone portion. Were these not present, the tooth would show a striking resemblance to the M3 of Desmatoclaenus hermaeus. It is feasible that the hypocone of the Hyracotherium lineage arose from a strong posterior labial cingulum similar to the one present on M3 of Desmatoclaenus hermaeus. Comparative Rectangularity of Teeth Hyracotherium seekinsi is morphologically similar to H. angustidens and temporally near this species. The problem of whether H. seekinsi is ancestral to the later H. angustidens or is representative of a variant population of early perissodactyls cannot be solved from a few teeth. Kitts (1956) demonstrates that H. angustidens consists of several intergrading populations closely related but not of proven contemporaneity. Aside from central measurements of size frequency, definitive molar characteristics have not been recognized for these populations. Molar comparison of H. seekinsi with H. angustidens shows a marked difference in rectangularity. That may reasonably be of phylogenetic importance as well as of diagnostic value. Rather than using scatter diagrams to portray rectangularity, a more direct method of comparison has been used (Figs. 4, 5). Linear measure- ments were made for each specimen and the mean length and mean width computed when warranted by the size of the sample. A line from the origin to a point determined by the coordinates of mean length and width was drawn 8 Contributions in Science No. 151 RECT ANGULARITY - H. ongustidens (Four Mile) and H. seekinsi H. ongustidens - MEAN DEVIATION (Drawn on 2 sigma points) Figure 4. Diagram illustrating graphic method of compiling comparative rectangu- larly measures as presented in Figure 5. Upper molars of Hyracotherium seekinsi (solid line) are compared with H. angustidens (dashed and dotted lines) from the Four Mile Creek localities; for H. angustidens, M1 N = 7, M2 N = 34, M3 N = 32. Upper left diagram shows parameters for diagram A-C. Figure 5 contains only the rectangularity scale. The measure of rectangularity adopted is comparative only when the same scale is used for all included teeth. Note that in size, as indicated by the dotted, dashed and solid outlined rec- tangles, the H. seekinsi molars fall within or close to two standard deviations of H. angustidens, whereas the measure of rectangularity more clearly expresses a difference between these teeth. and extended a convenient distance. An arbitrary scale from plus 1 to minus 1 was adopted as a measure of rectangularity. Fig. 4 illustrates the graphic method used in establishing the measure of rectangularity. Comparisons of the rectangularity of taxa under consideration are presented in Fig. 5. When 1968 A New Tertiary Perissodactyl 9 H. anguslidens (Four Mile) N=ll H. anguslidens (Sand Coulee) N = 7 Tetraclaenodon puercensis N = 3 H. anguslidens (Gray Bull) N = 7 Desmaloclaenus paracreodus N- H. seekinsi N = Tetraclaenodon puercensis N = 3 H. anguslidens (Sand Coulee) N = 7 H. anguslidens (Gray Bull) N = I4 "H. anguslidens (Four Mile) N = 34' H. seekinsi N = I Desmaloclaenus hermoeus N = I Desmaloclaenus paracreodus N = l A. UPPER Ml UPPER M2 H. seekinsi N = l H. angustidens (Gray Bull) N = II H. angustidens (Four Mile)N = 32 H. angustidens (Sand Coulee) N = 6 Desmaloclaenus hermoeus N = I Desmaloclaenus paracreodus N = l - 1.5 J C. UPPER M3 Figure 5. Relative rectangularity of selected species of Early Tertiary perissodactyls and condylarths. Desmatoclaenus paracreodus is included to show the closer mor- phological similarities of D. hermaeus to Hyracotherium seekinsi. rectangularity is presented in this manner, it is independent of size as long as the same scale is used in the comparison. M1 and M2 of Hyracotherium seekinsi are more rectangular than those of H. angustidens, although in both species the teeth are short and wide. This difference in rectangularity appears to be a consistent one, as shown in H. angustidens samples from Four Mile Creek (University of California, Berke- ley, V5365E, V5350, V5396, V5424, V5421, V5422, V5357), Sand Coulee and Graybull populations (measurements taken from Kitts, 1956). There is no significant difference in rectangularity between M3 of Hyracotherium seekinsi and H. angustidens. Three specimens of Tetraclaenodon from the upper Nacimiento Forma- tion (AMNH 15927), one of Desmatoclaenus hermaeus, and one of D. para- creodus (Gazin, 1941), were plotted. While certainly not conclusive, the plots provide interesting comparisons with Hyracotherium seekinsi (Fig. 3). Rectangularity exhibited in M1 and M2 of Tetraclaenodon does not differ from that of Hyracotherium angustidens. It does differ considerably from the rectangularity of M1 and M2 of H. seekinsi. Desmatoclaenus hermaeus and D. paracreodus have the most rectangu- 10 Contributions in Science No. 151 lar M1 and M2 when compared to Hyracotherium angustidens, H. seekinsi and Tetraclaenodon. Hyracotherium seekinsi may well belong in the direct lineage leading to H. angustidens, and, in turn, could more reasonably be de- rived from Desmatoclaenus than from Tetraclaenodon. This supposition is suggested by the rectangularity of the anterior molars and strengthened by the biostratigraphic positions of the genera. It is also possible that Hyraco- therium seekinsi is a representative from an isolated population evolving from desmatoclaenid condylarths, while H. angustidens evolved from condylarths closer to tetraclaendontids. M3 of Tetraclaenodon, Hyracotherium angustidens, and H. seekinsi has about the same degree of rectangularity. Desmatoclaenus, on the other hand, has a very rectangular M3. This may be indicative of a primitive characteristic that was lost when the perissodactyl level of organization was attained. Conclusions Hyracotherium seekinsi differs from H. angustidens, its closest morpho- logical relative, in a number of details of cusp geometry in the upper molars, as well as in rectangularity of M1 and M2. Hyracotherium seekinsi, or a species much like it, could very well be ancestral to H. angustidens. On the other hand, Hyracotherium seekinsi may represent a population not on the lineage from a condylarth progenitor to H. angustidens. These are important considerations, as Hyracotherium seekinsi is certainly, morphologically, more like Desmato- claenus than like Tetraclaenodon. If Hyracotherium seekinsi is ancestral to H. angustidens, then Desmatoclaenus is a more satisfactory morphological an- cestor for these Early Tertiary perissodactyls than is Tetraclaenodon. Radinsky (1966, p. 408) states that if “. . . protoperissodactyls and Tetra- claenodon were independently derived from a still more primitive common ancestor, [this hypothesis] requires an additional complicating factor— an independent acquisition of molar hypocones by perissodactyls and phenaco- dontids.” Gazin (1941) however, does regard Desmatoclaenus hermaeus as having a weak hypocone, but not well developed when compared to that of T etraclaenodon . Hyracotherium angustidens is morphologically more like Tetraclaenodon, and Hyracotherium seekinsi more like Desmatoclaenus hermaeus. Either the hypothesis of a desmatoclaenid-//. seekinsi-H . angustidens lineage is tenable or independent derivation of H. seekinsi and H. angustidens appears neces- sary. Resumen Un nuevo perisodactilo encontrado en depositos que pertenecen a inicios del Terciario se describe como una nueva especie, Hyracotherium seekinsi. Este esta relacionado de cerca con H. angustidens. La rectangularidad de los molares superiores sugiere diferencias morfologicas entre H. angustidens, H. seekinsi y 1968 A New Tertiary Perissodactyl 11 Desmatoclaenus. De acuerdo con este criterio de rectangularidad, Desmato- claenus parece ser un ancestro de H. seekinsi, sin embargo, no parece ser un an- cestro directo de H. angustidens. Esta description de H. seekinsi se basa en unico molar colectado aproxi- madamente 25 Km. al Sur de Punta Prieta, Baja California. Se incluyen ilus- traciones de los dientes. Literature Cited Beal, Carl H. 1948. Reconnaissance of the Geology and Oil Possibilities of Baja California, Mexico. Geol. Soc. Amer., Mem., 31 : 1-138. Barton, N. H. 1921. Geologic Reconnaissance in Baja California. Jour. Geol., 29:720-748. Gazin, C. Lewis. 1941. The Mammalian Faunas of the Paleocene of Central Utah, with Notes on the Geology. U. S. Natl. Mus., Proc., 91 (3 121 ) : 1-53. Heim, Arnold. 1922. Notes on the Tertiary of Southern Lower California (Mex- ico). Geol. Mag., 59:529-547. Kitts, David B. 1956. American Hyracotherium (Perissodactyla, Equidae). Amer. Mus. Nat. Hist., Bull., 110(1): 1-60. Morris, William J. 1966. Fossil Mammals from Baja California: New Evidence on Early Tertiary Migrations. Science, 153:1376-1378. Patterson, Bryan. 1939. New Pantodonta and Dinocerata from the Upper Paleo- cene of Western Colorado. Field Mus. Nat. Hist., Geol. 6:351-384. Radinsky, Leonard B. 1966. The Adaptive Radiation of the Phenacodontid Con- dylarths and the Origin of the Perissodactyla. Evolution, 20:408-417. Simons, E. L. 1960. The Paleocene Pantodonta. Amer. Phil. Soc., Trans., 50:3-81. Vaughan, T. W. 1929. Descriptions of New Species of Foraminifera of the Genus Discocyclina from the Eocene of Mexico. U. S. Natl. Mus. Proc., 76 (2800) :1- 18. Wood, Roger C. 1967. A Review of the Clark Fork Vertebrate Fauna. Harvard Univ., Mus. Compar. Zook, Breviora, 257:1-30. Accepted for publication September 3, 1968. MUSEUM ANGELES COUNTY LOS CONTRIBUTIONS IN SCIENCE jjMBER 152 December 31, 1968 A NEW CENTRAL AMERICAN SAND FLY BREEDING IN CRAB HOLES (DIPTERA, CERATOPOGONIDAE) By Charles L. Hogue and Willis W. Wirth Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8V2 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor <4 EH P3 W r P E ERRATA lhe Editor and Editorial Committee of the Los Angeles County Museum of Natural History Contributions in Science regret that three references to scale of illustra- tions were incorrectly designated in OCCURRENCE OF A GIANT BISON, Bison latifrons, AND A SLENDER-LIMBED CAMEL, Tanupolama , AT RANCHO LA BREA, Los Angeles County Museum Contributions in Science, number 147. The corrections should read as follows : . Page Figure Published scale Corrected scale We also take this opportunity to stress that you make these changes in your copy of Contributions 147 rather than insert this mimeographed sheet. We regret these inconveniences. 2 5 6 1 5 5 1 Dorothy M. Halmos, Editor Robert J. Lavenberg, Managing Editor A i A NEW CENTRAL AMERICAN SAND FLY BREEDING IN CRAB HOLES (DIPTERA, CERATOPOGONIDAE) By Charles L. Hogue1 and Willis W. Wirth2 Abstract: The larva, pupa, and adult of a new species of sand fly, Culicoides cancer, are described and figured from mate- rial collected in Costa Rica. The species is a member of the Fur- ens Group but is unique in its lack of anthropophily and in its breeding site, the burrows of the land crab, Cardisoma crassum. Sand flies of the genus Culicoides almost universally utilize, as their larval habitat, the shore lines of tidal swamps and estuaries where they often become extremely numerous. If the females of the species show a preference for human blood, they may become almost intolerable pests when man invades this environment. One group of salt marsh sand flies which has become notori- ous in the American tropics is the Furens Group, which contains four species, C. furens (Poey), C. barbosai Wirth and Blanton, C. gorgasi Wirth and Blan- ton, and C. alahialinus Barbosa, all causing annoyance to man. The senior author recently discovered a fifth member of this group living in crab holes on the Pacific coast of Costa Rica. During the six-week period of these investigations, this crab hole species was never observed to bite man, although C. gorgasi and C. barbosai (during other periods in the study area) were frequently taken while biting. Culicoides furens has been collected while biting man at Golfito, also on the Pacific coast of Costa Rica. At Boca de Barranca, Golfo de Nicoya, and at Playas del Coco, Peninsula de Nicoya, the new species, which we are naming C. cancer because of its association with land crabs, was collected and reared from 36 separate collec- tions from burrows of the mouthless crab, Cardisoma crassum Smith, and from one burrow of the wide red land crab, Ucides occidentalis (Ortmann) . It was not found on the Atlantic coast, but two collections from burrows of Cardisoma guanhumi Linnaeus at Cahuita (22 miles south of Limon) yielded a species of another group, Culicoides reticulatus Lutz, which has been reared from crab holes in Brazil (Forattini, Rabello, and Pattoli, 1958). A third species, C. arubae Fox and Hoffman, has been reported from crab holes in the Dutch West Indies (Fox, 1942: 420; Fox and Hoffman, 1944: 109), but according to Jones and Wirth’s (1958) observations in Texas, it is more likely to be found in open salt water pools. 1Senior Curator of Entomology, Los Angeles County Museum of Natural History, Los Angeles, California. Systematic Entomology Laboratory, Entomology Research Division, Agricultural Research Service, USD A, c/o U. S. National Museum, Washington, D. C. 20560. 1 2 Contributions in Science No. 152 Culicoides cancer new species (Figs. 1-13) Diagnosis : This species is most similar to Culicoides gorgasi Wirth and Blanton, but may be separated as follows: Antennal sensorial gorgasi cancer pattern 3, 8-10 3, 6-10 Halter color dark pale Leg bands Apices wing veins faint dark M2 and M3+4 dark pale Female eyes Female spermathecae contiguous narrowly separated necks not tapered tapered Male parameres no ventral lobe large ventral lobe Male aedeagus Male apicolateral simple tip 3-pointed tip processes short long Description (terminology and abbreviations follow Jamnback, 1965: 15-20): FEMALE. Length of wing 0.91 mm, breadth 0.45 mm. Head: Eyes (Fig. 4) bare, separated by a distance equal to the diameter of an eye facet. Antenna (Fig. 3) with lengths of flagellar segments in propor- tion of 16-12-13-13-14-13-13-13-19-20-23-26-35, AR 1.14; distal sensory tufts present on segments three (two per segment), six (one), seven to ten (three per segment). Palpal segments (Fig. 2) with lengths in proportion of 5-10-12- 5-5; third segment moderately swollen, with a small, round, moderately deep, sensory pit; PR 2.4. Proboscis moderately long, P/H ratio 1.2; mandible with 16 teeth. Thorax: Yellowish brown; scutum with prominent pattern (Fig. 5) of dark brown punctures at bases of the coarse, sparse, setose vestiture; scutellum pale on sides, dark brown in middle; postscutellum and pleuron brown. Legs brown, knee spots blackish; trochanters and bases of femora pale; femora with subapical, and tibiae with subbasal, narrow pale rings, hind tibia pale on distal fourth; hind tibial comb with four spines, the one nearest the spur the longest. Wing (Fig. 1): Pattern as figured, pale spots distinct, creamy colored; second radial cell dark to tip; base of wing broadly pale except at basal arculus; large pale spot over r-m crossvein extending broadly from costa to media; post-stigmatic pale spot small, confined to anterior wing margin just past second radial cell; a second, separate, elongate pale area lying behind it on anterior side of vein and extending proximad for length of second radial 1968 A New Central American Sand Fly 3 Figures 1-8. Culicoides cancer n. sp. 1-6, Female: (1) wing; (2) palpus; (3) antenna; (4) eye separation; (5) thoracic pattern; (6) spermathecae. 7-8, Male: (7) para- meres; (8) genitalia, parameres removed. cell; distal pale spot in cell R5 large, quadrate to trapezoidal in form, extending broadly from anterior wing margin to vein M4; cell with two elongate pale spots, the second continued as a narrow line to wing margin; cell M2 with pale streak from base to level of middle of cell M4, a separate, large, rounded pale spot apically at wing margin; cell M4 with large rounded pale area broadly meeting wing margin; anal cell with two distal pale spots and pale streaks basally and at margin on anal angle; vein M4 pale bordered more than halfway to base, apices of veins M2 and M3+4 with small pale spots at wing margin. Macrotrichia moderately dense and long on distal portion of wing, continued very sparsely to bases of cell M2 and anal cell; CR 0.6; radial cells with distinct lumens. Halter pale. Abdomen: Brownish. Spermathecae (Fig. 6) two, plus rudimentary third 4 Contributions in Science No. 152 Figures 9-13. Culicoides cancer n. sp. 9-11, Larva: (9) head and thorax; (10) ter- minal segments and setae; (11) frontoclypeus. 12-13, Pupa: (12) thorax and head sclerites; (13) terminal abdominal segments. 1968 A New Central American Sand Fly 5 and sclerolized ring; slightly unequal in size, measuring 0.065 mm by 0.038 mm and 0.053 mm by 0.035 mm; ovoid in shape with very long, tapering, slender necks. MALE. Similar to the female with the usual sexual differences; antennal plumes well developed. Genitalia (Figs. 7-8) : Ninth sternum with moderately broad and deep caudomedian excavation, the ventral membrane bare; ninth tergum moderately long and slightly tapering, with moderately long and slen- der, pointed, slightly divergent, apicolateral processes, the caudal margin between them slightly convex. Basistyle moderately slender, slightly tapering, ventral root foot-shaped, with well-developed caudal heel and sharp antero- median toe; dorsal root slender; dististyle slender and gradually curved, with abruptly bent, sharp tip. Aedeagus with broad, rounded, anterior basal arch extending to 0.6 of total length, the basal arms slender; distal portion broad on shoulders, tapering distally, with three subequal sharp points apically, the lateral pair abruptly bent ventrad. Parameres (Fig. 7) each with small basal knob, stem slender and sinuate, subapically with a small ventral lobe; distal constricted portion bent ventrad and mesad and tapering to sharp point with fringe of four to five sharp lateral spines. MATURE LARVA (Figs. 9-11). Head pale yellow, elongate, length of frontoclypeus 130 (n = 3) microns; comb with 5-8 (average 6.8, n = 9) teeth on each side. Thorax without pronounced pigmented areas. Setae on last segment as figured. PUPA (Figs. 12-13). Respiratory horn light brown, similar in color to rest of pelt except slightly darkened apically and basally, typically with four apical and three lateral spiracular openings on pronounced protuberances, with weak transverse convolutions at midlength, without spines; horn slender, widest near base, Length/Width ratio 6. 3-7. 2. Operculum with short, broad- based spines abundant over most of surface; am setae about one-half as long as maximum width of operculum. The d tubercles arranged as figured, setae 1 and 2 short and heavy, not overlapping; seta 4 short. Abdomen with Imp tubercles bifid with sharp points and a subapical seta. Last segment with sparse weak spines cephalad, none on disc; caudal apicolateral processes with a few spines caudad, apex only darkened, directed caudad at an angle of approxi- mately 40 degrees to the longitudinal axis of the body. Distribution : Pacific coast of Costa Rica. Material : HOLOTYPE. Female, allotype male, Boca de Barranca, Golfo de Nicoya, Puntarenas Prov., Costa Rica, 27 June 1967, C. L. Hogue and D. B. Bright, reared from burrow of Cardisoma crassum, LCBA 111 (deposited in Los Angeles County Museum of Natural History). PARATYPE. 21 males, 18 females, same data as holotype, except dates 20 June to 11 July 1967 (LCBA 2, 11, 112, 126, 147, 135, 142, 177, 111, 160, 142). Also one male, one female, same data but from burrows of Ucides 6 Contributions in Science No. 152 occidental is, 10 July 1967 (field no. CR 240). Also five males, one female, Playas del Coco, Peninsula de Nicoya, Guanacaste Prov., 18 July 1967, C. L. Hogue and D. B. Bright, from burrows of C. crassum (LCBA 230) . ADDITIONAL MATERIAL. Numerous specimens, primarily of larvae and pupae, Boca de Barranca, Golfo de Nicoya, Puntarenas Prov., Costa Rica (LCBA collections as follows: 2, 3, 11, 104-106, 111-112, 123-126, 135, 137- 139, 141-144, 147-150, 159-161, 163-165, 167-168, 170-171, 177) and Playas del Coco, Peninsula de Nicoya, Guanacaste Prov., Costa Rica (LCBA 205 and 230) . All collected by C. L. Hogue and D. B. Bright. Biology : All of the present material was collected during mid-rainy season in burrows of moderate to large diameter (1.5 to 5 inches) occupied by half-grown to mature crabs. The larvae were extremely numerous at times, breeding in the water which collects from ground water seepage and rainfall. The burrow water varied considerably in solute concentrations, depending on the proximity of a particular burrow to, and the nature of, a nearby water source (fresh-water stream, tidal inlet, tidal mangrove thicket or the open sea). Im matures were collected primarily from burrow water with low salt concentrations (ranging from 15-580 ppm NaCl) though they were frequently found, sometimes thriving in dense numbers, in water with moderate (1900- 8600 ppm NaCl) or even high (13,000-26,000 ppm NaCl) salt concentra- tions. A few burrows emitted strong sulfide odors and contained black water, indicating that considerable organic decomposition was taking place therein; though in most cases the water was less putrid, it was always turbid (dark brown to grey in color) and rich in organic matter (principally plant fragments brought into the burrow by the crab host). The foregoing indicates that the larvae and pupae of this species have a wide range of tolerance in regard to variation in the chemistry of its medium. This tolerance is shared by most of its associates and appears to be a general characteristic of estuarine and crabhole invertebrates, doubtless an adaptation permitting them to survive the extreme fluctuations, mainly in salt concentra- tion, common to their environment. Associated with Culicoides cancer at Boca de Barranca and Playas del Coco were crabhole mosquitoes ( Deinocerites pseudes Dyar and Knab, and Deinocerites sp. A. Belkin and Hogue) , Culex inflictus Theobald, unidentified water mites, and the larva of an unidentified marsh beetle (Helodidae). Adult Culicoides cancer rest on the walls of the burrow throat and mouth. They are agitated readily by debris falling into the burrow and by any object (such as an aspirator) being thrust into the burrow. Under these circumstances they crawl rapidly over the soil inside the burrow and fly swiftly out of the burrow mouth, landing on the soil surface or vegetation very close by. They do not hover about the opening attempting to return, a habit often displayed by Deinocerites. On no occasion did specimens attempt to bite the senior author while he was collecting. 1968 A New Central American Sand Fly 7 Acknowledgments : The material upon which this paper is based was collected as a part of a general study of the biology of land crabs and their burrow associates (LCBA) being conducted by the senior author and Donald B. Bright, California State College, Fullerton, with the support of a grant from the American Philosophical Society. The collections of Culicoides barbosai reported upon herein were made incidentally to a mammalian ectoparasite survey of Costa Rica (Los Angeles County Museum of Natural History) under grant numbers DA-MD-49-193-62-G54 and -63-G94, U. S. Army Medical Research and Development Command. SUMARIO Las larvas, pupas y adultos de una nueva especie de jejen, Culicoides cancer, han sido descritas y dibujadas basadas en ejemplares colectados en Costa Rica. Este especie es un miembro del Grupo Furens pero se diferencia en su falta de antropofalia y en el medio que usa para desarrollarse, el cual es en los agujeros del cangrejo terrestre, Cardisoma crassum. Literature Cited Forattini, O. P., E. X. Rabello, and D. Pattoli. 1958. Culicoides da regiao neo- tropical (Diptera, Ceratopogonidae). II — Observances sobre biologia em con- digoes naturais. Arqiv. Fac. Higiene e Saude Publica da Univ. de Sao Paulo, 12: 1-52. Fox, I. 1942. The respiratory trumpet and anal segment of the pupae of some species of Culicoides (Diptera: Ceratopogonidae). Puerto Rico J. of Public Health and Trop. Med., 17: 412-425. Fox, I., and W. A. Hoffman. 1944. New Neotropical biting sandflies of the genus Culicoides (Diptera: Ceratopogonidae). idem, 20: 108-111. Jamnback, H. 1965. The Culicoides of New York state (Diptera: Ceratopogonidae). N. Y. State Mus., Sci. Ser., Bull. 399: 1-154. Jones, R. H., and W. W. Wirth. 1958. New records, synonymy, and species of Texas Culicoides (Diptera, Heleidae). J. of the Kan. Entomol. Soc., 31: 81-91. Accepted for publication October 3, 1968. LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE umber 153 December 31, 1968 r.-'\ V A NEW PICULET FROM SOUTHEASTERN PERU By Kenneth E. Stager i Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8V2 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor A NEW PICULET FROM SOUTHEASTERN PERU By Kenneth E. Stager1 Abstract: A new species of Picumnus from southeastern Peru is described and named Picumnus subtilis. The relationships of Picumnus subtilis and Picumnus castelnau are discussed. During a revisional study of the neotropical pygmy woodpeckers of the genus Picumnus , an undescribed species has been found and, in anticipation of the complete revision of this genus, is named as follows: Picumnus subtilis, new species Holotype : Adult male, Figs. 1-2, collected at Hacienda Villacarmen, De- partment of Cuzco, S. E. Peru, at an elevation of 1000 meters (3280 ft), July 9, 1958, by Emmet R. Blake. FMNH 251769. Diagnosis : Closely resembles and previously confused with Picumnus castelnau of northern and central Peru. Is readily distinguished from this spe- cies, however, by possession of distinct white spotting on the crown and nape and by the presence of barring on the chest. These characters are in direct con- trast to the immaculate crown and chest of Picumnus castelnau. Description of Type : (Capitalized color terms are those of Ridgway, 1912.) Crown and nape deep black with each feather tipped with white or Orange Chrome. Feathers tipped with Orange Chrome extend from nasal area posteriorly for at least two-thirds of crown, with the amount of Orange Chrome tipping increasing posteriorly. Deep black feathers of the nape tipped with a circular dot of pure white. Lateral feathers of crown tipped with smaller dots of white, with a concentration of white dots above the eye, forming a small supraocular stripe. Entire back, rump and shoulders Yellowish Olive. Each feather double barred with Olive Yellow, imparting a pattern of faint barring to the dorsum. Primaries, secondaries and tertials Fuscous Brown with outer edge of secondaries and coverts Olive Yellow. Chin and throat dull white. Feathers of breast gray, barred and tipped with Straw Yellow. Proximal halves of abdominal feathers gray with distal halves heavily washed with Straw Yel- low. Upper tail coverts barred like the back; undertail coverts Straw Yellow like the abdomen; tail black with middle pair of rectrices clear white on inner webs; underwing coverts and axillaries Straw Yellow. Wing, 50.0 mm; tail, 26.5 mm; culmen, 13.0 mm. Measurements: Total of 12 specimens, including holotype. Adult males (7), wing 50.0-55.0 mm (average 51.9 mm); tail 25.0-29.0 mm (average 26.7 mm); culmen 12.0-14.5 mm (average 13.0 mm). Adult females (1), wing 52.0 mm; tail 27.0 mm; culmen 12.5 mm. Immatures (3 $ $ , 1 ?), wing 50.0- 5 Senior Curator of Ornithology, Los Angeles County Museum of Natural History. 1 Contributions in Science No. 153 Figure 1. Dorsal view of Picumnus subtilis sp. nov. and Picumnus castelnau. From left to right, $ P. subtilis (type), $ P. subtilis, $ P. castelnau, and $ P. castelnau. 54.0 mm (average 51.6 mm); tail 24.0-27.0 mm (average 25.9 mm); culmen 12.0 mm. Range : Known from a total of 12 specimens from the tropical foothills (Yungas) of east central Peru (Santa Rosa, upper Rio Ucayali, Dept, of Lo- reto, 200 m elev.) to southeastern Peru (Hacienda Villacarmen and Hacienda Cadena, Dept, of Cuzco, 1000 m elev. and Candamo, Dept, of Puno, 302 m elev.). Remarks: Since the collection of the first specimen of this new piculet by H. Watkins at Candamo, Dept, of Puno, Peru, in 1916, its identity has re- mained obscure because of the tendency of workers to assign it to the species Picumnus castelnau, which it superficially resembles. The spotting of the crown and the barring of the chest of P. subtilis, however, immediately distin- guish it from the latter (see Figs. 1 and 2). Despite the superficial resemblance to P. castelnau mentioned above, it is believed that this new piculet shows a definite affinity for the spot-crowned piculets rather than the immaculate 1968 New Piculet from Peru 3 crowned group consisting of P. castelnau and P. fuscus. At present P. subtilis and P. castelnau are known to be sympatric only in the area of the upper Rio Ucayali near its confluence with the Rio Urubamba. In the area of its great- est known abundance, in southeastern Peru, P. subtilis shares an area of sym- patry with Picumnus aurifrons. The type locality, Hacienda Villacarmen (12° 51' S, 71° 15' W), is de- scribed by E. R. Blake as “—lying just within the Dept, of Cuzco, being sepa- rated from the Dept, of Madre de Dios by the Rio Pina Pina, a minor tributary of the Rio Alto Madre de Dios— on the road from Cuzco to Manu.” (pers. comm.) The subtle yet distinct differences between Picumnus castelnau and this new piculet make the specific name subtilis seem appropriate. Specimens Examined : (Holotype and 11 paratypes) American Museum of Natural History (AMNH), 6 (2 $ $ , l $ , 3 imm.), Santa Rosa, upper Rio Ucayali, Department of Loreto, Peru; Candamo, Department of Puno, Peru. Figure 2. Ventral view of Picumnus subtilis sp. nov. and Picumnus castelnau. From left to right, $ P. subtilis (type), $ P. subtilis, $ P. castelnau, and $ P. castelnau. 4 Contributions in Science No. 153 Field Museum of Natural History (FMNH), 2 3 3, Hacienda Villacarmen, Department of Cuzco, Peru (holotype) and Hacienda Cadena, Department of Cuzco, Peru. Peabody Museum of Yale University (PMYU), 4(333,1 2 imm.), Hacienda Cadena, Department of Cuzco, Peru. For the loan of comparative material of Picumnus subtilis and Picumnus castelnau, I am indebted to the American Museum of Natural History, the Car- negie Museum, the Field Museum of Natural History, the Louisiana State Uni- versity Museum of Zoology, and the Peabody Museum of Yale University. This study has been supported in part by a grant from the Frank M. Chap- man Memorial Fund of the American Museum of Natural History. SlJMMARIO En un estudio de revision de los pequenos Pajaros Carpinteros del genero Picumnus de Centro y Sur America se encontro una especies no descrita en especimens de la region sur-este del Peru. Esta nueva especie ha sido nombrada Picumnus subtilis, y es superficial- mente parecida a Picumnus castelnau, aunque no estan cercanamente relacio- nadas. El macho y la hembra de P. castelnau presentan crestas sin manchas, mientras que tanto el macho como la hembra de P. subtilis tienen manchas blancas en la cresta. P. subtilis fue identificada estudiando 12 especimens, en- cluyendo el holotipo. Literature Cited Ridgway, Robert. 1912. Color standards and color nomenclature. Washington, D. C., 1-44, pis. 1-53. Accepted for publication October 3, 1968. LOS ANGELES COUNTY MUSEUM CONTRIBUTIONS IN SCIENCE TBER 154 December 31, 1968 k 13 A NEW SPECIES OF EURHOPALOTHR1X FROM EL SALVADOR (HYMENOPTERA: FORMICIDAE) By Roy R. Snelling Los Angeles County Museum of Natural History • Exposition Park Los Angeles, California 90007 CONTRIBUTIONS IN SCIENCE is a series of miscellaneous technical papers in the fields of Biology, Geology and Anthropology, published at irregular intervals by the Los Angeles County Museum of Natural History. Issues are numbered sepa- rately, and numbers run consecutively regardless of subject matter. Number 1 was issued January 23, 1957. The series is available to scientific institutions on an ex- change basis. Copies may also be purchased at a nominal price. Inquiries should be directed to Robert J. Lavenberg, Los Angeles County Museum of Natural History, 900 Exposition Boulevard, Los Angeles, California 90007. INSTRUCTIONS FOR AUTHORS Manuscripts for the LOS ANGELES COUNTY MUSEUM CONTRIBU- TIONS IN SCIENCE may be in any field of Life or Earth Sciences. Acceptance of papers will be determined by the amount and character of new information and the form in which it is presented. Priority will be given to manuscripts by staff members, or to papers dealing largely with specimens in the Museum’s collections. Manuscripts must conform to CONTRIBUTIONS style and will be examined for suitability by an Editorial Committee. They may also be subject to critical review by competent specialists. MANUSCRIPT FORM.— (1) the 1964 AIBS Style Manual for Biological Journals is highly recommended as a guide. (2) Typewrite material, using double spacing throughout and leaving ample margins, on only one side of 8V2 x 11 inch standard weight paper. (3) Place tables on separate pages. (4) Footnotes should be avoided if possible. (5) Legends for figures and unavoidable footnotes should be typed on separate sheets. Several of one kind may be placed on a sheet. (6) A factual summary is recommended for longer papers. (7) A brief abstract must be included for all papers. This will be published at the head of each paper. ILLUSTRATIONS. — All illustrations, including maps and photographs, should be referred to as “figures!’ All illustrations should be of sufficient clarity and in the proper proportions for reduction to CONTRIBUTIONS page size. Permanent ink should be used in making line drawings and in lettering (do not type on drawings); photographs should be glossy prints of good contrast. Original illustrations will not be returned unless specifically requested when the manuscript is first submitted. PROOF. — Author will be sent galley proof which should be corrected and re- turned promptly. Changes after the paper is in galley will be billed to the author. Un- less specially requested, page proof will not be sent to the author. 100 copies of each paper will be given free to a single author or divided equally among multiple authors. Orders for additional copies should be sent to the Editor at the time corrected galley proof is returned; appropriate forms for this will be included when galley is sent. Dorothy M. Halmos Editor A NEW SPECIES OF EURHOPALOTHRIX FROM EL SALVADOR (HYMENOPTERA: FORM1CIDAE) By Roy R. Snelling1 Abstract: A new species of basicerotine ant is described from two worker specimens from El Salvador, Central America. This new species, placed in the genus Eurhopaloihrix, is a mem- ber of the E. bolaui group and is most closely related to E. speci- osa. It differs from this and other species in the number and arrangement of specialized hairs on the head and body. Several years ago I was given a number of miscellaneous ants by Dr. R. O. Schuster, University of California, Davis. A single vial of Berlese sample material from El Salvador yielded six basicerotine ants; two specimens each of Octostruma balzani (Emery), a new species of Eurhopalothrix described below, and a new genus and species. The latter has been described by Brown and Kempf (1968). Dr. Brown first recognized the following species as new and returned it to me for description, along with specimens of E. speciosa Brown and Kempf for comparison. I wish to thank both Dr. Brown and Dr. Schuster for their assistance. Eurhopalothrix apharogonia new species Fig. 1 Diagnosis : A new species of the E. bolaui group as defined by Brown and Kempf (1960). It may be distinguished from other members of the group by the following combination of characters: four erect specialized hairs present on cephalic dorsum, no clavate hairs on posterior occipital angles, no erect clavate hairs on pronotal dorsum and posterior part of mesoscutum. The abbreviations used in the following description are those of Brown and Kempf (1960). Holotype worker: TL, 3.87; HI, 0.93; HW, 0.84 (Cl, 90); scape L, 0.53; maximum diameter of compound eye, 0.03; WL, 1.12 mm. Form of head and body as shown in figures. Appressed and subappressed ground pilosity similar to E. speciosa, i.e., consisting largely of simple hairs, which are rather dense on mandibles and clypeus, sparse on petiolar nodes and gastric dorsum, very sparse and incon- spicuous on vertex, occiput and thoracic dorsum; simple appressed hairs of tibiae replaced largely by appressed and decumbent spatulate or broadened hairs; larger specialized hairs thick-squamiform, smaller than corresponding hairs in E. speciosa ; reduced in number on head, consisting of four rectangu- larly arranged hairs medially on occiput. Humeral pair absent in both speci- 1Entomology Section, Los Angeles County Museum of Natural History. 1 2 Contributions in Science No. 154 Figure 1. Eurhopalothrix apharogonia, new species, a, lateral aspect; b, frontal as- pect of head; c, right mandible, enlarged; d, dorsal aspect of alitrunk and gaster. Il- lustrations by Ruth Ann DeNicola. 1968 New Species of EURHOPALOTHRIX 3 mens, presumably a real condition; a single pair present near middle of meso- notum, posterior pair lacking; postpetiole without dorsal clavate hairs. Median hairs lacking on first gastric segment of holotype (paratype has one median clavate hair; presumably one has been rubbed off). Promesonotal suture obsolete; metanotal groove present but very poorly defined. Body somewhat shiny between moderate, rather close punctures; gastric pubescence sparser than that of thoracic dorsum; cephalic punctures denser on cephalic dorsum; finer on clypeus and distinctly separated. Mandi- bles shiny, finely punctate; masticatory margin with thirteen teeth, an outer set of ten, and an upper, inner set of three longer, spikelike teeth (Fig. 1). Color medium ferruginous, legs and antennae more yellowish. Holotype : University of California, Davis, worker, EL SALVADOR, 4 miles north of Quetzaltepec, 7 July 1961, collected by M. E. Irwin. Paratype : Worker, with same data as holotype, Los Angeles County Museum of Natural History. Although E. apharogonia is one of the larger members of the E. bolaui group, it agrees with the smaller, more specialized species in the reduction of its erect pilosity. It is especially close to E. speciosa and E. floridana Brown and Kempf. From both of these it may be separated by the characters given in the key below, which is a modification of the appropriate portion of the key of Brown and Kempf (1960). The apical margins of the gastric segments are devoid of erect clavate hairs in the two specimens available to me. However, since these are consistently present in all other species of the group, as a group character, it seems logical that they are normally present in E. apharo- gonia also. 7. Dorsum of head with a single pair of erect clavate hairs on vertex, three pairs of erect clavate hairs on alitrunk, including one pair on pronotum (S Florida) floridana Brown and Kempf Dorsum of head with two or three pairs of erect clavate hairs; one or two pairs of erect clavate hairs on alitrunk, none present on pronotum 8 8. Dorsum of head with three pairs of erect clavate hairs, one on each posterior occipital corner and two pairs arranged in a close rectangle on vertex; mesonotum with two pairs of erect clavate hairs (SE Brazil) speciosa Brown and Kempf Dorsum of head with two pairs of erect clavate hairs, arranged in a close rectangle on vertex; mesonotum with a single pair of erect clavate hairs (El Salvador) apharogonia Snelling The name of this new taxon is derived from the Greek aphares, unclad, and gonia, angle or corner, in allusion to the lack of clavate hairs on the posterior occipital angles. 4 Contributions in Science No. 154 Resumen Una nueva especie de hormiga basicerotina, Eurhopalothrix apharogonia, es descrita como nueva especie basado en el estudio de dos ejemplares de hormigas obreras. Estas hormigas fueron obtenidas cuatro millas al Norte de Quetzaltepec, El Salvador, America Central. Esta nueva especie ha sido colo- cada en el genero Eurhopalothrix. Es un miembro del grupo E. bolaui y esta estrechamente relacionada con E. speciosa. Esta difiere de todas otras especies del grupo E. bolaui en la combinacion de caracteres que sigue: cuatro pelos especializados y erectos en el dorsum cefalico, ausencia de pelos en forma el remo en los angulos occipitales posteriores y ausencia de pelos erectos en forma de remo en la region pronotal dorsal y la parte posterior del mesoscutum. Literature Cited Brown, W. L., Jr., and W. W. Kempf. 1960. A world revision of the ant tribe Basi- cerotini. Studia Entomologica, (n.s.) 3:161-250. Accepted for publication August 14, 1968. m 2 ' _i 2 -J 2 ',v INSTITUTION NQU.niU.SNI NVINOSHiUMS S3 1 avaa IT LIBRARIES SMITHSON 21 r* i— 2 O S3 1 avaan li brar i esu Smithsonian institution ^NOiiniiiSNi nvinoshi 5 o®' g 2 vV,- in z „ . 1 1 I w i institution NouniiisNi NviNosHiiws^sa i ava a nzLt brar i es ^smithson 00 n on ~ co UJ s?a saiavaan libraries Smithsonian institution Noiimiism nvinoshi m iNSTiTUTioN^NoiiniusNi ""nvinoshiiws^sh i a va an “LI BRAR I es^smithson INSTITUTION NOlinillSNfiNVINOSHiIWSZ S3 I a Va 3 I T * L ! BRAR I ES^SMITHSON ^ — J 2 _ J 2 — JTUTION NOU.fUIJ.SNr NVlNOSHilWS S3ldVM8n LIBRARIES SMITHSONIAN I 2 r* ^ r- 2 r* “ PTI ' * ' ~ x o __ m ^ rn avaan LIBRAR I ES^SMITHSONIAN INSTITUTION ^OlxnillSNrNVINOSHlIWS^S " 2: v.,. 00 2: ^ c o v < s .< ftv 2: t\ o CO .■?; x 5 ''Wf S > (/) ** 2 ’ CO 2 ITUTION NOUfUUSNI NVINOSHJ.IWS S3iaVdail LIBRARIES SMITHSONIAN I w “* <0 “CO- CO CO O " , — o ^D1-/ “ '3^ O { V B 9 11 LIBRARI ES SMITHSONIAN^INSTITUTlON^NOliniliSNI^NVINOSHlIINS^S 2 r~ 2 r* > z 03 > ter 33 % KAsv^i — m \^yosv^x zj, rn x^vosv^ x^vasvvx' m co £ co £ co __ TUTION NOIJLfUllSNI NVIN0SH1IWS SHiaVaail LIBRARIES SMITHSONIAN If * 92 2 co T I, > ' S X^vosh^X > S' > , , 2 CO 2 CO 2 cn Vaan_UBRARIES SMITHSONIAN INSTITUTION NOIlfUllSNI NViNOSHillNS s 2 ^ to — <0 2 \ ^ co xT^^oX w fh uj V uj TUTION NOlilUllSNI NVINOSHillNS SBiaVaail LIBRARIES SMITHSONIAN If 3D X x 403 3 -^6# ? m ^ g xioms^x m vgy. dS a*. 92 2 _ co v- 2 . <£ Xv^IP^aX 2 X 2 *;& Jpjy O /££X x ■ 60 wZr to CO _ 2 *— Vcl^^^v 2 > X^iiis^X ^ X > -«fP^ 2 d°^X ^ 2 (/3 * 2 CO 2 TUTION^NOUfUliSNI NVINOSHXIWS SBiaVaail LIBRARIES SMITHSONIAN If S ..... co = co —