HARVARD UNIVERSITY »» m LIBRARY Museum of Comparative Zoology s - .^ ^^ JUL 7 195d TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY VOLUME XI Printed from the W. W. Whitney Publication Endowment .s^^^'ii'^^^^" v^ i^li SAN DIEGO, CALIFORNIA Printed for the Society 1946-1953 COMMITTEE ON PUBLICATION Joshua L. Baily Carl L. Hubbs Laurence M. Klauber Charles C. Haines CONTENTS OF VOLUME XI 1. The Gopher Snakes of Baja Cahfornia, with Descriptions of New Subspecies of Pituophis Catenifer. By Laurence M. Klauber. Published August 26, 1946 1-40 2. A New Gopher Snake (Pituophis) from Santa Cruz Island, California. By Laurence M. Klauber. Published August 26, 1946 -.- 41-48 3. An Undescribed Race of the Mangrove Warbler from Baja California, Mexico. By A. J. van Rossem. Published February 25, 1947 49-52 4. Three New Races of Pocket Gophers (Thomomys) from Baja California, Mexico. By Laurence M. Huey. Pub- lished January 31, 1949 53-56 5. The Relationship of Crota'us Ruber and Crotalus Lucasen- sis. By Laurence M. Klauber. Published January 31, 1949 . - - 57-60 6. Some New and Revived Subspecies of Rattlesnakes. By Laurence M. Klauber. Published January 31, 1949 61-116 7. A New Fossil Chilopod from the Late Cenozoic. By Ralph V. Chamberlin. Published April 29, 1949 117-120 8. The Subspecies of the Ridge-Nosed Rattlesnake, Crotalus Willardi. By Laurence M. Klauber. Published Septem- ber 30, 1949 121-140 9. The Shovel-Nosed Snake, Chionactis, with Descriptions of Two New Subspecies. By Laurence M. Klauber. Pub- lished April 30, 1951 - 141-204 10. The Kangaroo Rats (Dipodomys) of Baja California, Mex- ico. By Laurence M. Huey. Published April 30, 1951.... 205-256 11. The Vogdes Collection of Trilobites. By B. F. Howell. Published April 30, 1951 257-328 12. Two Californian Mountain Snails of the Genus Helmintho- glypta — a Problem in the Relationship of Species. By S. Stillman Berry. Published April 17, 1953 329-344 13. Mollusks from Clipperton Island (Eastern Pacific) with the Description of a New Species of Gastropod. By Leo George Hertlein and William K. Emerson. Published July 22, 1953 - 345-364 14. The Amphibians and Reptiles from Rancho La Brea. By Bayard H. Brattstrom. Published August 4, 1953 365-392 15. West American Razor-Clams of the Genus Ensis. By S. Stillman Berry. Published August U, 1953 393-404 16. Notices of New West American Marine Mollusca. By S. Stillman Berry. Published September 1, 1953 405-428 17. A Pleistocene Invertebrate Fauna from the Southwest Cor- ner of San Diego County, California. By William K. Emerson and Warren O. Addicott. Published November 10, 1953 429-444 TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 1, pp. 1-40, plates 1-2, map THE GOPHER SNAKES OF BAJA CALIFORNIA, WITH DESCRIPTIONS OF NEW SUBSPECIES OF PITUOPHIS CATENIFER f* laaR J BY Laurence M. Klauber Curaior of Reptiles and Amphibians, San Diego Society of Natural History SAN DIEGO, CALIFORNIA Printed for the Society August 26, 1946 TABLE OF CONTENTS Page Introduction ■ ^ Summary of Relationships 1 Descriptions of New Forms - / Pituophis cdtenijer bimaris subsp. nov - 7 Pttuophis catenijer insidanus subsp. nov 11 Pituophis catenijer fuliginatus subsp. nov ...14 Pituophis catenijer coronalis subsp. nov 19 Redescriptions of Old Forms 22 Pituophis catenijer vertebralis (Blainville) 22 Pituophis catenijer annectens Baird and Girard - 25 Pituophis catenijer ajjims Hallowell 28 Key to Subspecies of Pituophis catenijer in Baja California 31 Acknowledgments ,- — 32 Summary " 33 Bibliography 33 THE GOPHER SNAKES OF BAJA CALIFORNIA, WITH DESCRIPTIONS OF NEW SUBSPECIES OF ^ff ..--"'•^ PITUOPHIS CATENIFER 4'* Zoolofly *^t StPtOUlfi I BY Laurence M. Klauber Curator of Reptiles and Amphibians, San Diego Society of Natural History Introduction t There have been lately added to the collection of the San Diego Society of Natural History, and to my own, specimens of gopher snakes of the genus Pituophis from various localities in Baja California, Mexico. Some of these are from areas of importance in determining relationships, and it therefore has seemed desirable to resurvey the snakes of the peninsula. Through the courtesy of a number of insti- tutions I have had the privilege of examining their collections from Baja California. As a result, I find four new subspecies to be dis- tinguishable, one mainland and three bland. These I shall describe, following which the previously known forms, as newly delimited, will be summarized. These determinations have been facilitated by recently acquired material from the southwestern United States not available to those who have previously made revisions of the genus (Van Den- burgh and Slevin, 1919; Van Denburgh, 1920; Miss Stull, 1932, 1940). I expect to discuss the classification of some of the forms not occurring in Baja California in a subsequent paper. Summary of Relationships Before undertaking the descriptions of the new forms, it is necessary to summarize my reasons for considering all of them, and in fact all the gopher or bull snakes of the region west of the Mississippi River (except rutbveni), to be subspecies of a single form, Pituophis catenifer. I find an unbroken chain of intergradation connecting them all, either evident or strongly indicated. First, the subspecies catenifer catenifer intergrades with annectem north of Santa Barbara, California. These two subspecies are not sharply differentiated and therefore the zone of intergradation is broad. Catenifer intergrades with deserticola in the western Mojave Desert at the foot of the Tehachapi Mountains, and in the Lassen and the Klamath 6 San Diego Society of Natural History areas. Similarly, muiectens intergrades with deserticola where the San Gabriel and San Bernardino ranges descend into the Mojave Desert. Deserticola, in turn, intergrades with affinis along the northern border of Riverside County, California, and across the Grand and Marble canyons, and to the south of Kayenta, in northern Arizona. Affinis and sayi intergrade along a broad contact from central New Mexico south across western Texas info Chihuahua and Coahuila. Deserticola may intergrade directly with sayi at the headwaters of the Green River in western Wyoming, but I have no final proof of this. Annectens and affinis may contact or overlap in eastern San Diego County, California; there is no evidence here of intergradation, although fairly good series of both forms are available from closely contiguous points, which, in this area, involve a steep ecological gradient. Two peculiar specimens from the Sierra Juarez in Baja California suggest the possibility of annectens-affinis in- tergradation in the mountain passes there. In any case, the annectens-affinis interconnection is complete through deserticola. Summarizing, in the United States we have the unbroken chain, annectens- catenifer-deserticola-affinis-sayi, together with the short circuit, annectens-deserti- cola, and possibly deserticola-sayi. In Baja California the connection annectens- affinis is suggested but not proven. The weakest strand in the web is that connecting vertebralis with the rest. Vertehralis, through the new mid-peninsular subspecies bimaris, with which it undoubtedly intergrades, overlaps annectens, based on USNM 37536, a speci- men of bimaris from Alamo (Lat. 31°30' N.) on the coastal side of the Sierra Juarez. This specimen was collected by Nelson and Goldman and thus there is every reason to accept the accuracy of the locality data. Should further collecting in this vicinity substantiate this bimaris-annectens overlap, it will be a pertinent argument against considering vertebralis a subspecies of catenifer. However, there is another route by way of which there may be intergradation. Along the desert side of the mountains in northeastern Baja California, there is no specimen of Pituophis available from between El Marmol (Lat. 30° N.), where bimaris occurs, and the vicinity of El Mayor (Lat. 32° N.), where affinis has been collected. All of the intervening country is undoubtedly in- habited by Pituophis, but much of it is quite inaccessible, so that no specimens are available yet. Bimaris resembles deserticola — particularly specimens from the western Mojave Desert — in many ways. It appears to me that an intrusion of affinis into the Salton or Cahuilla Basin has broken a contact between them; but just as deserticola intergrades with affinis in Riverside County, so I think it probable that bimaris and affinis will be found to intergrade in the Gulf- coast desert between El Marmol and El Mayor, no doubt somewhere in the vicinity of San Felipe (Lat. 31° N.). Hence, partly because of this expected intergradation, and also by reason of the obvious resemblance between deserti- cola and bimaris (although territorially separated by the intrusion of affinis), I have considered vertebralu to be a subspecies of catenifer. If future collecting along the northeastern desert coast of Baja California should eventually show Klauber — Gopher Snakes of Baja California 7 an overlap between bimaris and ajfinis, then the Cape form must be known as rertebralis vertebrcdis, with bimaris and insidanus as its subspecies. As to the island forms tnsulanus, fuliginatus, and coronalts, so obvious are the relationships of the first to bimaris, and of the two latter to amiectens, together with some overlapping in the distinguishing characters, that I think subspecific rather than specific segregation is to be preferred. This carries with it no suggestion of weakness with respect to the vaUdity of the new forms. Since I am considering all of the forms discussed to be subspecies of Pitu-ophis catemfer (Blainville), 1835, I shall simplify the discussions by usually employing only the subspecific names, without repeating the generic and specific names or initials. Descriptions of New Forms The first new form to be described is one which occupies the middle section of the peninsula. The snakes of this area have hitherto been considered to belong to the species Pituophis vertebralis (Blainville), whose revised status I shall touch on later. Pituophis catenifer bimaris subsp. nov. Baja California Gopher Snake 1899. Coluber catenifer, var. vertebralis Mocquard, Nouv. Arch. Mus. Nat Hist., ser. 4, vol 1, p. 320. 1900. Pttyophis vertebralis (part) Cope, Rept. U. S. Nat. Mus. for 1898 p. 879. 1917. Pituophis vertebralis (part) Stejneger and Barbour, Checklist of N. A Amph. and Rept., ed. 1, p. 86. Type. — No. 32621 in the collection of LMK. Collected at Santa Ger trudis, near EI Arco (Lat. 28° N.), Baja California, Mexico, by Robert S Hoard, August 1939. Diagnosis. — A subspecies closely allied to vertebralis, from which it differs in having black (sometimes blackish-brown) anterior body blotches, whereas in the Cape form the anterior dark blotches are red or red-brown. From deserti- cola, of the Great Basin and the Mojave Desert, bimaris may be segregated by its fewer body blotches. It more often has a dark subcaudal stripe than deserticola, and less often shows dark streaks in the anterior dorsal light areas or interspaces, these streaks being characteristic of most deserticola. Affinis has brown anterior blotches as compared to the black of bimaris. The narrow and raised rostral in sayi is quite distinct from the broader and flatter scale in the Baja California form. Bimaris has fewer blotches than instdanus, annec- tens, catenifer, coronalis, or fidiginatus. Description of the Type. — An adult male with a length over-all of 1728 mm., and tail length of 231 mm., ratio .134. The body is of normal Pituophis 8 San Diego Society of Natural History form, the head wedge-shaped when viewed from above, and moderately distinct from the neck. The scale rows are 31-31-22, the ventrals number 243, and the subcaudals 65, all divided. The anal is entire. The central dorsal scale rows are strongly keeled; the ridges become fainter on the sides and the 5 lower lateral rows are smooth. The two lowest are considerably enlarged. There are paired apical scale pits. The rostral is triangular and is somewhat wider than high. It is slightly raised above the surrounding scales, and, viewed from above, is moderately convex. It contacts the first supralabials, prenasals, and internasals, indenting the latter for about half their depths. There are four prefrontals, the inner long and slim; they are widest anteriorly. The outer prefrontals curve down over the canthus to contact the loreals. The frontal is widest anteriorly. The parietals are fused into a single wrinkled and irregular scale. The nasals are subequal. The loreal is small, longer than high, and pointed posteriorly. There are two preoculars, the upper much the larger. There are three postoculars, the middle somewhat larger than the other two. The tem- porals are not in regular rows and are of uneven size; they are 3+3, 3+3. The supralabials number 8-8, the next to the last being the largest, and the fourth touching the eye. The infralabials number 13-12, the seventh being much the largest. The mental is small and triangular; it is followed by the first infralabials, which are in contact medianly. There are two long anterior genials, followed by a shorter, somewhat divergent pair separated by two rows of gulars. The head is brown above, almost without marks. Below it is immaculate cream-color. The dorsum is marked with 40 dark blotches, and the tail with 10. The anterior blotches are jet-black. Beginning at the eighth the blotches turn brownish until at mid-body they are reddish-brown, the posterior tips of the scales being darker. Then again the blotches turn darker; about thirteen anterior to the vent they are jet-black and remain so to the tip of the tail. Anteriorly, the blotches are highly irregular and confluent. They are somewhat longer than wide. On each side there is an alternating series of long, thin black blotches which form lateral boundaries of the interspaces, and almost surround them. Below are two other series, the lower and smaller engaging the edges of the ventrals, which are otherwise clear. The anterior light interspaces are not streaked with dark. Posteriorly, the main dorsal blotches become somewhat shorter and more regular. On the tail they are oval in shape, and are about equal to the inter- spaces. Here there is a single row of secondary blotches on either side. Mid- ventrally on the tail there is a longitudinal dark streak with serrated edges. The body surface between blotches is cream-colored to white, as is also the ventrum. However, it is probable that there was once considerable orange- color dorsally and anteriorly between blotches. Paratypes. — Sixteen paratypes are available, the localities being listed from south to north as follows: LMK 2934-5 EI Refugio (northeast of Magdalena Bay); USNM 15157 Ballenas Bay: LMK 32523 18 miles north of Punta Klauber — Gopher Snakes of Baja California 9 Prieta (Lat. 27° N.); LMK 3813, SDSNH 11553, MVZ 10673-4, MZUM 76462 San Ignacio; LMK 31032 Thurtoe Bay (1 mile south of Turtle Bay, on the west coast); SDSNH 17562-3 Rancho Las Flores (12 miles east of EI Arco); LMK 1129, LMK 1181, CAS 62957 El Marmol (= Onyx Mine, Lat. 30° N.); USNM 37536 Alamo (= El Alamo, Lat. 31i/2° N.). In addition, two island specimens are at hand, USNM 37537 from Santa Mar- garita, and CAS 59390 from Magdalena, which are tentatively assigned to this subspecies; however, they are not included in the statistics which follow, although the type is. The scale rows at mid-body vary from 31 to 35, being 33 in half the specimens, and less often 35 in the remainder than 31. The ventrais in the males (13) range from 238 to 249, mean 243.5; and in the females (4) from 253 to 257, mean 255.5.* The subcaudals in the males range from 63 to 72, mean 67.0, and in the females from 56 to 63, mean 58.8. The supralabials are usually 9, sometimes 8, and rarely 10; they average slightly under verte- brcilis, which has a mean of 9, the highest in the genus. The infralabials in bimaris range from 11 to 14, being most often 13. The preoculars always number 2; and the postoculars usually 3, but occasionally 4 and in one case 5. One labial contacts the eye in every specimen. The first temporals vary from 3 to 6 and the second from 3 to 5; they are highly irregular in arrangement. The scales on top of the head are rather constant for Pituophis. An azygos, which I should define as a central scale at the junction of the frontal and prefrontals, is not present in any specimen. There are four prefrontals in all specimens but one, in which each outer scute is fused to an inner. Although the longest specimen I have seen measured about 1800 mm., I am told that this snake grows to a larger size, especially in the vicinity of San Ignacio. The ratio of the tail length to length over-all averages about .140 in adult males, and .127 in adult females. The body blotches vary from 34 to 46, mean 40.8; and the tail spots from 8 to 13, mean 11.3. These numbers are lower than in almost any other sub- species. The pattern of bimaris is characterized by the presence of jet-black blotches both anteriorly and posteriorly. Between, the northerly specimens may be black throughout, but more often the blotches at mid-body are brown, streaked wtih dark-brown or even with black. Toward the south, as the range of vertebralis is approached, the number of the anterior black blotches decreases, there being a greater expanse of brown, or even red blotches, in the mid-section. The secondary side-blotches are usually confluent with the m.ain series anteriorly, thus isolating the first few interspaces into light spots within a solid black band, in a manner characteristic of deserticola; however, these light spots are not streaked with dark, as is so marked in most deserticola, although sometimes lacking in specimens from the western Mojave. In general, the streaking, both dorsally and laterally, is less evident in bimaris than deserticola. * In comparing these figures with those given by Miss Stull for vertebralis (1940, Table 9, p. 92), it should be noted that all specimens in her table are of the sex opposite to that listed. 10 San Diego Society of Natural History Ventrally b'tmaris is unmarked cream-color or white, except that the outer edges of the scutes are frequently blotched by the lowest row of lateral dark spots. All but four specimens have a dark caudal mid-ventral line, particu- larly evident posteriorly, but this is not so uniform as in vertebralis. The ground color is cream or white, and is clearer and brighter than in any other subspecies except vertebralis and the most southwesterly deserticola. The dorsal interspaces and the head in bimaris are usually orange or red, particularly in the southerly specimens, but this color virtually disappears in preservation. Except for a dark subocular vertical line, the head of bimaris is usually unmarked. Sometimes the dark labial sutures so characteristic of most Puuophis are present; a few inconspicuous marks are occasionally seen on the frontal and parietals, but they are not strongly evident as in insulanus. The two specimens from Santa Margarita and Magdalena islands do not seem to be peculiar in any way, except that the one from Santa Mar- garita has an unusually high ventral scale count; it is a female with 263 scutes compared with a record of 257 among the mainland specimens. Range. — Bimaris occurs in the peninsula of Baja California from about Lat. 24° 20' N. to Lat. 30° N., and from coast to coast. One record is known north of the latter line, this being Alamo, Lat. 31V'2° N. It is like- wise found on Santa Margarita and Magdalena islands. In addition to the localities given in listing the types and paratypes, the following have been cited in the literature: Arroyo de Santa Agueda, San Bartolome Bay, and Comondu. Remarks. — Bimaris certainly intergrades with vertebralis to the south, and the area of intergradation is probably fairly broad. Nevertheless, the geo- graphical segregation is remarkably consistent for snakes as variable in pat- tern as these gopher snakes. If we take, as an arbitrary division, a line drawn across the south end of the peninsula at its narrowest point, where indented by the Bay of La Paz (about Lat. 24° 20' N.), we have the following segre- gation : North of South of Bay of Bay of La Paz La Paz Anterior dark blotches black or dark-brown 21 2 Anterior dark blotches red or red-brown 1 33 In this table I have included the three specimens mentioned by Mocquard, 1899; v/ithout them none north of the line would have failed to key correctly. As it is, allowing for an intergrading area which may be more accurately de- lineated when additional material has been obtained, probably 90 per cent or more of the specimens will key out correctly. The absence of specimens from critical areas renders the northerly con- Klauber — Gopher Snakes of Baja California 11 tacts of bimaris much less certain. 1 think it likely that it will eventually be shown to intergrade with affinis somewhere along the desert (Gulf of Cali- fornia) coast northward between El Marmol and the delta of the Colorado River, where affinis is known to occur. The intervening territory is well suit- ed to either of these desert subspecies. The single specimen from Alamo remains somewhat of an anomaly. Intergradation with the coastal-mountain annectens seems very doubt- ful, not only because of the presence of the undiluted Alamo specimen in annectens territory, but because of the lack of bimaris tendencies in the most southerly annectens. The distance separating this annectens (collected at the mouth of the Rosario River) from the nearest bimaris at EI Marmol is some sixty miles. El Marmol is almost due east of Rosario and lies on the opposite side of the rolling hills at the south end of the Sierra San Miguel. If intergradation occurs, the modificatica must be quite abrupt. That insulanus of Cedros Island is an off -shoot of bimaris is suggested by a number of characteristic features of pattern. Nevertheless the differ- ences, which will be discussed in connection with the description of the island subspecies, have now become consistent enough to warrant a subspecific segre- gation. There is no overlap in number of body blotches, although a slight overlap may be found when larger series of both forms become available. Pituophis catenifer insulanus subsp. nov. Cedros Island Gopher Snake Plate 1, fig. 1. 1926. Pituophis catenifer annectens (part) Slevm, Proc. Cal. Acad. Sci., ser. 4, vol. 15, no. 3, p. 206. 1932. Pituophis vertebralis (part) Stuli, Occ. Papers Mus. Zool., Univ. Mich., no. 250, p. 3. Type. — No. 56353 in the collection of the California Academy of Sciences. Collected on Cedros (Cerros) Island off the west coast of Baja California, Mexico, by J. R. Slevin, August 7, 1922. Diagnosis. — A subspecies nearest to P. c. bimaris, from which it differs in having more than 50 body blotches while bimaris has less than this num- ber. From vertebralis and affinis it may be segregated by its black anterior blotches, while they have brown or red. Insulanus has a shorter tail and fewer dorsal blotches than annectens; is darker than catenifer, and has more ventral scales; lacks the subocular scales of coronalis; and is without the paired subcaudal dark stripes of fuliginatus. It differs from deserticola in pattern, and from sayi in having a flatter and less protruding rostral. Description of the Type. — A female with a length over-all of 1210 mm., and tail length of 151 mm., ratio .125. The body is of norm.al Pituophis shape, the head wedge-shaped when viewed from above, and moderately dis- tinct from the neck. The snout is neither as sharp as in affinis nor as blunt 12 San Diego Society of Natural History as in annectens. Th^, scale rows are 32-33-23, the ventrals number 247, and rhe subcaudals 58, all divided. The anal is entire. The central dorsal scale rows are keeled, but the 9 lower lateral rows are smooth. The lowest row is somewhat enlarged. Paired apical scale pits are present and in some scale rows are accentuated by dark dots. The rostral is triangular and is wider than high. It contacts the first supralabials, prenasals, and internasals. It indents the latter for about half their depths. There are four prefrontals, the inner long and slim; they are widest anteriorly. The outer curve down over the canthus to contact the loreals. The frontal is widest anteriorly. The parietals are wrinkled and irregular, and are separated by a crooked suture. Each supraocular is partly sutured posteriorly, this being characteristic of this subspecies. The nasals are subequal. The loreal is pointed posteriorly; there is a small subloreal on the right. There are two preoculars, the upper much the larger. There are three postoculars, the middle somewhat larger than the other two. The temporals are not in regular rows; they are 3 + 5, 5+5. The supralabials number 8-9, the next to the last being the largest; the fourth touches the eye on the right and the fifth on the left. The infralabials num- ber 13-13, the seventh being the largest. The mental is small and triangular; it is followed by the first mfralabials which are in contart medianly. There are two long pregenials, followed by a shorter pair, separated by two rows of gulars. The head is tan above (it was red in life), heavily mottled with black from the frontal posteriorly. There is a faint vertical dark mark below the eve and a few spots remain of a postocular dark stripe. The chin is immacu- late cream. The dorsum is marked with 57 dark blotches, and the tail with 12. The anterior blotches are jet-black; they are confluent laterally, thus accentuating the interspaces as small buff spots on a black background. At the third or fourth dark blotch back of the head a few lighter streaks become evident at the scale centers. These gradually increase in size and number until, at mid-body practically all the scales in the dark blotches are gray, only a few scale edges re- maining black. The interspaces, however, remain cream and virtually clear, except that some of the scales toward the sides have dark central streaks. Then posteriorly there is again an increase in dark pigment; the grayish scales are increasingly edged with black until, some 10 blotches anterior to the vent, they are solid-black and continue so to the tip of the tail. On the sides there are subsidiary alternating blotches. These begin as black streaks on the neck, but at mid-body become an irregular mixture of scales with black central streaks and others punctated with gray. These side marks are confluent with the main dorsal series so that throughout there is more regularity in the light dorsal interspaces, which are essentially a series of light blotches, than in the dark blotches. Posteriorly the lateral auxiliary blotches become more regular; they are black and it can be seen that there are four series on each side, the lowest engaging the outer edges of the ventral scutes. These ventral marks, on an otherwise clear cream background, are darkest posteriorly. The underside of the tail is much mottled, with a mid-ventral dark line posteriorly. Klauber — Gopher Snakes of Baja California 13 Paratypes. — Six paratopes, all from Cedros Island, are available: USNM 14088 and 54764; MCZ 19732-4; and CAS 59568. The following statistics will serve to summarize the subspecies. The type has been included in these data. There are three of each sex available and one, a skin, which is indeterminate. The scale rows at mid-body are usually 33, occasionally 31. The ventrals in the males are all 240, and in the females 247 or 248. The subcaudals in the males range from 58 to 65, mean 62.3; and in the females from 57 to 59, mean 58. The supralabials are usually 9, occasionally 8, and rarely 10. The infralabials are generally 12 or 13, but sometimes 14. The rostral is wider than high, with the upper lateral edges somewhat concave. The snout is blunt and the rostral is not raised above the adjacent scales. The inden- tation of the internasals is usually to about half their depths, but may be less. The loreal is wider than high and pointed posteriorly; in half the counts there is a second smaller loreal in the upper angle between two supralabials (the second and third, or third and fourth) in a manner characteristic of many specimens of deserticola. There are usually 2 preoculars, but some- times 1 or 3. The postoculars are most often 3 but may number 4 or 5. The temporals are 3 to 5 in the first row, and 4 or 5 in the second; they are too irregular in size and arrangement to be of any interest diagnostically. An upper labial (the fourth or fifth) contacts the eye in every specimen. The scales on top of the head are rather constant. Every specimen has 4 pre- frontals; none has an azygos. The greatest irregularities are in the supra- oculars, which are likely to have partial or complete posterior diagonal sutures, although this is not true of every specimen. The parietals are wrinkled and uneven, with no well-defined sutures separating them from the succeeding head scales. This snake grows to a length of at least 1400 millimeters. The adult tail proportionalities (to length over-all) are about .143 in the males and .129 in the females. The body blotches vary from 52 to 64, mean 55.3; and the dorsal tail spots from 11 to 17, mean 13.6. The pattern of insulanus comprises a series of irregular dorsal blotches which are bla.-k anteriorly and posteriorly, but gray (rarely brown) at mid- body. The change from black to the lighter color is produced by the gradual lightening of scale centers. At mid-body the posterior ends of the scales tend to be darker. On the sides there are auxiliary alternating blotches which are often confluent with the dorsal series and with each other; they are highly irregular and there is extensive dark streaking within and between them, thus producing a much more mottled appearance than is evident in bitnans or the more southerly deserticola. Anteriorly the first side-blotches are joined to the main series so that the light interspaces are themselves isolated into a series of light blotches in the manner characteristic of bimaris and aeserticola. The edges of these blotches are rather irregular in insulanus, and they are less marked by dark streaks on the scale keels than is usual in deserticola. Toward the tail the dorsal blotches are more regular and well-defined. Judging from the type, which I saw alive through the courtesy of Mr. 14 San Diego Society of Natural History J. R. Slevin, these snakes are brilliantly marked with red anteriorly, between the black blotches, but this color is lost in preservation, becoming yellow or cream. The outer edges of the ventral s are blotched with black or gray by the lowest series of lateral body spots. The tail is much mottled with black on the underside, and posteriorly there is sometimes a black line like that which characterizes bimans and vertebralis. The head, which is probably red in life, is spotted above with a large num- ber of conspicuous round or oval black spots, which are especially evident on the frontal, supraoculars, and parietals. These marks are characteristic of this subspecies and are present in all well-preserved specimens. The post- ocular dark dash and the dark lines which mark the labial sutures in most forms of Pituophis are absent or only faintly evident in wsulanus; their absence is in peculiar contrast to the unusual and accentuated spotting on top of the head. Below, the head is unicolor cream. Range. — Insnlanus is found only on Cedros Island, off the Pacific Coast of Baja California, Mexico. Remarks. — Insnlanus is most closely related to the nearest mainland form, bimaris, through which, in turn, it is allied to vertebralis and to deserti- cola, rather than to either annectens or ajjinis. This is evident from the tail proportionality and various characteristics of pattern, yet it is more mottled than either. In almost all cases the black spots in the frontal area serve as a key character that will readily segregate insnlanus. There is no overlapping in number of bodv blotches between insulanus and bimaris, for the former has from 52 to 64 with an average of 55.3, while the latter ranges from 34 to 46 with an average of 40.8. Even when much larger series have become available, it may be expected that only rarely will a specimen of one form fall within the limits of the other. Insulanus does overlap deserticnla in blotch counts, yet it is readily distinguishable by the lack of regularity of its anterior blotches, the lack of dark streaks in the interspaces (which may, however, be absent in some Mojave .specimens of deserticola) , and the absence of postocular and subocular dark vertical lines or streaks. Pituophis catenifer fuliginatus subsp. nov. San Martin Island Gopher Snake 1877. Pityophis sayi bellona Streets, Bull. U. S. Nat. Mus., no. 7, p. 40. 1895. Pituophis catenifer deserticola Van Denburgh, Proc. Cal. Acad. Sci., ser. 2, vol. 5, p. 149. 1905. Pituophis catenifer (part) Van Denburgh, Proc. Cal. Acad. Sci., . ser. 3, vol. 4, no. 1, p. 21. 1919. Pituophis catenifer annectens (part) Van Denburgh and Slevin, Proc. Cal. Acad. Sci., ser. 4, vol. 9, no. 6, p. 216. Klauber — Gopher Snakes of Baja California 15 Type.— No. 17449 in the collection of the San Diego Society of Natural History. Collected on San Martin Island, off the west coast of Baja Cali- fornia, Mexico, by Lewis W. Walker, July 11, 1939. Diagnosis. — A subspecies closely allied to P. c. annectens of the adjacent mainland, from which it differs in having a darker pattern, much black spot- ting on the head, a lower average number of body blotches, usually single instead of two preoculars, a high frequency of aberrant prefrontals, and in the possession of paired, dark, longitudinal streaks on the ventral surface of the tail, the latter being rarely and less perfectly present in annectens, where- in these streaks, if present, comprise series of separate, dark triangles. The same differences of blotch counts, pattern, and tail length which segregate annectens from the other subspecies of the genus will apply as well to fuligin- dtus, except that the latter is nearer catenijer in blotch counts. Description of the Type — A young adult female. Length over-all 960 mm.; length of tail 153 mm.; ratio of tail to length over-all .159. The body is of normal Pituophis proportions. The head is moderately distinct from the neck, and is wedge-shaped, but blunt anteriorly, when viewed from above. The scale rows are 31-35-25. The dorsal rows are much smaller than the lateral and are strongly keeled; about 8 lateral rows on either side are smooth. There are paired apical scale pits, which are moderately evident posteriorly. There are 245 ventrals, and 72 paired subcaudals. The anal is entire. The rostral is large, recurved, deeply indented below, and contacts the first supralabials, the prenasals, and the internasals. It is wider than high and is not raised above the adjacent scales. The internasals are narrowly in contact on the median line and diverge posteriorly. There are four pre- frontals, with an azygos at the point where they contact the frontal. The outer prefrontals diverge anteriorly and curve downward over the canthus, where they touch the loreals, postnasals, and, narrowly, the internasals. The frontal is widest anteriorly and only slightly indents the parietals. The supra- oculars are narrow anteriorly and touch the outer prefrontals at a point. The parietals are short and posteriorly irregular. The nasals are subequal and are narrowest at the nasal suture. The loreals are longer than high and are pointed posteriorly where they abut the preoculars. There is a small extra loreal below on the right. There are two preoculars on either side (not characteristic of this subspecies), the upper large, the lower very small. There are four subequal postoculars on either side. The temporals are highly irregular in size and arrangement; they are 4+4, 4-1^ 5. There are 8 supralabials on the right and 9 on the left, the next CO the last being the largest, followed by the next anterior. The fourth touches the eye on the right and the fifth on the left. The lower labials are 13-13, the seventh being the largest. The mental is small and triangular. It is followed by the first infralabials, which are in contact medianly; then a pair of long genials medianly in contact, and a second shorter pair separ- ated by two rows of gulars. The head is brown above, heavily mottled with black, especially posterior 16 San Diego Society of Natural History to the central prefrontals. On the sides it is buff, with a wide black streak from the eye to the angle of the mouth, a wide vertical line below the eye, and narrower black lines following the sutures between the lower labials. The ground color of the sides of the head is buff to gray. The lower surface is cream-color. All of these colors refer to the specimen as preserved in alcohol. The dorsum is marked by a series of square black blotches about 7 scale- rows wide and SVz scales long (end to end). The interspaces are buff and about V2 scale wide anteriorly, becoming brown and one scale wide poster- iorly. There are about 70 blotches in the central series. Alternating with the main series there is a somewhat smaller series on either side, and these in turn are followed by two others, each smaller than that above, and there- fore with wider interspaces. In general effect the light elements of the pat- tern form a net. Laterally the scales in the interspaces are centrally streaked or punctated with gray. The lateral blotches and interspaces are rather irregular. The ventral scales are yellowish-buff, and are blotched and stippled with dark gray, especially posteriorly. On the tail there are 19 squarish blotches above, with an alternating secondary series on either side. The interspaces are buff above and stippled-gray on the sides. The lower surface of the tail has a pair of dark-brown (almost black) streaks with an even, light stripe be- tween. The outer edges of the dark streaks are serrated, since each subcaudal is marked with a spot which is widest posteriorly. Each spot contacts its posterior fellow, hence the dark streaks are continuous, although somewhat irregular posteriorly. Paratypes. — Fourteen paratypes, all from San Martin Island, are avail- able: SDSNH 17463-4; USNM 8565 and 24395; MVZ 9703-6: LA 633; and CAS 8678, 43588-9, and 59678-9. The following summary is based on the type and paratypes: The scale rows at mid-body, while most often 33, may be 31 or 35. The ventrals in the males (10) vary from 224 to 234 with an average of 229.7; and in the females (5) from 236 to 245. average 242.2. This is a greater sexual difference than is to be expected when more specimens become available so that the probable population meaiis may be more accurately known. The subcaudals in the males vary from 73 to 82, with an average of 76.5; in the females 67 to 72, average 70.3. The supralabials are 8 or 9, with the latter slightly predomin- ating. Every specimen has at least one supralabial (the fourth or fifth) in contact with the eye, thus differing from coronalis; in 7 cases out of 30 two labials contact the eye, showing a resemblance in this character to deppei and Imeaticollis of southern Mexico and Central America. The infralabials are most often 13, but may be 12, or, less frequendy, 14. The rostral is penta- gonal in shape and is wider than high. It usually indents the internasals for about half their depths, but may completely separate them. The loreal is longer than high, with a posterior point indenting the upper preocular. In several specimens subloreals are present, and in one the main loreal is divid- ed vertically. Fuliginatus usually has a single preocular (23 out of 30 counts), a condition reversing that found in the peninsula specimens of annectens, in which a single preocular was noted in only 26 counts out of 102, the majority Klauber — Gopher Snakes of Baja California 17 having 2. This difference in proportions is, of course, significant. The post- oculars generally number 4, but sometimes 3 or 5. Azygos scales are present in 6 specimens out of 15, a higher proportion than in any other western Pituophis; and, in general, this subspecies is notable for the irregularity of the plates in the prefrontal area. Only 3 specimens out of 15 have the arrange- ment which is characteristic of most mainland snakes — that is, 4 regular prefrontals. In these island snakes there seems to be no standard, each individ- ual differing in some degree from its fellows. Thus there are central pre- frontals fused to each other, or to either or both outer prefrontals, in a variety of ways. Several specimens have a diamond-shaped central prefrontal, sur- rounded by internasals and outer prefrontals. These anomalies occasion- ally include the frontal. The parietals are also highly irregular but this is not so unusual in Pituophis. The supraoculars are not sutured as in insiilanus. This snake grows to a length of at least 1400 mm. The smallest speci- men measured 327 mm. The adult tail proportions, in relation to length over-all, average .175 in the males and .157 in the females, thus showing an affinity to annectens rather than bimaris. The body blotches vary from 55 to 70, mean 62.6, and the tail spots from 14 to 22, mean 19.1. The blotches are significandy fewer than in main- land annectens but there is too much overlap to afford a useful key character. The dorsal pattern of fuliginatus comprises a series of central blotches which are black anteriorly and at the tail, and may be either black, brown, or red- brown at mid-body. Anteriorly the main dorsal series is not regular in form, or well separated from the first lateral series on either side; the interspaces are narrow, much streaked with dark pigment, and comprise an irregular, light net-work. Posteriorly the blotches tend to become better separated and more uniform. The interspaces are cream anteriorly, and buff or brown toward the tail. There are about 6 auxiliary series of blotches on either side, the low- est engaging the ventrals. Anteriorly some of the light areas between these series tend to form light longitudmal lines, but these are not always present. Posteriorly the sides, between blotches, are suffused with gray or brown. The ventral surface is yellowish, much motded or blotched with black or gray. On the underside of the tail is the most characteristic feature of the pattern of this snake — a mid-ventral, even-edged light stripe, bordered on either side by a black or dark-brown stripe. While interruptions in these stripes are not rare, they are quite regular in 12 specimens out of 15. One of the others has lost its tail so that no determination can "be made; in a second the stripes are highly irregular, and in the last quite imperfect. It is to be understood that paired dark stripes are occasionally met with in annectens, deserticola, and in catenifer as well, but when present they usually comprise rows of dark triangles, the upper points of which barely touch their fellows on the next scale. In fuliginatus the dark mark on each scale contacts its neighbor rather broadly, although the mark usually widens posteriorly, thus causing the outer edges of the dark streaks to be serrated. Occasionally specimens of cate?iifer or deserticola will be found, particularly in central Washington, Oregon, and northeastern Gilifornia, in which the sub- caudal tail stripes are as perfect as in fuliginatus. While this complicates the 18 uSan Diego Society of Natural History preparation of a key, it does not invalidate fidiginatus, which differs from the northwestern snakes in other characteristics of scale counts and tail propor- tionality. The excess of black pigment in fidiginatus, which results in the body being darker than in most annectens, also affects the head marks. The usual dark marks in western Pituophis — a line across the prefrontals, a stripe from the eye to the angle of the mouth, a vertical subocular streak or triangle, and dark marks in both the upper and lower labial sutures, are all present. In addition the frontal, supraoculars, and parietals are much mottled with black. In annectens such marks in the fronto-parietal area, if present, are generally brown. The ground color of the head of fuligmatus is brown above and buff below. The lower jaw is unmarked except for the dark lines between infra- labials. Range. — Fidiginatus occurs only on San Martin Island, off the Pacific Coast of Baja California, at Lat. 30°29' N. Remarks. — San Martin is a small volcanic island only a mile in diame- ter, and but IV2 miles from the coast. The nearest point on the mainland is the cape along the west side of San Quintin Bay. Notwithstanding the proximity of the island to the mainland, the animal population must have been long isolated. All of the four reptiles thus far collected there show differences from the forms of the adjacent mainland. First there is Uta martinensis, which belongs to the stansburiana group and has been considered a valid species for forty years. Then there is an alligator lizard, Gerrbonotus midttcarinatus, which Van Denburgh described as a new subspecies, ignavus. in 1905. Although this is not currently recognized. Fitch (1938, p. 399) states that its validity may well be demonstrated by additional material. One specimen of a ring-necked snake, Diadophis amahilis similis, has been collected on the island. Blanchard (1942, p. 46) has pointed out that this differs in certain particulars from the mainland form, and more specimens may justify segregation. Thus it might be expected that the gopher snakes also, of which an adequate series is available, would show differences warranting nomenclatorial distinction. As would be anticipated, fidiginatus most closely resembles annectens of the adjacent mainland. It is not greatly different in pattern, although more completely pigmented with black, and generally darker, conditions probably not unrelated to the color of the lava on which it lives. It has somewhat fewer blotches on body and tail. The tail-length proportionality in fuliginatus further verifies its close relationship with annectens; in fact, it may have the greater ratio of the two, although it would take more material to prove this. In any case, it differs, in this important character, from the shorter-tailed forms, deserttcola, af finis, bimaris, insulanus, and vertebralis. As to scaiation, fidiginatus has 9 (instead of 8) supralabials somewhat oftener than annectens, has a higher percentage of single preoculars, and more often has fused and otherwise irregular prefrontals with azygos scales present. Klauber — Gopher Snakes of Baja California 19 With respect to the frequency of having 2 labials in contact with the eye, the proportion in juliginatus was 7 out of 30 counts, or 23.6 per cent. In 94 counts on specimens of annectens from northwestern Baja California, there were 5, or 5.3 per cent with 2 labials in contact, and, in 174 counts on San Diego County specimens. 2.9 per cent. Because of overlapping, none of these characters is as useful in a key as the parallel dark strikes under the tail. As these stripes are occasionally imperfect in fidigmatus, and sometimes appear (although seldom as continu- ous lines) in annectens, catemjer, and deserticola, I deem only a subspecific distinction to be warranted for the newly described form. Laurence M. Huey tells me that the gopher snakes are quite common on the island, although they are not easy to catch, since they readily escape into the lava crevices. No doubt they feed on the two mammals found there, an endemic wood rat, Neotoma niartinensis, and a white-footed mouse, Peromyscus maniculatuf exiguus. J. R. Slevin found one eating a young bird- Pitiiophis catenifer coronalis subsp. nov. Coronado Island Gopher Snake 1914. Pttuophis catenifer Van Denburgh and Slevin, Proc. Cal. Acad. Sci., ser. 4, vol. 4, no. 5, p. 141. 1919. Pitiiophis catemjer annectens (part) Van Denburgh and Slevin, Proc. Cal. Acad. Sci., ser. 4, vol. 9. no. 6, p. 216. Type. — No. 20229 in the collection of LMK. Collected on South Cor- onado Island, Mexico, by Philip M. Khuber, June II, 1933. Diagnosis. — A subspecies closely allied to the mainland P. c. annectens, from which it differs in usually having suboculars interposed between the supra- labials and the eye, and in the frequent abnormalities of the head scales. There are some differences in pattern; among others the island form has fewer blotches, but the overlap would prevent this being employed as a key character. The presence of suboculars will distinguish coronalis from other catenifer sub- species. In some areas suboculars occur with considerable frequency in deserti- cola, but this subspecies has fewer subcaudals and body blotches than coronalis, and differs in other ways. Description of the Type. — A female with a length over-all of 1150 mm., and tail length of 171 mm., ratio .149. The body is of normal Pttuophis shape, the head rather blunt when viewed from above, and moderately distinct from the neck. The scale rows are 33-33-23, the ventrals number 232, and the subcaudals 71, all divided. The anal is entire. The central dorsal scale rows are keeled, but the 10 lower lateral rows on each side are smooth. The three lowest are considerably enlarged. Apical scale pits seem to be virtually absent, although a few pairs may be discerned. The rostral is triangular and is wider than high. Viewed from above it is slightly convex; it is not raised above the other scales. It contacts the 20 San Diego Society of Natural History first supralabials, prenasals, and internasals, and deeply indents the latter. There are four prefrontals, the inner triangular, and the outer small, with their anterior halves split off to form an extra outer prefrontal on either side. The frontal is widest anteriorly, the supraoculars posteriorly. The parietals are wrinkled and irregular. The nasals are subequal. The loreal is small and pointed posteriorly. There are two preoculars, the upper much the larger. There are four post- oculars, of which the two lower are the smaller. The two lower are really to be considered suboculars, since they completely prevent any contact between the upper labials and the eye, a condition rare in Pituophis. The temporals are not in regular rows; they are 4+4, 4+4. The supralabials number 10-9, the next to the last being the largest. The infralabials number 13-14, the seventh being much the largest. The mental is small and triangular; it is followed by the first infralabials which are in contact medianly. There are two long anterior geniais, followed by a shorter pair which are separated by two or three rows of gulars. The head is light-brown above, with an irregular dark spot on the parietals. There is a dark streak in the suture below each eye, and the penultimate supra- labial is marked posteriorly. Some of the sutures between the lower labials are darkened, otherwise the lower surface of the head is immaculate cream. The dorsum is marked with about 64 dark blotches, and the tail with 20. They are highly irregular and often confluent anteriorly, so that opinions might differ as to the number. The anterior blotches are black; at mid-body they are brown, speckled with dark-brown. Po.steriorlv they are darkened, again be- coming black on the tail. In the posterior part of the body they are square, well separated, and are about equal to the interspaces. The ground color is grayish-buff, laterally suffused wtih brown, but becoming lighter posteriorly. The ventrum is buff, the posterior edge of each scute being somewhat darkened. There is no distinctive pattern on the underside of the tail. Pctratypes.—T^tre are three paratypes, SDSNH 11365 and CAS 13588-9. The latter are from South Coronado Island as is the type, but the locality of No. 11365 is given merely as "Coronado Islands." I have no record of gopher snakes having been taken on any of this group but South Island; however, there may be some significance in the fact that this is the only specimen of the four in which the labials contact the eyes. ' ' The following statistics summarize this rather inadequate series, in which I have included the type. There are three females and one male. The scale rows at mid-body vary from 31 to 35. The male has 222 ven- trals, the females 229 to 233. The subcaudals number 82 in the male, 69 to 71 in the females. The supralabials vary from 7 to 10, and the infralabials from II to 14. Both series of labials are characterized by occasional fusions of adja- cent members into scales of abnormal length. The rostral is wider than high and is not raised. The loreal is longer than high and is single in all cases. There are usually 2 but, in one case, 3 preoculars; the postoculars number 3 to 5. As the labials contact the eye in only, one specimen out of four, and then nar- towly, the lower of these postoculars are, in fact, suboculars. The temporals Klauber— Gopher Snakes of Baja California 21 vary from 3+4 to 4+5. The prefrontals are quite irregular; in no specimen are there 4 in what is the normal arrangement for annectens. Usually the posterior ends of what would ordinarily be the outer prefrontals are fused to the inner, while the outer ends, separated by sutures, might be considered supraloreals. In the single specimen wherein the labials contact the eye, the contact is quite narrow, just the opposite of the condition in the other annectens island relative, fuligtnatus, in which it is particularly wide. No specimen has an azygos. The longest available specimen is the type, 1150 mm., the shortest is 353 mm. We have too few specimens, only one an adult, to determine the tail proportionality; it is apparendy similar to that of annectens. The body blotches vary from 64 to 70 and the tail spots from 20 to 24. These are lower than average figures for annectens. In pattern coronalis is much like annectens, except that the body blotches are narrower, more circular, and more widely separated, especially toward the tail; there is also less streaking between blotches and on the sides. The anterior and posterior blotches are black, those between brown; in this there is no difference from most annectens. The ground color is buff or yellowish, suf- fused with gray or brown punctations laterally. The ventrum is buff with scattered suffusions or blotches of gray or black. The tail of one specimen has an imperfect double row of triangles on the underside; the others are virtu- ally unmarked. The head is not conspicuously marked — less so than in most annectens. A subocular dark line is always present. ^ange. — Coronalis is found only on South Coronado Island, off the northwest coast of Baja California, Mexico. One specimen is labeled only "Coronado Islands", and may have come from one of the others of the group. These Coronado Islands* are not to be confused with those in the Gulf of California, or with Coronado, a city on the broad section of the peninsula which forms the southwesterly outer boundary of San Diego Bay. Remarks. — Coronalis is obviously derived from annectens, from which it has been isolated long enough to have acquired differences in pattern and head scales. The former, while useful when comparative material is available, will not serve as a key character, although it may be noted that the four specimens of coronalis available have an average of 67.3 body blotches, while annectens in northwestern Baja California has 75.5, and in San Diego County 76.8. More important are the irregularly fused head plates in the island form; and, above all, the lack of contact between labials and eye in three out of four specimens. Out of 47 specimens of annectens from northwestern Baja California only two (both from. San Jose) have suboculars preventing a contact of a labial with the eye, and in these this condition is found only on one side of the head. Thus the proportions are 6 out of 8 (75 per cent) on the island compared with 2 out of 94 (2.1 per cent) on the mainland. In 115 annectens * More properly Coronados Islands. 22 San Diego Society of Natural History selected at random from a San Diego County series, 7 counts out of 230 (3.0 per cent) proved positive. Only two specimens out of the 115 had suboculars on both sides. In some areas a considerable proportion of deserticola speci- mens have suboculars. This does not invalidate coronalis, since they differ in other characters, but it does complicate the preparation of a key. A juvenile specimen of coronalis had eaten two fledgling birds. Redescriptions of Old Forms Pituophis catenifer vertebralis (Blainville) San Lucan Gopher Snake Plate 1, fig. 2. 1835. Coluber vertebralis Blainville, Nouv. Ann. Mus. Hist. Nat., vol. 4, p. 293. Type locality "Calif ornie", which included Baja California; type specimen in Paris Museum. 1854. Pituophis vertebralis Dumeril and Bibron, Erp. Gen., vol. 7, p. 238. 1860. Pilyophis haematois Cope, Proc. Acad. Nat. Sci. Phila., vol. 12, p. 342. Type locality Cape San Lucas; type specimens USNM 4682 (2). 1861. Pityophis vertebralis Cope, Proc. Acad. Nat. Sci. Phila., vol. 13, p. 300. 1884. Pityophis catenifer vertebralis Garman, Bull. Essex Inst., vol. 16, p. 27. Diagnosis. — A subspecies characterized by a short tail, high ventral scale counts, anterior red or red-brown blotches changing to black posteriorly, and a black mid-ventral band under the tail. Its nearest relatives are bimaris, msulanus, and deserticola in that order. The first two may be segregated from vertebralis by their black anterior blotches; the third is also black anter- iorly, although occasionally specimens from the western Mojave Desert have reddish or brown blotches in front. But these usually have dark streaks on the keels of the light scales in the anterior interspaces, as well as dark frontal and postocular streaks; also, they rarely have the dark subcaudal stripe character- istic of vertebralis. Applicability of Name. — Blainville's original description (1835. p. 293), and his figures (pi. 27), would lead one to question seriously whether the type really represented the snakes from the Cape region of Baja California that have so long been designated by the name vertebralis. However, the re- descriptions by Dumeril and Bibron (1854, p. 238) and Bocourt (1888, p. 672), as well as the figure in Jan's Iconographie (1867, liv. 22, pi. 1), leave little doubt as to the applicability of the name, stressing as they do, such recognizable characters as the chainlike anterior pattern of light blotches, the black posterior blotches, and the subcaudal dark streak. Material. — Of vertebralis, as now restricted by the segregation of bimaris, I have had available for study 33 specimens, as follows: LMK 3814, 20509-10, Klauber— Gopher Snakes of Baja California 23 20555, 20879, 20945, 21500, 21513, 23128, and ANSP 3791 Cape San Lucas; SDSNH 17660 10 mi. s. of Miraflores; USNM 12631 (2), 12644 La Paz, 37538 Cape San Lucas, 37539 San Jose del Cabo, 37540 Santa Anita, 64583 Miraflores; AMNH 5588-90 Miraflores; CAS 45874 San Pedro, 45875 San Antonio, 45876-7 San Bartolo, 45878 Agua Caliente; Stanford 4113 San Jose del Cabo; MVZ 11839-42 Todos Santos, 11843 Santa Anita, and 11844 Eureka. Counts are available on two other specimens, USNM 4682 (2) Cape San Lucas (the types of Cope's haematois). Of these specimens 21 are males, 14 females. Description. — The following summary is based on the above material. The scale rows at mid-body usually number 33, sometimes 35 and rarely 31. Paired apical scale pits are evident. The ventrals in the males range from 238 to 251, mean 243.5; and in the females 246 to 257, mean 250.4. The anal is entire. The subcaudals vary in the males from 60 to 69, mean 65.0; and in the females from 57 to 63, mean 59.2. A few of the anterior scales are sometimes undivided. The supralabials are usually 9 but are occasion- ally 8 or 10; the average is almost exacdy 9, the highest m the genus. The next to the last is the largest; the fifth touches the eye. The infralabials are most often 13, sometimes 12 or 14, rarely 11 or 15. Usually the seventh is the largest. The rostral is wider than high, is only moderately convex when viewed from above, and is not raised above the surrounding scales. The apex of the rostral is rounded and indents the internasals to about half their depths. The loreal is wider than high and is posteriorly pointed, where it abuts the upper preocular. Rarely a small extra loreal may be noted above or below. The preoailars are 2, or rarely 1 or 3. The postoculars usually number 3, occasion- ally 4, and in one instance 5. The temporals are highly irregular in size and arrangement; they vary from 2+4 and 3 + 3 to 6+6. The scales on the top of the head are quite regular for Pituophis. No specimen has a true azvgos (at the junction of the prefrontals and frontal), although in one specimen there is an extra scale between the internasals and the central prefrontals. Another has one outer prefrontal divided transversely. With these exceptions all specimens have 4 prefrontals. The supraoculars usually contact the outer pre- frontals. The parietals are somewhat irregular, especially posteriorly; on the sides they are usually indented by a small scale back of each upper post- ocular. The longest specimen I have seen measured about 1700 mm. The ratio of the tail to length over-all averages about .135 in the adult males and .123 in the females. Thus the tail is slightly shorter than in bimaris and closely approaches affinis; vertebrctlis is a much shorter-tailed snake than annectens or catentfer. Vertebralis is the handsomest of the gopher or bull snakes, with the possible exception of tnsulanus. Preserved specimens give little idea of the brilliant anterior red color in life, as this fades to brown or may almost dis- appear. This is not the case with the black posterior blotches, which have remained unchanged in museum specimens collected nearly 100 years ago. 24 San Diego SociEry of Natural History The pattern comprises a series of blotches which are red and saddle- shaped (that is, with narrow centers and wide lateral wings) anteriorly, gradu- ally changing through red-brown and brown to jet-black somewhat anterior to the tail. On the neck the wings of the dorsal blotches tend to merge, as in deserticola, completely enclosing diamond-shaped light areas (the true inter- spaces) which are pink or som.etimes orange. Posteriorly the blotches become square and well separated. The dorsal interspaces may continue pink to the tail, or they may become orange or yellowish posteriorly. The auxiliary lateral blotches are usually elongated anteriorly, and are rather indefinite in outline. Often the anterior lateral streaks may be darker than the main dorsal blotches; sometimes they are dark-brown or black. There are 2 or 3 rows of auxiliary lateral blotches on the neck. Posteriorly, as these blotches darken, they also become more evenly outlined; usually there are 3 or 4 alternating series at mid-body, the lowest engaging the edges of the abdominal scutes. The ground color, both laterally and ventrally, is yellow, cream, or buff. Below, the ventrum is usually clear anteriorly, but marked with gray or spots of black punctations caudad. On the tail there is a series of dorsal black spots which may be square, round, or elliptical, with a single secondary series on each side anteriorly. On the underside of the tail there is usually a dark streak, often somewhat ir- regular and always with serrated edges. This stripe is quite characteristic of this subspecies (it is somewhat less frequent and perfect in bimaris); and, in the series available to me, is evident in all but 3 out of 33 specimens, although in some cases only on the posterior part of the tail. The head is mottled or suffused with red above and is unmarked below. This color description has been based largely on specimens from Cape San Lucas and San Jose del Cabo. Some of those from further north have darker scales, or scales with dark posterior tips, scattered within the anterior dorsal red blotches, or in the lateral series. CAS 45875 from San Antonio is quite dark anteriorly, although not as black as typical bimaris. Another dark specimen is MVZ 11842; this is black anteriorly and is much more like bimaris than vertebrahs, although four other specimens from the same vicinity (Todos Santos) are typical vertebrahs, except for more than the usual amount of dark streaking forward. These are the two specimens to which attention has been called in the table on page 10. Mainly, in distinguishing verte- brahs from bimaris, one must look for a considerable contrast in color between the anterior and posterior dorsal body blotches in the former, while there is little or none in the latter. The body blotches vary from 38 to 57, although all but 4 specimens fall between 39 and 49. The mean is 44.0. The tail spots vary between 8 and 15, with a mean of 11.3. Range. — Vertebrahs. as newly defined, occurs only in the Cape region of Baja California, south of the isthmus formed by La Paz Bay (Lat. 24°20' N.). Here it has been collected at La Paz, San Pedro, San Antonio, San Bartolo, Eureka, Todos Santos, Agua Caliente, Miraflores (also 10. mi. s.), Klauber — Gopher Snakes of Baja California 25 Santa Anita, San Jose del Cabo, and Cape San Lucas. It seems to be quite plentiful and well-distributed throughout this area. Remarks. — Vertebralis intergrades with bimaris, the next subspecies to the north, which in turn relates it to instdanus, and, more distanriy, to deserti- cola. The fact that vertebralis has light anterior blotches, as opposed to the dark bimaris, together with its slightly shorter tail, suggests a possible influence of ajfinis, across the Gulf of California in Sonora. Such relationships are found in other genera such as Crotaltis, Lichanura, and Coleonyx. Pituophis catenifer annectens Baird and Girard San Diegan Gopher Snake Plate 2. fig. 1. 1853. Pituophis annectens Baird and Girard, Cat. N. Amer. Rept., part 1, p. 72. Type specimen USNM 1839; type locality San Diego, Cali- fornia. 1859. Pityophis annectens Baird, Reptiles, in Explor. and Surv. for Railroad Route etc., vol. 10, part 3, p. 15. 1919. Pituophis catenifer annectens (part) Van Denburgh and Slevin, Proc. Cal. Acad. Sci., £er. 4, vol. 9, no. 6, p. 216. Diagnosis. — A subspecies differing from all others except catenifer, fuligin- atus, and coronalis in the high number of body blotches and the high tail-length ratio. On the average it has more body blotches and ventral scutes than catenifer, and is more often black and with confluent blotches anteriorly. It lacks the well-defined and continuous subcaudal stripes of fuliginatus and also the conspicuous black mottling on the head of the latter. From coronalis it differs in usually having a labial in contact with the eye. Material. — Good series of this common snake are available for study, 52 from Baja California, over 300 from San Diego County, and 150 from elsewhere in southern California. Description. — The following summary is based entirely on the Baja Cali- fornia specimens. The scale rows at mid-body are usually 31 or 33, occasion- ally 35 and rarely 37. The central dorsal rows are sharply keeled; these ridges are less accentuated laterally and the lowest 6 or 7 rows are smooth. The low- est 3 or 4 rows are larger than the rest. Paired apical scale pits are in evidence, often emphasized by tiny dark spots. The ventrals in the males (27) vary from 223 to 242, mean 232.6; and in the females (21) from 230 to 253, mean 240.6. The anal is entire. The subcaudals number from 75 to 89 in the males, mean 81.1. One specimen has 68 but it is impossible to tell whether the tail is complete. The females range from 73 to 82, mean 76.5. The supra- labials most often number 8, sometimes 9, and rarely 10. The next to the last is the largest. Usually the fourth enters the eye, although if there are 9 it may be the fifth. In 2 counts out of 94 there is a complete row of sub- 26 San Diego Society of Natural History oculars so that no labial touches the orbit. In 5 counts there were two labials touching the eye. The infralabials are usually 12 or 13, but are occasionally 14, or less often 11 or 15. The seventh is the largest; the first pair meet on the median line. Behind there is a pair of large chin shields, followed by a smaller, divergent pair, the latter separated by 2 or 3 rows of gulars. The rostral is as wide as high, or, more often, slightly wider. It is not raised above the surrounding scales. Viewed vertically the front edge is almost flat, giving the snout a blunt appearance. The rostral parts the inter- nasals for about half their depths; they are posteriorly divergent. There are usually 4 prefrontals, but sometimes they are fused into 3 scales, and, in six specimens, into 2. Six specimens out of 50 have an azygos, a central small scale posterior to the inner prefrontals. The supraoculars are widest posteri- orly and the frontal anteriorly. The supraoculars contact the outer prefrontals. The parietals are wrinkled, and with uneven borders. The nasals are subequal, with the nostril at the upper end of the suture. The loreal is longer than high with a lower posterior point. Occasionally there is an extra loreal in the angle be- tween the nearest supralabial sutures. One (juarter of the specimens have a single preocular, the remainder two; when there are two the lower is much the smaller. The postoculars are most often 3, but there may be 4 or, less often, 5. The temporals, which are irregular in size and arrangement, number from 2+4 and 3+3, to 6+6. The longest specimen I have seen from Lower California measures 1700 mm. This is the longest-tailed of all the gopher snakes (with the possible exception of two of the new island subspecies) ; adult males have a tail to length over-all ratio of about .170 and the females .155. As to pattern, annectens is rather drab, lacking the bright and contrasting colors evident in the other subspecies. The dorsal blotches are usually black anteriorly and posteriorly; they may be black throughout, but more often are brown or red-brown at mid-body. Occasionally (6 cases out of 54) the anterior blotches are brown, although in such instances the blotches are border- ed with black; or each scale may have a brown center with a black edge. Brown anterior blotches are more apt to be found in juveniles. Non-black blotches posteriorly occur more often. Sometimes the last blotches have gray or brown scales with black edges. Anteriorly the dorsal blotches are quite irregular and indefinite; they are often confluent with each other and with the first lateral series. They are usually longer than wide. The interspaces comprise rows or streaks of yellow scales. Posteriorly rhe blotches become wider and rounder; the interspaces are buff or gray, streaked with brown, and the contrast remains imperfect. Finally, toward the tail the blotches become more distinct, the interspaces being lighter. There are usually 4 rows of alternating spots on each side anteriorly. The sides between blotches are so suffused with gray or brown punctations, or streaks on individual scales, that it is difficult to assign a color to the back- ground. The lower surfaces are buff or yellow, spotted with black or punc- tated with gray^. The outer edges of the ventrals are often marked by the lowest lateral series of blotches. The underside of the tail is usually stippled KiAUBER— Gopher Snakes of Baja California 27 with gray. Occasionally there are blotches suhcaudally, and sometimes double tows of brown or black triangles, the analogues of the twin dark stripes in fultginatus. The head is brown above, sometimes mottled in the parietal area with darker-brown or black. There is generally a sharp contrast between the most advanced black body-blotch and the brown of the head. Usually there is a slightly darker-brown streak across the anterior ends of the supraoculars and the frontal. A postocular dark streak, ending at the angle of the mouth, is sometimes faintly in evidence. The dark streaks in the labial sutures, so characteristic of many Pituophis, are generally present, particularly the one immediately below the eye. The head is buff on the sides, and, except for the darkened labial sutures, is immaculate below. Young specimens are usually brighter, with clearer spots and more color contrast. The body blotches in these Baja California annectens vary in number from 57 to 90, mean 75.5; and the tail spots from 17 to 29, mean 23.0. Range. — The subspecies annectens ranges from Santa Barbara County, California, south to the southern end of the San Pedro Martir Mountains in Baja California (Lat. 30^ N.) . In this area it is found from the coast, through the interior valleys and mesas, to the crests of the mountains, and well down the desert slopes, but not on the desert itself. Baja California localities are as follows: Tijuana, Rosarita Beach, Re- dondo (S. D. and A. E. Ry.), Lindero (S. D. and A. E Ry.), half way between Tijuana and Tecate, Tecate, Zacatosa, Descanso Point (also 1 m. n.), Mesquite Point, San Miguel Mission, 6 m. s. of Jatay, El Tigre grade, Ensenadn (also 3, 8, 10, 12, 15, and 23 mi. n., and 7 mi. s.), Guadalupe Valley, Laguna Hanson, Sierra Juarez. 8 mi. n. of Alamo, Santo Tomas, San Antonio del Mar ( = Johnson Ranch, also 10 mi. n.), Johnson Canyon (n. of San Antonio del Mar), Trinidad ( inter gr ade ) , San Jose (=^Rancho San Jose, Lat. 31° N.), Socorro Mine. La Encantada, San Pedro (Martir) Mt., San Quintin, and mouth of Rosario River (Lat. 30° N.). Remarks. — Annectens. as it occurs in northwestern Baja California, does not differ in any notable way from the snakes found in the vicinity of the type locality, San Diego, California. The Baja California specimens have slighdy higher ventral and subcaudal scale counts, and fewer blotches. As might be expected in an area extending from a foggy coast, through dry, interior val- leys to mountains exceeding 10,000 feet in altitude, there is a considerable local variation in scale counts and pattern. Some of the mountain speci- mens, from such localities as Laguna Hanson and La Encantada, have few- er body blotches, but this is not a uniform, tendency. There are two aberrant specimens from annectens territory, not including USNM 37536 from Alamo, which I have discussed elsewhere. The first is SDSNH 16863 from 8 miles north of Alamo. This has an annectem tail ratio and is, in fact, annectens in all particulars except that it is red or red- brown throughout, without any black, and the blotch count is low. Tentatively, 28 San Diego Society of Natural History I consider it an aberrant annectens with the black element of the coloring absent. It could hardly be considered an annectem-affinis intergrade in view of the tail length. The other queer specimen is CNHM 1394 from Trinidad (Valle de la Trinidad) . This is black both anteriorly and posteriorly, with a coloration and pattern reminiscent of both bimaris and annectens. It has a bimarts or affinis tail ratio; the blotch count (56) is too high for the former. I presume it is an intergrade, possibly bimaris-cinnectens. This can only be decided when much more material is available from this area. The relationships of ctnnectens to the island forms jidiginatus and coronalis have been discussed under those subspecies. Of the mainland subspecies, annectens intergrades in northern Santa Barbara County with catenifer, from which it is not sharply differentiated. It intergrades with deserticola at the foot of the San Gabriel and San Bernardino ranges, where they merge into the Mojave Desert in Los Angeles and San Bernardino counties. Direct intergradation with ajfinis or bimaris is as yet uncertain. Pituophis catenifer affinis Hallowell SoNORAN Gopher Snake Plate 2, fig. 2. 1852. Pituophis ajjims Hallowell, Proc. Acad. Nat. Sci. Phila., vol. 6, p. 181. No type specimen mentioned; type locality New Mexico, later (1854, p. 146) stated to be the Zuni River, New Mexico (which included Arizona) . 1920. Pituophis catenifer rutilus Van Denburgh, Proc. Cal. Acad. Sci., ser. 4, vol. 10, no. 1, p. 24. Type specimen CAS 33869; type locality Tucson, Pima County, Arizona. 1932. Pituophis sayi affinis Stull, Occ. Papers Mus. Zobl., Univ. Mich., no. 250, p. 4. 1943. Pituophis catenifer affinis Smith and Mittleman, Trans. Kans. Acad. Sci., vol. 46. p. 248. Diagnosis. — A subspecies which can be distinguished from catenifer, annectens, fuliginatus, and coronalis by its proportionately shorter tail and fewer body blotches. It can be segregated from deserticola, bimaris, and insidanus by its brown anterior blotches, compared with black on the others. It may be separated from vertebralis since it lacks the black subcaudal stripe which chararterizes the latter; also, vertebralis has black blotches at the base of the tail, while those of affinis are brown or dark-brown. Affinis has a rostral which is relatively wider and not so prominently raised above the surrounding .scales as in sayi; when the rostral is viewed from above the anterior curve is flatter in affinis than in sayi. Material. — Only two specimens of this subspecies from Baja California are available to me. These are USNM 22035 from Gardners Laguna, Salton Klauber — Gopher Snakes of Baja California 29 River (Mearns, 1907, p. 130), and MVZ 9562 from 23 miles north of El Mayor. Fortunately, adequate material is at hand from nearby Imperial County on the United States side of the international boundary, where the ecological conditions, involving both primitive desert and irrigated fields, are quite the same. In the days before irrigation these lowlands were occasionally watered by the overflow from the Colorado River, which periodically filled a number of shallow lakes. Similarly, the conditions in the Borego Valley of San Diego County match those at the foot of the Sierra Juarez and Sierra Cocopah below the line. Therefore, I have based the description of the subspecies afjinis, which follows, on the two Baja California specimens, together with 12 from San Diego County and 73 from Imperial County. Of these specimens, 38 are males, 39 are females, and the rest are heads only. The Arizona material has been omitted because of territorial variations. Description. — The scale rows at mid-body are most often 33 but may vary from 29 to 35. They are strongly keeled dorsally, but less so on the sides; from 5 to 7 of the lowest lateral rows are generally smooth. Paired apical scale pits are evident, particularly at the posterior end of the body, some being accent- uated by dark spots. The ventrals in the males vary from 232 to 251; most of them fall between 234 and 244; the mean is 239.6. In the females the variation is from 234 to 257, with the major portion falling between 241 and 252; the mean is 246.0. The anal is undivided. The subcaudals vary from 61 to 70 in the males, with an average of 65.1; and in the females from 50 to 63, with an average of 57.7. The supralabials usually number 8, but are occasionally 9, and, rarely, 7 or 10. Usually the fourth touches the orbit, and the next to the last is the largest. The infralabials generally number 13, but are not infrequently 12 or 14, and rarely 11, 15, or even 16. The seventh is the largest. The first pair meet on the median line; they are followed, first by a pair of large genials in contact, and then by a pair of smaller divergent genials which are separated by two or three rows of gulars. The rostral may be a litde higher than wide or the dimensions may be equal. Viewed from above it is convex, the snout being slightly sharper than in annectens. Occasionally it is slightly raised above the surrounding scales. It indents the internasals for about half their depths. Most specimens have four prefrontals, although in some specimens an outer and an inner may be fused. Occasionally the outer end of a prefrontal is cut off to form what might be considered an upper loreaj. The supraoculars are narrowest anter- iorly; they usually contact the outer prefrontals. The parietals are triangular and have irregular outer edges, particularly because of the invasion of a scale behind the upper postoculars. The anterior nasal is slightly larger than the posterior. The loreals, of which there is usually only one on each side, are longer than high. The preoculars are ordinarily 2-2 with the lower quite small compared to the upper; in 14 per cent of the counts there is a single preocular. The postoculars are subequal; they are most often 3-3 but there may be 4 or even 5 on a side. The temporals are highly irregular in shape; they vary from 2 to 6 in the first row, and 3 to 6 in the second. The longest specimen I have from Imperial County measures 1708 mm., 30 San Diego Society of Natural History and the shortest 415 mm. This snake grows to a large size in some sections of Arizona, probably well exceeding 2000 mm. The ratio of the length of the tail to the length over-all in the adults is about .136 in the males and .125 in the females. It is a short-tailed snake compared with deserticola, or especially with annectem. I have discussed tail-length equations and comparisons else- where (Klauber, 1943; see particularly p. 41 and table 13, but note that the series referred to as P. c. deserticola I, Imperial County, should be called P. c. affinis) . Ajfinis is primarily a brown-blotched snake with much dark streaking of individual scales in the interspaces. The main dorsal series is usually light- brown anteriorly, becoming darker toward the tail until, at the end, the blotches may be almost black. The anterior blotches are often outlined with dark- brown and sometimes even with black. The anterior blotches are not saddle- shaped, but the interspaces frequendy are. The interspaces are usually buff, but within them there are many scales which are conspicuously darkened by brown streaks on their keels; this is particularly evident laterally, the dorsal interspaces being often clear. There are about four series of auxiliary and alternating blotches on the side. Both the main series and the auxiliaries tend to become more widely separated, and more even and clearly outlined poster- iorly. The color below is cream or yellow, with some spotting on the outer edges of the ventrals. The subcaudal surface is usually spotted; rarely these spots may form longitudinal lines. The head is light-brown or buff above, and lighter on the sides. There is usually a wide dark-brown line across the fronts of the supraoculars and the frontal. Back of this line there are dark-brown spots scattered in the frontal and parietal areas. There is generally a dark line from the eye toward the angle of the mouth, with a vertical termination between the last two supra- labials. There are the usual Pttuophis streaks which follow the labial sutures, the one under the eye being widened and accentuated. The head is unspotted below. The body blotches in these Calif ornian specimens range from 41 to 62; with 4 exceptions out of 78 they fall between 43 and 58. The mean is 49.1. The tail spots vary from 10 to 19 with a mean of 14.1. Range. — Since only two specimens are available from Baja California, collected at Gardners Laguna and near El Mayor, the range in the peninsula is imperfectly known. Reasoning from the situation in San Diego and Imper- ial counties, it may be expected to occur from the desert foothills of the Sierra Juarez eastward, and along the Gulf of California coast southward until it meets and intergrades with, or overlaps, btmaris, probably in the vicinity of San Felipe. The general range of ajfinis is from central New Mexico, western Texas, Chihuahua, and southern Coahuila, westward through Arizona (excluding the section north of the Colorado River and extreme northern Navajo and Apache counties inhabited by deserticola) to central Riverside, Imperial, and northeastern San Diego counties in California; Sonora; extreme northern Sinaloa; and northeastern Baja California. Klauber — Gopher Snakes of Baja California 31 Remarks. — Mv reasons for assigning the gopher snakes of northeastern Baja Cahfornia to the subspecies P. c. affmis rather than to P. c. desertirola, and the relationship between these two subspecies will be discussed in a sub- sequent paper on the gopher snakes of the Pacific Coast and Great Basin. Key to Subspecies of Pttuophis catenifer IN Baja California la Supralabials usually not in contact with the eye P. c. coronalis South Coronado Island lb At least one supralabial usually touching the eye 2 2a Two parallel dark stripes formed of contigu- ous scale blotches on the underside of the tail P. c. fuliginatus San Martin Island 2b No parallel continuous dark stripes on under- side of tail, although there may be rows of adjacent triangles 3 3a Dark blotches on the body (exclusive of the tail) usually exceed 63; posterior lateral ground color suffused with gray P. c annectens Northwestern coastal, foothill, and mountain area from the U. S. Border south to Lat. 30° N. 3b Dark blotches on the body (exclusive of the tail) usually 63 or less; posterior lateral ground color not suffused with gray 4 4a Anterior body blotches black 5 4b Anterior body blotches red or brown 6 5a Body blotches 50 or more P. c. insulanus Cedros Island 5b Body blotches less than 50 P. c. bimaris Central area of the peninsula, from coast to coast, from Lat. 24°2r N. to Lat. 30° N. with an occasional specimen as far north as Lat. 31°30' N. Also Santa Margarita and Magdalena islands. 32 San Diego Society of Natural History 6a Posterior body blotches (at base of tail) brown; no subcaudal dark stripe P. c. ajfinis Delta of the Colorado River and the north- eastern desert area. 6b Posterior bodv blotches (at base of tail) black; a dark subcaudal stripe usually present P. c. vertebralis Cape region south of Lat. 24°20' N. Note: It has been possible to simplify the key considerably by excluding subspecies which do not occur in Baja California. The gopher snakes are so variable that an infallible key cannot be devised. Successful determinations must be premised on the use of locality data as well, for the complete differentiation of subspecies is based on the summation of more characters than can be included in a key. Acknowledgments I am very grateful to the following individuals and institutions for the loan of material and for other assistance: Mr. Charles M. Bogert, American Museum of Natural History; Mr. Joseph R. Sievin, California Academy of Sciences; Messrs. Karl P. Schmidt and Clifford H. Pope, Chicago Natural History Museum; Dr. Howard R. Hill, Los Angeles County Museum of History, Science and Art; Mr. Arthur Loveridge, Museum of Comparative Zoology, Harvard University; Mr. Thomas L. Rodgers, Museum of Verte- brate Zoology, University of California; Miss Margaret Storey, Natural His- tory Museum, Stanford University; Dr. Waldo L. Schmitt and Dr. Doris M. Cochran, United States National Museum; Dr. Raymond B. Cowles, Univer- sity of California at Los Angeles; Mrs. Helen T. Gaige, Museum of Zoology, University of Michigan; Dr. Emmett R. Dunn, Academy of Natural Sciences of Philadelphia. I was assisted in making scale counts by Mr. Charles E. Shaw. I am thankful to the late Mr. Clinton G. Abbott and to Mrs. Alice K. Miller for their courtesv in making translations. The map was prepared by Mr. Leslie C. Kobler. I am greatly indebted to Mr. C. B. Perkins for criticisms and editorial assistance. The following abbreviations have been used in the designation of museum specimens : AMNH American Museum of Natural History ANSP Academy of Natural Sciences of Philadelphia CAS California Academy of Sciences CNHM Chicago Natural History Museum LA Los Angeles Countv Museum of History, Science and Art LMK Collection of L. M.' Klauber MCZ Museum of Comparative Zoology, Harvard University Klauber— Gopher Snakes of Baja California 33 MVZ Museum of Vertebrate Zoology. University of California MZUM Museum of Zoology, University of Michigan SDSNH San Diego Society of Natural History SU Natural History Museum, Stanford University USNM United States National Museum Summary This paper comprises a resurvey of the gopher snakes of the genus Pituo- phis, found in Baja California, Mexico, based on additional material which has recently become available. All the gopher snakes of the area are deemed to be subspecies of Pituophis catenifer (Blainville), 1835. Four new subspecies are described: P. c. bimaris from the central area of the peninsula; P. c. insulanus from Cedros Island; 7^. c. jiiliginatus from San Martin Island; and P. c. coronalis from South Coronado Island. The first two are nearest to P. c. vertebralis of the Cape region, the others to P. c. annecfens of the San Diegan faunal area. The only other form found in the peninsula is P. c. affinis of the northeastern desert and Colorado River delta. The previously known forms, as newly circumscribed, are summarized. Relationships are discussed. Bibliography Baird, S. F. 1859. Reptiles: In "Explorations and Surveys for a Railroad Route from the Mississippi River to the Pacific Ocean," vol. 10, Gen. Rept., pp. 13-16, plates 24-36. Baird, S. F. and Girard, C. 1853. Catalogue of North American Reptiles and Amphibians in the Museum of the Smithsonian Institution, part 1 — Serpents, pp. 1-172. Belding, L. 1887. Reptiles of the Cape Region of Lower California. West. Am. Sci., vol. 3, no. 24. pp. 97-99. Blainville, H. D. de 1835. Description de Quelques Especes de Reptiles de la Calif ornie. Nouv. Ann. Mus. d'Hist. Nat., vol. 4, pp. 233-296. Blanchard, F. N. 1942. The Ring-neck Snakes, Genus Diadophts. Bull. Chicago Acad. Sci., vol. 7, no. 1, pp. 1-144. BocouRT, F. (with Dumeril, a., and Mocquard, F.) 1870-1909. Mission Scientifique au Mexique. Les Reptiles, pp. 1-1012. Atlas. BOGERT, C. M. 1930. An Annotated List of the Amphibians and Reptiles of Los Angeles County, California. Bull. So. Calif. Acad. Sci., vol. 29, part 1, pp. 3-14. 34 San Diego Society of Natural History Cope, E. D. c • r 1860. Notes and Descriptions of New and Little Known Species ot American Reptiles. Proc. Acad. Nat. Sci. Phila., vol. 12, pp. 339-345. 1861. Contribution to the Ophiology of Lower California, Mexico, and Central America. Proc. Acad. Nat. Sci. Phila., vol. 13, pp. 292- 306. 1887. Catalogue of Batrachians and Reptiles of Central America and Mexico. Bull. U. S. Nat. Mus., no. 32, pp. 1-98. 1900. The Crocodilians, Lizards, and Snakes of North America. Report of U. S. Nat. Mus. for 1898, pp. 153-1294. DuMERiL, A. M. C. and Bibron, G. 1854. Erpetologie Generale, vol. 7, part 1, pp. XVI + 780. Fitch, H. S. 1938. A Systematic Account of the Alligator Lizards (Gcrrhonotus) in the Western United States and Lower California. Amer. Midi. Nat., vol. 20, no. 2, pp. 381-421. Garman, S. 1884. North American Reptiles and Batrachians. Bull. Essex Inst., vol. 16, pp. 1-46. Hallowell, E. 1852. Descriptions of New Species of Reptiles Inhabiting North Amer- ica. Proc. Acad. Nat. Sci. Phila., vol. 6, pp. 170-182. 1854. Reptiles: In "Report of an Expedition down the Zuni and Colorado Rivers" by Capt. L. Sitgreaves, pp. 106-147. Hydrographic Office, U. S. Navy 1928. Mexico and Central America Pilot (West Coast) . H. O. no. 84, seventh edition, pp. 1-446, Jan, G. (with Sordelli, F.) 1860-1881. Inconographie Generale des Ophidiens, pp. 100 + 3 vol. Atlas. Milan and Paris. Klauber, L. M. 1931. A Statistical Survey of the Snakes of the Southern Border of California. Bull. Zool. Soc. San Diego, no. 8, pp. 1-93. 1939. Studies of Reptile Life in the Arid Southwest. Part 1. Night Collecting on the Desert with Ecological Statistics. Bull. Zool. Soc. San Diego, no. 14, pp. 6-64. 1941a. Variations and Relationships in the Snakes of the Genus Pituophis [A Review of Miss Stull's paper of 1940]. Copeia no. 1 of 1941, pp. 57 60. 1941b. The Correlation between Scalation and Life Zones in San Diego County Snakes. Bull. Zool. Soc. San Diego, no. 17, pp. 73-79. 1943. Tail-length Differences in Snakes wtih Notes on Sexual Dimor- phism and the Coefficient of Divergence. Bull. Zool. Soc. San Diego, no. 18, pp. 1-60. KiAUBER — Gopher Snakes of Baja California 35 LiNSDALE. J. M. 1932. Amphibians and Reptiles from Lower California. Univ. Calif. Pubs, in Zo6l., vol. 38. no. 6, pp. 345-386. Mearns, E. a. 1907. Mammals of the Mexican Boundary of the United States. Part 1 : Bull. U. S. Nat. Mus., no. 56, pp. XV + 530. Meek, S. E. 1905. An Annotated List of a Collection of Reptiles from Southern California and Northern Lower California. Field Col. Mus., Zobl. Ser., vol. 7, no. 1, pub. 104, pp. 1-19. MOCQUARD, F. 1899. Contribution a la Faune Herpetologique de La Basse Californie. Nouv. Arch. Mus. Nat. Hist., ser. 4, vol. 1, pp. 297-344. Mosauer, W. 1936. The Reptilian Fauna of Sand Dune Areas of the Vizcaino Desert and of Northwestern Lower California. Occ. Paper Mus. Zobl., Univ. Mich., no. 329, pp. 1-21. Nelson, E. W. 1921. Lower California snd Its Natural Resources. Mem. Nat. Acad. Sci., vol. 16, first memoir, pp. 1-194. Schmidt, K. P. 1922. The Amphibians and Reptiles of Lower California and the Neigh- boring Islands. Bull. Am. Mus. Nat. Hist., vol. 46, art. 11, pp. 607-707. Slevin, J. R. 1926. Notes on a Collection of Reptiles and Amphibians from the Tres Marias and Revillagigedo Islands, and West Coast of Mexico, with Description of a New Species of Tanttlla. Proc. Cal. Acad. Sci., ser. 4, vol. 15, no. 3, pp. 195-207. Smith, H. M. and Mittleman, M. B. 1943. Notes on the Mansfield Museum's Mexican Reptiles Collected by Williamson. Trans. Kan. Acad. Sci., vol. 46, pp. 243-249. Smith, H. M. and Taylor, E. H. ^ 1945. An Annotated Checklist and Key to the Snakes of Mexico. Bull. U. S. Nat. Mus., no. 187, pp. IV+239. Stejneger, L. and Barbour, T. 1917. A Check List of North American Amphibians and Reptiles, pp. pp. IV +125. Cambridge. 1923. Same. Second Edition, pp. X + 171. Cambridge. 1933. Same. Third Edition, pp. XIV + 185. Cambridge. 1939. Same. Fourth Edition, pp. XVI + 207. Cambridge. 1943. Same. Fifth Edition, pp. XIX + 260. Cambridge. (Bull. M.C.Z., vol. 43, no. 1) 36 San Diego Society of Natural History Streets, T. H. 1877. Contributions to the Natural History of the Hawaiian and Fanning Islands and Lower CaHfomia. Bull. U. S. Nat. Mas., no. 7, pp. 1-172. Stull, Olive G. 1932. An Annotated List of the Forms of the Genus Pituophis. Occ. Papers Mus. Zool., Univ. Mich., no. 250, pp. 1-5. 1940. Variations and Relationships in the Snakes of the Genus Pituophis. Bull. U. S. Nat. Mus., no. 175, pp. VI + 225. Terron, C. C. 1921. Datos Para una Monografia de la Fauna Erpetologica de la Penin- sula de la Baja California. Sociedad Cientifica "Antonio Alzate". Mem. vol. 39. pp. 161-171. Van Denburgh, J. 1895. A Review of the Herpetology of Lower California. Part 1- Reptiles. Proc. Cal. Acad. Sci., ser. 2, vol. 5, pp. 77-162. 1897. The Reptiles of the Pacific Coast and Great Basin. Occas. Papers Calif. Acad. Sci., no. 5, pp. 1-236. 1905. The Reptiles and Amphibians of the Islands of the Pacific Coast of North America from the Farrallons to Cape San Lucas and the Revilla Gigedos. Proc. Cal. Acad. Sci., ser. 3, vol. 4, no. 1, pp. 1-27. 1920. A Further Study of Variation in the Gopher-Snakes of Western North America, Proc. Cal. Acad. Sci., ser. 4, vol. 10, no. 1, pp. 1-27. 1922. The Reptiles of Western North America. Occas. Papers Calif. Acad. Sci., no. 10, 2 vols., pp. 1-1028. Van Denburgh, J. and Slevin, J. R. 1914. Reptiles and Amphibians of the Islands of the West Coast of North America. Proc. Cal. Acad. Sci., ser. 4, vol. 4, no. 5, pp. 129-151. 1919. The Gopher Snakes of Western North America. Proc. Cal. Acad. Sci., ser. 4, vol. 9, no. 6, pp. 197-220. 1921. A List of the Amphibians and Reptiles of the Peninsula of Lower California, with Notes on the Species in the Collection of the Academy. Proc. Cal. Acad. Sci., ser. 4, vol. 11, no. 4, pp. 49-72. Ki.AUBER— Gopher Snakes of Baja California Plate 1 Fig. 1. Pituophis catenijer insidaniis Adult female (type specimen) from Cedros Island. Collected and photo- graphed by J. R. Slevin, California Academy of Sciences. Fig. 2. Pituophis catenijer vertebralis Specimen from Cape San Lucas, Baja California. Collected by Fred Lewis. Photograph by L. C. Kobler. Klauber— Gopher Snakes of Baja California Plate 2 Fig. 1. Pituopbis catemjer annecteris Adult female from Mesquite Point, Baja California. Collected by Dr. C. L. Hubbs. Photograph by L. C. Kobler. Fig. 2. Pituophis catenifer affinis ' Adult male from 10 miles west of Benson, Cochise County, Arizona. ] Collected by Robert Hoard. Photograph by L. C. Kobler. TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 2, pp. 41-48, plate 3. August 26, 1946 A NEW GOPHER SNAKE (PITUOPHIS) SANTA CRUZ ISLAND, CALIFORNIA Laurence M. Klauber of Reptile! and Amphibians, San Diego Society of Natural History In a recent study of the gopher snakes of the genus Pituophis in Baja CaUfomia,* it was found that three island populations differed sufficiently from those of the mainland to warrant recognition as new subspecies. This naturally led to an investigation of the snakes found on the islands off the coast of the State of California, Pituophis being definitely known to occur on two, Santa Cruz and Santa Catalina, and rumored on a third, San Clemente. I have been unable to find any important differences between the Santa Catalina specimens and Pituo- phis catenifer annectens of the nearby mainland; but the Santa Cruz Island snakes, while overlapping, in some degree, both P. catenifer catenifer and P. c. annectens, possess average differences which warrant nomenclatorial segregation. Pituophis catenifer pumiliis subsp. nov. Santa Cruz Island Gopher Snake Plate 3, fig. 1. 1914. Pitiiophis catenifer Van Denburgh and Slevin, Proc. Cal. Acad. Sci., set. 4, vol. 4, no. 5, p. 136. 1919. Pituophis catenifer catenifer (part) Van Denburgh and Slevin, Proc. Cal. Acad. Sci., set. 4, vol. 9, no. 6, p. 211. Type. — No. 17238 in the collection of the San Diego Society of Natural History. Collected on Santa Cruz Island, Santa Barbara County, California, by Norman Bilderback, May 5, 1938. * The Gopher Snakes of Baja California, with Descriptions of New Subspecies of Pituophis catenifer. Trans. San Diego Soc. Nat. Hist., Vol. 11, No. 1, pp. 1-40, 1946. 42 San Du.go Scx:n:TY of Natural History Dia^nosii.— A subspecies most closely resembling P. c. catenifer and P. c. annectcns of the mainland, and differing in tail length and pattern from all other mainland speaes and subspecies of the genus, as do these two. It most nearly resembles ctnmctcns in pattern, including color and blotch counts, but is nearer catcnijcr in the number of ventral and subcaudal scales. It differs from both in having fewer scale rows — although with some overlapping — and in being stunted in size. In comparison with other island forms, pumilns has more blotches and a proportionately longer tail than msulanus, lacks the parallel darks subcaudal stripes and extensive dark head blotches of fuliginatus, and is without the subocular scales of coronalis. Description of the Type. — A male, probably a young adult. Length over- all 646 mm.; length of tail 111 mm.; ratio of tail to length over-all .172. The body is relatively slim for Pttuophis. The head is moderately distinct from the neck, and is wedge-shaped, with a rounded snout. The scale rows are 25-29-19. There are 29 scales only at one or two places forward of mid-body, where there are short runs of extra scales; elsewhere 27 rows are the maximum. The dorsal rows are conspicuously keeled, and are smaller than the lateral. About 8 of the lowest rows on either side are smooth. The lateral row next to the ventrals is considerably larger than those above. There are 217 ventrals, and 72 subcaudals, most of which are paired, al- though the third to fifth after the anal are undivided. The rostral is the same in height and width, and is not conspicuously raised above the contiguous ."^cales. Ir is recurved over the top of the snout and almost completely separates the internasals. The internasals are triangular and pos- teriorly divergent. There are four prefrontals, the inner long and triangular, the outer curved over the canthus to contact the loreals. The frontal is quad- rangular and is widest in front; it indents the parietals for less than Va of their length. The supraoculars are narrow in front and do not touch the prefrontals. The parietals are relatively short and even-edged; they are not conspicuously indented by the uppermost temporals, as in many western Pituophis. The nasals are subequal, with large nares at the upp)er end of the suture which separates them. There is a small oval loreal on either side, with a tiny sublorea! on the left. The eye is large. Tlie preoculars are 2-2, the upper much the larger; the postoculars are 3-3 and are subequal. The temporals are 3 + 3, 3 + 3; they are more regular than in many Pituophis. The lowest in the first row on the right is verv small. The supralabials are SS:. the seventh is the largest but does not greatly exceed the sixth; the fourth touches the eye. The infralabials are 12-12, the first pair meeting on the median line; the seventh is the largest, with the sixth next in order of size. The mental is small and triangular. The anterior genials are long and narrow: they are medianly in contact; the posterior are still slimmer and are separated by 2 to 4 rows of gulars. The head is light-brov\Ti. The posterior edges of the prefrontals are slightly darkened, and the parietals are spotted with black. On the sides the usual Pituo- phis postocular stripe is irregularlv in evidence. There is a heavy black mark under each eve in the suture between the fourth and fifth supralabials, and the Klaubfr — New Gopher Snake 43 two sutures anterior to these are marked with thin black lines. A few of the infralabial sutures are also lined with black, otherwise the ventral surface of the head is unmarked cream-color. There is a series of 78 median blotches on the body, and 25 spots on the tail. Anteriorly there are four alternating secondary series on either side. The main series is highly irregular and somewhat indefinite, for although many of the included scales are black, partiailarly in front, others are centrally streak- ed with brown or gray, and the light scales in the interspaces are punaated with darker. Thus there is a general effect of mottling which interrupts the regularity of the pattern. The first lateral series on either side is more or less confluent with the dorsal blotches anteriorly. The interspaces, where evi- dent, are buff; but the sides are much suffused with gray or brown, upon which the black secondary series of spots are scattered. There are black blotches on the outer edges of the ventrals, with another paired series of brown spots mid- ventrally. The ground color of the under surface is buff. Posteriorly the main dorsal series is more solidly black, with the interspaces clearer, so that they are also more definite. On the whole the snake gives the Impression of being grayer than annectens, especially on the sides, where there is sufficient gray to suggest longitudinal stripes. Paratypes.—Ei'^t Datatypes are available: CAS 36120-1. 45131; Stan- ford 8389;' MCZ 12679;' and MVZ 1 1131-3. The last three are from Johnson's or Johnson's Landing, near Gull Rock; the localities of the others are given merely as Santa Cruz Island. The scale rows vary from 23-27-19 to 27-29-21. Together with the type, whose counts are included in the following statistics, 2 specimens have 27 rows at mid-body, one has 28, and 6 have 29. This is a lower average than in any other subspecies. The ventrals vary from 209 to 226 in the males (7 counts), mean 215.0; in the females from 212 to 221 (2 counts), mean 216.5. The subcaudals in the males range from 69 to 73 with a mean of 70.9; and in the females from 55 to 61, mean 58.0. These averages are somewhat below those of mainland P. c. catenifer. Several specimens have a block of 3 to 5 undivided subcaudals be- ginning 2 or 3 scales back of the anal plate, but this is not universal. The snout is blunt compared to mainJand Pttuophts. The supralabials are usually 8 but may be 9 or rarely 10. Only one supra! abial touches the eye; this is the fourth where the total is 8, or the fifth if there are 9 or 10. The infralabials are generally 12 but range from 10 to 13. The loreals are usually single but there is a small subloreal in 3 counts out of 18. The pre- oculars are single in 9 counts out of 18, and paired m the others; the second pre- ocular, if present, is always small and is the lower of the two. The postoculars number 3 in 12 out of 18 counts, 4 in the others. They are subequal in size. The temporals vary from 2 + 3 to 3^4 or 4-3. They average fewer and are more regular than in the mainland snakes. There are 4 prefrontals m all specimens. They are not laterally divided or irregular in any other way, nor has any specimen an azygos. The supraoculars touch the prefrontals in some 44 San Diego Socifty of Natural History specimens but not in others. The parietals are regular and are triangular in shape. The ratio of the tail length to length over-all averages about .173 in the males and .156 in the females. The longest specimen, a male, measures 788 mm.; the shortest 228 mm. The body blotches vary from 58 to 85 (mean 75.6), and the tail spots from 17 to 26 (mean 22.2). The pattern comprises a series of dorsal spots or blotches which are usually black from head to tail, although they may be brownish, or the scales compris- ing them may have brown centers, at mid-body. Anteriorly the blotches are generally longer than wide, being about 5 scale rows wide and 3 scales (end to end) long. They are quite irregular, and often confluent with each other, or with the first subsidiary series on either side, especially on the neck. Poster- iorly they are round or oval, and are better separated and more distinct. There are generally 4 or 5 lateral series of blotches, the lowest of which marks the outer edges of the ventrals. The tail also is somewhat spotted below; in some specimens there are a few triangular spots in parallel rows. The ground color is brown, or gray, middorsally and laterally, with a gray streak between, which is rather characteristic of this subspecies. The ventrum has a cream or buff ground color. On the head a brown transverse bar between the prefrontals and supra- oailars is sometimes present, but in other specimens is obsolete; the same may be said of a postocular dark streak which seems to be lost with age. There are generally a few black or brown spots on the frontal and parietals. There is always a vertical dark bar, or triangle, in the supralabial suture below the eye; and other labial sutures are often touched with dark. The head has a brown ground color above, and cream or buff below. Rnnge. — Pumilus is found only on Santa Cruz Island, one of the Santa Barbara group, off the coast of California. Remarks. ^Pumilus most nearly resembles the mainland snakes in the vicinity of Santa Barbara, which are annectens, with tendencies toward catenijer. From these it does not differ greatly in pattern, although slightly more streaked anteriorly, and often (but not always) with narrower anterior dorsal blotches. With respect to scutellation, in comparison with these mainland snakes pumilus has fewer ventrals and subcaudals on the average, and more often has single preoculars. Pumilus more frequently has 2 temporals in the first row than is the case in the mainland specimens. While the difference is significant, there is over- lapping. The temporals do not usually constitute a satisfactory character for identification in Pituophis because of the irregularity of their arrangement. The most consistent difference from the mainland Pituophis is in scale rows. Snakes with only 29 rows are rather unusual on the mainland of Cali- fornia, while on Santa Cruz Island all available specimens have 29 or less. Comparing our nine specimens of pumilus with a series of eight annectens, Klauber — Neu" Gopher Snake 45 kindly sent me from Santa Barbara by Mr. Waldo G. Abbott, we have the following comparative distribution: Scale Santa Cruz Santa Rows Island Barbara 11 2 28 1 29 6 30 31 3 32 2 33 2 34 1 Or we may make the foilow-ing comparison with all available scale-row data on catemfer catenifer and c. annectens from the mainland: Scale Rows P. c. pumilus P. c. catenijer P. c. annectens 27 1 28 1 29 6 20 9 30 3 10 31 172 139 32 7 40 33 128 242 34 3 25 35 18 59 36 3 37 1 5 38 1 9 352 533 It will be observed that only about 5.7 per cent of catenijer catenijer and 1.7 per cent of c. annectens have 29 scale rows, whereas all c. pumilus have 29 or less. Although 9 is a relatively small sample from which to estimate the actual papulation composition of pumilus. there is no doubt as to the signi- ficance of this difference. Twenrv-nine scale rows occur most freauently in catenijer catenijer in snakes from the Sacramento Valley. Specimens of deserti- cola with 29 scale rows are quite prevalent in some areas, but this subspecies differs from pumilus in other characters of lepidosis, pattern, and proportionate tail length. Another important difference in pumilus is its size. We have reason to 46 San Diego Society of Natural History believe it to be a stunted form, although this is difficult to prove. The disper- sion curve of sizes in a population of snakes does not lend itself to mathemati- cal analysis, so we are unable to make estimates of the degree to which size limits will be increased as more specimens become available, although we know there will be such an increase. We have the following evidence concerning the relative sizes of pumilus and other subspecies: 1. Comparisons with other island forms, of which relatively small samples are available: Number of Size Range Subspecies Specimens mm. P. c. pumilus 9 228 - 78S P. c. insulanus 7 460 - 1213 P. c. juliginatus 15 ' 327 - 1390 P. c. coronctlis 4 353-1150 P. c. amiectens (Santa Catalina) 3 489 - 1051 2. A similar comparison may be made with the Abbott series of main- land snakes from Santa Barbara: Number oj Size Range Locality Specimens mm. Santa Cruz Island 9 228 - 788 Santa Barbara 8 361 - 1478 3. Out of over 300 catenijer catenijer the shortest normal specimen is 310 mm. There is one specimen, MVZ 2314, which is only 202 mm., but this is clearly abnormal. It has a peculiar navel which divides some ten of the adjacent ventrals, although it was found on a lawn, and therefore it is not an unhatched embryo. One may doubt whether it would have survived. The shortest pumilus (228 mm.) is not a freak . It was collected in April, and therefore must have been more than six months old. The next to the shortest specimen, CAS 36121, also collected in April (but by a different collector, in a different year) measures 272 mm., considerably smaller than any catenijer catenijer. 4. The smallest annectens, out of some 416 specimens collected in the wild, measures 350 mm.; the next smallest 359. Out of 73 young annectens hatched by Mr. C. B. Perkins at the San Diego Zoo, the two shortest measured 322 mm. each. This was in a brood of 10, all notable for their small size (range 322 to 346 mm., mean 333); the smallest in any other brood measured 355 mm. Summing up, I think the evidence rather strong that pumilus is a stunted subspecies, distinrtly smaller than the mainland subspecies catenijer and annectens. Neither of the latter, by the way, is particularly large among Pitiiophis; they are considerably exceeded in size by ajjinis, bimaris, and sayi. Klauber — Niiw Gopher Snake 47 The only contrary evidence is a comment by Mr. E. Z. Rett, of the Santa Barbara Museum of Natural History, to the effect that he saw only one snake on the island in many visits. This was a gopher snake which he tried to capture but failed. It appeared to him to be 4 or 5 feet in length. In view of the overlapping in scale rows, I consider only a subspecific designation warranted, even though actual intergradation is not possible. One pumilus contained the remains of a small mammal. Acknowledgments. — I am greatly indebted to Mr. J. R. Slevin, California Academy of Sciences; Mr. Thomas L. Rodgers, Museum of Vertebrate Zoology, University of California; Miss Margaret Storey, Natural History Museum, Stanford University; and Mr. Arthur Loveridge, Museum of Comparative Zoology, Harvard University, for the loan of specimens in die collections of which they have charge. Mr. Waldo G. Abbott kindly furnished specimens from Santa Barbara, which have been useful for comparison. To Mr. C. B. Perkins I wish to e.xpress my thanks for critical suggestions. Siwimary. — Pituophis catenijer pumilus is described as a new subspecies from Santa Cruz Island off the coast of southern California. It differs from its nearest relatives of the mainland, P. c. catemfer and P. c. annectens, in having fewer scale rows, on the average, and in probably being a stunted form. Klauber — New Gopher Snake I'l-ATL 3 Fig. 1. Pituophis ccttenifer pumilus Santa Cruz Island, California. Photograph by J. R. Slevin, California Academy of Sciences. :i^f9^ MAR 14 1S4Z TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 3, pp. 49-52 February 25, 1947 AN UNDESCRIBED RACE OF THE MANGROVE WARBLER FROM BAJA CALIFORNIA, MEXICO BY A. J. VAN ROSSEM Department of Zoology, University of California, Los Angeles In April, 1927, Laurence M. Huey collected a series of nine breed- ing Mangrove Warblers at San Ignacio and Pond lagoons on the west coast of central Baja California. The locality (Lat. 27" N) is the northernmost point at which mangroves (Rhizophora) occur on the Pacific coast and is the only one known north of the extensive "mang- lares" at Magdalena Bay where this ecologically restricted warbler is to be found. A description of these isolated lagoons has been written by Huey (Condor, 29, 1927: 239-243) and they need not be further dis- cussed here. Although the specimens referred to have been recorded by the col- lector and later cited by Grinnell (Univ. Calif. Pub. ZooL, 32, 1928: 199), their taxononic status had never been determined until I recently had the opportunity to do so when investigating the Baja California collections at the Natural History Museum. They are quite distinct from Dendroica erithachorides castaneiceps of the southern and eastern portions of the peninsula and are described below. In the bestowal of a formal name I am mindful not only of a long and friendly association but also of Mr. Huey's generosity in communicating to me much of his extensive knowledge of the distribution of birds in Baja California, 50 San Diego Society of Natural History Dendroica erithachorides hueyi new subspecies San Ignacio Mangrove Warbler Type. — Breeding male adult, number 11471, San Diego Society of Natural History; San Ignacio Lagoon, Pacific coast of Baja California, April 18, 1927; collected by Laurence M. Huey. Subspecijic characters. — Most nearly similar to Dendroica erithachorides castaneiceps Ridgway of southern Baja California and resembling that race in relatively large size, restriction of yellow on inner webs of rectrices, and obscure or obsolescent ventral streaking of adult males. It differs from castaneiceps, however, in decidedly darker and more olive (less yellowish) green coloration of the upper parts, and still further restriction of yellow on inner webs of rec- trices. Range. — San Ignacio and Pond lagoons on the Pacific coast of central Baja California. Remarks- — The specific name erithachorides is used only after some con- sideration. I am well aware that current concept favors the consolidation of the three closely related groups, the Yellow Warblers {aestiva) , the Golden Warb- lers {petechia), and the Mangrove Warblers {erithachorides) under the single specific name of petechia. But it seems to me that what is perhaps only fortui- tous overlap of j'of/ze characters has been overemphasized and, conversely, that more basic differences including rigidly restricted environment (during the breeding season at least) have correspondingly been minimized. Whether or not the Mangrove Warblers which breed northerly in Sonora and Baja California are migratory is a moot question. The weight of evidence at this time is that such is the case. There is the comment by Frazar (Brewster, Birds of the Cape Region, 1902, p. 182), of a slight seasonal fluctuation in numbers at La Paz. More definite is the observation by Bancroft (Condor, 32, 1930: 42), of the "appearance, in migration, in early May" at San Lucas (27° 14'), the most northerly point of occurrence on the Gulf side of the peninsula. Finally, there is my own field experience which has failed to find the species at all during the winter months of December to mid-March in Sonora, and the detection of but a single individual at Concepcion Bay on the Gulf coast of Baja California in January, 1932. I believe from this evidence, which is perhaps indicative rather than conclusive, that it is reasonable to suppose a withdrawal from the most northerly areas in winter. Among the 78 specimens examined from the range of castaneiceps there is one individual, a female (13533 Dickey Coll.), taken at La Paz on October 2, 1923, which has the characters of hueyi in such positive degree that it cannot be included in the known range of variability (which is considerable) of female castaneiceps. On this basis I tentatively include La Paz as within the dispersal range of hueyi. Further evidence would be desirable, but in view of the limited breeding range and consequent numerical rarity of hueyi its detection in local- ities where castaneiceps is common is pretty much a matter of chance. To make sure that the type of castaneiceps might not be such a specimen I submitted two Van Rossem— New Mangrove Warbler 51 of the San Ignacio males to Dr. Alexander Wetmore. He informs me that the type (89940 U. S. Nat. Mus.), an adult male collected at La Paz, December 16, 1882, by Lyman Belding, is representative of the population resident at La Paz. The breeding ranges of the three races of the Mangrove Warbler which occur in the Gulf area are summarized here. Distribution is not continuous but is restricted to estuaries, bays, and shores where an adequate growth of man- groves occurs. Dendroica erithachorides rhizophorae van Rossem SoNORA Mangrove Warbler Dendroica erithachorides rhizophorae van Rossem, Trans. San Diego Soc. Nat. Hist., 8, No. 10: 67-68, August 24, 1935. (T6bari Bay, Sonora, Mexico). Coasts of Nayarit (fide Wetmore), Sinaloa, and Sonora, north to Tepopa Bay at lat. 29" 18' N. Dendroica erithachorides castaneiceps Ridgway La Paz Mangrove Warbler Dendroica bryanti castaneiceps Ridgway, Proc. U. S. Nat. Mus., 8, No. 22: 350, Sept. 2 [Sept. 17], 1885. (La Paz, Lower California, [Mexico]). Coasts of Baja California, north on the Gulf to San Lucas, lat. 27° 14' N, and on the Pacific to Magdalena Bay and San Jorge, lat. 25*^ 45' N. Dendroica erithachorides hueyi van Rossem San Ignacio Mangrove Warbler Pacific coast of central Baja California at San Ignacio and Pond lagoons, lat. 27''' N. Specimens examined. — rhizophorae, 29 from Sinaloa (Mazatlan) 4, and Sonora (Agiabampo; Tobari Bay; Guaymas; Kino Bay; Tepopa Bay), 25. castaneiceps, 78 from Baja California (La Paz; San Jose Island; Espiritu Santo Island; Concepci6n Bay; San Lucas at 27° 14' N; Magdalena Bay, including Santa Margarita Island, Santa Magdlena Island, and San Jorge), hueyi, 10 from Baja California (San Ignacio Lagoon, 6; Pond Lagoon, 3; La Paz, 1). I am indebted to the Curators at the Chicago Natural History Museum, the Los Angeles Museum, the Museum of Vertebrate Zoology (Berkeley), the Museum of Zoology (Ann Arbor), and the Natural History Museum (San Diego) for use of the material under their care, and to Laurence M. Huey and Max M. Peet for the privilage of inspecting their private collections. MUS. COI^P. ZQOL LIBRARY FEB 21 I9ii| UNiyERSITY TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 4, pp. 53-56 January 31, 1949 THREE NEW RACES OF POCKET GOPHERS (THOMOMYS) FROM BAJA CALIFORNIA, MEXICO BY Laurence M. Huey Curator of Birds and Mammals, San Diego Society of Natural History Field work in the more remote sections of central Baja California on both the Gulf and Pacific Coast sides has revealed three unknown races of Tbo}no7nys. When the writer published his account, 'The Pocket Gophers of Baja California, Mexico, with Descriptions of Nine New Forms" (Trans. San Diego Society of Natural History, Volume 10, pp. 245-268, Map, August 31, 1945), he suggested the possible existence of yet undiscovered races of Thomomys in remote unexplored sections of the peninsula. This suggestion has been verified. Small collections made on two short trips have brought to light the following undescribed races: Thomomys bottae ruricola subsp. nov. Santa Catarina Pocket Gopher Type.— From 4 miles north of Santa Catarina Landing, Baja Californid, Mexico, lat. 29° 35' N., long. 115° 17' W.; No. 15944, collection of The San Diego Society of Natural History; adult male; collected by Laurence M. Huey, June 27, 1947. Characters. — Compared with Tlfomomys bottae abbotti, T. b. ruricola is less brightly colored on the back and sides, tending toward a grayer cast. The black auricular spot is much smaller than that of either T. b. abbotti or T. b. catavinensis, the ranges of which are north and eastward. Cranially, the spread of the zygomatic arches of T. b. ruricola, when viewed dorsally, gives the appearance of being more nearly circular and not angular, as in abbotti. The rostrum is shorter and has an accentuated ridge at the junction of the premaxilla-zygomatic plate. The molariform teeth are smaller, with the pterygoid fossa wider and more U-shaped. The bullae are flatter, less inflated, but flare outward anteriorly. Compared with T. b. catavinensis, T. b. ruricola 54 San Diego Society of Natural History is much lighter in color. This race has a larger, more rugose skull, with a wider spread anteriorly, but having less inflated bullae. Measurements. — Type: Total length, 223; tail, 70; hind foot, 28; ear, 5. Skull (type) : Greatest length, 38.9; spread of maxillary arches, 24.7; length of nasals, 13.8; interorbital constriction, 5.9; alveolar length of upper molar series, 8.9. Range. — Known only from the type locality. Remarks. — The region in which this specimen was taken has the appear- ance of being uninhabitable by gophers and it was with considerable surprise that this specimen was captured with a mouse trap set for kangaroo rats, Dipodomys. Further search failed to reveal the presence of a single gopher mound. It has been the writer's experience that in arid regions a fairly large Thomomys population can spend long periods underground without showing any surface activities, such as dirt mounds, that would reveal their presence to a mammalogist. During these subsurface periods they exist on roots of different shrubs and larger plants and often a dead cactus will prove to have had its root system completely eaten away while over the nearby surface of the ground not a single gopher indication is visible. Specimens examined. — 1 (the type) from 4 miles north of Santa Catarina Landing, Baja California, Mexico. Thomomys bottae abbotti, 17 from 1 mile east of El Rosario, Baja California; 2 from San Fernando Mission, Baja California. Thomomys bottae catavinensis, 12 from Catavifia, Baja California. Thomomys bottae rhizophagus subsp. nov. Los Angeles Bay Pocket Gopher Type. — From Las Flores, 7 miles south of Bahia de Los Angeles, Baja California, Mexico, lat. 28° 50' N., long. 113° 32' W.; No. 15710, collec tion of The San Diego Society of Natural History; adult male; collected by Laurence M. Huey, April 11, 1947. Characters. — In dorsal coloration this form differs little from its nearest relative T. b. borjasensis, the range of which lies on the Pacific slope to the westward, but on the sides it is slightly more tawny. It is a member of the more dark-colored group of Thomomys, differing decidedly from the gray T. b. russeolus found to the southward over the Viscaino Desert. Cranially. however, there is wide divergence. The skull of T. b. rhizophagus, when compared with that of T. b. borjasensis, is seen to have a longer, heavier, and wider rostrum, with a rounder, more inflated braincase. The angle of the zygomatic arches is less flaring, giving the skull a longer, more streamlined appearance. The bullae are more compressed and less inflated. T. b. rhizophagus has a longer tooth row and a larger hind foot than T. b. bor- jasensis. Measurements. — Type: Total length, 222; tail, 76; hind foot, 30; ear 5. Skull (type): Greatest length, 37.1; spread of maxillary arches, 21.9; length of nasals, 13.8; interorbital constriction, 6.2; alveolar length of upper molar series, 8.1. HuEY— Pocket Gophers (Thomomys) 55 Range.— Known only from the type locality, where it was extremely abundant, living in sandy soil, and feeding on the roots of desert apricot mallow, Sphaeralcea ambigua, which was growing in thicket-like patches over the dry, alluvial valley floor. Remarks. — This is the first race of Thomomys to be found close to the Gulf shore north of La Paz. Further exploration, either north or south of Bahia de Los Angeles, may reveal other isolated gopher populations that will be of decided interest to the mammalogist. Specimens examined. — 11 from Las Flores, 7 miles south of Bahia de Los Angeles, Baja California (type locality). Thomomys bottae borjasensis, 1 (the type) from San Borjas Mission, Baja California. Thomomys bottae homorus subsp. nov. Rancho Lagunitas Pocket Gopher Type. — From 1 mile east of Rancho Lagunitas, Baja California, Mex- ico, lat. 28° 20' N., long. 113' 15' W.; No. 15689, collection of The San Diego Society of Natural History; adult male; collected by Laurence M. Huey, April 6, 1947. Characters. — Compared with T. b. russeolus, its nearest comparatum, T. b. homorus is slightly darker in color, having a tinge of brownish in the pelage, though it belongs to the gray-colored races usually found in more arid desert sections. Cranially, the differences are well pronounced. T. b. homorus has a narrower skull, with a rounder, more elongated braincase. The incisors are less procumbent and the premaxillary tongues extend, well beyond the nasals, much farther than any specimen of T. b. russeolus exam- ined. The bullae are slightly flatter with less inflation and the hamular processes much more fragile than those of russeolus. Measurements. — Type: Total length, 227; tail, 73; hind foot, 30; ear, 5. Skull (type) : Greatest length, 38.5; spread of maxillary arches, 23.2; length of nasals, 12.9; interorbital constriction, 6.4; alveolar length of upper molar series, 8.1. Range. — Local populations found in suitable localities in hilly sections of extreme northeastern Viscaino Desert from vicinity of Calmalli (alt. 1200 feet) eastward to the summit of the Peninsular backbone near the type locality of this race, Rancho Lagunitas (alt. about 1900 feet). Remarks. — Specimens from Calmalli were determined as not being typi- cal of the race T. b. russeolus when the earlier paper, herein mentioned, was written. These specimens are now found to be nearer T. b. homorus and represent the western limits of the range of this race. Specimens examined. — 4 from Calmalli, Baja California; 8 from Rancho Union, 15 miles east of Calmalli, Baja California; 8 from Rancho Lagunitas, (lat. 28° 20') Baja California. T. b. russeolus, 19 from Campo Los Angeles, Viscaino Desert, Baja California. MUS. eO?>^P. ZQOL FEB 21 19^.9 UH!VERSiTY TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 5, pp. 57-60 January 3L 1949 THE RELATIONSHIP OF CROTALUS RUBER AND CROTALUS LUCASENSIS BY Laurence M. Klauber Curator of Reptiles and Amphibians, San Diego Society of Natural History The close relationship of Crotalus ruber and Crotalus lucasensis has long been recognized, for both are obviously derived from Crotalus atrox, as shown by lepidosis, pattern, and hemipenes. The problem of their relationship would have been settled long ago were it not for the fact that the gap that lies between their ranges is almost inaccessible. This is the Sierra de la Giganta in southern Baja California, along the Gulf of California coast. This range of mountains is quite suitable in character to occupation by either ruber or lucasensis, but the only north-south road in that section of the peninsula skirts far to the west of these mountains, in a desert not so much to the liking of these diamondbacks. Even so, lucasensis has been collected in this desert as far north as Yrais (Hiray), and 8 miles south of El Refugio; and there is a questionable record from 40 miles south of Comondu. Ruber, for its part, is found along the shores of Bahia de la Concepion, the most southerly specimen having been taken at the head of the bay, near La Cruces. In July, 1938, Robert S. Hoard collected a series of 7 diamondbacks, which obviously show intergradational tendencies, at Loreto, near the north end of the Sierra de la Giganta about 60 miles southeast of Las Cruces. I had hoped that further specimens might be forthcoming from elsewhere in these mountains, but, after waiting 10 years without securing more, have now decided to see how strong a case for intergradation may be built upon them. Fortunately, both ruber and lucasensis are well represented in collec- tions, so that their characters can be firmly established; the scale counts are available for 372 specimens of ruber and 348 of lucasensis. Of the ruber series, 72 are from Baja California, the rest from southern California. 58 San Diego Society of Natural History C. ruber and lucasensis differ in scutellation and pattern. None of the scale differences is sufficiently restrictive or consistent to be useful as a key character, for there is much overlapping, but we can draw inferences from the trends in the Loreto specimens. C. lucasensis has fewer scale rows, on the average, than ruber: 76 per cent of lucasensis have 17 or fewer, whereas 83 per cent of ruber have 28 or more. It is to be noted, however, that the Baja California specimens of ruber have a reduced percentage with 28 and more, only 65 per cent having the higher number. Of the Loreto specimens, 6 have 27 rows and one 29, thus favoring lucasensis in this character. On the average, ruber has about 5 ventrals more than lucasensis, the figures being: ruber males 193.8, females 197.3; lucasensis males 188.8, females 192.6. The Loreto males have 188, 190, 192, 192; and the females 193, 195, and 196. In this respect the Loreto specimens are intermediate, with a slight leaning toward lucasensis. The differences between the species in subcaudals and labials are of insufficient extent to be of value in diagnosis. C. ruber has a somewhat higher percentage of divided first infralabials than lucasensis (91 to 87), and all of these scales in the Loreto specimens are divided; however, lucasensis, itself, runs so high in having this division that this cannot be considered of importance. The minimum scales between the supraoculars are slightly higher in ruber than lucasensis (averages 6.5 and 5.9), and here the Loreto specimens somewhat favor ruber, for 3 speci- mens have 7 scales in this series and the others 8. The majority of ruber specimens have single loreals (mean 1.22 per side), while lucasensis usually has 2 or more (mean 1.96 per side). The Loreto specimens have single and paired loreals in equal numbers, so that they are intermediate in this character. The body blotches in ruber average 36.3, in lucasensis 29.9. The Loreto specimens number 28, 29, 30 (4 specimens), and 31, thus definitely resemb- ling lucasensis. As to pattern and color, ruber tends toward red or red-brown, lucasensis toward yellow- or olive-brown. Of the two, lucasensis is lighter and more brightly marked; the light scale rows bordering the dorsal diamonds are more accentuated,* the head marks clearer, and the lateral angles of the blotches sharper, more diamond- like, and less often hexagonal or round as in ruber. C. lucasensis. more often than ruber, has light areas in the centers of the dorsal blotches. Light cross-marks on the supraoculars are more apparent in lucasensis than in ruber: and the light preocular stripe is wider, being usually 3 or more scales wide *This is also true of the desert specimens of ruber in San Diego County, California. Klauber — Crotalus Ruber and Lucasensis 59 (at the second row of scales above the supralabials) , rather than 2 or fewer as in ruber. Surveying the 7 Loreto specimens with respect to these criteria, we find them intermediate. In color they are dark red-brown, somewhat more Hke ruber than lucasensis, but the light blotch-borders are more prominent than is usual in ruber. While the blotch shapes favor ruber, the light blotch centers characteristic of lucasensis are present, although less accentuated than is customary in the southern species. The light supraocular cross-marks are clearly present in 3 specimens, but faint or absent in the others. The light preocular stripes are clear in all specimens, and in none is the scripe less than 3 scales wide, thus showing an affinity for lucasensis. Summarizing, it is my opinion that the Loreto specimens are inter- mediate, with a slight leaning toward lucasenns. Loreto is only 60 miles southeast of the head of Bahia de la Concepcion, where pure ruber occurs, although admittedly not precisely the same in all characters as the ruber of San Diego County. But at least we now have evidence of an unbroken cline from the Santa Ana Mountains of southern California to Cape San Lucas at the southerly tip of Baja California, and only a single species should be recognized. The northern race should be known as Crotalus ruber ruber Cope, 1892; and the southern as Crotalus ruber lucasensis Van Denburgh, 1920. Crotalus r. lucasensis is more like C. atrox than is C. r. ruber, notwith- standing their separation by the Gulf of California, whereas the ruber and atrox ranges approach each other closely at several points. Indeed, it is quite likely that they overlap in the vicinity of La Quinta and Indian Wells, Riverside County, California. But there are no suggestions of intergradation either here or at other points where the ranges are closely contiguous. How- ever, I should not be surprised were the future to bring forth an occasional hybrid. Despite my recent treatment of certain island forms, in considering them subspecies of mainland forms with which they can no longer intergrade, I do not believe it desirable to join exsul of Cedros Island with ruber of the main- land as a single species. Although .the relationship between the two is readily apparent, I think that the known character differences are sufficient in extent to justify a continued specific separation. C. exsul is a better-differentiated form than the other island races that I have considered subspecies. ;> ( 1 ' > b^-..^^- LIERASY FEB 21 19^49 HAflVARO UNIVERSiTY ' TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 6, pp. 61-116, plates 4-6. SOME NEW AND REVIVED SUBSPECIES OF RATTLESNAKES Laurence M. Klauber Curator of Reptiles and Amphibians, San Diego Society of Natural History SAN DIEGO, CALIFORNIA Printed for the Society January 31, 1949 COMMITTEE ON PUBLICATION Joshua L. Baily Carl L. Hubbs Laurence M, Klauber m%. ^m?. ZQOL LIBRARY FEB 2 1 I9i^9 Kany^RO TABLE-DF CON IHNTS \Sa-^ b \ <: o, Introduction A Subdivision of the Pacific Rattlesnake Crotalus viridis oregantis Holbrook Diagnosis Nomenclatorial and Systematic Problems Redescription of the Type Specimen Material Description of Subspecies Intrasubspecific Trends Intersubspecific Relationships and Areas of Intergradation Range Crotalus viridis helleri Meek Diagnosis Nomenclatorial and Systematic Problems Redescription of the Type Specimen Material Description of Subspecies Intrasubspecific Trends Intersubspecific Relationships , Range Crotalus viridis cerberus (Coues) Diagnosis Nomenclatorial and Systematic Problems Redescription of the Type Specimen Material Description of Subspecies Intrasubspecific Trends Intersubspecific Relationships Range and Locality Records A New Island Subspecies of Crotalus viridis 90 Crotalus viridis cdiginis subsp. nov 90 Type Specimen 90 Diagnosis 90 Description of the Type 90 Material ^^ Description of Subspecies 92 Intersubspecific Relationships 94 Food Habits 95 A New Island Subspecies of Crotalus mitchellii 96 Crotalus mitchellii muertensis subsp. nov 97 Type Specimen 97 Diagnosis 97 Description of the Type 97 Material 98 Description of Subspecies 98 Intersubspecific Relationships 100 Field Notes 102 The Black-tailed Rattlesnake of San Esteban Island 103 Crotalus molossus estebanensis subsp. nov 104 Type Specimen 104 Diagnosis 104 Description of the Types 104 Intersubspecific Relationships 105 Acknowledgments 107 Summary 107 Bibliography 107 SOME NEW A MUS. COMP. ZOOL FEB 21 19^91 ND ^ RATTt: ^I^^D SUBSPECIES OF RkTTtESNAKES BY Laurence M. Klauber Curator of Reptiles and Amphibians, San Diego Society of Natural History Introduction For the past several years I have been engaged in writing a paper on the habits and Ufe histories of the rattlesnakes. While the proposed paper will not involve a taxonomic study of these snakes, I do hope to include a key and nomenclatorial summary; and in order that the larger work may contain the most recent views respecting the subspecies to be recognized, I think it advisable to publish, first, some conclusions entailing the further subdivision of several hitherto valid forms into additional subspecies, some of which are island races. A Subdivision of the Pacific Rattlesnake For some time the Pacific Rattlesnake, Crotalus viridis oreganus, has been considered a single subspecies, ranging along the Pacific slope from southern British Columbia into northern Baja California, with an isolated population inhabiting the mountains of central Arizona. I am of the opinion that there would be a practical advantage, facilitating reports and discussions, if further nomenclatorial subdivisions were made in this widespread sub- species, and that such a segregation is warranted by consistent regional dif- ferences. No new names are required except for an island form, it being only necessary to retrieve from synonymy some of the names hitherto pro- posed. The subspecies of the northern Pacific Coast, from south-central California northward, will continue to be known as Crotalus viridis oreganus Holbrook, 1840; the southern California and Baja California population will become C. v. helleri Meek, 1905; and the Arizona subspecies will be C. v. cerberus (Coues), 1875. With regard to the segregation of these populations, it may be said that differences in pattern separate the first pair. When from their centers of distribution — say central Washington and San Diego County, California — it is possible to make allocations that will prove accurate in better than 98 per cent of the individuals, employing the width and color of the posterior tail rings as distinguishing criteria. But the zone of intergrad- ation (and therefore of uncertainty) between the two, in south-central Cali- fornia, is fairly broad, especially in the Coast Range. Also, the differences in the life zones inhabited by these snakes have led to the formation of local varieties, which, while not justifying nomenclatorial recognition, are sufficient- ly divergent to lead to some confusion and overlapping in defining each subspecies as a whole. The differences between helleri and cerberus are based somewhat less on pattern and color, and to a greater extent on average differences in head scales, especially the loreals and the contacts between the first supralabials and the prenasals. In this case there is no intergrading, for the ranges of these 66 San Diego Society of Natural History two subspecies, at their nearest points, are separated by 200 miles of desert, uninhabitable to them; however, there is some overlapping of the characters used in segregation, and only a subspecific differentiation is warranted, despite the impossibility of territorial intergradation. The differences between oreganus and helleri are probably as important and consistent as those between either and other subspecies of Crotalus viridis, such as lutosus and viridis. The helleri-cerberus differences are somewhat less consistent, but in this case we have the added weight of a complete territorial separation. To indicate the similarity of the three subspecies in scale-counts and pattern, I have set forth in Table 1 the statistics of three example populations: a series of 615 specimens of oreganus from near Pateros, Okanogan County, Washington; 645 of helleri from the vicinity of San Diego, California; and 97 of Cerberus from Arizona. Unfortunately, I have no large series of the latter from a single locality, so that the last series involves some regional variability. Altogether, of the forms previously considered to comprise the single Pacific ratdesnake subspecies C. v. oreganus, over 2000 specimens have been available for this study. In Table 1, it will be observed that there are no important subspecific divergences in these primary characters. The differences are relatively great- est between the helleri and cerberus tail rings, the coefficients of divergence being, for the males, 34.5 per cent, and for the females 33.8. Actually, how- ever, these differences are of little practical interest, as the tail rings of these subspecies are often too ill-defined to permit accurate counting. TABLE 1 Comparison of Example Series of Rattlesnakes: Mean Scale and Blotch Counts Pateros San Diego Co. Arizona Character oreganus helleri cerberus Scale rows 25.51 25.45 25.12 Ventral scutes, males 173.31 173.24 171.23 females 177.98 177.97 175.41 Subcaudal scutes, males 22.99 24.00 23.58 females 18.84 19.44 19.07 Supralabials 15.22 15.17 15.47 Infralabials 16.18 16.03 16.03 Body blotches, males 33.20 34.84 35.24 females 33.19 35.39 35.71 Tail rings, males 5.50 4.27 6.05 females 4.47 3.29 4.63 Pateros San Diego Co. Arizona oreganus helleri cerberus -2.7 -2.7 -2.4 19.8 21.0 21.1 0.0 -1.6 -1.3 20.6 25.9 26.6 Klauber— New Rattlesnakes 67 The coefficients of sexual divergence (the difference between the means of each sex, divided by half their sums, expressed as a percentage), are as follows : Character Ventral scutes* Subcaudal scutes Body blotches* Tail rings It will be observed that these sexual differences are quantitatively con- sistent throughout the subspecies. The sexual dimorphism in body blotches is found to be negligible, and therefore no sexual segregation will be made in this character in the subsequent discussion. Since these three subspecies have not hitherto been segregated in pub- lished accounts, I present the following descriptions of the forms as newly delimited: Crotalus viridis oreganus Holbrook Northern Pacific Rattlesnake! Plate 4, fig. 1. (Alternative names: Pacific Rattlesnake, Oregon Ratdesnake, North- western Rattlesnake) . 1840. Crotalus oreganus Holbrook, North American Herpetology, ed. 1, vol. 4, p. 115. Type locality: Banks of the Oregon or Columbia River [probably between Walla Walla, Wash., and the Pacific Coast]. Type specimen ANSP 7158. 1842. Crotalus oregonus Holbrook, North American Herpetology, ed. 2, vol. 3, p. 21. 1852. Crotalus luafer Baird and Girard, Proc. Acad. Nat. Sci. Phila., vol. 6, p. 177. Type locality: Oregon and California. Type specimen: USNM 7762 (Oregon). 1859. Crotalus lecontei (part) Hallowell (not of Hallowell, 1852), Pac. RR. Surv. (Williamson route), vol. 10, pt. 4, no. 1, p. 18. 1860. Caudisona lucijer Cope, in Mitchell, Smithsonian Cont. Knowl., vol 12, art. 6, p. 121. 1868. Crotalus Hallowellt Cooper, ^in Cronise, Nat. Wealth Calif., p. 483 (nomen nudum). '■'Negative values indicate that the female means are higher than the male. + Among the vernacular names used locally are diamond-back rattler, black dia- mond, timber rattler, mountain rattler, green timber ratder, green rattler, and gray rattler. The belief is quite general throughout its range that there are two separate species — a lighter, slimmer, lowland form, and a darker, heavier, and more pugnacious mountain species. In many areas the young are called sidewinders, while in others only the lighter, lowland specimens are referred to as sidewinders. However, this sub- species only approaches (but does not overlap) the range of the true sidewinder, Crotalus cerastes cerastes, at the southeastern foot of the Tehachapi Mountains. 68 San Diego Society of Natural History 1883. Crotalus oregonus var. oregonus Garman, Mem. Mus. Comp. Zool., vol. 8, no. 3, p. 173. 1883. Crotalus oregonus var. lucijer (part) Garman, Mem. Mus. Comp. Zool., vol. 8, no. 3, p. 173. 1883. Crotalus confluentus lucijer Cope, Proc. Acad. Nat. Sci. Phila., vol. 35, p. 11. 1896. Crotalus confluentus (part) Boulenger, Cat. Snakes British Mus., vol. 3, p. 576. 1929. Crotalus confluentus oreganus (part) do Amaral, Bull. Antivenin Inst. Amer., vol. 2, no. 4, p. 92. 1936. Crotalus viridis oreganus (part) Klauber, Trans. San Diego Soc. Nat. Hist., vol. 8, no. 20, p. 191. Diagnosis. — This subspecies is characterized in most areas by a generally lighter color than helleri or cerberus; and by hexagonal or circular blotches, compared with the diamonds prevalent farther south. It has more sharply outlined head marks and lateral secondary blotches. The dark tail rings are of uniform width and are clearly defined, whereas in the other two sub- species the last dark ring is about twice as wide as the others and is ill- defined. In oreganus, especially the juveniles, the terminal tail ring is usually darker than those that precede it, and the matrix of the rattle button is generally black; in helleri and most specimens of cerberus the terminal tail ring in the juveniles is yellow and so is the button. The rattles are of smaller average dimensions in oreganus than in helleri. From viridis viridis and nuntius, oreganus differs in having a wider postocular light stripe and fewer tail rings. It is not vermilion or orange in color like abyssus, and has less widely separated body blotches than lutosus. It is larger and darker than decolor. Nomenclatorial and Systematic Problems. — For about forty years (1860- 1898) the rattlesnakes of the Pacific Coast were referred to the species lucifer Baird and Girard, 1852. However, in 1898 (p. 141) Van Denburgh had an opportunity to examine the type of Holbrook's oreganus, 1840, and correctly determined that this name must supersede lucifer. Garman had de- duced the close relationship between the two as early as 1883 (p. 173); and in the same year Cope (p. 11) became convinced of the subspecific relation- ship between the prairie and Pacific rattlers, as subsequently verified by do Amaral (1929, p. 92) and the writer (1930a, p. 123). Gill (1903, p. 910) called attention to the previously overlooked fourth volume of the first edition (1836-1840) of Holbrook's North American Herpetology, in which the name of this snake is spelled oreganus, instead of oregonus as in the second edition of 1842. Holbrook's work has been further discussed by Schmidt and Davis (1941, p. 12) and by Schmidt (1942, p. 53). It is evident that Holbrook was dissatisfied with the imperfections of the first edition of his work and sanctioned the destruction of most of the copies. It may be presumed, from the correction that Holbrook himself made, that the oreganus of the first edition is a lapsus calami and ought to be changed to the oregonus of the second, in accordance with Art. 19 of the International Klauber^New Rattlesnakes 69 Rules. However, since oreganns has been used almost universally during the past quarter of a century, and as a proposal to modify the rules in cases such as this has recently been discussed,* I deem it inadvisable to make the change at this time. Redescription of the Type Specimen. — The type specimen is Acad. Nat. Sci. Phila. 7158 (old number 840). It is a juvenile male, badly dried and with the surface skin flaked away. No scale counts can be made with com- plete assurance of accuracy. There are 25 scale rows at mid-body, about 175 ventrals, and 24 subcaudals. The supralabials are 16 — 16, and infralabials 15 — 15; the first infralabials are undivided. The scales contacting the rostral probably number 9 and the scales around the canthus 8. There are probably 4, but may be 3 internasals; the canthals are 2 — 3. There are about 20 scales on the top of the snout anterior to the supraoculars, and the minimum bridge between is 4+5. There may be a row of small scales between the prenasal and the first supralabial on the right only. The nasals are 2 — 2 and the loreals 1 — 1. The postnasals contact the upper preoculars, which are undivided. There are no intergenials or submentals. The rostral is higher than wide. The length over-all is about 337 mm., and the tail length 20 mm. The rattle retains only a single ring, which is badly shrunken. The body blotches number 32 and the tail rings 3 or 4, all of which are black. While the marks are still to be seen, it is impossible to determine what the dorsal color once was; at present it is a light bluish-gray, with black lines between scales. It was probably heavily mottled below. The postocular light stripe is two scales wide. Holbrook's figure shows a pattern of irregular dark-edged hexagons on a gray-green ground color. The specimen contains mammal hair. Material. — This study of the subspecies oreganus, as newly defined, is based on the scale counts of 1038 specimens, of which 559 were males, 450 females, and the rest indeterminate. Pattern notes were made on about 150 additional specimens. Because of the large series from near Pateros, Wash- ington,— 326 males, 289 females, 615 total — the numerical statistics con- tained in the description that follows have a northern complexion. I have seen well over 1000 rattlesnakes of this subspecies alive. Description of Subspecies. — -This is a snake of the usual viridis habitus. Like the other viridis subspecies, it differs from all other rattlesnakes in usually having more than 2 internasals; out of 998 specimens only 13.4 per cent had less than three internasals. The longest specimen of the series I have studied was a female measur- ing 1320 mm.; a male of proportionate size would be about 1550 mm. Without doubt 5-foot specimens occur occasionally in some areas. The shortest specimen measured was 221 mm., but specimens under 250 mm. are unusual. It is believed that in most areas the young average about 270 mm. at birth. * R. E. Blackwelder, J. B. Knight, and C. W. Sabrosky, Science, vol. 106, no 2753, pp. 315-6, Oct. 3, 1947; vol. 108, no. 2793, pp. 37-38, July 9,.J948. Wichter- man, Science, vol. 106, no. 2760, p. 491, Nov. 21, 1947. 70 San Diego Society of Natural History In this subspecies individuals having 25 scale rows at mid-body pre- dominate, the distribution in per cent (in parentheses) of those examined being 23(1), 24(1), 25(71), 26(7), 27(20), 28(*), and 29(*). The mean scale rows number 25.44, the coefficient of variation being 3.5 per cent. Specimens having even scale rows at mid-body comprise 7.9 per cent of the total. The ventrals in the males vary from 161 to 190, although most speci- mens fall between 168 and 182. The interquartile range is 170.9 to 176.6, the mean 173.72, and the coefficient of variation 2.4 per cent. The ventrals in the females vary from 167 to 194, although most specimens fall between 170 and 189. The interquartile range is 175.4 to 181.1, the mean 178.24, and the coefficient of variation 2.4 per cent. The subcaudals in the males vary from 18 to 29, but most specimens have from 20 to 27; the interquartile range is 22.3 to 24.6, the mean 23.42, and the coefficient of variation 7.2 per cent. The females range from 15 to 24, although few have less than 16 or more than 22; the interquartile range is 17.9 to 20.0, and the mean 18.95, with a coefficient of variation of 8.1 per cent. The rostral is higher than wide. There are from 1 to 7 internasals, most specimens having 4 (59 per cent), or 3 (20 per cent). Thirteen per cent have 2, and one per cent 1, and would thus fail to key out as belonging to a viridis subspecies. The canthals are usually 3 — 3, but may be 2 — 2 or 4 — 4. The scales on the snout, anterior to the supraoculars, range from 10 to 46, the mean being 24.3, and the coefficient oi variation 23.0 per cent. These scales cannot be counted with accuracy in this or any other viridis subspecies since there is no well defined line across the anterior edge of the supraoculars such as characterizes C. scutulatus, C. tnolossus, and certain other forms of rattle- snakes. The minimum scales bridging the gap between supraoculars vary from 2 to 9, most specimens having 6 (33 per cent), 5 (30 per cent), or 4 (17 per cent). The loreals most often number 2 (54 per cent), or 1 (45 per cent); rarely there are 3, 4, or more. The rostral, first supralabial, and prenasal meet at a point in 80 per cent of the specimens; in 6 per cent there is a group of small extra scales at this point; and in the other 17 per cent these scales are carried back far enough to separate completely the prenasals from the supralabials. The supralabials range from 11 to 18, but most specimens have from 13 to 17; the interquartile range is from 14.5 to 15.7, the mean 15.08, and the coefficient of variation 5.8 per cent. The most prevalent counts are 15 (48 per cent), 16 (24 per cent), or 14 (20 per cent). The infralabials range from 13 to 20, with an interquartile range of 15.4 to 16.7, a mean of 16.08, and a coefficient of variation of 5.9 per cent. The most frequent counts are 16 (41 per cent), 17 (27 per cent), or 15 (23 per cent). The first infralabials are undivided, and there are normally neither intergenials nor submentals. Usually 3 or 4 infralabials contact the mentals . The body blotches range from 20 to 41, although most specimens have from 26 to 39. The interquartile range is 31.3 to 34.8, the mean 33.05, * Present, but less than one per cent. One freak specimen has 33 scale rows. Klauber— New Rattlesnakes 71 and the coefficient of variation 7.8 per cent. The tail rings of the males range from 3 to 10, but nearly all fall between 4 and 7, with a mean of 5.61. The females range from 3 to 8, although only one specimen has 8 and only 4 have 7. The mean is 4.50. The following pattern description is based on this subspecies as it occurs in the vicinity of the type locality in eastern Washington. In the juveniles the head is brown on top, somewhat mottled with gray posteriorly. There are two light cross-marks, the first just posterior to the upper edges of the postnasals, the other crossing the supraoculars and the frontal area between. On the sides there is a broad dark stripe from the eye to a point above the commissure. Below this, the side of the head, in- cluding the supralabials, is light, except that there are aggregations of brown dots from the prenasal, backward and downward, past the pit, to the labials below the eye. Above, the ocular dark streak is bordered by a light line two scales wide that begins at the upper corner of the eye. The lower jaw is mostly clear, but the mental and first infralabials are punctated with brown. There is usually a clump of dots on the fourth and fifth infralabials, and another group on the fourth and fifth anterior to the commissure. In the adult snakes the light cross-lines disappear, the head becoming uniformly brown or olive on top. The upper postocular light stripe also tends to disappear, so that the postocular dark streak merges with the dark color of the upper surface. The side of the head below the postocular dark streak remains light, although considerably punctated anteriorly about the nasals and the pit. The dark postocular streak terminates rather abruptly above the commissure, and there is usually, at the angle of the jaw, a white patch that encroaches onto the top of the head. This, together with the light area anteriorly, gives a characteristic "white-faced" appearance to the snakes of this subspecies in many areas. It is, for example, conspicuous in the adult snakes of the Coast Range of northwestern California, being accentuated in the largest adults. The body pattern in the juveniles comprises a series of brown dorsal blotches about 11 scale rows in width, separated by light interspaces 1 to 2 scales long. The blotches are darkest at their edges, and also the light areas surrounding them are lightest where they border the blotches. The blotches may be round, square, or hexagonal. Laterally, between each pair of dorsal blotches, there is a slightly darkened area of the ground color, but not dark or definite enough to be considered a subsidiary series of spots. Below these, there are two series (the lower smaller than the one above) of dark auxiliary spots; these are, in fact, darker than the dorsal series. The lowest of these lateral spots mark the outer edge of the ventrals. Between blotches, the lateral areas are buff or clay-color, punctated with brown. The ventrum is cream, buff, or yellow, punctated or mottled with gray or brown, especially posteriorly. The dorsal and lateral blotches merge posteriorly to become cross-bars, and the tail is also crossed or ringed by these bars, the last being 72 San Diego Society of Natural History the darkest of the series. The last tail ring is no wider than those that precede it. The button matrix is dark, at least anteriorly. In the adult snakes the lateral secondary spots merge into the ground color, especially anteriorly, so that the lateral areas become almost uniform gray, often with a greenish tinge. The lack of lateral blotches makes the dorsal series the more conspicuous; they are usually brown, and may be round, square, or hexagonal. The light line of scales that edges the dark blotches in the juveniles tends to disappear laterally in the adults. To this extent the snakes of this area resemble cerberus rather than helleri. Posteriorly, the blotches become darker (dark-brown or even black) and join the lateral series to become cross-bars, or even complete rings. The interspaces are lighter caudad, so that these posterior rings are clearly and definitely out- lined, giving a black-and-white effect not evident anteriorly. The last ring is no wider than its fellows, and is, in fact, often narrower, thus differing from helleri: The matrix of the proximal rattle is usually black. The ventrum is heavily mottled or punctated with dark-gray, brown, or black, especially toward the tail. Intrasubspecific Trends. — In this subspecies the snakes having the lowest ventral scale counts are to be found in the vicinity of San Francisco Bay, and those with the highest in the central Sierra Nevada; the difference between the areas is about 12 scales in either sex, the average coefficient of divergence being 7 per cent, which is quite material for an intrasubspecific difference. The snakes of other areas have ventral counts that fall between these two extremes. Specimens from California more often have a single loreal than two, while the reverse is true in Oregon and Washington; also the scales between the supraoculars average lower in California than in the more northerly snakes. While oreganus is quite variable in pattern and color, there are several local phases that are rather consistent and worthy of mention. Along the northwestern coast of California there is a white-faced variety that is quite distinctive. Probably the center of this phase, where the most conspicuous representatives are found, includes northeastern Sonoma, Lake, and western Colusa counties. In these snakes the area below the dark diagonal line that crosses the eye is unmarked, whereas, in specimens from most other areas, this space on the side of the head is grayish.* Also, there is an ex- panded light patch posterior to the angle of the mouth, which even encroaches on the normally dark area on top of the head. The dark ocular streak in these snakes merges uninterruptedly with the dark fronto-parietal area, whereas, in most populations of oreganus, there is usually a light streak be- tween. This white-faced effect is not evident in the juveniles, but comes on gradually with age. It is usually present in the snakes of Oregon and Wash- ington, but not to so conspicuous a degree as. in northwestern California. Another color phase is to be found at tfie' south end of the San Joaquin Valley, particularly along the more arid west side. Here the snakes are * This darkening is particularly evident near the pit, below the light streak that comprises the lower border of the dark ocular stripe. Klauber — New Rattlesnakes 73 considerably lighter than elsewhere, the ground color being yellowish or tan, and the blotches brown. There is a brown patch on top of the head, with ill-defined outlines. In both color and pattern these snakes are reminiscent of lutosus. They merge with the more normally paterned oreganus in the foot- hills or mountains bordering the valley. In the Marysville Buttes, Sutter County, there is a phase that is markedly green in color, and white-faced as well. The tail rings, however, are sharply contrasting black and white. I have heard of a red phase occurring in the Trinity Mountains, but have not seen any that answered this description, although some of the snakes of that area have large red-brown dorsal blotches. In the southern Sierra Nevada the specimens from the higher altitudes are darker than those from the foothills; the dorsal blotches are black, just as is the case with helleri in the mountains of southern California. An altitude of at least 11,000 feet is reached by oreganus in Fresno and the adjacent counties of the southern Sierra Nevada. hitersubspecijic Relationships and Areas of Inter gradation. — C. v. oreganus intergrades with three of the other viridis subspecies — viridis viridis, lutosus, and helleri. All of these are much alike in scale counts and scale arrange- ments, the differences in squamation between them being no greater than some of the differences found to exist between certain of the local population segments that comprise each subspecies as an entity. The dependable sub- specific differences are largely matters of pattern, such as the narrow post- ocular light stripe and numerous tail rings that characterize viridis; the drab ground color, short dorsal blotches, and obsolescent lateral spots of lutosus; the larger and darker blotches of oreganus, and its wider postocular light stripe (compared to viridis) ; and the wide posterior tail ring of helleri. From these criteria it is evident that oreganus is intermediate between lutosus and helleri. While the northeastern specimens show some slight effect from the direct viridis contact, the oreganusMridis relationship is not as close as that of oreganus with lutosus. C. v. helleri was probably derived from oreganus, although it may represent an independent invader from" an ancestral form in southern Arizona, with a subsequent meeting and re-amalgamation in the central California area where oreganus and helleri now intergrade. The direct contact between oreganus and viridis viridis is a rather tenuous one, the only certain connection being along the Salmon River in central Idaho. The subspecies viridis is found in the Lemhi Valley, but in a phase already somewhat modified from the more typical viridis of the main body across the Continental Divide in extreme western Montana. As we pass down the Lemhi and Salmon rivers, along which rattlers occur, there is noted a gradual change from viridis to oreganus, until, eventually, quite typical oreganus is encountered in the vicinity of Florence and Riggins, Idaho County, Idaho. There may be a similar connection down the Selway River. The forest rangers of that area agree that the rattlesnakes are largely restricted to the water courses. 74 San Diego Society of Natural History The band of intergradation between oreganus and lutosus is a much broader one. It begins in western Idaho near Council and Payette, and runs westerly through eastern Oregon, past Malheur Lake to the Cascades. Thence it turns south along the crest of the Cascades to the California line, and from there southeastward following the crest of the Sierra Nevada. Along this boundary, where the mountain passes are low, there is intergradation; where the barriers are high there is none. The broadest contacts are probably to be found in western Modoc and Lassen counties, California. At any rate, the continuing subspecific relationship of oreganus with lutosus is established by the presence of intergrades at a number of localities along this boundary, such as Council, Adams County, Idaho; between Pleasant Valley and Durkee, Baker County, Owyhee, Malheur County, Camp Harney and Princeton, Harney County, and Summer Lake, Deschutes County, Oregon; and Alturas, Modoc County, California. It should be understood that these localities are not to be taken as establishing a sharp line between the subspecies; rather, they are localities from which intergrades have been noted, but they represent only a few points in a broad band of intergradation and may lie near either the oreganus or lutosus edge of this band. Whether oreganus filters through any of the mountain passes south of Lassen County, California, to intergrade with lutosus, I do not know. This could occur as far south as Mono County, for lutosus, in typical form, is common in the foothills west of Mono Lake. Rattlers are said to be absent from the vicinity of Lake Tahoe, but I have records that indicate the continued presence of a few specimens in that vicinity, and I suspect that intergradation may occur northwest of the lake, and also to the south. One small specimen in the collection of the University of California (MVZ 17022), from 6 miles north of Fernley, Washoe County, Nevada, has a pattern somewhat more like oreganus than lutosus; however, this is far into lutosus territory and, in the absence of others showing this peculiarity, is to be deemed only a non- modal lutosus. To determine the areas of intergradation between oreganus and helleri, it is necessary to restate the differences and note that there are several sequences by which one may be transformed into the other. At all ages oreganus has uniformly dark and narrow terminal tail rings, and the anterior lobe of the rattle matrix is black. In helleri the last tail ring is about twice as wide as the rings that precede it, from which one may deduce that it really comprises two rings fused together, as, indeed, is verified by a study of intergrades. In addition, this wide terminal ring is bright-yellow when the snakes are young, turning to gray, or even black, as they age. The same color change affects the anterior lobe of the rattle matrix. In general, in oreganus the final tail rings are darker than those that precede them, while in helleri the final ring, at least, is lighter. In a Sierran area of intergradation, particularly in Kern County but also to some extent in Fresno County, the intergrades, when juvenile, have dark terminal rings and rattle matrices — in color they are oreganus. But there is a tendency here toward a narrowing of the light space between the last two Klauber — New Rattlesnakes 75 tail rings, this light space being invaded by dark punctations, particularly on the dorsum. As the snakes age, this separating light ring is further narrowed and obscured by the dark adjacent rings, with the result that a snake, which as a juvenile was clearly oreganus, might, as an adult, be allocated to helleri. Specimens showing this tendency, particularly the partial obliteration of the final light interspace, became increasingly evident from Madera County southward, although they are hardly an appreciable part of the population until at least as far south as Kern County. I have examined 37 specimens from various points in Kern County and found that only 4, whether from the mountain areas or the floor of the San Joaquin Valley, were nearer helleri than oreganus. The Antelope Valley, a westerly extension of the Mohave Desert, forms a wedge in the oreganus-helleri range, for neither form is found on the floor of the desert. Since almost pure oreganus inhabits the Tehachapi Mountains to the northwest of this valley, and pure helleri the San Gabriel Mountains to the southwest, it is logical to expect intergradation at the point where these mountains meet, that is, in the vicinity of Fort Tejon and Lebec in extreme southern Kern County, and this is found to be the case. This, then, may be taken as a starting point of the boundary between the sub- species. In the Coast Range, the helleri influence is important much farther north than in the Sierra Nevada. This is especially true in the ranges bordering the Pacific — that is, the Santa Cruz and Santa Lucia mountains. A consider- able proportion of the juvenile snakes found in San Mateo, Santa Cruz, and western Santa Clara counties have yellowish tails, although this character is virtually absent on the eastern side of San Francisco Bay and the Santa Clara Valley, in Contra Costa, Alameda, and eastern Santa Clara counties. These northern specimens possessing this helleri character usually have the yellow tail bar split transversely by a white ring, with the result that they would key out as oreganus when adult, after the yellow has turned to gray or black, and would, in fact, as juveniles also, if the width of the last ring were given precedence over the color of the last two. Tabulating the available material (48 specimens), I find that, while some specimens from San Mateo, Santa Cruz, and Santa Clara counties will key out as helleri, especially the juveniles, the majority are oreganus, and I therefore assign these areas to the latter subspecies. From Monterey and San Luis Obispo counties I have had an insufficient number of specimens (18) to draw any final conclusions. Probably these represent areas of a fairly even balance between the forms. The specimens from along the edge of the San Joaquin Valley are usually oreganus, and those that I have seen from Santa Barbara County were all helleri. Tentatively I should fix the approxi- mate boundary between the two subspecies as a line from Lebec, Kern County, via the Carrizo Plain to Shandon, San Luis Obispo County, and thence to the northwestern comer of that county. By such a line I do not, of course, suggest that this really separates the subspecies; rather, it is my intention to indicate that south of this line helleri will predominate, and north, oreganus. 76 San Diego Society of Natural History One might presume, from the stress that I put on the pattern of the tail, that this is a minor difference upon which to justify the segregation of two subspecies. But other differences of pattern are also apparent, although less easily defined and therefore less useful in a key. And, although the tail-band criterion is troublesome in intermediate territory, as is always the case with characters defining intergrading forms, it will correctly segregate practically every specimen from the areas where the subspecies are typical, that is, oreganus in northern California, Oregon, Washington, British Col- umbia, and western Idaho; and helleri in southern California and Baja Cali- fornia. Range. — C. r. oreganus ranges along, or close to, the Pacific Coast, from British Columbia to south-central California. It is the only rattlesnake oc- curring in British Columbia or Washington; and it is the only rattler found throughout the rest of its range — neglecting intergradation with other virid'is subspecies — although there may be a narrow overlap with C. m'ltchelli'i stephensi through the Sierra Nevada passes along the southern border between Tulare and Inyo counties, and a similar fringe overlap with C. s. scutulatus and C. c. cerastes in the desert foothills southeast of the Tehachapi Range in Kern County, California. In British Columbia, oreganus occurs from the line Lillooet-Hat Creek- Kamloops-Shuswap Lake, south to the U. S. border, with an eastward ex- tension just north of the border to Christina Lake and possibly Waneta. I have had no reports of specimens in the vicinity of Yale or Hope, or from these points south or southeast to the border. In Washington, oreganus occurs only to the east of the Cascade Range. Along the Columbia River it has penetrated as far west as Carson, Skamania County. East of the Cascades it seems to range throughout the state; how- ever, it is absent from the extreme northeast; at least I have no records from Pend Oreille County or eastern Spokane County. In Idaho, oreganus ranges along the western border from the southern end of Lake Coeur d'Alene south to near Council and Weiser, in the vicinity of which there is intergradation with lutosus. Also, from this westerly fringe there are extensions eastward up the Clearwater and Salmon rivers and their tributaries toward narrow areas of intergradation with viridis virid'is. In Oregon, oreganus occupies the northern and western sections of the state; the southeast, with the limits that I have discussed under intergrada- tion, being inhabited by lutosus. In the northern part of the state, oreganus is plentiful on the lower east slopes of the Cascades, and thence eastward across the state, but to the west it is not so prevalent, although scattered through the Willamette Valley, at least from Cottage Grove north to Amity. It is apparently absent between the Willamette Valley and the coast. South of Cottage Grove it is increasingly plentiful in the mountains that lie between the crest of the Cascades and the coast, the unoccupied coastal area gradually narrowing until it almost disappears at the California line. In California, oreganus occurs everywhere west of the Sierra Nevada Klauber — New Rattlesnakes 77 and its area of intergradation with lutosus through the passes of that range, as discussed elsewhere, and north of its area of intergradation with helleri. It is seldom found in the coastal redwood belt from Sonoma County north- ward through Mendocino, Humboldt, and Del Norte counties; yet an oc- casional stray, possibly carried seaward by a river flood, is picked up in this section, sometimes within 10 miles or less of the coast. The subspecies, how- ever, seems unable to establish itself permanendy in the coastal redwood belt. There are large areas in the state, especially in the Central Valley and the coastal valleys, and about centers of population, where rattlers have now become exterminated; but these were once occupied, as their continued presence in adjacent parks and wildlife refuges demonstrates. This subspecies occurs on Morro Rock, a small rocky island off the coast of San Luis Obispo County, now connected by an artificial causeway with the mainland. Such specimens as I have seen from the island were ap- parently stunted. Crotalus viridis helleri Meek Southern Pacific Rattlesnake* Plate 4, fig. 2. 1859. Crotalus lecontei (part) Hallowell (not of Hallowell, 1852). Pac. RR. Surv. (Williamson route), vol. 10, pt. 4, no. 1, p. 18. 1859. Crotalus luajer Cope, Proc. Acad. Nat. Sci. Phila., vol. 11, p. 337. 1863. Crotalus adamanteus var. lucijer Jan, Elenco. Sist. degli Ofidi, p. 124. 1870. Crotalus Hallowellt Cooper, Proc. Cal. Acad. Sci., vol. 4, part 2, pp. 64, 68 (nornen nudum). 1883. Crotalus conjluentus var. lucijer (part) Cope, Proc. Acad. Nat. Sci. Phila., vol. 35, p. 11. 1883. Crotalus oregonus var. lucijer (part), Garman, Mem. Mus. Comp. Zobl., vol. 8, no. 3, p. 173. 1898. Crotalus oregonus (part) Van Denburgh, Proc. Amer. Philos. Soc, vol. 37, no. 157, p. 141. 1905. Crotalus helleri Meek, Field Col. Mus., Zobl. Ser., vol. 7, no. 1, pub. 104, p. 17. Type locality: San Jose [Lat. 31° N.}, Baja Cali- fornia. Type specimen: CNHM 1272; paratypes: CNHM 1727 1917. Crotalus oreganus (part) Grinnell and Camp, Univ. Calif. Pubs, in Zobl., vol. 17, no. 10, p. 194. 1929. Crotalus conjluentus oreganus (part) do Amaral, Bull. Antivenin Inst. Amer., vol. 2, no. 4, p. 92. 1936. Crotalus viridis oreganus (part) Klauber, Trans. San Diego Soc. Nat. Hist., vol. 8, no. 20, p. 191. Diagnosis. — This subspecies usually has diamonds on the back, instead of hexagons as in oreganus. The last dark tail ring in the adults is more than * Has such alternative local names as black diamond-back, gray diamond-back, Pacific rattler, San Diegan rattler, timber rattler, black rattler, and mountain rattler. Young specimens are often referred to as sidewinders. 78 San Diego Society of Natural History iy2 times as wide as the preceding dark ring, while in oreganus the rings are equal. In the juveniles the end of the tail in helleri has a wide yellow ring, gradually turning to gray or black as the snakes age; in oreganus the terminal cross-band is dark from birth and is no wider than the rings that precede it on the tail. The button matrix is usually yellow in juvenile helleri and black in oreganus. C. V. helleri differs from cerberus in having a single row of light scales completely bordering the dorsal diamonds, whereas in cerberus the dorsal blotches laterally merge directly into the ground color. A majority of speci- mens of helleri have single loreals, while cerberus has two. Most helleri specimens have no scales between the prenasal and the first supralabial at the rostral; in cerberus such scales are generally present, at least at the rostral. The differences from the other viridis subspecies are similar to those cited under oreganus. Nomenclatorial and Systematic Problems. — If the subspecies of Crotalus viridis found in southern California and Baja Cilifornia is to be separated from the snakes of the northwest and Arizona, helleri Meek, 1905, is the oldest available name, although Meek described it as new only because of his unfamiliarity with the snakes of the San Diegan region; for those of the type locality differ in no important particular from the San Diegan snakes, which were well known (as C. lucifer) in 1905. It is to be regretted that the long-used name liicijer cannot be revived for the southern race, but al- though Baird and Girard did not originally specify a type in 1852 (the type locality was given as Oregon and California), and specimens from southern California may have been available to them, they did set up a type, from Oregon, in 1853 (p. 8) . C. lecontei of Hallowell, 1859, although in part from southern California, is not available, as the original description (Hallo- well, 1852) was based on a specimen of C. v. viridis from Cross Timbers, in what is now Oklahoma (Stejneger, in do Amaral, 1929, p. 87). Crotalus hallowelli of Cooper, 1868, cannot with certainty be assigned to southern California, and besides is a nomen nudum. This I have cited under the syn- onymy of oreganus, since he mentions its presence in the Sierras. Cooper's second mention of hallowelli (1870, pp. 64, 68) leaves little doubt that he had in mind the Pacific Coast subspecies of viridis, since only this snake is present in both the Sierras and the San Diegan region. As the first mention (1868) referred to the Sierran range, it is probable that the name, if valid, would have to be assigned to the northern race, for which it would be preceded by both oreganus and lucifer. But, in any case, it is a nomen nudum for lack of a type specimen, type locality, or any description. This is to be regretted, as Hallowell's distinguished work on the herpetology of the West might well be recognized by associating his name with this snake. Cooper, in his 1870 paper, by the use of the initials Cp. after the name Crotalus Hallowelli, in- dicated that he considered himself the describer. He must have known that neither the mention of the snake in Cronise's book nor in his 1870 paper constituted description, from which one may infer that he actually presented the description in a paper that remained unpublished or that has been over- looked. Klauber — New Rattlesnakes 79 Redescription of the Type Specimen. — The type specimen is CNHM (formerly Field Museum) 1272, from San Jose (Lat. 31° N.), Baja Cali- fornia. Comparing it with specimens from Mehng's Ranch at San Jose, I should say, from its dark color, that it was probably collected on the moun- tain slope to the east. It is an adult male, with a length over-all of 957 mm., and a tail length of 60 mm. The scale rows at mid-body are 25, vent- rals 173, and subcaudals 25. All supra- and infralabials number 15. The rostral is higher than wide; it contacts 8 scales. The internasals number 4 and the canthals 2 — 3. The minimum scales between the supraoculars are 5 + 5. There are no small scales touching the rostral between the first supra- labials and the prenasals. The junction of the postcanthals and the loreals pre- vents a contact between the postnasals and the upper preoculars, which are undivided. There are 2 to 3 scale rows between the supralabials and the orbit. The head is almost black, both the supraocular and postocular light marks being obsolete. There are 32 body diamonds, and 4 rings on the tail, the last being wide and black. The body blotches are quite black anteriorly and somewhat brownish toward the tail; they are outlined by straw-colored single rows of scales. There are 5 rattles, a broken string. Material. — The description which follows is based on 892 mainland specimens, of which 645 were from San Diego County. There were 473 males and 399 females, the rest being heads, skins, or otherwise indeterminate. There have been 22 specimens available from Santa Catalina Island. I believe I have seen well over 3000 rattlesnakes of this subspecies alive. Description of Subspecies. — This is a snake of the usual viridis shape. Like other subspecies of viridis, it differs from all other rattlesnakes in usually having more than 2 internasals; in 893 specimens, 81, or 9.1 per cent, had 2 internasals or less. The longest specimen I have measured was 1371 mm. It is thought that fully adult males are usually about 1200 mm. in length. At birth the young specimens normally measure about 275 mm., although occasionally they are as short as 225 mm. Most specimens of this subspecies have 25 scale rows at mid-body, the distribution in per cent being 23 (1), ^24(1), 25(71), 26(5), 27(21). 28(*), 29(1). The mean is 25.43 and the coefficient of variation 3.6 per cent. The ventrals in the males vary from 162 to 184, but all but a very few specimens fall between 166 and 181. The interquartile range is 171.2 to 176.1, the mean 173.67, and the coefficient of variation 2.1 per cent. The ventrals in the females vary from 166 to 189, although most specimens fall between 170 and 187. The interquartile range is 175.7 to 180.7, the mean 178.23, and the coefficient of variation 2.1 per cent. The subcaudals in the males vary from 19 to 29, all but a few specimens falling between 21 and 27; the Present, but less than 1 per cent. 80 San Diego Society of Natural History interquartile range is 23.0 to 25.1, the mean 24.08, and the coefficient of variation 6.3 per cent. The females range from 15 to 25, with all but a few- specimens falling within the range 16 to 23; the interquartile range is 18.3 to 20.5, the mean 19.39 and the coefficient of variation 8.4 per cent. The rostral is higher than wide. There are from 1 to 8 internasals; most specimens have 4 (73 per cent), 3 (15 per cent), or 2 (9 per cent). The canthals usually number 2 on each side, but are often 3, or rarely 4. The scales on the snout, anterior to the supraoculars, range from 7 to 45, although most specimens fall between 13 and 33. The mean is 23.4 and the coefficient of variation 22.1 per cent. These scales cannot be counted with particular accuracy in this subspecies since there is no well-defined line at the anterior edge of the supraoculars. The minimum scales bridging the gap between the supraoculars vary from 1 to 9, most specimens having 4 (34 per cent), 5 (28 per cent), 6 (17 per cent), or 3 (12 per cent). The loreals most often number 1 (82 per cent), or 2 (18 per cent). An occasional specimen may have 3. In this subspecies, in 69 per cent of the specimens, the prenasal touches the first supralabial at their common point of contact with the rostral; however, in 17 per cent of the specimens there is an extra scale or two at this tripartite junction, and in the other 14 per cent a complete row of small scales separates the prenasal from the initial supralabial. The supralabials range from 12 to 18; the interquartile range is from 14.8 to 16.1, the mean 15.14, and the coefficient of variation 6.3 per cent. The most prevalent counts are 15 (46 per cent), 16 (26 per cent), or 14 (20 per cent). The infralabials range from 13 to 20, with an interquartile range of 15.3 to 16.6, a mean of 15.96, and a coefficient of variation of 6.1 per cent. The most frequent counts are 16 (41 per cent), 15 (25 per cent), or 17 (24 per cent). The first infralabials are undivided, and neither inter- genials nor submentals are usually present. The body blotches vary from 27 to 43, nearly all specimens falling between 29 and 41. The interquartile range is 33.5 to 36.8, the mean 35.14, and the coefficient of variation 7.0 per cent. The tail rings in the males range from 3 to 8. with a mean of 4.52, and in the females from 2 to 6, with a mean of 3.44. In the young of this subspecies the colors are sharply contrasting. The head is dark-brown above, with a light mark across the supraoculars and the frontal area. There is often a second thinner line across the head at the posterior ends of the central internasals, and a light longitudinal line joining the two cross-lines. The rostral is usually edged with light, but is otherwise dark-brown. On the side of the head a wide light stripe passes backward and downward, just touching the lower edge of the orbit; it widens at the suprala- bials, so that these are light from below the center of the eye backward almost to the commissure. Bordering this light streak above, there is a wide dark streak from the eye to the commissure, and this in turn is bordered above by a light streak two scales wide, which starts at the posterior outer edge of the supraocular, and is directed backward and downward to the angle of the jaw. On the underside, the mental and first supralabials are dark, and so Klauber — New Rattlesnakes 81 are the genials along their Hne of contact. The posterior infralabials may all be spotted, or some may be clear. There is often a collection of spots marking the gulars, anterior to the first ventrals. All dark areas on the head are usually chocolate-brown; the light are cream or buff. As the snakes age, the light head marks gradually disappear, until, in the adults, only the preocular light streak is left and even this is often obscured by punctations. Thus the head becomes uniform dark-brown or black above, and quite dark on the sides. The markings of the lower jaw are little changed with age. A somewhat similar change takes place in the body marks, the juveniles being clearly marked, while the adults are duller and darker. In the young the pattern comprises a series of dark-brown blotches, about 9 or 11 scale rows wide, separated by interspaces of buff or light-gray extend- ing over iVi to 2 scales end-to-end. The blotches are rather irregular, but are usually diamonds or hexagons. They are slightly lighter in the centers than on the edges, and are bordered with light scales, laterally as well as dorsally. On the sides, between blotches, there are brown areas which may be considered a first secondary series, or may be viewed as patches of ground color. Below these there are one or two other series of darker-brown and more evenly edged secondaries. The lowest touch the outer edges of the ventrals, which are mottled or punctated with dark, especially toward the tail, on a buff or yellow background. Toward the tail the dorsal blotches tend to merge with the secondaries to become dark cross-bands. However, the last tail band, having about twice the width of those that precede it, is bright-yellow in color, and so is the matrix of the button. As the snakes age the contrast between the light and dark areas is less accentuated, only the light scales that border the dorsal blotches remaining fairly clear, although even these are often darkened at the anterior end of each scale. These light blotch-borders remain in evidence, not only middor- sally, but laterally as well; in this respect, this subspecies differs from cerbenis. The yellow terminal tail ring of the juveniles, so conspicuously different from the rest of the coloration, gradually darkens, beginning at about the time the second rattle is acquired, until, when the snake is fully grown, it is dark-gray or black, and thus does not differ from the preceding rings, except that it retains its double width. In the adults, in some areas, a faint light ring may suggest the beginning of a division of this terminal ring, thus suggesting the relationship with oreganus. The blotches in the adults are often black, compared with the dark-brown of the juveniles; and the reduction of the light areas between blotches makes the adults generally darker. This is accentuated in mountain specimens. This description is based primarily on the snakes of the San Diegan area. Intrasubspecific Trends. — C. v. helleri has a smaller and more uniform range, ecologically speaking, than either oregantis or cerberus; and, as might 82 San Diego Society of Natural History be expected, is more homogeneous. There is a sHght tendency toward a higher number of ventral plates and a greater subdivision of the head scales, as one proceeds northward and westward. A similar moderate increase is to be observed in the number of body blotches. Contrasting with those from the lowlands, specimens from the mountains are generally darker, often almost black, thus justifying the popular name, black diamond. The few individuals available from the desert foothills are decidedly light-colored. One of the two southernmost specimens — that from Playa Maria Bay, Baja California — has some peculiarities of pattern, particularly in the head marks and the way in which the tail rings are interrupted laterally in a manner reminiscent of exsul. It is conspicuously light in color. The other, from northwest of Bahia Angeles on the Gulf coast, is peculiar in having long, narrow blotches on the neck. Intersiibspecific Relationships. — C. r. helleri contacts no subspecies of viridis other than oreganus, intergradation with which has already been dis- cussed. Although helleri no longer has a physical contact with cerberus, its relationship with that subspecies seems as close as with oreganus, since occa- sional specimens will be found in the territory of either that are quite like the mode in the other. In fact, it is to be noted that helleri and cerberus are alike in the character that I have used to separate helleri from oreganus, that is, the width and color of the posterior tail rings. These and other similarities are evidence that there must at one time have been a direct contact of the two populations across the area now occupied by the Colorado and Yuma deserts. C. v. helleri is more closely related to oreganus and cerberus than to any other viridis subspecies. Range. — C. v. helleri is found only in southern California and northern Baja California. The area of intergradation with oreganus has already been discussed (p. 74). C. v. helleri is the sole subspecies of the western rattle- snake, Crotalus viridis, occurring in southern California. Neglecting areas of intergradation with oreganus, it is found from the north line of Santa Barbara, Ventura, and Los Angeles counties, southward through southwestern San Bernardino, Orange, western Riverside, San Diego, and extreme southwestern Imperial counties, into Baja California, where it occurs as far south as Lati- tude 29 N. In this area, except where it has been exterminated by the encroachment of cities and intensive agricultural developments, it is quite common from the coast across the mountains. In the desert foothills it becomes increasingly rare as the descent is made to the desert itself, and in the desert it is absent. Altitudinally it ranges from sea level to the peaks of the highest mountains. It has been collected a few feet below the peak of San Jacinto (10,805 ft.), and as only one mountain (San Gorgonio, 11,485 ft.) within the helleri range is higher, it may be assumed that altitude is nowhere a serious impediment to its occupancy. In Baja California it has been taken at a large number of points from the coast to the crest of the Sierra Juarez and Sierra San Pedro Martir, and as far south as Socorro (Lat. 33° 20' N.). No specimens are yet available from the desert slopes of these mountains, but no doubt it occurs there spo- Klauber — New Rattlesnakes 8? radically, as it does in California. South of Socorro two specimens have been taken; one at Playa Maria Bay, some 120 miles below Socorro, on the Pacific side; the other, 10 miles northwest of Bahia Angeles on the Gulf of Cali- fornia side, a rather surprising find in such arid surroundings, for it was taken by Lewis W. Walker in desert mountains, amid an elephant-tree asso- ciation. C. V. helleri occurs on Santa Catalina Island, Los Angeles County. Crotalus viridis cerberus (Coues) Arizona Black Rattlesnake* Plate 5, fig. 1. 1866. Caudisona confluenta var. confluenta (part) Cope, Proc. Acad. Nat. Sci. Phila., vol. 18, p. 307. 1866. Caudisona luctfer Cope, Proc. Acad. Nat. Sci. Phila., vol. 18. p. 307. 1875. Crotalus luctfer (part) Yarrow, Surv. W. of 100th Mer. [Wheeler], p. 529. 1875. Crotalus confluentus (part) Yarrow, Surv. W. of 100th Mer. [Wheeler], p. 530. 1875. Caudisona confluenta (part) Coues, Surv. W. of 100th Mer. [Wheeler], p. 604. 1875. Caudisona lucifer var. cerberus Coues, Surv. W. of 100th Mer. [Wheeler], p. 606. Type locality: San Francisco Mountains, Coconino County, Arizona. Type specimens: Field numbers 509 (=ANSP 7085?) and 511 (=ANSP 7088). 1883. Crotalus luctfer (part) Yarrow, Bull. U. S. Nat. Mus., no. 24, p. 76. 1883. Crotalus oregonus var. cerberus Garman, Mem. Mus. Comp. Zool., vol. 8, no. 3, p. 173. 1913. Crotalus oregonus Van Denburgh and Slevin, Proc. Cal. Acad. Sci., ser. 4, vol. 3, p. 428. 1917. Crotalus oreganus (part) Grinnell and Camp, Univ. Calif. Pubs, in Zo6l., vol. 17, no. 10, p. 194. 1930. Crotalus confluentus oreganus (part) Klauber, Trans. San Diego Soc. Nat. Hist., vol. 6, no. 3, p. BOf 1936. Crotalus viridis oreganus (part) Klauber, Trans. San Diego Soc. Nat. Hist., vol. 8, no. 20, p. 191. Diagnosis. — Cerberus is a subspecies characterized by its dark color and a marked subdivision of the scales of the snout. It differs from the newly delimited oreganus and helleri, with which it was formerly merged under the name oreganus, in usually having paired loreals (Klauber 1930b, plate 1, fig. 2), while they have single loreals (figs. 1, 3, or 4). Also, most specimens of cerberus have one or more scales at the rostral between the anterior nasal and the first supralabial {loc. cit., fig. 6 or 7), while such * Has the alternative local names of brown rattlesnake, black rattlesnake, black diamond, mountain diamond-back. t Do Amarai (1929, p. 95) considered the Arizona specimens to be confluentus {^viridis) x oreganus intergrades. 84 San Diego Society of Natural History scales are usually absent in helleri and oreganus (fig. 5). C. v. cerberus has a wider terminal dark tail ring than oreganus. The dorsal body blotches are without lateral light borders, such as are generally present in helleri. Nomenclatorial and Systematic Problems. — While Coues did not describe his new subspecies, cerberus, except to call attention to its black head, he did specify type specimens, so there is no question as to the validity of his name, if segregation from the California forms be justified, as I now believe to be the case. There is some uncertainty with respect to these types of cerberus. Coues had at least six specimens from the San Francisco Mountains and Fort Whipple. Four he assigned to Caudisona liirijer. At least one of these must have been molossus, judging by his description; of the others, two, ANSP 7086 (original number 510) and ANSP 7087 (original number 572?), are cerberus, but were not assigned by Coues to his new subspecies, as they were browner than the types. But it is now known that there is some color variation in the snakes of the type locality. Coues mentioned only two types by number, although stating that he had others from the type locality. Of these two cotypes. No. 511 is now ANSP 7088, and 509 is believed to be ANSP 7085. Since Coues did not describe his types, no comparisons with his descriptions can be made. The following descriptions are of these two, imperfect as they are. Redescription of the Type Specimens. — ANSP 7085 is a skin in rather poor condition. It is a male with a length of about 850 mm. The scale rows at mid-body are 27, ventrals 167, subcaudals 26. The supralabials number 17, the internasals 6; there are 14 scales in contact with the rostral. Although the prenasals touch the rostral, they are separated from the first supralabials, as is characteristic of this subspecies. There are two loreals on each side. The intersupraoculars number 7+9. There are 5 tail rings, all black. The body blotches cannot be counted accurately. They are quite black, and, in fact, the entire dorsal surface is almost black, except for a few straw-colored scales between the blotches, mid- dorsally. There are six rattles, an incomplete string; the width of the proximal rattle is 15.5 mm. Of ANSP 7088 only the head and neck remain. The supralabials are 16 — 15 and infralabials 15 — 15. There are 9 scales around the rostral, which is higher than wide. There are 5 internasals, which come well down the sides but do not completely separate the rostral from either prenasal. The prenasals are not separated from the anterior supralabials, nor are there extra scales at the rostral-prenasal-supralabial junction. The canthals arc 4 — 4; there are about 45 scales before the supraoculars, and 3+5 scales between. The loreals are paired. The head is brown, the usual marks being obsolete. The initial dorsal blotches are brown, with dark punctations. Klauber — New Rattlesnakes 85 Material. — Of this subspecies 98 specimens have been available for study, one from New Mexico, the rest from Arizona. Several additional specimens with questionable locality data have been omitted from the statistics of the description that follows. Some 50 snakes have been seen alive. Description of Subspecies. — This is a snake of average viridis propor- tions. Like others of the viridis subspecies, it differs from all other rattle- snakes in usually having more than 2 internasals; only 2 per cent of this subspecies have 2 internasals. The longest specimen of the series I have studied is a male measuring 1032 mm.; the shortest 287 mm. I have a brood that is presumably of this subspecies, but not certainly, as there were several mothers in the cage when they were born. Of this brood of 13, the shortest is 265 mm., the longest 288 mm., and the mean 273 mm. Most specimens of this subspecies have 25 scale rows at mid-body, the distribution in per cent (in parentheses) being 23(4), 24(6), 25(76), 26(2), 27(12). The mean scale rows are 25.12, the coefficient of variation is 3.4 per cent. The ventrals in the males vary from 161 to 180, although most specimens fall between 166 and 176. The interquartile range is 168.6 to 173.9, the mean 171.23, and the coefficient of variation 2.3 per cent. The ventrals in the females vary from 164 to 184, although most specimens fall between 171 and 183. The interquartile range is 172.1 to 178.7, the mean 175.41, and the coefficient of variation 2.8 per cent. The subcaudals in the males vary from 20 to 26, interquartile range 22.6 to 24.5, mean 23.58, coefficient of variation 5.9 per cent; the females range from 16 to 24, interquartile range 17.7 to 20.5, mean 19.07, coefficient of variation 10.9 per cent. The rostral is higher than wide. There are from 2 to 8 internasals; most specimens have 4 (59 per cent), 3 (20 per cent), or 5 (11 per cent). Only two per cent have 2, and would thus fail to key out as a viridis subspecies. The canthals are usually 3 — 3, but are often 2 — 2 or 4 — 4. The scales on the snout, anterior to the supraoculars, range from about 20 to 50, the mean being 35, and the coefficient of variation 20.4 per cent. These scales cannot be counted with accuracy in this subspecies since there is no well-defined line at the anterior edge of the supraoculars, thus differing from C. sciitulatus and certain other forms of rattlesnakes that have definite boundaries between the prefrontal and frontal areas. The minimum scales bridging the gap be- tween the supraoculars vary from 2 to 9, most specimens having 6 (26 per cent), 5 (24 per cent), or 7 (21 per cent). The loreals most often number 2 (65 per cent) , followed by 1 (23 per cent) , or 3 (9 per cent) ; the extreme variation is 1 to 6. While the prenasals usually contact the rostral, this contact is prevented by rows of small scales in 17 per cent of the specimens. The percentage having this mitchellii-like characteristic is higher than in any other viridis subspecies. Cerberus is also exceptional in * This characteristic is evident in some of the diamond-backs, notably the Cedros Island species, C. exsul, and to a less extent in C. r. ruber, and C. r. lucasensis. 86 San Diego Society of Natural History another scale arrangement, the presence of a small row of scales interposed between the prenasals and the first supralabials.* This row is complete in 72 per cent of the counts, and in an additional 8 per cent there are one or more extra scales at the junction of the rostral, prenasal, and first supra- labials, thus partly interrupting the contact. The supralabials range from 13 to 18; the interquartile range is from 14.8 to 16.1, the mean 15.47, and the coefficient of variation 6.2 per cent. The most prevalent counts are 15 (44 per cent), 16 (33 per cent), or 17 (10 per cent). The infralabials range from 13 to 19, with an interquartile range of 15.3 to 16.8, a mean of 16.03, and a coefficient of variation of 6.8 per cent. The most frequent counts are 16 (38 per cent), 15 (25 per cent), or 17 (21 per cent). The first infralabials are undivided, and there are neither intergenials nor submentals. The body blotches vary from 25 to 46, although nearly all specimens fall between 29 and 42. The interquartile range is 32.9 to 37.9, the mean 35.39, and the coefficient of variation 10.3 per cent. The tail rings of the males range from 3 to 9; most fall between 5 and 8, with a mean of 6.05. The females range from 3 to 7, although only one specimen has the latter number; the mean is 4.63. The head marks in the adults of cerberus are virtually obsolete, the heads being usually uniform, or mottled, brown or black. The preoculars are often lighter and occasionally a preocular light streak, to the angle of the mouth, may be faintly seen. A postocular light stripe, while ordinarily obsolete, is sometimes present. These marks are usually faintly evident in the specimens from the Santa Catalina and Rincon mountains, in gray against a black background. The infralabials, mental, and genials are punc- tated with brown; the lower jaw is otherwise clear buff, although there may be aggregations of brown dots scattered about. In the juveniles the head marks are not unlike those of the coastal specimens. There are preocular and postocular light streaks sloping down- ward and backward, separated by a dark ocular stripe. The postocular mark is two scales wide. A light supraocular cross-dash is also present. The body pattern in the adults consists of a middorsal series of round, elliptical, or hexagonal blotches. These may be brown, red-brown, dark- brown, or black, the edges usually somewhat darker than the centers. The blotches are from 9 to 15 scale rows wide, most often 11 or 13. They are close together, being separated by one to two transverse lines of lighter scales, which range in color from yellow to light-brown. On the sides there are two rows of subsidiary blotches, smaller and less conspicuous than the dorsals, for they are not sharply differentiated from the ground color, which is usually brown, or dark-brown, but may be gray. The light scales which separate the dorsal blotches are clearest and most evident middorsally; later- ally these light bordering scales become increasingly clouded, so that, in adult specimens, the blotches on the sides are without the light bordering row, and thus merge direcdy into the ground color. This is a distinguishing characteristic from helleri, in which the light lateral borders of the blotches Klauber — New Rattlesnakes 87 are usually evident. In fully adult specimens from the Catalina and Rincon mountains the light separating lines may be reduced to a half-dozen light scales middorsally; these are often bright-yellow in life. The tail is crossed by brown or dark-brown bars, imperfectly set off by the ill-defined lighter areas between. The last bar is usually about twice as wide as the others, but this may not be true, especially in the northwest area of the range; here the last ring may be no wider, but may be distinctly darker than the others. The lower surface is buff or yellow, heavily blotched or mottled with gray, brown, or black. This mottling is accentuated posteriorly. The young specimens are lighter and brighter, with the pattern better defined. The light edges of the dorsal blotches are evident laterally. The secondary spots, especially the lower row, are more contrasting with the ground color. The final tail ring is yellow, or gray with a tinge of yellow, much like the coastal helleri. Intrasubspecific Trends. — There is a considerable pattern and color var- iation in Cerberus, particularly in the montane-island groups in the southern part of the range, in the Santa Catalinas, Rincons, and Pinalenos. Probably the most distinctive are the black rattlers of the Santa Catalina Mountains. When adult these are handsome jet-black snakes, the dorsal blotches being distinguishable only through the presence of small groups of yellow middorsal scales, which represent the obsolescent remnants of the usual light interspaces. These snakes, and, in fact, nearly all cerberus, change considerably on preservation, the interspaces and the lateral areas becoming somewhat lighter, so that both the main dorsal series and the auxiliaries on the sides become more evident than in life.* I have found that this same color change may be effected to a lesser extent, by increasing the temperature of a live snake; I have seen a specimen under varying conditions, with the dorsal blotches at times much more conspicuous than at others. In some areas the snakes are dark-brown, rather than black, with large round or oval dorsal blotches, set close together. The sides are lighter-brown. * The following color notes were made Aug. 2, 1930, in the course of preserving 4 Cerberus, 2 from Gleed and the others, somewhat larger, from Iron Springs and Hillside, Yavapai County, Arizona: "They vary from quite black with no evident dorsal blotches and only a few light-tipped scales — in the larger specimens — to gray- brown in ground color with dark-brown blotches outlined in punctated buff. The head is almost unicolor, except that the" side stripes can be faintly seen in the smaller specimens. After their deaths on the following morning — by drowning in alcohol — the color changes are almost beyond belief, especially in the two larger specimens. The black dorsal blotches have become brown, interspersed with some gray. The lateral ground color has become light-gray. The blotches are now very clearly outlined, whereas before they could be distinguished only when viewed with the light at a par- ticular angle. The yellow scale tips along the dorsum rem.ain unchanged as an evidence of the only light areas that were once apparent, for all else was black before. The ventrums have also become lighter." From these notes it is evident that preserved specimens of this subspecies do not afford an accurate idea of the coloration in life. It explains why cerberus is so generally called the "black rattlesnake" in Arizona, whereas preserved specimens are often blotched with brown or gray. 88 San Diego Society of Natural History This is true of most specimens of the Skull Valley-Prescott area, although some from here are black. I have seen an occasional specimen with bright- red (not red-brown) on the lower lateral scale rows. This color disappeared completely in preservation. Toward the northwest the snakes are often grayish. In the San Fran- cisco Peaks they are black or brown. I have too few specimens to demonstrate any important geographical trends in ventrals, subcaudals, or labials. The Santa Catalina and Rincon mountain snakes have head scales less divided than in those from the center of the range; that is, they have fewer paired loreals, fewer scales in the inter- supraocular area, and less often have the prenasals separated, or partly sep- arated, from the first supralabials. All of these divisions are characteristic of Cerberus as compared with helleri. The Santa Catalina snakes, in spite of their dark color, retain evidences of the pre- and postocular light streaks to a greater extent than specimens from the central area. The snakes of the northwestern part of the range quite often have terminal tail rings no wider than the anterior, possibly an effect of a lutosus or niintius infusion. However, the light separating rings are not greatly different in shade from the rings themselves; there is not the sharp contrast evident in most oreganus. Intersubspecific Relationships. — It is uncertain whether, today, cerberus intergrades with any other viridis subspecies; yet its relationship with helleri, which is a member of an uninterrupted viridis chain, is so close and obvious that it should not be considered a separate species, regardless of its present isolation. There is a possibility of intergradation between nuntius, as found at Valle, Coconino County, and cerberus at Crookton and deed, Yavapai County, Arizona (see map, Klauber, 1935). A number of the latter show some nuntius tendencies, particularly in head scales and narrow posterior tail rings. These localities are only about 35 miles apart, and the intervening territory offers no barrier to rattlesnakes. I should expect intergradation to be eventually demonstrated here, were it not for the fact that there may be an overlap without intergradation at another point. This is in Medicine Valley, just northeast of the San Francisco Peaks, where nuntius occurs, while cerberus has been taken at a slightly higher elevation along the southern border of the valley, and these specimens do not show intergradient tend- encies. Further collecting in this vicinity will be necessary to settle these problems of overlap vs. intergradation.* Of the other subspecies, hitherto considered oreganus, cerberus seems to be nearest the southern specimens of helleri, particularly to the darker indi- viduals found at the higher elevations of the Peninsula ranges of southern California and Baja California, including the Sierra Juarez and Sierra San Pedro Martir. Surprisingly, the snakes of the Santa Catalina and Rincon * In the previous discussion of this problem, loc. cit. p. 86, read supralabials instead of supraoculars, near the end of the second paragraph. Klauber — New Rattlesnakes 89 mountains of Pima County, Arizona, although more distant from CaUfornia than those from near Prescott, are more like helleri in color and head scales. Range and Locality Records. — C. v. cerberus is found only in Arizona and extreme western New Mexico, from the Hualpai Mountains and Cotton- wood Cliffs, southeastward to the Santa Catalina, Rincon, Pinaleno, and Blue mountains of southeastern Arizona, and Steeple Rock, New Mexico. The range is not continuous, for this snake is absent from intervening low- lying arid areas, it being essentially a resident of the Transition Zone. Al- though I had previously reported cerberus in Sonora, Mexico (1930a, p. 131), this record is to be judged highly questionable, and is to be suppressed unless verified by additional specimens. The detailed locality records* available are as follows — ARIZONA: Apache County — White Mountains, 5 mi. s. of Greer, Apache Forest (may be Greenlee County) ; Coconino County — Tule Basin (6 mi. sw. of Coleman Lake), Coon Hill (9 mi. s. of Williams), Sunshine Spring (10 mi. s. of Williams), Round Mountain (14 mi. s. of Williams), 2 mi. n. of JD Dam, San Francisco Mountains (type locality) , Medicine Valley (ne. side of San Francisco Peaks), Oak Creek, Oak Creek Lodge, Stoneman Lake (6500 ft.), Apache Maid Mountain at 6200 ft., Long Valley, Canyon Creek (26 mi. n. of Young on Holbrook road — may be Navajo County) ; Gtla County — 2 mi. ne. of Pine, near Payson (elev. 6000 ft.), Tonto Forest (north of Roosevelt Dam), Workman Creek, Workman Creek Trail at 6500 ft. (Mt. Aztec), Sierra Ancha, Salt River (15 mi. n. of McMillanville), bet. Roose- velt Dam and Globe, 6 mi. w. of Globe on US 180; Graham County — Pinalerio Mountains near Fort Grant, Mount Graham, Shannon Creek (2 mi. ne. of Arcadia Ranger Station, Pinalerio Mountains); Greenlee County — near Hannagan, KP Creek (White Mountains, at 8000 and 8200 ft.), north slope of Rose Peak (Blue Mountains), Eagle Creek at 5000 ft. (20 mi. nw. of Clifton); Maricopa County — Cave Creek; Mohave County — Hualpai Mountains, 10 mi. e. of Hackberry, near Peach Springs; Pima County— Mount Lemmon at least to 9000 ft., (Santa Catalina Mountains), Kellogg Peak (Santa Catalina Mountains), near Spud Rock Ranger Station (Rincon Mountains at 5500, 7000, and 7400 ft.). Manning Camp (Rincon Mountains at 8000 ft.); Pinal County — Superior, 2 mi. n. of Pinal Ranch (midway be- tween Superior and Miami), 6 mi. se. of Oracle (n. slope Santa Catalina Mountains) ; Yavapai County — Nelson, Crookton, Gleed, Drake, Hillsdale, Grand View, Kirkland, Skull Valley, foothills bet. Skull Valley and Rams- gate, Ramsgate, Iron Springs, Prescott, near Prescott at 6000 ft.. Fort Whipple, Crown King, Senator Mountains (=Senator Mine?, 10 mi. s. of Prescott). NEW MEXICO: Grant County— Steeple Rock. The following localities are to be considered quite doubtful unless verified by additional specimens: Florence, Pinal County; Fort Buchanan, Santa Cruz County; Sonora, Mexico, 20 mi. se. of Sasabe, Pima County, Arizona. The Florence specimen could well have come from the mountains to the * The locality records are given for this subspecies only, because of its discon- tinuous range. 90 San Diego Society of Natural History northeast, in the vicinity of Superior. Of the Fort Buchanan specimen, only the head is available; it was sent to the U. S. National Museum by Dr. J. B. D. Irwin and comprises one of a troublesome accession of speci- mens that have led to unwarranted range extensions of other reptile species in the past; for while they are credited to Fort Buchanan, from which place they were sent to the Museum, they were probably collected in southern California, since they include such other non-Arizona forms as Masticophis lateralis and Coluber constrictor mormon. The Sonora specimens were taken by a collector whose records later proved to be unreliable. A New Island Subspecies of Crotalus viridis I believe the stunted rattlesnake found on South Coronado Island, Mex- ico, to be worthy of subspecific recognition and therefore describe it as Crotalus viridis caliginis''' subsp. nov. Coronado Island Rattlesnake Plate 5, fig. 2. 1877. Crotalus adamanteus atrox Streets, Bull. U. S. Nat. Mus., no. 7, p. 40. 1895. Crotalus atrox (part) Van Denburgh, Proc, Calif. Acad. Sci., ser. 2, vol. 5, p. 156. 1895. Crotalus lucifer (part) Stejenger, Ann. Rept. U. S. Nat. Mus. for 1893, p. 445. 1896. Crotalus confluentus (part) Boulenger, Cat. Snakes. Brit. Mus., vol. 3, p. 576. 1905. Crotalus oregonus Van Denburgh, Proc. Calif. Acad. Sci., ser. 3, vol. 4, no. 1, p. 18. 1922. Crotalus oreganus (part) Van Denburgh, Occ. Papers Calif. Acad. Sci., no. 10, vol. 2, p. 930. 1929. Crotalus confluentus oreganus (part) do Amaral, Bull. Antivenin Inst. Amer., vol. 2, no. 4, p. 92. 1936. Crotalus viridis oreganus (part) Klauber, Trans. San Diego Soc. Nat. Hist., vol. 8, no. 20, p. 191. Type Specimen.— No. 2800 in the collection of L. M. Klauber. Collec- ted June 2, 1930, on South Coronado Island, off the northwest coast of Baja California, Mexico, by E. H. Quayle, and preserved June 10, 1930. Diagnosis.— A stunted form closely related to, but smaller than, helleri of the nearby mainland, and having those features of body proportionality characteristic of stunted races (Klauber, 1938, p. 29) as compared to theii larger congeners. Description of the Type. — An adult male. The fact that it is adult is shown by its possession of an incomplete string of 8 rattles, the oldest of which is no smaller than the proximal ring. The length over-all— measured * A dweller in the fog. I have Dr. Carl L. Hubbs to thank for this appropriate for the rattlesnake of this fog-enshrouded island. name Klauber— New Rattlesnakes 91 before shrinkage in preservative — was 674 mm. and the tail length 52 mm.; ratio of tail to total length .077. The head dimensions were 33 mm. long by 29 mm. wide. The scale rows are 29-27-25-27 (mid-body) -25-23-22. There are 17 rows at the middle of the tail. The middorsal rows are strongly keeled, the lat- eral rows less so, and the three lowest rows on each side are practically smooth. The ventral plates number 171; the anal is entire; the subcaudals are 25, only the last being divided. The 8 elements of the rattle string average 11.3 mm. in width; they are substantially equal in size, the greatest differ- ence between any two being 0.2 mm. The supralabials number 16 — 15, the infralabials 14 — 15. The first infralabials are undivided, and there are neither submentals nor intergenials. The rostral is higher than wide. Ten scales contact the rostral, there being 4 internasals, and a small extra scale on either side at the rostral-prenasal-supralabial junction. The canthals number 2 — 3. There is no line of demarcation between the scales in the prefontal and frontal areas; the former number about 23. The minimum scales between the supraoculars number 6+6. The supraoculars are pitted but not sutured. The nasals are divided, the anterior being much the larger. There is but one loreal on either side, the upper preocular contacting the postnasal above it. There are 6 small scales on the right and 5 on the left between the nasals and the pit border. The 2 preoculars on either side are undivided. Nine scales on each side contact the eye. There are 3 to 4 scales between the supralabials and the orbit. The head is uniform dark-brown above. A preocular light stripe begins at the upper preocular and passes diagonally downward, to end at the angle of the mouth; its anterior section is finely stippled with brown and is less apparent than the posterior, which is almost clear buff. The remnants of what was once a postocular light stripe are to be seen on either side below the canthus. The underside of the head is buff, but the mental, anterior infralabials, and the inner edges of the genials are heavily punctated with gray. The body pattern comprises some 34 dark-brown dorsal blotches, hexag- onal to round anteriorly, but changing into cross-bands toward the tail. The dark dorsal blotches are edged with rows of gray scales, the blotches being separated dorsally by about iVz light-colored scales. On the sides, in the angles between the main dorsal series of blotches, the scales are darkened, forming an irregular secondary series of blotches of smaller size; below these there is another row of spots that are somewhat darker and more regular than those above. Posteriorly these become confluent with the main dorsal series to form rings. The undersurface is buff, heavily punctated with gray, becom- ing increasingly dark posteriorly. The tail is crossed by 4 dark-brown rings, separated by narrower light-gray interspaces. The last ring, in characteristic helleri fashion, is more than twice as wide as the anterior 3 and is a trifle lighter. The anterior lobe of the rattle matrix is punctated with brown; the posterior lobes are buff. The hemipenes are of characteristic viridis form with short spines on the outer shoulders. 92 San Diego Society of Natural History Material. — In addition to the type, 30 specimens have been available from the Coronado Islands. Twenty-six specimens are designated as having been collected on South Coronado Island, one on East Coronado, and the other 3 merely from the "Coronado Islands". However, all probably came from South Coronado, for this, the largest island in the group — it is about 1% miles long and reaches a peak 672 feet above the ocean — is not only the most southerly, but the most easterly as well (see U. S. Navy Hydrographic Sheet No. 5195). North Coronado, which is also the westernmost of the group, and the second largest, is iVi miles from South Coronado. The other two islands, currendy known as Big Middle and Little Middle, are but rocks lying between the larger pair. I doubt that rattlesnakes have ever been collected on any but South Coronado, and that East Coronado was merely another name applied to that island. The paratypic series includes the following: California Academy of Sciences 13476, 13583-7, 13603, 63734; LMK 2801-4, 4924-6, 7538-40, 20077-8; Museum of Vertebrate Zoology, University of California 5404; San Diego Society of Natural History 11177-8, 13711-5; Stanford University 6681; U. S. National Museum 8564. Description of Subspecies. — The following description is a summary of the data on 31 specimens, including both the type and paratypes. There are 11 males and 20 females. Fortunately, both juveniles and adults have been available. This is a snake of normal viridis proportions. Like others of the viridis subspecies, it is distinguishable from all other rattlesnakes in usually having more than 2 internasals; only 2 out of 31 specimens of caliginis have 2 inter- nasals. The longest available specimen is a male measuring 683 mm.; the longest female is 647 mm. The shortest juvenile measures 220 mm. The ratio of the tail length to the length over-all in the adults averages .079 in the males, and .067 in the females. Most specimens have 25 scale rows at mid-body, the distribution being 25(23), 26(1), 27(6), 30(1).* The average is 25.58, and the coefficient of variation 4.0 per cent. The ventrals in the males vary from 167 to 174; the interquartile range is 168.5 to 171.3, the mean 169.90, and the coefficient of variation 1.3 per cent. The ventrals in the females range from 171 to 179. The inter- quartile range is 173.3 to 176.8, the mean 175.06, and the coefficient of variation 1.5 per cent. The subcaudals in the males vary from 22 to 28, with an interquartile range of 22.7 to 25.4, mean 24.0, and coefficient of variation 8.3 per cent; the females range from 15 to 23, interquartile range 17.6 to 20.1, mean 18.85, and coefficient of variation 10.1 per cent. The rostral is higher than wide. There are from 2 to 6 internasals; most specimens (23 out of 30) have 4. Two specimens out of 30 have only 2 * High for any viridis subspecies; evidently a freak. Klauber — New Rattlesnakes 93 internasals and would fail to key out as belonging to a viridis subspecies. The canthals are usually 2 — 2, but there are occasionally 3 on one or both sides. The scales on the snout, anterior to the supraoculars, range from 18 to 30, the mean being 22.8. Since there is no sharp line between these scales and those in the frontal area, observers might differ in counting them. The minimum scales across the space between the supraoculars vary from 3 to 7, most specimens having 4, 5, or 6. Usually there is a single loreal on each side, but sometimes there are 2, and, rarely, none. The prenasal usually contacts the first supralabial, but this contact is prevented by a row of small scales in 3 counts, and in 16 other cases out of 59, there are one or more small scales at the junction of the prenasal, first supralabial, and rostral. The supralabials range from 12 to 16; most specimens have 14 to 16, the mean being 14.53. The infralabials range from 13 to 18, but most specimens have 15 or 16, the mean being 15.14. The body blotches vary from 28 to 37; the interquartile range is 31.2 to 34.7, the mean 32.90, and the coefficient of variation 7.9 per cent. The tail rings of the males range from 4 to 6, with a mean of 5.18; and the females from 3 to 5, with a mean of 4.20. In the juveniles, the head is brown on top, becoming somewhat darker posteriorly. A faint light line crosses the middle of the prefrontal area, and another, better defined, crosses the supraoculars and the frontal area between. A faint middorsal light line usually connects these two cross-lines. On the sides a preocular light stripe begins at the upper preocular, and passes back- ward and downward to engage the posterior supralabials. A second light band, about 2 scales wide, arises behind the eye and runs back to end at the hinge of the jaw. Otherwise the side of the head is dark-brown, the dark stripe lying between the two light stripes being somewhat darker than the rest. Below, the head is light-gray, except that the mental and first supral- abials are stippled with dark-gray. As the snakes age, the head marks lose much of their definition. The two transverse lines on the crown virtually disappear; and the light streaks on the sides are invaded by gray and brown punctations, so that they become ill-defined and less striking. The body pattern comprises a series of dark-brown blotches on a gray background. The blotches are quite irregular in shape, but approach hexa- gons. They are darker on the edges than in the centers. They are about 11 or 12 scale rows wide and 3 scales long (end to end). The gray inter- spaces are 1 to iy2 scales long. In the lateral angles formed by the light blotch borders, there is a series of secondary brown blotches on each side; and, below these, another darker, better-defined series that engages the edges of the ventrals. Posteriorly, these combine with the main series to form transverse bands. The tail is crossed by similar bands, the last of which is much wider than the others. The ventral scutes are punctated or mottled with gray or brown, especially toward the tail, otherwise the ground color below is buff. 94 San Diego Society of Natural History The juvenile body pattern is somewhat brighter and more contrasting, between the Ught and dark areas, than the adult. Also, the wide terminal tail ring is yellow in the juveniles but becomes brown in the adults. Of the 20 females, 4 were gravid and contained 2, 3, 3, and 4 eggs. A hemipenial lobe contains from 36 to 42 short spines and terminates in from 30 to 32 fringes. The organ is typically viridis in character. Intersubspecific Relationships. — C. v. caliginis is obviously derived from heller i of the nearby mainland, yet it has evidently been separated from the mainland population for a considerable time. This is to some extent sub- stantiated by the presence on South Island of two other reptiles peculiar to Los Coronados*; Gerrhonotus midticarinatus nanus — found on the other three islands as well — and Pituophis catenijer coronalis also of South Island. South Coronado Island is somewhat over 8 miles from the nearest point on the Baja California mainland; and the intervening channel, which is rough and cold, is about 100 feet deep. While rattlesnakes are occasionally found floating in the Pacific a short distance off shore, it is to be doubted whether one could successfully survive this crossing or make a landing upon arrival, for the island is rocky, with precipitous cliffs, and the beaches are few and limited in extent. The winds blow almost continuously from the island toward the mainland. The conditions are quite different from those existing off Florida and Texas, where rattlesnakes (adamanteus and atrox, respectively) are often found swimming at considerable distances from shore, and there is an unquestioned exchange of population elements between mainland and islands, for the water there is warm and the islands flat, so that landings may be easily made. This difference is substantiated by the number of other islands off Baja California, which, notwithstanding their being only a few miles off shore, are inhabited by reptile forms not occurring on the mainland. Of these. South Todos Santos and San Martin may be cited as examples. But while the evidence points to a long isolation of caliginis, I am unable to find any consistent differences between it and helleri in squamation or pattern. The former may be a trifle lighter and grayer — as to ground color — and the latter browner and darker. C. v. helleri retains the light head marks of youth at a size at which caliginis has lost them. But this is a mere incident of stunting. This raises the question of how real and important these size differences are. Every evidence points to the fact that they are real, and when con- sistent in such an isolated form as this, I think they should be recognized just as they have been for many years in the case of birds and mammals. Here- tofore it has been said that snakes are different from mammals and birds, in that they do not reach a size limit upon becoming adults. It has, in fact, been believed that snakes grow continuously until death, and that the largest * Not to be confused with Coronado, a city on the Coronado Peninsula, across the bay from San Diego, California. South Coronado Island is 20 miles south of Coronado, and is on the Mexican side of the international boundary. Klauber— New Rattlesnakes 95 specimens are merely the oldest. But with the acquisition of much larger series in collections, and the observations on specimens kept alive for long periods in zoos, we now know that this is true only to a minor degree. Reptiles certainly do not attain a growth limit as suddenly as mammals; they may, indeed, continue to grow slowly throughout life. But in any case, this adult growth is minor in extent, so that the longest snakes of a subspecies are not necessarily the oldest, but are merely variants of the same character as tall men. No caliginis, regardless of age or intrasubspecific variation, would ever grow to the size of an average adult mainland helleri. Six specimens out of 31 caliginis exceed 600 mm. and none reaches 700. In helleri 1100-mm. specimens are not at all exceptional, and considerably longer ones have been recorded. If we were to compare, visually, a 650-mm. caliginis with a 1100- mm. helleri, we would find the difference far more striking than is indicated by these figures, for the helleri would have 5 times the bulk of the smaller snake. Further evidence of the stunted nature of caliginis is to be found in the sizes of the gravid females among the paratypes. These have lengths of 528, 547, 560, and 647 mm. Gravidity does not generally appear in helleri until a considerably larger size is reached; the smallest gravid specimen I have seen in a very much larger series was 596 mm. in length. Another evidence of the stunted character of caliginis is head propor- tionality, which I have discussed elsewhere (1938, p. 29). Another is the size of the rattle when parallelism is reached; this is at a width of about 10 mm. in female caliginis, and 1 1 mm. in the males. The corresponding dimensions, in normal mainland helleri, are about 15 mm. in females and 17 in males. The average widths of the first 6 rattles in an unbroken string in caliginis are 6.3, 7.2, 8.2, 8.9, 9.5, 9.9,* whereas in helleri they average 6.6, 8.2, 9.7, 11.2, 12.5, 13.3 mm. It will be noted that there is an increasing proportional divergence with age, the helleri rattle being only 5 per cent larger at the button stage, but is 34 per cent larger at the sixth-rattle stage. I have examined 22 rattlesnakes from Santa Catalina Island, Los Angeles County, California. They are somewhat larger and darker than the snakes of South Coronado, and I therefore prefer to consider them helleri. They have typical helleri tail rings. The ratde dimensions, however, are more like oreganus than helleri. Food Habits.— C. v. caliginis differs from the mainland helleri in feeding primarily on lizards, and this in spite of the fact that South Coronado Island is well populated with rodents that an adult caliginis could cope with quite successfully. I am told by L. M. Huey that the Coronado Island white- footed mouse (Peromyscus maniculatus assimilis) is relatively abundant on the island. Helleri, on the mainland, feeds primarily on mammals, although lizards comprise an important part of the diet of the juveniles. But cahgtms * Sexual dimorphism does not become important until the seventh or eighth rattle. 96 San Diego Society of Natural History does not modify its propensity for lizards even when it has reached a size adequate to eat mice. Of the 20 specimens examined that contained recog- nizable food remains, only one had mammal hair, and this is the more re- markable since hair is recognizable until the feces are voided, which is not always true of lizard scales. The other 19 contained lizard remains; 6 could be recognized as Gerrhonotiis multicarinatus nanus, 3 as Eumeces skiltonianus, and one as Uta stansburiana hesperis, the other 9 being too far digested to be identified. A large alligator lizard is a full meal for an adult caligmis. Mr. Huey advances the plausible theory that the rattlesnakes may be forced by the climate — which is cold and foggy — to be largely diurnal, as are the liz- ards, while the mice are nocturnal. The mice have their daytime refuges in rock crevices and cactus, and are probably not so easy for the snakes to catch as are mice on the mainland, where the rattlers may seek their nocturnal prey down ground holes in the daytime, during those seasons in which the snakes themselves are not nocturnal. A New Island Subspecies of Crotahs mitchellii For years I had heard, from ornithologists and others visiting the islands of the Gulf of California, that rattlesnakes were quite common on one island from which no museum had secured specimens, this being "San Luis". In 1946 arrangements were made with Messrs. Lewis W. Walker and Charles H. Lowe, Jr., to visit this island, Mr. Lowe, particularly, having the primary purpose of securing rattlesnakes. In this, the expedition was quite success- ful, many interesting herpetological specimens, including rattlesnakes, being collected. However, the ratdesnakes do not occur on the island designated as San Luis on the maps, although they are on one of the islands that have been called the San Luis group*, which lies in the Gulf of California, just off the Baja California coast, at Lat. 30" N. The island known to the Mexican fishermen as San Luis is off Willard Point (see U. S. Navy Hydrographic Sheet No. 0619, 53d edition) in San Luis Gonzaga Bay; the one named San Luis on the maps is farther off-shore; it is the largest island of the group and is called La Encantada Grande by the fishermen. Proceeding along the archipelago, in a northwesterly direction from La Encantada Grande, the other islands, according to Mr. Lowe and other recent visitors, are known to the Mexicans by the following names: Pomo, Islote, Cholludo, Coloradito, El Muerto, and Huerfanito. El Muerto,t although much smaller than La Encantada, is next in size; and this is the island where the rattlesnakes occur. It lies about 3 miles from the nearest mainland, east of El Marmol. It is probably the island referred to in the previous accounts, since rattlers have not been discovered on any of the other islands by recent expeditions and they are quite plentiful on El Muerto. Subsequent to the * The fishermen usually refer to this group as Las Islas Encantadas, while some maps call it the San Luis and others the Salvatierra group. t El Muerto is not designated by this name on any map known to me. When named at all, it is called Miramar, Link, or, to add to the confusion, La Encantada. The other islands of the group are also assigned several alternative names, but I shall not compound the perplexity by listing them. Klauber— New Rattlesnakes 97 Walker-Lowe expedition, Joseph R. Slevin and Wallace F. Wood visited there in May, 1947, and secured specimens for the CaHfornia Academy of Sciences. The EI Muerto rattlesnake proves to be a dwarfed race of Crotalus mitchellii. I deem it worthy of subspecific recognition and describe it as Crotalus mitchellii muertensis subsp. nov. El Muerto Island Speckled Rattlesnake Plate 6, fig. 1. Type Specimen. — No. 37447 in the collection of L. M. Klauber. Col- lected June 6 or 7, 1946, on El Muerto Island, Gulf of California, Mexico, by Charles H. Lowe, Jr., and preserved June 22, 1946. Diagnosis. — A dwarfed island race, allied to C. ??2. pyrrbus of the main- land, from which it differs in its small size and in having 24 or fewer scale rows more often than 25, while the contrary is true of pyrrbus; also in having a higher average number of body blotches. In adult head proportions, muer- tensis resembles C. m. pyrrbus rather than C. m. mitchellii. It differs from all other rattlesnakes, except pyrrbus and m. mitcbeliiu in usually having the rostral separated from the prenasal by a row of granules. Description of the Type. — An adult male. The length over-all, as measured before shrinkage in preservative, was 633 mm. and the tail length 42 mm., the ratio being .0663. The head (also measured in life) was 26.9 mm. long, being contained 23.5 times in the length over-all. The scale rows number 27-23-19, with 14 around the middle of the tail; all rows are keeled, except the lowest lateral row, which is also the largest. The ventrals number 177 and the subcaudals 22, of which the last 2 are divided. The anal is entire. There are 17 — 17 supralabials and 16 — 16 infralabials. The first infralabials are undivided and there are neither inter- genials nor submentals. The rostral is triangular, and of equal height and width. As is usual in the mitchellii subspecies, the scales on the snout are much broken up so that few can be assigned the customary names with as- surance. The rostral is separated from the prenasal on both sides by rows of granules; also, the small scales anterior to the pit are carried forward to the rostral so that the anterior nasal is separated from the supralabials. The area usually occupied by the upper- preoculars in most rattlesnakes contains 4 scales on each side, indicating that this scale has been split both horizon- tally and vertically. There are 3 to 4 scale rows on either side between the supralabials and the orbit. There are about 26 small scales in the inter- nasal-prefontal area; the minimum scales across the frontal area number 6+6. The supraoculars are by far the largest scales on the head; they are indented by a crease that may be the beginning of a suture on the right, and a pit reminiscent of stepbensi on the left. The mental is triangular. The first infralabials meet on the median line; altogether, 3 infralabials contact the genials on either side. 98 San Diego Society of Natural History The head is gray above, somewhat darkened with irregular brown blotches anteriorly. A few of the posterior scales on the head contain black dots. There is some evidence of a dark-gray postocular stripe, with a lighter stripe below, but these are by no means definite. The mental and infralabials are punctated with gray, but otherwise the undersurface of the head is cream. The ground color of the dorsum is light-pink, marked by 37 gray-brown blotches that become cross-rings posteriorly. The edges of the blotches are not well defined. The blotches have brown centers and gray edges; the brown areas are speckled with dark-brown or black dots, and the gray with darker-gray. Anteriorly, a series of secondary lateral blotches is in evi- dence; posteriorly these spots merge with the dorsal series to form trans- verse bars. The sides are much suffused and speckled with gray. The ventral surface is cream anteriorly and buff toward the tail, the outer edges of the ventrals being often stippled with gray, especially posteriorly. The tail is gray, with two gray-brown rings, followed by a pair of irregular black marks. The rattle matrix is motded with black and gray. Material. — In addition to the type, 18 specimens have been available, all from El Muerto Island. The paratypes are as follows: LMK 37442-4. 37446-49, 38040; and CAS 81354-63. Description of Subspecies. — The following description is based on 19 specimens, the type and paratypes taken together. There are 9 males and 10 females. All are adults or adolescents, no juveniles being available. This is a rattlesnake of the usual mitcbellii form, characterized by the interposition of a row of small scales between the rostral and prenasal, and the extreme subdivision of the other head scales, those on the crown being convex. The longest individual is a male measuring 637 mm.; the longest female is 534 mm. Four males out of 9 are above 600 mm., and 3 females out of 10 over 500 mm. The shortest snake is a female measuring 431 mm. The ratio of the tail length to the length over-all averages .067 in the adult males, and .053 in the females. The majority of specimens have 23 scale rows at mid-body, the distri- bution being as follows: 11 or 58 per cent have 23; and 8 or 42 per cent have 25. The average is 23.8. The head-length equation is approximately H=.0336L+6.5, where H is the head length and L the length over-all, both expressed in millimeters. The ventrals of the males range from 175 to 184 with a mean of 179.7; and in the females from 174 to 181 with a mean of 178.3. It is to be doubted whether larger series would validate this superiority of the males, for in all other ratdesnake subspecies of which adequate series are available the females average from 1 to 7 ventrals higher than the males. It is to be noted that the sexual dimorphism is lower in mitcbellii than in other species and subspecies. Returning to muertensis, the subcaudals in the males vary Klauber— New Rattlesnakes 99 from 21 to 24 with an average of 22.9; and in the females from 16 to 18 with a mean of 17.4. The rostral is about as high as wide, being sHghtly higher in some in- dividuals and wider in others. All specimens except two have the rostral separated from the prenasals by small scales or granules in the manner so characteristic of all mitchellii subspecies except stephensi; in the exceptional specimens the contact is complete on one side. The separation of the pre- nasal from the supralabials, by the extension to the rostral of the small scales anterior to the pit, is not so universal, for at least a partial contact is made in 17 out of 38 cases. The crescent-shaped postnasal, characteristic of all ratdesnakes, is sometimes divided into a small upper and a larger lower part. The loreal-preocular area is usually occupied by a number of small scales or granules, so that it is no longer possible to identify the large upper preocular, or the single or paired loreals characteristic of most rattlesnakes; where homologues can be recognized, it is evident that the upper preocular has usually been divided both vertically and horizontally. The narrow crescent-shaped lower preocular, that comprises the upper border of the pit in most species of Crotaliis, is sometimes in evidence, but is often broken up into smaller scales. In all except one specimen out of 19 there are blemishes on the supraoculars. Usually these take the form of pits or small sutures at the outer edges. In only 2 specimens out of 19 are these as prominent as is normal in stephensi; nevertheless they are more prevalent in muertensis than in in. mitchellii or pyrrhus. The scales in the internasal-prefrontal area are so broken up that they cannot be counted with accuracy. Most of them are decidedly convex. The minimum scales between the supraoculars vary from 4 to 8, with 5, 6, or 7 predominating; the average is 5.7. The supralabials range from 14 to 18, most specimens having 14 to 17; the mean is 15.9. The last supralabial is about as long as the others. The infralabials number 14 to 19, most specimens having 15 to 18, with a mean of 16.5. The mental is triangular and is often sharply pointed posteriorly. The first infralabials meet on the median line and are followed by a single pair of enlarged genials. No first infralabials are divided, and no specimen has submentals or intergenials. The supralabials are usually smaller than the row of scales above them. The considerable enlargement of the last supralabial that characterizes most m. mitchellii is not evident. The body blotches vary from '32 to 39 with a mean of 35.7. The tail rings in the males vary from 3 to 6, with a mean of 4.3; and the females from 2 to 5, with a mean of 3.3. C. m. muertensis gives a general impression of grayness, the color being applied to a considerable extent in the form of small spots or punctations characteristic of the mitchellii subspecies, and with the indefiniteness of blotch outlines evident in all the subspecies except stephensi. The head is usually gray above, occasionally with a pinkish tinge, and with some evidences of irregular brown spotting. There are always some 100 San Diego Society of Natural History black dots scattered over the crown. On the sides there usually remain some traces of a postocular dark stripe bordered with lighter streaks above and below. The mental, the infralabials, and the fronts of the genials are speckled with gray, otherwise the underside of the head is cream, although there may be a few gray dots scattered about. The dorsal pattern comprises a series of indefinitely outlined blotches, tending toward hexagons anteriorly and toward cross-bands posteriorly, where they become confluent with the lateral series and are somewhat more contrasting with the ground color. The blotches usually have brown centers and gray edges, but both areas are often speckled with dark-gray, dark- brown, or black dots. Dorsally, between blotches, there are usually pinkish spaces, which are sometimes moderately clear of punctations. Laterally there is a gray suffusion, so that the secondary lateral series is often ill- defined, especially anteriorly. The ground color of the lowest rows of lateral scales is cream or pink, and the punctations here become darker, being often black. The undersurfaces of the body vary from cream to buff, always becoming darker posteriorly, and with considerable dark motding or stippling, particularly at the edges of the ventrals. The tail is gray, with an anterior ring or two of brown or dark-gray, stippled with black, and several posterior black rings that are generally irregular in shape. The rattle matrix is mottled gray and black. The rattle widths in the adults, after parallelism is attained, vary from about 10 to 11 mm. in the males, and 8.5 to 9.5 mm. in the females. I have seen only two buttons in muertensis; one measured 6.1 mm. in width, the other 5.5 mm. Inter subs pecijk Relationships. — ^That muertensis is a dwarfed form is indicated in four ways. First, a sample of 19 specimens collected by two parties failed to locate any individual longer than 637 mm. (25 inches). A similar collection on the mainland would certainly have resulted in larger specimens being found. For comparative purposes I give the record lengths known to me of the three mainland subspecies: mitchellii, 939 mm.; pyrrhus 1295 mm.; and stephensi, 943 mm. It should be recalled that a snake 50 per cent longer than another is more than 3 times as bulky. Second, we have the evidence of the gravid females, of which 6 muer- tensis are at hand, measuring respectively 431, 456, 482, 489, 515, and 533 mm. The smallest gravid mitchellii I have seen measured 797 mm., the smallest pyrrhus, 786; and the smallest stephensi, 674 mm., and this in much larger series. Then there are the rattles, which reach parallelism in the male muertensis at a width of about 10 to 11 mm., and in the females at 8.5 to 9.5 mm. In mitchellii, which is noted for its large rattles, proportionate to the size of the snake, the corresponding figures are about 15 mm. for the males and 14 mm. for the females; and in pyrrhus about 14 and 13 mm. Finally, the head proportions indicate stunting; for while the adult head-to-body ratio in muertensis is about the same as that of pyrrhus, at Klauber — New Rattlesnakes 101 similar body lengths the muertemis head is smaller than that of pyrrhus, as is always true of dwarfed races for reasons that I have discussed elsewhere (1938a, p. 29). For example, an average muertensis head in a 600-mm. snake would have a length of 26 8 mm., while a pyrrhus of this size would have a head measuring about 29.2 mm. As might be expected from geographic considerations, miiertensu more nearly resembles pyrrhus than stephensi or mitchelln in most characters. Data on the previously recognized subspecies will be found in Klauber. 1936a. Although superficially like mitchelln in color and pattern, muertensis trends strongly toward pyrrhus in other important features. It resembles that subspecies in adult head and tail proportions, mitchellii having a rela- tively smaller head and longer tail. The spines in the cleft of the hemipenes in muertensis are like those of pyrrhus, in that t'.iey do not approach as close to the sulcus as they do in mitchellii. C. m. muertensis shows no particular likeness to stephensi, except that the frequency of supraocular pits or suture:, characteristic of stephensi is much higher in the island form than in mainland pyrrhus. But muertensis is sharply differentiated from stephensi in the almost universal separation of the rostral from the prenasals. The muertensis pat- tern, also, is much more like that of pyrrhus than stephensi, in that the blotch outlines are vague. As to the particular geographical phase of pyrrhus that muertensis re- sembles, it apparently is m.ore closely allied to the Arizona than to California or Baja California snakes, judging from the frequency of 23 as opposed to 25 scale rows, for the lower number is more prevalent in the Arizona snakes than in those from California. On the other hand, muertensis is high in ventral scale counts, and in this respect more nearly approaches the California snakes— which is all the more notable since stunted forms usually have a tendency toward fewer ventrals. The average number of body blotches in muertensis somewhat exceeds the mean in either California or Arizona speci- mens. Another criterion in which muertensis is closer to the Arizona pyrrhus than to the California stocks — particularly those that come from south of the Riverside-San Bernardino county line — is in the splitting of the preoculars both horizontally and vertically. This occurs in almost every muertensis and it is characteristic of the Arizona pyrrhus as well. The San Diego County snakes often have these scales entire, or split only horizontally or vertically, rather than in both directions. The ratdesnake species mitchellii occurs off Baja California on islands other than El Muerto, but on none of the other islands do specimens show any tendency toward muertensis; all resemble, as might be expected, the nearest mainland forms. Those from Ceralvo, Espiritu Santo, San Jose, and Santa Margarita islands are to be classified as m. mitchellii, since they have the head and tail proportions of mitchellii. It is interesting to observe that, although only a single specimen representing each island has been avail- able from Ceralvo, San Jose, and Santa Margarita, and 3 from Espiritu Santo, yet every one of these islands has produced a larger snake than the largest of 102 San Diego Society of Natural History the 19 specimens from El Muerto, another proof of the relatively small size of miiertensis* Beside El Muerto, Angel de la Guarda Island is the only other island off pyrrhiis territory known to be inhabited by a initchelli't subspecies. The snakes of this island are clearly allied to pyrrhus and are very large, which is unusual for island inhabitants; they may, in fact, prove to be larger than any mainland pyrrhus and are thus quite different from muertensts. Field Notes. — The following field notes were compiled by Charles H. Lowe, jr.. when he collected the first specimens of muertensis on June 6 and 7, 1946: "Of the 9 specimens taken, 5 were collected during evening twilight; 3 were crawling when found and 2 coiled. The other 4 specimens were found in the morning, from shortly after sunrise until as late as 7:30 A.M.; all were coiled among rocks. This snake seems, therefore, to be crepuscular, which may be correlated in part with its lizard feeding habits. "It is interesting to note that all 9 specimens, when found, were within appro.ximately 20 feet (usually less) of safe retreats in rocky crevices or holes. The intense daytime heat and the lack of shading vegetation makes it imperative that they stay near satisfactory shelters from the sun. An indication of the intense daytime heat in the open sunlight is seen in the fact that the two lizards inhabiting the island, Uta mearnsl and Uta stans- biiriana, when observed sitting on rocks after 8:30 to 9:30 A.M., were always in the shade, never sunning themselves. "The night of June 6-7, 1946, was warm. The air temperature that night on El Muerto probably dropped but little, if any, under 80'' F., as indicated by the 9:30 P.M. (June 6) temperature of 86' F., and the 4:30 A.M. (June 7) temperature of 80' F. The rattlers, observed coiled on rocks and apparently asleep early in the morning just as the rising sun was reaching them, appeared to have been there for some time, awaiting the sun to warm them. Perhaps they had remained coiled in the same place since the previous evening. "All 9 rattlesnakes showed vicious dispositions. They rattled contin- uouslv from the moment they saw or heard someone approaching, and for a considerable time after being sacked. When touched, they thrashed their bodies about wildly. One snake bit the dirt three or four times when held down by the neck. Another thrashed so violendy and suddenly after being picked up by the neck that it threw itself out of my grasp and onto the ground. Their actions upon the approach of an intruder leads one to believe that they have some important enemy; perhaps a bird of prey. Ospreys and duck hawks live on the island. "The rattlesnakes were found from the beaches to the ridge of the island, some 626 feet in elevation. Likewise, at least one of the two species * The Santa Margarita specimen was destroyed in the San Francisco fire but the length is of record. Klauber — New Rattlesnakes 103 of lizards, both of which the rattlers somewhat greedily ate, is found on all parts of the island where the rattlers were taken. Uta stansburiana is restricted to the beach area and Uta mearnsi to the sides and tops of the island, not being found on the beach rocks. "A recently dropped sample of snake feces, found near one of the rattle- snakes taken on the beach, contained rodent hair. A few other indications that mice inhabit the island were found. None could be trapped by using peanut butter as bait. There were indications that one rodent on the island is a pocket mouse. The following incidents, on June 7th, indicate that the snakes feed upon the two species of lizards inhabiting the island, when pos- sible: "At 5:05 A.M., a Uta meartisi of medium size was shot and placed (dead) in a white snake-sack. At 5:30 a medium-sized Crotaliis was placed in the same sack with the Uta, and the dead Uta was eaten sometime that morning before 9:30. "At 5:25 P.M., a large Crotalus was caught on the southeastern beach. Five minutes later an adult male Uta stansburiana was shot a few feet from where the rattler was found. At 6:15 P.M. the Uta, having been dead for Va hour, was placed in the sack with the rattlesnake, whereupon the rattler swallowed the lizard in the next 10 minutes. "The rattlesnakes are hard to see, although the rocks on which they are found are multicolored." J. R. Slevin reported that he got one ratder on a boulder-strewn beach as the snake was moving inland to escape the tide. Two specimens contained mice {Peromysciis) . Upon examining the snakes comprising the type series, I found mammal hair in 6 and lizard scales in 4 others. Evidently muertensis welcomes either food and does not strongly prefer lizards, as does caliginis on South Coronado Island. The Black-Tailed Rattlesnake of San Esteban Island Schmidt (1922, p. 697) was the first to mention the presence of Crotalus molossus on San Esteban Island in the Gulf of California. From the single specimen then available he pointed out some of the differences as compared to the mainland snakes. Subsequendy, when I examined this specimen, I thought that some of its peculiarities might be due to its poor state of pres- ervation, which later proved to be the case when a live specimen was secured in 1937, through the courtesy of Capt. G. Allan Hancock. Evidently these snakes are not common on the island, as several other expeditions, asked to be especially on the lookout for them, have failed to find any. Some 10 years ago (1938b, p. 193) I expressed the opinion that the subspecific recognition of this snake was not justified. However, since then I have revised my opinions with regard to island subspecies, the taxonomic 104 San Diego Society of Natural History importance of stunting in snakes, and the practical benefits accruing from the nomenclatorial segregation of island races. For these reasons I have decided to name the San Esteban black-tail Crotalus molossus estebanensis subsp. nov. San Esteban Island Rattlesnake Plate 6, fig. 2. 1922. Crotalus molossus (part) Schmidt, Bull. Amer. Mus. Nat. Hist.. vol. 46, art. 11, p. 697. 1936. Crotalus molossus molossus (part) Klauber, Trans. San Diego Soc. Nat. Hist., vol. 8, no. 20, p. 249. Type Specimen. — No. 26792 in the collection of L. M. Klauber. Col- lected on San Esteban Island, Gulf of California, Mexico, by an expedition under Capt. G. Allan Hancock, and preserved April 17, 1937. Paratype USNM 64586, April 18, 1911, C. H. Townsend, collector. Diagnosis. — A stunted island inolossus, related to the nearby mainland form, from which it differs in pattern and in body proportions. The blotches are lighter, smaller, and more numerous than in typical molossus; it also lacks the darkening of the internasal-prefrontal area. Description of the Types. — The holotype is an adult female. The fact that it is adult is shown by the presence of eggs, and its having a broken string of 5 rattles that are uniform in size. As measured before setting in preservative, the length over-all was 737 mm., the tail length 41 mm., and the head 33.7 mm. The rattles vary from 9.6 to 9.8 mm. in width. The scale rows are 33-27-21, with 13 at the middle of the tail. The dorsal rows are keeled; the 3 lowest lateral rows are smooth; the row adjacent to the ventrals is larger than any others of the dorsal series. The ventrals number 192; the anal is entire; there are 22 subcaudals, of which the first and last 2 are divided. The scales in the prefrontal area total 6, comprising 2 long slim internasals, each of which is almost broken to produce an inter- nasal and a canthal; a pair of small scales between the internasals; and a pair of large prefrontals. The scales between the supraoculars are quite small, the minimum bridge being 5, followed by 6. The supraoculars are undivided. The rostral is slightly wider than high. The prenasal is almost separated from the first supralabial by the small scales anterior to the pit on the right, and is separated on the left; these small scales number 8 on each side. There are 2 loreal scales on each side, and 2 extra scales in the sub- canthal area; these are difficult to classify in molossus. There are 2 pre- oculars on either side; and 5 post- and suboculars on the right and 7 on the left. There are 3 to 4 rows between the supralabials and the orbit, these counts including the suboculars. The supralabials number 18 — 18 and the infralabials 16 — 17. The first infralabials are undivided and there are no intergenials or submentals. Three infralabials contact the mentals on either side. Klauber— New Rattlesnakes 105 The head is unicolor olive-brown on top, although some evidences remain of darker blotches. On the sides there is faint evidence of an ocular dark stripe passing backward above the angle of the mouth. Below this there is an almost obsolete light streak. The supralabials are punctated with brown anteriorly, but become clearer toward the angle of the mouth. The mental and infralabials are also punctated with brown, although the 6th to 8th on each side are almost clear; the undersurface of the head is otherwise buff. The dorsal pattern comprises 41 olive-brown blotches with grayish inter- spaces.* The blotches are of the usual molossiis character, that is, they are open on the sides, are bounded by unicolor scales, and often have, a few light-colored scales at their centers, ' The first 8 blotches are closed by light borders laterally; but those that follow are open and are extended laterally, in the form of single rows of zig-zag brown scales, to the ventrum. Posteriorly. the light scales bordering the dorsal blotches become darker, while the blotches themselves become lighter, until the contrast has declined to such an extent that the posterior 10 blotches are virtually obsolete. They can only be counted by recourse to their lateral extensions, which remain in evidence. The belly is cream-colored, with brown punctations on the scutes adjacent to the lateral blotch-extensions. The punctations on the lower surface in- crease posteriorly. The tail marks comprise 6 rings, brown anteriorly, but changing to dark-brown, and then to black at the rattle. The last 3 inter- 'spaces, which are grayish, are evident only on the sides. Below, the tail is gray anteriorly, but otherwise black. The rattle matrix is black. The paratype is a juvenile male in a rather poor state of preservation. The length over-all is 466 mm., the tail length 35 mm., and the head 23y2 mm. The scale rows are 31-27-21, the ventrals 189, and the subcaudals 26. The rostral is wider than high. The supralabials number 17 — 19, and the infral- abials 17 — 16. The scales in the crown comprise 2 internasals and 2 pre- frontals. The minimum scales between the supraoculars number 4 — 4. The group of small scales anterior to the pit is carried forward to the rostral, thus preventing a contact between the prenasal and first supralabial. There are 2 preoculars and 2 loreals on either side. The upper preoculars are undivided. There are from 2 to 4 rows of scales between the labials and orbit. The general color is grayish. The dorsal blotches, which are olive-gray, are too faint to be counted with accuracy, especially posteriorly; this may, in part, be due to the condition of the specimen. The tail rings are faindy in evidence through the lightening of the interspaces; they are almost black. Inter subs pecijic Relationships. — San Esteban Island lies in the middle of the Gulf of California in Lat. 28° 40' N. It is a barren and rocky island about V/i miles in diameter. The nearest land is Tiburon Island 6^/2 miles to the northeast, and San Lorenzo Island I2V2 miles to the west. Crotalus atrox is found on Tiburon and C. r. ruber on San Lorenzo; so far as is known, C. molossus occurs on neither. Besides this new rattlesnake, San Esteban is * In life these colors at mid-body, using Ridgway's Standards, 1912, were Brown- ish Olive and Reed Yellow. 106 San Diego Society of Natural History known to harbor 5 other reptiles, 2 of which, Cnemidophorus estebanensii and Sauromalus varius, have differentiated sufficiently from their nearest relatives elsewhere to be deemed valid species. As might be experted, Crotalus m. estebanensis is much nearer C. m. molossus than C. m. nigrescens; it has the lighter color, higher number of ventral scales, and lateral blotch extensions that characterize the former sub- species. From molossus molossus the differences are only matters of degree, and some may not be substantiated when more specimens of estebanensis become available. It is stunted and has the morphological differences characteristic of dwarf races. Thus the head of the adult type is somewhat shorter than that of a molossus of similar size; and the rattle has reached parallelism at a width considerably less than that noted in adult females of molossus from Sonora and Arizona, the relative figures being about 9.7 and 14.5 mm. Furthermore, the rattles are peculiarly compressed both longitudinally and transversely, making them quite different in shape from the usual molossus rattle. This is another characteristic of some dwarfed forms. In pattern, the new form differs from molossus of the mainland in having smaller, lighter, and more numerous dorsal blotches. Lowland molossus are less brightly patterned, with less color contrast between the dorsal blotches and the interspaces, than those from the higher elevations, especially in the mountains of southern Arizona; but the type of estebanensis is particularly light-colored, and with the dorsal blotches quite obsolete posteriorly. Whether the same condition originally existed in the paratype cannot be determined because of its condition. Best of all the characters, from the standpoint of a key, is the high number of dorsal blotches. The type has 41; out of 144 specimens of m. molossus available, only one specimen reaches this number, and only 3 have more than 38, the mean being 31.3. A further difference in the type of estebanensis is the absence of a dark-brown patch on the crown of the head in the internasal-prefrontal section. This is evident, and even prominent, in most mainland molossus. Klauber — New Rattlesnakes 107 Acknowledgments I am appreciative of the courtesies extended to me by the many individ- uals and institutions whose collections I have been privileged to examine in the course of these studies, or who have furnished me with specimens or data. I desire particularly to mention Mr. Qiarles M. Bogert, American Museum of Natural History; Mr. Joseph R. Slevin, California Academy of Sciences; Messrs. Karl P. Schmidt and Clifford H. Pope, Chicago Natural History Museum; Dr. Howard K. Gloyd, Chicago Academy of Sciences; Dr. Howard R. Hill, Los Angeles County Museum of History, Science and Art; Mr. Arthur Loveridge, Museum of Comparative Zoology, Harvard University; Dr. Robert C. Stebbins, Museum of Vertebrate Zoology, University of California; Dr. George S. Myers and Miss Margaret Storey, Natural History Museum, Stanford University; Dr. Waldo L. Schmitt and Dr. Doris M. Cochran, United States National Museum; Dr. Raymond B. Cowles. University of California at Los Angeles; Mrs. Helen T. Gaige, then of the Museum of Zoology, University of Michigan; Dr. Emmett R. Dunn, Academy of Natural Sciences of Philadelphia; Dr. E. H. Taylor, University of Kansas; Dr. Henry S. Fitch, U. S. Fish and Wildlife Service; Mr. Charles H. Lowe, Jr., Mr. Lewis W. Walker, Captain G. Allan Hancock, Mr. Waldo G. Abbott, and Mr. R. M. Williams. I am especially grateful to Dr. Carl L. Hubbs, Dr. Joshua L. Baily, Mr. C. B. Perkins and Mr. Charles E. Shaw for their editorial criticisms and suggestions. I was assisted by Charles E. Shaw, Bruce Provin, and Richard Schwenkmeyer in making scale counts. The photographs were taken by Mr. Leslie C Kobler. Summary In a resurvey of the widespread Pacific rattlesnake, Crotalus viridis oreganus, geographical differences are found to warrant recognizing two addi- tional subspecies, one involving the southern California and northern Baja California population, the other that in Arizona. For these, names long in synonymy are now revived, the California subspecies to be known as Crotalus viridis helleri Meek, 1905; and the Arizona form as C. v. cerberus (Coues), 1875. Territorial boundaries and subspecific differences are discussed. Three new rattlesnake subspecies are described, dwarfing being considered an important justification for the recognition of three island subspecies: These are Crotalus viridis caliginis of South Coronado Island, off the north- west coast of Baja California, Mexico; Crotalus mitchellii muertensis of El Muerto Island, off Baja California near the head of the Gulf of California, Mexico; and Crotalus molossus estebanensis from San Esteban Island in the central Gulf of California. Bibliography Amaral, Afranio do 1929. Studies of Nearctic Ophidia. V. On Crotalus conjluentus Say, 1823, and its Allied Forms. Bull. Antivenin Inst. Amer., vol. 2, no. 4, pp. 86—97. 108 San Diego Society of Natural History Anon 1937. Sailing Directions for the West Coasts of Mexico and Central America. Eighth ed. Hydrographic Office, Washington, pp. vii+430. Baird. S. F. and Girard C. 1852. Characteristics of Some New Reptiles in the Museum of the Smithsonian Institution. Proc. Acad. Nat. Sci. Phila., vol. 6. pp. 173-177. 1853. Catalogue of North American Reptiles in the Museum of the Smithsonian Institution. Part 1 — Serpents, pp. xvi+172. BOULENGER, G. A. 1896. Catalogue of the Snakes in the British Museum (Natural History), vol. 3, pp. xiv+727. Brown. A. E. 1901. A Review of the Genera and Species of American Snakes, North of Mexico. Proc. Acad. Nat. Sci. Phila., vol. 53, pp. 10-110. Cooper. J. G. 1868. Zoology — Reptiles, in T. F. Cronise's The Natural Wealth oj California. San Francisco, pp. xvi + 696 [Zoology, Cooper, pp. 434-501]. 1870. The Fauna of California and its Geographical Distribution. Proc. Cal. Acad. Sci., vol. 4, part 2, pp. 61-81. Cope, E. D. 1859. Catalogue of the Venomous Serpents in the Museum of the Academy of Natural Sciences of Philadelphia, with Notes on the Families, Genera and Species. Proc. Acad. Nat. Sci. Phila.. vol. 11, pp. 332-347. 1860. An Enumeration of the Genera and Species of Rattlesnakes, with Synonymy and References. (Appendix A in Researches upon the Venom of the Rattlesnake, by S. Weir Mitchell.) Smithsonian Cont. Knowl., vol. 12, art. 6, pp. 119-145. 1866. On the Reptilia and Batrachia of the Sonoran Province of the Nearctic Region. Proc. Acad. Nat. Sci. Phila., vol. 18, pp. 300-314. 1883. Notes on the Geographical Distribution of Batrachia and Reptilia in Western North America. Proc. Acad. Nat. Sci. Phila., vol. 35, pp. 10-35. 1900. The Crocodilians, Lizards, and Snakes of North America. Rept. U. S. Nat. Mus. for 1898, pp. 153-1270. CouES, Elliott 1875. Synopsis of the Reptiles and Batrachians of Arizona. Explor. Surv. W. of 100th Merid., vol. 5, chapt. 5, pp. 585-633. Garman, Samuel 1883. The Reptiles and Batrachians of North America. Mem. Mus. Comp. Zool., vol. 8, no. 3, pp. xxxi+'185. Klauber — New Rattlesnakes 109 Gill, Theodore 1903. The First Edition of Holbrook's North American Herpetology. Science, N. S., vol. 17, no. 440, pp. 910-911. Gloyd, H. K. 1940. The Rattlesnakes, Genera Sistrurus and Crotalus, a Study in Zoogeography and Evolution. Chi. Acad. Sci., Spec. Pub. no. 4, pp. vii+266+[4]. Grinnel, Joseph and Camp, Charles L. 1917. A Distribution List of the Amphibians and Reptiles of Cali- fornia. Univ. Calif. Pubs, in Zo6l., vol. 17, no. 10, pp. 127-208. Hallowell, Edward 1852. Descriptions of New Species of Reptiles Inhabiting North America. Proc. Acad. Nat. Sci. Phila., vol. 6, pp. 177-182. 1859. Report upon Reptiles of the Route. Report of Explorations in California for Railroad Routes (Williamson Route) Pac. R. R. Surv., vol. 10, pt. 4, no. 1, pp. 1-27. HOLBROOK, J. E. 1840. North American Herpetology, ed. 1, vol. 4, pp. viii + 9-126. 1842. North American Herpetology, Philadelphia, ed. 2, vol. 3, pp. 1-128. Jan Georg 1863. Elenco Sistematico degli Ofidi. Milano, pp. 1-143. Klauber, L. M. 1930a. New and Renamed Subspecies of Crotalus confluentus Say, with Remarks on Related Species. Trans. San Diego Soc. Nat Hist., vol. 6, no. 3, pp. 95-144. 1930b. Differential Characteristics of Southwestern Rattlesnakes Allied to Crotalus atrox. Bull. Zobl. Soc. San Diego, no. 6, pp. 1-72. 1935. A New Subspecies of Crotalus confluentus, the Prairie Rattle- snake. Trans. San Diego Soc. Nat. Hist., vol. 8, no. 13, pp. 75-90. 1936a. Crotalus mitchellii, the Speckled Rattlesnake. Trans. San Diego Soc. Nat. Hist., vol. 8, no. 19, pp. 149-184. 1936b. A Key to the Rattlesnakes with Summary of Characteristics. Trans. San Diego Soc. Nat. Hist., vol. 8, no. 20, pp. 185-276. 1938a. A Statistical Study of the Rattlesnakes. V. Head Dimensions. Occ. Papers San Diego Soc. Nat. Hist., no. 4, pp. 1-53. 1938b. Notes from a Herpetological Diary, I. Copeia, no. 4, pp. 191- 197. Meek, S. E. 1905. An Annotated List of a Collection of Reptiles from Southern California and Northern Lower California. Field Columb. Mus., Zobl. Ser., vol. 7, no. 1, pub. 104, pp. 1-19. 110 San Diego Society of Natural History Schmidt, K. P. 1922. The Amphibians and Reptiles of Lower California and the Neighboring Islands. Bull. Amer. Mus. Nat. Hist., vol. 46, art. 11, pp. 607-707. 1942. The First Edition of Holbrook's North American Herpetology. Copeia, no. 1, pp. 53-54. Schmidt, K. P. and Davis, D. D. 1941. Field Book of Snakes. New York. pp. xiii+365. Stejneger, Leonhard 1895. The Poisonous Snakes of North America. Rept. U. S. Nat. Mus. for 1893, pp. 337-487. Streets, T. H. 1877. Contributions to the Natural History of the Hawaiian and Fanning Islands and Lower California. Bull. U. S. Nat. Mus., no. 7, pp. 1-172. Van Denburgh, John 1895. A Review of the Herpetology of Lower California. Part 1, Reptiles. Proc. Calif. Acad. Sci., ser. 2, vol. 5, pp. 77-162. 1896. Additional Notes on the Herpetology of Lower California. Proc. Calif. Acad. Sci., ser. 2, vol. 5, pp. 1004-1008. 1897. The Reptiles of the Pacific Coast and Great Basin. Occas. Papers Calif. Acad. Sci., no. 5, pp. 1-236. 1898. Herpetological Notes. Proc. Amer. Philos. Soc, vol. 37, no. 157, pp. 139-141. 1905. The Reptiles and Amphibians of the Islands of the Pacific Coast of North America from the Farallons to Cape San Lucas and the Revilla Gigedos. Proc. Calif. Acad. Sci., ser. 3, vol. 4, no. 1, pp. 1-40. 1912. Notes on a Collection of Reptiles from Southern California and Arizona. Proc. Calif. Acad. Sci., ser. 4, vol. 3, pp. 147-154. 1922. The Reptiles of Western North America. Occas. Papers Calif. Acad. Sci., no. 10, 2 vols., pp. 1-1028. Van Denburgh, John and Slevin, Joseph R. 1913. A list of the Amphibians and Reptiles of Arizona, with Notes on the Species in the Collection of the Academy. Proc. Calif. Acad. Sci., ser. 4, vol. 3, pp. 391-454. 1914. Reptiles and Amphibians of the Islands of the West Coast of North America. Proc. Calif. Acad. Sci., ser. 4, vol. 4, no. 5, pp. 129-152. Yarrow, H. C. 1875. Report upon the Collections of Batrachians and Reptiles made in Portions of Nevada, Utah, California, Colorado, New Mex- ico, and Arizona during the years 1871, 1872, 1873, and 1874. Explor. Surv. W. of 100th Merid., vol. 5, chapt. 4, pp. 509-584. 1883. Check List of North American Reptilia and Batrachia, with Catalogue of Specimens in U. S. National Museum. Bull. U. S. Nat. Mus., no. 24, pp. 1-249. Klauber— New Rattlesnakes Plate 4 ,^ Fig. 1. Crotalus viridis oreganus Northern Pacific Rattlesnake Young adult male from near Wenatchee, Chelan County, Washington. Fig. 2. Crotalus viridis helleri Southern Pacific Rattlesnake Adult male from Rancho Santa Fe, San Diego County, California Klauber — New Rattlesnakes Plate 5 Fig. 1. Cro talus viridis cerberus Arizona Black Rattlesnake. Adult male from Hillside, Yavapai County, Arizona Fig. 2. Crotalus viridis caliginis Coronado Island Rattlesnake. Young male from South G)ronado Island, Pacific Coast of Baja California, Mexico. Klauber — New Rattlesnakes Plate 6 Fig. 1. Crotalns mitchellii muertensis El Muerto Island Speckled Rattlesnake. Adult male from El Muerto Island, Gulf of California, Mexico. Fig. 2. Crotalus molossus estebanensis San Esteban Island Black-tailed Ratdesnake. Adult female (type specimen) from San Esteban Island, Gulf of California, Mexico. _J r».j «?:: ii *~l —t as ■s^ e/9 =3 =3 2 MUS. mi?. ZOOL UBRARY HAY 16 19^9 HARV?.?^0 UNIVERSITY TRANSACTIONS OF THE _SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 7, pp. 117-120, plate 7 April 29, 1949 A NEW FOSSIL CHILOPOD FROM THE LATE CENOZOIC BY Ralph V. Chamberlin The present account of a fossil chilopod is based upon a single specimen found in an onyx marble pen base prepared by the Southwest Onyx and Marble Company. The specimen was sent to me for study through Prof. A. Petrunkevitch of Yale University at the request of Mr. Clinton G. Abbott, late director of the San Diego Natural History Museum. The pen base as received was broken in such a way as to divide the elongate specimen into two parts. This of itself has in no way increased the difficulty of the study. It is, however, very unfortunate that the head as well as the caudal end of the body, the two regions affording ordinarily the most significant diagnostic characters, are so far disintegrated and the fragments lost that nothing can be found excepting portions of the two antennae. Nevertheless, the great interest and importance attaching to fossils front this new source seem to justify a description, even though incomplete, of the specimen and an indication of its probable systematic position. The fossil centiped, as well as those of an arachnid already described and named by Prof. Petrunkevitch^, was discovered by Mr. J. W. Fisher, president of the Onyx company mentioned above, through whose courtesy they have been m.ade available for study. For a detailed account of the quarry the interested person may consult the account by Prof. E. D. McKee quoted in the paper by Prof. Petrunkevitch here cited. This quarry is located "in a canyon on the north side of Black Mesa, about ten miles southwest of Ashfork, Arizona. . . . The age of the deposit is definitely 'post-faultmg', which means since the middle of Cenozoic time, but deposition might have been any time from then up to the present. Similar deposits of travertme are forming today in many parts of the region where there are permanent or semipermanent flows of water." * A. Petrunkevitch. "Calcitro ftsheri, a new fossil arachnid/" Amer. Jour. Sci., 1945, vol 243. p. 320. 118 San Diego Society of Natural History While the features of the mouthparts are unknown, the form of the antennae and the general structure of the body segments render its place in the Geophilidae as highly probable. The new form may thus be systematically treated as follows. Order GEOPHILIDA Family Geophilidae Genus CALCIPHILUS new Apparently lacking pleural suprascutella and otherwise agreeing with Geophilus in having the antennae filiform and the tergites bisulcate. Differing from that genus and related forms in having the legs proportionately longer. Generotype: Calciphilus abboti new species. The disintegration and nearly complete loss of the head and posterior end of the body prevents a more adequate diagnosis. Calciphilus abboti new species Long and slender, its length much exceeding that of any living species of Geophilus sens. str. of the Southwest and also apparently somewhat longer than the conspicuous Linotaenia laevipes (Wood) common in California and belonging to the closely related family Linotaeniidae. The total length of the holotype is estimated to be 115 mm. or more, with the width 1.5 mm., though in some parts of length less than this from shrinking. In comparison with this the legs of a typical segment of middle region of body are 2.3 mm. long. Because of the coiling and the partial disintegration of the posterior portion of the body the number of pairs of legs cannot be determined accurately but must be 115 pairs or more. The legs are proportionately long with joints of the relative proportions shown in the accompanying figure; terminal claw with two accessory claws or spines. Sternites with complete paired longitudinal sulci spaced as shown in the figure. It has been impossible in the present state of preservation of the specimen to detect any ventral pores although such may have been present. The tergites also seem to be bisulcate as in typical Geophilus. Of the head only portions of the two antennae remain, eight articles of the right and three of the left being preserved in situ while two other segments are in the matrix at a distance. The form and proportions of the middle and distal articles of the antennae are shown in the figure. The length of the eight segments of the right antenna is 1.8 mm., which would normally mean a total length of a complete antenna of close to 3.1 mm. "Type locality: Arizona: Bonner Quarry, near Ashfork. Holotype: San Diego Society of Natural History Chamberlin — A New Fossil Plate 7 Fig. 1. Eight distal segments of an antenna. Fig. 2. Segment from middle region of body, ventral view. MUS. COMP. 200L UBRARY OCF 17 19^9 HARVARD UNIVERSITY TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 8, pp. 121-140 THE SUBSPECIES OF THE RIDGE-NOSED RATTLESNAKE, CROTALUS WILLARDI Laurence M. Klauber Curator oj Reptiles and Amphibians, San Diego Society of Natural History SAN DIEGO, CALIFORNIA Printed for the Society September 30, 1949 Jv\>mu L r>v..\ L\ORKNVi M k. vl tV>R MUS COMP 7001 LIBRARY UCI I i* hi^H HARVARD UNIVEKSIIY ^T or roN-n N IS I'v. IntioJiution C"». ./.»/.-. n,ll,ty,i, ntlivJ, M.vk V\\\- Sp^l/l<^ I' ;//,/». /) w/m> miI>>i>, hov Ix pC S|>CilH>CI\ PcM npiion ol i!ic In p<- P.u.ilN pes IXvMiiplioii ol ll\i- SulvNpoiic's (■».'f.i/;(> nill.nJi »»;<-M./(.'fi.i/;. siihsp iio\- Is pc Spc-. iinni 1 >,,,,vM>OMs 1 \vsi iiplion ol llic I'vpc l\ll.l(\pcs lnt<-iMil'spc. ill. l'i. willardi. The light head marks so typical of the Arizona subspecies are not evident. Range. — This subspecies has been collected only in northeastern Sonora and western Chihuahua, at the following specific localities: SONORA: Above the Santa Maria Mine in El Tigre Mountains. CHIHUAHUA: Rio Gavilan (7 mi. sw. of Pacheco) at 5700, 6200, and 6700 feet; Sierra Madre (near the summit) at 2400 meters (7874 feet); near Colonia Garcia; Rio Piedras Verdes at 6900 feet (at the head of the canyon) ; half way between Nahuarachic and Las Varas at 7000 to 8000 feet; Tamarino; Distrito Guerrero at 2250 and 2365 meters (7382 and 7759 feet) ; Mojarachic. Crotalus willardi meridionalis subsp. nov. Southern Ridge-nosed Rattlesnake 1936. Crotalus willardi (part) Klauber, Trans. San Diego Soc. Nat. Hist., vol. 8, no. 20, p. 231. Type Specimen. — No. 6569 in the collection of L. M. Klauber. Collected in August, 1904, by Edmund Heller and Charles M. Barber at Coyotes, Durango, Mexico, at elevation 8000 ft. This is the Coyotes on the railroad to El Salto. The locality will be found on Am. Geogr. Soc. Millionth Map, sheet North F-13, where the altitude is given as 2574 m. (8445 ft.). The specimen was formerly one of a pair, carrying the single number 1493, in the Field Museum collection, now the Chicago Natural History Museum. Diagnosis. — This subspecies differs from w. willardi in having a propor- tionately longer tail, more scale rows, more subcaudals, more body blotches, but fewer ventrals. From silus it differs in having a conspicuous vertical light line on the rostral and mental, and other head marks not evident in siliii. On the average, it has fewer ventrals than silus. Description of the Type. — The type specimen is an adult female with 132 San Diego Society of Natural History a length over-all of 562 mm. and a tail length of 53 mm.; ratio of tail to total length .094. The head length is 301/2 mm., and the width 20y2 mm. The fang length (point to lower end of upper lumen) is 5.8 mm. The rattle width (3 segments of equal size remain) is 6.1 mm. The rattles are more rounded than in most other rattlesnakes, as is characteristic of willardi. The scale rows number 29-27-19, with 9 at the middle of the tail. The dorsal rows are strongly keeled, but the rows above the ventrals, which are also the largest, are smooth, and the next two rows only faintly keeled. Paired apical scale pits are faintly in evidence. The ventrals number 149; the anal is entire; the subcaudals number 30, of which the last 3 are divided, the final row being feathered out into a ring of pointed scales that jut over the proximal rattle. The rostral is flat, square-topped, and is higher than wide; it engages the first supralabials, a pair of large prenasals, and a pair of internasals. The latter are the most distinctive scales in this snake, for their anterior halves turn upward at right angles to form a ridge above the rostral, giving the snake its popular name. Behind each internasal there are 3 canthals; the first of each series is uptilted to form a posterior continuation of the ridge formed by the internasals; the second is the largest of the series and is slightly raised, outwardly; the last, which is quite small, engages the supraoculars. The area between the canthals is concave and contains about 25 small scales. The frontal area is also filled with small scales, there being a minimum of 7 bridging the space between the supraoculars. The supraoculars are the largest scales on the head; they narrow to points anteriorly. On each side there is a large anterior nasal, followed by a smaller postnasal; 2 loreals, the upper smaller; and 2 preoculars, the upper larger and the lower long and thin, and forming, with the lower loreal, the upper border of the pit. There are 6 sub- and postoculars on the right and 7 on the left. There are 3 rows of scales between the labials and the orbit. The supralabials number 14 — 14 and the infralabials 14 — 15. The mental is triangular. The first infralabials are undivided; they meet on the median line behind the mental and are followed by a pair of enlarged genials, which are contacted by the first 3 in- fralabials on either side. There are no submentals or intergenials. The condition of this specimen precludes a description of the body pattern with assurance, as the epidermis has been scraped from some areas. The head is brown above, the color being applied in the form of very fine stippling. Thjsre are scattered black dots on the supraoculars, with some smaller, less prominent ones on the canthus. The sides of the head are strikingly marked with white lines on a brown background. There is, first, a thin vertical white line down the center of the rostral; upon reaching the lower edge of the rostral this line branches to right and left and becomes a narrow white border of the upper lip; it continues back to the fifth supralabial, where it disappears. A second white stripe begins at the upper center of each pre- nasal and passes backward and downward across the pit and below the eyes, to terminate on the last supralabials at the angle of the mouth. This streak Klauber — Ridge-Nosed Rattlesnake 133 widens posteriorly and is lined with dark-brown dots on some scales. The underjaw is also brightly marked with white streaks. Down the center of the mental there is a narrow white streak, which is a continuation or that on the rostral. This widens as it follows the median line between the genials. At the first gulars it diverges to the gulars adjacent to the last supralabials. The area between is irregularly spotted and blotched with gray, so that some gulars are white while others are gray. A second light line on each side begins as a narrow white edge on the lip at the third infralabial and passes backward, widening along the lip until about the ninth infralabial. Here it turns down, following the first row of scales below the infralabials, until the commissure is reached, where it branches, one part turning upward to the commissure, while the other passes back to the neck. Except for the white areas, the underjaw is gray-brown anteriorly and gray toward the neck. The dorsal pattern comprises a series of 29 brown blotches on a lighter- brown ground color. The blotches are not well defined, although they are somewhat clearer medianly, where the contrast between the dark blotches and the light interspaces is accentuated. The lateral edges of the blotches are quite indefinite, but this may be due to the condition of the specimen. Pos- teriorly, the blotches form fairly well-outlined crossbands. There are a few widely spaced dark-brown or black dots scattered over the dorsum. The sides are somewhat grayish. The ventrum is buff anteriorly, increasingly punctated and blotched with gray toward the tail, so that two definite lines of gray mottling are formed. The tail is crossed by 3 brown bands anteriorly, with light-brown inter- spaces. Posteriorly it is gray-brown on the dorsum and gray laterally. On the sides many of the scales contain a black spot posteriorly, and the lowest lateral row is edged with white. Below, the tail is light-gray, speckled and spotted with dark-gray or black. The rattle matrix is light-gray with a few scattered black spots. Paratypes. — There are three paratypes: Chicago Natural History Mu- seum No. 1493, one of a pair from Coyotes, Durango, of which the other is the holotype; Chicago Academy of Sciences No. 13953 from Weicher Ranch, 50 miles west of Durango, Durango (State); and U. S. National Museum No. 46332 from Sierra Madre, Zacatecas, Mexico. The first is a male 510 mm. in length over-all with a tail 58 mm. long, and a head length of 29V2 mm. It has 27 scale rows at mid-body, 146 ventrals, and 31 subcaudals. The supralabials number 17—15, and the in- fralabials 15 — 14. The minimum scales across the frontal space number 8. There are 28 body blotches. The Chicago Academy specimen is a young male measuring 233 mm. over-all with a tail length of 27 mm. The head measures 16 mm. There are 27 scale rows at mid-body, 148 ventrals, and 34 subcaudals. The supra- labials number 14 — 15 and the infralabials 14 — 14. The minimum count of scales between the supraoculars is 8. The body blotches are indeterminate because the skin is torn at the neck. 134 San Diego Society of Natural History The National Museum specimen is a small female in rather poor con- dition. It is 257 mm. long, with a tail length of 28 mm. and a head length of 18.3 mm. There are 29 scale rows, 147 ventrals, and 30 subcaudals. The labials number 14 — 15, 14 — 15, and the minimum scales in the supraocular bridge, 8. The condition of the specimen is such that the blotches cannot be counted. The patterns of the paratypes, to the extent that they can be determined, are much the same as in the type. The heads are brown above, with some black specks, which are particularly evident on the supraoculars and canthals. On the sides and below, the pattern of white streaks, as described in the type, is strikingly evident, especially the vertical line on the rostral and mental, and the side stripes marking the labials. Below there is a pair of divergent gray-brown blotches, punctated with black, that cover the anterior gulars. The dorsum is marked with a series of square brown blotches separated by light-brown to buff interspaces. The anterior blotches are rather poorly defined, with little contrast between blotches and interspaces; posteriorly the contrast increases. All blotches are rather poorly defined on the sides. Below the main series on either side, 3 other series of small dark spots may some- times be distinguished; these become darker as the ventrum is approached. Fine black or brown punctations are scattered over the dorsum. Larger spots often mark the edges of the light interspaces. The ventrum is heavily mottled with dark-gray to black, although the mid-ventral line is somewhat clearer. The dorsal blotches are continued on the anterior third of the tail, but the final sertion is longitudinally striped, instead of being crossbarred. Lat- erally and below, the tail is gray. A thin serrated white line marks the outer edges of the subcaudals. Intersubspecific Trends Although the statistics of the subspecies meridionalis are uncertain be- cause of the meagreness of the material, it appears that the subspecific trends are geographically consistent in most characters. Scale rows, subcaudals, labials, body blotches, and tail-length proportionality all increase from north to south, that is, from w. willardi through stlus to meridionalis. An inconsistency is shown by the ventrals, for in ulus these scutes apparently average slightly higher than iu w. willardi, while in both the ventral count exceeds that of meridionalis. The most surprising inconsistency, speaking geographically, is to be found in the head pattern, for the two terminal forms have retained a striking pattern unique among the rattlesnakes, whereas the intermediate sub- species, although retaining some traces of what must have been the ancestral pattern, has reverted to a nondescript design without distinctive elements. Thus it has lost the vertical white lines on the rostral and mental; the clear white streak from the nostril to the last supralabial, except for a less-contrasting rear half; the second, but shorter, streak along the bottoms of the first five supra- labials; also the two pairs of brown ovals on the lower jaw. All of these color Klauber — Ridge-Nosed Rattlesnake 135 features serve to make w. wllUrcli and meridionalis the most strikingly marked of all rattlesnakes, as far as head pattern is concerned. It is surprising to ob- serve the similarity of w. willardi and meridionalis in head marks, separated as these subspecies are by 600 miles of mountain chain, much of which is occupied by the race ulus that has lost these marks. This is apparently an example of central rather than peripheral differentiation. Life History and Morphology Not much is known concerning the habits of wdlardi, since the relative inaccessibility of the mountain areas where it occurs has made it rare in col- lections. It is probably moderately common in some places, for such collectors as have reached its habitat — this being true particularly of sdus — have been rewarded with a half-dozen or more specimens. That it is consistently a mountain form, is indicated by the fact that none of the 50 or more specimens now available came from an altitude below 5700 feet. The maximum record is somewhat above 9000 feet. Kauffeld (1943, p. 355) refers to it as an alpine forest snake, and suggested that it is not as partial to rocky areas as Crotalus lepidus klciuberi. However, the two specimens that he did collect in the Huachucas were found near or under rocks. Dr. D. D. Brand, who captured three at between 7 and 8000 feet in Chihuahua, found them coiled in rock ledges. Crotalus willardi is known to feed on both mammals and lizards. Like most rattlesnakes of the smaller species, it is probable that lizards comprise a major part of the diet, particularly of the juveniles. A Sceloporus, probably S. ']. jarrovii, was found in a specimen of »'. wdlardi, and an alligator lizard, Elgaria kingii, in a silus. Mammal remains were found in all three subspecies. Kauffeld (1943, p. 355) had one strike and quickly kill a white mouse, from which he judged the venom to be relatively powerful. One of his two captive specimens took dead mice readily, but the other had to be force-fed. To Kauffeld it appeared that willardi, in captivity, showed some resemblance to specimens of Agkistrodon and Bothrops. It remained stretched out rather than assuming the typical resting coil of a rattler, and had the habit of turn- ing to bite. It seemed quite inquisitive. A specimen of w. willardi 481 mm. (19 in.) long contained 6 eggs; and a silus 452 min. (17^/4 in.) in length contained at least 2 well -developed em- bryos, which gives an idea of the size of the adult females. As in all rattle- snakes except Crotalus cerastes, the adult males exceed the females in length by about 10 to 15 per cent. The ratio of the size of the young at birth to the ultimate adult size (about .32) is high, as is usual in the smaller species of rattlesnakes, when compared to the larger. Proportionately, willardi has the largest head of any rattlesnake; the adult head length i.s about .052 tunes the body length. The typical subspecies has a slightly larger head than idiis Also, it has a somewhat longer fang (relative to head length) than most other small species of rattlers. The rattles are distinctive in both form and color. 136 San Diego Society of Natural History being more rounded and darker than in other species. The terminal body scales that edge the proximal rattle are particularly pointed in this species. Key to the Subspecies of Crotalus willardi Al No white vertical line on the rostral or mental silus A2 A white vertical line on the rostral* and mental (fig. 3) Bl Scale rows usually 25; ventrals 150 or more; sub- caudals in males 28 or fewer, in females 25 or fewer; body blotches 25 or fewer willardi B2 Scale rows more than 25; ventrals fewer than 150; subcaudals in males more than 28, in females more than 25; body blotches more than 25 meridionalis Acknowledgments The following individuals and institutions have kindly permitted me to examine the specimens of C. willardi contained in their collections: Mr. Charles M. Bogcrt, American Museum of Natural History; the late Dr. Charles T. Vorhies, University of Arizona; Mr. Joseph R. Slevin, Cali- fornia Academy of Sciences; Mr. M. Graham Netting, Carnegie Museum; Dr. Howard K. Gloyd, Chicago Academy of Sciences; Messrs. Karl P. Schmidt and Clifford H. Pope, Chicago Natural History Museum; Messrs. Arthur Loveridge and Benjamin Shreve, Museum of Comparative Zoology, Harvard University; Dr. Robert C. Stebbins and Mr. Wade Fox, Museum of Vertebrate Zoology, University of California; E>r. Edward H. Taylor, University of Kansas; Dr. James A. Peters, University of Michigan; Dr. Donald D. Brand, University of Texas; Dr. Emmett R. Dunn, Academy of Natural Sciences of Philadelphia; Mr. Carl F. Kauffeld, Staten Island Zoolog- ical Society; and Dr. Waldo L. Schmitt and Dr. Doris M. Cochran, United States National Museum. I am grateful to Dr. Carl L. Hubbs, and Messrs. C. B. Perkins and Charles E. Shaw for editorial criticisms and suggestions. I was assisted by Charles E. Shaw and Richard Schwenkmeyer in making scale counts. The sketches were prepared by Mr. Norman Bilderback; and the map and photo- graph by Mr. Leslie C. Kobler. Summary The ridge-nosed ratdesnake, Crotalus willardi Meek, 1905, a distinctive montane species restricted to the Sierra Madre Occidental, is now divided into 3 subspecies, based on differences in lepidosis and pattern. These are C. w. wil- lardi of the Santa Rita and Huachuca ranges of southeastern Arizona; and two new subspecies: C. w. silus, of northeastern Sonora and western Chihuahua, and C. w. meridionalis, of western Durango and southwestern Zacatecas, Mexico. * Sometimes abraded on specimens that have been long in captivity. Klauber — Ridge-Nosed Rattlesnake 137 Bibliography Hartman, Frank A. 1911. Description of a Little-lcnown Rattlesnake, Crotalus willardi, from Arizona. Proc. U. S. Nat. Mus., vol. 39, no. 1800, pp. 569-570. Kauffeld, Carl F. 1943. Field Notes on Some Arizona Reptiles and Amphibians. Amer. Midi. Nat., vol. 29, no. 2, pp. 342-358. Klauber, L. M. 1936. A Key to the Rattlesnakes with Summary of Characteristics. Trans. San Diego Soc. Nat. Hist., vol. 8, no. 20, pp. 185-276. Meek, Seth E. 1905. An Annotated List of the Collection of Reptiles from Southern California and Northern Lower California. Field Col. Mus. pub. 104, Zo6l. Ser., vol. 7, no. 1, pp. 1-19. Stejneger, Leonhard and Barbour, Thomas 1917. A Check List of North American Amphibians and Reptiles. [Ed. 1]. Cambridge, pp. iv, 5-125. Swarth, H. S. 1921. The Type Locality of Crotalus willardi Meek. Copeia, no. 100, p. 83. Klauber— Ridge-Nosed Rattlesnake Plate 8 Fig. 1. Cro talus willardi willardi Arizona Ridge-nosed Rattlesnake. Adult male from Ramsey Canyon, Huachuca Mountains, Cochise County, Arizona. Fig. 2. Cross-section of head showing internasal ridge. Fig. 3. Snout showing vertical light line on rostral and mental. «e» o :i MUS. COMP. ZO&L UBaARt HAY 14 I9S TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 9, pp. 141-204, plates 9-10 THE SHOVEL NOSED SNAKE, CHIONACTIS WITH DESCRIPTIONS OF TWO NEW SUBSPECIES Laurence M. Klauber Curator of Reptiles and A mphibians, San Diego Society of Natural History SAN DIEGO, CALIFORNIA Printed for the Society April 30, 1951 COMMITTEE ON PUBLICATION Joshua L. Baily Carl L. Hubbs Laurence M. Klauber MUX. COMP. ZOGl UmRY MAY 14 19S1 TABLE OF CONTENTS Page ...145 Introduction Historical Summary 146 Material 147 Character Variation and Sexual Dimorphism 148 Genus Chionactis 151 Species and Subspecies Descriptions 153 Chionactis occipitalis occipitalis (Hallowell) 153 Chionactis occipitalis annulata (Baird) 161 Chionactis occipitalis klauberi (Stickel) 170 Chionactis occipitalis talpina, subsp. nov 172 Chionactis palarostris palarostris (Klauber) 175 Chionactis palarostris organica, subsp. nov 178 General Relationships and Intrageneric Trends 181 Phylogeny 183 Ecology, Life History, and Habits 184 Ecological and Temperature Preferences 184 Relative Populations 190 Locomotion 191 Fossorial Activity 192 Food 194 Reproduction 194 Defense 194 Enemies 195 A Key to the Genus Chionactis 195 Acknowledgments 197 Summary 197 Bibliography 198 i THE SHOVEL-NOSED SNAKE, CHIONACTIS, WITH DESCRIPTIONS OF TWO NEW SUBSPECIES BY Laurence M. Klauber Curator of Reptiles and Amphibians, San Dief^o Society of Natural History Introduction The shovel-nosed snakes of the genus Chwnactis are small, noc- turnal inhabitants of the deserts of the southwestern United States and northwestern Mexico. Like many other desert snakes, they were once thought to be rare, but within the past 20 years it has been found that they are quite common in many places. This discovery has resulted from a new method of collecting, that of driving on paved desert roads at night, when the nocturnal snakes crossing the highway are brightly set off against the dark background of the pavement by the headlights of the car, and may readily be caught. Through this device, the Chionactis material in collections has greatly increased in recent years, so that it is now possible to determine subspecific differences with greater assurance than heretofore; and specimens from areas not previously known to be within the range of the genus are at hand. For some years past, Mr. William H. Stickel has had under way a survey of the snakes of the genus Sonora. The shovel-nosed snakes were considered to belong to this genus until Mr. Stickel pointed out differences from Sonora warranting their generic separation, and involving the revival for them of the genus Chionactis Cope. In the course of his study of Sonora, Mr. Stickel made important advances in our knowledge of Chionactis. Notwithstanding this work, he courteously agreed to confine his activity to the much more widespread genus Sonora, thus permitting me to undertake this study of Chionactis without infringing on his program. This generosity on Mr. Stickel's part is much appreciated. 146 San Diego Society of Natural History Historical Summary The first shovel-nosed snake to be classified was collected by A. L. Heermann, the surgeon and naturalist who accompanied Lieut. R. S. William- son's railroad-route exploring expedition in California in 1853. It was taken in the Mojave Desert, the exact locality being unrecorded. Although the type specimen is no longer available, a drawing of it is at hand (Hallowell, 1859, plate iv, fig. 2). Judging from the narrow crossbars, relative to the width of the interspaces, and the fact that the posterior bars do not cross the ventrum, we may presume it to have come from the western part of the Desert, no doubt from the general area of the Mojave River, in western San Bernardino County, where several other reptiles taken by the expedition are known to have been collected. The new species was first described by Hallowell in 1854 (p. 95) as Rhinostoma ocapitale. The genus Rhinostoma had been set up by Fitzinger in 1826 (p. 56) with the type species rufofiisca from Dominica. In 1856, Hallowell decided that the new snake was not congeneric with Fitzinger's Rhinostoma, for reasons that he made clear later (1859, p. 16), and therefore he set up the new genus Lamprosoma for the shovel-nosed snake (Hallowell, 1856, p. 310). Subsequently, Cope (1860, p. 241) pointed out that the name Lamprosoma was preoccupied by a genus of Coleoptera and therefore substi- tuted the name Chionactis, "having allusion to the refulgent whiteness of the scales" {Chion Gr. snow; actis Gr. a ray or beam of light) . Garman (1883, pp. 91, 164) placed the shovel-nosed snakes in the genus Contia, but this assignment did not receive wide acceptance. Subsequently, Van Denburgh and Slevin (1913, p. 412) combined them with the ground snakes of the genus Sonora and this view was almost universally adopted for the next 30 years. Finally, Stickel (1943, p. 122) revived Chionactis, re-segregating the shovel-nosed snakes from Sonora by reason of a number of fundamental character differences, including the nasal valve, a feature of Chionactis that he himself had discovered, as well as the angled abdomen, the spadelike snout, fewer maxillary teeth, and hemipenial divergences. To the present writer, Stickel's re-establishment of Chionactis appears fully justified. As has been stated, the first spade-nosed snake was collected in the Mojave Desert in 1853. To this Hallowell (1854, p. 95) assigned the specific name occipitalis. The earliest specimens from the Colorado Desert to reach scientific collections were 2 taken by A. Schott, while on the Mexican Boundary Survey in 1855; these became no. 2105* in the collection of the U. S. National Museum (Smithsonian Institution) . They may have come from anywhere along the boundary, in what is now Imperial County, California, between the Colorado River on the east and the foot of the coastal mountains to the westward. While Kennicott (in Baird, 1859a, p. 21) considered them to * Although listed as nos. 2105-6 in the report, both specimens were apparently catalogued under no. 2105. Klauber— Shovel-Nosed Snakes 147 belong to Hallowell's species occipitalis, Baird himself (p. 22) pointed out certain differences, particularly the completeness of the body rings, and assigned them to a new species that he called Lamprosoma annulatum. Since then the Colorado Desert form has had a varied career, sometimes being viewed as a subspecies of occipitalis, as first suggested by Cope in 1875 (p. 36), but usually being considered synonymous with the Mojave Desert form. Finally, Stickel (1941, p. 135; 1943, p. 122), aided by new material, and making a more thorough study than any previously undertaken, revived annulata as a subspecies of occipitalis. Two additional forms have been described recently. In 1937 (p. 363) I named palarostris from central Sonora as a full species, although this was subsequendy relegated to subspecific status by Stickel (1941, p. 137). Finally, Stickel (1941, p. 138) described the subspecies C. o. klauberi from the Tucson area of Arizona. Thus four subspecies of Chionactis occipitalis are currently recognized: occipitalis, annulata, palarostris, and klauberi. Material The present study is based on specimens geographically derived as follows: San Diego County Imperial County Riverside County San Bernardino County Kern County Los Angeles County Inyo County 345 58 126 94 14 2 3 Total California 642 Clark County Nye County Esmeralda County 5 1 2 Total Nevada 8 Yuma County Mojave County Pinal County Maricopa County - Pima County 45 2 10 11 27 Total Arizona 95 Total Sonora 19 Uncertain 6* Grand Total 770 "Including one doubtfully stated to be from Utah. 148 San Diego Society of Natural History Of these specimens, about 60 per cent are contained in my own collection. I have had some field experience with shovel-nosed snakes, and have seen several hundred live specimens. Character Variation and Sexual Dimorphism Subspecific differences in Chionactis are largely confined to pattern and color, the squamation being relatively constant. Even the pattern is much less variable than in most genera of snakes. There is little doubt that this con- sistency is related to the circumstance that the ecological range of the genus is one of the most restricted of any snake inhabiting the southwestern United States. As a preliminary to investigating specific and subspecific differences, I have made it a practice to determine the extent of intrasubspecific variations, using, as a basis of study, the largest available homogeneous population, in this case that from eastern San Diego County, California. This population, as a matter of fact, is not completely homogeneous, since the area inhabited, although including only the desert section of the county, varies somewhat in physiographic character, and likewise in altitude (from 85 to 2300 feet) . Also, the population is not pure annulata, but shows some occipitalis influences. Nevertheless, I have used it in determining degrees of variation and the extent of sexual dimorphism, since this is the only series containing a really adequate number of females. The data on these specimens are as follows: Number of specimens, males 234 females 106 Coefficients of variation, per cent Ventrals, males 2.12 females 2.06 Subcaudals, males 5.38 females 6.23 Body bands, males 9.40 females 10.64 Tail rings, males 12.69 females 13.88 Coefficients of sexual divergence, per cent* Ventrals -5.94 Subcaudals 12.00 Body bands -9.43 Tail rings 5.23 Correlation coefficient between body bands and tail rings, per cent 42.2 These results are quite consistent with those disclosed ■ i" other genera of snakes, for the coefficients of variation of the ventrals in homogeneous series *Positive if the males are higher, negative if the females are higher. Klauber— Shovel-Nosed Snakes 149 nearly always adhere closely to 2 per cent, the subcaudals to 5 or 6 per cent, the body bands or blotches to 6 to 10 per cent, and the tail rings or spots to 12 per cent or more. The sexual differences are all highly significant, but with an order of significance corresponding inversely to their coefficients of variation — that is, the ventral-scale dimorphism is most significant and that of the tail rings least. In the subspecific descriptions and discussions that follow, I shall separate the sexes in presenting the statistics of ventrals and subcaudals, but shall seldom do so in pointing out differences in the number of body bands and tail rings, since sexual dimorphism in these characters is not statistically useful unless many specimens of the two populations to be compared are available, and adequate numbers of females are often lacking. Pattern differences in form and color are, in many cases, as useful as key characters as is the numerical variability in the number of bands across the body. The scale rows are nearly always 15-15-15, and the ventrals, though showing some territorial variation, involve such inadequate differences as to be of little value in keys. The subcaudals also exhibit some subspecific differences, but these are not extensive, and, as key characters, suffer likewise from the fact that this genus does not possess a conspicuous terminal cone, and hence it is not always possible to determine with certainty whether the tail is complete. Neither the labials nor the other head scales are subject to any subspecific differences, at least none that I have found. With this uniformity in squamation, taxonomic differentiation is largely reduced to a consideration of pattern, which, fortunately, is subject to much more variation, yet has sufficient constancy in each area to afford valuable differential characters. Essentially, the Chionactts head pattern comprises a dark transverse crescent (black to brown) engaging the parietal region, and thence passing down and forward along the sides until the horns of the crescent terminate in the region of the nostril or the eye. The light ground color of the head is substantially the same as that on the body. The body pattern involves a series of dorsal crossbands of black to brown, on a light ground color varying from white to cream or yellow. These cross- bands may or may not extend to the ventrum to form partial or complete rings. The posterior bands much more often form complete rings than the anterior. The proportion that is complete is an important subspecific criterion. The dorsal bands usually narrow laterally; however, those that cross the belly re-widen ventrally. This is particularly true of the posterior bands. In most specimens of Chionactis the widths of the bands are surprisingly uniform, especially at mid-body and when measured middorsally. One may take a pair of dividers, set it for a single band, and find that it measures accurately the widths of a number of adjacent bands. The same is true of the interspaces. Bilateral asymmetry in pattern is rare in Chionactis, specimens with U- or Y-shaped blotches, such as are highly characteristic of Hypsiglena or Arizona, 150 San Diego Society of Natural History see a being evident in less than 1 per cent of the shovel-noses. I have yet to Chionactis with non-confluent spots on the two sides, a condition frequently observed in the two genera above mentioned. The middorsal joining of two successive blotches to form dumbbells is likewise rare in Chionactis. Although well defined, the tail rings, which are nearly always complete ventrally, are irtferior to the body bands as differential characters for two reasons: First, there is uncertainty as to the completeness of the tail; and, secondly, doubt often exists as to whether the small dark spot that usually terminates the tail should, or should not, be counted as a ring. For this reason, though I have given some tail-ring statistics, I prefer to omit them in deriving key characters. There is seldom a questionable blotch at the base of the tail, which cannot be allocated to the body or tail series, depending on its position with respect to the anal plate. Although the proportion of dorsal bands that encircle the body is a diagnostic character of value, it is often difficult to tell whether a body ring is, or IS not, complete. This is particularly true laterally, where the dark Dior may be carried only by the scale edges, and observers might differ as to whether the thread-like connecting maculations are, or are not, interrupted. In other specimens these lateral connections are punctated and faint. I have found the presence of a dark mark (however incomplete) on the ventrum opposite the position of a dorsal band, to be a much more objective criterion than the completeness of a ring. Stickel (1941, p. 138) found that occipitalis has, on the average, more crossbands than annulata; also, that the anterior rings in occipitalis are much less likely to be complete ventrally than the corresponding rings of aiumlata. To take advantage of both differences in a single character, he introduced the combination: "Total number of bands (on body and tail) plus the number of body bands not entirely encircling the body." In order to eliminate the uncertainty as to the completeness of the tail, to make incomplete- tailed specimens useful, and to avoid decisions involving doubtfully complete rings, I prefer to substitute the following combination character: "Total number of bands on the body plus the number of unmarked band positions on the ventrum." This criterion sharpens the differences as well as does Stickel's and is more objective in application. Color is important in the classification of Chionactis subspecies, but unfortunately is too evanescent in preserved material to be as fully useful as might be hoped. Three colors are involved: the ground color, the band color, and the color of any suffusion or maculations in the interspaces. The ground color may vary in live specimens from white to various shades of cream, yellow, or buff, but all of these hues usually revert to white or nearly white in preservative. The black or brown of the bands is also invariably lightened by time and the preservative, particularly if the specimen is not kept in darkness. But, as far as I know, neither black nor brown ever disappears, and thus the shape and arrangement of bands and maculations may always be determined, though the original hues may have changed. It may be noted that in this genus, as in many others, the dark marks are darker and more contrasting with the ground color in juveniles than in adults. Klauber — Shovel-Nosed Snakes 151 The most important and disappointing changes in the Chionactis colors take place in such suffusions as may color the interspaces, particularly their centers. These suffusions, often so definitely outlined that they may be deemed secondary blotches or saddles, may be yellow, orange, vermilion, salmon, or deep red. All of these colors disappear in preservative, the yellows earlier in formalin, and the reds earlier in alcohol, but in either preservative they vanish in time, leaving hardly a trace. A useful criterion is thus lost, even though the specimens have been kept in darkness. Black or brown interspace maculations constitute another type of markings. These are important, both with respect to their extent and character; they may be dorsal, lateral, or ventral, or all three, and they may engage scale centers, scale edges, or both. As far as I know, these maculations never dis- appear in preservative, so that one may always know their character, if not their original color, and thus this valuable criterion is in part retted. When the ventrum is spotted, an additional type of marking sometimes present, it is usually not difficult to recognize the first blotch that is the ventral analogue of a dorsal band, this blotch being independent of the general spotting. Dorsal bands are usually of solid color and even-edged. However, in the snakes from some areas the dark part of any scale that is half, or less than half, dark is lighter than the adjacent, completely dark scales. The blotch edge is thus, given a serrated appearance. In some specimens the dark color of a band is carried beyond the normal edge of the band along the interstices between scales. In determining the relative widths of dorsal bands and interspaces, I have usually employed a draftsman's bow dividers, measuring a sample series of 3 or 4 bands at mid-body. Counting scales end-to-end may also be useful, particularly if adults are to be compared with juveniles. It will sometimes be found of interest to study the position and extent of the head crescent, particularly its forward edge with relation to the supra- oculars and frontal, and whether posteriorly it extends beyond the parietals, or does not completely cover them. There is also some difference in the extent of the horns of the crescent, which in some subspecies terminate at the eye, but in others reach as far forward as the nasal scales. Another character sometimes useful is the number of light scales (end-to-end) between the pos- terior edge of the head crescent and the anterior edge of the first body band. Genus CHIONACTIS Lamprosoma Hallowell, Proc. Acad. Nat. Sci. Phila., vol. 8, p. 310, Dec. 1856. Type species occip'ttale. Chionactis Cope, Proc. Acad. Nat. Sci. Phila., vol. 12, p. 241, June 1860. Type species occipitale. The snakes of this colubrid genus are small and of moderate body pro- portions. The maximum length attained in any area is somewhat over 400 mm. (15% in.). The head is only slightly distinct. The pupil of the eye is round. The tail length varies from 19 to 22 per cent of the length over-all in the 152 San Diego Society of Natural History adult males, and from 15 to 19 per cent in the females. The snout is quite sharp, with a horizontal cutting edge to facilitate progress through sand; other modifications favorable to an arenaceous existence are the presence of a nasal valve, and a protrusion at the tip of the mental plate which serves to close the apperture through which the tongue is thrust out. There is a sharp angle, represented by a bend in the ventrals, between the abdominal crawling surface and the sides of the body. The rostral is sharply recurved and pointed posteriorly, but does not separate the internasals. It is somewhat raised above the adjacent plates and is deeply cleft on the under side. The plates on the crown follow the usual colubrid pattern. The parietals are the largest of the head plates, followed by the frontal. The nostril is placed in an undivided nasal, usually slightly forward and above its center. A small loreal, longer than high, is usually present; it is widest anteriorly. There is ordinarily a single preocular and 2 postoculars, of which the upper is the larger. The temporals are usually 1 followed by 2. The upper labials almost always number 7, the next to the last being the largest; the third and fourth touch the eye. The lower jaw is deeply inset. The lower labials usually number 7; the fourth is the largest. The quadrangular mental has an anterior protuberance that is centrally ridged, with a depression on either side. The anterior genials meet on the median line; the posterior, much reduced in size, are separated by gular scales. The dorsal scales are smooth and rarely in other than 15-15-15 rows. Single apical scale pits are faintly evident on some scales, particularly on the sides of the body. The ventrals vary from 141 to 165 in the males, and 153 to 178 in the females. The anal is divided. The subcaudals range from 39 to 57 in the males, and 34 to 51 in the females. Normally all subcaudals are divided. The pattern comprises a series of black or brown crossbands on a white to yellow ground color. These bands vary from 10 to 40 on the body and from 3 to 13 on the tail. The bands narrow laterally; some completely encircle the body, widening again on the ventrum. A characteristic black or brown crescent on the head usually engages the parietals, with the horns carried forward on the sides to the eyes or even to the nostrils. The dorsal bands may be either wider or narrower than the interspaces. The interspaces may be marked dorsally in various ways, either suffused with pink or red, or maculated with black or brown, so that in some specimens the secondary blotches or saddles are almost as definite as the primary. The ventrum is sometimes spotted, in addition to being marked by such bands as may be carried to the under surface. There are 3 maxillary foramina and the maxillary teeth number 8+3 or 9+3 (Stickel). The hemipenis has a distal section of fine and almost uniform reticulations or flounces. The outermost part comprises two low mounds, otherwise the organ is undivided. The basal spinous section contains 2 large spines, one on each side of the sulcus, and about 30 much smaller spines. The transition from reticulations to spines is quite sharp. The sulcus passes diagonally from the spinous base across the reticulations to the top of Klauber — Shovel-Nosed Snakes 153 one of the terminal mounds. It ends among the reticulations, there being no smooth area. In the southwest, Chionactis is most often confused with Chilomemscus, or with some of the crossbanded subspecies of Sonora. The separation from the first-named is simple, for in Chtlomenisciis the nasals are merged with the internasals, whereas in Chionactis they are not. Chilomeniscus almost invariably has 13 scale rows and Chionactis 15. From Sonora, Chionactis differs in having a flatter snout, an angled abdomen, nasal valves, a protrusion on the mental, and a proportionately shorter tail. The bands on Sonora are less definitely outlined. In practice I have found that these characters serve readily to segregate Chionactis from banded forms of Sonora — some subspecies of Sonora are striped or unicolor — except when the specimens are badly preserved or dried juveniles. Species and Subspecies Descriptions Chionactis occipitalis occipitalis (Hallowell) Mojave Desert Shovel-nosed Snake Plate 9, fig. 1 1854. Rhinostoma occipitale Hallowell, Proc. Acad. Nat. Sci. Phila., vol. 7, p. 95. Type specimen unknown (figured in Hallowell, 1859, plate 4, figs. 2a, 2b, 2c) ; type locality Mohave Desert. 1856. Lamprosoma occipitale Hallowell, Proc. Acad. Nat. Sci. Phila., vol. 8, p. 311. 1859. Lamprosoma occipitale (part) Baird, Gen. Rept. Zool. Pac. R.R. Routes, vol. 10, part 3, p. 16. 1860. Chionactis occipitalis Cope, Proc. Acad; Nat. Sci. Phila., vol. 12, p. 241. 1870. Chionactis occipitalis (part) Cooper, Proc. Calif. Acad. Sci., vol. 4, part 2, p. 66. 1875. Chionactis occipitalis occipitalis Cope, Bull. U. S. Nat. Mus., no. 1, p. 35. 1883. Contia occipitalis var. occipitalis Garman, Mem. Mus. Comp. Zool,, vol. 8, no. 3, pp. 91, 164. 1894. Contia occipitalis (part) Botilenger, Cat. Snakes Brit. Mus., vol. 2, p. 266. 1901. Contia occipitale (part) Brown, Proc. Acad. Nat. Sci. Phila., vol. 53, part 1, p. 68. 1916. Sonora occipitalis Camp, Univ. Calif. Pubs, in Zool., vol. 12, no. 17, pp. 503-544. 1917. Sonora occipitalis (part) Grinnell and Camp, Univ. Calif. Pubs, m Zobl., vol. 17, no. 10, p. 182. 1941. Sonora occipitalis occipitalis Stickel, Bull. Chicago Acad. Sci., vol. 6, no. 7, p. 135. 154 San Diego Society of Natural History Diagnosis. — A subspecies of occipitalis differing from the subspecies of palarostris in having more crossbands on the body and tail. It is distinguishable from the other occipitalis subspecies as follows: from klciuberi and talpina in usually lacking dark maculations in the interspaces between the primary bands or, at least, in having such spots as may exist in these spaces generally small, scattered, and restricted to scale edges; and from annulata in having brown, rather than black, or nearly black, crossbands, and in having more crossbands, fewer of which, however, are carried to the ventrum anteriorly. Notnenclatorial and Systematic Probleiiis. — The type locality of this subspecies is not known definitely, having been recorded only as the "Mohave Desert" (Hallowell, 1854, p. 95). Whether the type specimen is still in existence is not known; fortunately, it was figured by Hallowell rather com- pletely in a later paper (1859, plate 4, figs. 2a-2c) . One may hazard the guess, based on the known itinerary of the collecting expedition and the character of the crossbands — their narrowness and their failure to cross the ventrum — that it probably came from the western Mojave, in the general vicinity of the Mojave River in the southwestern part of what is now San Bernardino County, California. The most interesting problem with respect to the distribution and variations in this subspecies is the unusual relationship between the populations of the Colorado and Mojave deserts. In most reptile species, if a difference is evident between such populations, the Coachella Valley (Riverside County) population is found to belong to the southern (Colorado Desert), rather than to the northern (Mojave Desert), form, as would be expected, for the Coachella Valley is a northwesterly extension of the Colorado Desert. Of such a territorial pattern, the separation of the Mojave Desert forms Pituophis catetiifer deserticola and Crotalus cerastes cerastes from the corresponding Colorado Desert P. c. affinis and C. c. laterorepens are examples. In both instances the Coachella Valley inhabitants belong to the southern race. But in Chionactis, the Mojave Desert form occupies the Coachella Valley and the eastern part of Riverside County as well. Under such circumstances one should expect at least some differences between the Mojave and Coachella populations, and I had hoped to find a character that might separate the two and thus interpose an additional subspecies between occipitalis and annulata. Average differences have been disclosed, but the overlaps are such that segregation does not seem advisable, at least there is no justification on the basis of any character that I have been able to discover. Such differences as there are will be discussed later. Material. — The description of the subspecies occipitalis that follows is based on about 240 specimens; however, there is far from an adequate repre- sentation of all areas, since much more than half the material is from two sections, the western Mojave Desert and the Coachella Valley. The eastern and extreme southeastern sections of the range are poorly represented in collections. Description of the Subspecies. — This widespread subspecies, which inhabits more than half of the total area assigned to the species occipitalis as a whole, is quite variable in pattern, but much less so in squamation. Klauber — Shovel-Nosed Snakes 155 The largest specimens that I have seen were 369 mm. (141/2 in.) long; I have had both a male and female of this length. In the western Mojave, males 320 mm. (121/2 in.) long are not uncommon; in the Coachella Valley they frequently reach 340 mm. (131/2 in.). Having in mind the fact that the males in collections usually outnumber the females by 2 to 1 or more, I should say that there is no conspicuous difference between the sexes in the ultimate length attained. The shortest juveniles seen were Mojave specimens 121 and 122 mm. (4% in.) long. The ratio of the tail length to the length over-all in adult males ranges from 17 to 20 per cent, with a mean of 18.5 per cent; and in the adult females from 15.5 to 17.5 per cent, with a mean of 16.7 per cent. Juveniles have somewhat shorter tails, proportionately, than adults. The scale rows number 15 at mid-body, with only one exception among all the specimens of this subspecies that I have examined. The aberrant specimen was one from the Coachella Valley having 14 rows. There are frequently 17 rows on the neck, but they are reduced to 15 at or near one head-length behind the head. Similarly, when there are, rarely, 13 rows just before the anal plate, this number persists for only a short distance. The ventrals range from 146 to 165 in the males, mean 155.9; and from 154 to 176 in the females, mean 165.9. The subcaudals vary from 39 to 50* in the males, mean 44.7; and 37 to 48 in the females, mean 41.9. Occasionally a few subcaudals at the base of the tail are undivided. The supralabials are normally 1-1, less than 2 per cent having 6. The third and fourth touch the orbit and the sixth is the largest. The infralabials are usually 7 on each side, but occasionally number 6 or 8. Because of the position of the last infralabials, it is sometimes difficult to decide whether or not the last scale should be included. There is usually a single loreal on each side; however, about 5 per cent of the specimens have one or both loreals fused with the prefrontals, and one specimen has 2 loreals on one side. The preoculars are rarely divided (about 1/2 of 1 per cent), and a very few specimens have single or 3 postoculars, instead of the customary 2. The usual temporal formula is 1+2, but some specimens (about 2 per cent) have 1 + 1, and still fewer have 1+3 or 2+2. The crossbands on the body vary from 25 to 40, mean 31.2; and on the tail from 6 to- 13, mean 9.2. The bands may be wider than, equal to, or narrower than the interspaces; there is some territorial variation in this character, as is discussed below. Spotting in the interspaces is rather infrequent; when present it usually takes the form of dark edges on some of the lateral scales. Ventral spotting between the positions opposite the dorsal bands is unusual but occasionally noted. This subspecies is characterized by the relative infrequency with which the anterior dorsal crossbands become complete rings, or involve correspondingly located dark blotches on the ventrum. The contrary is true in anmdata. As annulata also has fewer bands than occipitalis, the combination criterion, dorsal bands plus unmarked blotch positions on the ventrum, serves as the best ^Omitting two aberrants with 35 and 54, respectively. 156 San Diego Society of Natural History numerical character in differentiating the subspecies. The statistics of occipitali\ with respect to this character are: range 31 to 80, mean 54.7. About 20 per cent of the specimens number 44 or fewer, and thus fall within what may be considered the annulata range (p. 164) . Essentially, occipitalis is a snake with brown crossbands on a white or. cream ground color, at least, this is the appearance of preserved specimens. The band color may vary from medium- to dark-brown. Rarely does it give the impression of being black, except in juveniles. Dorsally the bands do not differ greatly from the interspaces in width — they may be either somewhat wider or narrower. Laterally the bands narrow conspicuously, usually to terminate on the first or second row of scales above the ventrals. Even pos- teriorly, where they are often complete, the rings narrow laterally, although they may re-widen on the ventrum. The band color is darker dorsally than that it is below. The blotches are often even-edged, but in many specimens thin threads of the dark color of the blotches run out into the interspaces along the interstices between scales. Also, on the edges of the bands, scales that are partly light and partly dark often are not as dark in their dark section as are the adjacent scales that are entirely dark, thus giving an effect of serrations. In such partly dark scales the dark color consists of punctations. The head crescent usually engages only the posterior halves of the frontal and supraoculars. The lateral horns are not often carried farther forward than the eye. Regardless of whether the ventrum is marked on the body, the tail bands are nearly always complete rings. The anal, however, is rarely marked, even though there may be a dorsal band immediately above it. As to live specimens, the following Ridgway colors are typical of adults from the western Mojave Desert and from the Coachella Valley: Western Mojave Desert Coachella Valley Snout Sea-foam Green Olive Buff Head crescent Blackish Brown ( 1 ) * Bone Brown Dorsal bands Aniline Black Bone Brown Interspace centers Ochraceous-Buff Pale Chalcedony Yellow Interspace edges Marguerite Yellow White Ventral bands Light Seal Brown Natal Brown Ventral interspaces Ivory Yellow Pale Olive Buff The snout is usually darker and slighdy greener than the interspaces on the body. The ventral surfaces are somewhat lighter than the dorsal, a characteristic of both the bands and the interspaces. *The numeral is a part of the color name, as Ridgway distinguishes between 3 tones of blackish brown. Klauber— Shovel-Nosed Snakes 157 In general, these live snakes may be described as having brown crossbands on a yellow or orange ground color. Intrasubspecijic Trends. — Of the subspecies occipitalis only two series that are territorially concentrated are available in sufficient numbers to justify numerical comparisons. These are a Coachella Valley series from Riverside County, and a series from the western Mojave Desert, within a radius of about 40 miles of Kramer Junction, San Bernardino County. The first series comprises 100 specimens, the second 65; males greatly predominate in both. The statistics follow: Ventrals, males, range mean females, range mean Subcaudals, males, range mean females, range mean Crossbands on body, range mean Crossbands on tail, range mean Crossbands on body plus unmarked ventral positions, range mean Per cent of specimens with no ventral blotches on body 39.5 29.3 It will be observed that the numerical differences between these series are relatively minor. Only in the percentage of snakes without blotches on the ventrum is there a distinct difference, the Mojave series being less maculated. The same is true if we take the proportion of the dorsal band positions that have opposite marks on the ventrum. In the Coachella series, of those with some ventral marks, specimens having from 40 to 70 per cent of the ventrum clear approximately equal in frequency those having 70 to 99 per cent clear. But in the Mojave series nearly all fall in the 70 to 99 per cent class, and very few below. Thus, in these characters indicating the relative absence of ventral maculations, the Mojave specimens are clearer and more sharply differentiated from antudcita than are those of the Coachella Valley. As to colors, the most important, from the standpoint of general appear- ance, are obviously those of the dorsal bands and the centers of the interspaces. The following Ridgway colors were noted in 20 live specimens from the western Mojave Desert and 12 from the Coachella Valley: •stern Mojave Coachella Valley desert series series 14a-162 151-165 155.3 156.8 159-169 160-176 163.0 167.4 40-49 41-49 44.3 44.7 37-43 39-48 41.0 42.2 24-39 25-40 29.8 32.2 6-11 6-12 8.7 9.5 37-78 31-76 57.6 59.0 158 San Diego Society of Natural History Dorsal Crossbands Western Mojave Desert Natal Brown Blackish Brown (1)* Blackish Brown (3)* Sooty Black Chaetura Black Plumbeus Black Fuscous Black Centers of Interspaces Coachella Valley Light Seal Brown Seal Brown Hay's Brown Bone Brown Dusky Brown Warm Sepia Blackish Brown (2) * Warm Blackish Brown Aniline Black Black White Deep Colonial Buff Cinnamon Buff Buff Yellow Sea-foam Yellow Pale Chalcedony Yellow Primrose Yellow Chamois Colonial Buff Salmon Buff Warm Buff Apricot Yellow Reed Yellow Honey Yellow Naples Yellow Straw Yellow Salmon Orange Ochraceous Orange Capucine Orange Vinaceous Cinnamon Ochraceous Salmon It will be observed that there is not much difference in the depth of the browns and dark-browns of the bands, but the interspaces are a brighter yellow in the Mojave snakes, with more of a tendency toward orange. The dark bands, middorsally, are narrower than the interspaces more frequently in the Mojave specimens than in those from the Coachella Valley, for, in the latter, the dark bands are wider than, or equal to, the interspaces more often than narrower. There is no conspicuous difference in the number of uncolored dorsal scales between the head crescent and the first dark band on the body; in both series the range is from 3 to 6, many specimens having 4. In both series the head crescent usually extends to the posterior edge of the parietals, or slightly beyond; and, in both, the lower preoculars are ordinarily darkened by the horns of the crescent. Dark scale edges in the lateral inter- ^Different colors in Ridgway's notation. Klauber — Shovel-Nosed Snakes 159 spaces are of moderate frequency in both series; and in each an occasional specimen is found showing considerable interspace maculations, a klauberi or talpina tendency. A single specimen from an altitude of 3000 ft. on the Palms-to-Pines highway, above the Coachella Valley, has particularly wide dark bands. There is no marked difference between the Mojave and Coachella series in the relative tail lengths. Both have shorter tails than annulata. The absence of important differences between the western Mojave and Coachella groups is the more surprising in view of the differences in their surroundings. The Coachella is a flatter, sandier basin, much of it below sea level. The western Mojave is rockier; the mean altitude of the section from which the example series came is about 2200 ft. It has a longer, colder winter. Both areas are extremely windy at night in the spring. It is to be presumed that occipitalis gained access to the Coachella Valley from the Mojave via the Morongo Valley to the north, or via the pass between the Cottonwood and Orocopia mountains to the east. Since the western Mojave and Coachella snakes more nearly resemble each other, in their lack of ventral maculations, than either resembles the snakes farther east, the Morongo route appears the more logical. One may presume that the differences between the Colorado Desert annulata and Mojave Desert occipitalis are either not importantly fitted to ecological necessities, or that the invasion of the Coachella Valley by occipitalis has been comparatively recent. I judge the former to be the real reason why the Coachella snakes are not more like the snakes of the lower Colorado Desert, although I do think that the meeting of occipitalis and annulata along the westerly shore of the ancient Lake LeConte took place rather recently, possibly at the time of its latest subsidence. Specimens from the vicinity of Boulder City, Clark County, Nevada, and from eastern San Bernardino and Riverside counties in California, are not available in sufficient numbers to permit a detailed analysis of territorial trends. They average slightly lower in the number of body bands, and the ventrums are more maculated — that is, the ventral blotches begin nearer the head, and fewer specimens are without at least some ventral marks on the body. In this character they show a tendency toward annulata. In the Clark County speci- mens, the dorsal bands are wider than the interspaces, whereas the contrary is true in those from the western Mojave. It may be concluded that the^ population of occipitalis most widely differentiated from annulata, as well as from talpina, klctuberi, and palarostris, is that inhabiting the extreme western section of the range, that is, the Mojave-Kramer-Adelanto-Palmdale quadrangle in the western part of the Mojave Desert. Geographically this is quite logical. Suhspecific Relationships. — C. o. occipitalis intergrades with two other subspecies, talpina and annulata, and possibly with a third, klauberi. With the new subspecies talpina, it intergrades along the southern border of eastern Inyo County, California, and near the eastern edge of Nye County, Nevada, as will be discussed in more detail under talpina. 160 San Diego Society of Natural History Although many specimens of the subspecies occipitalis and annidata are at hand, and the fact that they intergrade is unquestioned, the boundaries between their ranges cannot be defined with assurance, partly because of lack of specimens from critical areas, and pardy because of the composite character of the material from eastern San Diego County. Pure occipitalis occurs in the Coachella Valley of Riverside County, and pure annulata in the Kane Spring- Westmorland section of Imperial County. The two subspecies apparendy intergrade in the neighborhood of Travertine Rock and Fish Springs in extreme northwestern Imperial County. This is logical, especially if we assume that the two populations were originally separated by prehistoric Lake LeConte, which once filled what is now the Salton Basin, and that they re-merged when the lake dwindled to leave a flat margin between its western shore and the mountains. However, the snakes of eastern San Diego County are not pure annulata, as one might expect, but show a considerable infusion of occipitalis characters. Whether this has resulted from a southward migration of occipitalis around the lower end of the Santa Rosa Mountains into the Borrego Valley, or whether these characters were developed independently in the San Diego population because of its desert-mountain habitat is uncertain. For the present, I shall consider the San Diego County population to be annulata, but of a non-typical character, and the western end of occipitalis- annulata line of intergradation to be near Travertine Rock, at the Imperial- Riverside county line. Eastward of the Salton Sea, we may assume the boundary to follow the crest of the Chocolate Mountains southeasterly to the Colorado River. Across the river, in Arizona, the dividing line may be tentatively placed at the Bill Williams River. Of the few specimens available from this section of Arizona, those from the vicinity of Parker and farther south in Yuma County are nearer annulata, while the only two from Mohave County favor occipitalis. The only two specimens now at hand from northern Maricopa County are annulata or annulata-klauberi, rather than occipitalis-klauberi intergrdes. But it is not impossible that occipitalis and klauberi may intergrade somewhere along the line Yucca-Alamo- Wickenburg, if Chionactis occurs, as it may, in that relatively inaccessible region, in which there has been little or no collecting. Range. — Chionactis o. occipitalis is found throughout the Mojave Desert of southwestern Inyo County, eastern Kern County, northeastern Los Angeles County, San Bernardino County (except the mountain and transmontane area in the southwest). Riverside County (from the desert slope of the San Jacinto Mountains eastward, including the Coachella Valley), and northeastern Imperial County (beyond the Chocolate Mountains), California; southern Clark County, Nevada; and southwestern Mohave County, Arizona. Locality Records. — CALIFORNIA: Inyo County — ^Owens Lake, Goler Canyon (Panamint Mts.) (talpina intergrade), 11 mi. s. of Shoshone (talpina intergrade); Kern County — Brown, Garlock, Mojave (also 4, 6, 11, 17 mi. e.), Muroc (also 4 mi. se.), Muroc Dry Lake, Boron (also 1, 11, 12 mi. w.); Los Angeles County — 20 mi. e. of Lancaster, 3 mi. n. of Palmdale, county line e. of Llano (about 9 mi. e.); San Bernardino County — Trona (also 2 mi. n.), 10 mi. s. of Salt Wells Jnct. ( = 5 mi. n. of Searles Sta.), Lava Mts., Klauber— Shovel-Nosed Snakes 161 Red Mt. (also 13 mi. s.), Atolia (also 3 mi. n. and 11 mi. s.), Fremont ( = 14 mi. n. of Kramer Jnct.), Kramer (also 1 mi. w.), Kramer Jnct. (U.S. 395 and U.S. 466) (also 1 mi. w. and 1, 3, 4, 6, 11, 12, 13, 14 mi. s.) Kramer Hills, Adelanto (also 1, 3, 5, 6, 7, 8, 9, 10, 12, 13, 14, 16, 17 mi. n., 7 mi. sw., and 11 mi. s.), Oro Grande (also 12 mi. n. = 1 mi. s. of Helendale), Victor- ville (also 3 mi. s.), Phelan (also 5 and 6 mi. nw.), 16 and 22 mi. s. of Shoshone (Inyo Co.) (= Ibex Pass and 7 mi. s.) {talpina intergrades) , 1 mi. e. of Leach Lake {talpma intergrade), 16 mi. e. of Camp Irwin (near Red Pass Dry Lake), Baker, Midway (also 4 mi. e. on U.S. 91), Cronise, Alvord Mts., Jim Grey, Hawes, Hinkley (also 2 mi. w.). Mace, Lenwood, Barstow (also 2, 6, mi. w., and 15 mi. sw.), Minneola, Hector, 2 mi. n. of Mt. Pisgah Crater, Arimo, Fenner (also 10 mi. w.), Goffs (also 5 and 12 mi. e.), Homer, Piute Valley, 25 mi. s. of Needles on Cal. 95, 8 mi. n. of Rose (= Vidal Jnct.), 1/^ way bet. Rose and Grommet, Rice (= BIythe Jnct.) (also 1 mi. s.), Warren Wells (= Lone Star), Twentynine Palms (also 3 mi. w. and 19 mi. e.), Monroe Dry Lake (not located); Riverside Comity—Desert Hot Springs (also 3 mi. s.), near Seven Palms, Garnet (also 4 mi. n. and 5 mi. se.), Whitewater (also 3 and 5 mi. se.), Palm Springs R.R. Sta., Palm Springs (also 4, 5, 10 mi. n., 7 mi. nw., 4 mi. e., and 10 mi. se.). Palm Springs Airport, Tahquitz Canyon, Rimlon, Thousand Palms P.O. (also 5 mi. n. and 6 mi. w.). Thousand Palms, Edom, Cathedral City (also 1 mi. e. and 2 mi. s.), Date Gardens, Palm Village (also 3 mi. e.), Palms-to-Pines Highway (Cal. 74) just above 3000 ft. contour, also 2 mi. s. of Palm Village, Indian Wells (also 2 and 3 mi. e.), sand dunes 5 mi. n. of Indian Wells, Myoma, La Quinta (also 2 mi. e.), Indio (also 2, 4, mi. n., 3 mi. e., 6 mi. s., 5, 7, 10, 11, 16 mi. nw., and 5 mi. w.), Pushawalla and Berdoo canyons (Little San Bernardino Mts.), Coachella (also 5 and 6 mi. s.). Thermal, Mecca (also 2 mi. e. and 3 mi. sw.), 10 mi. n. of Riverside-San Diego county line (= Martinez), Caleb Siding, Oasis (also 7 mi. nw.), 1 mi. nw. of Travertine Rock (annulata intergrade), 15 mi. w. of Freda (San Bernardino County), Desert Center (also 11 and 16 mi. e.), Hopkins Well (also 2 mi. w. and 7 mi. e.), 6, 7, 10, 12, 16, 17, and 22 mi. w. of BIythe, 10 mi. n. of BIythe, Camp (= Dry Camp Siding?), Banning*; Imperial County — Travertine Rock {annulata intergrade). Fish Springs {annulata intergrade). NEVADA: Clark County — near Indian Spring {talpina intergrade), Boulder City. ARIZONA: Mohave County— :Fort Mohave, Yucca, Hualpai Valley. Chionactis occipitalis annulata (Baird) Colorado Desert Shovel-nosed Snake Plate 9, dg. 2 1859. Lamprosoma ocapitale Kennicott in Baird, Reptiles of the Boundary, United States and Mexican Boundary Survey (Emory), vol. 2, p. 21. *An MCZ specimen; this locality should be considered doubtful until verified by additional specimens. 162 San Diego Society of Natural History 1859. Lamprosoma anniilatum Baird, Reptiles of the Boundary, United States and Mexican Boundary Survey (Emory), vol. 2, p. 22, plate 21. Type specimens (2) USNM 2105*; type locality Colorado Desert. 1859. Lamprosoma ocdpitale (part) Baird, Gen. Rept. Zool. Pac. R.R. Routes, vol. 10, part 3, p. 16. 1870. Chionactis occipitalis (part) Cooper, Proc. Calif. Acad. Sci., vol. 4, part 2, p. 66. 1875. Chionactis occipitalis annulata Cope, Bull. U. S. Nat. Mus., no. 1, p. 36. 1883. Contia occipitalis var. annulata Carman, Mem. Mus. Comp. Zobl., vol. 8, no. 3, pp. 91, 164. 1894. Contia occipitalis (part) Boulenger, Cat. Snakes Brit. Mus., vol. 2, p. 266. 1901. Contia ocapitale (part) Brown, Proc. Acad. Nat. Sci. Phila., vol. 53, part 1, p. 68. 1913. Sonora occipitalis Van Denburgh and Slevin, Proc. Calif. Acad. Sci., ser. 4, vol. 3, p. 412. 1917. Sonora occipitalis (part) Grinnell and Camp, Univ. Calif. Pubs, in Zo6l., vol. 17, no. 10, p. 182. 1941. Sonora occipitalis annulata Stickel, Bull. Chicago Acad. Sci., vol. 6, no. 7, p. 136. 1943. Chionactis occipitalis annulatus Stickel, Proc. Biol. Soc. Wash., vol. 56, p. 128 [= 123}. Diagnosis. — C. o. annulata is a subspecies differing from klauberi and talpina in usually lacking maculations in the interspaces between the primary dorsal bands; such spotting as there may be is usually restricted to the sides of the body and to scale edges. From the subspecies occipitalis, annulata differs in having black, rather than brown, dorsal bands, and also in having a greater proportion of anterior bands carried to the ventrum. C o. annulata has more crossbands and a sharper snout than the subspecies of C. palarostris. Nomenclatorial and Systematic Problems. — The type locality of annulata was generalized as the "Colorado Desert", but it may be inferred, both from the itinerary of the collector and from the morphology of the types, that they were collected somewhere along the U.S.-Mexican boundary between the Colorado River on the east and the desert base of the coastal mountains on the west, along what is now the southern border of Imperial County, California. The important problems of this subspecies concern not only its relation- ships with palarostris, klauberi, and occipitalis, but also the variations, within the subspecies, between the snakes from the east and west sides of the Colorado River, that is, between those from the Yuma and the Colorado deserts. Another ^Although the original description in the Mexican Boundary Survey report lists two specimens as numbers 2105-6, I am advised by Dr. Doris M. Cochran that it is doubtful whether there ever was a number 2106, since both types were included under the number 2105. Klauber— Shovel-Nosed Snakes 163 problem is to consider whether the differences between the snakes inhabiting the flat expanses of the Colorado Desert and those from the desert foothills of the San Diego mountains seem to have been caused by ecological differences, or by an infusion of occipitalis around the southern end of the Santa Rosa Mountains into the Borrego Valley and beyond. Material. — The data on this subspecies are derived from somewhat more than 400 specimens from California, 50 from Arizona, and 18 from Sonora. Unfortunately, a large part of the California material is from San Diego County, where the population does not represent pure and typical annulata. Description of the Subspecies. — Like the other occi pit alts subspecies, this is a snake of rather constant scale characters, but with observable intrasubspecific differences in pattern and color, some of which differences are territorially consistent. The longest specimen thus far accurately measured is a female 407 mm. (16 in.) in length over-all.* The shortest is 130 mm. (51/8 in.) ; two others are 132 mm. In Imperial County, where the genus seems to attain its maximum length, the females grow slightly larger than the males; females of 370 mm. (141/2 in.) are not exceptional, whereas the longer males average only 350 mm. (133/4 in.). The tail length in the adult males varies from 17.9 to 20.2 per cent of the length over-all, with a mode of 19.7 per cent. In the adult females, the range is from 15.7 to 18.4 per cent, with a mode of 17.0. The juveniles have proportionately shorter tails, the mode for the young males being 18.2 and for the females 15.0 per cent. In the San Diego County material, the regressions are fairly well represented by the following straight lines, in which T is the tail length and L the length over-all, both expressed in millimeters: males, T= .207 L- 3. 35; females T= .185 L- 5.00. The snakes of the Imperial- Yuma area have slightly longer tails, proportionately, than those from San Diego County. The scale rows are almost always 15 at mid-body; among more than 400 specimens, two have 16, and one has 14. There may be 17 or 16 for a short distance behind the head, and 14 or 13 just before the anus, but almost all specimens are 15-15-15 according to the usual method of scale-row expression. The ventrals in the males vary from 143 to 164, mean 153.2; and in the females from 153 to 178, mean 163.5. The subcaudals vary from 40 to 57, with a mean of 47.4 in the males; and from 34 to 51, mean 42.4 in the females. These averages for the subspecies as a whole are somewhat reduced by the high proportion of San Diego County specimens. The supralabials nearly always number 7, about one per cent having 6, and still fewer 8. As in the other subspecies, the third and fourth enter the orbit, the sixth being the largest. The infralabials also generally number 7, although a count of 8 on either or both sides is not unusual, and a few have 6. ^I have a record of a live specimen about 425 mm. (16% in.). 164 San Diego Society of Natural History It is often difficult to count the infralabials accurately, since the posterior termination is not definite. In most specimens there is a single loreal on each side, but about 10 per cent are without loreals, these scales being fused to the prefrontals. Two loreals per side are rare. Although a single preocular is normal, 1 per cent of the specimens have 2. One postocular, instead of the normal 2, occurs in about 1 per cent; 3 on a side are rarely observed. The usual temporal formula is 1+2; some 3 per cent of the counts are 1 + 1, about 1 per cent 2+2, and still fewer 1+3. The dark crossbands on the body vary from 18 to 35, with a mean of 24.9; and the tail rings from 5 to 12, mean 7.7. The bands in this subspecies may be either narrower or wider than the interspaces, but in most specimens the bands are the narrower. There may be some spotting in the interspaces, on the sides especially; this spotting is usually restricted to darkened scale edges, and becomes somewhat more frequent as the klauberi territory is approached. There is also occasionally present a type of ventral spotting that is independent of the body rings. The subspecies annulata differs from occipitalis in the greater frequency with which the dark dorsal bands become complete rings by crossing the ventrum. In both subspecies the anterior bands are less often complete than the posterior, but in annulata the ventral blotches that complete the rings make their appearance nearer the head than in occipitalis. For reasons discussed elsewhere, I prefer to use, in differentiating the subspecies, the following combina- tion criterion: the dorsal bands plus the anterior unmarked blotch positions on the ventrum, counting up to the first apparent ventral blotch. Only about 20 per cent of the specimens of occipitalis fall below 45 in this criterion, while in annulata only 9.6 per cent have 45 or more. In annulata the range is 22 to 59 and the mean 35.1. These values are rendered somewhat higher, both in range and average, because of the higher proportion of San Diego County specimens, which deviate from the subspecific mode (see below) . Superficially, preserved specimens of annulata are black-ringed snakes with a white or cream ground color. Actually, if one compares the bands of fresh specimens with the darkest colors of the Ridgway series, it will be found that they are seldom truly black; and they fade to a still lighter color on exposure to light. But on a casual examination, without comparative color guides, the rings in well-preserved material appear black. Many live specimens have varying degrees of pink or red suffusions in the interspaces, and some of these colors may be retained for a time after preservation. The body bands may be even-edged or somewhat serrated. Anteriorly they usually end at the first to third scale row above the ventrals, their termina- tions being rather blunt; then, as one proceeds posteriorly, it will be observed that the bands become more pointed laterally, as if reaching out to contact the ventral blotches which correspondingly rise higher on the sides. Finally the dorsal and ventral sections of the bands become confluent and the rings are complete, usually on the anterior quarter of the body. But since the first lateral contacts between the dorsal and ventral bands are rather tenuous, it is difficult Klauber — Shovel-Nosed Snakes 165 to decide which ring is to be counted as the first complete one. It is for this reason that I prefer to use the first ventral spot in the differential criterion, rather than the first complete ring. The ventral marks are lighter than the dorsal. Posteriorly the ventral blotches widen considerably, sometimes darkening as many as 4 adjacent ventral scutes. The black head crescent is rather narrow in this subspecies. Often the frontal remains unmarked, and the supraoculars are darkened only along their outer edges. Posteriorly the tips of the parietals may remain unmarked, or the dark color may be carried to the first dorsal scales. Laterally the horns of the crescent rarely extend forward of the eye. The light dorsal scales (end-to- end) between the head crescent and the first dark band on the neck vary from 3 to 9, with a mode of 5 or 6. As to the typical representative colors of live specimens of this subspecies, I present the Ridgway colors of specimens from the mesas east and west of the Imperial Valley, in Imperial County, California Dixieland Yellowish Glaucous Grenadine Dull Purplish Black Dull Purplish Black Grenadine Sea-foam Green Vinaceous Slate Sea-foam Yellow Sand dunes 14 mi. w. of Yuma Pale Chalcedony Yellow Peach Red Black Black Peach Red Pale Chalcedony Yellow Warm Blackish Brown Pale Chalcedony Yellow Snout (except rostral) Rostral Head crescent Dorsal bands Interspace centers Interspace edges Ventral bands Ventral interspaces In general, live specimens of anmdata, when from the center of the range of this subspecies, are black-ringed, with bright suffusions of pink or red in the interspaces, on a cream or white background. The red is often even-edged, leaving a single row or light scales between the red and black. The red, however bright and clear it may be, and however much like a true ring dorsally, is never complete on the ventrum, but always terminates laterally one or more scale rows above the ventrals. Intrasubspecific Trends. — The area inhabited by the most typical specimens of anmdata, that is, those most widely differentiated from the other occipitalis subspecies, comprises the eastern and western borders of the Imperial Valley, in Imperial County, California; and the Yuma Desert as far east as Mohawk, in Yuma County, Arizona. Outside of this area klauberi and occipitalis influences are shown in several ways, particularly by the interblotch maculations in snakes from the eastern border, and a change from black to brown bands, with a suppression of the red interband suffusions or saddles in western specimens. If the scale counts be restricted to the snakes of the central area, we have the following data: Ventrals, males, range 147 to 164, mean 156.6; females, 166 San Diego Society of Natural History range 161 to 177, mean 168.1; subcaudals, males, range 43 to 57, mean 50.5; females, range 41 to 51, mean 45.5. Comparing these counts with those of snakes from border areas, we find that typical annulate has more ventrals and subcaudals, on the average, than the fringe populations, whether to the west in San Diego County, or to the east in southern Maricopa and western Pima counties. The differences in numbers of scales from the San Diego averages are as follows, the Imperial- Yuma specimens being always higher: Ventrals, males 4.0, females 6.2; subcaudals, males 3.6, females 3.9. These differences may be caused by the higher temperatures to which the Imperial-Yuma snakes are subjected, or to their larger size; it is known that both of these conditions lead to average increases in the number of scales. There are no conspicuous differences in ventral and subcaudal scales between the Imperial-Yuma snakes and those of northwestern Sonora. The divergence in pattern between the populations from the fringe areas and those from the central Imperial-Yuma area is more marked than in squamation, the differences being not only numerical but of form and color as well. The Imperial-Yuma snakes are characterized by well separated, black crossbands. The bands are almost never as wide as the interspaces; in fact, in the majority of specimens the interspaces are almost twice as wide as the bands. Also, in most specimens, an interblotch suffusion of pink or red is present in life to a marked degree, so definite, in fact, as to form a well-defined saddle. There is usually a spot of pink on the rostral. The head crescent is narrower than it is in the fringe-territory specimens; it often leaves both the frontal and the posterior edges of the parietals clear, and the lateral horns rarely reach the preoculars. The statistics of the bands are given in the following summary, the San Diego County data being supplied for comparative purposes: Imperial- Yuma Counties San Diego County Body bands, range mean 18-35 23.9 19-35 24.8 Tail rings, range mean 5-11 7.7 5-10 7.6 Body bands plus unmarked ventral blotch positions. range mean 20-40 28.0 20-59 36.7 It will be observed that the only important difference is in the combination category of body bands plus unmarked ventral blotch positions, for the first ventral marks in the San Diego series usually occur farther back than in the Imperial-Yuma snakes. This is apparently an occipitalis influence, as are some of the other divergences of the San Diego County snakes from the more typical Klauber— Shovel-Nosed Snakes 167 Imperial-Yuma population. Although the San Diego specimens usually have black dorsal bands, dark-brown bands are not rare. The dorsal patterns of the San Diego snakes are, in fact, particularly variable, the bands being diverse not only in color, but also in relative width. The superiority of the interspaces in width, as compared with the bands, as evident in the Imperial-Yuma series, is not so manifest in the San Diego population, wherein bands equal to, or exceeding, the interspaces in width are quite common. Snakes with wide bands are most frequent at the higher altitudes on the extreme western fringe of the range, at such mountain-foothill localities as Mountain Spring, La Puerta, and lower Sentenac Canyon. In some of these, the bands are so wide as to comprise large dorsal blotches; for example, in a specimen from Yaqui Pass, the dorsal blotches are 5 scales wide (end-to-end) and the interspaces only 2 scales. In contrast, a specimen from Clark Dry Lake, out on the flat area of the Borrego Valley, is exactly reversed, with bands 2 scales wide and interspaces 5. Another specimen has a superabundance of melanin resulting in a blackened snout, dark scale edges in the interspaces, and a heavily mottled ventrum between the normal ventral bands. The interspace colors are also highly variable in the San Diego County series. The following Ridgway shades were observed in 20 live specimens; these may be compared with typical Imperial County colors listed on p. 165. Interspaces Dorsal edges White Ivory Yellow Sea-foam Yellow Naphthalene Yellow Light Chalcedony Yellow Marguerite Yellow Martius Yellow Massicot Yellow Dorsal bands Centers Dusky Purplish Gray Colonial Buff Blackish Brown (1) Warm Buff Blackish Brown (2) Cinnamon Buff Blackish Brown (3) Ochraceous Buff Dull Violet-Black Sea-foam Yellow Fuscous Black Honey Yellow Aniline Black Flesh-Ochre Black Mikado Orange Bitter Sweet Orange Apricot Orange Ochraceous Orange Light Salmon Orange Zinc Orange Grenadine Ochraceous-Salmon Carrot Red Carnelian Red Just as the bands become wider, with the transition from a flat, sandy habitat to one of rocky foothills, so also the reds of the interspaces are changed to yellow or buff. 168 San Diego Society of Natural History On the opposite side of the annulata range, in the Gila Bend-Ajo area, there is, as might be expected, a klauber't influence. Compared with the Imperial- Yuma series, the ventrals and subcaudals are low, and the body bands somewhat high, in number. There is usually an orange suffusion in the interspaces with considerable dark spotting on the sides (the most evident klauberi influence). The interspace saddles, whether orange, or maculated with brown scale edges, are usually 1 scale wide dorsally, and 2 wide laterally. Along the Sonoyta-Punta Periasco road in northwestern Sonora, somewhat the same tendencies are to be observed; however, the snakes of this area differ from those of the Gila Bend-Ajo section in having more ventrals, in which character they are nearer to typical annulata. They are like the Gila Bend-Ajo series in the color and shape of the body bands, and in the presence of macula- tions in the lateral interspaces. The red or orange interspace saddles, whether or not carrying the brown maculations relating them to klauberi, are quite narrow— 1 scale wide dorsally and 2 or 3 laterally. The following were the dorsal body colors of 3 specimens from the vicinity of EI Papalote, Sonora: Dark bands— Dull Violet Black, Dull Purplish Black, Black; interspace centers — Japan Rose (only 1 scale wide, with dark spots on anterior tips of scales), Carnelian Red (only 1 scale wide, mostly on the anterior scale tips), Grenadine Red; interspace edges — Naples Yellow, Naphtha- lene Yellow, Straw Yellow. Specific and Subspecific Relationships. — The subspecies C. o. annulata intergrades with both C. o. occipitalis and C. o. klauberi in ways that have been indicated, both in discussing intrasubspecific variation in annulata and also in the treatment of the other two subspecies. There remain only the subspecies of C. palarostris to mention. Of all the subspecies of C. occipitalis, annulata most nearly resembles C. palarostris in pattern, for annulata is characterized by black rings, which are both narrower and fewer in number than in the other occipitalis subspecies, and has more red in the interspaces. Both of these characters are carried to a still further extreme in palarostris, although some- what less so in p. organica than in p. palarostris. Were these details of pattern the only differences, one might expect an intergrading population to be found between annulata near Colfred and organica in the Organ Pipe Cactus National Monument, by way of Mohawk Valley. But organica is notably low in ventral scale counts, whereas annulata is high. In addition, the convex crown of organica contrasts with the flatter and sharper snout of annulata. For these reasons, although intergradation between organica and annulata is a possibility — which would make Chionactis monotypic — the present evidence is against the idea. Range. — Chionactis o. annulata is found in the desert foothills and desert areas of San Diego County,* and in Imperial County (except northeast of the Chocolate Mountains), California; Yuma County, western and southern *USNM 59465 from the "San Diego River Valley" almost certainly has an incorrect locality assigned to it. In spite of the most intensive collecting during the last 30 years, no other cismontane specimen, not known to have escaped from captivity, has come to light in southern California. Klauber— Shovel-Nosed Snakes 169 Maricopa County, and northwestern Pima County, Arizona; and in Sonora, Mexico, from Sonoyta west and southwest. Although annulata has not yet been reported from Baja California, it certainly occurs there, as it has been taken within a mile or so of the border at several points between Calexico and Winterhaven. Tracks, quite probably of Chionactis, have been seen at San Felipe about 120 miles south of the border, on the Gulf of California coast. Locality Records. — CALIFORNIA: Riverside County — 1 mi. nw. of Travertine Rock (occipitalis intergrade) ; San Diego County — Clark Dry Lake (also 1.2 and 2 mi. s.), Beatty's (Borrego Valley), Borrego Spring, Christmas Tree Circle (Borrego Valley), Yaqui Pass (also 2 mi. ne.), foot of Sentenac Canyon, Yaqui Well, The Narrows, Bensons Dry Lake ( = Ocotillo) (specimens have been collected on Cal. 78 every hundred yards or so from Yaqui Well via The Narrows to the County Line east of Ocotillo, a distance of I8I/2 mi.), San Felipe Wash, HalfhiU Dry Lake (old San Felipe townsite), La Puerta, Vallecito, Agua Caliente Spring, Carrizo Spring; Ij7iperial County — Travertine Rock (occipitalis intergrade), Sea View, San Felipe Wash, San Diego-Imperial County line at Cal. 78 (also 2 and 5 mi. e.), Kane Spring (also 11/2, 3, and 5 mi. nw., and 4, 5, and 8 mi. w.), Calipatria, Alamorio, Brawley, Imperial, Mountain Spring (also 2 and 3 mi. e.), Coyote Wells, Plaster City, Dixieland (also 1 mi. w.), Seeley (also 3 mi. w.), Calexico (also 3 mi. w.), Bonds Corner, Meloland, Holtville (also 14 mi. se.). Drop 3 (on the All- American Canal), Midway Well (junction U.S. 80 and Cal. 98) (also 1 and 8 mi. w., and 6 mi. ne.). Grays Well (also 1, 2, and 5 mi. w.). Chocolate Mts., Winterhaven (also 10 and 14 mi. w.), 14 mi. w. of Yuma (Arizona), Pilot Knob, Pilot Knob R.R. Sta., Colorado River (near Pilot Knob) . ARIZONA: Yuma County — 6 mi. s. and 15 mi. se. of Parker, 9 mi. s. of Quartzsite, 2 mi. e. of Dunn (=10 mi. e. of Brenda), Yuma (also 5 mi. e.), Yuma Mesa, Monument 200 (= Arizona-Sonora boundary, 15 mi. e. of Colorado River), Dublin (also 7 mi. w.), south Gila Valley, Fortuna Wash (15 mi. e. of Yuma), Wellton Mesa, Wellton (also 4 mi. e.), Tacna (also 5 mi. e.), Lugar Bonita (=12 mi. e. of Wellton), Pembroke, Mohawk (also 5 mi. e. and 7 mi. w.), Chrystoval (= Stoval), bet. Roll and Maricopa County line; Maricopa County — 21/2 mi. e. of Aguila, 7 mi. sw. of Wicken- burg. Cactus Garden ( = 3 mi. nw. of Morristown), Sentinel (also 2 mi. e.), Gila Bend (also 7, 15, and 281/2 mi. s., and 12 mi. e.), 7 mi. s. of Black Gap, Midway (Black Gap and Midway are stations on the Tucson, Cornelia, SC Gila Bend R.R.), 24 mi. n. of Ajo (Pima Co.); Pirna County- — 7 mi. se. of Ajo, Gunsight Junction (= Ajo-Tucson-Sonoyta road junction) (also 2 mi. nw.). SONORA: 25 mi. s. of San Luis, I/2 way bet. Sonoyta and Punta Peiiasco, 23 mi. sw. of Sonoyta, 1 1 and 20 mi. sw. of Pozo Sipiano, EI Papalote (also 2, 6, and 14 mi. ne., and 1 and 4 mi. sw.), 6I/2, 9, and 16 mi. ne. of Punta Peiiasco, 8 mi. n. of Rocky Point (= Punta Peiiasco). With regard to intergrades between anmdata and klauberi, it should be stated that all of the snakes in the Wickenburg, Gila Bend-Casa Grande, Giia 170 San Diego Society of Natural History Bend-Ajo, and Sonoyta-Punta Penasco sections show klauberi tendencies in various degrees. Specimens from the following places show these tendencies to a sufficient degree to be considered integrades: Maricopa County — 21/^ mi. e. of Aguila, Gila Bend (also 15 mi. s.), and 24 mi. n. of Ajo; Pinal County— near Casa Grande (locality somewhat indefinite) ; Pima CoMn^y— Gunsight Junction; Sonora—\ and 2 mi. sw. of El Papalote, and 6I/2 mi. ne. of Punta Penasco ( =■ 71/2 mi. sw. of El Papalote) . It appears that the klauben influence is somewhat spotty, being concentrated in some areas of the general border between the subspecies. Even in these the effect is not uniform, for it is evident in some specimens but not in others collected at the same place. Chionactis occipitalis klauberi (Stickel) Tucson Shovel-nosed Snake 1941. So7iora occipitalis klauben Stickel, Bull. Chicago Acad. Sci., vol. 6, no. 7, p. 138. Type specimen LMK 29647; type locality Tucson, Pima County, Arizona. 1943. Chionactis occipitalis klauberi Stickel, Proc. Biol. Soc. Wash., vol. 56, p. 124. Diagnosis. — This subspecies of Chionactis occipitalis is characterized by dark secondary crossbands occupying the centers of the interspaces between the primary black crossbands. In the typical specimens the secondary bands are complete moddorsally. They widen laterally but do not reach the ventrum. The new subspecies C. o. talpina is the only other one having dark (non-red) secondaries; in klauberi the primary bands are black, or nearly so; in talpina they are brown. C. o. talpina has more ventral plates than klauberi. Systematic Problems. — In the area from Tucson, Pima County, Arizona, northwest to Picacho, and thence north to Florence and Florence Junction, in Pinal Co., most specimens of Chionactis are typical klauben. Intergradation with annulata begins at Casa Grande, and is noted as far west as Gila Bend, north to Aguila, and south to Ajo. Although most of the specimens from the Sonoyta-Punta Penasco road in northwestern Sonora are to be considered annulata, some evince klauberi tendencies to a noticeable degree, which is all the more surprising, since a population of palarostris organica occupies the Organ Pipe Cactus National Monument, situated between these Sonoran intergrades and the klauberi headquarters in the Tucson-Florence area. Material. — Of typical klauberi I have had 17 specimens available, of which 7 are females. Description of the Subspecies. — This is a snake of the usual Chionactis form, with the inset lower jaw and sharp rostral characteristic of the genus. The shortest specimen at hand is a female 157 mm. in length; the longest, a male and a female each measuring 339 mm. The tail in the adult males is about 19 per cent of the length over-all, and in the females about 16 per cent. The scale rows number 15. The ventrals in the males range from 141 to 151, mean 145.8; and in the females from 153 to 159, mean 156.0. The Klauber — Shovel-Nosed Snakes 171 corresponding subcaudal counts are 42 to 47, mean 44.4; and 38 to 43, mean 40.4. The labials, both upper and lower, number 7. The nasals, loreals, and preoculars are single; the postoculars paired, except in one specimen that has a single postocular on one side. The temporals are 1+2, except in one specimen having 1 + 1 on the left. The black head crescent is quite broad on top; it not only marks the supraoculars, but is carried back to the first dorsal scales behind the parietals. At the frontal, there is often a deep light indentation or notch that serves to lighten the frontal itself. On the sides the points of the crescent are carried forward to mark the nasals. The ground color of the snout, anterior to the crescent, is somewhat darkened, compared with the rest of the ground color. The primary rings on the body range in number from 23 to 29 with a mean of 26.7, and on the tail 7 to 11, mean 8.1. The body rings plus the unmarked ventral ring positions vary from 25 to 40, mean 30.1. The rings are usually narrower middorsally than the interspaces, but may be equal, or, rarely, wider. The dorsal marks narrow laterally, and then widen again on the ventrum. The first ring completely encircling the body ranges from the fifth to the twelfth ring. On the ventrum the rings again widen, more so, in fact, than in any other subspecies; posteriorly, they usually engage 3 or 4 ventrals. Between the primary rings there is a secondary dark series, more conspicuous in this than in any other subspecies. The secondary rings are narrowest mid- dorsally; they are usually separated from the primary series by a single row of immaculate scales, and thus the secondaries widen on the sides where the primaries become narrower. Laterally, the secondaries end on the first or second row of scales above the ventrals. The dark color of the secondaries is never as dense as that of the primaries; often only the centers of the scales are darkened and thus some of the ground color may show through, or a pink suffusion may do so, particularly on the tail. Breaks in the secondary rings occur most often middorsally. The basic color of the secondaries may be black or brown, but the pink suffusion often gives them a purplish cast. The ground color is white, cream, or light-yellow. Intrasubspecific Trends. — In the Tucson-Marana-Picacho-FIorence- Florence Junction area, klauberi seems to be a rather consistent form. With a single exception, all of the specimens of Cbionactis that I have seen from this territory clearly belong to this subspecies. True, only about half the specimens have complete secondary bands middorsally, but all have darkened scale centers in the lateral scales comprising the secondaries, whereas it is characteristic of most annulata-klauberi intregrades that only the scale edges are darkened. In the klauberi territory, the secondary rings are more often complete dorsally in the Florence area than farther south at Marana. The single specimen from klauberi territory failing to agree with the others in pattern is No. 422 in the Woodin collection. This is said to have been taken on the San Xavier Road near Tucson (it was not collected by Mr. Woodin himself). It has none of the characteristics of klauberi, or even of annulata, but resembles occipitalis from Riverside Co., California. Unless there is some error in the locality record, it is a very queer specimen indeed. 172 San Diego Society of Natural History Relationships with Other Subspecies — To the northwest of the kiauberi territory comparatively few specimens are available. There are 2 from the general vicinity of Casa Grande, Pinal County, — the exact points of collection are not known — one of which favors kiauberi, while the other shows an annulata affinity. From the Wickenburg area only 2 specimens are at hand; one is from Cactus Garden, northwest of Hot Springs Junction, and the other is from 21/2 miles east of Aguila. Both show kiauberi influences, the latter somewhat more than the former, although it was collected about 25 miles farther away from the kiauberi territory. From southern Maricopa County, south of Gila Bend, and thence south to somewhat beyond Ajo, in Pima County, a fairly adequate series is available. The interspaces are almost always spotted to some degree. I consider this to be an area of intergradation, in which most of the specimens favor annulata more than kiauberi. Southwesterly in Sonora, along the road between Sonoyta and Punta Penasco, occurs another intergrading population. In these snakes, the dark color in the interspaces is nearly always restricted to scale edges. All fresh specimens show pink suffusions in the interspaces. Though the kiauberi affinity is clearly evident, I consider this series nearer annulata. Farther west in the vicinity of Yuma an occasional specimen shows faintly the kiauberi secondaries; and even in California, as I have discussed elsewhere, this darkening to form secondary bands is sometimes evident to a moderate degree. Locality Records. — Specimens of kiauberi have been collected at the following points: ARIZONA: Pima County — Tucson (type locality), Marana (30 and 32 mi. nw. of Tucson) ; Pinal County — 6.4 mi. se. of Red Rock, 3 mi. se. of Pichacho, 3, 4.7, and 10 mi. n. of Florence (7 mi. s. of Florence Junction), Magma Junction, 5.5 and 8.3 mi. s. of Florence Junction. To complete the locality list, the following are to be considered records of annulata-klauberi intergrades, as discussed more fully under the former sub- species; ARIZONA: Maricopa County — 21/^ mi. e. of Aguila, Gila Bend (also 15 mi. s.), and 24 mi. n. of Ajo; Pinal County — near Casa Grande; Pima County — Gunsight Junction; SONORA — 1, 2, and 71/2 mi. sw. of El Papalote. Chionactis occipitalis talpina* subsp. nov. Northern Shovel-nosed Snake Plate 10, fig. 1 There have lately been collected in Nevada, considerably to the north of any specimens hitherto known from that state, and about 75 miles north- easterly of the nearest specimen from California, 3 specimens of Chionactis that justify subspecific segregation from their territorially closest relatives. Two of the 3 specimens are from a higher altitude than any previously known to be inhabited by Chionactis. *Mole-Iike, to denote its fossorial habit. Klauber— Shovel-Nosed Snakes 173 Type 5pecrmeri.— California Academy of Sciences number 81364. Found dead on the road by Joseph R. Slevin and Wallace Wood, 50 miles south of Goldfield on the highway to Beatty, in Nye County, Nevada, June 3, 1947. Diagnosis. — A subspecies differing from the subspecies of palarostris, and from the other subspecies of occipitalis except klauberi, in having, in the interspaces between the main series of the crossbands, dark marks that give the effect of secondary brown crossbands. From klauben it differs in having more ventrals, and in having brown, instead of black, primary bands. Description of the Type. — An adult male. The length over-all is 284 mm. and the tail length 52 mm.; ratio .183. The body is of normal Chionactis configuration, and it has the important Chionactis characters of nasal valves, abdominal angle, and protruding flap on the mental. The pupil is round. The dorsal scale rows are 15-15-15, all smooth, and with single apical pits. There are 153 ventrals, the anal is divided, and the subcaudals, all divided, number 44. The head scales are normal for the species, with one loreal on each side, one preocular, two postoculars (the upper larger) and the temporals 1+2. The supralabials number 7-7; the third and fourth enter the orbit, and the sixth is the largest. The infralabials are 7-7. The body pattern comprises a series of dark-brown crossbands, 27 on the body and 7 on the tail. They are considerably wider than the interspaces, being about 4 scales long (end-to-end) while the interspaces cover 2. All bands except the first on the neck touch the ventrals, and all but the first 2 cross the ventrum and thus become complete rings. The bands narrow laterally; the posterior ones again widen at the ventrum. The primary bands are somewhat darker dorsally than on the sides and ventrum. In the interspaces the ground color is cream, but the central scales are heavily blotched with dark-brown, although the edges of the maculated scales are usually clear. On the sides, the secondaries widen somewhat, but the maculations become less intense at the lowest 2 rows of the lateral scales; on the ventrals the interspace macula- tions are represented by scattered punctations. The head crescent is quite wide, extending from the middle of the supra- oculars and frontal to the posterior edge of the parietals. The snout is some- what punctated above, and there are several spots on the labials and lower jaw, where other forms of Chionactis are usually clear. Paratypes. — There are 2 paratypes, LMK 39520-1, collected by Eric Comstock and Verne Larson 10 miles north of Goldfield, Esmeralda County, Nevada, on the road to Tonopah, in July, 1949. Although taken 60 miles north of the type locality, and at a higher altitude (about 5500 ft.), they are not as extreme in pattern as the type, for the characteristic brown maculations in the interspaces do not cross the dorsum. However, they have higher ventral counts. No. 39520 is an adult male, length over-all 335 mm., tail length 66 mm. The scale rows are 15-15-15, the ventrals number 162 and the subcaudals 51, all divided. All labials are 7-7. The head crescent extends across the parietals; its lateral horns extend 174 San Diego Society of Natural History forward to the nasals. There are 32 primary crossbands on the body and 9 on the tail. The anteriormost band marking the ventrum is the fourth, and the first comprising a complete ring is the sixth. The dorsal bands, wider than the interspaces, extend longitudinally for slightly less than 3 scales, end-to-end, while the interspaces measure 21/2. Scales less than half maculated are usually lighter than their fellows in the bands. The bands narrow laterally, and are both lighter and narrower on the ventrum. Posteriorly, the bands on the ventrum become wider and darker. Between the main dorsal blotches the interspaces are maculated with brown, particularly at scale centers. The colors in life (Ridgway) were as follows: The snout was Reed Yellow, the crescent Mummy Brown. The primary dorsal bands were mostly Warm Sepia, but Rood's Brown at their lower edges. The interspaces were Colonial Buff with spots of Light Ochraceous Salmon; the maculations were Rood's Brown. The ventral surface was Pale Olive-Buff and the rings Rood's Brown. There were Maize Yellow spots on the tail between the primary blotches. The eye was Black. The second paratype, an adult female containing 3 eggs, has a length over-all of 340 mm. and a tail length of 56 mm. The ventrals number 162 and the subcaudals 43. There are 33 primary bands on the body and 11 on the tail. None of the primary bands crosses the ventrum; instead, the ventrals are generally speckled with brown. However, the rings are complete on the tail. The colors in life were much the same as in the other paratype, but the maculations in the interspaces are heavier. Relationships. — Although talpina superficially resembles klauberi in having dark bands in the interspaces, it is doubtful whether the two are closely related, or that there is or ever has been any territorial interconnection between them. As klauberi intergrades with, and was evidently derived from annulata, so talpina intergrades with, and is apparently a darkened derivative of occipitalis. Southward and southeasterly of the talpina territory, intergrades are already at hand that comprise a logical geographic pattern. Southwesterly none is yet available. MVZ 40977, from Indian Springs, Clark County, Nevada, is a faded specimen with some maculations in the interspaces. Farther to the southeast, at Boulder City in the same county, the snakes seem to be pure occipitalis. Southward of the talpina area, the nearest available specimens are Stanford 9218 and 9219 from 11 and 16 miles south of Shoshone, the first locality being in Inyo County, the second in San Bernardino County, California. Also there is CAS 65380 from Goler Canyon, Panamint Mountains, Inyo County, and LMK 35549 from 1 mile east of Leach Lake, San Bernardino County. All of these have brown spots in the interspaces laterally, although seldom middorsally; they are to be considered talpina-occipitalis intergrades. No other specimens are yet available from this area immediately below Death Valley, but to the extent that specimens are at hand, the pattern trend is quite consistent. From the southwest of the talpina area only a single Chionactis (CNHM 1714, from Owens Lake, Inyo County) is available. Since this rather faded Klauber— Shovel-Nosed Snakes 175 specimen shows no interspace maculations, it is to be considered pure occipitalis. The question may well be asked as to the frequency of occurrence of talpma-type maculations among specimens of occipitalis available from southern and western San Bernardino County, and eastern Kern County, from which area I have at hand about 100 specimens. I find that snakes with a few scattered dots in the interspaces are by no means unusual; probably 25 per cent of the population have such dots. Of snakes with marked talpina tendencies, strong enough to be considered intergrades, were they from fringe areas, there are 4: LMK 31686 from Minneola, LMK 31687 from Hector, LMK 39961 from 6 miles north of Adelanto, and UCLA 506 from 2 miles west of Barstow. Of these, the Hector specimen is the most surprising, as it is heavily maculated across the dorsum. But I do not consider that these territorially inconsistent, aberrant specimens of occipitalis invalidate talpina, any more than klciuberi is invalidated by the circumstance that scattered specimens of annulata, in San Diego and Imperial counties, have klatiberi tendencies. One other specimen, MCZ 9806, from a questionable locality, should be mentioned. It is reputed to have been collected in southern Utah, but Mr. Arthur Loveridge informs me that this is to be considered doubtful. No other Chionactis has yet been reported from that state. Superficially it looks like talpina, having brown primary bands, and lighter brown secondaries. Yet it could be a badly faded klauben, since the ventrals are low for talpina (it is a male with 146), and it has two other klauberi characteristics — a frontal notch in the head crescent, and marked widening of the posterior rings on the ventrum. Localities. — The subspecies talpina is known only from the type locality: 50 miles south of Goldfield, in Nye County, Nevada, on the Beatty road; and from 10 miles north of Goldfield, Esmeralda County, Nevada, on the Tonopah road. Areas of intergradation with C. o. occipitalis are indicated by the following localities where intergrades have been collected: near Indian Springs, Clark County, Nevada; 11 miles south of Shoshone, and Goler Canyon, Panamint Mountains, Inyo County, California; and 16 and 22 miles south of Shoshone, and 1 mile east of Leach Lake, San Bernardino County, California. Chionactis palarostris palarostris (Klauber) SoNORAN Shovel-nosed Snake 1937. Sonora palarostris Klauber, Trans. San Diego Soc. Nat. Hist., vol. 8, no. 27, p. 363. Type specimen LMK 26771; type locality 5 miles south of Magdalena, Sonora, Mexico. 1941. Sonora occipitalis palarostris Stickel, Bull. Chicago Acad. Sci., vol. 6, no. 7, p. 137. 1943. Chionactis occipitalis palarostris Stickel, Proc. Biol. Soc. Wash., vol. 56, p. 128 [= 123]. Diagnosis. — A subspecies having a more convex snout than any of the subspecies of occipitalis, also fewer ventrals (on the average), and fewer 176 San Diego Society of Natural History crossbands on the body and tail. It has fewer crossbands and relatively wider interspaces than the subspecies palarostris organica. Systematic Problems. — I originally described this form as a full species. Subsequently Stickel (1941, p. 137) reduced it to subspecific status, on the inference that specimens from intermediate territory would demonstrate inter- gradation. At that time only two specimens of palarostris were available to him — the type and MCZ 36890 from Costa Rica Ranch, some 40 miles to the westward of Hermosillo, Sonora. Recently an additional population of palarostris has been discovered in the United States, through the activity of W. R. Supernaugh, M. Max Hensley, and others collecting in the vicinity of the Monument headquarters at the Organ Pipe Cactus National Monument, on the road from Ajo, Pima County, Arizona, southward to Sonoyta. These U. S. specimens, although not as extreme as the type in low blotch counts, do have the convex snout of palarostris, and have fewer body bands than annulata. They differ in these characters from 12 specimens of annulata (or amuilata-klaiiberi intergrades) from Gila Bend to Gunsight, as well as from another series of 15 specimens of annulata collected between Sonoyta and Punta Pefiasco, in Sonora. This newly discovered palarostris population is known to extend at least 20 miles along the highway through the Monument, from 2 to 22 miles north of the U.S.-Mexican border. The nearest specimen of annulata available from Sonora was collected only 26 miles distant from the southernmost of this palarostris population, yet there is no indication of inter- gradation between the two. In the interval of about 7 miles separating the populations of palarostris and annulata north of the Monument, there are some suggestions of intergradation in pattern, but the snout difference remains uncompromised. Therefore, I conclude that palarostris should be retained as a valid species, unless further collecting in this area should demonstrate inter- gradation. An actual overlap between the two populations is by no means impossible. Of the two specimens of palarostris hitherto available from Sonora, the type, from 5 miles south of Magdalena, shows certain well-marked differences from both the Costa Rica Ranch specimen and the Organ Pipe population previously mentioned. During the past summer an additional west-central Sonoran specimen was collected by R. G. Zweifel and K. S. Norris, at a point 40.7 miles south of Hermosillo, or about 150 miles (straight-line distance) south of the type locality.* This has all of the outstanding pattern features of the type, thus indicating that these are characteristic of the population of that area. I have therefore decided to consider the Magdalena-Hermosillo popula- tion as subspecifically distinct from the Organ Pipe, assigning the new name organica to the latter. Material. — Two specimens of the subspecies palarostris palarostris are available. Description of the Subspecies. — This is a snake of the usual Chionactis configuration, except that the rostral is blunter, and the top of the head, from *The nearest named place to this locality is the hamlet of El Pocito some 3 miles northward. I shall hereafter refer to this as the Pocito specimen. Klauber— Shovel-Nosed Snakes 177 the center of the frontal to the snout, is distinctly convex, whereas in the subspecies of occipitalis it is virtually straight, or slightly concave, at the prefrontals. Of the two specimens available, the type is an adult male with 144 ventrals and 39 subcaudals; the Pocito specimen is a female with 152 ventrals and 42 subcaudals. The scale rows in both number 15, the supralabials 7, infralabials 8, the nasals are undivided, the loreals and preoculars single, the postoculars paired, and the temporals 1+2 or 1+3. The ratios of the tail length to length over-all (both are adults) are .183 in the male and .166 in the female. The pattern comprises alternating black and deep red blotches or saddles, separated by narrow borders of yellow ground color, which becomes white after preservation. In both specimens there are 10 black blotches on the body and 3 on the tail. The black blotches have a longitudinal extent of only i/^ of the interspaces, the latter having a length of about 10 to 12 scales (end-to-end), whereas the black blotches cover 3 to 31/2 scales middorsally, where they are widest. The red saddles have about 4 times the extent of the narrow yellow stripes that border them. Both the black and red blotches narrow on the sides, so that laterally the ground color is more in evidence than dorsally. The black blotches rewiden ventrally, but the red saddles fade out at the first row of scales above the ventrals. Laterally the red saddles are maculated with black spots; these are much less prominent in the type than in the Pocito specimen, in which they are carried to the ventrum. In this subspecies the snout is cream-colored. There is a large black parietal blotch covering the posterior 2/3 of the frontal and extending to the rear edges of the parietals; on the sides this blotch engages the eyes and the upper edges of the posterior supralabials. Although this blotch is analogous to the crescent characteristic of all subspecies of occipitalis, it is more rectangular in form. The underside of the head is cream. There are 14 to 16 scales, end-to-end, between the head blotch and the first dorsal body band. Shortly after preservation, the type exhibited the following Ridgway dorsal colors: Black, Brazil Red, and Maize Yellow, with a Cream Color ventrum. Intrasubspecific Trends and Relationships. — With only 2 typical specimens available, no conclusions respecting trends can be drawn. The most conspicuous difference between the Magdalena and Pocito specimens is the interspace maculations of the latter. Based on the available material, palarostris palarostris is subspecific- ally distinct from the snakes of the Organ Pipe Cactus National Monument. I shall discuss this relationship under the new subspecies. There remains MCZ 36890 from Costa Rica Ranch, Sonora, a locality some 40 miles west and slightly south of Hermosillo. In all important charac- ters, particularly in the number and relative spacing of the blotches, this specimen resembles the new subspecies organica, rather than p. palarostris. It is a juvenile and the preservation is not good. It is not possible to tell whether the snout is sharp as in occipitalis or blunter as in palarostris. Only additional collecting in this relatively inaccessible area will determine whether this is a 178 San Diego Society of Natural History palarostris of unusual pattern, an organica involving a very long range extension, or an atinulata far from its nearest relatives and with a sufficiently low blotch count to warrant a separate subspecific status. The latter is not an impossibility, having in mind the equally surprising territorial relationship between C. o. annulata and C. p. organica in western Pima County and northwestern Sonora. Thus far, Chionctis has not been collected between a point only a few miles south of Tucson and Magdalena, although considerable night collecting has been tried on the road from Tubac south to Nogales. It seems rather doubtful whether a continuous Chtonactis population exists over this direct line from Tucson to Magdalena via Nogales, hence there is little probability of a direct contact between the populations of klauberi and p. palarostris as repre- sented by their most typical forms. Range. — Chionactis p. palarostris has been collected 5 mi. s. of Magdalena, and 40.7 mi. s. of Hermosillo (near El Pocito) , Sonora, Mexico. There is a specimen of uncertain status from Costa Rica Ranch, Sonora. Chionactis palarostris organica* subsp. ndv. Organ Pipe Shovel-nosed Snake Plate 10, fig. 2 1950. Chionactis occipitalis palarostris Hensley, Trans. Kansas Acad. Sci., vol. 53, no. 2, p. 283. Type Specimen.— LMK 40673. Collected by William R. Supernaugh and Grover E. Steele at 9:42 P.M., May 22, 1950, on the Sonoyta-Ajo road, 9 miles north of the U.S.-Mexican Border, in the Organ Pipe Cactus National Monument, Pima County, Arizona. Preserved July 19, 1950. Diagnosis. — A subspecies differing from all occipitalis subspecies in having a blunter snout and fewer body blotches. From palarostris palarostris it differs in having more body blotches, and in possessing red saddles shorter than, or only equal to, the black blotches, whereas in p. palarostris the red saddles are more than twice the length of the black. Description of the Type. — An adult male. The length over-all was 303 mm. and the tail length 58 mm.; ratio .191 (before preservation). The body is of the usual Chionactis shape, and it possesses the nasal valve, abdominal angle, and protruding flap on the mental characteristic of the genus. The snout is blunter and more convex on top than in occipitalis. The pupil is round. The dorsal scale rows are 17-15-15, all smooth. There are 146 ventrals, a divided anal, and 42 subcaudals, all divided. There are 7 supralabials on each side; the third and fourth touch the eye, and the sixth is the largest. The infralabials also number 7. The nasals, loreals, and preoculars are single, the postoculars paired. The temporals are 1+2. The body pattern comprises a series of 14 black blotches or saddles; there are 4 blotches on the tail. Middorsally the saddles engage 4 scales. *To indicate its habitat in the Organ Pipe Cactus National Monument. Klauber— Shovel-Nosed Snakes 179 end-to-end, while the interspaces comprise 6 scales. The anterior edge of the first blotch is 51/^ scales back of the parietals. The black saddles narrow laterally and then rewiden on the ventrum. The first saddle to mark the ventrum is the third from the head, and the first to comprise a complete ring is the fifth. Between each pair of black blotches there is a vermilion saddle extending for about 31/2 scales. Although bright and clear, the vermilion blotches are not as sharp-edged as the black; laterally they fade out on the second or third row of scales above the ventrals. A few of the vermilion scales have darkened edges. The head crescent is quite wide, extending from the anterior third of the frontal and supraoculars almost to the posterior edges of the parietals. The anterior lateral points of the crescent reach the loreals. The ground color is cream, somewhat darker dorsally than below. Paratypes. — There are 14 paratypes available, all collected on the Ajo- Sonoyta road between 2 and 221/^ miles north of the International Boundary. These are Cornell 4073, Woodin 651, Illinois 5602-6, 5906, 10625, and 10630, and 4 specimens belonging to the Monument Museum. The following data were compiled from the type and paratypes taken together, comprising 12 males and 3 females. This is a snake of the usual Chionactis configuration but with the blunt snout and convex crown characteristic of palarostris, as compared to occipitalis. The longest specimen is a female 391 mm.; the longest male is 308 mm. The shortest is 166 mm. The tail in the adult males varies from 18.5 to 20.8 per cent of the body length over-all, with an average of 19.7; in the females it averages 16.5 per cent. The scale rows number 15. The ventrals in the males range from 141 to 150, mean 145.6; and in the females from 156 to 161, mean 158.3. The sub- caudals in the males range from 42 to 47, mean 44.5; and in the females are either 42 or 43. Both upper and lower labials usually number 7, but rarely there may be either 6 or 8. The nasals, loreals, and preoculars are single, and the postoculars paired. The temporals are 1+2. The species palarostris is the most beautiful of all the forms of Chionactis because of the extent and depth of the red coloration in the spaces between the black crossbands. The subspecies organica is somewhat less spectacular than p. palarostris. The black head crescent, so characteristic of Chionactis, becomes almost a rectangle in organica; its anterior edge crosses the frontal and supraoculars at, or just behind, their forward edges, and it extends back as far, or nearly as far, as the posterior edges of the parietals. The lateral borders of the black mark follow the upper edges of the supralabials. The black engages the eyes and is carried forward to darken the preocular, particularly its lower half, and sometimes the loreal. There are 4 to 6 scales between the parietals and the anterior edge of the first black crossband; this compares with 14 and 16 in the two specimens of p. palarostris. The black crossbands on the body are widest (measured along the snake) middorsally, narrowing laterally, but again widening — this is especially true posteriorly — on the ventrum. The bands are from 3 to 5 scales long (end-to- 180 San Diego Society of Natural History end) middorsally, and usually are narrower than the interspaces, which vary from 4 to 7 scales in length. All but the first 1 or 2 black bands are represented by complete or partial rings crossing the belly, except for one specimen, wherein the fifth blotch is the first one carried to the ventrum. Dorsally, the interspaces are colored with deep-red for lengths of 2 to 4 scales; these red saddles widen laterally (where the black bands narrow) but terminate 1 to 3 scale rows above the ventrals. The lowest red scales are often edged with brown. Dorsally and laterally there is a single row of ground-colored scales between the red and black; and, since the ground color is white or yellow, the snake is tricolored in bright, contrasting hues.* But the beautiful red of the live specimens is soon lost in preservation. In organica there is undoubtedly sexual dimorphism in the number of black body blotches, for the 3 females have 18, 18, and 20 respectively, while the males vary from 13 to 16 with a mean of 14.2. The tail blotches number either 4 or 5 in both sexes. Using the criterion of body bands plus the unmarked ventral spaces opposite the bands, the variation in organica is from 14 to 22, with an average of 16.5. This is considerably lower than in any subspecies of occipitalis. Just after it had shed, an adult female of the Organ Pipe series had the following Ridgway colors: snout. Sea-foam Green anteriorly, Deep Sea-foam Green posteriorly; head crescent, Aniline Black; dorsal crossbands, Dull Violet- Black; interspace centers. Scarlet; interspace edges, Pale Chalcedony Yellow; ventral dark marks. Dull Purplish Black; ventral interspaces. Pale Chalcedony Yellow to Pale Dull Green- Yellow, depending on the color of the organs within, for the skin here is slightly translucent. A male, somewhat smaller, exhibited the following life colors: snout. Pale Chalcedony Yellow in front to Pale Dull Green- Yellow behind; head crescent. Dusky Purplish Gray; dorsal crossbands, Black; interspace centers. Scarlet; interspace edges. Pale Chalcedony Yellow; ventral dark marks, Dull Purplish Black; ventral interspaces, Pale Glass Green. Relationships with Other Subspecies. — Chionactis p. organica differs from the subspecies of C. occipitalis in having a blunter snout, fewer ventrals, and fewer dorsal crossbands. Territorially, the nearest occipitalis specimens avail- able, which belong to the subspecies annidata, are from two areas: those from along the Sonoyta-Punta Peiiasco road (15 available specimens), and those from the road from Gila Bend south via Ajo to Gunsight. The former show no organica affinities; to the extent that they are not typical annidata, they tend toward klauberi, for they have a high number of narrow crossbands and maculations in the interspaces. The specimens of annidata from the Ajo-Gunsight area, distant only 7 to 15 miles from the most northerly Organ Pipe locality of organica, are less conclusive, there being only 3 specimens available from below Ajo. For *For this reason Chionactis, particularly this species and C. o. annidata, is popularly confused with the Sonoran coral snake {Micruroides euryxanthus) . Besides differences in squamation, it may be noted that the red bands of Micruroides, unhke those of Chionactis, cross the ventrum. Klauber— Shovel-Nosed Snakes 181 annulata, they are somewhat low in ventrals (males with 147, 149, and 156) but not in body bands (28, 27, 22). The snouts are fairly sharp, and the head marks more crescentic than in organica. The red bands in the interspaces are narrower and more maculated (a klaubert trend) than in organica. Al- together, I deem intergradation or hybridization between the snakes of the Organ Pipe and those from near Gunsight to be improbable, but not impossible. There are strong morphological reasons for believing palarostris and organica to be conspecific, yet they are widely separated geographically and no intergrades are at hand. If organica is, as has been inferred, a north- westerly intrusion of palarostris between two annulata populations (Pozo Sipiano-Papalote and Ajo-Gunsight) then eventually palarostris-organica intergrades should be found in the vicinity of Altar, Caborca, and Tajitos. These places lie in a relatively inaccessible area, which will for some years lack the paved roads so necessary to facilitate the collection of Chionactis, and the solution of this problem may be long delayed. MCZ 36890 from Costa Rica Ranch has many organica characteristics and comes from near the range of palarostris, yet it is from a point that is territorially inconsistent with its being considered a palarostris-organica intergrade. Locality Records. — At present organica is known only from the Sonoyta- Ajo road through the Organ Pipe Cactus National Monument, where it has been taken 2, 4, 4.3, 4.5, 6.5, 8.2, 9.8, 12, 16, 17.5, 18.4, and 22.4 miles north of the International Boundary.* Generic Relations and Intrageneric Trends It is probable that Chiloineniscus is the nearest present-day relative of Chionactis. I say this, not because of the really remarkable superficial re- semblance in color and pattern between Chilomeniscus cinctus and Chionactis occipitalis, but because the two genera have certain important characters in common that distinguish them from most other Colubridae, namely, possession of nasal valves, an angled abdomen, a notably sharp snout with a deeply inset lower jaw, and a mental protrusion to close the tongue orifice. These are characters in which both genera differ from Sonora; they are characters fitting them for a life on or beneath the surface of sand. Chilomeniscus is a stouter-bodied, shorter-tailed snake than Chionactis, and the snout is even sharper with a more deeply inset under jaw. Although Chionactis has head plates conforming to the normal of most genera of Colubridae, Chilotneniscus has several distinctive deviations, including (usually) a contact between the rostral and prefrontals, the merging of the nasals and internasals on each side, and the prevailing absence of loreals. Thus, the two genera are quite distinctive; nevertheless, their mutual possession of certain unusual characters suggests a common ancestry. *The localities recorded by the collectors sometimes refer to the Boundary, and sometimes to the Monument Headquarters or Ajo. To secure uniformity I have referred all localities to the Boundary. The Monument Headquarters are 5 miles north of the Boundary, and Ajo is 38 miles north, these distances being measured along the road. 182 San Diego Society of Natural History The nasal valve in Chionactis is hinged to the rear edge of the nostril. As it opens, it pulls back toward the eye and slightly inward. The movement can be easily seen in a live specimen. I am unable to determine whether the opening and closing are coincident with breathing. The valve was never seen to close entirely, as it may when the creature is submerged in sand. A drop of water does not cause complete closure. The relative bluntness of the palarostris snout, as compared with occipitalis, is somewhat more evident in live than in preserved specimens. The point on the mental seems somewhat less conspicuous in the former species. The action of this tip in occipitalis indicates that there is some flexibility of the tip independent of the mental itself. There is a slot in the rostral into which this point fits, so that the aperture through which the tongue is extruded may be tighdy closed. The jaws are slightly opened when the tongue is protruded. The tongue of both species is black or dark-brown, with white tips back almost to the bifurcation. When the snake is appraising its surroundings, the tongue is vibrated through 180 degrees, from vertically upward to downward. The pupil is round, the iris black, thus making the eye appear unicolor. C. palarostris may have a slighdy larger eye, proportionately, than occipitalis. As for intrageneric trends, the following will summarize the major differ- ences: The form with the proportionately longest tail is typical anniilata from Imperial and Yuma counties; the snakes of this subspecies have relatively shorter tails in the fringe areas both to the east of Yuma County, and in San Diego County to the west. C. o. occipitalis, alike in the Coachella Valley and in the western Mojave, is shorter-tailed than annulata, even shorter than the San Diego County population. In ventral scale counts, the maxima are reached in the warmest areas, the central Colorado and Yuma deserts, the Coachella Valley, and the western Mojave. This variation conforms to the general rule that snake populations inhabiting warmer areas have more ventral scutes than individuals of the same subspecies in cooler places. The foothill snakes in San Diego County have lower counts, and the same is true of the Arizona population east of Yuma County. C. p. palarostris, C. p. organica, and C. o. klauberi have fewer ventral plates than the other groups. The subcaudals follow tail proportionalities, annulata being high, and occipitalis low, with the San Diego County and south-central Arizona populations intermediate. The greatest variations are in pattern and colors. The snakes with the highest numbers of crossbands are those from the Coachella Valley, with the western Mojave occipitalis only a few below. C. p. palarostris has the fewest marks, annulata being intermedite. In the number of bands there is little difference between the Yuma, Imperial, and San Diego county populations. Farther east, in Arizona, including klauheri, the tendency is upward, so that the counts fall between those of annulata and those of occipitalis from the western Mojave. In the color of the primary series of bands, occipitalis and talpina are generally brown; annulata, klduberi, organica, and palarostris black. The Klauber— Shovel-I^osed Snakes 183 interspace suffusions run to yellow in occipitalis and to red in annidata, organica, and palarostris. Orange is occasionally observed in the western Mojave popula- tion. The most perfect annidata interspace reds are found in Imperial and western Yuma counties; here they are deep in color, and with even edges sharply contrasting with the white or buff ground color. Westward in San Diego County the interspace colors tend to be lighter, narrower, and less sharply bordered; at the higher elevations they are often absent. Eastward of Yuma County the interspace secondary bands become narrower and darker, reaching a climax in klauben, in which the red is almost completely masked by maculations nearly as dark as the primaries. The same trend is noted from occipitalis northward toward talpina, except that in this case the maculations are brown. Both subspecies of palarostris have wider and deeper red secondary bands than any annulata. The total dark pigmentation (whether black or brown) seems to increase at the higher altitudes. Thus, in San Diego County, the dorsal bands are wider, proportional to the interspaces, than in Imperial County, and the bands widen ventrally to a greater degree. In south-central Arizona the same tendency is manifested by increased maculations in the interspaces, plus a widening of the ventral bands; it reaches the ultimate in klauben. In the same way the snakes to the north of the Mojave become darker at the higher altitudes, reaching the extreme in talpina. C. a. occipitalis tends, with considerable consistency, toward more dorsal but fewer ventral bands, as compared with annulata. Phylogeny I visualize the present geographical arrangement of the subspecies of Chionactis as, possibly, the result of three migrations from Sonora northward and westward. The first postulated migration was accompanied by the evolution of the brown, short-tailed snakes {occipitalis), of the Mojave Desert, and included an invasion of the Coachella Valley down to Lake LeConte by way of the Morongo route. C. o. talpina was a later northward extension and differentiation from the main occipitalis population in the south. The second inferred migration, that of annulata into the Yuma Desert and into the Colorado Desert (except the Coachella Valley), probably occurred not long before the termination of the ancient main stage of Lake LeConte (Hubbs and Miller, 1948, p. 104). The lack of difference between the snakes of the east and west sides of what was once the bed of this lake — now repre- sented by the East and West mesas bounding the Imperial Valley — indicates that these populations were not separated for any great length of time. The occipitalis tendencies observed in the San Diego County population of annulata suggest a merging of the two populations by reason of a drift of occipitalis around the southern end of the Santa Rosa Mountains during one of the Lake LeConte recessions. It seems probable that occipitalis was the later arrival in this area of composites since the isolated pockets, at La Puerta, for example, are nearer annulata. C. o. klauberi is an eastern offshoot of annulata. The final invasion was that of organica, which has advanced to the point of producing an intrusion into the annulata area in the vicinity of Organ Pipe 184 San Diego Society of Natural History Cactus National Monument, separating, around Ajo and Pozo Sipiano, two sections of an annidata population having some klauberi tendencies. The close similarity of the two annulata populations, particularly their klauberi peculiarities, suggests that they have not long been separated by the organica wedge. Ecology, Life History, and Habits Ecological and Temperature Preferences. — As we survey the snake fauna of desert areas, we note various criteria that indicate the degree to which the several genera have become addicted to or tolerate the arduous conditions found there. Some are clearly fringe invaders; that is, they filter into the edges of our southwestern deserts but are usually absent where the dry and barren conditions are most extreme. Typical of these are such forms as Lampropeltis getulus and Lichanura roseojuica. Others endure desert conditions, but show, by population increases when the conditions are ameliorated by agricultural development and irrigation, that it was toleration rather than preference which led them to an acceptance of the desert. Such, for example, are Pituophis catenijer affinis and Masticophis flagellurn piceus, snakes that, though always present in the Imperial Valley, have increased greatly in numbers since the coming of irrigation. Of the snakes that really thrive in the desert, some are represented by desert subspecies of much more widely distributed forms, among them Arizona elegans, Rhinocheilus lecontei, and Crotalus scutulatus. All three flourish in Upper Sonoran areas, Arizona and Rhinocheilus in California (and elsewhere as well), and C. scutulatus in Arizona. Of those whose greatest populations — whose headquarters, as it were — are in the desert, there are three: Phyllorbynchus decurtatus, Chionactis occipitalis, and Crotalus cerastes. All three, as shown by roadside observations in the Coachella, Imperial, Yuma, and Blythe agricultural areas, are driven out by irrigation. And we are struck by the virtual identity of their ranges. If we assume that decurtatus, perkinsi, and nubilus are all subspecies of Phyllorbynchus decurtatus — of which I am beginning to be rather doubtful — that species has somewhat the greatest range, for it reaches the Cape Region of Baja California, whereas the other two may not extend much below San Felipe on the Gulf coast. In San Diego County, Phyllorbynchus has reached San Felipe Valley, where it is found at the top of Sentenac Canyon, whereas the others stop at the bottom of the Canyon, some 4 miles farther down the slope. Phyllorbynchus is also found somewhat higher up the Palms-to-Pines grade in Riverside County than the other two. Yet on the Mountain Springs grade, on the San Diego-Imperial Border, Chionactis reaches a higher elevation than the others. At the northeastern corner of the range, in southwestern Utah, it is probable that Crotalus cerastes considerably outranges the other two; on the other hand, in the Tucson area of Arizona, the C. cerastes range stops at Cortaro while the others continue into Tucson. The ranges of the species are Klauber— Shovel-Nosed Snakes 185 imperfectly known in Sonora, but it appears that C. cerastes is the more restricted. But, by and large, despite these slight sorties of one or another species at the fringes, the congruence of their ranges is remarkable. In general, these three desert snakes prefer dry and sandy areas. Phyllorhynchus is likely to predominate where scattered stones abound, whereas the others prefer a looser soil. Since all are small snakes, an extensive uniformity of terrain is not required; to whatever extent sand is preferred, a sufficient amount will be found in the dry washes threading their way amid rocky hills or sun-baked mesas. Areas of extreme barrenness, with an almost complete lack of vegetation, are not densely populated, probably owing to a scarcity of food. Thus we find these sand-dwellers to be rare where there is a super- abundance of sand, in such places as the belt of sand hills some 17 miles west of Yuma, in Imperial County. They are virtually absent from some particularly barren stretches along the shores of the Salton Sea, although there is much sand about. The population of all three becomes more sparse as one proceeds eastward from Bensons Dry Lake, in eastern San Diego County, toward Kane Spring, Imperial County, a route along which the vegetation also becomes increasingly attenuated. Scattered sand hummocks, crowned with mesquite or other desert shrubs, are favorite refuges of both sidewinders and shovel-noses. Hillocks of coarse sand constitute the typical habitats of sand-loving forms according to Cowles (1941, p. 125). Since much hunting of Chionactis is done at night when the exact nature of the surroundings cannot be appraised, most specimens in my collection are not accompanied by ecological data. However, the following records will indicate the frequency of different surroundings under some rather broad generalizations : Sandy desert 55 Barren desert 54 Brushy desert 39 Rocky desert 6 Grassy desert 3 Some of the particular situations noted were: Date grove, rocky gorge, creosote bush, burro weed, mesquite hummocks. One of the earliest experiences tliat I had in Chionactis collecting was on the night of April 30, 1927, when 8 of us, with flashlights and Coleman lamps, hunted the mesquite hillocks in the vicinity of the old San Felipe Hotel in eastern San Diego County for two hours or more. We secured 3 of these little snakes, a meagre catch for the effort, and far less than would have rewarded the road-collecting method that was first tried a year or so later. The dark crossbands of these little snakes are admirably suited to blend with the shadows of the twigs scattered about on the sand, and we found them difficult to discover. On a dark, smooth pavement, picked out by the headlights of a car, they have no such protection. 186 San Diego Society of Natural History These little snakes are sometimes trapped in fresh road tar. Hubbs and Walker (1947, p. 464) reported finding 3 in a stretch of 5.8 miles of road in the eastern Mojave Desert, among many other creatures snared the night before. The air temperatures at which Chionactis has been found active on the roads at night are indicated in the following table: The Temperature DEGREES F. C. Nui 0. occ MBER OF SPECIMENS ipitalis C. 0. annulata 62-3 1 64-5 2 2 66-7 1 6 68-9 1 3 70-1 5 72-3 2 9 74-5 5 10 76-7 1 11 78-9 7 6 80-1 4 13 82-3 2 16 84-5 12 86-7 1 9 88-9 1 8 90-1 3 92-3 2 5 94-5 3 96-7 1 3 98-9 1 31 125 illecting times were as follows: Time When OBSERVED ] C. 0. MUMBER OF SPECIMENS occipitalis C. o. annulata 5:00- 5:29 P.M. 1 5:30- 5:59 6:00- 6:29 1 Klauber — Shovel-Nosed Snakes 137 6:30- 6:59 1 7:00- 7:29 4 7 7:30- 7:59 7 14 8:00- 8:29 4 34 8:30- 8:59 8 31 9:00- 9:29 4 26 9:30- 9:59 2 21 10:00-10:29 1 17 10:30-10:59 2 9 11:00-11:29 1 4 11:30-11:59 3 12:00-12:29 A.M. 1 1 12:30-12:59 1:00- 1:29 1:30- 1:59 2:00- 2:29 1 2:30- 2:59 34 171 These records indicate that Chionactis is most active at air temperatures between 70° and 90° F., and from 7:00 to 10:00 P.M., but certain cautions respecting the interpretation of the tables are necessary. As those of us who have written on the habits of our southwestern reptiles have frequently pointed out, the recorded air temperature is only an approximate indication of the temperature of the snake's body, for the creature is in contact with the ground and is subject to conduction of the ground temperature, and it is partly protected from the convection effect of the wind. In early evening, particularly in spring, the ground is almost invariably warmer than the air. It is probable that the snakes are abroad partly because of this fact. As to the time of activity, though it is unquestionably true that in the spring the snakes avoid the late hours when the temperature is suboptimum, the lack of late-hour records in the summer, when such hours are likely to be most suitable in temperature, is merely the result of inadequate collecting, for our collecting forays rarely extend beyond midnight. Our records certainly do prove definitely that all forms of Chionactis are almost exclusively nocturnal, for I have driven thousands of miles on desert roads in the daytime, and, in contrast with the many specimens recorded alive at night, those active diurnally have been rare enough to warrant individual mention. I found one on the east base of a large mesquite hummock at 6:10 P.M., just as the sun was cutting the horizon. This was in Imperial 188 San Diego Society of Natural History County, 4 miles west of Kane Spring Junction, April 20, 1935. A week later another was caught crossing the road at 5 P.M., in bright sunlight, 1 mile west of Grays Well, Imperial County. Charles E. Shaw found a C. o. annulata crossing the road at 9:40 A.M., May 14, 1947, 20 miles southwest of Pozo Sipiano, Sonora. The air temperature was 69° F. and a strong wind was blowing. He also found a juvenile issuing from a hole near Clark Dry Lake, San Diego County, at 5:10 P.M., when the sum was still shining brightly. This was in the early spring, on March 29, 1941. Dr. R. B. Cowles informs me that in many hundreds of desert trips he has seen Chionactis abroad in the daytime only twice. After sundown catching them is not so unusual, yet even in the spring they do not reach maximum activity until dusk has turned to darkness. That these little snakes reach their maximum activity in the early evening in the spring, when the later temperatures are much below the optimum, is easy to demonstrate, for the live specimens are found during the early hours, while later only DOR's remain as evidence of their earlier prevalence. In summer the best temperature conditions are undoubtedly reached just before sunrise, for, earlier in the evening, ground and air temperatures are too high for comfort and may even be lethal. However, so little collecting has been done after midnight in summer that we cannot say whether there is any considerable early-morning activity at that season. Since the snakes' prey is also forced to adopt a subterranean existence, food, to the limited extent that it is required, may be found below ground, without the risk of surface activity. Collecting up to midnight on several hot nights in June, C. B. Perkins had meagre results, but next morning a number of DOR's were found, indicating that there had been an accelerated post-midnight activity. Although statistics based on air temperatures are not trustworthy proofs of optimum body temperatures, I should judge from field and laboratory experiences with Chionactis that such temperatures are probably in the range of 26° to 28° C. (78.8° to 82.4° F.). Cowles and Bogert (1944, p. 286) report the highest voluntarily accepted temperature in captivity as 31° C. (87.8° F.), the critical maximum at 37° C. (98.6° F.), with recovery appear- ing at 33° C. (91.4° F.). Under laboratory conditions, the snakes lost the power of co-ordinated action at 38° C. (100.4° F.). The minimum voluntary tolerance is given as 20° C. (68° F.) . The desert is usually windy on spring nights, but this does not completely discourage the nocturnal forays of these little snakes. I find among my notes the following designations of wind conditions at the time of the capture of Chiotiactis on the road: Windy, very windy, hard wind, strong wind, cold strong wind. Occasionally one sees these and other little snakes blown across the road, so strong is the gale. Although specimens of Chionactis have been brought to the San Diego Zoo or Natural History Museum during every month of the year, it is probable that those representing the months of November to February, inclusive — about a dozen in 27 years — were not indicative of surface activity, but were dug out of their winter refuges in the course of agricultural or road work. 4 2 16 75 161 49 107 23 5 5 2 Klauber — Shovel-Nosed Snakes 189 Probably a better criterion of seasonal activity is furnished by the following table showing when specimens have been found alive or dead on the road at night: Number of specimens Month C. o. occipitalis C. o. annulata January February March April May June July August September October November December 126 323 Of course, these figures, like our other statistics, are premised on the coincident activities of snakes and collectors, and it must be admitted that collecting is neglected in seasons other than spring. However, the reason for our summer inactivity is mainly the result and not the cause of the strong spring peak, for experience soon taught us that collecting was virtually fruitless after the end of June. A more accurate criterion of relative seasonal activity may be derived from statistics of the shovel-noses encountered on the road (dead or alive) per mile of night travel. The following statistics are based on 20,768 miles of travel on desert roads at night in the months of March to October, inclusive, during which 324 specimens of Chionactis were recorded.* The resulting statistics in snakes per 100 miles of travel are: Specimens of Chionactis per Month 100 miles March .00 April ^ .47 May 2.00 June 2.09 July .86 August .69 September .27 October .00 *The difference between the total number of snakes listed in this and the previous table results from the fact that the first table includes data on snakes found on night trips when mileage records were not kept. 190 San Diego Society of Natural History It will be seen that there is a sudden rise in activity in May and June (the actual peak is probably during the first week of June), followed by a decline through the rest of the summer. Our experience indicates that the season is somewhat earlier in the Borrego area of eastern San Diego County than in the Mojave Desert. At the peak season in Borrego, and under favorable conditions of weather, the take will often exceed 10 shovel-noses per 100 miles of desert travel. The much lower average values result from unfruitful trips during the many unseasonable cold nights that are encountered on the desert in spring and early summer. That autumn collecting is unrewarding, even when spring weather condi- tions are duplicated, has lately been verified by Charles E. Shaw. He cruised the road between Scissors Crossing and Bensons Dry Lake, in the Borrego area of eastern San Diego County, at weekly intervals from early spring to autumn. He found the road to be virtually barren of snakes on the autumn nights (2 snakes in 138 miles of travel on Sept. 23, 30, and Oct. 7) compared with the spring nights (40 snakes in 138 miles, on May 27, June 2 and 8), notwithstanding a similarity in temperature and other weather conditions. These statistics refer to snakes of all kinds, not Chionactis exclusively. As an indication of the concentration sometimes reached, it may be noted that Shaw and the writer collected 10 specimens of C. o. annidata in 17 miles along the Yaqui Well-Bensons Dry Lake road in the Borrego area of eastern San Diego County on June 9, 1939. The best Mojave Desert record was made in the Mojave-Kramer-Adelanto area, where Shaw found 10 individuals of C. o. occipitalis in 43 miles of travel, on June 5, 1949. The following night 8 were taken. William R. Supernaugh, Superintendent of Organ Pipe Cactus National Monument, reports that the C. p. organica specimens collected in that vicinity were found prior to June 20 and between the hours of 8:30 and 9:45 P.M. I believe that Chionactis is somewhat less tolerant of unfavorable nocturnal weather conditions in the spring than are the other, larger common snakes of the desert, namely, Rhinocheilus , Arizona and especially, Crotalus cerastes. Our collecting experiences on the desert indicate that all these nocturnal snakes prefer the dark of the moon to moonlight, though the available statistics do not yet supply absolute proof of this theory. Relative Populations. — In many desert areas, Chionactis is one of the commonest snakes. The following statistics indicate the relative frequency of occurrence of the 5 most plentiful nocturnal snakes found in three desert areas of southern California, as indicated by records of specimens observed alive or dead on the road during the past 20 years: Western Mojave Desert Crotalus cerastes cerastes 132 Arizona elegans Candida 96 Rhinocheilus lecontei lecontei 86 Klauber— Shovel-Nosed Snakes 191 Crotalus scutidatus scutulatus 79 Chionactis occipitalis occipitalis 61 COACHELLA VaLLEY, RiVERSIDE CoUNTY Crotalus cerastes laterorepens 59 Phyllorhynchus decurtatus perkmsi 51 Arizona elegans eburnata 50 Chionactis occipitalis occipitalis 42 Rhinocheilus lecontei clarus (or lecontei-clarus intergrades) 15 BoRREGO Area, San Diego County Phyllorhynchus decurtatus perkmsi 523 Chionactis occipitalis annulata 397 Crotalus cerastes laterorepens 196 Arizona elegans eburnata 153 Rhinocheilus lecontei clarus (or lecontei-clarus intergrades) 105 The numbers of specimens recorded from the different areas are not to be taken as indicating the relative snake concentrations between districts; the high records from the Borrego area merely represent many more trips than were made to the Mojave or Coachella sections. It will be observed that Chionactis falls within the first 5 in each area. It is probable that this small snake (and Phyllorhynchus as well) represents a somewhat higher percentage of the snake population than these figures indicate, since the small snakes are more likely to be overlooked, whether alive or dead, than the larger forms, Crotalus, Arizona, and Rhiiiochedus. Of course, these figures involve generalities based on areas of considerable extent and ecological diversity. Within these areas there are smaller, more uniform sections in which one or another species may predominate. For example, there is a short stretch of road at the top of the Sentenac Canyon, in San Diego County, where the worm snakes (Leptotyphlops humilis in the form of L. h. humilis-cahuilae intergrades) are exceedingly common, far outnumbering all other snakes taken together. So it is with the other small desert snakes. It is probable that the Chionactis population exceeds Phyllorhynchus in the sandier sections of the Colorado Desert. For example, Cowles (1941, p. 134) reported that in an area of sand hummocks on the northeast side of Coachella Valley, out of 96 reptiles (including lizards) excavated by land grading operations during the season of hibernation, 41 were C. o. occipitalis. This census, probably representative of the population, indicated this snake, locally, to be much the most common form. Locomotion. — Normally Chionactis uses the horizontal undulatory type of locomotion. In soft sand it leaves a sinusoidal track with wide side loops. 192 San Diego Society of Natural History The sand is piled up on the rearward edge of each crest, so that it is easy to determine the direction of travel. Mosauer (1932, p. 11) pointed out that during motion the ventral plates are drawn in through action of the abdominal muscles and that this serves both to form a concave ventral surface and to sharpen the angle formed by the ridge at the outer ends of the ventrals, this ridge being characteristic of the genus. The ridge is thought to facilitate motion in the sand by reducing side slippage. Chionactis will sidewind if badly frightened or stimulated by the dis- comfort of an uncomfortably warm surface. It will also attempt this type of locomotion on a smooth surface such as glass or paper. The loops formed are not as deep as those of the sidewinder {Crotalus cerastes), and though the writhing is rapid, the motion is not particularly effective in net travel. Even so, on the smooth surface of a paved desert road, Chionactis is more often successful in escaping than any other night snake encountered. In its nightly cruises, it is evident that Chionactis spends most of its time on the surface. Although it can burrow with rapidity, and can progress below the surface of sand with good efficiency, it will not normally seek to escape by burrowing, as do some other sand-dwellers, Chilomeniscus and Anniella for example. Placed in a box containing yielding sand it will spend much time crawling about on the surface searching for a chance to escape up the sides. In so doing, the anterior part of the body is raised well above the ground, so that only the posterior part is used for progression. These observations of locomotion are equally applicable to occipitalis and palarostris. Although the latter has a blunter snout, it was observed on several occasions to enter sand quite rapidly and with little apparent effort. Fossorial Activity. — Mosauer (1932, p. 11) first expressed the opinion that Chionactis travels only short distances on the surface, spending more of its time in the sand, where it crawls below the surface using a horizontal undulatory motion. But later (1933, p. 15; 1935, p. 20; 1936, p. 13) he decided that most, if not all of its night cruising is done on the surface, as it travels from one shrub-covered hillock to another. He doubted whether more than 100 feet was usually covered in one night. Mosauer's observations were based on tracks, and it is now known that Chionactis tracks cannot always be distinguished from those of other small desert snakes, as has been verified by Cowles (1941, p. 139) . However, there is no reason to doubt Mosauer's conclusion that although Chionactis burrows quite readily and efficiently in sand, often spending the day thus buried, and though it progresses through sand with ease, it does not normally move for as great distances below the sand surface as do some sand-dwellers among the snakes. In seeking refuge in the daytime, Chionactis does not always burrow in sand; it may take to mammal holes or hide under debris. I have found shovel-noses under stones, boards, and cardboard signs. Wade Fox, Jr., advised me that he had dug two specimens from holes, an occipitalis near Garnet, Riverside County, and an annulata in the Sand Hills, 14 miles west of Winterhaven, Imperial County. He found them by following the tracks that Klauber— Shovel-Nosed Snakes 193 showed in which burrows the snakes had taken refuge. He judged the holes to have been made by scorpions. These little snakes are frequently plowed out during road work or land- leveling for agriculture, but in such cases it is usually impossible to tell whether they had been down holes or buried in sand. Cowles (1941, p. 134), in the experience previously mentioned, reported that the 41 specimens of Chionactis were taken during 41/2 days of brushing or rough grading of hummocks and small sand dunes with a tractor and scraper near Edom, Riverside County. This was in the early spring during the season of hibernation. The depths of the few specimens whose location could be determined varied from 2 feet to only an inch. Most of them were within a foot of the surface. More recently Cowles has stated (letter of March 3, 1949) : "As I suggested in one of my earlier papers, we discovered that Chionctis inhabits the warm, near-surface layers of the sand during the daytime, and thus secures ideal temperatures in total darkness. The snakes were found 2 or 3 inches below the ground surface, at temperatures varying from 26° to 28° C. (78.8° to 82.4° F.)." In another letter. Dr. Cowles remarks: "We caught ChionactK through a new technique of raking down the warm surface of the sand early in the Spring and I think that, properly applied, it might be a profitable method for collecting spade-nosed snakes." This suggests a sort of subter- ranean basking that is even more efficient and certainly safer than surface basking, when air temperatures are below the optimum and a reptile seeks the benefits of direct solar radiation. For, by this habit there is no loss of heat through convection — particularly important if there be a breeze — and the decline in temperature from the surface downward permits the choice of whatever depth will produce the most comfortable body temperature. The methods of burrowing used by Chiondctis may be readily observed with captive specimens, especially when kept in a box with one glass side. First, it will be noted that the snakes do not immediately seek refuge in the sand; on the contrary, they may crawl about on the surface for an hour or more before going below. They cannot be frightened into the sand. However, if they once go into the sand, and then are disturbed by a probe jabbed near them, they will move elsewhere in the sand without emerging. If picked up and the fore part of the body is stuck into the sand, the snake will usually go in. Once in the sand, a snake will often remain for hours at a time. When crawling below the surface, Chionactn usually progresses at a depth of an inch or so; faint movements of the sand surface can often be seen, but the snake is too deep to leave a surface track as does Chilomeniscus, which often crawls just below the surface. Shovel-noses progress in the sand by an undulatory motion and seemingly with little more effort than when on the surface; the polished scales are no doubt a favorable factor in this type of locomotion, as pointed out by Mosauer (1935, p. 20). Chionactis can back up in sand without difficulty. If a snake has been buried in the sand and is dragged to the surface, it will usually crawl about the cage without going into the sand again. When watching the snakes through a glass sidewall of a cage, some will be seen to descend to a depth exceeding 6 inches. 194 San Diego Society of Natural History Food. — I have records of scorpion, centipede, and insect remains in Chionactis. Richard Schwenkmeyer has advised me that, in his studies of the stomach contents of southwestern reptiles, he found the following in annulata: Uta (egg containing an embryo), Blattaria (Phyllodromidae), spiders, ab- dominal plates of Coleoptera, Coleoptera larvae, Hymenoptera (Formicidae), Lepidoptera pupa, unidentifiable insect pupa. Cowles (1941, p. 134) reported one that contained some form of Blattidae; in 1949 (p. 212) he stated that the diet comprises insects and other arthropods, including fairly large and powerful scorpions. Captive specimens ate centipedes. David Regnery found that shovel-noses would eat termites offered on the end of a stick. The type specimen of palarostris contained a large spider. Charles E. Shaw has devised a successful method of keeping Chionactis alive in captivity. He reports: "At the Reptile House of the San Diego Zoo specimens of Chionactis occipitalis annulata are kept in a glass terrarium measuring 20 in. X 10^ in. X 10 in. A glass partition 7 in. high is inserted transversely to separate one-third of the tank from the remainder. About 6 in. of sand are placed in this smaller compart- ment, while the rest of the tank is filled with 6 in. of yellow cornmeal well stocked with mealworms {Tenebrto molitor) . Chio7iactis will eat the larvae and pupae of Tenebrto. The snakes are liberated on the sand and are provided with cover in the form of a small board under which they may conceal themselves without burrowing, if they prefer. Most of the time, however, they remain beneath the surface of the sand. The snakes have no difficulty crossing the one inch of glass partition which extends above the surface of the sand and cornmeal and serves primarily to discourage the mealworms from crossing over into the sand. Kept in the manner described, 2 adults of Chionactis o. annulata are still in good condition after 36 months in captivity."* It is somewhat difficult to cause Chionactis to drink; however, it will do so if the head be held under water. Reproduction. — I have examined 7 females that contained enlarged eggs, the number varying from 2 to 4, average 3.1. The smallest gravid annulata measured 289 mm. (11% in.). The largest eggs measured 14X41/2 and 13X5 mm. Cowles (1941, p. 134) reported one occipitalis containing 6 and another 9 ova. A specimen of organica contained 4 eggs.f On May 27, 1939, a live male C. o. occipitalis was found hovering over a DOR female as if endeavoring to mate. This was at 8:41 P.M., air temperature 24° C. (75.2° F.). Dejejise. — When annoyed Chionactis stands its ground and lunges for- ward violently and repeatedly from a striking coil. Sometimes the mouth is *One of these was still alive at 42 months. fl find no evidence of embryos as suggested by Hensley, 1950, p. 283. Chionactis is probably oviparous. Klauber— Shovel-Nosed Snakes 195 opened at the end of the strike, but often not. In daylight it can hit one's finger with considerable accuracy and will even follow it about, striking continuously. Presumably the strike might intimidate a small creature; the teeth are much too small to puncture the skin of one's hand. As do many other snakes, Chionactis sometimes voids excrement when handled. Little is known concerning the senses of Chionactis. It can see well in bright light, since it will strike at a finger with fair accuracy, and will dodge hand movements. It can see such motions at a distance of a foot in dim light. Enemies. — Presumably Chionactis is preyed upon by Lampropeltis, Salva- dor a, Rhinocheilus, and Arizona, which feed on snakes, among other kinds of food. Owls and coyotes may be assumed to eat them. One was killed by a tame cat at La Puerta. I have only one definite record of a shovel-nose becoming the prey of a wild creature; at Palm Springs Station, Riverside County, a red racer (Masticophis flagellum ptceus) was found that had swallowed an adult Chionactis o. occipitalis. It is not known whether this diurnal snake had caught an early shovel-nose above ground, or had found it down a hole. A Key to the Genus Chionactis la Snout convex above; dark bands on the body usually fewer than 21; bands on the body plus unmarked anterior band positions on the ventrum usually fewer than 23* C. palarostris (for subspecies continue on to 2) lb Snout flat-topped; dark bands on the body usually 21 or more; bands on the body plus unmarked anterior band positions on the ventrum usually 23 or more* C. occipitalis (for subspecies continue on to 3) 2a Black bands on the body fewer than 12, and less than half as wide (middorsally) as the inter- spaces; more than 8 scales (end-to-end) between the parietals and the anterior edge of the first black body band C. palarostris palarostris West-central Sonora, Mexico - 2b Black bands on the body more than 11, and more than half as wide (middorsally) as the interspaces; fewer than 9 scales (end-to-end) between the parietals and the anterior edge of the first black body band C. palarostris organica Southwestern Pima County, Arizona *The first criterion will place most specimens. If the dark hody bands number 18 to 22, however, the second criterion should be used, as it will result in a higher accuracy of determination. With either criterion, tail bands are not mcluded. 196 San Diego Society of Natural History 3a 3b 4a 4b 5a 5b Definite black or brown secondary bands in the interspaces between the primary bands, usually with scale centers maculated, and often crossing the dorsum Without black or brown secondary bands in the interspaces between the primary bands, such dark marks as there are being usually restricted to scale edges and rarely evident dorsally Primary bands black or very dark-brown; ventrals usually fewer than 152 in males and fewer than 160 in females. Southwest Maricopa, and Pima and Pinal counties, Arizona Primary bands brown; ventrals usually 152 or more in males and 160 or more in females Tonopah-Indian Springs area of Nevada, and southwesterly into the Death Valley region of California. Dark bands on the body plus unmarked anterior band positions on the ventrum usually fewer than 45; bands usually black or almost black Eastern San Diego, and western and southern Im- perial counties, California; Yuma, western and southern Maricopa, and northwestern Pima counties, Arizona; and northeastern Baja Cali- fornia and northwestern Sonora, Mexico. Dark bands on the body plus unmarked anterior band positions on the ventrum usually 45 or more; bands brown The Mojave Desert, including southwestern Inyo, eastern Kern, northeastern Los Angeles, and San Bernardino counties in California; also desert part of Riverside County and northeastern Im- perial County, California; southern Clark County, Nevada; and southwestern Mohave County, Arizona. C. occipitalis klcinberi C. occipitalis talpina C. occipitalis annulata C. occipitalis occipitalis Klauber — Shovel-Nosed Snakes 197 Summary This is a survey of the shovel-nosed snakes of the genus Chionactis (Colubridae) , small nocturnal snakes inhabiting sandy areas in the deserts of the southwestern United States and northwestern Mexico. There are 2 species, C occipitalis and C. palarostris; C. occipitalis is, in turn, divided into 4 subspecies: occipitalis, annulata, klauberi, and talpina. The latter, occurring in southern Nevada and the Death Valley region of California, is newly described. C. palarostris comprises 2 subspecies, palarostris and organica. The latter is a new subspecies found in the Organ Pipe Cactus National Monument of southern Arizona. Relationships are discussed and life-history notes given. Acknowledgments To the following individuals and institutions that have granted me the courtesy of examining specimens contained in their collections, I wish to extend my thanks: Mr. Charles M. Bogert, American Museum of Natural History; the late Dr. Charles T. Vorhies, University of Arizona; Mr. Joseph R. Slevin, California Academy of Sciences; Mr. M. Graham Netting, Carnegie Museum; Dr. Howard K. Gloyd, Chicago Academy of Sciences; Mr. Karl P. Schmidt and Mr. Clifford H. Pope, Chicago Natural History Museum; Dr. A. H. Wright, Mrs. Kay Kapp, and Mr. M. Max Hensley, Cornell University; Dr. Hobart M. Smith and Mr. W. L. Burger, University of Illinois; Mr. Victor H. Housholder of Phoenix, Ariz.; Mr. M. Sullivan, Lake Mead Recreational Area; Dr. Howard R. Hill, Los Angeles County Museum of History, Science and Art; Dr. Edward H. Taylor, University of Kansas; Dr. Norman Hartweg, University of Michigan; Mr. Arthur Loveridge, Museum of Comparative Zoology, Harvard University; Dr. Robert C. Stebbins, Museum of Vertebrate Zoology, University of California; Dr. Raymond B. Cowles, Mr. Charles H. Lowe, Jr., Mr. Kenneth S. Norris, and Mr. Richard G. Zweifel, University of California, Los Angeles; Dr. Emmett R. Dunn, Academy of Natural Sciences of Philadelphia; Mr. William R. Supernaugh, Organ Pipe Cactus National Monument; Mr. Wade Fox, Jr., University of Southern California; Dr. George S. Myers and Miss Margaret Storey, Natural History Museum, Stanford University; Dr. Waldo L. Schmitt and Dr. Doris M. Cochran, United States National Museum; and Mr. William H. Woodin III, of Tucson, Arizona. 1 wish to thank Dr. Carl L. Hubbs, Mr. C. B. Perkins, Mr. Charles E. Shaw, and Dr. Joshua L. Baily for critical comments and suggestions. I was assisted in making scale counts and also in the compilation of field notes by Charles E. Shaw, Richard Schwenkmeyer, and Bayard Brattstrom. Mr. Leslie C. Kobler prepared the map and took the photographs. 198 Saj-i Diego Society of Natural History Bibliography Baird, Spencer F. 1859a. Reptiles of the Boundary, in: United States and Mexican Boundary Survey, etc. (Emory), vol. 2, pp. 1-35. 1859b. General Report upon the Zoology of the Several Pacific Railroad Routes, vol. 10, part 3, Reptiles, pp. 1-16. Boulenger, George A. 1894. Catalogue of the Snakes in the British Museum (Natural History) , vol. 2, pp. xi+382. Brown, Arthur E. 1901. A Review of the Genera and Species of American Snakes, North of Mexico. Proc. Acad. Nat. Sci. Phila., vol. 53, part 1, pp. 10-110. Brown, John S. 1920. Routes to Desert Watering Places in the Salton Sea Region, California. U. S. Geol. Surv., Water-Supply Paper no. 490-A, pp. v+86 Bryan, Kirk 1922. Routes to Desert Watering Places in the Papago Country, Arizona. U. S. Geol. Surv., Water-Supply Paper no. 490-D, pp. vi+4+3 17-429. Camp, Charles L. 1916. Notes on the Local Distribution and Habits of the Amphibians and Reptiles of Southeastern California in the Vicinity of Turde Mountains. Univ. Calif. Pubs, in Zobl., vol. 12, no. 17, pp. 503-544. Cooper, J. G. 1870. The Fauna of California and its Geographical Distribution. Proc. Calif. Acad. Sci., vol. 4, part 2, pp. 61-81. Cope, Edward D. 1860. Catalogue of the Colubridae in the Museum of the Academy of Natural Sciences of Philadelphia, with Notes and Descriptions of New Species. Part 2. Proc. Acad. Nat. Sci. Phila., vol. 12, pp. 241-266. 1875. Check-list of North American Batrachia and Reptilia. Bull. U. S. Nat. Mus., no. 1, pp. 1-104. Klauber— Shovel-Nosed Snakes 199 CowLES, Raymond B. 1920. A List and Some Notes on the Lizards and Snakes Represented in the Pomona College Museum. Journ. Ent. & Zool., vol. 12, no. 3, pp. 63-66. 1941a. Evidence of Venom in Hypsiglena ochrorhynchus. Copeia, no. 1, pp. 4-6. 1941b. Observations on the Winter Activities of Desert Reptiles. Ecology, vol. 22, no. 2, pp. 125-140. 1949. Tracks in Desert Dunes. Natural History, vol. 58, no. 5, pp. 206-212. CowLES, Raymond B. and Bogert, Charles M. 1944. A Preliminary Study of the Thermal Requirements of Desert Reptiles. Bull. Amer. Mus. Nat. His., vol. 83, art. 5, pp. 261-296. Fitzinger, Leopold J. 1826. Neu Classification der Reptilien. Wien, pp. [vi}+66. Garman, Samuel 1883. The Reptiles and Batrachians of North America. Mem. Mus. Mus. Comp. ZooL, vol. 8, no. 3, pp. xxxi+185. Gloyd, Howard K. 1937. A Herpetological Consideration of Faunal Areas in Southern Arizona. Bull. Chicago Acad. Sci., vol. 5, no. 5, pp. 79-136. Grinnell, Joseph and Camp, Charles L. 1917. A Distributional List of the Amphibians and Reptiles of Cali- fornia. Univ. Calif. Pubs, in ZooL, vol. 17, no. 10, pp. 127-208. Hallowell, Edward 1854. Descriptions of Reptiles from California. Proc. Acad. Nat. Sci- Phila., vol. 7, pp. 91-97. 1856. Description of a New Genus of Colubriform Serpents from Cahfornia. Proc. Acad. ^ Nat. Sci. Phila., vol. 8, pp. 310-311. 1859. Report upon the Reptiles Collected on the Survey, in: Report of Explorations for a Railway Route, etc. (Williamson), vol. 10, pp. 1-27. Hensley, M. Max 1950. Results of a Herpetological Reconnaissance in E.xtreme South- western Arizona and Adjacent Sonora, with a Description of a New Subspecies of the Sonoran Whipsnake, Masticophis biUneatus. Trans. Kansas Acad. Sci., vol. 53, no. 2, pp. 270-288. 200 San Diego Society of Natural History HuBBS, Carl L. and Miller, Robert R. 1948. The Great Basin, with Emphasis on Glacial and Post Glacial Times. II. The Zoological Evidence: Correlation between Fish Distribution and Hydrographic History in the Desert Basins of Western United States. Bull. Univ. Utah, vol. 38, no. 20, pp. 17-166. HuBBS, Carl L. and Walker, Boyd W. 1947. Abundance of Desert Animals Indicated by Capture in Fresh Road Tar. Ecology, vol. 28, no. 4, pp. 464-6. Klauber, L. M. 1924. Notes on the Distribution of Snakes in San Diego County, California. Bull. Zobl. Soc. San Di€go, no. 1, pp. 1-25. 1937. A New Snake of the Genus Sonora from Mexico. Trans. San Diego Soc. Nat. Hist., vol. 8, no. 27, pp. 363-366. 1939. Studies of Reptile Life in the Arid Southwest. Bull. Zool. Soc. San Diego, no. 14, pp. 1-100. 1943. Tail-length Differences in Snakes with Notes on Sexual Dimorphism and the Coefficient of Divergence. Bull. Zobl. Soc. San Diego, no. 18, pp. 1-60. 1945. Herpetological Correlations. I. Correlations in Homogeneous Populations. Bull. Zool. Soc. San Diego, no. 21, pp. 1-101. 1946. The Glossy Snake, Arizona, with Descriptions of New Subspecies. Trans. San Diego Soc. Nat. Hist., vol. 10, no. 17, pp. 311-398. Mendenhall, Walter C. 1909. Ground Waters of the Indio Region, California, with a Sketch of the Colorado Desert. U. S. Geol. Surv., Water-Supply Paper no. 255, pp. 1-56. MosAUER, Walter 1932. Adaptive Convergence in the Sand Reptiles of the Sahara and of California: A Study in Structure and Behavior. Copeia, no. 2, pp. 72-78. 1933. Locomotion and Diurnal Range of Sonora occipitalis, Crotalus cerastes, and Crotalus atrox as Seen from their Tracks. Copeia, no. 1, pp. 14-16. 1935. The Reptiles of a Sand Dune Area and its Surroundings in the Colorado Desert, California: A Study in Habitat Preference. Ecology, vol. 16, no. 1, pp. 13-27. 1936. The Reptilian Fauna of the Sand Dune Areas of the Vizcaino Desert and of Northwestern Lower California. Occ. Papers Mus. Zobl., Univ. Mich., no. 329, pp. 1-21. Klauber— Shovel-Nosed Snakes 201 Perkins, C. B. 1949. The Snakes of San Diego County with Descriptions and Key. Second Edition. Bull. Zool. Soc. San Diego, no. 23, pp. 1-77. Richardson, Charles H., Jr. 1910. Notes on a Litde-known Species of Snake, Chionactis occipitalis. Science, vol. 32 (n.s.), no. 820, pp. 383-4. RiDGWAY, Robert 1912. Color Standards and Color Nomenclature. Washington, pp. iii+43, 53 plates. Ross, Clyde P. 1922. Routes to Desert Watering Places in the Lower Gila Region, Arizona. U. S. Geol. Surv., Water-Supply Paper no. 490-C, pp. iv+315. Stickel, William H. 1938. The Snakes of the Genus Sonora in the United States and Lower California. Copeia, no. 4, pp. 182-190. 1941. The Subspecies of the Spade-nosed Snake, Sonora occipitalis. Bull. Chicago Acad. Sci., vol. 6, no. 7, pp. 135-140. 1943. The Mexican Snakes of the Genera Sonora and Chionactis with Notes of the Status of Other Colubrid Genera. Proc. Biol. Soc. Wash., vol. 56, pp. 109-127. Sykes, Godfrey 1937. The Colorado Delta. Am. Geogr. Soc, Spec. Pub. no. 19, pp. vii+193. Thompson, David G. 1929. The Mohave Desert Region, California. U. S. Geol. Surv., Water-Supply Paper no. 578, pp. xi+759. Van Denburgh, John 1922. The Reptiles of Western North America. Occ. Papers Calif. Acad. Sci., no. 10, 2 vols., pp. 1-1028. Van Denburgh, John and Slevin, Joseph R. 1913. A List of the Amphibians and Reptiles of Arizona, with Notes on the Species in the Collection of the Academy. Proc. Calif. Acad. Sci., Ser. 4, vol. 3, pp. 391-454. Klauber — Shovel-Nosed Snakes Plate 9 Fig. 1. Chionactis occipitalis occipitalis Mojave Desert Shovel-nosed Snake Adult male from the Alvord Mountains, San Bernardino County, Cahfornia. Fig. 2. Chionactis occipitalis annulata Colorado Desert Shovel-nosed Snake Adult male from 2 miles south of Clark Lake, Borrego Valley, San Diego County, California. Klauber — Shovel-Nosed Snakes Plate 10 / «%•• i Fig. 1. Chionactis occipitalis talpina Northern Shovel-nosed Snake Adult female from 10 miles north of Goldfield, Esmeralda County, Nevada. Fig. 2. Chionactis palarostris organica Organ Pipe Shovel-nosed Snake Adult male from 9 miles north of U.S.-Mexican Boundary on Ajo-Sonoyta road, Pima County, Arizona (type specimen). H -^ y J)\c^Ci TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 10, pp. 205-256 Figs. 1-2, Maps 3-6, Tables 1-3 THE KANGAROO RATS (DIPODOMYS) OF BAJA CALIFORNIA, MEXICO BY Laurence M. Huey Curatar o/ Birds and Mammals, San Diego Society of Natural History mi. COMP. ZOOL MAY 14 1951 SAN DIEGO, CALIFORNIA Printed for the Society April 30, 1951 COiMMITTEE OlN PUBLICATION Joshua L. Baily Carl L. Hubbs Laurence M. Klauber UBfiART MAY ' ' l^''' TABLE OF CONTENTS Merriami Group Range Habitat Climatic Influences Page .. 210 .. 210 .. 212 .. 213 Variation 214 Dorsal Color Variations (fig. 1) 215 Tail Tuff Color (fig. 2) 216 Material and Acknowledgments 218 Dipodomys merriami parvus 220 liolus 220 eremvagus quintinensis trinidadensis 220 221 222 223 223 224 225 226 226 227 228 229 " " semipallidus " " platycephalus " " annulus " " brunensis " " melanurus " " llanoensis " insularis " margaritae Distributional Map (Merriam Group) Table of Measurements (table 1) 230-31 Agilis Group 232 Range and Habitat 232 Climatic Influences 233 Material - 234 Dipodomys agilis cabezonae 234 " " simidans 234 Distributional Map (Agilis Group) 237 Table of Measurements (table 2) 238 Dipodomys agilis martirensis 239 " plectilts 240 " paralius 241 TABLE OF CONTENTS (Continued) Page Distributional Map (Peninsularis Group) - 243 Peninsularis Group 244 Range 244 Climatic Influences, Variation 244 Material and Acknowledgments 246 D'lpodomys peninsidans peninsularis 246 " " pedionomus 247 " " eremoecus 248 " " australis 249 Table of Measurements (table 3) 250 Distributional Map (Deserti and Heermanni Groups) 251 Deserti Group 252 D'lpodomys deserti 253 Heermanni Group 254 Material 255 D'lpodomys gravipes 255 THE KANGAROO RATS (DIPODOMYS) OF BAJA CALIFORNIA, MEXICO Laurence M. Huey Curator of Birds and Mammals, San Diego Society of Natural History During the past two decades representative collections of kangaroo rats (Dipodomys) from many localities in Baja California have been made by collectors from the San Diego Society of Natural History. The localities range from the international boundary on the north to Cape San Lucas on the south. There are few sections of Baja California wherein some members of this genus have not now been found. The writer has seen their tracks obliterated by the tides from the beach sands at San Felipe on the Gulf of California; he has dug out living specimens from their shallow burrows beneath driftwood on the sandy beaches within reach of the highest tides at Rosarita on the Pacific side near Tijuana; and his traps, set on the higher slopes of the Sierra Juarez and Sierra San Pedro Martir, have captured representatives of this genus from among the pine trees. Kangaroo rats are to be found along the international boundary from the Colorado River to the Pacific Ocean in the north, and in an irregular checkerboard pattern over the length of the peninsula to Cape San Lucas in the south. Two islands, one on either side of the peninsula, are known to be inhabited by kangaroo rats. The island on the western side, Santa Margarita, iForms part of the seaward boundary of Magdalena Bay. The other one, San Jose Island, is located a few minutes of latitude northward, but on the Gulf side. Its shores are washed by the tepid tropical waters of the Gulf. The only areas on the peninsula that seem to be avoided by kangaroo rats are the rock-bound or lava-strewn mesas, like those over the length of the Sierra de la Giganta, sterile alkali plains, and the higher jungle- covered reaches of the Sierra de la Victoria in the southern Cape District. Careful work in the Cape sierra may still reveal remote pockets where these mammals occur. Although the collections are not as yet complete, sufficient material has been accumulated to reveal the general systematic and distributional 210 San Diego Society of Natural History con- picture. The facts determined appear especially significant when sidered in the Ught of the study that has been made of similar collections from California (Grinnell, Univ. Calif. Pub. Zool., 24: 1922). Of the six groups of this genus listed by Grinnell from California, four have been found in Baja California, and a new group, peninsularis , is segregated from the agilis group. This newly proposed group is based on the form described by Merrian as Perodipus simulans peninsularis (Proc. Biol. Soc. Wash., 20: 1907). With but little material on which to reach his conclusion, Grinnell (Journ. Mam., 2: 1921) referred this on the form described by Merriam as Perodipus simulans peninsularis Today the writer, with magnificient collections at hand, believes that the elevation of peninsularis to specific rank, as the type of a new group, markedly simplifies the classification. The five groups thus constituted occur in Baja California in the following order of probable population abundance: merriami, agilis, peninsularis, deserti, and heermanni. Future collections may well change this order and reveal other races as yet unknown. Setzer (Univ. Kans. Pub. Mus. Nat. Hist., 1 (23) : 1949) has recently proposed some radical changes in Grinnell's classification. Until the proper material for the newer technique has been accumulated and studied, the writer believes that his purpose will be best served by adhering to Grinnell's natural groups. When the assembled collections are examined, the range pattern of each of these groups proves to be of great interest. This is especially true of the two largest groups. The following account deals separately with each group. Merriami Group Range The Dipodomys merriami group is most widespread in Baja Cali- fornia (Fig. l). It extends from along the desert section of the international boundary south to the southernmost parts of the peninsula and occurs on at least two of the nearby coastal islands. However, high elevations are unsuitable; seldom is it found in areas above 4500' altitude. Although it is essentially a desert mammal, Dipodomys merriami ranges onto the more humid Pacific slope west of the high coastal mountains in two places, one in southern California, and the HuEY— Kangaroo Rats of Baja California 211 other in northern Baja CaUfornia. In California, San Bernardino Valley, with its contiguous desert-like areas, is the only area of such range expansion. A parallel extension is to be found in Baja California, where, by way of the San Matias Pass, the group invaded the semi-arid El Valle de la Trinidad and thence continued northwestward into the more arid sections of El Valle de San Rafael, which lies against the western base of the Sierra Juarez. Below the southern extent of the higher Sierra San Pedro Martir, which terminates the long chain of lofty coastal mountains, the range of Dipodomys merriami reaches the Pacific side of the peninsula. From this section its range extends northward in a wedge along the coast to a point north of San Quintin. The coastal plain west of Santo Domingo Mission is, to date, the northermost coastwise record station. South- ward, Dipodomys merriami occurs over the plains and valleys along the Pacific coast to about 24° latitude in the Cape district. Paradoxically, below latitude 29° along the Gulf coast to La Paz, the positions of the desert and the more humid areas are reversed. In this central Baja California region the main stem of Dipodomys popula- tion is largely on the Pacific slope. Along the middle part of the Gulf side of the peninsula, the inhabitable areas are interrupted by rocky precipitous terrains where the higher mountains descend almost abruptly to the waters of the Gulf, leaving only occasional sections along the coastal shelf that are suitable for occupation by Dipodomys merriami. Three such areas are now known. The most extensive one extends from Bahia Los Angeles in latitude 29° southward to Barril near latitude 28°. The topography of the northern region, between Bahia Los Angeles and San Felipe, is, as yet, little known, and sufficient specimens from that section have not been available for this study. From Bahia Los Angeles and proximity a good series of Dipodomys merriami is at hand, as are also specimens from Valle de Agua Amarga, a mid-way inland locality along the gently sloping pass that connects with the Pacific watershed. Specimens from the Bahia Los Angeles, Barril and San Fran- cisquito area represent an unnamed race. Farther south, near Santa Rosalia, is to be found another of these areas. Here, two occupied localities are known : one, a hanging valley, El Valle de Yaqui, lies a few miles west of the town, and the other, a much larger area, El Llano de San Bruno, is south of it. Near the 212 San Diego Society of Natural History southern end of Concepcion Bay is the third region. Two inhabited localities are known in this region of open gravelly terrain, one, near the southern end of the bay and the other at Canipole, ten miles farther south. The two latter localities appear to be isolated, but future explora- tion may reveal connecting channels. The systematic status of these populations is dealt with on page 218. Westward from this central section of the Gulf coast rise the long chain of rugged lava mountains, La Sierra de la Giganta. Owing to the rough character of this area, transportation facilities, other than by pack animal, are almost nil. As a result, this section is but little known. However, when any near approach can be made, the general aspect of the terrain would seem to offer little that is suitable for habitation by this species. Near latitude 24° the merriam'i chain again leaves the Pacific side and is found at La Paz and thence at different localities of low elevation along the Gulf side of the peninsula until, at the extremity, the range of the species extends westward again to the westernmost point on the tip, Cape San Lucas. There seems to be no record of this or of any other Dipodomys along the Pacific slope section of the peninsula west of the Sierra Victoria, Habitat Soil conditions in the western desert regions below 4500' altitude seem to be one of the main factors controlling the range of Dipodomys meniami. It is commonly known to mammalogists who are acquainted with the habits of this species that its members are not good "diggers", that is, they cannot burrow into very hard soil. Some types of stone-bearing soil are tolerated, especially desert pavement, which is so common over the arid sections of the southwest. This terrain is most abundant over plains or llanos where the wind has blown the liner soil away, leaving a rock-bound surface that guards the arenaceous sub-surface soil. This is capital habitat for Dipodomys meniami; in fact, this species is the most abundant mammal in such regions. However, the areas that support the densest populations are those where the soil is loose and sandy, and in which both annual and perennial growths exist. In these places burrows are easily dug, food is adequate, and protection from enemies is supplied. HuEY— Kangaroo Rats of Baja California 213 Climatic Influences The highly diversified climate of the long narrow peninsula of Baja California is reflected in the divergent development of this rather homogeneous species. The climatic differences, however slight in some places, are in no small part attributable to the influences exerted by the adjoining coastal waters. Per unit of land area Baja California has a greater proportion of coast line than most like-sections of the globe. The bordering waters differ considerably in temperature. The eastern coast line is washed by semi-tropical seas and the general aridity of this part of the peninsula is outstanding. Off the western coast a cool current sweeps southward from the more northerly climes to meet a north flowing warm current in the general region of Point Eugenio and much of the coastline is greatly cooled by upwelling. The ocean influences along this western coastline are shown in the littoral and near-shore ocean flora: the mangroves of the tropics are found where the warmer waters lie, whereas the cool northern waters harbor fields of kelp. Both end abruptly where the ocean temperatures change most abruptly. The land flora of the peninsula likewise is affected, though the ranges are not so sharply terminated. The diverting influences of mountain chains on the atmospheric conditions created by the warmer or cooler ocean waters cause the areas of transition to be wider. In the northern section of the peninsula the higher mountains separate the influences of the Gulf and Pacific waters, resulting in the two greatest extremes of color development in the entire merriami group. The darkest race, Dipodomys merriami quintinensis (newly de- scribed herein), lives in the relatively cool and humid San Quintin region, whereas directly eastward on the torrid, arid desert east of the Sierra San Pedro Martir, occurs the most pallid race of the group, Dipodomys merriami areniragus. Farther south, where the elevation of the backbone of the peninsula decreases, the climatic conditions are more evenly balanced by the combination of the influences of the Gulf and Pacific waters. This results in a more widely spread flora and fauna. Here the lighter 214 San Diego Society of Natural History Dipodomys merges with the darker form of the cooler humid coastal section. In the Vizcaino desert region the lighter strains of Dipodomys merriami reach the Pacific side to replace the darker races of the northerly coastal section. To illustrate these points and other interesting facts brought to light by this study, I assembled a series of topotypical specimens repre- senting the different races from within the geographical limits of this report, along with two from north of the boundary in California. Com- mencing in the north with a series of topotypical specimens of D. m. parvus from the San Bernardino Valley region on the Pacific slope, and with a like series of D. m. simiolus from the northwestern Colorado desert section on the eastern side of the Coast Range of Southern California, representative series of specimens from all the type localities to the end of the peninsula of Baja California were arranged in their geographical position. Variation The kangaroo rats of the merriami group form three closely aligned color chains, which conform loosely to the geographical features of the regions under discussion (Fig. l). The races occupying the Pacific drainage slope, where the aridity is ameliorated by the cool coastal waters, are darkest and have a general dorsal color tone of " Avellaneous" (Ridgway, Color Standards and Color Nomenclature, 1912). Palest in dorsal coloration are the kangaroo rats in the region that extends from the Colorado desert area south along the eastern side of the high Coast Range, and thence, at a mid-peninsula point past the southern extent of the Sierra San Pedro Martir, veers diagonally south- westward across the peninsula to the vicinity of Point Eugenio on the Pacific coast. This general region includes most of the arid parts of the Southern California and Baja California deserts, and conforms loosely to the area that is influenced largely by warm ocean waters. The dorsal color tone of the races occupying this arid belt is "Pale Ochraceous Bnjf". The races living in the Cape district present yet another dorsal tone of coloration, "Pinkish Bujf". Along each chain the specimens from the successive type localities give the impression that each race is a step in the color gradient. The D. M. TRINIDADENSIS MEIDIUM D.M.QUINTINENSIS VERY DARK DM. SEMIPALLIDUS LIGHT D. M. ARENIVAGUS VERY LIGHT D.M. LLANOENSIS LIGHT DORSAL COLOR VARIATIONS OF THE RACES OF DIPODOMYS MERRIAMl FROM BAJA CALIFORNIA. MEXICO COLORS FROM RIDGWAY-I9I2 D.M. ANNULUS VERY LIGHT D.M.BRUNENSIS LIGHT D.M.MELANURUS DARK Fig. 1 .VICINITY SAN BERNARDINO. CALIF. EL VALLE DE TRINIDAD, MEIDIU SAN QUINT MEDIUM 7 MILES N. SAN LLANO DE SAN BRUNO MEDIUM"'>v I. LIGHT COLORED SERIES 2. MEDIUM COLORED SERIES 3. BLACK SERIES BUENA VISTA-;.; VERY dark'- DENSITY OF THE TAIL TUFT COLOR OF THE RACES OF DIPODOMYS MERRIAMI from baja california, mexico (type localities) SAN JOSE'^DEL CABO VERY DARK Rg.2 HuEY— Kangaroo Rats ov Baja Cai.ipornia 217 true situation is very different, for each race blends into the other, with broad overlap. Only from the vicinity of the key points along the chain can the clear color cast for the races be found. Other characters than dorsal coloration help to define the races. Some have well-defined cranial characters. This is particularly true of the auditory bullae. In some races the bullae are bulbously inflated (Table I). In several the maxillary arch has developed into wide proportions. Other cranial characters, although minute or of only average significance, with much overlap, in the aggregate help to set the races apart and help to define their ranges with greater clarity. Another character of importance shown by the assembled series is the color of the tail-tufts (Fig. 2) when the races are placed in the same geographical and type-locality sequence as was done for dorsal coloration. The color pattern of the tail tufts, like the dorsal color, presents three chains. The gradients do not parallel those shown by the dorsal coloration in one outstanding feature; only two colors are involved, black and white. The varying density of these two colors covers the entire range of variation of all the races. The variation shown by the tufts follows definite courses. As is to be expected, the specimens from the northern coastal section are darker than those from the more arid deserts, but, paradoxically, the specimens from the central Gulf coast and from the Cape districts have the darkest tails of all. There seems to be no logical explanation why the races occupying the central east-coastal section southward to the extreme peninsular position should have developed such an extremely black tail tuft, with a moderately light dorsal color, while the north-coastal races have de- veloped the darkest pelage color, with less contrasting tail tufts. Just what relationships this unique transposition of color intensity indicates is not answerable. In his "Geographical Study, of Kangaroo Rats of California", (Univ. Calif. Pub. Zool., Vol. 24, 1922, pp. 12-18) Grinnell advanced the thought, with rather convincing evidence, that cloudiness is partly responsible for color intensity. Regarding pelage colors, this relationship would seem to hold as true in Baja California as Grinnell demonstrated it did in California. The dorsal coloration of the specimens at hand, together with a very vague knowledge of weather conditions in Baja California, would seem to substantiate this relationship, but the color of the tail tufts, particularly in the Cape San Lucas series, would seem 218 San Diego Society of Natural History to open the "cloudiness theory" to some question. Presumably some other important factor, as yet unappreciated, is involved. In measurements as well as in coloration there is divergent develop- ment in the merriami group. Two measurements, total length of the specimen, made in the field at the time of preparation, and the greatest length of the cleaned skull, measured with visible dial calipers registering to 0.1 mm., amply confirm the variable development that has already been shown in the color comparisons. Dipodomys merriami is a composite of many similar races. There are no extremes of size, dwarfs or giants, in the assembly. The adherence of the species to certain circumscribed range conditions seems to have prohibited such development, yet there is a plasticity in the species that, as here demonstrated, responds, within limits, to certain environ- ments. Baja California seems to provide, by far, more varied influences on the species for racial development than is done by any other similar area in the extremely large section of southwestern North America occupied by the species. The changes are gradual, that is, there are no skeletal or color characters in different racial populations that change sharply (Fig. 2, Map 3 ) . The transition from lighter coloration to darker or from smaller size to larger is always a matter of degree. However, in the three chains of relationship there are boundable areas wherein certain character combi- nations are definable and constant. In such areas, even though the characters are minute and even though the race is restricted in its range, designation of this population by name seems justifiable. Material and Acknowledgements In preparing this report, 753 specimens of Dipodomji merriami from 83 localities in Baja California and 56 specimens from 7 localities in California have been examined. Of these 809 specimens 117 were borrowed from outside sources and 692 are in the Museums' collection. Fifty-one specimens from the Huey Collection were used to augment the Museum's series from localities in the north-central section of the peninsula. A much larger number of Californian specimens were available in the Museum's collection, but were not used. They were HuEY — Kangaroo Rats of Baja California 219 from many intermediate localities between the two California type localities and the international boundary. This is particularly true of the desert race D. m. simiolus, but no purpose would have been accomplished by their inclusion. Through the generosity of the late Major E. A. Goldman of the Biological Survey, now known as the United States Fish and Wildlife Service, 39 specimens from 12 localities, including the type of Dipo- domys platycephalus, taken on the famous trip made by Nelson and Goldman over the full length of the peninsula in 1905-1906, were loaned to the writer. Twenty-two specimens from two critical localities were loaned by the Museum of Vertebrate Zoology of the University of California at Berkeley by Dr. Alden H. Miller, and 3 specimens and 1 skull without skin, from Margarita Island, and 2 specimens from Santa Teresa Bay on the Gulf coast, were borrowed from the Dickey Collection at the University of California at Los Angeles through the kindness of the late Mr. A. J. van Rossem. To these institutions and the personnel, the writer wishes to express his gratitude. The writer also wishes to acknowledge his indebtedness to the several departments and officials of the Mexican Government. Without their cordial cooperation in granting permits for collecting in Baja California and the many courtesies given at the international boundary at the different ports of entry, this work would not have been possible. Lastly, the writer wishes to express his gratitude to Dr. Carl L Hubbs for his interest in this paper and the many helpful suggestions in the preparation of this manuscript. The present study has demonstrated the presence of two species and ten races of the Dipodomys merriami group within the confines of Baja California. Five races are herein described as new. That so many forms should be found in this limited area seems at first incredible. In fact, more races of Dipodomys merriami live in Baja California than in all the rest of the vast range of the species. Without question such variety is due in no small measure to three factors — geographical, meteorological, and sequestrational. No better region than Baja California can be found to illustrate the effects of these factors on the differentiation of a plastic species. 220 San Diego Society of Natural History Dipodomys merriami parvus Rhoads San Bernardino Kangaroo Rat Original Description. — Dipodomys parvus Rhoads, Am. Nat., Vol. 28, p. 70, January, 1894. Type locality.— '"San Bernardino, California" (= Reche Canyon, 4 miles southeast of Colton), California. The type and the small series of paratypes were taken by R. B. Herron at his old ranch in this canyon. Distribution. — This race is limited to a few localities west of the Coast Range in San Bernardino and Riverside counties, California. It is not found in Baja California, but reference to this form in this account makes its inclusion desirable. Specimens examined. — California: Herron's Ranch, San Bernardino Valley = Reche Canyon, 4 miles southeast of Colton, San Bernardino County (type locality), 4; 4 miles southwest of Ferris, Riverside County, 6; Ethenac, Riverside County, 2; Aguanga, Riverside County, 19. Dipodomys merriami simiolus Rhoads Allied Kangaroo Rat Original description. — Dipodomys simiolus Rhoads, Proc. Acad. Nat. Sci. Philadelphia, p. 410, January 27, 1894. Type locality.— Agua. Caliente (^ Palm Springs), Riverside County, California. Distribution. — Over the Colorado Desert in California east of the Coast Range at least to the Colorado River. In Baja California this race occupies a fringe along the international boundary from Jacumba Valley eastward to the Colorado River. Cultivated land in the desert section has restricted its range somewhat, but specimens from localities barely north of the boundary in California as at Pilot Knob, near the Colorado River in eastern Imperial County, Carrizo Creek and Jacumba Valley in eastern San Diego County, substantiate this statement. A short distance below the boundary this subspecies intergrades with D. m. arenivagus, particularly over the western desert section. Specimens examined. — California: Palm Springs, Riverside County (type locality), 24 (10 from Huey Collection). Baja California: few yards south of international boundary near Jacumba, California, 1; Signal Mountain, 2; southeast base Signal Mountain, 1. Dipodomys merriami trinidadensis subsp. nov. Trinidad Kangaroo Rat Type. — From Aguajito Spring, El Valle de la Trinidad, Baja California, Mexico; No. 11531, San Diego Society of Natural History; adult male; collected by Laurence M. Huey, March 13, 1936. HuEY— Kangaroo Rats of Baja California 221 Characters. — Compared with Dipodomys merriami arenivagus, D. m. trinidadensis is darker in dorsal color, more closely resembling D. m. parvus and showing distinctly the effect on coloration of the Pacific slope environmental conditions (Fig. 2, page 00) . The tail of trinidadensis is heavily striped above and below, in fact a greater area is covered by the darker dorsal and ventral stripes than by the white side stripes. This is in sharp contrast to the relative area covered by the ventral and dorsal stripes on the tail of arenivagus, in some specimens of which the ventral stripe is almost lacking. D. m. trini- dadensis is smaller then arenivagus and larger than parvus. Cranially the relationship with its desert relative is apparent, as there is a wide gap in general shape of the skull between trinidadensis and parvus. This is shown in a limited degree by the measurements (Table I, page 227), though the contour of the skull and the inflation of the bullae do not lend themselves well to measurements. Compared with arenivagus, trinidadensis has a more elongated skull, due to the more compressed mastoid bullae. The auditory bullae are larger, more inflated, and less trumpet shaped. Measurements of type. — Total length, 243; tail, 144; hind foot, 38; ear, 10. Skull: greatest length, 35.5; width across bullae, 22.1; spread of maxillary arches, 19.5; greatest length of nasals, 13.0; width of maxillary arch at middle, 5.0. ^ange. — Dipodomys merriami trinidadensis is found in El Vale de Li Trinidad and the more arid parts of El Valle de la San Rafael, at least to Sangre de Cristo along the western base of the Sierra Juarez. In pelage characters, specimens from San Matias Pass show definite intergradation with the desert race, D. m. arenivagus, though their size places them with D. m. trinidadensis. Specimens examined. — Baja California: Sangre de Cristo, 31 (6 from Huey Collection); EI Valle de la Trinidad (type locality), 29; summit of San Matias Pass, Diablito Spring, 9. Dipodomys merriami arenivagus Elliot San Felipe Kangaroo Rat Original description. — Dipodomys (merriami) arenivagus Elliot, Field Colum. Mus., Publ. 87, Zool. ser.. Vol, 3, p. 249, December, 1903. Type locality. — San Felipe, Baja California, Mexico. Distribution. — ^The known range of Dipodomys merriami arenivagus covers the arid area east of the Sierra San Pedro Martir and Sierra Juarez from San Felipe nordiward almost to the boundry, excepting the region about the Colorado River delta. Specimens are at hand from the northern locality of Gaskill's Tank, which is located on the eastern base of the Sierra Juarez within 25 miles of the international boundary. No specimens are available to determine 222 San Diego Society of Natural History the southern extent of the range of this race. It intergrades with D. m. simiolus near the international boundary and with D. m. trinidadensis on the desert slopes of San Matias Pass. Specimens examined.— Q^a. Cahfornia: San Fehpe (type locahty), 69 (11 from Huey Collection); 5 miles north of San Felipe, 43; 30 miles north of San Felipe, 1; 40 miles north of San Felipe, 2 (1 from Huey Collection); De Mara's Well, western side of Laguna Salada (Laguna Maquata on some maps), 12; Gaskill's Tank, east base Sierra Juarez, 17. Dipodomys merriami quintinensis subsp. nov. San Quintin Kangaroo Rat Type. — From 5 miles east of San Quintin, Baja California, Mexico; No. 4205, San Diego Society of Natural History; adult male; collected by Laurence M. Huey, April 9, 1923. Characters. — The pelage color of D. m. quintinensis, when viewed in series, is darker than in D. m. parvus, its nearest comparatum. This form furnishes a good demonstration of the need of mass series to arrive at color values in species that show considerable variation in this respect. Compared with D. m. parvus, quintinensis averages larger; its tail tuft is lighter in density of black, and it has a larger ear. It differs from trinidadensis in being darker and in having a slightly smaller skull. Compared with semipallidus , its intergrading relative and apparent ancestor to the southward, quintinensis is darker, both in pelage color and in tail tuft. In both body size and skull size it is smaller than semipaUidus. Its differentiation, like that of trinidadensis, seems to have involved a reduction in size. Measurements of type. — Total length, 240; tail, 148; hind foot, 39; ear, 10. Skull: greatest length, 36.4; width across bullae, 22.6; spread of maxillary arches, 20.1; greatest length of nasals, 13.0; width of maxillary arch at middle, 5.1. Range. — Found over the coastal plain above San Quintin Bay, and as far north as the vicinity of Santo Domingo, which lies some 25 miles north of San Quintin. Santo Domingo is the northernmost point where any member of the merriami group closely approaches the Pacific Ocean. All the more northerly races live inland in areas that do not approach the coast. Specimens taken below El Rosario River are all referable to the more southerly race. Remarks. — This form, like parvus and trinidadensis, provides another example of parallel development. All three branched separately from the desert stock line, with no known interconnecting links of relationship along the Pacific slope. In other words, they were separate invaders into an as yet completely unoccupied region of different climatic conditions. The efTect of the change in climate is recorded in their differentiations. Could meteorological data be taken in the three localities for an adequate period and the differences correlated with the three races, parallel correlations would presumably be indicated between the climatic gradients and character gradients. Trans Mexico HuEY— Kangaroo Rats of Baja California 223 Specimens examined.— Biji California: Santo Domingo (lat. 30° N), 11 (3 from Huey Collection); 1 mile south San Ramon, 4 (1 from Huey Collec- tion) ; north end San Quintin Plain, 7 (1 from Huey Collection) ; San Quintin, 8 (1 from Huey Collection); Santa Maria near San Quintin, 1; 5 miles east San Quintin (type locality), 1; mouth of Agua Chiquita Canyon, 2 (1 from Huey Collection); 7 miles southeast of San Quintin, 1; 3 miles east of El Rosario, 2; 8 miles east of El Rosario, 1. Dipodomys merriami semipallidus Huey Mid-peninsula Kangaroo Rat Origmal description.— Dipodomys merriami semipallidus Huey, San Diego Soc. Nat. Hist, Vol. 5, No. 5, pp. 65-66, July 6, 1927. Type locality. — 7 miles north of Santa Catarina, Baja California, Lat. 29° 45' N., Long. 115° 10' W. Now (1948) known as Rancht La Ramona. Distribution.— This race is found over a large triangular area having its northern boundary near El Rosario River Canyon on the west and the southern- most foothills of the Sierra San Pedro Martir on the east, thence tapering southward to an apex on the Pacific Coast somewhere west of Punta Prieta. Specimens from this coastal point are not available. However, specimens are at hand from Punta Prieta and vicinity and northward in an oblique line toward the Gulf. Over this long section all samples show characters blending toward the race D. m. platycephalus. The specimens from the section just mentioned suggest that a very broad band of intergradation exists between the two races. The specimens demonstrate, as would be expected, that the range of this darker race extends farther south on the Pacific side and veers from the more arid regions near the Gulf side, with no hard or fast lines of demarcation. Specitnens examined .—Ba']a. California: San Fernando, 1; 5 miles south- east of San Fernando, 1; 3 miles west of El Marmol, 3; Onyx (= El Marmol), 1; San Agustin, 16; 7 miles north of Santa Catarina (type locality), 14; 8 miles north of Santa Catarina, Rancho Ramona, 9; Santa Catarina Landing, 7; 10 miles north of Catavina, 8; Cataviiia, 6 (3 from Huey Collection); 13 miles northwest of Chapala, 2; 2 miles northwest of Chapala, 3; 25 miles north of Punta Prieta, 13 (not typical); Punta Prieta, 16 (6 from Huey Collection); San Andres, 8 (not typical; 2 from Biol. Surv. Collection). Dipodomys merriami platycephalus Merriam Calmalli Kangaroo Rat Original description.— Dipoc/om}/^ platycephalus Merriam, Proc. Biol. Soc. Washington, Vol. 20, p. 76, July 22, 1907. Type locality.— Calmalli, Baja California, Mexico. Distribution. — Dipodomys merriami platycephalus occupies a larger area 224 San Diego Society of Natural History of the peninsula than any of the other races. Specimens have been examined from a wide range, mainly on the Pacific slope. The northernmost locaUty is San Francisquito, N. Lat. 29° 48' (a Nelson-Goldman locahty, not to be confused with San Francisquito Bay, N. Lat. 28° 30'). The range extends from this northeastern locality diagonally southwestward to Santa Rosalia Bay on the Pacific side and over the great Viscaino Desert section. Specimens of D. m. pldtycephalus have not been examined from localities south of Campo Los Angeles on the eastern side of the desert nor south of the south entrance of Scammon's Lagoon, though there are no geographic barriers that would interrupt a population until the area near 26° N. lat. is reached. Specimens examined. — Baja California: San Francisquito, 3 (from Biol. Surv. Collection) ; mouth of Calamahue Canyon, 4 (from Biol. Surv. Collec- tion); Ubai (= Yubay), 30 miles south of Calamahue, 2 (from Biol. Surv. collection); San Borjas Miscsion, 2; Santa Rosalia Bay, 1; La Lomita Maria (Miller's Landing), 1; Santo Domingo Landing (N. Lat. 28° 15'), 11 (4 from Biol. Surv. Collection) ; mainland on south side of Scammon's Lagoon, 19; Mesquital, 3; 5 miles west of El Canon (10 miles west of Calmalli), 2 (1 from Huey Collection); Calmalli (type locality), 39 (16 from Huey Collection); 4 miles east El Arco, 1; Pozo Altamirano, 1 (from Biol. Surv. Collection); Campo Los Angeles, 3 (1 from Huey Collection); 1 mile east of Rancho Lagunitas, 1; Valle de Agua Amarga, 17 (not typical). Dipodomys tnerriatni annulus Subsp. nov. San Francisquito Kangaroo Rat Type. — From Barril, Gulf of California, Baja California, Mexico, lat. 28° 20' N., long. 112° 50' W.; No. 15522, San Diego Society of Natural History; adult female (2 foeti, 35 mm.); collected by Laurence M. Huey, March 22, 1947. Characters.- — Compared with Dipodomys merriami platycephalus, D. m. annulus is nearly equal in size, but is paler dorsally and has a darker crest on its bicolored tail. In color tone annulus resembles D. m. arenivagus but is several shades darker. Here the resemblance ends, for annulus is larger and has the black tail of the Cape district races. Compared with Dipodomys merriami brunensis, D. m. annidus is more pallid and has a bicolored tail. In fact, as the name implies, it is the connecting link in the two racial chains, and has dominant characters of both. The skull of annulus, compared with that of platycephalus. is similar in size, but has more inflated mastoid bullae and several other minor unmeasureable characters, such as very sharply angled zygomatic arches and truncated auditory bullae, which set annidus apart from either platycephalus or brunensis. Measurements of type. — Total length, 251; tail, 150; hind foot, 35; ear, 10. Skull: greatest length, 37.9; width across bullae, 23.8; spread of ma.xillary arches, 21.0; greatest length of nasals, 14.1; width of maxillary arches at middle, 5.5. HuEY — Kangaroo Rats of Baja California 225 Range. — In typical form this race lives on the Gulf side of the peninsula, over the coastal plain bordering San Francisquito and Santa Teresa bays and thence northward into a nontypical population at Los Flores, near Los Angeles Bay. Remarks. — Oddly enough, the type locality of the darker form D. m. platycephalus on the Pacific slope is less than 40 miles distant and almost directly west from Barril. The presence of this pallid race so near, reflects the decidedly arid character of this Gulf coast region, and is further evidence of the differentiating influence of the elements within short distances along the climatically variable peninsula. It is of further interest to note that certain plants typical of the Cape region of the peninsula find their northernmost limit in this Gulf coastal area. Their wedge-shaped ranges taper northward from Cape San Lucas to this apex. Specimens examined. — Baja California: Barril, at 28° 20' N .Lat., (type locality), 25; Santa Teresa Bay, 2 (collection of Donald R. Dickey); San Francisquito Bay, 6; 7 miles west of San Francisquito Bay, 19; Los Flores, near Los Angeles Bay, 37 (not typical). Dipodomys merriami brunensis Subsp. nov. San Bruno Kangaroo Rat Type. — From Llano de San Bruno, Baja California, Mexico: No. 6904, San Diego Society of Natural History; adult male, collected by Laurence M. Huey, March 24, 1928. Characters. — In size D. m. brunensis closely resembles D. m. platycephalus, its nearest relative, at least geographically. In dorsal coloration, however, it is darker, in general color tone tending toward Dipodomys merriami melanurus of the extreme Cape region. The tail tuft color is blacker than that of platycephalus. Both the mastoid and tympanic bullae are larger and more inflated. Compared with D. m. melanurus, brunensis has the same color tone, but is lighter and has a lighter colored tail. The difference in dorsal color is attributable to the climatic influences shown in the ranges of the two races. Both brunensis and melanurus occupy sections of the peninsula that are climatically different than the regions occupied by their relatives to the north and west. The mastoid bulla of the skull of D. m. brunensis is more inflated than that of D. m. melanurus, and a number of the specimens have smaller ears. This ear character in Dipodomys, especially when the whole genus is considered, is directly related to the inflation or deflation of the mastoid bullae. As a general rule, the species having large inflated mastoids have smaller ears than do those whose mastoid bullae are moderate in size. In Dipodomys merriami this character is too variable and has too little racial stability to warrant its use as a major differentiating character between races. Measurements of type. — Total length, 252; tail, 152; hind foot, 37; ear, 10. Skull: greatest length, 37.6; width across bullae, 23.8; spread of maxillary 226 San Diego Society of Natural History arches, 19.8; greatest length of nasals, 13.8; width of maxillary arch at middle, 4.8. Range. — ^Three areas, all on the Gulf slope of the peninsula and in the Concepcion Bay district, are known to be inhabited by this race. The hanging valley, El Valle de Yaqui, located about 8 or 10 miles northwest of Santa Rosalia, is the northernmost point of record, and incidentally, the highest in altitude. The Llano de San Bruno, south of Santa Rosalia, is the largest area inhabited by brunensis. The gradually rising region from the south end of Concepcion Bay to Canipole, is the third region populated by this form. Specimens examined. — Baja California: El Valle de Yaqui, northwest of Santa Rosalia, 14 (6 from Huey Collection) ; Llano de San Bruno (type locality), 9 (3 from Huey Collection); San Bruno, 4 (Biol. Surv. Collection); south end of Concepcion Bay, 15; Canipole, 9. Dipodomys merriami melanurus Merriam Cape San Lucas Kangaroo Rat Original description. — Dipodomys merriami melanurus Merriam, Proc. Calif. Acad. Sci., ser. 2, Vol. 3, p. 345, June 5, 1893. Type locality. — San Jose del Cabo, Baja California, Mexico. Distribution. — In its most typical form this race occupies a crescentic area, fringing the southern base of the higher foothills of the Sierra Victoria from Cape San Lucas to San Jose del Cabo, thence northward up the very wide valley at least to Agua Caliente. Between Agua Caliente and La Paz a mountainous area seems to interrupt the range, but isolated spots adjacent to the Gulf coast may be occupied. A good series from La Paz shows slight deviations, but within the degree allowable for this race. This is true also of a single available specimen from Tres Pachitas on the northwestern base of the Sierra Victoria. This specimen, though still considered to D. m. melanurus, has characters tending toward the population of D. m. llanoensis occupying the northern end of the vast Magdalena Plain. From Tres Pachitas southward to Cape San Lucas there seem to be no areas from which Dipodomys has been reported. This precipitous, hilly area, in which the mountains abut the sea, is similar to the region just noted on the opposite side of the middle section of the peninsula, which also appears to be devoid of kangaroo rats. Specimens examined. — Baja California: Cape San Lucas, 2 (Biol. Surv. Collection); San Jose del Cabo (type locality), 19 (9 from Biol. Surv. Collec- tion and 2 from Mus. Vert. Zool. Collection); Santa Anita, 5 (Biol. Surv. Collection); 7 miles south of Miraflores, 6; Agua Caliente, 10 (Mus. Vert. Zool. Collection); La Paz, 15; Tres Pachitas, 1 (Biol. Surv. Collection). Dipodomys merriami llanoensis subsp. nov. Magdalena Plain Kangaroo Rat Type. — From Buena Vista, Magdalena Plain, Baja California, Mexico, Lat. HuEY— Kangaroo Rats of Baja California 227 24° 50' N., Long. 111° 50' W.; No. 14676, San Diego Society of Natural History; adult male: collected by Laurence M. Huey, November 15, 1941. Characters. — This race differs from D. in. melanurus in having more pallid dorsal color and a very wide hip stripe, which on some specimens almost obliterates the hip spot. It differs from D. m. platycephalus in being slightly more pallid dorsally, resembling specimens from the western section of the Viscaino Deseert in this respect, and in having the tail and tail tuft decidedly black as in melanurus. D. m. llanoensis is intermediate between melanurus and platycephalus in that it exhibits characters found in both. However, it has more inflated mastoid bullae than either. In some specimens this develop- ment has pinched the interparietal into extremely narrow proportions. This one character, large inflatted mastoid bullae, set llanoensis apart sharply from the nearby island species, Dipodomys margaritae. Measurements of type. — Total length, 250; tail, 144; hind foot, 38; ear, 10. Skull: greatest length, 37.2; width across bullae, 24.0; spread of maxillary arches, 20.5; greatest length of nasals, 13.1; width of maxillary arch at middle, 5.7. Range. — This race is known only from San Jorge, on the northern end of Magdalena Plain, south to the area south of El Refugio, but without doubt it will be found farther southward, at least to the region of Arroyo Seco below Magdalena Bay. Specimens examined.— Bai^ California: 9 miles south of El Refugio, Magdalena Plain, 8; Buena Vista, Magdalena Plain (type locality), 5; Matan- cita, Magdalena Plain, 1 (Biol. Surv. Collection); Santo Domingo (Lat. 25° 30' N.), Magdalena Plain, 7; San Jorge, 6. Dipodomys insularis Merriam San Jose Island Kangaroo Rat Urtginal description.— Dipodomys insularis Merriam, Proc. Biol. Soc. Washington, Vol. 20, p. 77, July 22, 1907. Type locality. — San Jose Island, Gulf of California, Baja California, Mexico. Distribution. — Known only from San Jose Island. Remarks- — The writer was tempted to treat this species and the one following as subspecies amid the chain of mainland races. In certain basic characters, however, such as the shape of the bullae, which are smaller than those of the mainland races, and the correspondingly large ears, general pelage color, and robust appearance, it contrasts so strongly with the geographically nearest mainland relative brimensis, that intergradation cannot be assumed. The series of this species examined was ample to justify these statements and shows constancy in all its characters. Specimens examined.— Ba']2L California: 8 from "southwest side" and 4 from "southeast end" of San Jose Island, all from collection of Museum of Vertebrate Zoology, Berkeley. 228 San Diego Society of Natural History Dipodomys margaritae Merriam Margarita Island Kangaroo Rat Original description. — Dipodomys margaritae Merriam, Proc. Biol. Soc. Washington, Vol. 20, p. 76, July 22, 1907. Type locality. — Santa Margarita Island, Baja California, Mexico. Distributio7i. — Known only from this island. Remarks. — The range of variation shown by the limited available material of this species stands well beyond that shown by a good series from the nearby mainland. The bullae are much smaller and less inflated and the pelage is decidedly pale, as is the tail tuft. The ratio of tail to body length is smaller than that of the nearby mainland species. There seems to be no valid reason to reduce either this species or D. insiilaris to subspecific standing, for there is no evidence of intergradation. Specimens examined. — Baja California: 3 skins with skulls and 1 skull from Santa Margarita Island, all from Dickey Collection of the University of California, Los Angeles. Map 3 Distribution of The Merriam Kangaroo Rats of Baja California Name 1. Dipodomys merriami simiolus 2. Dipodomys merriami trinidadensis 3. Dipodomys merriami arenivagus 4. Dipodomys merriami quintinensis 5. Dipodomys merriami semipallidus 6. Dipodomys merriami platycephalus 7. Dipodomys merriami anmdus 8. Dipodomys merriami briinensis 9. Dipodomys merriami melanurus 10. Dipodomys merriami llanoensis 11. Dipodomys insularis 12. Dipodomys margaritae Type Locality Palm Springs, Riverside Co., Calif. Aguajito Spring, EI Valle de la Trinidad, Baja California. San Felipe, Baja California. 5 Miles east of San Quintin, Baja California. 7 Miles north of Santa Catarina (Lat. 29° 45' N.), Baja California. Calmalli, Baja California. Barril, Gulf of California (Lat. 28° 20' N.), Baja California. Llano de San Bruno, Baja California. San Jose del Cabo, Baja California. Buena Vista, Magdalena Plain, Baja California. San Jose Island, Gulf of California, Baja California. Santa Margarita Island, Baja California. Map 3 jc=:-c:S ^H' 1.^ r^ o o rs; VO o K « ^ •^ ir< « ,r< P u-s « ^ ^ '^ '—[ \r\ '^ S* p \r\ ^ .X " s ■^ 1 ■"J- 1 U-N 1 •^ 1 ■^ 1 ITN oi ITN 1 ir\ U-v 1 O rv VO \r\ rsl rsj \r\ rj^ ■^ Tt- ■^ rt ■^ Tt^ \r\ rt v\ O ^ ^ rs. ■^ ^ "^ d <^ d ^ d ^ (N 1 fN 1 ^ i r-s :q r-; ^ r^ rs; P Ov rA iy-\ r<\ fN rvi rr\ rA rsj •^ rA ■^ rt^ fA 1—1 rvl ■^ (N irs rsl rA rsl rA rsl 0\ r^ rsj rsl ^ rg <^ rsj ^ rsl ?i ^ ^ "h rsl oi ^ :^ 5^ P rA rsl 1 rA 1 rsl "^ rA rsl 1 ?Q q r-j rsj rsl rsi rA fS (N (N rsj rsj rsl rsj rsj rsj rj o ITS Ov t\ ITN rs| u-; rsl vo rs cN ;q ^ ^ u^ rA 00 vd O ^ o i:;^ ^ rA ^ ?^ O rs rA rA ^ 3; ^ !^ r^ rA -^ rA p '^ ^ ^ «t rA •^ ^ rA rA 1 rA ^ 3 ro ■^ ITN ITS rA Tt^ ir\ ir\ VO lA ro rA rA rA rA rA rA rA ^x ^^ ^ ^ ^ ^ ^ £ ^ r^ o\ VO z 00 0\ 1/^ oo ITS T^ ^ -^ ^ ^ ^ ■^ iy-\ ■^ 36.4 5-38 ^ 0\ ■ rA l^ 1 00 rA On rA M3 1 rA u-\ Tj- 00 j^ rA rA vi rA 00 rA ■^ o O^ 1 rA oi O 00 rA ^ ci rA rs '^ rA rA rri rA rA rA rA 00 s ^ u-\ 00 ir\ ir\ O 00 o 00 2; vo ^ 00 ^ VO U-N Tj- ;^ O w-\ ^^ i/^ vo ir\ VO VO VO ^ <> 1 On" I rA V VO ■— ' ■^ 1 00 T 00 ■— ' l/~N ""T Tt" d T ^ g 2; O ^ s ■^ i =^ § ITN 3 O 00 o 00 >/> O ^ r^ -^ 00 00 i/^ r^j 3 rsj rsj 1/^ ■^ 00 r. § ^ 7 ^ 3 ^ .«. a r 1 1 1 .2 c 1 'i-g Q >s rt s S in t ■ S s R .2 'c .5- 1 1 ill O rt rt c a If S c o -^ .2 'c d S a S w C^^O QScQ Q ^^ Q>CQ Q r^CQ Qu CQ Qcau^ QScQ Qc^oQ u il ill 113 III ll III i q si y i ^ ll w i - i 1 ■IK i QEdS - i 1 1 1 1 i III B i 1 l.s m Q>(S - 5 s s § § - S ^ 1 tl 11. 1^1 § 1 - i 1 |i It il 2 3 q q 5 I i i ii i i |1| s I til 2 2 ■^ i 1 ^ s - i i 1 1 s 113 i I ill ill- (5W£ if" QWDQ 232 San Diego Society of Natural History A gilts Group Range and Habitat Four races and one closely allied species of the agilis group are to be found in Baja California. Their ranges are confined to the north- western section of the territory and are limited to the Pacific slope, except in a very small area on the northern end of the Sierra Juarez, where the drainage is toward the desert side. No member of the agilis group, so far as is known, occurs on the arid desert slopes. Specimens of this group have been collected by the writer from the beach just above the tide line at Rosarito upward to the pine-studded meadows of Sierra San Pedro Martir, at an elevation of over 7500 feet. The species occupies each of the three life zones on the Pacific slope of northern Baja California — Lower and Upper Sonoran and Transition. Thus the gamut of range is in bold contrast to that occupied by D. tnerriami, which is almost wholly confined to the Lower Sonoran zone. It is perhaps more significant to state that agilis is essentially an inhabitant of the chaparral belt or brush-covered regions, for the range of the species and its races is limited mainly by this factor. In Baja California this type of ground cover, largely dominated by Adenostoma, reaches southward coastwise to a point some 50 miles below Ensenada, thence inland along the foothills of the higher mountains, where it gradually pinches out against the southern slopes of the Sierra San Pedro Martir. A great basin-like area south of the termination of the Californian chaparral belt is covered with more xerophilous growths. A species of wild rose {Rosa minuti folia) is the dominant plant of this belt, and in places forming large thickets almost excluding other species. This region, which might be called the San Quintin Subfaunal District, ends near the long and rather narrow El Rosario Valley, which extends eastward from El Rosario some 20 to 30 miles. South of this valley great thickets of mescal. Agave Shawi, form another type of ground cover, which is sparser than that of either of the two other types mentioned above. It is in this area that the agilis finds its southernmost and most arid habitat. Two of the races of agilis found in Baja California range also over large sections of southern California. One, Dipodomys agilis HuEY — Kangaroo Rats of Baja California 233 cabezonae, which occupies an inland range south from San Gorgonio Pass in Riverside County along the backbone of the Coast Range, barely gets below the international boundary, where its range ends against the northern slopes of the Sierra Juarez. The other, Dipodomyi agilis simulans, the most widespread race of the agilis group, occupies a great coastal foothill area of San Diego County, thence southward across the California brush cover into the San Quintin region, ending near El Rosario Valley. South of this valley Dipodomys agilis plectili< is found. The range of this difficult race is not well known and is in an inaccessible region where little collecting has been done. Coastwise, in the vicinity of Santa Catarina Landing, D. a. plectilis occupies common ground with Dipodomys paraliiis. This too is an area of little brush and is the southernmost known limit of the agilis group. In northern Baja California the range occupied by the agilis races is not as large, territorially, as is that occupied by the peninsularis group to the southward, but because of the nature of the vegetative cover it probably comprises the greater population. Climatic Influences The climatic influences which are reflected in the physical variations of the populations of five-toed kangaroo rats of the agilis group in northern Baja California are not as marked as are those which affect the populations of the two other major groups, merriami and penin- sularis, living on the peninsula to the southward. True, the regions which form the habitat of agilis range in elevation from sea level to over 7500 feet and are thus subjected to the more or less severe winter- summer extremes found in such location. However, this montane climatic variation does not influence the physical characters of the races of the agilis species as much as the varying oceanic temperatures influence the forms of either merriami or peninsularis. These latter groups are more diverse, despite the fact that the habitats they occupy are of much lower elevation and are more sparsely vegetated. Sufficient weather data are not available from Baja California to demonstrate these relations which, however, are amply demonstrated by an examination of the different characters of the several races among the three groups, and bv a plotting of the contrasting botanical associations of the different regions in which the various faces live. 234 San Diego Society of Natural History Material In the account of the agilis group the writer has examined 680 specimens from 42 locaUties; 73 specimens are from two locaUties in CaUfornia and 607 are from 40 locaUties in northwestern Baja CaUfornia. Dipodomys agilis cabezonae (Merriam) Cabezon Kangaroo Rat Original description. — Perodipus cabezonae Merriam, Proc. Biol. Soc. Washington, Vol. 17, p. 144, July 14, 1904. Type locality. — Cabezon, San Gorgonio Pass, Riverside County, Californis. Distribution. — The metropolis of this race is in the higher semi-arid valleys and over the chaparral-covered mesas along the eastern elevated desert slopes of the mountains of Riverside and San Diego counties in California. It reaches the southernmost extent of its range a short distance below the international boundary, against the northern and northwestern slopes of the Sierra Juarez in Baja California. This area comprises the southwestern part of Jacumba Valley, part of Nachoguero Valley, and surrounding brush-covered mesas. Specimens are available from three localities within this limited section. Specimens examined. — California: Cabezon, Riverside County (type locality), 51. Baja California: Just south of the international boundary near Jacumba, California, 12; 3 miles northwest of Neji, 2; Tres Pirios Mine, near Juarez, 3. Dipodomys agilis simulans (Merriam) DuLZURA Kangaroo Rat Original description. — Perodipus streatori simulans Merriam, Proc. Biol. Soc. Washington, Vol. 17, p. 144, July 14, 1904. Type locality. — Dulzura, San Diego County, California. Synonym. — Dipodomys agilis latimaxillaris Huey, Proc. Biol. Soc. Wash- ington, Vol. 38, p. 84, May 26, 1925 (type locality, 2 miles west of Santo Domingo Mission, Baja California, Mexico, Lat. 30° 45' N.). Distribution. — In Baja California this race ranges over a great coastal foothill area, from the western base of the Sierra Juarez to the Pacific Ocean and from the international boundary directly south of Campo, California, as far south as Rancho San Pablo, 10 miles south of Alamo, thence diagonally southwest to San Quintin and the Pacific Ocean. All is more or less heavily overgrown with dense chaparral. The region from the mountains to the sea and from the international boundary south to San Vincente is clothed with several species of brush commonly found over this race's range in California, namely red shank {Adenostoma sparsijolium),, chemise (Adenostoma fascicu- HuEY — Kangaroo Rats of Baja California 235 latum), and some manzanitas {Arctostaphylos sp.). Many valleys filled with heavy stands of live oak trees bisect this area. Some of these valleys are lightly farmed, hay or grain being the principal harvests. Such conditions are con- ducive to a large population of kangaroo rats. From San Vicente to the southernmost extent of the range, in the vicinity of San Quintin, a different type of brush cover is found. Here considerable cactus, a peculiar species of wild rose (Rosa minutjoldia), and other xerophilous shrubs form the chaparral. Remarks- — Dipodomys agilis simulans has the densest population of any form of the agilis group living in northern Baja California and the center of agilis population of Baja California is to be found within its bounds. Around the outer perimeter of this great range considerable variation is prevalent. The population on the eastern fringe from just below Campo southward, has a more pallid dorsal coloration, tending toward that of D. a. cabezonae. Farther south, along this brush-covered region through Valle de San Rafael to Rancho San Pablo, the cabezonae tendencies give way toward a blending with D. a. martirensis. The population of the Rancho San Pablo mesa region has a dark coloration, most nearly like that of topotypic simulans from Dulzura, but it shows an average tendency toward more inflated bullae. The writer was tempted to describe and name this population, but decided that its characters were inadequate to justify subspecific separation. A similar population from the Santo Domingo-San Quintin coastal district to the southwestward had been named by him in 1925 and now, with specimens at hand from a. wider area, the race is deemed untenable. This population about the San Quintin coastal area does average darker in dorsal coloration than does the general series of topotypes from Dulzura when viewed in mass, and there is a tendency toward an inflation of the bullae, but both characters are of two slight divergence to warrant subspecific designation. This "darker coloration tendency" is paralleled in the mernami group, but as agilis is in reality a coastal species, whose general habitat to the northward extends through zones of various coastal influences, the climatic effects of the slightly different San Quintin section were not as great upon it as they were on the desert living merriami. One specimen, a male (No. 11496, San Diego Society of Natural History), collected by the writer 3 miles east of Ojos Negros on March 12, 1936, is unique in having a white tail tip. In the large series examined, this is the only specimen that shows a markedly divergent character. No specimen yet examined lacks the fifth toe or hallax, which is occasionally revealed in some Californian species of Dipodomys. Specimens exdminec/.— California: Dulzura, San Diego County (type locality), 22. Baja California: Rosarita Beach, 1; north side of Descanso Bay, 2; Ensenada, 2; 14 miles south of Ensenada, 1; Santo Thomas, 6; Santo 236 San Diego Society of Natural History Map 4 Distribution of The Baja California Kangaroo Rats of the Dipodomys agilts Group. Type Locality 1. Dipodomys agiiis cabezonae. Cabezon, Riverside Co., California. 2. Dipodomys agiiis simidajis. Dulzura, San Diego Co., California 3. Dipodomys agiiis martirensis. La GruIIa, Sierra San Pedro Martir, Baja California. 4. Dipodomys agiiis plectilis. Canyon San Juan de Dios (Lat. 30° 7' N), Baja California. 5. Dipodomys peralius. Santa Catarina Landing (Lat. 29° 31' N), Baja California. Map 4 C/^2 ^H' a:ai s s OS i\ o OS o o - ° <^ Tt^ vq rr ^ .r; ^. ^ •^ ^ ^ S r^ u^ vq • rss ^ '^ -i. ■^ ■^ ^ 1 ^ oi ^ 1 ^ 1 ^ 1 ■^ rA vrs ^ 1 rr\ ■*^ ^ Tt^ Tf rt M- Tt- ■^ t ro °9 o 00 vo >rs rvj P <^ c OS S OS .—1 Os 0\ o^ 0\ OS ^ ^ '"' x ITS Tt^ Tt" ir\ — ( rsl ^ ^ ^ rr\ 00 (N ^ ^ i/^ (N T^ rsl I\ (N rsl rsj U~; ?Q" ^ (N 1 ^ r^ ^ ^ ^ ^ ?Q 'A rs| ^ rsj q ^ ^ rsl ^ oo ro rr-\ -^ T^ CO rA Tt" t- (N r-J r^ (^1 (N rs) rsj rsj rsi ■*- rg r\ o ITS QO r-s O O 00 00 ^ r^ ? >^ ^ ■<^ rA ^ o^ rA ^ ^ > ^ 00 00 1 ^ ^ ^ Js ^ ^ ^ ^ 00 rA oi O^ rA 1 ^ ^ o6 rA s f\ i< 00 00 I^ 00 K 00 K r^ rr\ ro rA ro rr, rA t^ t^ ^ ^ ^ § tK t^ 00 VO -H 1^ OJ T^ O c irs ITS ■^ m Tj- ITS ^ ITS ■^ ^ ^ p r-i - ^ ? 2 ^ ? ^ rsl ^ T^ p ^ 00 ? ^ ^ ^ i ^ i ^ i ^ 5^ ^ ? ? ol rA O 1 ^ OS rA o^ rA VO rs| rr\ o i/^ rA o Os irs OO § i 00 i^ 00 2 00 t O CO 1 00 1 00 S rA 1 u-s ^ B rv. g ^ ITS irs ci —1 ITS ^ § rss. § ^ OS u-s - i u-s !^ VO ITS O o o 00 rsl rsl o o 2^ o r-. T^ 00 TJ- OS rsl O ■^ OS o OS Th 00 o 292. 85-3 Csl 1 00 f5 o ^ 3 rA ""? rsl § 00 fs. rsl rsl 1 ^ 3 u-s rs fN r-4 c^ rs| (N rsj rsl 1 s 1. 1 III .-I lit E2 < w C S ll 2 < 1^1 5 c' agilis aliforn ) ■ 5 '2 11 ^ 2 ^^ So ri « — ' IH Dipodomys Cabezon, C, (Topotypes 1^' I 1^5 Qub -1 1 I. 5 1-3 Q to CQ Q to CQ O N QQ o S Q LO HuEY— Kangaroo Rats of Baja California 239 Domingo (Lat. 30° 45' N.), 28; 1 mile south of San Ramon, 2; north end of San Quintin Plain, 1; San Quintin, 11 (2 from Huey Collection); El Valle de las Palmas, 7; Ojos Negros, 17; 3 miles east of Ojos Negros, 26; 10 miles southeast of Alamo, Rancho San Pablo, 33 (7 from Huey Collection); San Jose, 5; (not typical); Las Cabras, 2 (not typical). Dipodomys agilis martirensis Huey Sierra Kangaroo Rat Original description.— Dipodomys agilis martirensis Huey, Trans. San Diego Soc. Nat. Hist., Vol. 5, pp. 7-S, February 20, 1927. Type locality.— La GruUa (east side of valley) , Sierra San Pedro Martir, Lower (= Baja) California, Mexico, altitude 7500 feet. Distribution. — Known from the higher parts of the Sierra Juarez, south- ward to the type locality in the Sierra San Pedro Martir and westward along the higher foothill slopes of the range to the vicinity of San Jose, where the specimens taken are nearer to D. a. simulans. The southern extent of the range of this race is, as yet, undetermined. Specimens examined from localities east of El Rosario, near San Fernando Mission and El Marmol, are all assignable to other races or species. Remarks. — In a series of 240 specimens from El Valle de la Trinidad, there is considerable variation, but no series could be picked out with characters that would justify separation from D. a. martirensis. This valley bisects the two mountain-top populations of this race. As a large portion of the valley is within the Lower Sonoran zone, D. a. martirensis inhabits three life zones. In this respect it is unique. This long valley with its abundant Dipodomys population is perhaps one of the few outstanding examples of coastal-desert intergrading areas left untouched, biologically, by man. He has tampered some with his poisons (Huey, Journ. Mam. 18: 74, 1937), but agriculture and dense human setde- ment have made no scars. Starting in the extreme western end of the valley amid coastal willow, flowing streams and Adenostroma associations, one can drive directly eastward over a gently rising valley floor, passing, within 20 miles, through associations of screwbean, mesquite, and creosote, thence over the gradual lift of San Matias Pass, through varied cacti and other xerophilous vegetation, downward through creosote and cacti into one of the most arid, torrid deserts of North America. This valley, still unspoiled, is one of the few western spots left in its pristine state, an ecologist's paradise. It was through such associations that this large series was collected. The greater portion could be placed within the known characters of D. a. martirensis, in spite of the fact that topotypical material was meager. It is to be regretted that the race martirensis was named from a locality in the higher parts of the Sierra San Pedro Martir. La Grulla Meadows is 240 San Diego Society of Natural History most inaccessable and lies on the very edge of the range of the race, in a region where it does not show its average characters. This factor has miHtated against a clear understanding of the characters of D. a. martirensis and has caused no end of difficulties in aHgning the various five-toed kangaroo rat populations sampled over this interesting terrain. Specimens examined. — Baja California: Laguna Hanson, 16; 1 mile east of Laguna Hanson, 1; El Rayo, Sierra Juarez, 19; Sangre de Cristo, 31; El Valle de la Trinidad, extreme western end, 14; (not typical) ; El Valle de la Trinidad, Aguaito Spring, 230; Summit of San Matias Pass, Diablito Spring, 35; La Grulla, Sierra San Pedro Martir, 5 (including type) . Dipodomys agilis plectilis' subsp. nov. El Rosario Kangaroo Rat Type. — From mouth of canyon San Juan de Dios (Lat. 30° 7' N.). Baja California, Mexico: No. 4721, San Diego Society of Natural History; adult female; collected by Laurence M. Huey. May 1. 1925. Characters. — This medium sized kangaroo rat has the lightest and most buffy color of any of the agilis forms. In dorsal coloration it approaches Dipodomys paralius, but it is larger and has a much larger ear. Its relationship is to the northward, with Dipodomys agilis simidans and D. a. martirensis, from which it differs in its smaller size and more buffy coloration. D. a. plectilis differs further from the two northern races in having a smaller skull with pro- portionately more inflated bullae. Measurements of type.— Total length, 275; tail, 165; hind foot, 41; ear 13. Skull: greatest length, 38.8; width across bullae, 24.5; spread of maxillary arches, 21.2; greatest length of nasals, 13.6; width of maxillary arch at middle, 4.6. Range. — Specimens are available from El Rosario Valley as far eastward as the type locality, the mouth of Canyon San Juan de Dios, a distance of about 10 miles airline, thence southward to Aguaito and El Marmol and thence coastwise to near Santa Catarina and Santa Catarina Landing. A broad area between these localities, consisting of foothill, mesa and coastal shelfland, is yet to be explored and will no doubt reveal considerable data for a better understanding of the incompletely known range of this perplexing race. Specimens examined. — 1 mile east of El Rosario, 5; 4 miles east of El Rosario, 6; 10 miles east of El Rosario, 4; mouth of canyon, San Juan de Dios (type locality), 20; Aguaito, 5; 5 miles southeast of San Fernando, 9; 3 miles west of El Marmol, 1; 8 miles north of Santa Catarina (Rancho La Ramona), 3; 7 miles north of Santa Catarina, 7; 4 miles north of Santa Catarina Landing, 5. 1. Latin — complicated, involved, intricate. HuEY — Kangaroo Rats of Baja California 241 Dipodomys paralius^ sp. nov. Santa Catarina Kangaroo Rat Type. — From Santa Catarina Landing (Lat. 29° 31' N.), Baja California, Mexico: No. 4250, San Diego Society of Natural History; adult female; collected by Laurence M. Huey, April 13, 1923. Characters. — This kangaroo rat is similar in color to Dipodomys peninsu- laris pedionomus , but it is smaller and has smaller ears. Cranially, it is widely divergent, with smaller, proportionately flatter, more inflated bullae and with slightly more angular and more widely spreading maxillary arches. This latter character is prominent and places this species very near to the broad-faced group of kangaroo rats. Compared with Dipodomys agilis plectilis, D. paralius is lighter in dorsal coloration and smaller in size, and further differs in the cranial characters mentioned above. The general oudine of the skull is more nearly that of an equilateral triangle than that of an acute triangle, such as charactetizes D. a. simulans, D. a. plectilis and other members of the agilis group. However, paralius is nearer to the D. a. simulans-plectilis chain than it is to the D. peninsularis group and it is best left under the agilis series. Measurements of type. — Total length, 250; tail, 150; hind foot, 38; ear, 11. Skull: greatest length, 38.8; width across bullae, 24.3; spread of maxillary arches, 20.5; greatest length of nasals, 13.2; width of maxillary arch at middle, 4.7. J(ange. — So far as known, from near Santa Catarina and around Santa Catarina Landing (type locality) . Both localities are in a great basin-like coastal section of Baja California. Specimens examined. — Baja California: 8 miles north of Santa Catarina (Rancho La Ramona), 12; 7 miles north of Santa Catarina, 2; 4 miles north of Santa Catarina Landing (type locality), 12 (including type). 1. Latin — that grows (lives) by the sea. 242 San Diego Society of Natural History Map 5 Distribution of The Baja California Kangaroo Rats of the Dipodomys peninsularis Group. Type Locality 1. Dipodomys peninsularis peninsularis. Santo Domingo Landing (Lat 28° 51' N), Baja California. 2. Dipodomys peninsularis pedionomus. 2 Miles north of Chapala Dry Lake (Lat. 29° 30' N), Baja California. 3. Dipodomys peninsularis eremoecus. 7 Miles west of San Francisquito Bay (Lat. 28° 30' N) , Baja California. 4. Dipodomys peninsidaris australis. Santo Domingo, Magdalena Plain (Lat. 25° 30' N), Baja California. ( CALIFORNIA J I 1 \ ] I iT A R IZONA OJ o \^ o o ■z- a ••■j'-<'*'3 S ON O R A { I 1 \ f DISTRIBUTIOMAL. MAP OF THE PENINSULARIS GROUP OF KANGAROO RATS OF- BAJA CALIFORNIA, IVIEXIOO TYPE LOCALITIES O SPECIMENS EXAMINED BY AUTHOR • BROKEN LINE SHOWS APPROXIMATE BOUNDARY OF RANGE. FOR NAMES INDICATED BY NUMBERS. SEE LIST ON OPPOSITE PAGE [ [ [ Map 5 244 San Diego Society of Natural History Peninsularis Group Range The peninsularis group is the third largest group of kangaroo rats in the peninsula. At present it is comprised of four races of one species. It ranges from just south of latitude 30° to the middle section of the Magdalena Plain, near latitude 25° and thus occupies a large portion of the length of the peninsula but reaches neither end. With two exceptions, this kangaroo rat is confined to the western or Pacific drainage slopes and the species range is entirely within the Lower Sonoran Zone covering some of the most sparsely brushed sections of the peninsula. In but very few places are the five-toed kangaroo rats of this group abundant. Through the Chapala region in the central northern section of its range, it apparently finds the center of abundance. Southward, toward the Magdalena Plain, it is very scattered in occurence and often rare. Through these southern portions of its range, areas that appear ideal for Dipodomys habitation are often devoid of these rats. No insular occurence of this species is known, nor has the writer found it close to the sea coast, as he has found Dipodomys ctgilis to the northward. So far as is known, D. peninsularis has not been found over the llanos approaching the Sierra Victoria though much territory on the north and northwestern sides seems to offer good ground for habitation. Climatic Influences Along the Pacific slope, this range, like that of the merriami group (Map 3, p. 227), spans the regions respectively influenced by cool and by warm coastal waters. The effect of this climatic influence is reflected largely upon the vegetative growth on and within which the mammal population live and which in turn is shown upon the mammalian life itself. By way of illustration, the color of the dorsal pelage and cranial development, shown in the two extreme races of this group, australis in the south and pedionomus in the north, are outstanding. Variation This group of kangaroo rats is outstanding in several characters. All four races are rather large and have thick, decidedly bicolored tails of medium length and extremely inflated mastoid bullae. The inflation HuEY— Kangaroo Rats of Baja California 245 of the bullae increases markedly from north to south, following the trend shown in other Dipodomys species, which range through brush cover grading from dense to sparse. It has been shown by Grinnell that the more open the brush habitat, the greater the bullae inflation. Three races of peninsularis are excellent examples of this trend. Some specimens from the southern end of the range of the species have such swollen, inflated mastoids that the interparietal and supraoccipital are almost crowded out, leaving only a narrow sulcus between the left and right buUi. That this character is not always due to age development is illustrated by a specimen just beyond the juvenal stage, in whose skull the interparietal is almost obliterated by the crowding of the inflated bullae. The fourth race D. p. eremoecus, provides an exception to the general trend, for it lives in very sparse vegetation, yet has the bullae less inflated than in either D. p. australis or D. p. peninsularis. The dorsal color pattern of the peninsularis group progressively lightens as the habitat extends over the peninsula, from the cooler to the warmer more arid sections. The tail tuft likewise darkens, but from north to south. As in merriami the darkest and most striking pattern is found at the southern end of the range, in D. p. australis. The lightest tail tuft and dorsal color pattern, however, is found in the race D. p. eremoecus, which lives on the Gulf side of the central peninsula, in the most arid desert climate within the range of this species. Certain basic characters found throughout the different populations of five-toed kangaroo rats living within the inland area of the peninsula from El Marmol southward, such as extremely inflated bullae, brightly colored and heavily boned tail and average dorsal color tones, set them apart from all forms of the Dipodomys agilis group living from Monte- rey in California to the coastal locality of Santa Catarina (Lat. 29° 40') in Baja California. The writer has been wholly unable to adjust these outstanding characters into the agilis subspecies group and believes a clearer delineation is provided by elevating Dipodomys agilis peninsu- laris from the agilis races into specific characterization, as Dipodomys peninsularis peninsularis. This form typifies a closely related group of four races, with characters that are compatable with the geographic and ecological conditions under which they live. Specimens of D. agilis plectilis and of D. peninsularis pedionomus have been taken in fairly close proximity, but not yet together. The inter-relationships of agilis and pedionomus may be determined when 246 San Diego Society of Natural History collections are made in a large area lying northward of the Llano Buenos Aires, on which San Agustin and El Marmol are situated. Friends who have flown over this region report having seen canyons in which large groves of fan palms were growing and state that this inland territory appeared extremely desert-like from the air. Material and Acknowledgements In this study of the peninsularis group of kangaroo rats, 268 speci- mens have been examined, from 34 localities in Baja California. Of this number, 242 are in the collection of the San Diego Society of Natural History, 14 are in the Huey Collection, and 10, including the type of Perodipus simuUns peninsularis, have been borrowed from the United States Biological Survey (now the Fish and Wildlife Service), through the generosity of the late Major E. A. Goldman, and 2 have been borrowed from the Dickey Collection, now at the University of California in Los Angeles, through the courtesy of the late A. J. Van Rossem. Dipodomys peninsularis peninsularis (Merriam) ViscAiNo Desert Kangaroo Rat Original description. — Perodipus simulans peninsularis Merriam, Proc. Biol. Soc. Washington, Vol. 20, p. 79, July 22, 1907. Type locality.— Santo Domingo (= Santo Domingo Landing), Lower (= Baja) California, Mexico, Lat. 28° 51' N. Nelson recorded (Mem, Nat. Acad. Sci., Vol. 16, p. 30, 1921) that he and Goldman were camping on the date when this type specimen was collected, at the sight of the old well near the edge of a mesa-like shelf, some three miles inland from the landing beach, elevation about 50 feet. Distribution. — The area inhabited by D. p. peninsularis is principally the Viscaino Desert of the central part of the peninsula, from below Punta Prieta (at about Lat. 28° 40') south to the vicinity of San Ignacio (Lat. 27° 20'), thence over the peninsula divide to El Valle de Yaqui, which lies between San Ignacio and Santa Rosalia on the Gulf slope. Specimens from coastal regions west of San Ignacio are not available and to the writer's knowledge, no Dipodomys peninsidaris has ever been collected coastwise between the type locality of D. p. peninsularis at Santo Domingo Landing, latitude 28° 15', and San Jorge, latitude 25° 40', the northernmost point of Dipodomys peninsularis australis. Remarks. —Specimens from Santa Gertrudis Mission are slightly paler than those from the type locality and have slightly larger, more inflated skulls. Sufficient material might reveal an isolated race. A notable example of mastoid inflation is found in specimen No. 7010, San Diego Society of Natural HuEY — Kangaroo Rats of Baja California 247 History, from Mesquital, April 13, 1928. These bullae are so expanded that the supraoccipital and interparietal are almost completely enveloped. The dorsal interspace between the left and right mastoids, a deep sulcus, measures less than 0.3 mm. Another specimen. No. 579, collection of L. M. Huey, from Campo Los Angeles, March 31, 1928, has the cranium heavily perforated, almost like a sieve, by parasite borers. Similar examples have been found by the writer at EI Valle de la Trinidad in northern Baja California and at Pellisier's Ranch in northern Owen's Valley, Mono County, California. Specimens examined. — Baja California: 11 miles south of Punta Prieta, 1; La Lomita Maria, 2 (1 from Huey Collection); Mesquital, 17 (3 from Huey Collection); Santo Domingo Landing, latitude 28° 15' (type locality), 8 (2 from Biological Survey, including type); 5 miles west of El Canon, 4 (1 from Huey Collection) ; Calmalli, 14 (5 from Huey Collection, 2 from Biological Survey Collection) ; 4 miles east of El Arco, 4; 5 miles east of El Arco, 4; 12 miles east of El Arco, Rancho Mira Flores, 8; Santa Gertrudis Mission, 3; Rancho Union, 15 miles east of Calmalli, 3; 1 mile east of Rancho Lagunitas (Lat. 28° 30'), 3; Poso Altimisano, 1 (from Biological Survey Collection); Campo Los Angeles, 9 (4 from Huey Collection); San Ignacio, 7 (1 from Huey Collection, 3 from Biological Survey Collection) ; 18 miles east of San Ignacio, 1; 10 miles west of Santa Rosalia, Valle de Yaqui, 3. Dipodomys peninsularis pedionomus' subsp. nov. Chapala Kangaroo Rat Type. — From 2 miles north of Chapala Dry Lake, on Llano de Santa Ana, lat. 29° 30' N., long. 114° 35' W., Baja California, Mexico; No. 8363. San Diego Society of Natural History; adult male; collected by Laurence M. Huey, October 17, 1930. Subspecific characters. — Compared with Dipodomys peninsularis peninsu- laris, D. p. pedionomus is darker on the sides, back and ears and has larger hip spots and correspondingly narrower hip stripes. The tail stripes are blacker and are about equal in color and size above and below. The white side striping is narrower than that found on the tail of D. p. peninsidaris. D. p. pedionimus has a narrower, more elongated skull than peninsularis with less inflated mastoid bullae and more truncated, less bulbous tympanic bullae. Compared with Dipodomys agilis martirensis and D. a. plectilis, D. p. pedionomus shows the characters of the nominate race, namely, brighter color, heavily boned tail, brighter, more buffy dorsal color tones of pelage and extremely inflated bullae. Material examined by the writer so far shows no overlap in these characters. Measurements of type. — Total length, 290; tail, 175; hindfoot, 42; ear, 13. Skull: Greatest length, 41.0; width across bullae, 25.0; spread of maxillary arches, 20.1; greatest length of nasals, 14.4; width of maxillary arches at middle, 5.0. Range. — This is the northernmost race of the peninsularis group. It is 1. Greek — dwelling on plains. 248 San Diego Society of Natural History found over the inland llanos from southeast of San Fernando Mission south to the valley region below Punta Prieta and eastward to Valle de Agua Amarga and San Borjas Mission. Specimens examined. ^^a]^ California: San Agustin, 22; 2 miles north Catarifia, 5; 7 miles south Catarina, 11; 13 miles northwest of Chapala, 15; 2 miles northwest of Chapala (type locality, including type) 48; 25 miles north Punta Prieta, 9; Punta Prieta, 14; El Valle de Agua Amarga, 1; San Andres, 1; San Borjas Mission, 6. Dipodomys peninsularis eremoecus' subsp. nov. Gulf-coast Kangaroo Rat Type. — From 7 miles west of San Francisquito Bay, lat. 28° 30' N., Gulf of California, Baja California, Mexico; No. 15619, San Diego Society of Natural History; adult male; collected by Laurence M. Huey, March 31, 1947. Subspecific characters.— Compared with Dipodomys peninsularis peninsu- laris, D. p. eremoecus is more pallid dorsally, with decidedly paler tail stripes and blacker terminal tuft. The under tail stripe is much lighter and narrower than the upper one and on some specimens is almost obsolescent. The hip stripe is very wide and the hip spot is proportionately reduced. The colored portion of the soles of the hind feet are much reduced in width and narrower than in any other member of the peninsularis group yet seen. The white spot above the eye is much larger than that of D. p. peninsidaris. In cranial characters D. p. eremoecus resembles D. p. peninsularis but has slighdy less inflated mastoid bullae with more elongated and less inflated tympanic bullae. The tooth row is shorter and molariform teeth are smaller. Measurements of type.— Total length, 288; tail, 170; hindfoot, 44; ear, 14. Skull: Greatest length, 39.2; width across bullae, 25.2; spread of maxillary arches, 21.7; greatest length of nasals, 13.1; width of maxillary arch at middle, 4.7. ^ange. — Known so far from the wide llano-like region that borders Santa Teresa and San Francisquito bays on the Gulf side of Baja California, near latitude 28°. Remarks.— This form is the lightest colored race of the peninsularis group and is the only member of the five-toed kangaroo rats whose population lives near the Gulf shores. The area in which this race is found is the northernmost growing range of a number of plants belonging to the flora of the Cape region. Some of the most noteworthy are as follows: Palo Blanco Lysiloma Candida; Fig, Ficus palmeri; Euphorbia xanti; Blue morning glory, Jacquemontia abutiloides and Desmanthus jruticosus. The presence of these plants belonging to the cape flora indicates a definite climatical influence northward along the Gulf coast strong enough to permit their invasion. That this same condition has changed the characters of Dipodomys peninsidaris is demonstrated in the pallid coloration of D. p. eremoecus. 1. Greek — desert inhabiting. HuEY— Kangaroo Rats of Baja California 249 Specimens examined. — Baja California, Mexico: Santa Teresa Bay, Gulf of California, 2 (from the Dickey Collection) ; 7 miles west of San Francisquito Bay, 21 (type locality, including the type). Dipodomys peninsularis australis subsp. nov. Magdalena Plain Kangaroo Rat Type. — From Santo Domingo, Magdalena Plain, Baja California, Mexico, lat. 25° 30' N.; No. 14734, San Diego Society of Natural History; adul: male; collected by Laurence M. Huey, November 21, 1941. Subspecific characters. — As compared with Dipodomys peninsularis peninsularis, its nearest relative to the northward, D. p. australis is slightly darker dorsally with darker ears and with the upper and under black tail stripes much blacker and well defined, extending into the tail tuft. The under tail stripe of D. p. australis in this character is far more constant than it is in the races to the northward. Soles of hind feet and ankles are black. This character too, like the undertail stripe, is decidedly more variable on specimens from the northern and central sections of the peninsula where the soles of the feet are often narrowly colored or of a faded grayish cast. Cranially Dipodomys peninsularis australis differs from D. p. pemnsidaris in having more massive, inflated mastoid bullae and more enlarged and swollen auditory bullae. The mastoid bullae protrude farther backward, leaving a deeper occipital sulcus between them, thus depreciating the size of both the supraoccipital and the interparietals. Measurements of type.— Total length, 298; tail, 184; hindfoot, 42; ear 12. Skull: Greatest length, 40.2; width across bullae, 26.6; spread of maxillary arches, 21.1; greatest length of nasals, 13.1; width of maxillary arch at middle, 4.9. Range. — So far as known this subspecies ranges over the Magdalena Plain from San Jorge in the northernmost section, south to Matancita, a ranch inland from the mouth of Magdalena Bay at about latitude 24° 40', the southernmost station yet recorded. Remarks. — Toward the Cape district the merriami and peninsularis group show a remarkably parallel development of the black or blackish coloring of the tail tufts. The black coloring is particularly intense in this newly named race, however most of the races of peninsidaris have dark or very dusky tail tufts but australis has the blackest tuft of all. The extreme inflation of the mastoid bullae of this southernmost race is likewise notable, for it is another example of mastoid enlargement in races or species of Dipodomys living in sparsely brushed or brush-less regions, as compared with those living in heavily brush-covered areas. The progressive inflation of the bullae in races of the five-toed species from the international boundary southward through ever decreasing chaparral cover is very notable. Specimens fXdm/nec/.— Baja California: San Jorge, 1; Santo Domingo, lat. 25° 30' N., 13 (including type); 7 miles north El Refugio, 1; 9 miles south El Refugio, 4; Matancito, 2 (from the Collection of the U.S. Biological Service, now Fish and Wildlife Service) . to s g u ^3 HO 05 CO CO !L! ^ w SQU ffl ,5_, O ^^ '^ tL. [i, qO Zc/5 < w §^ zo mO > < -a 6JS. o-J^ Xti a I Z^ ^ ^ o 00 ^ ? : ^ o 00 -r ? ^ 5 ^H c P— I ^-^ • w rt o 5_SH K-gl IS -5 ^ s "3 0 t !§- ^H re ^-^ ti V4 ^.2 5 S £ l^^ Ci, tr! "73 ?N ^". Q « (F •^ JiCQ ^ § >^ Q r.CQ «L, -n .2 cqI 18° 17° 16' 15° 14° 13 2" iT lo" ' 1 ' / ' ' 1 I I "'7^ — : 3 4" \ CALIFORNIA j 3/ \ J A R IZ ON A \ ' • y"\ ' 32° \ •' "Vn ^^^^-^ 31° ^.^ \^ ^^^_ ■ -- ' 7 Y > > Nv 30° \ \(\ '^ \ * 29° \ \ V7 ^ SON ORA / ^ \ V^ nl /' ~ \ ^ ) 1 "o^ /- 28° s ( o , y \ ^ \ o N, V \ "^ 1 o ) ^ \ i 27° ^ ^ ^ \o o L -^ V^ \ \ \ % r'v 26° \ o ^ V"^ \ Y " \^ 8 ^ jy 25° DISXRIBUTIOMAL MAP OF DIPODOMYS DESERTI ^ > ^v AND DIPODOMYS GRAVIPES \l ~ irsi NORXHERfSI BA_IA C Al_ IFORNI 1 A , MEX ICO \ 24° 1 DIPODOMYS DESERTI V v'^ 2 DIPODOMYS GRAVIPES \ - TYPE LOCALITIES 0 \. v 23° SPECIMENS EXAMINED BY AUTHOR • BROKEN LINE SHOWS APPROXIMATE BOUNDARY OF RANGE fill 1 J 1 ( Map 6 252 San Diego Society of Natural History Deserti Group This group is one of the most speciahzed groups of the whole genus Dipodomys, if not the most speciahzed. It is formed of one species divided into two races. Dipodomys deserti is one of the largest of the kangaroo rats and the common name "Big Desert Kangaroo Rat" is thoroughly descriptive. A small portion of the range of the nominate race Dipodomys deserti deserti reaches into the most arid section of northeastern Baja California and the species is known to occur as far south as San Felipe. Dipodomys deserti is truly a child of the desert. Its habitat niche, always confined to the Lower Sonoran Zone, is restricted to the most silty or sandy soils or aeolean dunes of the hottest deserts. Wherever such soil conditions are found over the deserts of eastern California,, southern Nevada, western and southwestern Arizona, northeastern Baja California or western Sonora, this species is usually present. At some localities, where sand dunes border the extreme northwestern and the northeastern shores of the Gulf of California, the species can be found almost to the water's edge. A section of the range occupied by deserti in Sonora, Mexico, approaches the arid Tropical Zone and within this area, color characters sufficient for racial determination have been developed. This pelage trend is toward darker dorsal coloration and is paralleled in the develop- ment shown by populations of D. merriami that approach the arid tropical zone of the Cape of Baja California. A further resemblance to other members of the genus is shown in the predilection of deserti to inhabit an area towards a zone below its main zonal niche rather than the zone above it. This is a four-toed species, with ponderously developed auditory bullae and correspondingly small ears. Its pelage is extremely pallid. These outstanding characters all reflect the open, almost grassless, sandy regions of its habitat. Dipodomys deserti lives in loose colonies and occupies large con- spicuous burrows, the openings of which are never plugged. These open burrows invite many coinhabiting enemies which no doubt take great toll. Members of this species are great trail makers. Their padded tracks can be followed easily in the soft sand even by a novice, and HuEY— Kangaroo Rats of Baja California 253 prove to be a positive index to the inhabited burrows in which, according to the writer's experience, but a single adult Uves. This habit also offers some light on the amount of inter-burrow social calling that is made nightly by the various colony members. So persistently do these big kangaroo rats follow their trails, which, as before stated, are most often made on soft wind blown sand, that an observer is forced to believe that they are possessed of rather keen eyesight, for surely no other faculty would guide them so unerringly along these trails on dark moonless nights. At this point, it might be well mentioned that moonless nights are the nights of greatest activity amongst all forms of Dipodomys. This statement has been proven by the writer on many occasions when trap lines set during the full moon would yield poor catches and lines set over the same ground on dark nights would have the opposite results. Material In the account of this group, 14 specimens of Dipodomys deserti deserti from 4 localities in Baja California have been examined (Map 6) . Dipodomys deserti deserti Stephens Big Desert Kangaroo Rat Original descriptwn.^Dipodomys deserti Stephens, Am. Nat., Vol. 21, No. 1, p. 42, January 1887. Type locality.— Mo\ave River, California (= Mojave River, 3 or, 4 miles from and opposite Hesperia, San Bernardino County, California.) Distribution.— In Baja California, where proper arenaceous ground is to be found, from the vicinity of Pilot Knob westward along the international boundary to the base of the Coast Range Mountains, thence southward along the eastward edge of the Pattie Basin (See Map, Plate 2, in The Salton Sea, MacDougal, Carnegie Inst. Wash. Pub. 193, June 1914), skirting the desert base of the Sierra San Pedro Martir at least to the latitude of San Felipe. The great salt flats and the vegetated regions of the Colorado River delta, determine its eastern limit of range in the peninsula. In all probability Dipodomys deserti will be found farther south in the narrowing, wedge-shaped, arid region bordering the Gulf of California below San Felipe. This section of the peninsula is exceptionally difficult of access and to the writer's knowledge, it remains, as yet, unexplored by mammalogists. Specimens examined.— Baja. California: 40 miles north of San Felipe, 1; 30 miles north of San Felipe, 1; San Felipe, 7; De Mara's Well on western side (old beach line) of Laguna Salada, 35 miles below international boundry, 5. 254 San Diego Society of Natural History Heerntanni Group This group of broad-faced kangaroo rats is represented in Baja California by a single endemic species, Dipodomys gravipes, which has an extremely liinited range. The heermanni group as defined by Grinnell, is composed of 8 species, one of which comprises nine races. All but two species of this lot are found north of latitude 34° 30' in California, north of the mountain chain which runs roughly in a west to east direction from Point Conception in Santa Barbara County to the San Bernardino Mountains in San Bernardino County. The broken range and scattered and varied geographical localities in which a number of species of the heermanni group live, when com- pared with the almost continuously occupied range of the more homo- genous agilis group, offers some evidence of fact that the heermanni group is older, that it had occupied its range long before the agilis group began its expansion. At present, mountain ranges and rivers offer unsumountable barriers that bisect the ranges of members of the heermanni group. Thus the region must have been occupied before these geological barriers came into existence, or at least, before the avenues of expansion were closed as they are today. One basic character, the broad face, is common to all members of the group. This, in itself, indicates that, at one time in their history, all forms of this group were closely interrelated, possibly as races. Yet today there are to be found species within this group having such diver- gently developed characters as: light colored pelage and dark colored pelage, white-tipped and non-white-tipped tail, and even the threatened loss of a toe, with one member of the group. These differences it would seem are attributable to isolation. The two species of the heermanni group that live south of the Sierra Madre barrier in California are Dipodomys stephensi, which is found in San Jacinto Valley in Riverside County, and Dipodomys gravipes, which occurs 250 niiles to the southward in the vicinity of San Quintin, Baja California. Both occupy extremely limited ranges and both are found in areas also inhabited by races of the agilis group. In fact, the agilis group, of which several races are involved, blanket the entire coastal area from the Sierra Madre in California south to the HuEY — Kangaroo Rats of Baja California 255 vicinity of El Rosario in Baja California. These two species of the heermanni group living in their midst are like small islands in a great sea. The superficial resemblance between Dipodomys stephensi and D. gravipes in some characters and the present distance between their ranges, would indicate that they had occupied the intervening area at some remote time, much as do the forms of agilis today. Thus it seems that the more virile agilis group had, either by more rapid reproduction or by food competition, reduced the range of this declining heermanni group until today it occupies small and remote areas, in which, through isolation, characters sufficient for species determination have been developed. In their invasion of Baja CaUfornia the representatives of the deserti and the heermanni groups have reached approximately the same latitude and both have apparently been checked in their advancement by some biological or geological barrier. The fact that the range of the heermanni representative is now decidedly restricted in bounds whereas that of deserti is still connected to the main body of its population would argue for the heermanni being the older group of the two. How ever, there are as yet no dependable bases from which the age of either could be computed. Material In this group 30 specimens of Dipodomys gravipes, from four locations in Baja California, have been used (Map 6). Dipodomys gravipes Huey San Quintin Broad-faced Kangaroo Rat Original description. — Dipodomys gravipes Huey, Proc. Biol. Soc. Wash- ington, Vol. 38, p. 83, May 26, 1925. Type locality.— 2 miles west of Santo Domingo Mission, Baja California, Mexico (lat. 30° 45' N.). Distribution.— -So far as known from the vicinity of the type locality south over the llano and foothills east of San Quintin, a distance of about 20 miles, and north over the Llano de Camalu to the San Telmo River. Specimens examinee/.— Baja California: 3 miles south of San Telmo, 2; 2 miles west of Santo Domingo Mission (type locality, 17 (6 from Huey Collection); mouth of Agua Chiquita Canyon, 11 (4 from Huey Collection); Santa Maria near San Quintin, 1; San Quintin, 1; mile south of San Ramon, 1. (For map see page 250.) ^|o::^ TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 11, pp. 257-328, plates 1-13 THE VOGDES COLLECTION OF TRILOBITES BY B. F. Howell Professor of Geology and Paleontology, Princeton University \m. coMP. zoaL MAY 14 195 SAN DIEGO, CALIFORNIA Printed for the Society April 30, 1951 COMMITTEE ON PUBLICATION Joshua L. Baily Carl L. Hubbs Laurence M. Klauber HAY 14 1951 m'!h^^ UN; TABLE OF CONTENTS . North American Families Page Order Hypoparia 26^. Family Trinucleidae ■^"■^ Order Opisthoparia 262 Family Olenidae 26^ Family Dikelocephalidae 263 Family Lloydiidae 263 Family Bathyuridae 264 Family Asaphidae 26^ Family Illaenidae 267 Family Proetidae 270 Family Odontopleuridae 277 Order Proparia 278 Family Encrinuridae 278 Family Calymenidae 280 Family Homalonotidae 284 Family Cheiruridae 284 Family Dalmanitidae 285 Family Phacopidae 288 European Families Order Hypoparia -. 290 Family Harpedidae 290 Family Trinucleidae 290 Family Remopleuridae 291 Order Opisthoparia 291 Family Olenidae 291 Family Asaphidae 291 Family Illaenidae 292 Family Scutellidae 293 Family Proetidae 293 Family Odontopleuridae 296 Order Proparia 296 Family Encrinuridae 296 Family Calymenidae 297 Family Cheiruridae 298 Family Dalmanitidae 298 Family Phacopidae 299 TABLE OF CONTENTS (Continued) Australian Families Page Order Hypoparia 301 Order Opisthoparia 301 Family Scutellidae 301 Family Proetidae 302 Family Odontopleuridae 303 Order Proparia 305 Family Encrinuridae 305 Family Cheiruridae 306 Family Dalmanitidae 307 Family Phacopidae 307 Explanation of Plates 308 THE VOGDES COLLECTION OF TRILOBITES by B. F, Howell Professor of Geology and Paleontology, Princeton University From 1904 to 1920 General A. W. Vogdes, the eminent biblio grapher on North American trilobites, served as President of the San Diego Society of Natural History. At his death in 1923 his large collection of trilobites was, in accordance with the terms of his will, presented to the Society. Some of the trilobites in this collection are type specimens and others are of sufficient import- ance to merit discussion and illustration. In order that interested students may know the location of these types and the other tri- lobites in the collection, an annotated list of them is here pre- sented, together with detailed descriptions and illustrations of the more important specimens. The writer is indebted to Dr. Joshua L. Baily, Jr., Research As- sociate of the San Diego Society of Natural History, to the late Mr. Clinton G. Abbott, Director of the Society, to Professor Arthur F. Fischer, the present Director, and to Professor Baylor Brooks, of San Diego State College (where this collection was deposited on loan when the writer examined it) for the opportunity to study these fossils. To these gentlemen he expresses his gratitude for the many courtesies which they extended to him. The writer is grateful also to Miss Alice W. Bar- low, former Librarian of the San Diego Scientific Library, for her co- operation. He made much use of the Vogdes Library of Geology and Paleontology, which was in her care. The Vogdes collection of trilobites consists of a miscellaneous as- sortment of North American Cambrian, Ordovician, Silurian, Devonian, and Caboniferous specimens (including the types of some of the species described by General Vogdes and of one species described by A. F. Foerste), a smaller number of specimens from the Cambrian, Ordovician, Silurian, and Devonian rocks of England, Scandinavia, Germany, and Bohemia, and some forty specimens from Silurian beds in New South Wales, identified and presented by the Australian paleontologist, John Mitchell. It includes representatives of 120 species. 262 San Diego Society of Natural History The North American specimens will be listed and discussed first. Two of them are described as new species. The Vogdes Collection catalog number of each series is given. NORTH AMERICAN SPECIES Order Hypoparia Family Trinucleidae Cryptolithus tessellatus Green Cryptolithus tessellatus Green, Mon. trilobites N. Amer. 1832, p. 73, cast 38, pi, 1, f.g. 4. Cryptolithus tessellatus Green, Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, pp. 295-296 (gives full bibliography to 1910) . Cryptolithus tessellasus Green, Raymond. Ottawa Naturalist, vol. 26, 1913, p. 4. Cryptolithus tessellatus Green, Shimer and Shrock, Index fossils N. Amer,. 1944, p. 643, pi. 267, figs. 2-5. Cryptolithus tessellatus Green, Wilson, Canadian Geo!. Surv. Bull. 9, 1947, p. 13, pi. 1, fig. 3. Four cephala, one a young shield, are labelled "Lower Silurian, no local- ity". As they are preserved in pieces of black limestone that also contain the brachiopod, Zygospira recurvirostris (Hall), they are probably from the middle Ordovician Trenton Formation of New York. No. 146. Several dozen other cephala, associated with a number of different species of bryozoans and an unidentified Zygospira on the upper surface of a block of grey limestone that is labeled as having come from the "Cincinnati Group" are probably from the Upper Ordovician of Ohio or an adjacent state. No. 141 Order Opisthoparia Family Olenidae Triarthrus becki Green Triarthrus becki Green, Mon. trilobites N. Amer. 1832, p 87, cast 34, pi. 1, fig. 6. Triarthrus bech Green, Bassler, U. S. Nat. Mus., Bull. 92, vol. 2, 1915, pp. 1286-1287 (gives full bibliography to 1910). Triarthrust becki Green, Shimer and Shrock, Index fossils N. Amer. 1944, p. 654, pi. 252, figs. 17, 18, pi. 276, fig. 22. Five cranidia (one of them very large) and an almost entire test, in black Upper Ordovician "Utica Slate" from New York, No. 151. A small cranidium, and a hypostome and associated thorax and pygidium, probably from the Utica Fonnation of New York. No. 152. Howell— VoGDES Collection of Trilobites 263 Several cranidia and entire tests, associated with tests of Ogygites cana- densis (Chapman), in gray weathering black "Utica Slate" (Collingwood For- mation) from Georgian Bay, Ontario. Nos. 139 and 140. Several small cranidia and pygidia and a small entire test, in Upper Ordov- ician gray shale form the vicinity of Cincinnati, Ohio. Nos. 147 and 152. Family Dikelocephalidae Subfamily Richardsonellinae Richardsonella unisulcata Rasetti Plate 1, f^g. 1. Richardsonella unmdcata Rasetti, Jour. Paleont., vol. 18, 1944, p. 256, pi. 39, figs. 54-56. A cranidium in a gray Upper Cambrian limestone from Point Levis, Quebec, appears to be referable to this species. It is one of five specimens from this locality, all in the same kind of limestone, that were associated in the Vogdes Collection. Each of the five is in a separate piece of rock; but all probably came from a single Upper Cambrian pebble in the Ordovician Levis Conglomerate that crops out at this locality. The two other species of trilobites found in the other four pieces of limestone are described below. All five speci mens are numbered 104 in the Vodges Collection. They all appear to be mem- bers of the late Late Cambrian, Trempealeauian, Hungaia tnagnijica Fauna. The specimen of Richardsonella unisulcata is numbered 104A. Loganellus macropleurus Rasetti Plate 1, f^g. 2. Loganellus macropleurus Rasetti, Jour. Paleont., vol. 18, 1944, p. 247, pi. 38, figs. 15-17. A cranidium, in gray limestone at Point Levis, Quebec, and presumably from the same Upper Cambrian pebble in the Ordovician Levis Conglomerate that yielded the cranidium of Richardsoiiella unisulcata that is described above, is believed to be referable to this species. This specimen is numbered 104B. Family Lloydiidae Lloydia parva, new species Plate 1, f^gs. 3-4. Three pygidia in gray limestone from Point Levis, Quebec, which probably came from the same Upper Cambrian pebble in the Ordovician Levis Con- glomerate as did the cranidia of the two species discussed above, appear not to be referable to any known species and are therefore here described as new. They are small, averaging 8 mm. long and 9 mm. wide. The entire pygidium is of moderate convexity, with the axis rising well above the pleural lobes and extending a little beyond the pleural lobes at the rear. There are four well 264 San Diego Society of Natural History defined segments in the anterior half of the axis and indistinct traces of two or three more in the rear half, the posterior part of which shows no segmenta- tion. There is a convex border, about 1 mm. wide, which is separated from the pleural lobes by a marginal furrow. The surface of the entire pygidium is smooth. Location of Types.-The pygidium chosen as the holotype is no. 104C. The other two pygidia, numbered 104D and 104E, are paratypes. Discussion-BecAUse of their association with the cranidia of Richardson- ella unisiilcata and Loganellus macro pleur us, species that have been described by Professor Rasetti as members of the Upper Cambrian, Trempealeauian, Hungaia magnifica fauna of the Levis region, it is believed that Lloydia parva is also a member of that fauna. It is possible that the three pygidia that are here referred to this new species, Lloydia parva, are in reality pygidia of one of the other species described by Professor Rasetti from Upper Cambrian pebbles in the Levis conglomerate in volume 18 of the Journal of Paleontology, for many of Professor Rasetti's species are known only from their cranidia. However, none of Professor Rasetti's cranidia appear to be referable to the genus Lloydia or to any other genus that has a cranidium similar enough to the cranidium of Lloydia so that it is likely to have had a pygidium like the pygidia here described. Lloydia is more commonly found in Lower Ordovician than in Upper Cambrian rocks; and L. parva is perhaps one of the oldest species of the genus yet discovered. Our pygidia are smaller than most of the pygidia of the Lower Ordovician species of Lloydia whose tail shields are known. It is just such a pygidium as one would expect to find in a very early form of the genus or in a genus ancestral to Lloydia. It is of about the same size as the pygidium of Lloydia seelyi (Walcott), of the Upper Cambrian Potsdam Sandstone of New York; but it has a smaller and less completely segmented axis and a wider border than the New York species. It also has a narrower axis and a wider border than L. brevis Raymond, of the Highgate Formation of Vermont. Family Bathyuridae Bathyurus extans (Hall) Plate 1, figs. 5-7. Asaphus? extans Hall, Pal., New York, vol. 1, 1847, p. 228, pi. 60, figs. 2. 2a-c. Bathyurus extans (Hall), Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, p 105 (gives full bibliography to 1913). Bathyurus extans (Hall), Shimer and Shrock, Index fossils N. Amer., 1944, p. 639, pi. 267, fig. 37. Bathyurus extans (Hall), Wilson, Canadian Geol. Surv. Bull. 9, 1947, pp. 16, 17, pi. 2, figs. 5, 6. Howell — Vogdes Collection of Trilobites 265 An associated thorax and pygidium and a free cheek and cranidium that may have belonged to the same individual, in black limestone of the Middle Ordovician Black River Formation from Great Bend, Jefferson County, New York. No. 122. Family Asaphidae Ogygites canadensis (Chapman) Plate 1, fig. 8; plate 2, figs. 1-2. Asaphus canadensis Chapman, Canadian Journ., vol. 1, 1856, p. 482. Ogygites canadensis (Chapman), Bassler, U. S. Nat. Mus. Bull. 92, vol. 2, 1915, p. 869 (gives full bibliography to 1913). Nine large and small entire tests, a cephalon, and a number of cranidia and pygidia, all in black shale of the Ordovician Collingwood Formation of Georgian Bay, Ontario. Two of the small tests and the cephalon are figured here. Nos. 121, 139, 140, 172, 176, 182, 184, 185, 190-192, 198-200. One slab of shale from the Collingwood Formation of Georgian Bay in the Vogdes Collection (no. 176) that contains pygidia and other fragments of Ogygites canadensis holds also a fragment of a thoracic segment of a huge trilobite, presumably an Isotelus and probably Isotelus gigas DeKay. Further mention of this is made in the discussion of /. gigas below. Isotelus gigas DeKay Plate 2, fig. 3; plate 3, fig. 1, plate 4, fig. 1-2. Isotelus gigas DeKay, Ann. Lyceum Nat. Hist. New York, vol. 1, 1824, p. 176, pi. 12, fig. 1, pi. 13, fig. 1. Isotelus gigas DeKay, Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, pp. 676-677 (gives full bibliography to 1914). Isotelus gigas DeKay, Shimer and Shrock, Index fossils N. Amer. 1944, p. 647 pi. 268, figs. 2-3. Isotelus gigas DeKay, Wilson, Canadian Geol. Surv. Bull. 9, 1947, p. 24, pi. 3, figs. 2a-c, 3. Two entire tests and one pygidium in black limestone of the Middle Ordovician Trenton Formation. One of the tests, the small one that is figured here, is from Herkimer County, New York (no. 183). The other is from Crab Island, Lake Champlain (no. 180) . The pygidium has no label, but is probably from somewhere in New York (no. 195). Two entire tests, two cranidia, two free cheeks, a hypostome, and several fragments from Ordovician limestones in Kentucky and possibly adjacent states. The largest test (no. 186) is from Lexington, Kentucky, and a note on its label states that it shows Panderian organs. A somewhat smaller one (no. 189) , which is here figured, is from Kentucky also (probably from Lexington) . The hypostome (no. 188) is labeled as having come from Lexington, and all the 266 San Diego Society of Natural History other specimens (nos. 179, 196 (a doublure), 197, and 200) except two are probably also from that city. Of these two, one, an enrolled test of a small individual (no. 178), is labeled as having come from Cincinnati, Ohio, and another, also a pygidium (no. 181), has no label, so that its locality is unknown. This last pygidium is accompanied by a sheet of paper on which are printed the figure and description of a cross section of the shield which are reproduced on plate 4. Three cross section slices of this shield are in the Vogdes Collection (no. 181). The writer has tried without success to discover whether the above printed figure and description were ever published. The box containing the pygidium and the sheet describing it were probably mailed to Professor James D. Dana some time during the eighteen-seventies, for it has, pasted on one side. Professor Dana's address and on another side postage stamps that were issued during that period. But a careful search of the bibliographies and the literature of that time has yielded no evidence of the publication of such a description or figure. The rock in which the specimen is preserved resembles that in which the specimens from Kentucky and the Cincinnati region were found, and the specimen probably came from that general area. The markings which are in- terpreted as traces of the internal anatomy of the trilobite may not actually be such. All the above specimens are referred to Isotelus gigas because none of their free cheeks bear spines, the feature which is said to be characteristic of the related species, Isotelus maximus Locke. As noted above, in the discussion of Ogygites canadensis, there is, in a slab of black shale from the Ordovician Collingwood Formation of Georgian Bay, Ontario, in the Vogdes Collection a single fragment of a thoracic segment of a huge species of trilobite that is probably Isotelus gigas. The trilobite of whose test this segment was a part must have been two feet or more in length. Isotelus maximus Locke Isotelus maximus Locke, 2nd Ann. Rept. Geol. Surv. Ohio, 1838, p. 246, figs. 8, 9. Isotelus maximus Locke, Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, pp. 678-679 (gives full bibliography to 1914). Isotelus maximus Locke, Wilson, Canadian Geol. Surv. Bull. 9, 1947, p. 26, pi. 3, fig. 1. Two fragments of hypostomes (no. 194) are preserved in a piece of gray limestone that also holds shells of the Lower Silurian (Richmondian) brachio- pod, Catazyga headi schuchertana (Ulrich) and the free cheek of an unidenti- fied species of trilobite. As Catazyga headi schuchertana is a characteristic species of the Richmondian Waynesville Formation, the two hypostomes are probably from that formation. As Isotelus gigas is said not to range up into Richmondian Howell — Vogdes Collection of Trilobites 267 beds, while 7. maximus has been recorded from such beds, there hypostomes are probably referable to /. maximus. Vogdesia vigilans (Meek and Worthen) Plate 2, f^g. 4. Asaphus (Isotelus) vigilans Meek and Worthen, Proc. Acad. Nat. Sci. Phila., 1870, p. 53. Nileus vigilans (Meek and Worthen), Bassler, U. S. Nat. Mus. Bull. 92, vol. 2, 1915, p. 857 (gives full bibliography to 1913). Vogdesia vigilans (Meek and Worthen), Shimer and Shrock, Index fossils N. Amer., 1944, p. 654, pi. 268, figs. 20, 21. Four enrolled tests, one of them a young one (figured here), from the Lower Silurian (Richmondian) Lower Maquoketa Formation of Fayette County, Iowa (no. 120). The eyes of the young test are less prominent than those of the two largest tests. The eyes of the test that is intermediate in size between the two largest tests and the small test are intermediate in their promi- nence also. Family lUaenidae Bumastus insignis (Hall) lllaenus insignis Hall, adv. sheets 18th Rept. New York State Cab. Nat. Hist., 1865, p. 27, figs. 5-6. Bumastus insignis (Hall), Bassler, N. S. Nat. Mus. Bull. 92, vol. 1, 1915, p. 141 (gives full bibliography to 1910). Bumastus insignis (Hall), Shimer and Shrock, Index fossils N. Amer., 1944, p. 639, pi. 269, figs. 10-12. Two cranidia in gray dolomite of the Middle Silurian (Niagaran Racine Formation, one from Bridgeport, Illinois (no. 105), and one from Wauwatosa, Wisconsin (no. 96), have the elongate, relatively narrow, form and long dorsal furrow that are characteristic of this species. Two pygidia from the same formation at Wauwatosa (no. 97) also appear to be referable to the species. Bumastus chicagoensis (Weller) Plate 3, fig. 3. lllaenus chicagoensis Weller, Chicago Acad. Sci. Bull. 4, pt. 2, 1907, pp. 220-222, pi. 16, figs. 10-12. A single pygidium (no. 105) in gray dolomite of the Middle Silurian (Niagaran) Racine Formation at Bridgeport, Illinois, agrees exactly with Weller's description of the tail shield of this species. As Weller did not figure the pygidium of B. chicagoensis and it has apparently not been illus- trated by any other author, a figure of our specimen is presented here, although the shield is not complete. 268 San Diego Society of Natural History Bumastus armatus (Hall) Plate 3, fig. 4-6. Illaenus armatus Hall, adv. sheets, 18th Rept. New York State Cab. Nat. Hist., 1865, p. 26, figs. 3-4. Bumastus armatus (Hall), Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, pp. 139-140 (gives full bibliography to 1910). Two cephala (no. 102) and a pygidium (no. 103) from gray dolomite of the Middle Silurian (Niagaran) Racine Formation at Bridgeport, Illinois, are referable to this species. The pygidium is evenly convex throughout and has a median ridge that extends one-third of the way forward from the rear edge of the shield. This pygidium and the two cranidia are figured here. The dorsal furrow is more deeply impressed toward the front on the smaller cephalon (which may be an immature shield) than it is on the larger cephalon. It is just possible that the pygidium here referred to B. armatus may be a tail shield of the species which Weller (Chicago Acad. Sci. Bull. 4, pt. 2, 1907, pp. 224-225, pi. 16, figs. 7-9) called Illaenus transvarsalis, for the pygidium of that species must have been very similar to that of Bumastus armatus. It is, however, not probable that this is the case, for cranidia of the Bamastus occur at the locality where our pygidium was found. The possibility that Weller's "Illaenus transver salts" may be a Bumastus rather than an Illaenus raises another question, however, for in 1903 Weller (Geol. Surv. New Jersey, Pal., vol. 3, p. 195, pi. 14. fig. 14) gave the name Bumastus transversalis to an Ordovician species from New Jersey, and, if the Niagaran species is also a Bumastus, its name will have been preoccupied and it will require a new one. The writer has not been able to locate in any museum a specimen of Weller's Illaenus transversalis that has any of the thorax pre- served, so he has not been able to determine whether that Illinois Silurian species is actually an Illaenus or a Bumastus. He will be grateful for information concerning any specimen that will afford evidence as to the true generic position of the species. Bumastus ioxus (Hall) Bumastus barriensis Hall (not Murchison), Geol. New York, pt. 4, 1843, p. 102, fig. 4, p. 101, tab. org. rem., 10, fig. 4, and 19, fig. 2. Illaenus ioxus Hall, 20th Rep. New Cork State Cab. Nat. Hist., 1867, p. 387, fig., pi. 22, figs. 4-11, pi. 23, fig. 1. Bumastus toxus (Hall), Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, p. 142 (gives full bibliography to 1910). Bumastus ioxus (Hall), Shimer and Shrock, Index fossils N. Amer., 1944, p. 639, pi. 269, figs. 13, 14. Two cephala, one large and one small, from gray dolomite of the Middle Silurian (Niagaran) Racine Formation at Racine, Wisconsin (no. 99) . Howell— VoGDES Collection of Trilobites 269 Bumastus cuniculus (Hall) Plate 3, fig. 7, plate 5, fig. 1. Illaemis cuntculus Hall, 20th Rep. New York State Cab. Nat. Hist., 1868, p. 337, pi. 22, fig. 12. Bumastus cuniculus (Hall), Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, p. 140 (gives full bibliography to 1907). A large cephalon, a small cranidium, a right free cheek, and a small pygidium, all from gray dolomite of the Middle Silurian (Niagaran) Racine Formation at Wauwatosa, Wisconsin. Both the large and the small cranidia have the elongate, narrow, glabella and short dorsal furrows that are characteristic of this species. The cephalon and cranidium and the pygidium are no. 101; the free cheek is no. 100. The large cranidium shows the doublure. The cephalon and the cranidium are figured here. All are topotypes. Unfortun- ately the pygidium is pardy covered by the small cranidium. Enough of it can be seen, however, to show that it, like the smaller cranidium, is of medium convexity throughout and is about as long as it is wide. Bumastus? vogdesi, new species Plate 3, fig. 2. A single pygidium from Dug Gap, Walker County, Georgia, that was presumably collected from the Middle Silurian Clinton beds present at that locality, would seem to be that of a new species of Illaenidae, probably a Bumastus. It is described below as such and is named in General Vogdes' honor. Pygidium, the only part of the test known, twice as wide as long, of low convexity, with the front corners evenly and obtusely rounded. No traces of an axis or dorsal furrow are visible on the single known specimen, which is exfoliated; but the axial portion of the front edge of the shield protrudes for- ward and clearly indicates the width of the axial part of the pygidium. The rear portion of the shield is not depressed, and there is no trace of a median ridge on our specimen. Location of type-The holotype pygidium is no. 95 in the Vogdes Collection. Formation and localtty-F'me grained brown sandstone of Clintonian (early Niagaran, early Medial Silurian) age at Dug Gap, Walker County, Georgia. Discussion-ln its general form this pyidium resembles the tail shield of Bumastus imperator (Hall), of the late Niagaran (late Medial Silurian) Racine Formation; but it differs from the pygidium of B. imperator in being longer in proportion to its width and in not having the anterior end of the axis defined by dorsal furows. If we may judge from our single shield, B. vogdesi was also a smaller species than B. imperator. 270 San Diego Society of Natural History A small piece of fine grained gray sandstone which accompanies the piece holding the holotype of Bumastus tmperator, and which therefore probably came from the same formation and locality, contains internal molds of two Orthoid brachiopods and the right half of the cranidium of what is probably an undescribed species of Addas pis. Unfortunately this last fragment is too incomplete and poorly preserved for identification or description. Genus and species undetermined A squeeze of an entire trilobite which may possibly be an illaenid of some kind, although its dorsal furrow is too well developed for that of a typical member of that family, is no. 50 in the Vogdes Collection. It bears no label, and its characters cannot all be clearly seen. It is possibly a squeeze of a specimen in some private collection that has never been figured. The writer hs not been able to identify it. Family Proetidae Proetus parviusculus Hall Proetus parviusculus Hall, 13th Rep. New York State Cab. Nat. Hist., 1860, p. 120. Proetus parviusculus Hall, Bassler U. S. Nat. Mus. Bull. 92, vol. 2, 1915, p. 1039 (gives full bibliography to 1895). An enrolled test from the Cincinnati region, and therefore presumably from the Upper Ordovician (Maysvillian) Corryville Formation (no. 24). The genal spines have been broken off because of their enrollment; but other- wise this is a good example of this small early proetid. Proetus, Species undetermined Plate 5, f^g. 2. A single small free cheek, referred by General Vogdes to Calymenella rostrata, but almost certainly the cheek of a small species of Proetus, is pre- served in fine grained gray sandstone of Clintonian, Medial Silurian age from Taylor's Ridge, near Catoosa Station, Catoosa County, Georgia (no. 129). It is impossible to determine the species to which this free cheek belongs, and it may be the cheek of an undescribed form. It is figured here to call attention to the presence of a small species of Proetus at this locality. Doubtless future investigators will discover cranidia of the species there. General Vogdes recorded "Proetus" from this locality (Amer. Jour. Sci. 3rd ser., vol. 18, 1879, p. 477), and his record may have been based on this free cheek, since there are no other examples of Proetus from Taylor's Ridge in his collection; but, if he did assign the cheek to that genus, he did not label it as such. Howell — Vogdes Collection of Trilobites 271 Proetus crassimarginatus (Hall) Plate 5, fig. 3. Calymene crassimarginatus Hall, Geol. New York, Geol. 4th Dist., 1843, p. 172, fig. 5. Proetus crassimarginatus (Hall), Bassler, U. S. Nat. Mus. Bull. 92, vol. 2, 1915, p. 1038 (gives full bibliography to 1909). Proetus crassimarginatus (Hall), Shimer and Shrock, Index fossils N. Amer., 1944, p. 651, pi. 274, figs. 18, 19. Two cranidia, a pygidium, and some fragments, labeled as having come from Vermont, but presumably collected somewhere in New York, appear to be referable to this species (no. 37) . They are preserved in a fine grained light gray limestone that must be of Devonian age and is probably from the Middle Devonian Onondaga Formation. Two other pygidia (nos. 40 and 41) in the same kind of rock (presumably also from the Onondaga Formation) are labeled as having come from the Devonian of Columbus, Ohio (no. 40) . and of Kelly Island, Lake Erie (no. 41), and the specimens labeled as having come from Vermont may also have come from one of these two localities. Proetus clarus Hall Proetus clarus Hall, 15th Rept. New York State Cab. Nat. Hist., 1862 pp. 99, 100. A single pygidium, which accompained the specimens of P. crassimargin- atus referred to above and is preserved in the same fine grained gray lime- stone, is believed to be a tail shield of this species (no. 37). Although its locality is given as "Vermont" and its formation is not recorded on its label, it is also probably from the Middle Devonian Onondaga Formation of New York. Proetus doris Hall Plate 3, figs. 8-9. Proetus dons Hall, 13th Rep. New York State Cab. Nat. Hist., 1860, p. 112. Proetus dons Hall, Weller, U. S. Geol. Surv. Bull. 153, 1898, p. 505 (gives full bibliographly to 1887). Pbillipsia dons (Hall), Herrick, Bull. So Labs. Denison Univ., vol. 2, 1887, p. 62. A cast of one of Hall's type pygidia of this species and an additional cast of an almost entire test (both no. 26) . Hall's types came from the Carbonif- erous "Goniatite Limestone" (Waverly Group) at Rockford, Indiana. The locality of the entire test is not indicated on the label, but is presumably also Rockford. These casts were probably used by General Vogdes in the preparation of his discussion of this species that was published in the Annals of the New York Academy of Sciences, vol. 4, 1887, pp. 90-91. In that paper 272 San Diego Society of Natural History General Vogdes referred the species to Phillipsia, as had been done earher by Alexander Winchell (Proc. Acad. Nat. Sci. Phila., 2nd ser., vol. 2, 1865, p, 133); but in his own personal copy of this paper, which is in the Vogdes Library, he has indicated that subsequent to the publication of the paper he decided that the species belonged in Proetus. His final opinion is probably the correct one, for Winchell states that the species has 11 thoracic segments, which is too many for a Phtllipsia, but not too many for a Proetus, and the pygidium, while it is long for a Proetus, is also short for a Phillipsia, Winchell appears to have assigned the species to Phillipsia because of the weakness of the glabellar furrows; but that feature alone is not diagnostic in doris, for its glabella has faint furrows, like those of some species of Proetus. P. doris is, in general, intermediate between a typical Proetus and a typical Phillipsia, as might be expected of an Early Carboniferous proetid. As this species appears never to have been figured, illustrations of our casts are presented here, although the casts are poor and difficult to photograph. The location of Hall's orginal types of the species is not known to the writer. They are not listed as being in either the collection of the New York State Museum or that of the American Museum of Natural History. Proetus haldemani Hall Plate 5, figs. 4-5. Proetus haldemani Hall, 15th Rep. New York State Cab. Nat. Hist., 1862, pp. 102-103. Proetus haldemani Hall, Hall and Clarke, Geol. Surv. New York, Palaeont. vol. 7, 1888, pp. 113-116, pi. 21, figs. 7-9, pi. 33, figs. 13-15 (gives full bibliography to 1887). Proetus haldemani Hall, Shimer and Shrock, Index fossils N. Amer. 1944, p. 653, pi. 74, fig. 6. A beautifully preserved enrolled example of this species, from the Middle Devonian Hamilton Formation at Judd's Falls, on the road between Cherry Valley and Sharon Springs, New York, is no. 39. The specimen is figured here. Proetus bairdensis Wheeler Plate 5, fig. 6. Proetus ellipticus Meek and Worthen, Vogdes, Proc. Calif. Acad. Sci. (meet- ing), Oct. 17, 1892, Zoe, vol. 3, 1892, p. 274. Proetus bairdensis Wheeler, Trans. San Diego Soc. Nat. Hist., vol. 8, no. 8, 1935, pp. 49-51, pi. 6, figs. 1-3 (gives full bibliography to 1894). There is in the Vogdes Collection a trilobite, preserved in a fine grained brown sandstone from the Middle Carboniferous Baird Formation at Baird, on the McCloud River, in Shasta County, California, that is labeled "Proetus ellipticus Meek and Worthen" (no. 31). The specimen is a mold of the under Howell — Vogdes Collection of Trilobites 273 side of the test and consists of the pygidium, most of the thorax, and the right free cheek. Although such parts of the test as are preserved indicate that the trilobite had the same general form as P. ellipticus, the following facts make is very probable that the Shasta County form is a distinct species: the axes of the thorax and pygidium appear to be narrower than the axes of that species (if one can judge from the published figures of the latter); the brim of the free cheek seems to have been wider; there must have been a brim on the pygidium (something which P. ellipticus did not have) ; ellipticus was described from Illinois, which is geographically so far distant from California that the species found in the two regions are not likely to be alike. Wheeler, who examined other specimens of this trilobite from the Baird Formation of Shasta County some years ago, was also of the opinion that they were not referable to P. ellipticus and proposed* for them the new name "Proetus bairdensis". Wheeler stated that reference to the specimen in the Vogdes Collection had been made in printf, but that the specimen was "no longer available for study", so he was apparently unaware that it was in the Vogdes Collection. As our specimen is in some respects a more complete one than the specimens on which Wheeler based his description of P. bairdensis, it is described in detail below. Of the various parts of the cephalon, only the free cheek is known. As nearly as can be determined from our specimen this cheek resembles that of P. ellipticus, as figured by Meek and Worthen (Geol. Surv. Illinois, vol 3, 1868, pi. 14, fig. 8) and by General Vogdes (Ann. New York Acad. Sci., vol. 4, 1887, pi. 3, fig. 3), except that the brim is wider. The posterior portion of the genal spine is broken off, however, and the spine may well have been longer that it was in ellipticus. The thorax of our specimen appears to have consisted of eight segments but presumably had nine, since Wheeler found nine in his specimens. The axial portion, which is more than one-third of the width of the thorax in ellipticus, is almost exactly one-third the width of the thorax in bairdensis. The axis of the pygidium is at its widest part (the front end) hardly one-third the width of the shield (it is fully one-third in ellipticus), tapers evenly backward more rapidly than in ellipticus, and is a little longer than in the Illinois species, so that the rear end has a more pointed appearance. There appear to have been seven ribs on the pleural lobes (the rear ribs are indistinct — Wheeler says there are eight or nine ribs on his specimens), and there must have been ten or more segments in the pygidial axis (the exact number in our specimen cannot be determined because of damage). Wheeler states that there are about thirteen segments in his specimens. There must have * Wheeler, Harry E., "New trilobite species from the Anthracolitic of northern California", Trans. San Diego Soc. Nat. Hist., vol. 8, 1935, pp. 49-51, pi. 6, figs. 1-3. t In Zoe, vol. 3, 1892, p. 274. The writer has not seen this reference. 274 San Diego Society of Natural History been a margin on the pyidium that was not only thickened (as the brim of ellipticus is said to be) but was also of some width. Location of plesiotype-No. 31 in the Vogdes Collection. Wheeler (op. cit., p. 51) states that a plastotype of one of his specimens is no. 272 in the trilobite collection of the San Diego Society of Natural History. Formation and locality-Ba.iTd Formation, Carboniferous, Baird, McCloud River, Shasta County, California. Wheeler (op. cit., p. 51) states that the beds in which P. bairdensis occurs are probably of Early Moscovian age, Middle Carboniferous. It is difficult to determine the exact proportions of the test of our specimen of P. bairdensis because the specimen is a m.old of the under side of the test and all the furrows and the brims are accentuated. But, when we make all possible allowance for this fact, the species would seem to be distinct from P. ellipticus. Other differences from that species will probably be disclosed when the entire cephalon of P. bairdensis is known. Cyphaspis christyi Hall Plate 5. fig. 7-8. Cyphaspis christyi Hall, Trans. Albany Inst., vol. 4, 1864, p. 220. Cyphaspis christyi Hall, Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, p. 339 (gives full bibliography to 1889). Cyphaspis christyi Hall, Shimer and Shrock, Index fossils N. Amer. 1944, p. 643, pi. 276, figs, 11-13. An enrolled test, labeled as having come from the Middle Silurian Niagaran Group (no. 156). The label does not indicate the locality, but the specimen looks as though it had been collected from the Waldron Formation of Waldron, Indiana. Brachymetopus lodiensis (Meek) Phillipsia (Grijjithides) lodiensis Meek, Pal. Ohio, vol. 2, 1875, p. 323, pi. 18, fig. 3. Brachymetopus lodiensis (Meek), Weller, U. S. Geol. Surv. Bull. 153, 1898, p. 149 (gives full bibliography to 1888) . Brachymetopus lodiensis (Meek) , Shimer and Shrock, Index fossils N. Amer. 1944 p. 639, pi. 275, fig. 28. An incomplete test in gray shale from the Lower Carboniferous Cuyahoga Formation at Lodi, Ohio, the type locality of the species (no. 19). Brachymetopus cuyahogae (Claypole) Plate 6, fig. 1. Dalmanites? cuyahogae Claypole, Geol. Mag. Dec. 3, vol. 1, 1884, p. 306, text-fig. A (p. 305). Howell — Vogdes Collection of Trilobites 275 An incomplete pygidium in black shale of the Lower Carboniferous Cuyahoga Formation from Akron, Ohio (no. 66) . This species was described by Claypole in 1884 on the basis of the pygidium alone. Claypole considered it to be a species of Dahnanites, but it is a Brachymetopus. Its pygidium is remarkable because each of the ribs of the pleural lobes ends in a spine; but other species of Brachymetopus, such as B. armatus Vogdes, in which the last pair of ribs on the pleural lobes are extended into spines, and B. spinosus (Herrick), which has short spines on each of the ribs of the pleural lobes, also have spinose pygidia. As Claypole's figure of the holotype pygidium of this species would appear to be inaccurate in that the spines are made to appear too long and the edge of the shield too uneven, the pygidium in the Vogdes Collection is figured here, even though it is incomplete. Phillipsia insignis Winchell Plate 5, f^g. 9. PhtlUpsia insignis Winchell, Proc. Acad. Nat. Sci. Phila. 1863, p. 24. Phillipsia msigms Winchell, Waller, U. S. Geol. Surv. Bull. 153, 1898, p. 421 (gives full bibliography to 1887). Casts of three pygidia that are labeled as being this species and probably are such, from Curryville, Missouri (no. 34) . As the species was described from the Lower Carboniferous Burlington Limestone these specimens probably came from that formation. Since this species appears not to have been figured an illustration of one of these casts is presented here. Phillipsia major Shumard Plate 6, figs. 2-3. Phillipsia major Shumard, Trans. St. Louis Acad. Sci., vol. 1, 1858, p. 226. Phillipsia major Shumard, Weller, U. S. Geol. Surv. Bull. 153, 1898, p. 422 (gives full bibliography to 1897). A complete pygidium and an almost complete cranidium from a white Upper Carboniferous limestone at Kansas City, Missouri (no. 33). The pygidium agrees with the description of the tail shields on which the species is based. The cranidium has the glabella almost quadrate in form, but some- what rounded in front and a little expanded at the rear. It is difficult to de- termine from our specimen, which is largely exfoliated, how many glabellar furrows are present; but there appears to be a pair of incomplete, backward- curving furrows opposite the front ends of the eyes and a pair of longer furrows that curve backward from a point opposite the middle of the eyes to the neck furrow to mark off large basal lobes. The preglabellar area is about as wide as the eye lobe, and the neck ring is about equally wide. The whole shield is moderately convex, and the test appears to have been smooth. 276 San Diego Society of Natural History The writer has not seen an entire example of this species; but General Vogdes has left a manuscript description of such a specimen on the margin of page 85 of his personal copy of his paper. "The genera and species of North American Carboniferous trilobites" (Ann. New York Acad. Sci., vol. 4, 1887, pp-105, pis. 2 and 3) that reads as follows: "Remarks on an entire specimen from collection of F. A. Sampson, Sedalia, Mo. Pygidium axis well marked with 16+ rings. Sides with 10 ribs, surface smooth. Thorax with 9 segments, test smooth. Glabella and fixed cheeks. Glabella convex, with nearly straight sides, slightly tumid, marked on both sides with 2 short furrows opposite the eyes and by well defined basal lobes. Occipital lobe prominent, below the plane of the glabella. Occipital furrow deep. Limb concave." Phillipsia sampsoni Vogdes Plate 6, fig. 4. Phillipsia sampsoni Vodges, Trans. New York Acad. Sci., vol. 7, 1888, p. 248, 2 figs. Phillipsia sampsoni Vodges, Keyes, Missouri Geol. Surv., vol. 4, 1894, p. 235. Phillipsia sampsoni Vogdes, Weller, U. S Geol. Surv. Bull. 153, 1898, p. 423. Phillipsia sampsoni Vogdes, Shimer and Shrock, Index fossils N. Amer. 1944, p. 651, pi. 276, figs. 1-3. A beautiful small entire test in gray limestone of the Lower Carboniferous (Waverly Group) Chouteau Formation at Banks, Pettis County, Missouri (no. 36). In his orginal description of this species General Vogdes stated that there were three pairs of glabellar furrows, althought his figure showed only two. A very faint trace of a fourth furrow can be seen on the glabella of our specimen. Whether this is due to the fact that our specimen is unusally well preserved, or whether it is due to its being somewhat smaller than (and therefore younger than) the specimen figured by General Vogdes, the writer is unable to say. Phillipsia stevensoni Meek Plate 6, fig. 5. Phillipsia stevensoni Meek, 3rd Ann. Rept. Board Regents West Virginia Univ., 1870, p. 73. Phillipsia stevensoni Meek, Vogdes, Ann. New York Acad. Sci., vol. 4, 1888, p. 88, pi. 3, fig. 6. Phillipsia stevensoni Meek, Weller U. S. Geol. Surv. Bull. 153, 1898, p. 424. A pygidium, with two thoracic segments attached, in gray shale of the Lower Carboniferous Chester Group from Monongalia County, West Virginia (no. 35). Howell — Vogdes Collection of Trilobites 277 Ditomopyge scitula (Meek and Worthen) Griffithides scitula Meek and Worthen, Proc. Acad. Nat. Sci. Phila., 1865, p. 270. Griffithides scittda Meek and Worthen, Weller, U. S. Geol. Surv. Bull. 153, 1898, p. 303 (gives full bibliography to 1897). Ditomopyge scitula (Meek and Worthen), Shimer and Shrock, Index fos- sils N. Amer., 1944, p. 645, pi. 275, figs. 4^6. Two incomplete pygidia in gray shale from the Upper Carboniferous at the "Willard Coal Mine, Carter County, Kentucky, south terminus of East Kentucky R. R.", appear to be referable to this species (no. 20) . Griffithides bufo Meek and Worthen Plate 6, fig. 6; plate 7, fig.l. Phillipsia (Griffithides) bufo Meek and Worthen, Proc. Acad. Nat. Sci. Phila., 1870, p. 52. Giffithides bufo Meek and Worthen, Weller, U. S. Geol. Surv. Bull. 153, 1898, pp. 301-302 (gives full bibliography to 1887). Giffithides bufo Meek and Worthen, Shimer and Shrock, Index fossils N. Amer., 1944, p. 647, pi. 275, fig. 9. An excellently preserved entire test (no. 32), the cast of a good and somewhat larger test (no. 25), and a poorly preserved entire test of still larger size (no. 32), all from gray limestone of the Lower Carboniferous Keokuk group of Crawfordsville, Indiana. The smaller test and the cast are figured. Family Odontopleuridae Ceratocephala anchoralis (Miller) Plate 7, fig.2. Acidaspis anchoralis Miller, Cincinnati Quart. Jour. Sci., vol. 2, 1875, p. 349, figs. 23, 24. Acidaspis anchoralis Miller, N. Am. Geol. Pal., 1889, p. 526, text. figs. 951, 952. Ceratocephala anchoralis (Miller) Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, p. 196. Two cranidia in gray shale of the Upper Ordovician Maysville Formation of the Cincinnati region, Ohio, are probably both referable to this species, although one of them is labeled ''Acidaspis cincin7iatiensis'\ The larger of these two cranidia (no. 109) has a very long and heavy neck spine, while the smaller one (no. HI) has a much shorter and more slender one. The figure of the type cranidium given by Miller has a neck spine that is intermediate in size between those of our two specimens. It 278 San Diego Society of Natural History would seem that if all three of these crandia are referable to Ceratocephala anchoralis there must have been much varation in the size of the neck spine in that species. Possibly the individuals with the heavier spines were males, those with more slender spines females. It is possible that one or both of these cranidia are referable to one of the other two species of Ceratocephala which have been described from the Upper Ordovician beds of the Cincinnati region, Ceratocephala ceralepta Anthony and C. cmcinnatiensis (Meek); but since these two species appear to have been based on pygidia alone and their cranidia seem not to be known, this question must be left open for the present. Both of these species were described before C. anchoralis, and the discovery of an entire test of either of them may demonstrate that the cranidium of C. anchoralis belongs with the pygidium of one of the other species. Order Proparia Family Encrinuridae Encrinurus americanus Vogdes Plate 7, figs. 3-4. Encrinurus americanus Vogdes, Descr. new Crust. Clinton of Georgia, 1886, p. 1. Encrinurus americanus Vogdes, Foerste, Bull. Sci. Lab. Denison Univ., vol. 2, 1887, p. 102. Encrinurus atnericanus Vogdes, Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, p. 477. The cotypes of this species, preserved in fine grained gray sandstone, are no. 12 in the Vogdes Collection. As their original description appeared without illustrations, in a private publication that is not available to most students and the species has not been figured since, the orginal description is reproduced here and both pygidia are figured. The original description, which appeared in a privately printed paper, entitled "Description of a new crustacean from the Clinton Group of Georgia, with remarks upon others", was published in New York City in 1886 (pp. 1-5, text-figs. 1-4). In addition to the description of Encrinurus americanus this paper contained discussions, with figures, of the trilobites Calymene rostrata Vogdes (Calymenella rostrata) and Calymene clintoni Vanuxem (Calymene vogdesi Foerste), both, like Encrinurus americanus, from Clintonian Middle Silurian beds in Georgia. The original description of Encrinurus americanus was as follows : "Pygidium elongated-triangular in form, strongly arched in the anterior border and very convex, produced behind forming a muronate (mucronate) extension. Length 10 mm., greatest width 8 mm., or nearly % as wide as long. "Axis narrowly elongated, flattened on the surface and sharply pointed behind, marked by 20 or more axial rings; the first 13 extend entirely across, Howell — Vogdes Collection of Trilobites 279 the others appearing only as notches on the sides, obscurely indicated in the specimens before us. The axis has a central row of nodes, which appear on the 1, 6, 9, and 12 ring (rings), and possibly on others. "The lateral lobes are marked by six pairs of ribs, the anterior three of which are distinct and free at their outer extremities, the other three being united at their outer ends, the posterior pair being parallel to and appearing almost as a part of the axis. The ribs are abruptly directed backward, and distinctly separated from each other by rather deep grooves, they gradually decrease in size as they approach the termination of the pygidium, being rather flat on the surface. The first pair originate almost opposite the first axial ring, the second between the 2 and 3, the third between the 5 and 6, the fourth between the 7 and 8, the fifth pair run sub-parallel to the axis and the sixth pair are only separated from it by a shallow dorsal furrow. "This species approaches Encnnurus elegantulus, Billings (Cat. Silurian Foss. Anticosti, p. 62), in general form, the axis of this species has 24 axial ring (rings) of which the first 8 or 9 extend entirely across it, the others being represented by elongated pits on each side, but it lacks the nodes. The lateral lobes of Encnnurus elegantulus are marked with five pair (pairs) of ribs, the first four being free at their extremities, the fifth pair being slightly curved outwards, and converging toward the axis they unite behind it, and continue as two short sharp spines. Between the fifth rib and the side of the axis there is a narrow smooth space with a faint groove on the inside, it extends round the apex of the axis and seems to be the rudiments of a sixth pair of ribs. "The principal differences between these species lies in the character of the lateral lobes, Encnnurus Americanus having, three free and three united ribs in the pygidium; whereas (in) Encnnurus elegantulus the first four are free and the other two unite at their extremities, the side ribs originate at the first axial ring, the second pair at the second, the third pair at the fourth, and the fourth pair at the sixth; whereas in the new species the third pair originate between the 5 and 6 axial ring, and the fourth pair between the 7 and 8. The minor differences with regard to the nodes on the axis may be due to the state of preservation. "The pygidia of ^ethus verrucosus and Z^ethus bellatula described and figured by Dr. Volborth (Russ. Mineral. Gesellsch. Verhandl., 1848, p. 4, pi. 1, figs. 3, 4, and 7), are in general form similar to those of Encrinurus elegantulus. They all have from 4 to 6 side ribs, the anterior ones being free, and the last two or three united at their extremities in rear of the axis. The axis in the Russian species has from 16 to 18 rings, the first 4 or 5 anterior ring (rings) only bearing side ribs, whereas the American species has from 20 to 24 rings, the first 8 to 13 rings extending across it, the side lobes extending down in one species at least to the 8 arial (axial) ring. "The Russian species have a notched axis, with nodes along the smooth central part. "Geological Position-CXmton Group, Taylor's Ridge, west of Catoosa Station, Catoosa Co., Georgia." 280 San Diego Society of Natural History Family Calymenidae Calymene vogdesi Foerste Plate 7, figs. 5-7. Calymene dintoni Vogdes (not Vanuxem), Proc. Acad. Nat. Sci. Phila. 1880, p. 178, figs. 3-4. Calymene vogdesi Foerste, Bull. Sci. Lab. Denison Univ., vol. 2, 1887, p. 95, pi. 8, figs. 12-16. Calymene vogdesi Foerste, Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, pp. 169-170 (gives full bibliography to 1910). Calymene vogdesi Foerste, Raymond, Bull. Mus. Comp. Zool., vol. 60, 1916, p. 27. An incomplete cranidium and a free cheek, both in fine grained gray sandstone, and a pygidium in fine grained dark red sandstone, all from Clintonian, Middle Silurian, beds at Dug Gap, Walker County, Georgia, are no. 133 in the Vogdes Collection. Three other cranidia in fine grained gray sandstone of the same age from Taylor's Ridge, near Catoosa Station, Catoosa County, Georgia, are no. 132. These are evidently some (but not all) of the specimens which General Vogdes identified as Calymene clintoni (Vanuxem) in 1880 (Proc. Acad. Nat. Sci. Phila., pp. 177-178, text-figs. 3 and four) and in 1886 (Description of a new crustacean from the Clinton Group of Georgia, with remarks on others, p. 5, text-figs. 3 and 4) and which Foerste (Bull. Sci. Labs. Denison Univ., vol. 2, 1887, p. 96) referred to his species, Calymene vogdesi. The cranidium on which text-figures 3 of both of General Vogdes' papers were based is in the Vogdes collection and is reproduced here. Our single free cheek does not appear to be the one from which figures 4 of those two papers were made (General Vogdes stated that he had three free cheeks), and it is therefore figured here also. Our pygidium, which is probably that of a young individual, was not figured by General Vogdes, but is illustrated here. It was said by General Vogdes to have no ribs on the pleural lobes; but, although it is exfoliated, faint traces of such ribs can be seen and they were probably more distinct in unexfoliated shields of the same size. Foerste described the adult pygidium as having six ribs on the pleural lobes. Calymene celebra Raymond Calymene blumenbachii ntagarensis Hall, Geol. Surv. Wisconsin, vol. 1, 1862, p. 432. Calymene niagarensis Hall, 18th Rept. New York State Cab. Nat. Hist., 1865, p. 30 adv. sheets. Calymene niagarensis Hall, 20th Rept. New York State Cab. Nat. Hist., 1868, p. 334; revised edition, 1870, p. 425. Calymene niagarensis Hall, Weller, Bull. Chicago Acad. Sci., no. 4, pt. 2, 1907, p. 261, pi. 23, figs. 9, 19. Howell — Vogdes Collection of Trilobites 281 Calymene magarensis Hall, Bassler (pars), U. S. Nat. Mus. Bull. 92, 1915, p. 168 (refers in part to true Calymene magarensis Hall). Calymene celebra Raymond, Bull. Mus. Comp. Zool., vol. 60, 1916, pp. 28-29, pi. 3, figs. 9, 10. Calymene celebra Raymond, Shimer and Shrock, Index fossils N. Amer. 1944, p. 641, p. 272, figs. 6, 7. Two entire tests (nos. 130 and 131) in gray Niagaran (Middle Silurian) dolomite from Preble County, Ohio, one enrolled test (no. 157) in yellowish dolomite of the Niagaran Racine Formation at Wauwatosa, Wisconsin, two entire tests (no. 127) in gray dolomite, probably of Niagaran age, from Illinois, an entire test, in gray dolomite (no. 163), labeled as having come from the Cincinnati Group at Cincinnati but probably from some Niagaran formation of the Mississippi Valley, and an entire test in reddish dolomite (no. 126), labeled as having come from a "Lower Silurian" formation but probably also from a Niagaran formation of the Mississippi Valley, all have the kind of wide pygidium which is characteristic of Calymene magarensis, with only six segments in the axis and with only four short ribs on the pleural lobes, the outer halves of the lobes being unribbed. Calymene platys Green Calymene platys Green, Mon. trilobites N. Amer., 1832, p. 32. Calymene platys Green, Hall, 15th Rept. New York State Cab. Nat. Hist., 1862, p. 82. Calymene platys Green, Hall, lUustr. Devonian fossils, 1876, pi. 1, figs. 1-9. Calymene platys Green, Hall and Clarke, Geol. Surv. New York, Palaeont., vol. 7, 1888, pp. 1-4, pi. 1, figs. 1-9, pi. 25, fig. 1. Calymene platys Green, Shimer and Shrock, Index fossils, N. Amer. 1944, p. 641, pi. 272, figs. 8-10. An incomplete cranidium, in black limestone, labeled as having come from the Middle Devonian Onondaga Formation of Genesee County, New York, would appear to be referable to this species (no. 134). It is much smaller than the cranidia of this species figured by Hall, but may be the head shield of a young individual. Flexicalymene senaria (Conrad) Calymene senaria Conrad, 5th Ann. Rep. Geol. Surv. New York, 1841, pp. 38, 49. Calymene senaria Conrad, Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, p. 169 (gives full bibliography to 1913). Flexicalymene senaria (Conrad), Whittington, Jour. Paleont., vol. 15, 1941, pp. 493-498, pi. 72, figs. 1-27, 31-34, 38-40, 42-47. 282 San Diego Society of Natural History Flexicalymene senaria (Conrad), Shimer and Shrock, Index fossils of North America, 1944, p. 645, pi. 272, figs. 3-5. A pygidium in black limestone of the Ordovician Trenton Group from Crab Island, Lake Champlain, has the furrowed ribs of the pleural lobes extending all the way to the rear edge of the shield, as is characteristic of this species (no. 163). Flexicalymene meeki (Foerste) Calymene senaria Meek, Geol. Surv. Ohio, Pal. 1, 1873, p. 173, pi. 14, figs. 14a-f. Calymene meeki Foerste, Bull. Sci. Lab. Denison Univ., vol. 16, 1910, p. 84, pi. 3, fig. 18. Calymene meeki Foerste, Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, p. 167 (gives full bibliography to 1913). Flexicalymene meeki (Foerste), Shimer and Shrock, Index fossils N. Amer., 1944, p. 645, pi. 272, figs. 1, 2. Apparently referable to this species are an enrolled test (no. 155), which has no orginal label but is almost certainly from some Upper Ordovician Maysvillian or some Richmondian formation of the Cincinnati region; a cephalon and attached thorax (no. 135), which are labeled as having been collected from Upper Silurian rocks in New York but are very probably also from one of the Maysvillian or Richmondian formations of the Cin- cinnati region; and four small enrolled tests (no. 138), which are labeled as having come from the Cincinnati region and are therefore likewise from some Maysvillian or Richmondian formation. All of them are excellently preserved. Liocalymene clintoni (Vanuxem) Hemicrypturus clintoni Vanuxem, Geol. New York, pt. 3, 1842, p. 79, Fig. 2. Calymene clinto?ii (Vanuxem) , Hall Palaeontology of New York, vol. 2, 1852, p. 298, pi. A66, figs. 5a-d. Calpmene clintoni (Vanuxem), Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, p. 166 (gives full bibliography to 1909). Liocalymene clintoni (Vanuxen), Raymond, Bull. Mus. Comp. Zool., vol. 60, 1916, p. 29. Two small cranidia with very narrow brims, preserved in fine grained gray sandstone (presumably from some Clintonian, Middle Silurian, bed) in Catoosa County, Georgia, agree in form with this species (no. 128). They are smaller than the average adult cranidium of C. clintoni and they may be a southern variety of that species; but it is more probable that they are head shields of young individuals of clintoni, itself Howell — Vogdes Collection of Trilobites 283 These specimens were presumably collected by General Vogdes, him- self, but he seems to have been in doubt as to their specific identity, for he labeled them merely "Calymene sp.?". Calymenella rostrata (Vogdes) Plate 7, figs. 8-10. Calymene rostrata Vogdes, Amer. Jour. Sci., 3rd ser., vol. 18, 1879, p. 477. Calymenella rostrata (Vogdes), Pompeckj, Neues Jahrb. Min., Geol., Pal., 1898, vol. 1 ,p. 243. Calymenella rostrata (Vogdes), Bassler, U. S. Nat. Mus. Bail. 92, vol. 1, 1915, p. 170 (gives full bibliography to 1910). An almost entire cranidium, a small fragment of a cranidium, part of a small cranidium, an incomplete adult free cheek, a very small free cheek, and two adult pygidia, one exfoliated and one not, are labeled "Calymene rostrata Vogdes" by General Vogdes (nos. 129 and 136). These are presumably the cotypes of this species, from the Middle Silurian Clinton beds at Catoosa Station, Georgia, for the almost entire cranidium (no. 129A) and the exfoliated pygidium (no. 136A) appear to be the specimens on which figures 1 and 2 of General Vogdes second description of Calymenella rostrata was based (Proc. Acad. Nat. Sci. Phila., 1880, pp. 176-177, text- figs. 1 and 2). Whether all of these specimens are actually referable to this species is doubtful, but this cranidium certainly is, and the figured pygidium probably is, also. The fragment of a cranidium (no. 136B) is so incomplete that its certain identification is not possible, and the large free cheek (no 136C) is possibly referable to Calymene vogdesi. The small free cheek (no. 129C) is probably not even the free cheek of a Calymenella, but is rather that of a Proetiis, as noted on an earlier page. The unexfoliated pygidium (no. 129B) differs somewhat in form and proportions from the exfoliated one, but seems to have the same number of axial segments (about eight) as the exfoliated shield and definitely has the same number of ribs on the pleural lobes-five, so it is probably a tail shield of C. rostrata. An additional small pygidium, which is comparable in size to the smaller of the cranidia that are assigned above to Calymene clintoni and was placed with those cranidia by General Vogdes, is, however, probably the tail shield of a young Calymene rostrata (no. 128), as it agrees in form with the much larger, exfoliated, pygidium of that species discussed above. Until entire tests of either Calymene vogdesi or Calymenella rostrata are discovered it will not be possible for us to be certain which of the calymenid pygidia found in the Clintonian beds of Georgia are referable to each of those species. The pygidium that is here assigned to Calymene vogdesi has the almost straight rear edge and subtriangular form of the pygidium assigned by Foerste to C. vogdesi (Bull. Sci. Labs. Denison Univ., vol. 2, 1887, pi. 8, fig. 16), but the unexfoliated pygidium assigned by General Vogdes to Calymenella rostrata has much the general appearance. However, the ex- 284 San Diego Society of Natural History foliated pygidium assigned by General Vogdes to Caiymenella rostrata has the rear edge more curved than it is in the small pygidium assigned to Calymene vogdesi, and the axis of the larger pygidium is narrower in porportion to its length than is that of the smaller pygidium of C. vogdesi. The larger (C. rostrata) pygidia are also more arched upward at the rear than is the smaller (C. vogdesi) pygidium, and in this respect they agree with the cranidia of C. rostrata, which are more convex in form than are those of C. vogdesi. Family Homalonotidae Trimerus delphinocephalus Green Trlmerus delphinocephalus Green, Mon. triobites N. Amer., 1832, p. 82, pi. 32, fig. 1. Homalonotiis delphinocephalus (Green), Murchison, Sil. System. 1839, p. 651. Homalonotus delphinocephalus (Green), Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, p. 632-633 (gives full bibliography to 1910). Trimerus delphinocephalus Green, Shimer and Shrock, Index fossils N. Amer. 1944, p. 654, pi. 272, fig. 33. An entire test, preserved in brown weathered limestone presumably of Medial Silurian age and from an unknown locality (no. 153), and a pygidium with thorax attached, in black limestone, also of Medial Silurian age, from Lockport, New York. Dipleura dekayi Green Dipleura dekayi Green, Monograph trilobites N. Amer., p. 29, pi. 1, figs. 8, 9. Homalonotus dekayi (Green), Vogdes, Occ. Pap. Calif. Acad. Sci., no. 4, 1893, p. 311 (gives bibliography to 1888). Dipleura dekayi Green, Shimer and Shrock, Index fossils N. Amer. 1944, p. 643, pi. 272, figs. 26-31. A cephalon and attached thorax, preserved in black limestone, labeled as having come from the Middle Devonian Hamilton Group of New Yofk (no. 144). Family Cheiruridae Ceraurus pleurexanthemus Green Plate 8, figs. 1-2 Ceraurus pleurexanthemus Green, Monthly Amer. Jour Geol., 1832, p. 560, pi. 4, fig. 10. Ceraurus pleuerexanthemus Green, Raymond and Barton, Bull. Mus. Comp. Zool., vol. 54, 1913, pp. 528-533, pi. 1, fig. 1, pi. 2, figs. 1, 2, 7. Howell — Vogdes Collection of Trilobites 285 Ceraurus pleurexaiithemus Green, Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, p. 203 (gives full bibliography to 1913). Ceraurus pleurexanthemus Green, Wilson, Canadian Geol. Surv. Bull. no. 9, 1947, pp. 50, 51, pi. 9, figs. 3-5. A young cephalon and attached thorax, preserved in black Middle Ordovician limestone (probably of the Trenton Formation) from New York (no. 137), and an adult cranidium and fragment of a young cranidium, also in black limestone and definitely from the Middle Ordovician Trenton Formation, of Trenton Falls, New York (no. 160). The glabella of the young cephalon is less convex than that of the adult cranidium, and this makes the glabellar furrows of the young shield more noticable, although they are no longer and no more deeply impressed than they are in the adult one. Sphaerexochus romingeri Hall Sphaerexochus mtrus Roemer, Die silurische Fauna des westlichen Tennessee, 1860, p. 81, pi. 5, fig. 20. Sphaerexochus romingeri Hall, Geol. Surv. Wisconsin, vol. 1, 1867, p. 434. Sphaerexochus romingeri Hall, Bassler, U. S. Nat. Mus. Bull. 92, vol. 2, 1915, p. 1167 (gives fully bibliography to 1910). Sphaerexochus romingeri Hall, Shimer and Shrock, Index fossils N. Amer. 1944, p. 653, pi. 272, figs. 17-22. Two young and three adult cranidia in light gray dolomite of the Middle Silurian (Niagaran) Racine Formation at Wauwatosa, Wisconsin (no. 162). Family Dalmanitidae Dalmanites limulurus (Green) Asaphus limulurus Green, Mon. trilobites N. Amer. 1832, p. 48. Dalmanites limulurus (Green), Billings, Geology of Canada, Geol. Surv. Canada, 1863, p. 320, text-fig. 340. Dalmanites limulurus (Green), Bassler, U. S. Nat. Mus. Bull. 92, vol. 1, 1915, pp. 384-385 (gives full bibliography to 1913). An entire test (no. 74) and a pygidium and attached thorax (no. 62), both in dark gray Middle Silurian (Niagaran) limestone from Lockport, New York. Odontochile concinnus (Hall) Dalmanites concinnus Hall, Illustrations of Devonian fossils, Crustacea, 1876, pi. 10, figs. 3-5. Dalmanites (Hausmannia) concinnus Hall and Clarke, Pal. New York, vol. 7, 1888, p. 30, pi. 11 A, figs. 9-11. 286 San Diego Society of Natural History A single incomplete pygidium in light gray limestone, presumably of Onondagan (Medial Devonian) age, from Columbus, Ohio (no. 81). Odontochile micrurus (Green) Asaphus mtcrurus Green, Mon. trilobites N. Amer., 1832, p. 56, cast 19, pi., fig. 3. Dalmama micrurus (Green), Hall, Pal. New York, vol 3, 1859, pp. 359-361, pi. 74, figs. 13-20. Odontochile mtcrurus (Green), Shimer and Shrock, Index fossils N. Amer., 1944, p. 651, pi. 273, figs. 3, 11. A pygidium in dark gray limestone, labeled "Upper Silurian, New York", but undoubtedly of Lower Devonian age (no. 60). Anchiopsis anchiops (Green) Plate. 8, fig. 3. Calymene anchiops Green, Mon. trilobites N. Amer., 1832, p. 35, cast 7. Dalmanttes anchiops (Green), Vogdes, Occ. Pap. Calif. Acid. Sci., no. 4, 1893, p. 297 (gives bibliography to 1888). Anchiopsis anchiops (Green), Delo, Jour. Paleont., vol. 9, 1935, p. 412. Anchiopsis anchiops (Green), Shimer and Shrock, Index fossils N. Amer., 1944, p. 637, pi. 274, figs. 20-22. Casts which appear to have been made from Green's holotype pygidium of this species are nos. 72 and 73 in the Vogdes Collection. Cast no. 72 is marked in pencil "16", and a paper label attached to it states that it is Green's cast no. 16, which is the cast of the holotype of Dalmanites limulurus; but this is an error. The original cast of the holotype of anchiops is Green's cast no. 7. As the cast of the holotype of A. anchiops seems not to have been figured, an illustration of cast no. 72 is presented here. Corycephalus dentatus (Barrett) Plate 8, fig. 4. Dalmanites dentatus Barrett, Amer. Jour. Sci. 3rd ser., vol. 11, 1876 p. 200, pi. Dalmanites dentata Barrett, Amer. Jour. Sci., 3rd ser., vol. 12, 1876, p. 70. Dalmanites (Corycephalus) dentatus, Barrett, Hall and Clarke, Pal. New York, vol. 7, 1888, p. 58, pi. 11 A, figs. 4-6. Two cephala and a pygidium (no. 59) and three pygidia and a cranidium (no. 58) , all in gray Lower Devonian limestone from Port Jervis, New York. The better preserved of the two cephala (no. 59) is figured here. Howell — Vogdes Collection of Trilobites 287 Coronura aspectans (Conrad) Asaphus aspectans Conrad, 5th Ann. Rept. Pal. Dept. New York Geol. Surv., 1841, p. 49, text-fig. 9. Dalmanites aspectans (Conrad) , Hall, Illustrations of Devonian Fossils, Crustacea, 1876, pi. 13, figs. 6-8. Dalmanites (Coronura) aspectans (Conrad), Hall and Clarke, Pal. New York, vol. 7, 1888, p. 33, pi. 13, figs. 1-11, 13. Dalmanites aspectans (Conrad) , Vogdes, Occ. Pap. Calif. Acad. Sci. no. 4, 1893, pp. 297-298 (gives full bibliography to 1888). Coronura aspectans (Conrad), Shimer and Shrock, Index fossils N. Amer., 1944, p. 643, pi. 274, fig. 27. Two pygidia in gray limestone of the Middle Devonian Onondaga Formation from Kelly Island, Lake Erie (nos. 71 and 76). Calliops callicephalus (Hall) Phacops callicephalus Hall, Pal. New York, vol. 1, 1847, p. 249, pi. 65, figs. 3a-i. Dalmanites callicephalus (Hall), Vogdes, Occ. Pap. Calif. Acad. Sci., no. 4^ 1893, p. 298 (gives full bibliography to 1855). Calliops callicephalus (Hall), Delo, Jour. Paleont., vol. 9, 1935, p. 417, text- figs. 42, 43 (p. 414). Calliops callicephala (Hall), Shimer and Shrock, Index fossils N. Amer. 1944, p. 639, pi. 273, figs. 20-23. Calliops callicephalus (Hall), Wilson, Canadian Geol. Surv. Bull. no. 9, 1947, pp. 56, pi. 10, figs. 5, 6. A single pygidium in black limestone, probably of the Ordovician Trenton Formation, from New York (no. 68) . Achatella car ley i (Meek) Plate 8, fig. 5. Dalmanites carleyi Meek, Amer. Jour. Sci., 3rd ser., vol. 3, 1872, p. 424. Pterygometopiis carleyi (Meek), Bassler, U. S. Nat. Mus. Bull. 92, 1915, p. 1066 (gives full bibliography to 1874). A cephalon, complete except for the ends of the genal spines, in gray calcareous shale, presumably from the Upper Ordovician (Maysvillian) Fairmount Formation at Cincinnati, Ohio (no. 64) . 288 San Diego Society of Natural History Family Phacopidae Phacops rana Green Plate 9, fig. 1. Calymene bufo rana Green, Mon. trilobites N. Amer., p. 42, casts 11, 12. Phacops rana (Green), Hall, Descriptions of new species of fossils from the Upper Helderberg, Hamilton and Chemung groups (advance sheets of 15th Rept. New York State Cab. Nat. Hist.), 1861, p. 65. Phacops rana (Green), Vogdes, Occ. Pap. Calif. Acad. Sci. no. 4, 1893, pp. 334-335 (gives full bibliography to 1888). Phacops rana (Green), Shimer and Shrock, Index fossils N. Am., 1944, p. 651, pi. 274, figs. 2-5. An entire test, two incomplete tests, and a cranidium, preserved in gray shale, and an entire test, enrolled and weathered to a buff color, all said to have been collected from the Middle Devonian Moscow Formation of the Hamilton Group of New York (no. 70), and two entire tests and parts of seven other tests, all from shale of the Moscow Formation type at London, Ontario (nos. 80 and 86). The specimens from London have had their under sides exposed by removal of the matrix so that the features of the lower side of the test can be observed, and the hypostomes and doublures can be seen in place in the two entire tests. Phacops cacapona Hall Plate 9, figs. 2-4. Phacops cacapona Hall, Description of new species of fossils from the Upper Helderberg, Hamilton and Chemung groups (advance sheets of 15th Rept. New York State Cab. Nat. Hist.), 1861, p. 68. Phacops cacapona Hall, Vogdes, Occ. Pap. Calif. Acad. Sci., no. 4, 1893, p. 333 (gives full bibliography to 1888) . An excellently preserved enrolled specimen of this large species, from Capon Springs, West Virginia, presumably from the Middle Devonian Hamilton Group (no. 82) . Phacops hudsonica Hall Plate 9, fig. 5; plate 10, fig. 1. Phacops hudsonica Hall, Pal. New York, vol. 3, 1859, p. 335, pi. 73, figs. 26-28 An excellently preserved enrolled entire test and a poorly preserved cranidium from the Lower Devonian of western Tennessee (no. 88) appear to be referable to this species. The cephalon agrees exacdy with the holotype cephalon. The thorax has 11 segments. The ends of the pleura are rounded. The pygidium has about 7 segments in the axis and six ribs on the pleural lobes. These ribs extend to the edge of the shield. Howell — Vogdes Collection of Trilobites 289 Phacops cristata Hall Phacops cristata Hall, Description of new species of fossils from the Upper Helderberg, Hamilton and Chemung groups (advance sheets of 15th Rept. New York State Cab. Nat. Hist.), 1861, p. 67. Phacops cristata Hall, Vogdes Occ. Pap. Calif. Acad. Sci., no. 4, 1893, p. 333-334 (gives full bibliography to 1888). Phacops cristata Hall, Shimer and Shrock, Index fossils N. Amer. 1944, p. 651, pi. 274, figs. 16, 17. Two cranidia and a pygidium from the Devonian Schoharie Formation of New York (no. 84) . Phacops cristata pipa Hall Phacops cristata pipa Hall and Clarke, Pal. New York, vol. 7, 1888, p. 18, pi. 8A, figs. 5-18. An incomplete pygidium in gray limestone, probably of the Middle Devonian Onondaga Formation, from Columbus, Ohio, is probably referable to this subspecies (no. 81). Eophacops catoosaensis, new species Plate 10, figs. 2-4. Three cranidia and a pygidium of a small species of Eophacops, all molds of the under side of the test and all preserved in fine grained sandstone of the Middle Silurian Clinton Group from beside the Chicamauga River, between Catoosa Station and Ringgold, Catoosa County, Georgia, are labeled by General Vogdes "Phacops catoosaensis MSS". They seem never to have been described by him, however, and are therefore described here. Cranidium similar to that of the Middle Silurian Clinton species, E. trisulcatus (Hall), but with only 2 pairs of glabellar furrows, and those poorly developed. Neck furrow well developed. Free cheeks unknown. Thorax unknown. Pygidium with a narrow axis which tapers slowly and extends almost to the rear edge of the shield. Only one rib can be seen, faintly defined, on each of the pleural lobes of the only known pygidium, and only the faintest traces of two segments can be discerned on the axis. Discussion-T\\e cranidium of E. catoosaensis resembles that of E. trisulcatus (Hall) except that it has fewer glabellar furrows. The pygidium of E. catoosaensis resembles that of the Middle Silurian Niagaran species E. handwerki Weller, except that the axis of the Georgia species is narrower and longer. Location of types-The holotype, the largest of the three cranidia, is no. 65a in the Vogdes Collection. The other two cranidia, paratypes, are nos 290 San Diego Society of Natural History 65b and c, and the paratype pygidium (the counterpart of which is also preserved), is 65d. Formation and locality-Clinton Group, Middle Silurian, bank of Chi- camauga River, at the Ringgold end of the bridge on the road from Catoosa Station to Ringgold, Catoosa Station, Georgia. EUROPEAN SPECIES Order Hypoparia Family Harpedidae Harpes venulosus Corda Harpes venulosus Corda. Prodrom einer Mongraphie der bbhmischen Trilo- biten, 1847, p. 164. Harpes venulosus Corda, Barrande, Systeme Silurien du centre de la Boheme, vol.1, 1852, pp. 350-351, pi. 8, figs. 11-15, pi. 9, figs. 11-19. A cephalon, in white limestone from Stage Ff2 (Silurian) at Konieprus, Bohemia (no. 169) . A smaller cephalon, in similar limestone from Bohemia, and therefore probably from the same stage (no. 170). Family Trinucleidae Cryptolithus goldfussi Barrande Trinucleus goldfussi Barrande, Nouveaux trilobites, supplement a la notice preliminaire sur le Systeme Silurien et les trilobites de Boheme, 1846, p. 39. Trinucleus goldfussi Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852, pp. 628-631, pi. 30, figs. 29-40, pi, 35, figs. 30, 31. Cryptolithus goldfussi (Barrande), Raymond, in Zittel-Eastman, Textbook of paleontology, 2nd edit., vol. 1, 1913, text-fig. 1361 (p. 711). An almost complete cephalon and a fragment of a cephalon, in lilac- colored fine-grained sandstone of Stage Dd2 (Ordivician) at Veseta, Bohemia (no. 148). Marrolithus ornatus (Sternberg) Trilobites ornatus Sternberg, Verb, des vaterl. Mus., 1833, p. 53, pi. 2, figs. 2a, b . Trinucleus ornatus (Sternberg) , Burmeister, Organisation der Trilobiten, 1843, pi. 67. Trinucleus ornatus (Sternberg), Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852, pp. 623-628, pi. 29, figs. 1-9, pi. 30, figs. 41-60. Eight cephala in fine grained gray sandstone of Stage Ddl (Ordovician) at Vosek, Bohemia (no. 150). Two cephala in fine grained gray sandstone, presumably of Stage D (Ordovician), from an unstated locality in Bohemia (no. 149). Howell — Vogdes Collection of Trilobites 291 Family Remopleuridae Caphyra radians Barrande Caphyra radians Barrande, Notice preliminaire sur le Systeme Silurien et les trilobites de Boheme, 1846, p. 32. Remopleurides radians Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852, p. 359-361, pi. 43, figs 33-39. A small cephalon in gray shale of Stage Dd5 (Ordovician) at Leskow, Bohemia (no. 167) . Order Opisthoparia Family Olenidae Protopeltura praecursor (Westergard) Peltura praecursor Westergard, Lunds Univ. Arskrift, N. P., Afd. 2, Bd. 5, No. 3, 1909, p. 48. Protopeltura praecursor (Westergard), Sveriges Geologiska Undersokning, Ser. Ca. No. 18, 1922, pp. 171-173, pi. 14, figs. 23-29, 30?, 31, pi. 15, fig. 1 (gives full bibliography to 1922). Two cranidia and a thorax, unnumbered and preserved in fine grained black limestone which, although it is accompanied by no label except one which reads "Dikelocephalus?", is almost certainly from the Upper Cambrian of Norway or Sweden, are believed to be referable to this species, which occurs in the Ctenopyge flagellifer Subzone of the Olenus Beds of Scandinavia. Aulacopleura konincki (Barrande) Arethusa koninckn Barrande, Notice preliminaire sur le systeme Silurien et les trilobites de Boheme, 1846, p. 48. Aulacopleura koninckn (Barrande), Corda, Prodrom einer Monographie der bbhmischen Trilobiten, 1847, p. 84, pi. 5, fig. 48. Aulacopleura angusticeps Corda, Prodrom einer Monographie der bbhmischen Trilobiten, 1847, p. 85. Arethusina konincki (Barrande), Systeme Silurien du centre de la Boheme, vol. 1, 1852, pp. 495-497, pi. 15, figs. 1-21. An excellent entire test and a cranidium in gray shale of Stage Ee2 (Silurian) from Lodenice, Bohemia (no. 168). Family Asaphidae Asaphus expansus (Linne) Entomolithus paradoxus expansus Linne Syst. Nat., 12th ed., vol. 3, p. 160. Entomostracites expansus (Linne,) Wahlenberg, Petrificata Telluris Svecanae, 1818, pp. 25-27. 292 San Diego Society of Natural History Asaphus expansus (Linne), Dalman, Vet. Acad. Handl., 1826, p. 241, pi. 3, fig. 3. Asaphus expansus (Linne), Angelin, Palaeontologia Scandinavica, 1878, pp. 52-53, pi. 28, fig. 1. Asaphus expansus (Linne), Raymond, in Zittel-Eastman, Text-book of paleontology, 2nd edit., vol. 1, 1913, fig. 1384 (p. 719). An excellent entire test from the Ordovician Orthoceras Limestone of Husbyfjol, Ostrogothia, Sweden (No. 177), and an equally good enrolled test (no. 98) from the Ordovician of "Gothland" (probably Ostrogothia), Sweden. Also a cast (no. 75) of an entire test that is very possibly an example of this species, although it bears no label. There are, however, other European species of Asaphus that are very similar to A. expansus, and the details of the cast are not clear enough to permit of certain identification. The specimen form Husbyfjol, Ostrogothia, is from Dalman's collection and must be a homeotype. Nileus armadillo (Dalman) Asaphus (Nileus) armadillo Dalman, Vet. Acad. Handl., 1826, p. 246, vol. 4, figs. 3a-e. Nileus armadillo (Dalman), Angelin, Palaeontologia Scandinavica, 1878, p. 19, pi. 16, figs. 5a-c. A good enrolled test from the Ordovician Orthoceras Limestone of Ostrogothia, Sweden (no. 107). This specimen is also from Dalman's collection, and is therefore presumably a homeotype. Family Illaenidae Wossekia katzeri (Barrande) Illaenus katzeri Barrande, Systeme Silurien du centre de la Boheme, suppl. to vol. 1, 1872, pp. 72-73, pi. 5, figs. 28-37, pi. 6, figs. 1-4, pi. 14, fig. 36. Wossekia katzeri (Barrande), Raymond, Bull. Mus. Comp. Zool., vol. 60, 1916, p. 12. A good cranidium from Stage D. (Ordovician) of Bohemia (no. 94). Warburg * has questioned the validity of the genus Wossekia. She thinks the species, katzeri, should perhaps be left in the genus, Illaenus. Illaenus esmarki (Schlotheim) Trilobites (Asaphus) esmarkn Schlotheim, Isis, 1826, p. 315. * Warburg, Elsa, The trilobites of the Leptaena Linestone in Dalarne with a discussion of the zoological position and the classification of the Trilobita, Bull. Geol. Inst. Univ. Upsala, vol. 17, 1925, pp. 99-100 (footnote). Howell — Vogdes Collection of Trilobites 293 Illaenus esmarkt'i (Schlotheim), Holm, Kongl. Svenska Vet. Akad. Handl., vol. 7, 1883, p. 55, pi. 3, figs. 1-10, pi. 6, fig. 8. Illaenus esmarkU (Schlotheim) , Holm, Mem. Acad. Imp. Sci. St.-Petersbourg, 7th ser., vol. 33, no. 8, 1886, pp. 47-54, pi. 1, figs. 1-6. An entire test from the Ordovician Orthoceras Limestone of Husbyfjol, Ostrogothia, Sweden (no. 106), a tophomeotype from Dalman's collection. Family Scutellidae Scutellum umbellifer (Beyrich) Bronteus umbellijer Beyrich, iJber einige bohmische Trilobitcn, 1845, p. 35, figs. 12, 13. Bronteus umbellijer (Beyrich), Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852, pp. 879-882, pi. 44, figs. 13-24, pi. 48, figs 28-30 (gives full bibliography to 1847) . A pygidium in black Silurian limestone, presumably of Stage F, from Bohemia (no. 45). Scutellum partschi (Barrande) Plate 10, figs. 5, 6. Bronteus partschi Barrande, Notice preliminaire sur le systeme Silurien et les trilobites de Boheme, 1846, p. 60. Bronteus partschi Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852, pp. 870-873, pi. 46, figs. 19-31 (gives full bibliography to 1847). A cranidium and a hypostome (no. 11), in black limestone of Stage Ec2 (Silurian) at Lochkow, Bohemia. As Barrande's figures of the cranidium and hypostome of this species do not agree exactly with our specimens and may be a little inaccurate, illustrations of our specimens are presented here. Family Proetidae Proetus concinnus (Dalman) Calymene concinna Dalman, Vet. Acad. Handl., 1826, p. 231, pi. 1, figs. 5a-c. Proetus concinnus (Dalman), Loven, Vet. Acad. Forhandl, 1845, p. 49, pi. 1, figs. 2a, b. Proetus concinnus (Dalman), Angelin, Palaeontologia Scandinavica, 1878, p. 21, pi 17, figs. 5-5b. Forbesia concinna (Dalman), Anglein, Palaeontologia Scandinavica, 1878, p. 22, pi. 17, figs. 5-5b. A single complete enrolled young test, which was part of a lot of 5 enrolled tests, the other four of which are believed to be young tests of Cyphaspts elegantula (all 5 were labled "Proetus elegantulus") , is believed to 294 San Diego Society of Natural History be a young shield of P. conannus, as its surface is smooth and the long glabella has 3 pairs of furrows and reaches forward almost to the rim (no. 38). The specimen is from Silurian limestone on the island of Gothland, Sweden. Proetus bohemicus Corda Proetus bohemicus Corda, Prodrom einer Monographie der bohmischen Trilobiten, 1847, p. 75, pi. 4, fig. 45. Proetus bohemicus Corda, Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852, pp. 452-454, pi. 16, figs. 1-15 (gives full bibliography to 1847). Proetus bohemicus Corda, Raymond, in Zittel-Eastman, Text-book of paleon- tology, 2nd edit., vol. 1, 1913, text-fig. 1389 (p. 72^). A cephalon in gray limestone, presumably of Stage F (Devonian), from an unrecorded locality in Bohemia (no. 79). Also a cranidium in red Lower Devonian linestone of Stage F2 (Hercynian) from Greifenstein, Hesse (no. 43). Proetus orbitatus (Barrande) Trilobites orbitatus Barrande, Notice preliminaire sur le systeme Silurien et les trilobites de Boheme, 1846, p. 78. Proetus orbitatus Barrande, Systeme Silurien du centre de la Boheme, vol. 1. 1852, pp. 444-445, pi. 15, figs. 29-32, pi. 16, figs. 16, 17, pi. 27, fig. 22 (gives full bibliography to 1847) . Two cranidia, a free cheek, and nine pygidia, in red Lower Devonian limestone of Stage F2 (Hercynian) from Greifenstein, Hesse, appear to be referable to this species (nos. 43 and 47). Two other cranidia, a free cheek, and 5 pygidia, in gray limestone of the same formation at Bicken. Germany (no. 42), appear also to be referable to this species. Proetus myops Barrande Proetus myops Barrande, Notice preliminaire sur le systeme Silurien et les trilobites de Boheme, 1846, p. 74. Proetus myops Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852, pp. 422-443, pi. 15. figs. 20-22 (gives full bibliography to 1847.) A free cheek in red Lower Devonian limestone of Stage F2 (Hercynian) from Greifenstein, Hesse (nos. 43), and a free cheek and pygidium in gray Lower Devonian limestone of the same formation at Bicken, Germany (nos. 42 and 47) seem to be referable to this species. Proetus eremita Barrande Proetus eremita Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852, p. 462, pi. 17, figs. 9, 10. Howell — Vogdes Collection of Trilobites 295 A pygidium in gray Lower Devonian limestone of Stage F2 (Hercynian) from Bicken, Germany (no. 47) . Proetus venustus Barrande Proetus venustus Barrande, Notice preliminaire sur le Systeme Silurien et les les trilobites de Boheme, 1846, p. 64. Proetus lejurus Corda, Prodrom einer Monographie der bohmischen Trilobiten, 1847, p. 75. Proetus venustus Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852, pp. 467-468, pi. 17, figs. 1-6. A small entire test, in gray limestone, probably of Stage E (Silurian) in Bohemia, bears no label or number, but agrees so exactly with Barrande's figures of this species that there can be little doubt about its identity or horizon and general locality. Cyphaspis elegantula (Loven) Proetus elgantulus Loven, Vet. Acad. Forhandl., 1845, p. 51, pi. 1 figs. 4a, b. Goniopleura elegantula (Loven) , Angelin, Palaeontologia Scandinavica, 1878, p. 23, pi. 17, fig. 7. Four enrolled young tests from the Silurian of Gothland, Sweden, have the tubercles, wide brim, and almost smooth pygidium of this species (no. 38). Cyphaspis megalops (McCoy) Harpes (?) megalops M'Coy, Synopsis of Silurian fossils of Ireland, 1846, pp. 54-55, pi. 4, fig. 5. An entire test in gray limestone of the Silurian Wenlock Formation of Dudley, England (no. 158). Dechenella verneuili (Barrande) Phillipsia verneuili Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852, p. 478. Dechenella verneuili (Barrande), Kayser, Zeit. Deutsch. Geol. Ges., vol. 32, 1880, pp. 705-706, pi. 37, figs. 1-5. A pygidium in gray limestone of the Eifel Formation (Devonian) of the Eifel, Germany, labeled as being this species (no. 22). It has only about 12 segments in the axis and about 8 ribs on the pleural lobes. This is fewer segments and fewer ribs than the average for this species; but Kayser's figures indicate that the number is less in some pygidia than in others. Kayser records the species from the Stringocephalus Limestone of the Eifel. 296 San Diego Society of Natural History Family Odontopleuridae Acidaspis crenata (Emmrich) Odontopleura crenata Emmrich, Zur Naturgeschichte der Trilobiten, 1844, p. 17. Ceraiirus cerenatus (Emmrich), Loven, Vet. Acad. Fbrhandl., 1845, p. 47, pi. 1, fig. 1. Acidaspis crenatus (Emmrich), Angelin, Palaeontologia Scandinavica, 1878, p. 34, pi. 21, fig. 6. Two cephala from the Silurian Gothland Limestone of the island of Gothland, Sweden (no. 110). Order Proparia Family Encrinuridae Encrinurus variolaris (Bongniart) Calymene? varwlarts Brongniart, Histoire naturelle des crustaces fossiles, 1822, pp. 14, 15, pi. 1, figs. 3A-C. Calymene variolaris Brongniart, Murchison, Silurian System, 1839, p. 655, pi. 14, fig. 1. Cybele variolaris (Brongniart), Fletcher, Quart. Jour. Geol. Soc. London, vol. 6, 1850, pp. 404-405, pi. 32, figs. 6-10 (gives full bibliography to 1850). Encrinurus variolaris (Brongniart), Murchison, Suluria, 3rd edit., 1859, text- fig. 64 (6), p. 261, pi. 18, fig. 9. An enrolled test from the Silurian Wenlock Formation of Dudley, England (13). Also an unnumbered pygidium, labeled "Dudley, England?", which is probably referable to this species. Encrinurus punctatus (Briinnich) Entomostracites punctatus Wahlenberg, Petrificata telluris svecanae, 1818, pp. 32-33, pi. 2, fig. 1. Cybele punctata ( Whalenberg) , Dalman, iJber die Palaeaden, 1828, p. 40, pi. 2, figs. 2a, b. Calymene punctata (Wahlenberg), Murchison, Silurian System, 1839, pi. 23, figs. 8a, b. Cybele punctata (Wahlenberg), Fletcher, Quart. Jour. Geol. Soc. London, vol. 6, 1850, pp. 403-404, pi. 32, figs. 1-5 (gives full bibliography to 1850). Howell — Vogdes Collection of Trilobites 297 Encrinurus punctatus (Wahlenberg) , Salter, Cat. Cambrian and Silurian fossils in Geol. Mus. Univ. Cambridge, 1873, p. 131, text-fig. (p. 130). Encrinurus punctatus (Wahlenberg), Raymond, in Zittel-Eastmen, Textbook of Paleontology, 2nd edit., vol. 1, 1913, text-fig. 1399 (p. 724). A cranidium with thorax attached and a pygidium in gray Silurian limestone from the island of Gothland, Sweden (no. 14). A pygidium (no. 15) preserved in fine-grained, red-weathering, buff sandstone, labeled as having come from Dudley, England, and therefore presumably of Silurian age, agrees with this species except that there is no medial spine on the rear end. The spine may have been removed by weathering, for the rear end of the shield is exposed. The rock in which this specimen is preserved is not that of the Wenlock Limestone of Dudley, and the locality indicated on the label may be wrong. The pygidium might well be that of the North American species, E. ornatus (Hall and Whitfield) which is very similar to E. punctatus, but lacks the spine on the end of the shield. Encrinurus bohemicus (Barrande) Cromus bohemicus Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852, p. 828, pi. 43, figs. 15-17. Encrinurus bohemicus (Barrande), Raymond, in Zittel-Eastmen, Textbook of paleontology, 2nd edit., vol. 1, 1913, text-fig. 1400 (p. 724). A pygidium in gray limestone of Stage Ee2 (Silurian) from Lochkow, Bohemia (no. 161). Encrinurus beaumonti (Barrande) Calymene? beaumonti Barrande, Notice preliminaire sur le Systeme Silurien et les trolobites de Boheme, 1846, p. 52. Amphion beaumonti (Barrande), Corda, Prodrom einer Monographie der bohmischen Trilobiten, 1847, p. 92. Cromus beaumonti (Barrande), Barrande, Systeme silurien du centre de la Boheme, vol. 1, 1852, pp. 826-828, pi. 43, figs. 6-14. A cranidium in gray Silurian limestone, presumable of Stage E. from Bohemia (no. 159). Family Calymenidae Calymene declinata Corda Calymene declinata Corda, Prodrom einer Monographie der bohmischen Trilobiten, 1847, p. 87. Calymene declinata (Corda), Barrande, Systeme silurien du centre de la Boheme, vol. 1, 1852, pp. 570-571, pi. 43, figs. 53-58. An incomplete cranidium preserved in a fine grained black sandstone is believed to be referable to this species (no. 159). It is probably from the 298 San Diego Society of Natural History Ordovician Stage D of Bohemia, since, although it bears no label, it was in a box with another specimen from Bohemia. Calymene blumenbachi Brongniart Calymene blumenbachtt Brongniart, Histoire naturelle des crustaces fossiles, 1822, p. 11, pi. 1, figs, lA-D. Calymene blumenbachi Brongniart, Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852 pp. 566-567, pi. 19, fig. 10, pi. 43, figs. 46-48 (gives full bibliography to 1850) . Calymene blumenbachi Brongniart, Salter (pars), Monograph British trilobites, 1863, pp. 93-95, pi. 8, fig. 7-16, pi. 9, figs. 1, 2. A good entire test from the Silurian Wenlock Limestone of Dudley, England (no. 125). Also two excellent adult, and two equally good young, enrolled entire tests from the Silurian of the island of Gothland, Sweden (no. 124). Family Cheiruridae Cheirurus sternbergi (Boeck) Trilobites sternbergi Boeck, Notizer til Laeren om Trilobiterne, Magazin for Naturwidenskaberne, vol. 1, 1827, p. 1. Cheirurus sternbergi (Boeck), Beyrich, iJber einige bohmischen Trilobiten, 1845, p. 15, fig. 4. Cheirurus sternbergi (Boeck), Barrande, Systeme Silurien du centre de la Boheme, vol. 1, 1852, pp. 795-798, pi. 41, figs. 29-39 (gives full biblio- graphy to 1847). A cranidium and a pygidium in gray limestone of Stage Ff2 (Devonian) at Koneprusy, Bohemia (no. 166) . Sphaerexochus mirus Beyrich Sphaerexochus mirus Beyrich, iJber einige bohmischen Trilobiten, 1845, p. 21. Sphaerexochus mirus Beyrich, Barrande, Systeme Silurien du centre de la Bohame, vol. 1, 1852, pp. 808-810, pi. 42, figs. 16-23a (gives full bibliography to 1847) . An excellent cranidium and an equally good pygidium in gray limestone of Stage Ee2 (Silurian) at Listice, Bohemia (no. 165). Family Dalmanitidae Odontochile hausmanni (Brongniart) Asaphus hausmanni Brongniart, Histoire naturelle des crustaces fossiles 1822, pp. 21-22, pi. 2, figs. 3 A, B. Odontochile applanata Corda, Prodrom einer Monographie der Bohmischen Trilobiten, 1847, p. 93. Howell — Vogdes Collection of Trilobites 299 Dalmanta hausmanni (Brongniart) , Barrande, Systeme Silurien . AchatclLi i\nlc\t (Meek). Cephalon. X 4. Upper Ordovician. Cin- cinnati, Ohio. San Diego Soc. Nat. Hist. Vogdes Coll.. no. 64. Pl.\TE 9 Fig. 1. Phacops ratu Green. X 2. Hamilton Group. Middle Devonian. London. Ontario. San Diego Soc. Hist. Vogdes Coll., no. 86. Fig. 2. Phiicops Cciai/v>fK/ Hall. X Uj- Hamilton Group, Middle Devonian, Capon Springs, West X'irginia. San Diego Soc. Nat. Hist. X'ogdes Coll., no. 82.^ Howell — Vogdls Colh;ction oi- Trilobitls 313 Fig. 3. Phacops cacctponci Hall Xll/'2. Same specimen as figure 2. Fig. 4. Phacops cacapona Hall. X 11 j. Same specimen as figures 2 and 3. Fig. 5. Phacops hudsonica Hall. X 2. Lower Devonian, western Tennessee. San Diego Soc. Nat. Hist. Vogdes Coll., no. 88A. Plati- 10 Fig. 1. Phacops hudsonica Hall. X 2. Lower Devonian, western Tennessee. San Diego Soc. Nat. Hist. Vogdes Coll., no. 88A. Fig. 2. Eophacops catoosaensis n. sp. Holotype. Cranidium. X 5. Clinton Group, Middle Silurian, near Ringgold, Georgia. San Diego Soc. Nat. Hist. Vogdes Coll., no. 65a. Fig. 3. Eophacops catoosaensis, n. sp. Paratype cranidium. X 5. Clinton Group, Middle Silurian, near Ringgold. Georgia. San Diego Soc. Nat. Hist. Vogdes Coll.. no. 65b. Fig. 4. Eophacops catoosaensis, n. sp. Paratype pygidium. X 5. Clinton Group, Middle Silurian, near Ringgold, Georgia. San Diego Soc. Nat. Hist. Vogdes Coll., no. 65d. Fig. 5. Scutellum partschi (Barrande). Cranidium. X 3. Stage Ec2, Silurian, Lochkow, Bohemia. San Diego Soc. Nat. Hist. Vogdes Coll., no. 11 A. Fig. 6. Scutellum partschi (Barrande). Hypostome. X 3. Stage Ec2, Sil- urian, Lochkow, Bohemia. San Diego Soc. Nat. Hist. Vogdes Coll., no. IIB. Fig. 7. Scutellum bownmgense (Etheridge and Mitchell). Plesiotype pygidium. X 11/^. Bowning Series, Upper Silurian, Bowning Creek, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 55. Fig. 8. Scutellum jenkinsi (Etheridge and Mitchell). Cranidium. X 2. Bowning Series, Upper Silurian, Bowning Creek, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 21. Plate 11 Fig. 1. Scutellum jenkinsi (Etheridge and Mitchell). Cranidium. X 4. Bowning Series, Upper Silurian, Bowning Creek, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 54. Fig. 2. Scutellum jenkinsi (Etheridge and Mitchell). Young test. X 21/2- Bowning Series, Upper Silurian, Bowning Creek, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 52. 314 San Diego Society of Natural History Fig. 3. Scutellum jenkinsi (Etheridge and Mitchell). Pygidium. XU/j. Bowning Series, Upper Silurian, Bowning Creek, New South Wales. San Diego Sec. Nat. Hist. Vogdes Coll., no. 46. Fig. 4. Scutellum longtspinifex (Mitchell). Cranidium. X 3. Bowning Series, Upper Silurian, Bowning, New South Wales, San Diego Soc. Nat. Hist. Vogdes Coll., no. 18. Fig. 5. Proetus rattei Etheridge and Mitchell. X 3. Bowning Series, Upper Silurian, Bowning, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 23. Plate 12 Fig. 1. Cyphaspis yassensis Etheridge and Mitchell. Cranidium. X 6. Bowning Series, Upper Silurian, Belle Vale, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 172A. Fig. 2. Cyphaspis yassensis Etheridge and Mitchell. Pygidium. X 6. Bown- ing Series, Upper Silurian, Belle Vale, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 172B. Fig. 3. Odontopleura rattei Etheridge and Mitchell. X 3. Bowning Series, Upper Silurian, Bowning Village, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 113. Fig. 4. Odontopleura jenkinsi Etheridge and Mitchell. X 3. Bowning Serie.i, Upper Silurian, Bowning Villiage, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 171. Fig. 5. Phacops serratus Foerste. X 3. Bowning Series, Upper Silurian, Bowning, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 90. Plate 13 Fig. 1. Ceratocephala longispina (Mitchell). Cranidium. X 2. Bowning Series, Upper Silurian, Bowning Village, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 108. Fig. 2. Ceratocephala longispina (Mitchell). Pygidium. X 3. Bowning Village, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 112. Fig. 3. Phacops crossleii Etheridge and Mitchell. Cephalon. X 3. Browning Series, Upper Silurian, Bowning Village, New South Wales. San Diego Soc. Nat. Hist. Vodges Coll., no. 91a. Fig. 4. Phacops crossleii Etheridge and Mitchell. Thorax and pygidium. X 3. Bowning Series, Upper Silurian, Bowning Village, New South Wales. San Diego Soc. Nat. Hist. Vogdes Coll., no. 91b. Plate 1 Plate 2 ^'^ ^^ ^^as^. Plate 3 THE AUXILIARY JAirKl^SSlUJSS DEKIVLI) PYGIDIUM OF ISUTILCS* PLATVCEPHAL'iJS, Stoke>. KUM THE i'hc geiieiul for, a of thispait in the .specie; is subtiigonal; the axis lias an obtuse teriuiiiatiou, and faintly marked witn eight arlic- uhitions. The dorsal sliell, covering the entire pygidium.is rtncly punctate. Between the outer and the inner siitll. there is a limited space, whichbeconies greater in the region bord< red by the concaved limb, surrounditig tie outer margin of the pygidium. ITie dorsal shell i>; only align tly concave over the limb ; but the second covering of t lis part is much more so, owing to the large mus- cular space w/ilch lies between the two inenibraneous shells. The limb is marked o.i the second cover by a sei i( s df bands, they com- mence on the anterior lateral margin of the pygidium and along the inner and outer nii/gins of the limb. The inner series consist of ieven terrao^d strip t., cj.u nev ing on the knee line and those which arise on the inner luiigiu of the limb. Tlie second, or outer series are entirely of a d.Herent c.iaracter; they commence in the same manner as the first series, and are increased by additional stripes aris- ing along the outer margin of tlie liinb; but this series consist of irregular ribboned stripes, which are covered with an irregular lines of minute nofk's. We have made a traiisvi-i-sc section of flic pygidium of a young specimen, which shows suuin interesting points with regard to the ii.ternal structure. The specimen has a deep internal plate, shown at fig. h. connected with the lower membraneous shell of the dorsal sur- face by a median band represented at fig. k, and nine supporting a,-ches— see fig. g, f, <"• Tlie central part of the internul plate is some what tliicken(!d, and connected .with the ventral side of the trilobite. Unfortunately our specimen only exhibits a portion of this side. The internal plate was probably connected with the second dorsal shell; such an attachment is indicated by its upward curvature. The dorsal futrows are shown at a a, and the limits of the limb, which sur- round the pygidium, at d d. In the above figure. Froip Prof. Holmes Cabinet. 2 Plate 4 Plate 5 ^-H1 Plate 6 '*^' %. Plate 7 Plate 8 Plate 9 ^ y / I Plate 10 Plate 11 ^ ^.*J. Plate 12 -^] ^ Plate 13 !^US. Zm?. Z80L UBRAXY APR 3 0 1953 TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 12, pp. 329-344, plates 24-25 UMiVEHSlTY TWO CALIFORNIAN MOUNTAIN SNAILS OF THE GENUS HELMINTHOGLYPTA— A PROBLEM IN THE RELATIONSHIP OF SPECIES BY S. Stillman Berry Redlafids, California SAN DIEGO, CALIFORNIA Printed for the Society April 17, 1953 COMMITTEE ON PUBLICATION Joshua L. Baily Carl L. Hubbs LaurencI' M. Klauber Charles C. Haines APR 3 0 1953 TWO CALIFORNIAN MOUNTAIN SNAILS OF THE GENUS HELMINTHOGLYPTA— A PROBLEM IN THE RELATIONSHIP OF SPECIES BY S. Stillman Berry Redlands, California comparisons, and dissections The field work, moulded into the present paper are portions of which have been a much more extensive undertaking dating back a number of years. As there has never arisen an opportunity to carry out the additional field explorations implicit in rounding out the original plan, this fragment of the study as thus far accomplished has been selected for publication as more or less unitary and complete in itself, while in it are involved several questions the answers to which must be of concern to all students of the more funda- mental problems of snail anatomy, the mechanics of speciation, and the evolution of a natural taxonomy in the group. As a groundwork for the consideration of these questions it first becomes necessary to investigate and discuss the anatomy of one of the established species which inhabits an isolated mountain top (Palomar) in southern California, and then with this in mind to delve into the characteristics of an hitherto unnamed species inhabiting a neighboring mountain (Hot Springs) which is iso- lated from the first by the interposition of the small but deep upper San Luis Rey Valley and a lower life zone. 1) Helminthoglypta lowei (Bartsch) PI. 24, figs. 1-3, text fig. 1. 1918. Epiphragmopbova ciiyainacensis lowei Bartsch, — Proc. U.S. Nat. Mus.,'54: 523, pi. 83, figs. 1-3. 1939. Helminthoglypta ciiyamacensis lowei Pilsbry, — Land Moll. N. Amer., 1 (1): 147, text fig. 74 (after Bartsch). 1940. Helminthoglypta ciiyamacensis lowei Gregg, — Naut. Material Studied: 1 Elev. 5,000 ft.. Upper Pauma Cr., Palomar Mt. 8 Observatory road, 3-4 mi. E. of Crestline, Palomar Mt. 11 Elev. 5,000 ft., near sawmill E. of Palomar P.O., S. ridge of Palomar Mt. iMr. Lowe infoinied the writer in 1929 that his exact type-locality was upper Pauma Creek, near the junction of Doane Creek, about three-fourths of a mile north of the highway of that time. H. N. Lowe E. P. & E. Chace M. S. S. Berry & W. G. Craig May 1919 April 1928 54 (1): 33. topotype' mostly juv. 2 living adults 332 San Diego Sociftv of Natural History (^ -sy.slenr. The vas Llefcrens is short, looping under the right ocular retractor, and forms a kink accompanied by a shght cnhirgement as it enters the epiphallus. The epiphallus is long and slender, rather evenly cylindrical until a point somewhat past the insertion of the retractor, but its final portion steadily diminishing in caliber except for a small but distinct oval expansion just before it enters the penis by way of a short and very narrowly restricted neck. The epiphallic caecum is very long and slender, longer than the penis, although not as long as the epiphallus. The large penis is elongate and sausage-shaped, abruptly expanding from the epiphallus at its proximal end and opening into the basal part of the atrium at its distal end ; it is about as long as the combined atrium and vagina, but much shorter than either the epiphallus or its caecum; its inner tube is very thick-walled and folded into its very narrow lumen ; the outer tube or sheath is closely applied and more or less adherent to the inner tube, very thin-walled, membranous, yet appearing in a small excised fragment of preparation 6548b as though composed of two or three layers. Retractor muscle slender and inserted quite high up (about at the distal third) on the epiphallus. ^-sysle))/: The vagina is comparatively slender, quite straightly aligned, and rather shorter than the penis. The spermatotheca is small, with a long duct, and is provided with an exceedingly long convoluted Bhrrv -Two Californian Mountain Sn m' ■; ^33 diverticulum of considerably greater diameter than is the duct at their point of junction and as great as or even slightly greater than that of the long common duct which does not effect juncture with the vagina until nearly or quite opposite the apex of the dart-sac. The atrium is large, saccular, thick-walled, nearly as long as the vagina, which enters it on one side well above the middle. The dart-sac is large, rotund, perched upon a pouch-like atrial sac or apical projection of the atrium a little larger than itself, and divided therefrom by a sharp constriction partly subtended in the specimen figured^ by a small collar-like pad which appears to be con- tinuous with the stalk-like common duct (a little longer than the dart-sac) of the two very large, moderately swollen, elongate-oval mucus-glands; the terminal appendices of the mucus-glands are not very extensive, but are fairly long, looped closely back on themselves, and furnished with copious enveloping membranous extensions. Couimeuts and Comparisons: Several of the anatomical features above described are remarkable among the species of the genus Helniin- thoglypta so far as these have been made known, — (1) the long saccular atrium, with (2) the vagina entering high on one side; (3) the extrem.e constriction of the epiphallus before it enters (4) the abruptly enlarged sausage-shaped penis; and (5) the small node-like inflation of the epi- phallus just precedent to its final narrowing. Less peculiar, but also worth noting, are the extremely long redundant diverticulum, the large rotund dart-sac, the long mucus-glands, and the long epiphallic caecum. Many of these features are prevised in Pilsbry's hitherto rather enigmatic second figure of the reproductive complex in his original H. ciiyamacens'is (1939: 146, fig. 73), and it was certainly because of this that Gregg (1940: 33) so promptly identified the Pilsbryan snail with H. lowei. That a relation- ship is demonstrated can hardly be denied, but that this is tantamount to actual identity is not at all borne out by my dissections. The similarity is a general one only, and a comparison of mjy Fig. 1 with that of Pilsbry at once reveals important and, I believe, significant details to which it does not extend. On the cf side my preparation reveals (1) a more robust penis; and (2) an epiphallus which is extremely long proximad to the insertion of the retractor (about twice as long as shown in Pilsbry's figure). On the $ side we find (1) nothing like the long pre-atrium shown by Pilsbry; (2) the vagina entering the atrial sac about a quarter of the way down instead of immediately under the dart-sac; (3) the vagina much longer and (4) the' mucus-glands much larger; (5) a much longer common duct of spermatotheca and diverticulum, consequent upon (6) its entrance into the vagina close to the atrium, and (7) the much higher juncture of the diverticulum relative to both spermatothecal duct and oviduct. It is of interest to note that both figures show the very peculiar distal dilatation of the epiphallus. 2) Helminthoglypta thermimontis new species PI. 24, figs. 4-9, PI. 25, Text Figs. 2-4 ^This structure is much more obscure in 654Sb. 334 San Diego Society of Natural History Description: Shell of moderate size and thickness, with a thin whitish or reddish callus internally, low-conic in outline; whorls 5l^ to 5%, convex, the suture distinct; last whorl produced anteriorly, moderately to quite strongly descending back of the lip. Aperture rounded-ovate, some- what flattened above, in profile oblique. Peristome quite distinctly thick- ened and everted, the columellar flare sufficiently developed slightly to indent the widely funicular umbilicus. Umbilicus permeable to apex, con- tained on the average about 6^/2 times in the major diameter of the shell, but this figure varying from but 51^ times in one specimen to as much as 7 or 8 times in two of the smaller shells. Spiral sculpture wanting. Initial half-whorl at first smooth and glassy, then with a few rather coarse concentric wrinkles and a few scat- tered, serially decurrent papillations superimposed; next succeeding turn very minutely wrinkly-papillose beneath a heavier, more scattered, but inconspiciious granulation which thereafter becomes stronger and is well developed over the main portion of the shell on both superior and infe- rior surfaces, the rounded papillae showing a somewhat inconstant ten- dency to arrangement in forward-slanting lines as far as the body-whorl, where they become coarser and more irregular, especially on the upper surface. Periostracum of spire Sayal Brown to Snuff Brown, the base of the shell paling almost to Tawny Olive, the conspicuous supraperipheral band Chestnut-Brown, rather obscurely bordered on each side by a somewhat narrower pale area (Ridgway matchings). Caliper Measurements: Max. Alhi. Diam. Number Diam. Diam. Alt. Umbilicus Whorls 6548^ mm. mm. mm. mm. Paratype 27.0 22.7 15.1 4.3 5 3/4 24.8 20.8 14.2 4.0 5 1/2 24.8 20.2 14.1 3.4 5 1/3 24.6 20.4 14.3 4.0 5 2/3 24.1 19.5 12.3 4.0 5 1/4 23.8 19.8 13.7 3.9 5 1/3 23.2 19.5 13.6 4.1 5 1/4 Paiatype 23.2 19.2 14.0 3.5 5 2/3 23.2 19.3 13.0 3.8 5 1/3 Holotype 22.5 18.4 12.3 3.5 5 1/2 Paratype 22.5 18.6 12.1 3.2 5 1/4 21.8 18.3 13.2 2.G 5 1/2 6547^ cf -system : The vas deferens is short, looping tightly under the right ocular retractor to bind itself and the associated organs close to the base Berry — Two Californian Mountain Snails 335 of the vagina, and enters the epiphallus on the inner side of its curve by a hook-hke kink. The epiphallus is quite evenly cylindrical, a little longer than the penis, bears a moderately long caecum (which is, however, a little shorter than the epiphallus itself), and insensibly enlarges into the penis. The penis too is long, club-shaped, attaining its maximum at about the proximal quarter, whence it tapers gradually to the slender stem-like exit; its inner tube is narrow and distinctly separate from the moderately thick-walled outer tube. The retractor muscle is very long, thin, flat, delicate, and inserted on the epiphallus but slightly proximad to its point of passage into the penis. 9 -system: The distal part of the oviduct is sharply narrowed, the lessening of its convolutions and pari passu its transition into the vagina being so gradual that it is not easy to decide upon any precise demarca- tion. The vagina is rather short, the distance distad to the emergence of the vas but about half as long as the penis. The spermatotheca is of mod- erate size, its duct slender, flattened, and quite long, almost equal in its greatest diamieter to the long convoluted diverticulum which conjoins it with about the same diameter as the long common duct; the latter swells nearly opposite the top of the atrium to form a sharp node-like enlarge- ment continuous with the rather abrupt basal expansion of the vagina. The atrium is a short, thick-walled, asymmetrical sac, most distended at the extreme base where both vagina and penis enter at one side; this sac is roughly rhomboid in form, and surmounted by a very large, oblique, rotund, heavy-walled dart-sac set off by a sharp basal constriction. The mucus-glands are large, elongate, moderately tumid, and provided with the usual convoluted membranous extensions apically; they open into a slender common duct perhaps half as long as the dart-sac. 336 San Diego Society of Natural History Holotype: to be deposited in the collection of the San Diego Mu- seum of Natural History. Paratypes: Cat. No. 6547 Berry Collection; others to be deposited in the collections of the United States National Museum, the Museum of Comparative Zoology of Harvard College, and the private collections of Allyn G. Smith, Dr. Wendell O. Gregg, and Emery P. Chace. Type-Locality: Alt. 5000-5500 ft., gulch on southeast side of Hot Springs Mountain (see PI. 25), Los Coyotes Indian Reservation, San Diego County, California; 11 mature and 7 immature living snails, 8 mature and 6 immature empty shells, S.S. Berry and Willis G. Craig coll., 5 April 1928; others taken in the immediate vicinity by Allyn G. Smith, 17 April 1919. Comments and Comparisons: This fine snail at once suggests its mountain neighbor across the valley to the northwest, H. lowei (Bartsch) of the Palomar Range. In shell characters alone the two species are quite weakly differentiated, H. thermiiiiontis being separable principally because of its wider umbilicus,- lesser number of whorls, and distinctly narrower and less inflated body-whorl. In fact were shells alone available to the student those of these two forms are so closely alike in sculpture and texture as well as in general form that he would be hard put to it to decide whether the two should be divided more than racially, it would take so little for connectent specimens to bridge the small gap between them. The reproductive anatomy, on the contrary, tells a completely differ- ent story. The truly astonishing and unexpected situation revealed here is best brought out by a recapitulation in parallel columns of the more salient features of the two species. ^This character does not seem important when comparison is made with Bartsch's fine figure of his holotype, but his specimen does not seem fully charac- teristic of its race in this particular, being more widely umbilicate than any topo- type or other specimen from Palomar Mountain that I have seen. Berry — Two Californian Mountain Snails 337 Epiphallus Epiphallic caecum Penis Retractor penis Vagina Atrium Uart-sac Mucus-glands H. loivei Very long, about twice the length of the penis; a small dilatation fol- lowed by a sharp con- striction distally. H. theym'imonth Short, about equal in • length to the penis ; no distal dilatation or con- striction. Very slender and much longer than penis. Inner tube thick-walled, folded into its narrow lumen, and completely filling the thin - walled outer envelope. Entire organ large, sausage- shaped, sharply enlarged from epiphallus ; base little constricted. Attached high on epi phallus. Narrow, of moderate length, entering high on Very large, ovate; atrial sac small. Ovate - spherical, moder- ate in size, smaller than atrial sac. Very large, elongate- ovate ; common duct longer than dart-sac. Shorter than penis. Inner tube slender, not filling the relatively thick-walled outer enve- lope. Entire organ club- shaped, gradually enlarg- ing from epiphallus; base stalk-like. Attached near junction of epiphallus with penis. Stout, entering low on atrium close to penis. Short, with wide basal expansion ; atrial sac large. Nearly spherical, very large, about as large as atrial sac. Moderately large, elon- gate ; common duct short- er than dart-sac. The only taxonomic divarication which has thus far been proposed within the genus Helm'nilbngl^pta is that of Pilsbry (1939: 68, 69, 170). On the foundation of a single character, the structure of the penis- wall, he divides the genus into two subgenera. In one subgenus, named by him Charodotes. this organ is solid, with its constituent layers of tissue closely adherent. In the typical subgenus, the outer layer is free so that the sac-like penis appears to be enclosed within an envelope. The actual significance of this difference in structure is not shown nor are any his- tological studies available which would help us in interpreting it. Further- more, so far as it has been traced through the various species, the tndi- adata. calijoiiiiensis. cinosa, and diipeiithoiiarsu groups appear to lie on 338 San Diego Society of Natural History one branch of the divarication and most of the Iraskii, and petricola group of species on the other, whereupon some troublesome disassociations (from the standpoint of groupings based upon other structures) quickly become apparent. For example, H. walkeria)ia, H. aytesiana, and H. pblyclaena do not find them.selvcs closely grouped, but split. On this basis too, though perhaps properly, despite the close similarity of their rough, papillose shells, the cHyaniace)2sis-SQnes affords a world of trouble. In the present instance, even though a careful histologic analysis of the tissues should show that the closely applied, adhering outer membrane of the penis in H. lowei is not quite Charodotean and not fundamentally different from the loose envelope in H. thevniimonth. we are left with all the radical differences evinced in the rest of the anatomy still staring us in the face. Thus, the most remarkable and surprising feature in all this is com- parative, that two snails so closely similar in shell and locale should offer such radical contrast in internal organization. On the other hand the most unique single peculiarity to be noted in H. therm'imonth probably lies in the curiously formed atrium and dart-sac. I know nothing elsewhere in the genus which is closely similar. The specific name chosen is derived from the L. thermae, hot springs, and nwus, mountain, and has reference to the current name of the emi- nence which at present provides its only known habitation. CoiicIhsiou Discovered in occupancy of precisely similar habitats, in the same Life Zone (Transition), well toward the summit of neighboring moun- tains, are two species of snails, so closely similar in all outward appear- ance that a certain amount of experienced meticulousness is required in their discrimination, yet so strikingly different in the anatomy of their reproductive systems that it seems fairly doubtful whether they could suc- cessfully mate, while according to the criterion adopted by the latest monographer of the genus, some might even place them in separate sub- genera. The single character relied upon in the establishment of Charo- dotes Pilsbry 1939 runs athwart so many other features apparently of at least equal import that one cannot avoid a certain skepticism as to whether this group in fact forms a natural segregate or even indeed a useful one. Certainly as at present constituted it quite fails to do so. The structure of the penis-wall has never in any species been subjected to precise his- tologic investigation, for instance by imbedding and sectioning, yet it is so manifest that this structure undergoes wide variation in the respective species and groups that the assumption of an infallible morphologic dichotomy founded upon it may be premature and not fully to be borne out by more intensive study. One must increasingly wonder how far, if at all, it is safe to trust to a single character in basing a distinction as wide as a subgenus. And surely when such division runs across the grain of other known characters within the group it becomes still more needful to be wary. Berry — Two Californian Mountain Snails 339 I have here entered into perhaps redundant detail for the express purpose of bringing into sharper focus a yet deeper question of speciation and distribution which is of the utmost interest, and which now, as Hmned by the outhne of these two species, is squarely within our view. What are their real relationships ? Are they so closely allied to one another that we may refer to them as "geminate species," which, as their shell characters would indicate, have eifected a comparatively recent diver- gence since their separation and isolation upon neighboring mountain tops? Or are their nearly identical shell characters the deceptive conse- quence of a convergent response to closely similar ecological influences, and their actual phylogenetic relationship consequently more remote as is so strongly indicated by the unexpected diversities evident in the repro- ductive anatomy ?i Although I must confess that the key to answering such far-reaching queries is one I have not yet come by, it seems to me that it must lie to a considerable degree implicit in the answer to yet another question — which characters in land-snails are correctly to be regarded as the more conservative and stable, and which as the more plastic, those of the hard shell with its more or less consistent sculptural detail, or those of the soft internal organs? The anatomical differences above described are so numerous, so great, and of so fundamental a character that it seems altogether incredible that they are not deeply significant; yet all the dis- tributional and ecological data at hand seem rather to support the old- fashioned dictum that the shell is the more reliable key (within certain limits of course) to natural affinity. The large number of varyingly iso- lated species and races scattered about southern California, offering shells of such closely similar, occasionally almost invariant type that at times they could hardly be separated successfully if deliberately mixed together, coupled with a veritable eruption of anatomical variation, offers a most tempting field for future research, upon the outcome and interpretation of which our whole understanding of the phylogeny and inter-relation- ships of the members of this important and fascinating genus may one day turn. S/ini>i/ctyy. An hitherto undescribed land-snail, Helniiiithoglypta therni'imont'is n. sp., which inhabits the Transition Zone on Hot Springs Mountain, the highest peak in San Diego County, California, is intensively compared with an isolated neighboring species, H. lowei (Bartsch) which occupies corresponding habitats on neighboring Palomar Mountain, the two ranges being separated by the Upper Sonoran San Luis Rey Valley. Although the shells in the two species exhibit only very minor differentiation, the repro- ductive anatomy proves to be strongly divergent in practically all details. Attention is called to the bearing of the data set forth on the taxonomic value of the subgenus Charodotes Pilsbry 1939, and on the more funda- mental question of the relative evolutionary plasticity of shell characters in land snails as against those of the internal organs. 340 San Diego Society of Natural History Literature Cited Bartscm, p. 1918. Two new land snails of the Ep/pbyj^i^iz/offboni trasku group. Proceedings United States National Museum, 54 (2246) : 523-524, PI. 83, 1918. Gregg, W. O. 1940. Anatomy of Helminthoglypta. Nautilus, 54 (1): 33, 23 July 1940. Pilsbry, H. a. 1939. Land Mollusca of North America (north of Mexico). Mono- graph 3, Academy Natural Sciences Philadelphia, i-xvii, 1-573, i-x, figs. 1-377, 6 December 1939. PLATE 24 Figs. 1-3. Helminthoglypta lowei (Bartsch). Shell of specimen 6548c from Palomar Mountain, California; about natural size. Figs. 4-6. Heh)iiuthoi^l)pta tbevniimoniis n. sp. Shell of holotype from Hot Springs Mountain, San Diego Co., California; about natural size. Figs. 7-9. Helniiuthoglypta theriiiiniontis n. sp. Shell of paratype 6547c from Hot Springs Mountain, San Diego Co., California; about natural size. Fig. 10. Helni'Dilhoglypla iherniniKnUis n. sp. Microphotograph of upper surface of latter portion of last whorl of liolotype, showing papil- lose sculpture; x. Fig. 11. Hehniuthoglypta ihenmnioiil'ts n. sp. Microphotograph of basal surface of latter portion of last whorl of holotype; same scale as preceding. Berry — Two Californian Mountain Snails Plati: 24 # A n 34^ San Diego Society of Natural History PLATE 25 View looking up trail on southeast side of Hot Springs Mountain, San Diego Co., California; the type locality of Helmiiithoglypta therm't- montis is among the trees on the right side of the small canyon shown in the center of the picture. (Photograph by S. S. Berry, "S April, 1928.) llxplauat'ion of Text Figures Fig. 1. Hehji'nilhoglypta lowei (Bartsch). Camera sketch of ante- rior portion of cT 9 system of specimen 6548a. Figs. 2-3. Heliii/ntboglypta tbernumnntis n. sp. Can-^ra outlines of shell of holotype. Fig. 4. Hehnintboglypta therni'iniontis n. sp. Camera sketch of anterior portion of cf 9 system of paratype 6547c; same scale as Fig. 1. Berry — Two Californian Mountain Snails Plate 25 TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 13, pp. 345-364, plates 26-27 MOLLUSKS FROM CLIPPERTON ISLAND (EASTERN PACIFIC) with the DESCRIPTION OF A NEW SPECIES OF GASTROPOD BY Leo George Hertlein and William K. Emerson SAN DIEGO, CALIFORNIA Printed for the Society July 22, 1953 mz. coMP. zooL LISRARY AUG 7 195c ffARVARD BMI'/€RSiTY COMMITTEE ON PUBLICATION Joshua L. Baily Carl L. Hubbs Laurence M. Klauber Charles C. Haines MOLLUSKS FROM CLIPPERTON ISLAND (EASTERN PACIFIC) with the DESCRIPTION OF A NEW SPECIES OF GASTROPOD' By Leo George Hertlein' and William K. Emerson' m^. COMP. ZOO LiBRARY AUG 7 195 HARVARD UHIVCRSITY INTRODUCTION The marine faunas of Clipperton Island are of particular interest to zoogeographers because of the presence of Polynesian elements in an area situated on the boundary between the Panamic and Indo-Pacific Faunal Provinces. Through the courtesy of Dr. Edward L. Hamilton* the writers were privileged to examine a small but significant collection of mollusks representing most of the biological contents of two dredge hauls from 100-200 fathoms off the east slope of Clipperton Island. The material was collected in May, 1952, by the scientific personnel of Expedition "Shuttle" of the United States Navy Electronics Laboratory and represents the first deep water collection from that area. Five species of mollusks, two gastropods, one of which is described as new, three pelecypods, and two species of stony corals (not included in this paper) were represented in the dredged material. The acquisition of this material has prompted the writers to prepare the present report. The authors wish to express their appreciation to Dr. J. W. Durham, Associate Professor, Department of Paleontology of the University of California, for helpful suggestions and for critical reading of the manu- script. Dr. G. D. Hanna, Curator of the Department of Paleontology, California Academy of Sciences, offered suggestions concerning the new. species described in the present paper. Mr. J. R. Slevin, Curator of the Department of Herpetology of the same institution, called our attention to some of the references pertaining to Clipperton Island and furnished a photograph of the island. The conchological collection in the Department ^Contribution from the Department of Paleontology, California Academy of Sciences, and the Museum of Paleontok>gy, University of California. ^Associate Curator, Department of Paleontology, California Academy of Sci- ences. ^Museum of Paleontology, University of California. ^United States Navy Elect.onics Laboratory, Point Loma, San Diego, California. 348 San Diego Society of Natural History of Geology at Stanford University was made available to us through the kindness of Dr. A. Myra Keen. Mr. Ted Dranga of Miami, Florida, fur- nished information concerning the occurrence of Conus ehraeus and sent specimens of Btirsd useful to us during the present study. Mr. Elton L. Puffer, Department of Paleontology, University of California, aided us in verifying certain of the bibliographic references in the University library. Mrs. Carma R. Zimmerman, Librarian, and Mr. Allan R. Ottley, Califor- nia Section Librarian, furnished references to articles concerning Clipper- ton Island in the California State Library at Sacramento, California. Mr. J. A. Gibbs, Jr., Seattle, Washington, gave information concerning the ship "Kinkora" which was wrecked on Clipperton Island. The photo- graphs used to illustrate the specimens of mollusks on the plate were made by Frank L. Rogers. General Remarks Concerning Clipperton Island Clipperton Island is a coral atoll about two miles in diameter, sur- rounded by a fringing reef, which lies in Latitude 10°17'00"N., Long. 109°13'00"W., about 670 miles southwest of Acapulco, Mexico. The general cartographic features are shown on United States Hydrographic Office Chart No. 1680. It is cited in some of the early literature and charts under the names of "La Pasion" and "Medano" or "Los Medanos." Some authorities (Cubas, 1909: 76-77) have suggested that the island, or evidences of it, may have been sighted as early as 1527 by Captain Alvaro de Saavedra Ceron, captain of a Spanish armada sent across the Pacific to the East Indies by Hernando Cortes. Others consider it likely that the island was not discovered before the eighteenth century. It was definitely sighted by Clipperton in 1705. Morrell (1832:219) who an- chored off the island in 1825 believed it to be of volcanic origin. He renriarked on the presence of immense numbers of sea birds, a few land birds, and a few fur seals and sea elephants. A little coarse shrubbery, some coarse grass and a plant resembling sarsasparilla were recorded as occurring on the atoll. Later Belcher (1843:255-257) viewed it from the masthead but, like Clipperton, did not land there. Belcher remarked on the presence of two openings in the atoll, both on the weather side. It appears that the name Clipperton was first applied to this island by Cap- tain Cook. Both Mexico and France have laid claim to this small islet; the latter has received the latest legal approval from an international tribunal to support her claim. The greater part of the atoll is composed of coral limestone which extends from about five to fifteen feet above sea level and varies in width from a few yards to one half mile. Hertlein-Emerson — MoLLiisKS FROM Clipperton Island 349 On the southeastern edge, a single mass of rock (known as Chpper- ton Rock), an altered trachyte'' (Teall, 1898; Elschner, 1913; Lacroix, 1931; Glasser, 1932) containing 54% of silicon dioxide, rises to a height of 62 feet (plate 26). This island furnishes nesting sites for a multitude of sea birds (Edwords, 1906; Gifford, 1913). There have been differences of opinion as to the origin of this atoll. Pease (1865:201) believed that it showed evidences of elevation. He cited the journal of Lieutenant Griswold who landed on the island and who recorded that in places the walls of the high rock mass were en- crusted with coral. Wharton (1898) inclined to the belief that the coral may have formed on the lip of a portion of a volcanic crater and that the highest portion of the island might be what remains of a volcanic plug. Others including W. M. Davis (1928), in one of the most comprehensive modern studies of coral reefs and atolls, considered Clipperton Island to be a "near-atoll" and probably a product of subsidence. A unique feature of the atoll is the considerable depth of the enclosed lagoon, reported to contain brackish water (Anon., 1938), which has been cited as ranging from one foot to fifty-five fathoms. This lone atoll in the eastern Pacific is seldomj visited. Its isolated position in the open sea results in a heavy surf which usually beats upon the shores. Abrupt changes in the weather may give rise to sudden squalls with attendant low visibility. This condition, together with lack of shelter and very poor anchorage, discourages the mariner from spending much time in that vicinity. The "Kinkora" was wrecked there in 1897 (Anon., 1897). Guano and the phosphatic coral rock, some of the latter said by Elschner (1913:92) to contain 78 percent phosphate, led at times to commercial exploitation of these products which were shipped to the mainland. Workers were brought to the island at various times in con- nection with the concession to engage in this industry. Some of the expe- riences of the inhabitants have been described by Perrill (1937) and others (Anon., 1893). Occasionally expeditions engaged in scientific research stopped briefly at Clipperton Island and collected a few natural history specimens. A few specimens from the island reached collectors through other sources. One of the earliest collectors to secure a number of species of mol- lusks on the island was W. H. Ochsner, a member of the expedition of ^Living specimens of Delectopecten zacae, dredged (Exped. "Shuttle") from 110-150 fathoms off the east slope, were attached to pebbles of trachyte. 350 San Diego Society of Natural History the California Academy of Sciences to the Galapagos Islands in 1905- 1906 (see Slevin, 1931:21-22). Eleven species and subspecies assembled by Ochsner were reported by Hertlein (1937). Dall (1910; 1911), Lowe (1933) and Bartsch & Rehder (1938) cited additional species from that locality. The species of Conns were recently included in a monograph by Hanna & Strong (1949) and the species of Cypraea by Ingram (1947; 1951). In view of the few collections that have been made in this area, it seems reasonable to conclude that systematic collecting would greatly enlarge the known fauna. Mollusks from Clipperton Island A list of species of mollusks known to occur on Clipperton Island, compiled from collections and the literature, follows. Those not known to occur along the mainland of the western Americas are indicated by an asterisk (*). Species and subspecies Pelecypoda "Acar species? (young)" Aiioiuid peruvianj d'Orbigny *Chama squamuligera rubropicta Bartsch & Rehder Very similar to C. squamuligera Pilsbry & Lowe, a Panamic species. Ctena clipperlonensis Bartsch & Rehde: Similar to C. cldrtonemis Hertlein & St-ong from Clarion Island, Mex- ico; also C. bella Conrad from the Hawaiian Islands, jide Bartsch & Rehder (1939:14). Delectopecten zacae Hertlein Gastropoda * Bursa granularis (Bolten) Roding^ Synon. -Ranella granijera Lamarck. Faunal Province Source of Record *Clanculus sis, n. sp. (Panocochlea ) clippertonen- Panamic'' CHpperton Island type locality Clipperton Island type locality. Also recorded from Maria Madre Is- land, Mexico, and Hannibal Bank, Panama, by Hertlein & Strong. Panamic Indo-Pacilic Clipperton Island type locality Bartsch & Rehder (1939:16) Here recorded Exped. "Shuttle." Bartsch & Rehder (1939:16) ; also Exped. "Shuttle." Bartsch & Rehder (1939:16) Here recorded, Exped. "Shuttle." Morrison (1949:11) Also in Calif. Acad. Sci. coll. Here recorded, Exped. "Shuttle." ^The Panamic faunal provmce here construed as including tropical west American mollusks from west Mexico to northern Peru. ''Tritonium jabick (Bolten) Roding has line priority and may be the correct name for this species. Iredale (Rec. Australian Mus., Vol. 18, No. 4, June 29, 1931, p. 213) stated: . . . "granijera Lamarck, apparently equivalent to granularis Bolten and jabick Bolten, the latter having priority is here made the type of a new genus Dulcerana." On page 232, "Dulcerana gen. nov.: type Ranella granijera Lamarck." This appears to be distinct from Bursa pustulosa jabick Fischer, from West Africa. HeRTLEIN-EmERSON MOLLUSKS FROM ClIPPERTON ISLAND 351 Conns brunneus Wood Conus ebraetis Linnaeus * Conns ebraeus verm'nulatiis Lamarck Conns gradatns Mawe Conns pnrpurascens regalitatus Dall Conns tiavatus Broderip Cyinat'inm vestitnm Hinds Cymatium vestitntn ins nl are Pils- bry occurs in the Hawaiian Is- lands. Both C. vestitnm and its subspecies closely resemble the Indo-Pacific C. pileare Linnaeus. *Cypraea depressa Gray ^•Cypraea isabella Lamarck Cypraea isabella mexicana Stearns *Cypraea senna Gmelin * Cypraea teres Gmelin *Drupa ricinus Linnaeus *Hipponix fimbriatns Bartsch & Rehder Veiy similar to H. serratus Carpenter, a Panamic species. Hipponix pilosns Deshayes Synon.-H. barbatns Sowerby *Littorina schmitti Bartsch & Rehder Related to L. pintado Wood from the Hawaiian Islands, jide Bartsch & Rehder (1939:10). Panamic Dall ( l'-)10:220) Panamic* and Hertlein ( 1937:3C6) ; Indo-Paciiic Hanna & Strong (1949: 312) Indo-Pacihc Hertlein (1937:306) Panamic Hanna & Strong (1949:279) Panamic Dall (1910:219) Panamic Dall (1910:220, as C. niiliaris Hwass) ; Bartsch & Rehder (1939:16, as C. rooserelti). (See Hanna & Strong, 1949:272). Panamic Hertlein ( 1937:307) Indo-Pacihc Hertlein (1937:307, as C. gillei Jousseaume) Ingram (1947:72; 1951:149). Indo-Pacific Hertlein (1937:307) ; Ingram (1947:73; 1951:150). Panamic Hertlein ( 1937 : 307 ) ; Ingram (1947:73; 1951:152). Indo-Pacific Hertlein (1937:307);. Int^ram (1947:77; 1951:157). Indo-Pacific ' Hertlein (1937:307) ; Ingram (1947:81 ; 1951:162). Indo-Pacific Hertlein (1937:308) Bartsch & Rehder (1939:16, as Dm pa "vicina"') . Clipperton Island Bartsch & Rehder type locality (1939:16). ^Panamic and Ba tsch & Rehder Indo-Pacific (1939:16, as H. barbata Sowerby ) . Clipperton Island Bartsch & Rehder type locality (1939:16). *Mr. Ted Dranga (written communication, January 14, 1953) recently col- lected a live specimen of Conns ebraens on the Pacific coast of Guanacaste Province, Costa Rica. It was taken "in a crevice on an extensive area of hard rock wave bench covered with short sea weed — just the sort of location that would have supported ebraeus in Hawaii. It was an area that is bared by the low tides, and the spot was not far from low water mark." 352 San Diego Society of Natural History *Mjii!liis robillardi Lienard Indo-Pacific '"Mitt a papalis Linnaeus Indo-Pacific ■^Morula ura Bolten (Roding) Indo-Pacific Synon. -Ric/nula morus Lamarck ; RicniuLi nodus Lamarck; Morula papulosa Schumacher. Nassarius catallus Dall Panamic *-Ne)ila plicata Linnaeus Indo-Pacific *Opeas oparanum Pfeiflfer Indo-Pacific A land snail Purpura nuttalli Conrad Californian Almost certainly ballast Valuta ancilla Solander Patagonian Almost certainly ballast Woluta deshayesii Reeva Indo-Pacific The foregoing list which one is a landsnail. Bartsch & Rehder (1939:16). Hertlein (1937:308). Hertlein (1937:308, as Drupa tnorus Lamarck) ; Bartsch & Rehder (1939: 16, as Morula nodus Bory St. Vincent). Here recorded, Exped. "Shuttle." Bartsch & Rehder (1939:16) Bartsch & Rehder (1939:16) Collected by W. H. Ochsner. Label by I. S. Oldroyd states "ballast shell?" Lowe (1933:112) Dall (1911:112); Lowe (1933:112) ; Hertlein (1937:308) is comprised of 31 species and subspecies of In addition one genus {Acar sp.) occurs there, the species of which is unknown. Four species and subspecies are pelecy- pods and 27 are gastropods. Of the latter, one Californian species. Purpura nuttalli Conrad, almost certainly arrived on the island adventi- tiously by ballast in a ship or by other means and it appears quite probable that the same applies to Voh/ta aticUla Solander, a species known to occur in the Straits of Magellan and the neighboring region. This list obviously denotes but a small portion of the total molluscan fauna of the area, but its composition, based upon known occurrences, is extremely significant. A number of the records of molluscan species cited from Clipperton Island are based upon shells lacking the live animal. However, in con- nection with this fact it is important to recall that a number of Indo- Pacific invertebrate species (moUusks and echinoids) have been taken alive in the eastern Pacific. Among these are Q/ioyula madreporarum (mainland; Marie Madre Island, Tres Marias Islands; Socorro Island, Revillagigedo Islands), Conns ebraeus (Costa Rica), Conus tessullatns Born (Synon. — C. edaphus Dall) (Clarion Island, Revillagigedo Is- lands), Hippon/x p'tlosus Deshayes (Synon. — H. barbatus Sowerby) (mainland; Tres Marias Islands; Revillagigedo Islands; Galapagos Is- lands), the echinoids Br/ssus latecarinatus Leske (Gulf of California to Panama), Metalia spatagus Linnaeus (Gulf of California), Ecbinometra oblonga Blainville (Gulf of California; Clarion Island, Revillagigedo Islands; Galapagos Islands) and Loven/a cordiformis A. Agassiz (Santa Barbara, California, to Guayaquil, Ecuador; Hawaii). Twelve species and subspecies (3 pelecypods and 9 gastropods), a little more than a third of the total number, are known to occur on the HI'Rtlkin-Emerson — MoLLUSKS FROM Clipprrton Island 353 mainland of the tropical Americas. Of these, nine represent truly Pa- namic faunal elements, one is closely allied to a species described from Clarion island, and two gastropods, H/ppon/x pilosi/s and Coiius ebvaeus have been found living both on the mainland and in the Indo-Pacific region. One pelecypod and 16 gastropods have not been recorded from the mainland. Three of these gastropods have been recorded as occurring in the Galapagos Islands. Of the four species (1 pelecypod and 3 gas- tropods) known only from Clipperton Island, three have affinities with Panamic and one with Hawaiian forms. H'lpponix pilosns is the only one of the Indo-Pacific moUusks from Clipperton Island which is known to occur as a fossil in the eastern Pacific. None of these species is known to be living in the Caribbean, but one species, Cypraea Isabella or a sub- species of it, has been recorded as a fossil in the Miocene in that region. At least 10 of the 14 typically Indo-Pacific gastropods from Clipperton Island also occur in the Hawaiian Islands. The vast expanse of uninterrupted abyssal marine water between the west American shores and the easternmost Polynesian Islands, the Eastern Pacific Barrier of Ekman (1935:105; 1953:72; Sverdrup, et al.. 1942: 860-861), is generally considered to be the prime obstacle to the distri- bution of littoral marine forms by transoceanic currents. It is an interest- ing fact, however, that the majority (approximately 60%) of the gas- tropods definitely known to comprise a portion of the Clipperton Island fauna are Indo-Pacific forms or have definite affinities with Indo-Pacific species. The great preponderance of gastropods (25, not including 2 believed to have arrived as ballast) as compared to the pelecypods (4 and One genus, the species unknown), the comparatively large number of species of Coi///.( (6), and Cypraea (5), as well as the decided Indo-Pacific affinity of the moUuscan assemblage from this island, all are in harmony with the results of studies made by Hertlein (1937). An analysis of the data afforded by the present list and those pre- sented by Hertlein (1937) on the fauna of the other ofl^ shore islands (Cocos Id., Galapagos Ids., etc.), would seem to indicate that such groups as the Cypraeidae, Conidae, Bursidae and Cymatidae have exceptionally long pelagic larval stages which can tolerate current transport of consid- erable duration. Dispersal may also be accomplished by the attachment of the adult or larval phases to floating objects, to the feet of birds, to nekton, that is, pelagic swimming organisms such as fishes, mammals, etc., and as transport by man. It seems probable that most of the Indo- Pacific gastropod species reached Clipperton in a larval state by way of the Equatorial Counter Current. Whatever the agent of transport, it is significant to note the predominance of gastropods over the pelecypods, the latter with one exception, are all attaching types of Panamic faunal association. Obviously, this suggests that in the mollusks the larval stages of the gastropods afford the most successful means for dispersal of littoral forms across a great expanse of deep water. Hedley (1899:391-417) pointed out that distribution in Polynesian islands indicates that deep 354 San Diego Society of Natural History water is not a barrier to Conns. Cypraea and Mitva. Powell (1933:154- 164) mentioned that triton shells pass through a free swimming stage in which the apical shell is allied to the type known as S'lnus'igera, a type of apex always associated with species of wide distribution. Iredale (1911: 319) noticed its presence in species of the Mitridae, Buccinidae, Nas- sariidae, Thaisidae, Pyrenidae, Coralliophilidae, Turridae, and Terebridae. Ostergaard (1950) found that forty of forty-one species (97 percent) of Hawaiian marine gastropods studied by him which attained the veliger stage were found to have a free swimming stage. The results of his work and that of others indicate that the greatest number of free swimming larvae occur in warm waters, the fewest in cold waters. Inasmuch as the Equatorial Counter Current reaches the American coast, it is difficult to explain why more of the Indo-Pacific species have not been able to become established on the continental shelf. Only a very few of the Polynesian species appear in American coastal waters, the ex- treme eastern range for most of those species being the islands lying far off shore, namely, Cocos Island, Clipperton Island, the Tres Marias Is- lands, Revillagigedo Islands, and the Galapagos Islands. This restrictive distributional barrier is undoubtedly created by a number of complex factors interplaying upon each other, one of which perhaps is the com- petition of the established Panamic fauna. Menard (1953:22) recently reported the discovery of a submarine escarpment trending between Christmas and Clipperton Islands, a dis- tance of approximately 3300 miles. This topographic high is recorded to be 1000 feet in altitude and is bounded in its eastern limits by the Aca- pulco Trench. This submarine ridge is another factor to be taken into account in considering a possible migrational route for Indo-Pacific faunal elements to cross the mid-Pacific deep to Clipperton Island. The deep water of the Acapulco Trench apparently forms a barrier to the establish- ment of many of the Indo-Pacific species (especially shallow water forms) upon the continental shelf. Questions connected with the processes of the formation of Clipper- ton Island, together with its isolated geographic location, the preponder- ance of Indo-Pacific mollusks there, continue to invest this easternmost atoll in the Pacific with a singular attraction, intriguing to students of natural history. Clanculus (Panocochlea) clippertonensis Hertlein & Emerson, new species Plate 27, Figures 19, 20, 22 Description: Shell rather small and consisting of about 5 depressed turbinate whorls, the sutures deeply channeled; nucleus smooth, followed HnRTLEIN-EMHRSON — MOLLUSKS FROM ClIPPERTON ISLAND 355 by about one and one half similar grayish whorls; the succeeding whorl gradually becomes somewhat flattened, colored red and developing a carinate shoulder posterior to which there are 3 or 4 fine spiral threads; on the succeeding whorls there are 4 coarse, nodose, spiral ribs which are nearly equally spaced and between these major ribs there are 3 or 4 fine spiral threads which become coarser on the later whorls, on the last whorl a nodose median riblet occurs in the interspace between the third and fourth rib and below the latter a weak nodose spiral rib occurs below the periphery, this is followed by 2 or 3 smooth, nearly obsolete spiral threads on the base; the base is broadly rounded, smooth and imperforate; whorls crossed by fine oblique lines of growth; aperture obliquely rounded, columella curved, a well developed node is present near the base, and below this a very small node is present on the base of the aper- ture, the interior of the outer lip is smooth; exterior of shell above the periphery red, the nodes white, base pinkish-cream and aperture of a pearly cream color. Dimensions: height, 8.6 mm., of aperture 5.9 mm., maximum diameter 10.5 mm. Holotype: No. 33341, Univ. Calif. Mus. Paleo. Coll. from Loc. A-8635 (U.C.M.P.), east slope of Clipperton Island, Lat. 10°17.5'00"N., Long. 109°11.5'00"W., dredged in 100-200 fathoms by Expedition "Shuttle," U. S. Navy Electronics Laboratory, Point Loma, California, May 14, 1952. This species bears a general resemblance to Clancnl/ts (Paiwcochlea) rubidus Dall,^ originally described from the Gulf of Panama, in 259 fathoms. The species here described as new diff^ers from C. (P.) rubidus in that the miajor spiral ribs are nearly equally spaced whereas in Dall's species, the third rib is separated by a wide interspace from the two rather closely spaced posterior ribs. The outline of the outer lip of the aperture of the new species is evenly rounded in comparison to that of C. (P.) rubidus in which the outline reflects the depressed interspaces which also bear more numerous spiral threads. The node at the base of the columella on the species here described from Clipperton Island is much smaller than the large callus mass shown in the figure of Dall's species. Furthermore, another very small node occurs on the interior of the base of the aperture, but no such node is mentioned on the Panamanian shell. The subgenus Pauocochlea Dall, type Clancuhis {Panocochlea) rubidus Dall, lacks the umbilical opening and denticulate aperture typical of the genus Clanculus Montfort. The type (by original designation) of Clanculus is Clanculus pharaoiiicus Linnaeus originally described as Trochus pharaonicus. ^Clanculus {Panocochlea) rubidus Dall, Bull. Mus. Comp. Zool., Vol. 43, No. 6, p. 346, pi. 8, figs. 3, 4, October, 1908. "U.S.S. 'Albatross,' station 3396, Gulf of Panama, in 259 fathoms, mud, bottom temperature 47°. 4 F. U.S.N. Mus. 122, 954." Hoffstetter (Bol. Inst. Cienc. Nat. (Quito), Ano 1, No. 1, June, 1952, p. 48) recorded the occurrence of "Clanculus cf. rubidus Dall" as a subfossil from Salinas, Ecuador. 356 San Diego Society of Natural History BIBLIOGRAPHY Anonymous 1893. Found them safe. Clipperton's two Crusoes home again. Back aboard the Viking. Morning Call, San Francisco, Vol. LXXIV, No. 153, Tuesday, October 31, 1893, p. 8, col. 3. [Article reports the experiences of two men on Clipperton Island who were employed by the Oceanic Phosphate Com- pany]. 1897. Were stranded on Clipperton. Shipwrecked crew of the Kinkora rescued by the cruiser Comus. The steamer Navarro gone after the men left behind on Island. San Francisco Call, Vol. LXXXII, No. 51, Wednesday, July 21, 1897, p. 8, col. 3. [Mr. J. A. Gibbs, Jr., of Seattle, Washington, fur- nished the following information concerning the "Kinkora." "The Kinkora was a British ship constructed of iron, of 1830 gross tons, 267 feet long, 39 feet of beam and a depth of 23 feet. She was built at Marysport, England, for P. Iredale & Sons in October, 1888. Her builders were Ritson & Com- pany, and her home port was Belfast, Ireland." (Written communication February 14, 1953)]. 1920. French Possessions in Oceania. Handbooks prepared under the direction of the historical section of the foreign office (London), No. 145, pp. 1-67. [Clipperton Island, pp. 8, 41-42]. 1938. Sailing directions for the west coasts of Mexico and Central America. Hydrographic Office No. 84 (Washington, D.C.), Ed. 8, 1937, pp. I-VII, 1-430, 2 charts. [Clipperton Island, pp. 52-53]. 1947. 5 in shipwreck enroute home. San Francisco Examiner, Vol. 187, No. 1, p. 14, July 1, 1947. Bartsch, P., and Rehder, H. A. 1939. MoUusks collected on the presidential cruise of 1938. Smith- son. Miscell. Coll., Vol. 98, No. 10, (Publ. 3535), pp. 1-18, pis. 1-5, June 13, 1939. [See especially list of species from Clipperton Island p. 16. A popular account of the itinerary of this cruise entitled "Third Presidential Cruise of the U.S.S. Houston," Captain G. N. Barker, U.S.N., Commanding, was printed on board the U.S.S. Houston. Pp. 1-75, illustrated, 1938. (Clipperton Island, pp. 20-21)]. Belcher, E. 1843. Narrative of a Voyage round the World, performed in Her Majesty's Ship Sulphur during the years 1836-1842, . . . Vol. 2, pp. I-VI, 1-474, 11 engravings, 9 vignettes. [Clip- perton Island, pp. 255-257, 364]. HhRTLF.IN-EmERSON MOLLUSKS FROM ClIPPFRTON ISLAND 357 CUBAS, A. G. 1909. Documento No. 29 [in] Isia de la Pasion llemada de Clip- perton (Secretaria de rclaciones de Mexico. Publ. Oficial. Mexico City), pp. 71-83, 2 maps. [The entire volume is composed of pages Mil, 1-96. (Documentos Nos. 1-3H), 6 figs., 5 maps}. Dall, W. H. 1910. Summary of the shells of the genus Conus from the Pacific Coast of America in the U. S. National Museum. Proc. U. S. Nat. Mas. Vol. 38, No. 1741, pp. 217-228, June 6, 1910. [See pp. 219, 220]. 1911. Notes on California shells (II). Nautilus, Vol. 24, No. 10, pp. 109-112, February, 1911. [See p. 112]. Davis, W. M. 1928. The Coral Reef Problem. Amer. Geogr. Soc. Spec. Publ. No. 9, pp. I-V, 1-596, frontispiece and 227 figs, in text. [Clip- perton Island cited pp. 38, 93, 97, 148, 167, 199, 377, 517, 540]. Edwords, C. E. 1906. Clipperton Island and its strange birds and crabs (and the origin of its pirate treasure legend). San Francisco Chronicle, Vol. LXXXIX, No. 58, September 9, 1906, Sunday Supple- ment, p. 3 and p. 4, col. 7, 9 illustrations. Ekman, S. 1935. Tiergeographie des Meeres. (Akademische Verlagsgesell- schaft M.B.H.; Leipzig), pp. I-XII, 1-542, 244 figs, in text. 1953. Zoogeography of the Sea. (Sidgwick and Jackson, Ltd.: Lon- don), pp. I-XIV, 1-417, 121 figs, in text. [See especially pages 40, 42, 72-77]. Elschner, C. 1913. Corallogene Phosphatinseln austral Oceaniens und ihre Prod- ukte. Beitrag zur Kenntnis der Korallen-Inseln austral- Oceaniens, unter besonderer Beriicksichtigung Naurus, eines Reprasentanten dolomitisicrter und phosphatisierter gehobener Atolle. (Verlag Max Schmidt: Lubeck), 8°, 118 pp., 19 pis. (1 map col.), 18 photographs. [Clipperton Island, pp. 91-92. Review by C. Gagel, Geol. Zentralblatt, Bd. 20, No. 1, pp. 15-17, December 15, 1913]. Gardinkr, J. S. 1931. Coral Reefs and Atolls. (Macmillan & Co.: London), 1931, pp. I-XIII, 1-181, pis. 1-15, 33 figs, in text. [Clipperton Island, p. 172]. 358 San Diego Society of Natural History GiFFORD, E. W. 1913. Expedition of the California Academy of Sciences to the Galapagos Islands, 1905-1906. VIII. The Birds of the Gala- pagos Islands, with observations on the Birds of Cocos and Clipperton Islands (Columbi formes to Pelecaniformes) . Proc. Calif. Acad. Sci., 4th Ser., Vol. 2, Ft. 1, pp. 1-132, pis. 1-7, August 11, 1913. Glasser, M. 1932. [in] Bertrand, L., F. Blondel, J. Bourcart, A. Demay, M. Dreyfuss, L. Dubertret, P. Fallot, M. Glasser, H. Hubert, C. Jacob, L. Joleaud, A. Lacroix, and L. Neltner. La Geologie et les mines de la France d'Outre-mer. (Paris). Societe d'editions geographiques, maritimes et coloniales. Publica- tions du Bureau d'etudes geologique et minieres coloniales, (Paris), pp. I-VIII, 1-604 [including maps and tables], 3 folding maps. [Clipperton Island, p. 480. Review by K. Keilhack, Geol. Zentralblatt, Bd. 49, No. 7, pp. 362-363, June 1, 1933}. Hanna, G. D., and Strong, A. M. 1949. West American mollusks of the genus Conus. Proc. Calif. Acad. Sci., 4th Ser., Vol. 26, No. 9, pp. 247-322, pis. 5-10, text figs. 1-4, January 28, 1949. Hedley, C 1899. Zoogeographic Scheme for the mid-Pacific. Proc. Linn. Soc. New South Wales, Vol. 24, Pt. 3, (No. 95), pp. 391-417, December 9, 1899. Hertlein, L. G. 1937. A note on some species of marine mollusks occuring in both Polynesia and the Western Americas. Proc. Amer. Philos. Soc, Vol. 78, No. 2, pp. 303-312, pi. 1. [See especially pp. 306-308]. Hertlein, L. G.,and Strong, A. M. 1939. Marine Pleistocene Mollusks from the Galapagos Islands. Proc. CaHf. Acad. Sci., 4th Ser., Vol. 23, No. 24, pp. 367- 380, pi. 32, July 20, 1939. Ingram, W. M. 1947. Fossil and Recent Cypraeidae of the western regions of the Americas. Bull. Amer. Paleo., Vol. 31, No. 120, pp. 43(5)- 124(82), pis. 5(l)-7(3), May 2, 1947. 1951. The living Cypraeidae of the Western hemisphere. Bull. Amer. Paleo., Vol. 33, No. 136, pp. 125 (1)-178(54), pis. 21(l)-24(4), March 24, 1951. HeRTLEIN-EmERSON MOLLUSKS FROM ClIPPERTON ISLAND 359 Iredale, T. 1911. On the Value of the Gastropod Apex in Classification. Proc. Malacol. Soc. London, Vol. 9, Pt. 5, pp. 319-322, June, 1911. Lacroix, a. 1931. La Mineralogic de la France d'outrc-Mer au Museum Na- tional d'Histoire Naturelle. Bull. Mus. Nat. Hist. Nat. (Paris), Ser. 2, Vol. 3, supplement pp. 1-137. [Clipperton Island, pp. 108, 125, 126]. Lowe, H. N. 1933. Cruise of the Petrel. Nautilus, Vol. 46, No. 4, pp. 109-115, pL 9, April, 1933. [See p. 112]. Menard, Jr., H. W. 1953. Shear Zones of the northeastern Pacific Ocean and anomalous structural trends of western North America. 1953 Official Program of the Geological Society of America, Cordilleran Section, Stanford University, Stanford, California, March 27- 28, 1953, p. 22. Morrell, B. 1832. A narrative of four voyages to the South Sea, North and South Pacific Ocean, Chinese Sea, Ethiopic and Southern Atlantic Ocean, Indian and Antarctic Ocean from 1822 to 1831 . . . ( J. &J. Harper: New York), 492 pp., frontispiece. [See pp. 219-220]. Morrison, J. E. P. 1949. Notes on the Florida species of Bursa. Amer. Malacol. Union, News Bulletin and Ann. Report. 1949, pp. 10-13. [Received at Calif. Acad. Sci., February 3, 1950]. Ostergaard, J. M. 1950. Spawning and development of some Hawaiian marine gas- tropods. Pac. Sci. (Univ. Hawaii), Vol. 4, No. 2, pp. 75-115, 42 figs, in text, April, 1950. Pease, W. H. 1865. On the existence of an atoll near the west coast of America, and proof of its elevation. Proc. Calif. Acad. Nat. Sci., Vol. 3, pp. 199-201, February, 1865. Perrill, C. K. 1937. Forgotten Island. U. S. Naval Inst. Proc, Vol. 63, No. 412, pp. 796-805, 3 pis. (on pp. 796-798), June, 1937. Powell, A. W. B. 1933. Notes on the Taxonomy of the Recent Cymatiidae and Nati- cidae of New Zealand. Trans, and Proc. New Zealand Inst., Vol. 63, Pt. 2, pp. 154-170, pi. 23, 22 figs, in text, February 28, 1933. 360 San Diego Society of Natural History Slevin, J. R. 1931. Log of the schooner "Academy" on a voyage of scientific research to the Galapagos Islands, 1905-1906. Calif. Acad. Sci. Oct. Papers No. 17, pp. 1-162, pis. 1-16, 1 map, Feb- ruary 14, 1931. [See pp. 21-22]. SvERDRUP, H. U., Johnson, M. W., and Fleming, R. H. 1942. The Oceans . . . (Prentice-Hall, Inc.: New York), pp. I-X, 1-1087, 265 figs, in text, 7 charts, 121 plus I-Vb. tables. Teall, J. J. H. 1898a. A phosphatized trachyte from Clipperton Atoll. Geol. Mag., New Ser., Dec. IV, Vol. 5, No. 5, p. 234, May, 1898. 1898b. A phosphatized trachyte from Clipperton Atoll (northern Pacific). Quart. Jour. Geol. Soc. (London), Vol. 54, pt. 2, pp. 230-232, pi. 23, May 2, 1898. [See also discussion p. 233]. Wharton, W. J. 1898a. Note on Clipperton Atoll. Geol. Mag., New Ser., Dec. IV, Vol. 5, No. 5, p. 233, May, 1898. 1898b. Note on Clipperton Atoll (northern Pacific). Quart. Jour. Geol. Soc. (London), Vol. 54, Pt. 2, pp. 228-229, pis. 20-22, May 2, 1898. HeRTLHIN-EmHRSON — MOLLUSKS FROM CliPPHRTON ISLAND PlATI- 26 View of Clipperton Rock, a cavernous mass of altered trachyte, look- ing northwest from coral reef across an arm of the lagoon. The birds in the foreground are noddy terns. Photograph taken by Rollo H. Beck, August 10, 1905. The original negative from which this picture was made remained undeveloped for over a year until after the Expedition of the California Academy of Sciences to the Galapagos Islands returned to San Francisco on November 29, 1906. 362 San Diego Society- of Natural History Explanation of the Plate PLATE 27 Fig. 1. Cypraea isabella mexkana Stearns. Hypotype No. 9878, Calif. Acad. Sci. Dept. Paleo. Type Coll. Length 40.4 mm., width 22 mm. Dorsal view. Fig. 2. Cypraea Isabella Linnaeus. Hypotype No. 9879, Calif. Acad. Sci. Dept. Paleo. Type Coll. Length 35.4 mm., width 18.2 mm. Fig. 3. Nassarius catallus Dall. Hypotype No. 33343, Univ. Calif. Mus. Paleo. Invert. Type Coll. Length 13.4 mm., maximum diameter 8.5 mm. Inspection of a series of specimens of this species reveals consider- able variation in the sculpture. See Figs. 4, 9, 13. The present specimen has fine axial sculpture in comparison to that shown in Figs. 9 and 13. Fig. 4. Nassarius catallus Dall. Hypotype No. 33344, Univ. Calif. Mus. Paleo. Invert. Type Coll. Length 16.5 mm., maximum diameter 11.1 mm. Fig. 5. Cypraea sciirra Gmelin. Hypotype No. 9880, Calif. Acad. Sci. Dept. Paleo. Type Coll. Length 47.2 mm., width 24.2 mm. Dorsal view. Fig. 6. Cypraea scurra Gmelin. Ventral view of specimen shown in Fig. 5. Fig. 7. Delectopecten zacae Hertlein. Hypotype, right valve, No. 33347, Univ. Calif. Mus. Paleo. Invert. Type Coll. Length 16 mm., height 16.5 mm. Fig. 8. Cypraea teres Gmelin. Hypotype No. 9881, Calif. Acad. Sci. Dept. Paleo. Type Coll. Length (somewhat worn) 28.2 mm., width 16.3 mm. Fig. 9. Nassarius catallus Dall. Hypotype No. 33345, Univ. Calif. Mus. Paleo. Invert. Type Coll. Length 14.4 mm., maximum diameter 9.2 mm. This specimen, somewhat eroded, has coarse axial sculpture in com- parison to that shown in Figs. 3 and 4. Fig. 10. Delectopecten zacae Hertlein. Hypotype, right valve. No. 33348, Univ. Calif. Mus. Paleo. Invert. Type Coll. Length 18.3 mm., height 19.6 mm. Individuals of this species show variation in the devel- opment of radial threads on the right valves. Axial sculpture is nearly lacking on the medial portion of the present specimen in comparison to that shown in Fig. 7. Fig. 11. Cypraea teres Gmelin. Ventral view of specimen shown in Fig. 8. Fig. 12. CyniatiuDi vest/t/n)i Hinds. Hypotype No. 9883, Calif. Acad. Sci. Dept. Paleo. Type Coll. Length 55 mm., maximum diameter 29.5 mm. Fig. 13. Nassarius catallus Dall. Hypotype No. 33346, Univ. Calif. Mus. Paleo. Invert. Type Coll. Length 15.4 mm., diameter 10.3 mm. Hertlkin-Emerson — MoLLUSKS FROM Clipperton Island 363 Fig. 14. Cypraeu teres Gmelin. Hypotype No. 9882, Calif. Acad. Sci. Dept. Paleo. Type Coll. Length (worn) 36.1 mm., diameter 20.9 mm. Dorsal view. Fig. 15. Cypraea teres Gmelin. Ventral view of specimen shown in Fig. 14. Fig. 16. Delectopecten zacae Hertlein. Hypotype, left valve, No. 33348, Univ. Calif. Mus. Paleo. Invert. Type Coll. Length 18.3 mm., height 19.6 mm. Fig. 17. Dr/ipa nanus Linnaeus. Hypotype No. 9885, Calif. Acad. Sci. Dept. Paleo. Type Coll. Length 20 mm., maximum width l6.8 mm. Fig. 18. Bursa granulans Bolten (Roding). Hypotype No. 9884, Calif. Acad. Sci. Dept. Paleo. Type Coll. Length 54.2 mm., diameter 35 mm. Fig. 19. Clanculiis {Panocochlea) clippertonensis Hertlein & Emer- son, new species. Holotype No. 33341, Univ. Calif. Mus. Paleo. Invert. Type Coll. Height 8.6 mm., maximum diameter 10.5 mm. Fig. 20. Clanciilus {Panocochlea) clippertonensis Hertlein & Emer- son, new species. Basal view of specimen shown in Figs. 19 and 22. Fig. 21. Conns ehraeus Linnaeus. Hypotype No. 9886, Calif. Acad. Sci. Dept. Paleo. Type Coll. Length (slightly worn) 19.5 mm., maximum diameter 9.8 mm. Fig. 22. Clanculus {Panocochlea) clippertonensis Hertlein & Emer- son, new species. Apertural view of specimen shown in Figs. 19 and 20. Specimens shown in Figures 3, 4, 7, 9, 10, 13, l6, 19, 20, 22, were dredged in 110-200 fathoms off the east end of Clipperton Island by the scientific personnel of Expedition "Shuttle," U. S. Navy Electronics Laboratory, Point Loma, San Diego, California. All the other specimens illustrated on this plate, except those shown in figures 1, 2 and 18 from the Henry Hemphill Collection (C.A.S.) were collected by W. H. Ochsner on Clipperton Island during the Expedition of the California Academy of Sciences to the Galapagos Islands, 1905-1906. HERTLEIN-EmERSON MOLLUSKS FROM ClIPPERTON ISLAND PLATE 27 #*\ ..r^^. ^ / & (. g.!^ wP \ X 12 /#^"^ I; 14 15 I^US. COMP. ZOOL LIBRARY AUG 1 7 195c TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 14, pp. 365-392, fig. 1-4 HARVARD y5^1V€RSITY THE AMPHIBIANS AND REPTILES FROM RANCHO LA BREA BY Bayard H. Brattstkom Department of Zoology, Uiiirerstty of California, Los Angeles SAN DIEGO, CALIFORNIA Printed for the Society August 4, 1953 COMMITTEE ON PUBLICATION Joshua L. Baily Carl L. Hubbs Laurence M. Klauber Charles C. Haines THE AMPHIBIANS AND REPTILES FROM RANCHO LA BREA * BY Bayard H. Brattstrom Department of Zoology, University of California, Los Angeles MIJS. mh?. ZOOL AUG 17 19S3 OHMRSITY INTRODUCTION Data and records of Pleistocene reptiles and amphibians from the West coast of the United States are rare. A study was therefore made of the herpetological material from Rancho La Brea in the Los Angeles County Museum and the Museum of Paleontology, University of Cali- fornia. Stock (1949) has recently reviewed the literature on Rancho La Brea and the reader is referred to that paper for any additional information on Rancho La Brea not presented herein. ACKNOWLEDGMENTS The writer wishes to thank Dr. Hildegarde Howard of the Los An- geles County Museum and Dr. Charles L. Camp, Museum of Paleon- tology, University of California, Berkeley, for the loan of Rancho La Brea herpetological material. The writer wishes also to thank Drs. Raymond B. Cowles, George A. Bartholomew, Daniel I. Axelrod, and Laurence M. Klauber for their criticisms and suggestions. ^Submitted in partial satisfaction of the requirements for the degree of Master of Arts in Zoology, University of California, Los Angeles. 368 San Diego Society of Natural History PART I. RANCHO LA BREA LOCATION Rancho La Brca is in the Wilshire district of the city of Los Angeles, Los Angeles County, California. The pits, when found, occupied a grass- filled open area near the base of the wide bajada descending from the nearby Santa Monica Mountains. The pits are now part of Hancock Park, maintained by the Los Angeles County Museum. AGE Geological evidence (Eaton, 1928; Grant and Sheppard, 1939; and Woodring, Bramlette, and Kew, 1946) and paleontological evidence (Stock, 1929, 1949; Merriam, 1906, 1911, 1912, 1914; A. H. Miller, 1929, 1937, 1947; and L. Miller, 1910, 1911, 1912, 1925) suggest that the pits at Rancho La Brea are probably very late Pleistocene; just how late requires further investigation by more modern techniques. Many of the characteristic animals found at Rancho La Brea are similar to or iden- tical with types described elsewhere from Pleistocene horizons. Many representatives of living species are also known from the asphalt. It is not known if there has been continuous deposition in the pits from Late Pleistocene into Recent. It seems possible that the various pits may have been exposed to the surface periodically for varying lengths of time from late Pleistocene to Recent. This view is supported by the reptilian fauna to be discussed below. The presence of the skull of Camelops in a cave in west-central Utah and remains of ground sloths, Nothrotheri/nn, in the dry caves of Nevada, Arizona and New Mexico suggest that those species, which have also been found in Rancho La Brea, were living in North America in relatively recent times. Radiocarbon dating (Arnold and Libby, 1951) of the dung of Nnthvother'nim found in the Gypsum Cave, Nevada, gives an age range of from 8,000 to 10,000 years for this material. The presence of human remains and a few extinct species of birds, with no extinct mammals in pit 10, suggests that this pit is an important link be::ween the Pleistocene and the Recent (Merriam, 1914; Howard and Miller, 1939). MATERIAL Herpetological material from Rancho La Brea has been found in Pits A, B, 3, Y, 81, and 101, and in University of California Museum of Paleontology localities 2051 and 2052. Some material in the Los Angeles County Museum is labeled "Rancho La Brea" with no other data. The approximate positions of the majority of these localities may be seen in Fig. 1. Herpetological material from the early excavations is rare or lacking. In these sites, the larger bird and mammal bones were saved, but the residue, probably containing many amphibian and reptile bones, was dis- carded. Recently, Dr. W. D. Pierce of the Los Angeles Museum, assisted by Drs. E. E. Hadley and G. A. Kanakoff, made extensive re-examina- Brattstrom — La Brea Amphibians and Reptiles 369 tions of the material from Pits A and B, primarily in search of entomo- logical material. Dr. Pierce has developed a special xylene technique for separating the oil and tar from the smaller organic material. The writer is indebted to him for saving and separating the herpetological material. Dr. Pierce has also washed out the brain cases of some of the Saber- toothed Cats from earlier excavations and has found herpetological re- mains in them. The fragility of small amphibian and reptilian fossils, even those from Rancho La Brea, cannot be over-emphasized. METHODS All the herpetological material taken from the tar pits has been compared with recent skeletons and checked with descriptions in the lit- erature. In some cases the differences would not be of major osteological character but are those of minor shapes, processes, contours, and propor- tions. In some cases specific identification was impossible, but in most, the species could be determined. LIST OF SPECIES The following is a list of species of Amphibians and Reptiles iden- tified from Rancho La Brea which will be discussed in the following section : AMPHIBIANS Bnfo n est or Bujo horeas Hyla sp. Raua cutroya LIZARDS Sceloponis iiiagister Scelopori/s occ'identaVis Uta stcmshur'iana PhrynosoDUi coronatuvi El gar Id DiuU'icar'inata yianlHsia vigil is Cnemidophoriis tigris Eumeces skiltonianiis SNAKES Coluber {Sensii latii) Pitiwphis Ccitenifer Lampropeltis getiilus Crotdh/s viridis TURTLES Clemmys mdrmordtd Nestor Toad Western Toad Tree Frog Red-legged Frog Desert Scaly Lizard Western Fence Lizard Side-blotched Lizard California Horned Lizard Alligator Lizard Yucca Night Lizard Whip-tailed Lizard Western Blue-tailed Skink Racer Gopher Snake Western King Snake Western Rattlesnake Pacific Pond Turtle 370 San Diego Society of Natural History PART II. SPECIES ACCOUNTS Bufo nestor Camp Camp (1917) described Biifo nestor from Rancho La Brea distin- guishing it primarily on the basis of cranial elements. Limb and girdle measurements are not useful in distinguishing B. nestor from B. h. halo- philus. The parasphenoids are, however, useful in distinguishing the two species. No cranial bones, such as those used by Camp in his description, have come to light since his description of B. nestor. Bufo nestor was found in U. C. locality 2051 and is known only from fossil material. Camp placed B. nestor in the horeas-canoriis group of western toads characterized by not having cranial crests and with the presence of narrow fronto-parietals. Its habits and habitat were probably similar to B. horeas. It probably still should be regarded as a distinct species, not intergrading with B. horeas. as no intermediates (at least in parasphenoid measure- ments) can be found. The species is extinct. Bufo horeas halophilus Baird and Girard Camp (1917) established the first record of a Pleistocene Bnjo horeas halophilus from Rancho La Brea. Since then, some 19 "Riker Mounts" of toad bones, all from Pit A, have been available for examina- tion. None of this toad material is different from skeletons of living B. h. halophilus. No material of Scaphiopus has, as yet, been found in the asphalt. Two fragments of vertebrae of a toad from Pit B are here referred to B. horeas. Camp (1917) gives an account of the comparison of B. nestor with Rancho La Brea B. h. halophilus, and recent B. h. horeas and B. h. halophilus, as well as presenting measurements of all of these forms. None of this material will be presented herein other than to add what material has come to light since Camp's work. The following is a sum- mary of the material from Pit A. A series of 24 parasphenoids, 15 pterygoids, 5 ectopterygoids, and 17 lower jaws (whole) are the only skull material available for examina- tion. The measurements of the parasphenoids fall within the range given by Camp (1917) though they average somewhat smaller than the average given by Camp. Average measurements of specimens from Pit A are as follows: total length, 10.5; total width, 14.5; greatest width of anterior arm, 2.1; and greatest width of wing, 2.2 mm. The ratio of the width of the wing to the total length is 23.8. The lengths and widths of four of the pterygoids are: 9.9-9.5, 9-4-7.0, 9.2-7.1, 8.4-5.5 mm. Lengths and widths of three of the ectopterygoids are: 11.6-7.0, 10.8-6.7, and 12.5-7.5 mm. Lengths of the lower jaw are from 15.1 to 23-3 mm. There are no teeth in the lower jaw. A total of 41 humeri of amphibians are here all referred to Bujo. The humeri of Kana, Bujo. or Scaphiopus are indistinguishable. Of 30 available amphibian radio-ulna bones, 28 are referred to B. horeas, and the other two are identified as belonging to the species following. Brattstrom — La Brea Amphibians and Reptiles 371 By means of size and snape, 49 tibia-fibula have been identified as B. borers. The minimum and maximum of all specimens were 10.3 and 30.0 mm. long. The tibia-fibula of a good-sized recent Rd/hi ct/irnrii ciray- tcn'i measures 50.6 mm. in length. Twenty-seven tarsals and astragali, 1 epicoracoid, 7 clavicles, 17 right and 11 left coracoids, 20 left and 19 right scapulas are all referred to B. boreas. The coracoids are much shorter than in Rdihi and the median process is not as wide as is that of Raiui. The following measurements have been made of amphibian femurs from Rancho La Brea: Largest fossil, 31.6 mm. long; large middle size, 26.2 mm.; small middle size, 20.2 mm.; small, 10.5 mm. as compared with a recent Rcvia a. draytoni which measured 49.4 mm. long. They thus all agree with B. b. halophiliis and with the measurements given by Camp (1917). Many miscellaneous tarsal and carpal bones as well as 20 ilia, 3 coccyx, and 10 ischia are all referred to B. boreas on the basis of size and shape, though actually the tarsal and carpal bones cannot be distinguished from Rana or Scaph'iopus. The ischia of this species can be distinguished from Rana and Scaph'iopus by its dorsal extension. On size and shape 14 sacral, 119 thoracic, and 10 atlas vertebrae are referred to B. b. halophiliis. It is difficult to distinguish differences between Rana and Biijo atlases, but the thoracic vertebrae of Bujn do not have the high neural spine, or the posterior extension of the neural spine, as in Rana. The two species of Biijo here discussed are apparently not related to the Pliocene forms of Kansas (Taylor, 1941, 1942). The material of Bujo boreas from Rancho La Brea was referred by Camp (1917) to B. b. halophiliis (?). No major differences in the sub- species of Bi/fo boreas are evident, though the Rancho La Brea material may well be of B. b. halophiliis due to the widespread geographical range of the subspecies today. Pleistocene B. boreas may not have yet differen- tiated into the subspecies known today. Only a large series of skeletons of living and fossil B. boreas and its subspecies will give the answer to this problem. Bujo b. halophiliis is found in the area of Los Angeles today and has a wide range extending from northern California through the San Joaquin Valley and south into northwestern Baja California, while the other subspecies, B. b. boreas. extends northward to Alaska and eastward to Montana, Wyoming and Colorado. Hyla sp. A small radio-ulna and a small tibia-fibula from Pit A are here referred to Hyla sp. The tibia-fibula measures 7 mm. in length as com- pared with 10.3 and 30 mm. as the minimum and maximum for fossil B. boreas. The radio-ulna is small (see Fig. 2), the distal end expands widely and the proximal end is weakly bifurcated. Comparative measure- ments are given below in relation to Rana aurora. There are no apparent differences between the bones of Hyla regilla and H. arenicolor. Length Width mm. mm. Anterior Posterior 14.9 4.5 6.0 17.0 6.0 7.1 16.0 4.3 4.8 15.6 4.0 5.1 6.6+ 2.5 3.4+ 372 San Diego Society of Natural History It is probable that in the spring, Hyld would congregate about the small temporary pools undoubtedly in the Rantho La Brea area and per- haps mistake one of the asphalt pits for a pool. The small fragile bones would probably not preserve well, or they may have been overlooked in collecting. Rana aurora Baird and Girard One radio-ulna from Pit A, with the distal ends of the bone not widely separated as in Bufo and with the shape of the anterior end thin as compared with Bufo, is here referred to Rana aurora. It is indistin- guishable from present-day skeletons of R. a. draytoni. Comparative measurements are as follows: Ratia aurora — Fossil Rana a. draytoni — Recent Bufo boreas — Fossil Bufo boreas — Fossil Hyla sp. — Fossil Drawings of the radio-ulnas of these three species are given in Fig. 2, A, B, and C. One large, isolated vertebra from Pit B is here referred to Rana aurora primarily on the basis of the shape of the neural spine. Today Rana a. draytoni occurs in the Santa Monica Mountains ap- proximately two miles north of Rancho La Brea. This subspecies ranges from northwestern California, through the Sierra Nevada and Coast Ranges south into northwestern Baja California. Sceloporus magister Hallowell Material from Rancho La Brea localities Pit A and "no data," con- sisting of 14 dentaries, 7 maxilla, 1 quadrate, 4 frontals, 1 occipital, and 10 parietals are here referred to S. magister. The Sceloporus magister quadrate from Rancho La Brea has a dorso- lateral thin margin and a deep depression posteriorly whereas 5". orcutti is shallow posteriorly. The dentary can be distinguished from that of S. occidenlalis by its larger lize and by the presence of a well developed flat area on the lateral side of the dentary just lateral to the teeth. The occipital may be separated from that of 5. occidentalis only on the basis of size and minor details of the processes. Sceloporus magister is not found in the Los Angeles city area today. It is present primarily on the fringes of the desert. The cause of its disappearance from the coastal area is not known. A possible increase in aridity of the area should not have afi^ected a desert-margin species. Pos- sible interspecific competition with S. occidentalis might be a cause of its elimination from the area. What effect the increasing human population in the area had on the distribution of the lizard is unknown. Brattstrom — La Brea Amphibians and Reptiles 373 Sceloporus occidentalis Baird and Girard Material of 5". occidentalis, taken from Rancho La Brea localities A, 101, Y, 81, and "no data," include 40 dentaries, 24 maxilla, 8 parietals, 15 frontals, 2 occipitals, 3 scapulas, 1 femur and several miscellaneous bones such as parts of pelvic and pectoral girdles, vertebrae, sacra, and hyoids. The parietals of S. occidentalis are quadrangular. The frontals do not expand anteriorly and are usually found to have a median process. The dentaries are larger than in Uta and smaller than in S. magister or S. orcutti. The posterior teeth on the dentary and maxilla are trifid, the anterior teeth are not. There is no flat area just lateral to the teeth on the dentary. The lower end of the dentary extends posteriorly more than the upper does. This species differs from Sceloporus graciosi/s in details, especially as to the parietals and frontals. The material from Pits A and 81 may be referred to the subspecies S. 0. hiseriatHs, living in the area today, to emphasize the close similarity of the fossil and living specimens. The two dentaries and the two maxil- lae from Pit 101 and the one dentary from Pit Y differ sufficiently so that they are identified as S. occidentalis only. The most anterior mental foramen in the dentary from Pit Y is elongate rather than round. Remains of this species are the most abundant lizard material found in the Rancho La Brea diggings to date and it is also the most abundant lizard in the Los Angeles area today. Uta stansburiana Baird and Girard Rancho La Brea material from Pits A, 81, and "no data" consisting of 18 dentaries, 18 maxillae, 17 parietals, 10 frontals, 9 scapulae and numerous girdle and limb elements can be referred to LUa stansburiana. Usually the bones of IJta can be distinguished by their small size, how- ever, characters do exist that differentiate between this and other genera. The frontal bone is expanded anteriorly and the posterior portion is extremely thin in the central area. In both fossil and recent specimens the parietal is very thin in the anterior central area and is often broken through. The teeth are usually not trifid. The teeth of IJta are small and numerous. The dentary is pointed anteriorly, at least more so than Sceloporus occidentalis. This lizard's remains are very abundant in the Rancho La Brea material and it is also a common species in the area today. Phrynosoma coronatum (Blainville) The osteology of the genus Phrynosoma has been discussed by Bryant (1911) and recently by Reeve (1952). Fortunately the Rancho La Brea material consists of two posterior temporal spines from Pit A which, by their character, places them with Phrynosoma coronatum. According to Reeve (1952) two subspecies, P. c. jrontale and P. c. blainvilli, occur in the Los Angeles area today. A comparison of measurements of these two subspecies and the one Rancho La Brea specimen given below suggests 374 San Diego Society of Natural History that the fossil form had not as yet differentiated into the two subspecies. No definite conclusions can be reached, however, based on only one specimen. Distance Height of spines between mm. Widtii tips of Species of base spines Posterior Anterior Number mm. mm. Fossil— La Brea 6.0 4.4 6.1 4.4 1 P. c. fyontcde 6.1 4.5 5.9 3.6 5 P. c. bhunvill} 6.8 5.5 6.3 4.2 10 P. p. phityrh'iuos 4.5 3.7 2.6 2.4 1 Elgaria multicarinata (Blainville) From Rancho La Brea localities Pit A and "no data," 7 maxillae, 12 dentaries, 5 parietals, 4 frontals, 15 occipitals, 3 pterygoids, 7 angulare and articular bones of the lower jaw, and 1 quadrate are referred to Elgaria mnltkarinata. Tihen's recent (1949) revision of the genera formerly assigned to Gerrhonotiis makes identification of this form simple and the reader is referred to his paper for discussion of the osteology. The frontal and parietal are particularly diagnostic and can be easily referred to £. mult'i- carinata. The pterygoids have teeth on them and the teeth of the maxillae and dentary are conical. The bones of £. in/ilt/carinata are larger than any of the Elgaria coernleus group, and perhaps on size alone (mean length and width of all Rancho La Brea dentaries 11.6 and 3.1 mm.), may be referred to the subspecies E. m. ivehbi. Xantusia vigilis Baird Portions of 2 maxillae and 3 fragments of dentaries from Rancho La Brea Pit 101 are here referred to Xantusia vigilis. The general shape of the maxilla of both the fossil and of recent X. vigilis is triangular with a wide, flat anterior surface. The maxillae of X. henshaivi and X. riversiana are quadrangular and have no wide, flat anterior surface. The teeth on the maxilla and dentary bones of Xantusia are trifid. Xantusia vigilis does not occur in the area of the tar pits today. It is found in the Mohave and Colorado Desert region just outside the barriers formed by the San Gabriel, San Bernardino, and San Jacinto Mountains. It is not known what could have caused its extinction from the Los Angeles area. It is not known if changes in the human popula- tion of the area has caused its elimination from the Los Angeles area. It is possible that the specimens in Pit 101 represent an isolated race or form, diverging from the main vigilis stock, with subsequent extinction of this form in the area for unknown reasons. Additional material is needed from Rancho La Brea before the taxonomic status (on the sub- specific level) can be determined. Pit 101 is one that suggests a relatively older age than Pit A. Herpetologically this is emphasized by the Xantusia and one specimen of Brattstrom — La Brea Amphibians and Reptiles 375 a CueDi'idophoriis (see below) from this pit. The Sceloporiis and Eumeces from this pit are less like recent specimens. Cnemidophorus tigris Baird and Girard Rancho La Brea localities Pit A, 101, and "no data," contain 2 dentaries, 1 maxilla, and 6 frontals that are referred to Cnemidophoviis tigris. Maxillary and dentary bones with bifid teeth are easily referred to this genus. The shape and size of the bones, especially the frontal, are also diagnostic (DuBois, 1943). The material is referred to this species on similarity of size and shape. The only other Teiid in the area today is Cnemidophorus hyperythn/s, from which C. tigris can be distinguished by size and general shape of the bones. The one dentary from Pit 101, when compared with recent C. tigris, differs in that the dentary is more robust, larger, and wider. The teeth appear taller. The junction of the fork of the bifurcation of the teeth is more pronounced, especially below the fork. The distance between the two posterior mental foramen is greater than in recent specimens (2.3 mm. compared with an average of 1.6 mm. for recent C. tigris). It appears to be an extra large specimen of C. tigris. This large size might be explained on the basis of a warmer climate, hence larger individuals, according to Bergmann's rule as modified by Cowles (1945) for cold- blooded vertebrates. It might also be explained on the chance that an unusually large specimen was trapped in the pits. Even though differ- ences are apparent, they do not seem to be significant enough to warrant separating this as a distinct species or subspecies. Cnemidophorus tigris is found in the Los Angeles area today and is common in the Arroyo Seco, Tujunga, and San Gabriel washes as well as in the Santa Monica Mountains. Eumeces skiltonianus (Baird and Girard) Rancho La Brea material from Pits A, 3, Y, 101, and "no data," consists of 11 dentaries, 1 maxilla, and 8 parietals. All are referred to Eumeces skiltonianus. The osteology of this genus is presented by Taylor (1935) and is discussed by him in regard to two Pliocene forms (1941). The specimens from the asphalt alLagree with the characters of this genus as presented by Taylor (1935, 1941) and with skeletons of recent E. skiltonianus and E. gilberti. E. skiltonianus can be separated from the closely related E. gilberti only on the basis of size, gilberti being the larger. There are no major osteological differences between the two species, but there are small differences of details of shape of the various bones. The material from Pits 101, Y, and 3 (the latter being washed out of the cranium of a saber-tooth tiger from this pit) seem to be less like recent E. skiltonianus than the material from Pit A or "no data." Eumeces skiltonianus is found living in the area of the tar pits today, whereas E. gilberti lives only on the top of the San Gabriel Mountains and in the San Bernardino Mountains. 376 San Diego Society of Natural History Coluber sp. (sensu latu) One partly damaged lower jaw of a snake from Pit 81 is here re- ferred to the genus Coh/ber by reason of the fact that the Meckel foramen is almost closed laterally. The two genera, Coluber and Mas tic op his (of Ortenburger, 1928), can be distinguished by the shapes of the nasals, frontals, parietals, and postfrontals; however none of these bones were found in the asphalt. Generic {sensu strictii) and specific determination of this snake must await additional material. Pituophis catenifer (Blainville) All of the colubrid vertebrae found in the tar pits, with one excep- tion mentioned below, can be referred to Pituophis cateuifer. The gopher snake material was found in Pit A (96 vertebrae, 3 premaxillae). Pit B (26 vertebrae), U.C. 2051 (1 vertebra), U.C. 2052 (1 vertebra), and Rancho La Brea "no data" (1 vertebra). These vertebrae are identified as P. catenifer on the basis of size and measurements. Pituophis lacks the strong subcentrum keel and the heavy top ridge of the neural spine as found in Lanipropeltis. Pituophis has small hook-like processes on the anterior ends of the zygosphene which are generally lacking in Lampropeltis. Lampropeltis characteristically has rounded anterior corners of the zygosphene. Pituophis vertebrae can be distinguished from Thamnophis by the absence of the hypapophysis pres- ent in that genus. From other genera it difi^ers in size. Measurements of La Brea, McKittrick, and recent skeletons of Pituophis catenifer are presented in another paper (Brattstrom, 1953). Three premaxillae from Pit A are referred to P. catenifer on the basis of size and shape. They are not similar to any of the other colubrid pre- maxillae seen, but they differ from recent P. cate)?ifer in small details. The vertebrae appear more like P. c. annectens than P. c. affinis, P. c. deserticola or P. c. catenifer. Lampropeltis getulus californiae (Blainville) On the basis of a strongly developed sub-centrum keel and in the presence of rounded anterior corners of the zygosphene, one vertebra bear- ing no other data than "Rancho La Brea" is identified as Lampropeltis getulus. It is distinguished from Lampropeltis zonata on size and is re- ferred to this subspecies on the basis of geographic distribution. Crotalus viridis Rafinesque Crotalus material is available from Pits A, B, 81, U.C. 2052, and "no data." Previously Gilmore (1938) mentioned that due to lack of material all fossil Crotalids should be referred to Crotalus sp. Since 1950 the writer has been engaged in a study of the comparative osteology of the Crotalidae including the species and subspecies of Crotalus. All of the species of Crotalus can be distinguished by means of osteological char- acters. It is therefore possible to refer most of the fossil material, includ- ing Rancho La Brea specimens, to recent or fossil species on the basis of Brattstrom — La Brea Amphibians and Reptiles 377 size, shape, and comparative measurements. The differences in vertebrae of the various species is sometimes very diagnostic, but some species show great variabiHty. As additional studies of the comparative osteology of the rattlesnakes are in progress, a final analysis will be presented later. Fifteen vertebrae, 1 pterygoid with 8 teeth, and a questionable presphenoid from Pit 81 can all be identified as Crotalus v'lr'id'is. The material from Rancho La Brea with no data are also of this species. From Pit A there are 7 ribs or fragments of ribs, 3 presphenoids, 1 fragment of a lower jaw and many vertebrae. The presphenoids are variable among snakes and these are only tentatively referred here to C. vmdis. The ribs are all small, the measurements being similar to present day C. v'lr'idis heller}. The Crotalus vertebrae from Rancho La Brea can be identified as C. viridis on the basis of relative proportions of the parts and on the shape and size of the zygosphene, centrum and zygapophyses. A summary of measurements of La Brea, McKittrick and recent skeletons are pre- sented elsewhere (Brattstrom, 1953). The La Brea specimens resemble skeletons of C. v. oreganiis and C. v. helleri more than any of the other subspecies of C. viridis. The rattlesnakes from the La Brea pits were larger than average for this species. This is suggested by the material from U.C. 2052 and a few vertebrae from Pits A and B. If Bergmann's rule (as modified by Cowles, 1945) is applicable, the large size of some of these rattlesnakes may indicate a warmer climate. Crotalus viridis helleri and the very closely allied C. v. oreganus occur today from British Columbia to Baja California (the break between the two races being essentially in Santa Barbara County, California). C. v. helleri is found today in the Santa Monica Mountains just north of Rancho La Brea. Clemmys marmorata (Baird and Girard) Material from Rancho La Brea localities Pit A and "no data" con- sists of 3 femora, 5 innomonate bones, 2 scapulas, 1 humerus, 1 tibia, 1 fibula, 34 pieces of plastron and carapace, and 41 pieces of marginal dermal scutes. Some turtle material from Pits 3 and 5 is listed in the Los Angeles County Museum catalogue, but has not been found and is appar- ently lost. When Hay (1903) described the two western fossil forms, Clemmys saxea (Pliocene) and Clemmys hesperia (Miocene), from the John Day country of Oregon, he had no skeletons of the recent C. marmorata avail- able. The material from Rancho La Brea is easily identifiable with C. marmorata and does not differ from it except possibly in size, as seen in Figure 4. Clemmys saxea, the Pliocene species, can be separated from C. marmorata by the elongate pygal (Fig. 4). Clemmys hesperia, the Miocene form, can likewise be distinguished from C. marmorata on the shape of the hyoplastron. The hyoplastron of C. marmorata is straight lat- erally whereas in hesperia it is curved inward. Whether C. hesperia and saxea differ is not known as the available material of each fossil does not contain the diagnostic bones of the other species (Hay, 1903, 1908). Clemmys marmorata is found today in areas of permanent water 378 San Dihgo Society of Natural History from Washington to 31° N. Latitude in Baja California. It occurs today, though rarely, in the Los Angeles area in places of permanent water such as the San Gabriel Mountains, the Arroyo Seco and Los Angeles Rivers (prior to the cementing of the beds of these rivers) and Malibu Creek in the extreme eastern Santa Monica Mountains. It does not occur today in that part of the Santa Monica Mountains just north of Rancho La Brea. PART III. CLIMATE, ECOLOGY, AND ZOOGEOGRAPHY FLORA Plants are usually good indicators of past climates, but unfortunately the material from Rancho La Brea is rather scarce and of the material available only the more obvious species have been determined. The plant material from Pit A has not as yet been studied. It is at present in the process of being separated from the asphalt (Templeton, personal com- munication). So far the following plants have been identified from Rancho La Brea: V'nms Diiirkata Bishop Pine F'nuis sab'imana Digger Pine Cupressus sp. Cypress Quercus agr'ijolia Coast Live Oak Arctostaphylos sp. Manzanita Xanthhim calvmn Cockle Burr Samhucus glaucn Blue Elderberry Celt'is D/ississ/ppiemis var. ret/culcifci Western Hackberry The Cupressus sp. and the P/iu/s niinicata come from relatively older pits in association with the Mastodon and Imperial Mammoth, whereas the other species are from various pits. Beside the two pines mentioned above, another fossil pine, P. Vingu'ijormis Mason, resembles the Rancho La Brea material (Mason, 1932 and personal communication). The taxonomic status of the Cupressus is still in doubt (Mason, 1927) and Mason has suggested (personal communication) that, "". . . all the de- scribed species for the state of California might possibly be reduced to just two, a northern species and a southern. Cones alone are not adequate to distinguish these entities, the characters of the bark also being impor- tant." The plants of the older pits would indicate a humid flora, drying slightly to a more arid association of Juniper, Quercus and Arctostaphylos characteristic of Upper Sonoran Life Zone, with perhaps subsequent dry- ing to today's present level of aridness. FAUNA A summary of the ecology of the mammalian faunas is given by Stock (1949) and a summary of the avifauna is presented by Miller and De May (1942). Pierce (1946, 1947, 1948) has been studying the inver- tebrate fauna of the asphalt. In general, the bird, mammal, and inverte- brate faunas suggest a moist climate warming and drying to the condition found today. Brattstrom — La Brea Amphibians and Reptiles 379 AMPHIBIANS AND REPTILES Table 1 gives a summary of the number of individuals of the various species of reptiles and amphibians from the various pits of Rancho La Brea. The relative number of individuals is also presented in Figure 4. The number of individuals was estimated by counting the highest number of any one bone available for that species (or in the case of paired bones, counting and dividing by two) to give the approximate number of indi- viduals thus far found in the diggings of Rancho La Brea. The number of reptiles and amphibians is small compared to that of the birds and mammals. This is probably due to their small size and to the methods employed in collecting. It may be seen that most of the material comes from Pit A. This is probably due to the fact that this material was carefully handled and separated from the tar by Dr. Pierce, and hence more of the smaller mate- rial was saved. The Eiimeces from Pit 3 was washed out of the skull of a saber-tooth tiger. Camp has suggested (personal communication) that U.C. locality 2052 is relatively recent. On the basis of herpetological material it is suggested that pits 101, 81, 3, Y, and U.C. 2051 are rela- tively older than Pits A, B, and 2052. Pit B may be slightly older than A. There is no herpetological material from Pit 9, which contained the Pinus murkata and the Imperial Elephant material. Little attention has been paid to the comparative study of the faunas of the various pits, though there are obvious differences. Relative ages for some pits are sug- gested in some cases (Howard and Miller, 1939), but no analysis by pits has been given in the literature. Conspicuous by their absence in the asphalt are such forms as the spadefoot toads, Scdph}op//s, and the garter snakes, Thamnoph'is. Perhaps some of the indistinguishable bones of the amphibian material are of Scaphiopus, but its identity in the asphalt must wait until diagnostic material is available. Thamnoph'is should have been found in the asphalt if it had been in the area, as the presence of water would suggest. The vertebrae of this genus are very characteristic with a wide hypapophysis on each centrum. Many of the smaller forms (Salamanders, Hyps/glena, Tant/lla, etc.) would without doubt have been lost or broken in the move- ments of the asphalt, or these may have been lost in the diggings that had not been handled with such care as is being done by Pierce. Certain forms not found in the tar pits may be diie to their absence from the area during the deposition of the pits. Only one species, B//fo nestor, is extinct. RELATION TO OTHER PLEISTOCENE LOCALITIES IN CALIFORNIA On the basis of herpetological data, Rancho La Brea contains mate- rial that appears to be equal in age, at least in part, to the McKittrick asphalt, but with deposition possibly continuing into Recent times. As judged from conclusions reached on the basis of avian and mammalian faunas (Stock, Merriam, Miller), Rancho La Brea is definitely younger than either Hawver or Potter Creek Caves. Herpetological material tends 380 San Diego Society of Natural History to support this view, though there appears to be less of a change in the herpetofauna than has taken place in the mammalian and avian fauna. A summary of the other California Pleistocene herpetological material is presented elsewhere (Brattstrom, 1953). ZOOGEOGRAPHY There appears to have been no conspicuous local change in the zoo- geography of the reptiles and amphibians of the Late Pleistocene as indi- cated by the forms found in the asphalt of Rancho La Brea. Most of the forms in the pits can still be found in the immediate vicinity today or in adjacent regions of Los Angeles County such as the Santa Monica, San Gabriel, or Palos Verdes Mountains (Bogert, 1930; Hill, 1948). Only one form, Bufo tiestor, is extinct. The single species of turtle, Clemmys manuoyata, is still found in the adjacent areas of Los Angeles County wherever there are permanent streams. Two species, Xautiisia vigilis and Scelopori/s niagister, occur typically at the edge of the desert today. The factors that caused the disappearance of these species from the area, or whether they were populations separated from the main stock, is un- known. The apparent decrease in rainfall would not affect these two as it would the turtle, nor would it be expected that a small increase or de- crease in temperature would affect them both. Strictly Lower Sonoran desert species {D'lpsosaurus, Callisaurus, etc.) have apparently not been in the Los Angeles basin during the time of the tar pits. Other forms might have been kept from entering the asphalt by having different eco- logical habitats than that which was present at Rancho La Brea {Law- propeltis zo>jata, Sceloporus grac/osus, Anniella pulchra, etc.). Other material, of very small size, is probably present but, as yet, undiscovered. CLIMATE AND ECOLOGY It is difficult to suggest paleoclimatic conditions based on the present day ecology of the same or similar forms and their associated faunas and floras, but this is the only means available in most cases. As a means of estimating the paleoclimate of the time of deposition of the animals in the asphalt, a summary of the optimum body tempera- tures for existent relatives is given below (Data from Stebbins, 1951; Mullally, m.s.; Cowles and Bogert, 1944; Cole, 1943; Zweifel, personal communication). Species Optimum body Mean or temperature Median Bufo boreas 16.7-22.2°C 19.5°C 22-24.5 23.2 Sceloporus magister 34-37 35.5 Sceloporus occidentalis 36.6 36.6 Uta stansbur'iana 35-38 36.5 Phrynosoma coronatum 34.9 34.9 Elgar/d multicarinata 27.4 27.4 Xantus/a vigilis 29 Total 29.0 242.6 Mean 30.32°C. (86.6°F.) Brattstrom — La Brea Amphibians and Reptiles 381 This is not to infer that the temperatures during the Late Pleistocene were as high as indicated by these data, as these forms can alter their body temperatures by behavior, so that they may maintain their optimum metabohc rates under seemingly unfavorable air temperatures. These animals can, of course, operate for a time at temperatures below their optimum, but not much above the optimum, as these temperatures of reptiles and amphibians are usually just a few degrees below their lethal temperature (Cowles and Bogert, 1944). Most of the present day reptiles and amphibians represented in the asphalt are characteristic of the Upper Sonoran Life Zone. A few of the species range into coastal Lower Sonoran and/or Dry Transition Life Zones. A few species {Crotahis viridis. El gar/a multicar'mata) range into higher zones and are probably not good indicators of past climate. Most of the forms are typically open valley, ground-dwellers, {Bufo boreas, Pituophis, Coluber. LciDipropeltis. Uta, Phrynosoma and Cnemidophorus), while some {Rana aurora, Crotalus viridis, Elgaria. and Eumeces) prefer rocky areas. A few {Bufo boreas, Xantusia vigil is) are nocturnal today, Pituophis, Crotalus viridis. and Lawpropeltis are both diurnal and crepus- cular, and the remainder are diurnal. A moist situation, with permanent streams or pools, is suggested by the presence of the Bufo. Hyla. Raua, Eumeces, and Clemmys. The Rana (Grinnell and Camp, 1917; Stebbins, 1951) and the Clemmys (Storer, 1930; Seeliger, 1945; Carr, 1952) are restricted today in southern California to areas of permanent water. Notes on the ecology of Eumeces skiltonianus according to Rogers and Fitch (1947) suggest that: "Although it seems to be adaptable to widely different habitats and climates, it appears to favor moderately humid climate. It is apt to be concentrated in localized colonies where there is an abundance of ground cover in the form of dead wood or flat rocks and a good growth of herbaceous vegetation. The localities where skiltonianus has been found in the greatest abundance are mostly in or near open woods of blue oak, garry oak, or coast live oak and bay {Umbellularia)." The presence of this lizard in the pits would suggest a more humid climate than in the area today. This is also supported, especially in the case of the older pits, by the presence of the Cupressus and Pinus muri- cata. This evidence would also suggest a pushing northward of the species as a result of increasing aridity of the southern part of its range. This suggestion is supported by the restricted range in the Cape region of Baja California of Eumeces lagunensis, which is very similar to E. skiltonianus, and which has, by some authors, been considered as a subspecies of skil- tonianus even though intergradation does not occur (see Rogers and Fitch, 1947, for a discussion). As based on the ecology of the reptiles and amphibians found in the asphalt and by the inclusion of data derived from the flora and other faunas, the ecological conditions surrounding the asphalt pits was prob- ably greater than at present, as indicated by the plants, aquatic amphibians 382 San Dihgo Society of Natural History and reptiles, and aquatic beetles and bugs. Undoubtedly small permanent streams or pools were present in the area. The temperature was probably the same as today or a little warmer. There was probably a local diminution of rainfall from a time in which Cypress and Pine were locally found through a time of open, inte- rior and warmer, Juniper-Oak association, which finally gave over to an Oak-woodland Savanna and Coastal Sage-scrub association. The topog- raphy was probably, from all geological evidence, the same as today. CONCLUSIONS Previous to this study, Bufo nestor, B. boreas and Clemmys sp. were the only herpetological species that had been identified as coming from the asphalt of Rancho La Brea. As a result of this study, 17 forms, rep- resenting approximately 135 individuals, have been identified. There is no evidence to prove that there has not been continuous or partial deposition in the pits since their formation. The fact that the pits may be of different ages, ranging from Late Pleistocene to Recent, is sug- gested by the fauna and flora of the various pits. Bujo nestor is the only extinct amphibian found. Most of the forms studied are identical with living species found in the area today. Two reptiles, Xantusia vigH'is and Sceloporus magister, are found only on the periphery of the desert today and not in the vicinity of the city of Los Angeles. The reason for the elimination of these two forms from the area is unknown. An aquatic turtle, Clemmys marmoyata, is absent in the immediate Los Angeles area today except in areas of permanent water. This would suggest a local diminution of yearly rainfall. This view is supported by the presence of amphibian {Rana) and botanical material. A summary of the ecology of the birds, mammals, amphibians, rep- tiles, and insects suggests an Upper Sonoran Life Zone. From the dis- cussion of the plants and with a possible differentiation of time of depo- sition of the various pits, it is suggested that from Late Pleistocene to Recent there was a local transition from a moist climate of Piniis and Cnpyessiis through a stage of decreasing rainfall and a vegetation of Oiierciis agr'ijolia and Juniper/is califoni/ca. to the present-day climate and vegetation of Oakwoodland Savanna and Coastal Sage-scrub with subsequent changes in the fauna. Brattstrom — La Brea Amphibians and Reptiles 383 LITERATURE CITED Arnold, J. R. 1951. Radiocarbon Dates. Science 113: 111-120. BOGKRT, C. M. 1930. An annotated list of the amphibians and reptiles of Los An- geles County Calif. Bull. So. Calif. Acad Sci. 29 (1): 1-14. Brattstrom, B. H. 1953. Pleistocene reptiles from California. Copeia 3: Bryant, H. C. 1931. The horned lizards of California and Nevada of the genera Phrynosoma and Anota. Univ. Calif. Publ. Zool. 9 (1): 1-84. Camp, C L. 1917. An extinct toad from Rancho La Brea. Univ. Calif. Publ. Bull. Dept. Gcol. 10 (17): 287-292. Carr, Archie 1952. Handbook of turtles. Cornell Univ. Press, xv-542 pp. CoLK, La Mont C. 1943. Experiments on toleration of high temperatures in lizards with references to adaptive coloration. Ecology 24 (1): 94-108. COWLES, R. B. 1945. Surface-mass ratio, paleoclimate and heat sterility. Amer. Nat. 79: 561-567. CowLES, R. B. and C M. Bogert 1944. Preliminary study of the thermal requirements of desert rep- tiles. Bull. Amer. Mus. Nat. Hist. 83 (5): 261-296. Du Bois, E. P. 1943. Osteology of the skull of Cnem/dophor/ts. Amer. Midi. Nat. 30 (2): 510-517. Eaton, J. E. 1928. Divisions and duration of the Pleistocene in Southern Cali- fornia. Bull. Amer. Assoc. Petrol. Geol. 12 (2): 111-141. GiLMORE, C. W. 1938. Fossil snakes of North America. Spec. Pap. Geol. Soc. Amer. 9: 1-96. Grant, U. S. and W. E. Sheppard 1939. Some recent changes of elevation in the Los Angeles basin of southern California and their possible significance. Bull. Seismol. Soc. Amer. 29 (2): 299-326. Grinnell, J. and C L. Camp 1917. A distributional list of the amphibians and reptiles of Cali- fornia. Univ. Calif. Publ. Zool. 17 (10): 127-208. 384 San Diego Society of Natural History Hay, O. p. 1903. Two new species of fossil turtles from Oregon. Univ. Calif. Publ. Bull. Dept. Geol. 3 (10): 237-241. 19O8. The fossil turtles of North America. Carnegie Inst. Wash. Publ. 75: iv-547. Hill, H. R. 1948. Amphibians and reptiles of Los Angeles County. Los An- geles Co. Mus. Sci. Ser. 12: 1-29. Howard, H. and A. H. Miller 1939 The avifauna associated with human remains at Rancho La Brea, California. Carnegie Inst. Wash. Publ. 514 (3): 39-48. Mason, H. L. 1927. Fossil records of some West American conifers. Ibid. 346: 139-160. 1932. A phylogenetic series of the California closed-cone pines sug- gested by the fossil record. Madrono 2 (6): 49-55. Merriam, J. C. 1906. Recent discoveries of Quaternary mammals in southern Cali- fornia. Science, n.s. 24: 248-250. 1911. The fauna of Rancho La Brea, Pt. 1. Occurrence. Mem. Univ. Cahf. 1 (2): 197-213. 1912. Ibid. Pt. 2. Canidae. Ibid. 1 (2): 217-272. 1914. Preliminary report on the discovery of human remains in an asphalt deposit at Rancho La Brea. Science, n.s. 40: 198-203. Miller, A. H. 1929. The passerine remains from Rancho La Brea in the paleonto- logical collections of the University of California. Univ. Calif. Publ. Bull. Dept. Geol. Sci. 19 (1): 1-22. 1937. Biotic associations and life-zones in relation to the Pleistocene birds of California. Condor, 39: 248-252. 1947. A new genus of Icterid from Rancho La Brea. Ibid. 49 (1) : 22-24. Miller, L. H. 1910. Fossil birds from the Quarternary of southern California. Ibid. 12: 12-15. 1911. A synopsis of our knowledge concerning the fossil birds of the Pacific coast of North America. Ibid. 13: 117-118. 1912. Contributions to avian paleontology from the Pacific coast of North America. Univ. Calif. Publ. Bull. Dept. Geol. 7 (5): 61-115. 1925. The birds of Rancho La Brea. Carnegie Inst. Wash. Publ. 349: 63-106. Miller, L. H. and I. S. DeMay 1942. The fossil birds of California; an avifauna and bibliography with annotations. Univ. Calif. Publ. Zool. 47 (4) : 47-142. Brattstrom — La Brea Amphibians and Reptiles 385 Ortenburger, a. T. 1928. The whip snakes and racers, Genera Masticophis and Coluber. Mem. Univ. Mich. Mus. I: 1-247. Pierce, W. D. 1946. Exploring the minute world of the California asphalt depos- its. Bull. So. Calif. Acad. Sci. 45 (3): 113-118. 1947. A progress report on the Rancho La Brea asphaltum studies. Ibid. 46 (3): 136-138. 1948. The carabid genus Elaphrus in the asphalt deposits. Ibid. Al (2): 53-55. Reeve, W. L. 1952. Taxonomy and distribution of the horned lizards genus Phrynosoma. Univ. Kansas Sci. Bull. 34 (14): 817-960. Rodgers, T. L. and H. S. Fitch 1947. Variation in the skinks (Reptilia: Lacertilia) of the skil- tonianus group. Univ. Calif. Publ. Zool. 48 (4) : 169-220. Seeliger, L. M. 1945. Variation in the Pacific Mud Turtle. Copeia 3: 150-159. Smith, H. M. 1946. Handbook of lizards. Comstock Publishing Co. xxi-557 pp. Stebbins, R. 1951. The amphibians of western North America. Univ. Calif. Press. 1-506 pp. Stock, C. 1929. A census of the Pleistocene mammals of Rancho La Brea, based on the collections of the Los Angeles County Museum. Jour. Mamm. 10 (4): 281-289. 1949. Rancho La Brea, A Record of Pleistocene life in California (4th Ed.). Los Angeles Co. Mus. Sci. Ser. 13: 1-81. Taylor, E. H. 1935. A taxonomic study of the cosmopolitan scincoid lizards of the genus Eumeces. Univ. Kansas Sci. Bull. 23: 1-643. 1941. Extinct lizards from Upper Pliocene deposits of Kansas. State Geol. Surv. Kansas Bull. 38 (5): 165-176. TlHEN, J. A. 1949. The genera of Gerrhonotine lizards. Amer. Midi. Nat. 41 (3): 580-635. WooDRiNG, W. P., N. 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. 386 San Diego Society of Natural History Table 1 Summary of the approximate number of individuals of reptiles and amphibians from the various pits in Rancho La Brea. No 2051 2052 data Total SPECIES A B 101 81 3 AMPHIBIANS Bufo nestor (extinct) Bufo boreas Hyla sp. Rana aurora 25 1 1 1 1 SNAKES Phuophh catenifer Coluber {sensu latu) Lanipropeltis getulus Crotalus rhidh 3 2 1 1 2 1 LIZARDS Sceloporus maghter Sceloporus occidentalis Uta stansburiana Phrynosoma coronatum Cnemidophorus tigris Elgaria multicarinata Eumeces skiltonianus Xantusia vigilis 6 14 11 1 6 6 3 2 4 1 1 1 1 1 turtles Clemmys marmorata 2 15 41 1 2 1 7 1 1 1 I 7 1 7 2 23 1 13 1 1 s 1 7 I 7 1 OG DE N DR. D — / } :d m 1 1 en CD / ; |- / < \ o 1 1 'T7 / > / / ^ •. / / ° / y j • \ CO X H > < 3AV Nosano Figure 1. Map of Rancho La Brea showing locality sites of major excavations. B C Figure 2. Amphibian radio-ulna bones from Rancho La Brea. A. Rana aurora, B. Hyla sp., and C. Bufo boreas. Figure 3. Pygal bones of Clemtnys marmorata and Clemmys saxea. A and B. C. marmorata from Rancho La Brea, Pit. A. C. C. marmorata from Rancho La Brea with no pit data. D. Recent C. marmorata from San Gabriel River, Los Angeles County, California. E. Clemmys saxea, Type, Pliocene of Oregon (Redrawn from Hay, 1908, pg. 294). Figure 4. Chart showing the relative abundance of the species of amphibians and reptiles thus far found in Rancho La Brea. TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 15, pp. 393-404, Fig. 1-4 WEST AMERICAN RAZOR-CLAMS OF THE GENUS ENSIS BY S. Stillman Berry Redlands, California mi. COMP. ZOOL. LIBRARY AUG ?8 1953 NAftVARO t UNIVERSITY SAN DIEGO, CALIFORNIA Printed for the Society August 14, 1953 COMMITTEE ON PUBLICATION Joshua L. Baily Carl L. Hubbs Laurence M. Klauber Charles C. Haines !AUG 2 8 1953 L UHiVERSITY WEST AMERICAN RAZOR-CLAMS OF THE GENUS ENSIS BY S. Stillman Berry Redlands, California Prior to the present study but a single species of the genus Ensis Schumacher (family Solenidae) has been recognized in the Recent fauna on the western coast of North America, although the related genus Solen is quite generously represented. That there are two species actually present and confused both in our collections and in the literature is the burden of the ensuing communication. Grateful acknowledgement is due to the collectors variously men- tioned in the following pages for the gift of helpful material, and espe- cially to Dr. Harald A. Rehder and his associates at the United States National Museum and to Dr. Joshua L. Baily, Jr., of the San Diego Society of Natural History, as well as to Dr. A. Myra Keen of the Depart- ment of Paleontology, Stanford University, for making available to my study the essential material in the collections under their care. I am indebted to Mr. Edgar R. Fisher of Redlands for making the photographs used in the plate. 1. Ensis californicus Dall 1899 1899. Ensis californicus Dall,— Proc. U. S. Nat. Mus., 22: 108 (except the Calif ornian record)-; 110. 1936. Ensis californicus Jordan, — Contr. Geol. Dept. Stanford Univ., 1 (4): 112 (rec. from Pleist. of Magdalena Bay). 1950. Ensis californicus Hertlein and Strong, — Zoologica, N.Y., 35 (4): 227 (except the Californian record). Holotype: U. S. Nat. Mus. Cat. No. 158891. Type-Locality: 14 fathoms, off San Pedro Martir Id., Sonora, Mex- ico; U. S. F. C. Str. ""Albatross," Sta. 3013. 396 San Diego Society of Natural History ^o ^ Eri&is C3k.\\4-OT'rtic.u.S tip Fig. 1. Map showing the presently known distribution and range of the two west American species of Ensis* * The Santa Barbara record was established too late for inclusion in the West American Razor-Clams of the Genus Ensis Material Examined : 397 No. Speci- mens Depth Locality Collector or Source Mus. No. Remarks 5 1 frag- ments 1 5/2 fms. Santa Maria Bay, Baja Calif. Cape San Lucas, Baja Calif. off Cape San Lucas, Baja Calif. La Paz, Baja Calif. H. N. Lowe Stearns Coll. ex Verrill "Albatross" Sta. 2835 Belding SDM 22549 USNM 74854 USNM 158892 USNM 34064 imper- fect 12 shore San Felipe, Baja Calif. H. N. Lowe 1933 SDM 22577 1 V. beach drift San Felipe, Baja Calif. Ralph Hav- ickhorst 1952 Berry 18355 2 V. 1 beach 14 fms. Miramar. Sonora off San Pedro Martir Id., Sonora Mary E. Long, Jan. 1951 "Albatross" Sta. 3013 Berry 19680 USNM 158891 holotype Fig. 2. Ensis calijornicus Dall. Anterior portion of a left valve from San Felipe, Baja California (No. 18355) in camera outline, showing the hinge- plate. Additional (probably valicT) Published Records: Pleist. — N. of vil- lage of Magdalena Bay, Baja California (Jordan). Recent — 30 fathoms, off Manzanillo, Colima (Hertlein and Strong). Commentary: This slender, exceedingly graceful, highly polished little species is one of the most attractive west American representatives of the Solenidae. It has been much misunderstood and frequently mis- identified, partly no doubt because it is far from common in collections, and partly because of the lack of any correct representation of it in the literature. So far as I can discover the true species is here illustrated for the first time. The finest series of the shells that I have seen is that from 398 San Diego Society of Natural History San Felipe, collected by the late H. N. Lowe, and now in the San Diego Museum of Natural History, where it has been reposing under the name Etisis mexicanus Dall; the latter properly a Sole)i. More detailed inter- pretations and comparisons will be found in the commentary upon the following species. 1892. Solen ensis var. minor Dall {not of Conrad) in Williamson, — Proc. U. S. Nat. Mus., 15: 184 (recorded from Long Beach, Calif.). 1899. Ensis calif ornicus Dall, — Proc. U. S. Nat. Mus., 22: 108 {pars, — the Calif ornian record only). 1907. Ensis calif ornicus Kelsey, — Trans. S. D. Soc. Nat. Hist., 1 (2): 39 (rec. from vicin. San Diego, Calif.). 1924. Ensis californicus Oldroyd, — Stanford Univ. Publ. Geol. Sci., 1 (1): 189 {pars, — the Californian record only); pi. 49, fig. 6 {not of Dall). } 1937. Ensis californicus Willett,— Trans. S. D. Soc. Nat. Hist., 8 (30): 391 (rec. from Upper Pleist. of Baldwin Hills). 1945. Ensis californicus Burch, — Minutes Conch. Club So. Calif. (43): 27. 1948. Ensis californicus Smith and Gordon, — Proc. Cal. Ac. Sci., 26 (8): 176 (Monterey Bay records). 1950. Ensis californicus Hertlein and Strong, — Zoologica, N.Y., 35 (4) : 227 {pars, — the Californian record). Fig. 3. Ensis myrae n. sp. Anterior portion of left valve of an immature paratype (No. 1256) having nearly the same length as the shell illustrated in Fig. 2 ; camera outline in same scale. West American Razor-Clams of the Genus Ensis 399 Fig. 4. Ensis myrae n. sp. Hinge-plate of left valve of holotype in camera outline; same scale as Figs. 2 and 3. Descript'/on: Shell of moderate size and slenderness, rather strongly falciform; valves narrowing slightly and rather sharply and squarely truncate in front, more rounded and narrowing a little more steadily and decidedly posteriorly. Hinge-plate strongly anterior in position, small and comparatively short; right major cardinal compressed, squarish in profile, strongly projecting; right posterior cardinal slender, short (appre- ciably less than one-tenth the length of the shell), laminar, terminating in a short free flange; left major cardinals strongly unciform and consid- erably heavier than that in the right valve; inner left posterior cardinal generally similar to its mate of the right valve, the weaker outer cardinal closely adnate to the shell margin. Exterior of shell mostly with a some- what silky sheen due to the numerous fine growth-striae underlying the thin shining Sayal Brown to Mikado Brown periostracum; in adult shells the periostracum is usually rubbed away in a blade-shaped swath extend- ing from the beaks to the posterior margin, this area being white except for some ruddy or purplish coloring, especially along the edges of the area and in the incremental rest-marks, where the Deep Rose Pink or Vinaceous tinting of the interior may shine through. Coffiparative Measurements : Max. Alt. Species Specimen - Cat. No. Long, mm. mm. % of length Diam. mm. californicus valve, Miramar 19680 58.2+ 6.9 11.8 " San Felipe 18355 54.8 5.8 10.6 — myrae holotype SU7852 81.3 11.9 14.6 5.8 " paratype 74.8 11.7 15.6 6.2 " 1256 56.+ 7.8 14.0 3.8 Holotype: Stanford University Paleo. Type Coll. No. 7582. Paratypes: No. 8533 Stanford Univ. Paleo. Coll. (Oldroyd, No. 359): Nos. 1256, 18370, and 18371 Berry Collection; No. 1151 W. O. 400 San Diego Society of Natural History Gregg Collection; others to be deposited in the collections of the San Diego Museum of Natural History, the Los Angeles Museum, Museum of Comparative Zoology, and the United States National Museum. Type-Locality. San Pedro Bay, California; mainly cast up by storms in the vicinity of Terminal Island; Mr. and Mrs. T. S. Oldroyd, Dr. R. H. Tremper, Mrs. M. Burton Williamson, and Mrs. W. H. Eshnaur, collectors. Additional Specimens Examined: Number Speci- mens Depth Locality Collector Repository Remarks 1 8-10 fms. off Monterey, Calif . W. H. Dall USNM 158893 fragment 1 Santa Barbara, Calif. L.A. Mus. 3 (+ Long Beach, Calif. Mrs. M. B. L.A. Mus. 4 V.) Williamson 16 Long Beach, Calif. H. N. Lowe SDM 22550 4 Point Loma, Calif. Mary G. Beckwith SDM 4847 juv. Additional Published Records (all as E. calijornicus but probably valid for E. myrae): California — 15 fathoms, off Pacific Grove, Mon- terey Co. (Smith and Gordon); 25 fathoms, off Redondo, Los Angeles Co. (Burch); Newport Bay, Orange Co. (Burch); vicinity of San Diego, San Diego Co. (Kelsey). Commentary: This has been a familiar species to Californian mala- cologists for many years, but has gone unrecognized through its incorrect identification with the more southern and tropical E. calijornicus Dall, a confusion the more strange since any direct comparison at once reveals the two as amply distinct. The error actually originated with Dall him- self when in his first publication of E. calijornicus (1899: 108) he in- cluded in its attributed range a locality as far north as Monterey Bay. A further natural bias toward the confusion arose from the name selected, since this, although perfectly apropos to any species described from the Gulf of California is resorted to more commonly by taxonomists when reference is intended to the State of that name. Happily Dall did not specify any Californian paratypes, otherwise he would neatly have pointed up the danger inevitably inherent in the unfortunate and increasingly prevalent present practice by some authors of denominating paratypes of a new species from a locality or horizon different from that of the holo- type. Now whatever the lack of discrimination prevailing in the literature of these two species, E. myrae is readily distinguishable from E. calijor- nicus by 1) its much greater size and robustness; 2) its definitely stronger arcuation; 3) its generally subdued polish, the conspicuous and close growth-striae giving it a somewhat silky lustre; 4) the squared-off rather than rounded truncation of the anterior end; 5) the nearly terminal rather than subterminal position of the relatively stronger hinge-plate, with West American Razor-Clams of the Genus Ensis 401 the posterior cardinals markedly shorter; and 6) a pallial Hne running appreciably nearer the margin. All of these appear to be reasonably good and dependable solenid characters. Superficially E. myrae does not appear greatly dissimilar to the somewhat larger European E. ensis (Linne), but there are numerous differences in detail, particularly in the hinge. An excellent and very characteristic figure of the exterior of one of the shells of the type-lot has been published by Oldroyd (1924: pi. 49, fig. 6). So far as I have been able to learn all specimens collected from San Diego north belong unquestionably to E. myrae, all the West Mexican examples from as far up the coast as Magdalena Bay to califoiuic/is. From the long outer shore of Baja California between these two widely sundered localities no species of the genus have as yet been seen, so it is not known in what precise area the two species meet or overlap, or whether they are separated by some hiatus in distribution. However this may be, their disentanglement resolves a somewhat anomalous minor distributional problem, the supposed penetration of a characteristically West Mexican and therefore tropical species as far north as the cool- temperate waters of Monterey Bay. The data here offered do indeed indicate that E. calif ornic/is is sufficiently thermal in its preferences to be placed without reserve in the tropical fauna, while £. tnyrae is as defi- nitely warm to cool temperate. This conclusion becomes of potential con- sequence in assisting to a correct interpretation of any fossil faunule in which either of these two species may prove to be represented. Up to the time of writing I have been able to find but two references to the occur- rence of any species of Ensis in the Californian and Mexican Pleistocene (Jordan, 1936: 112, and Willett, 1937: 391). As the fauna elucidated by Willett is a temperate one but with a strong warm-water element in- volved, it would be of much interest to re-examine his material and estab- lish its true identity. I can further find nothing on record concerning the natural history of E. myrae save a brief but suggestive comment by Burch (1945: 27) which I think is worth quoting in full; "Our experience has been to find this species exceedingly rare. We have dredged great numbers with the 25 fathom gravel off Redondo Beach but by the very nature of the bottom and the dredgings it was a very low ratio of recovered complete speci- mens. The species is quite fragile. We have sets in our collection labelled as from San Pedro Bay and from Terminal Island, collected by Mrs. W. H. Eshnaur and others in the years before the dredging of Deadman's Island, etc." This fine species is here dedicated to Dr. A. Myra Keen of Stanford University in deserved recognition of the major contributions she is steadily making to our knowledge of west American pelecypods. 402 San Diego Society of Natural History References BURCH, J. Q. 1945. [Family Solenidae]. Minutes Conchological Club Southern California, No. 43: 26-29 [in mimeograph]. Jan. 1945. Dale, W. H. 1899. Synopsis of the Solenidae of North America and the Antilles. Proceedings United States National Museum, 22: 107-112, 1899. Hertlein, L. G., and Strong, A. M. 1950. Eastern Pacific Expeditions of the New York Zoological So- ciety. XLII. MoUusks from the west coast of Mexico and Central America. Part IX. Zoologica (N.Y. Zool. Soc), 35 (4): 217-252, pis. 1-2, 30 Dec. 1950. Jordan, E. K. 1936. The Pleistocene fauna of Magdalena Bay, Lower California. Contributions Department Geology Stanford University, 1 (4): 103-173, pis. 17-19, 13 Nov. 1936. Kelsey, F. W. 1907. Mollusks and brachiopods collected in San Diego, California. Transactions San Diego Society Natural History, 1 (2): 31-55, 1907. Oldroyd, Ida S. 1924. The marine shells of the west coast of North America. Stan- ford University Publications Geological Sciences, 1 ( 1 ) : 1-247, pis. 1-57, 1924. Schumacher, C. F. 1817. Essai dun nouveau systeme des habitations des vers testaces. 4to, i-iv, 1-287, pis. 1-22. Copenhague, 1817. [The genus Eiisis is described on pp. 47, 143.] Smith, A. G., and Gordon, M., Jr. 1948. The marine mollusks and brachiopods of Monterey Bay, Cali- fornia, and vicinity. Proceedings California Academy Sci- ences, ser. 4, 26 (8): 147-245, text figs. 1-4, pis. 3-4, 15 Dec. 1948. WiLLETT, G. 1937. An Upper Pleistocene fauna from the Baldwin Hills, Los Angeles County, California. Transactions San Diego Society Natural History, 8 (30): 379-406, pis. 25-26, 15 Dec. 1937. Williamson, Mrs. M. B. 1892. An annotated list of the shells of San Pedro Bay and vicinity. Proceedings United States National Museum, 15: 179-219, pis. 19-23, 1892. TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 16, pp. 405-428, Plates 28-29, Figs. 1-10 NOTICES OF NEW WEST AMERICAN MARINE MOLLUSCA m%. COMP. ZQOL LIBRARY SEP 16 1953 HARVARD UtIfVERSlTY By S. Stillman Berry Redlands, California SAN DIEGO, CALIFORNIA Printed for the Society September 1, 1953 COMMITTEE ON PUBLICATION Joshua L. Baily Carl L. Hubbs Laurence M. Klauber Charles C. Haines m^. cos^p. zooL LIBRARY SEP 16 1953 HARVj^RD UtIIVERSITY NOTICES OF NEW WEST AMERICAN MARINE MOLLUSCA BY S. Stillman Berry Redlands, California Any long continued regional study of a fauna is likely to result in the gradual accumulation of currently unidentifiable specimens, some of which eventually demand publication as new. It is in the effort to clear away a few such odds and ends and to make possible intelligible reference to some of them in other work now impending that the ensuing paper has evolved. The eleven molluscan species to be described include several which have long been known but have been quite as long rather thor- oughly misunderstood, among them a most interesting new mussel (Volselhi), a Diplodo)7ta. and a fine new N as sarins. Of special interest among the remaining species, all of which are gastropods, are an exquisite oflF-shore Oceiiebra. a beautiful olive-shell {Agarouia) from the tropical fauna, a splendid species of the striking turrid genus Knejastia. and two minute but hardly less remarkable tectibranchs, each of these last consti- tuting the type of a new genus. In addition to the various collectors mentioned in connection with the specimens so generously furnished me by them, my very appreciative thanks are likewise due to Dr. Harald A. Rehder and his associates of the Division of MoUusks, United States National Museum, to Dr. A. Myra Keen, Department of Paleontology, Stanford University, and to Col. Arthur F. Fischer and Dr. Joshua L. Baily, Jr. of the San Diego Museum of Natural History for aflfording me every possible facility in the study and in some instances the utilization of relevant comparative material to be found in the collections under their care. Further grateful expression of obligation is due to Dr. G. D. Harris and Dr. Katherine V. W. Palmer of the Paleontological Research Institution, Ithaca, and especially to my good friend Mr. Edgar R. Fisher of Redlands for making the photo- graphs used in the accompanying plates. 1. Volsella sacculifer new species PI. 28, figs. I, 2, Text-fig. 1. Descriptiou: Shell of but moderate size for the genus, in outline broadly almond-shaped; highest at about the mid-point, thin, smooth, moderately inflated; hinge-line nearly straight to very slightly arcuate. Valves well rounded behind, more or less distinctly swollen ventrally in the byssal region and produced abruptly into a small obtuse lobe-like flare 408 San Dii:go Sociktv of Natural History' or pocket just under the umbones, the rounded angle of this pocket form- ing the anterior end of the shell; postero-dorsal area subalate, its angle obtuse. Hinge toothless except for a short, sharply conical, posteriorly directed process set off by a notch at the anterior insertion of the liga- ment. Periostracum light to deep brown (near Buckthorn Brown to Mummy Brown), smooth, polished, under the more or less dehiscent traces of some fairly considerable posterior shagginess. Measurements: Max. long, mm. Chace Coll. 1789 61.7 Paratype 20654 40.1 Paratype USNM 39.1 Holotype 38.0 Holotype: To be deposited in the collection of Stanford University, No. 7853. Paralypes: Cat. No. 20654 Berry Collection, together with a small series in the Oldroyd Collection in the Department of Paleontology, Stan- Max. alt. Diam. mm. mm. 38.2 24.5 23.2 17.6 22.8 17.6 20.0 15.8 Fig. 1. Vohella sacculijer n. sp. Camera .sketch of anterior portion of right valve of paratype 20654 to show anterior pouting and its relationship to umbo and hinge. ford University, besides one to be deposited in the collection of the United States National Museum. Type-Locality: San Pedro Harbor, California (Los Angeles County Museum collectors and others). Additional Locality: Long Beach, California; one large complete specimen washed in from deep water; Emery P. Chace no. 1789. Commentary: This species has been known and neglected for a number of years. It does not appear to be very common, but occasional specimens are to be seen in local collections, almost invariably having been collected in San Pedro Harbor or its near vicinity, and usually mas- querading under the name of the very different V. fornicata (Carpenter). The byssal swelling varies greatly in strength in different individuals (it is rather weak in the shell which happened to be selected as the holotype), but the anterior pouting is both a constant and characteristic feature which Berry — Notices of New West American Marine Mollusca 409 enables this species to be distinguished at a glance from any other pres- ently known west American member of the genus. It is probably not an animal of littoral habit, but a dweller in somewhat moderate depths off- shore, since almost all the specimens seen were taken at the time ot the major dredging operations in San Pedro Bay and few if any of them appear to have been picked up alive. It is not probable that the species is anywhere either abundant enough or sufficiently available to be of impor- tance as food, yet in its proper habitat it may occur more numerously than we at present know. The Long Beach example in the Chace Collection is far larger than any other examined and retains more of the pristine shag- giness.^ The specific name is derived from the L. sciccidiis, pouch, + -fer, bearer, and has reference to the peculiar sub-umbonal pouting. 2. Diplodonta impolita new species PL 28, figs. 3, 4. Text-fig. 2. Descyiplion: Shell of medium size, thin, crude, dull and more or less chalky in texture, subrotund, a trifle longer than high, tumid, equi- valve, distinctly oblique, the anterior end shorter but more narrowly rounded and projecting than the posterior. Surface pale brown, rough- ened by the coarse growth-striae which are lower and more distant on the earlier portion, but abruptly both more acute and more crowded during the subsequent half of development. Beaks narrow, bent strongly for- ward, quite angular and pointed for a D'/plodonta, the area in front somewhat excavated. Hinge plate and teeth small; posterior right cardinal strongly divided by a sharp groove throughout its length, each segment strongly and rather acutely cusped on its inner side; anterior right car- Fig. 2. D!l>locionUt tnipoltta n. sp. Interior view of umbonal region of right valve; holotype; camera outline. dinal smaller, simple, arising close under the valve margin and projecting but little below the supporting plate, its cusp either rounded or obtusely angular; anterior left cardinal simple, elongate, very oblique; posterior left cardinal duplex, little projecting ventrally, nearly vertical in align- ^During the printing of this paper and thus unfortunately too late to make it the type-series, a suite of three superb examples of this species were kindly sent to me by Dr. Howard R. Hill of the Los Angeles County Museum (Berry Coll. 18367). These were dredged alive in 25 fathoms, off Santa Catalina Id., Califor- nia, in June 1927 by the late George Willett. The largest measures: long. 74.8, alt. 40.4, diam. 33.6 mm. 410 San Diego Socii-tv of Natural History ment; hinge plate short, continuing, especially in front, as a low ridge running along the slopes of the valve close to the somewhat beveled and even slightly excavated margin; ligamental plate narrowly angled or ridged on its inner aspect, anterior adductor-scar subpyriform, concave or indented on the inner margin; posterior scar more ovate and without emargination. Periostracum thin, light brown, largely worn away in adult. Measurements: Largest paratype, long. 26.0, alt. 25.0, diam. (depth) 8.8 mm.; holotype, long. 23.0, alt. 21.2, diam. 8.6 mm. Holotype: To be deposited in the type collection of the Department of Paleontology, Stanford University, No. 7854. Paratypes: Cat. No. 3658 Berry Collection; others to be deposited in the collections of the San Diego Society of Natural History, the United States National Museum, and the Museum of Comparative Zool- ogy of Harvard College. Type-Locality: 15 fathoms, off Forrester Id., Alaska; dredged by George Willett, July-Aug., 1916. Commentary: This is the species which Willett (1918: 68) re- ported from Forrester Island as D. ovbella (Gould). He noted that it is "much less globose than California specimens," and indeed upon closer Fig. 3. Diplodonta orbellus (Gould). Interior view of umbonal region of right valve of specimen from San Pedro, California; no. 160, Berry Coll.; camera outline. study it appears clearly distinct from them specifically, differing not only in the very much narrower, more pointed, and less projecting umbones, but also in the earthy texture, the heavier incremental striae, the relatively minute hinge-plate, the smaller, more obtuse, and less projecting cardinals, and other more minor features. It may bear some relationship to D. aleiitua Dall (1901: 820), but this I have not seen. It would seem a distinctly less globose and more elongate species than either D. impolita or D. orbellus r I suspect that nearly or quite all of the more northern published records under Gould's name will prove referable to the present species. -As the Latin orl)is is masculine in gender, its diminutive should logically be spelled orbellus, for which Gould's use of o.bella might be regarded as a lapsus calami. We could do with either a sharper or a more inclusive definition of what properly constitutes lapsus calami within the purview of the International Code. Berry — Notices of New West American Marine Mollusca 411 Unfortunately all of the mature shells are incomplete and represent the same valve, the right. The specific name is the L. iinpolitiis. unpolished, and refers to the crude, inelegant appearance of the outer surface of the shell. 3. Lacuna succinea new species Text-fig. 4. Description: Shell small, thin, narrovidy acutely conic, umbilicate; spire attenuate, produced. Whorls about 6.3, rapidly descending and en- larging, moderately convex above the strongly corded peripheral carina which nearly always remains visible just above the rather deep suture almost from its inception; nepionic whorls at first whitish, smooth, pol- ished, eventually becoming slightly roughened by the increasingly strong incremental lines and the faint beginnings of spiral sculpture; later whorls corneous, translucent, strongly and copiously spirally striate; base con- cavely flattened. Aperture large, sub-pyriform or semi-lunar, acute poste- riorly, broadly rounded in front; lip thin, very sharp, efi^use anteriorly, very slightly everted near its junction with the parietal wall just below the carina, whilst in front continuing smoothly past the columella into the threaded keel which bounds the very open, crescentic, trimly reamed umbilical groove; columella narrow, solid, calloused, whitish, for the most part nearly straight, its attenuate lower third curving into the aperture. Color of shell usually a nearly uniform Deep Olive-Buff to Wood Brown without definite markings or mottlings, the umbilical keel a little Fig. 4. L^uunj succinea n. sp. Camera outline of holotype. darker, but the larger shells occasionally showing on the body-whorl a few dim brownish oblique lines, especially visible near the carina as it approaches the lip. Meas/ireme>7ts: Holotype, alt. 8.9, max. diam. 5.5, alt. aperture 4.7, diam. aperture 2.6, max. diam. umbilical chink from carina to inner face of columella 0.8, max. inner diam. of chink 0.4 mm. 4i: San Diego Society of Natural History Holotype: Cat. No. 11983 Berry Collection. Paratypes: Cat. No. 12886 Berry Collection; others to be deposited in the collections of the San Diego Museum of Natural History, Stanford University, and the United States National Museum. Type-Locality: San Pedro, California; Emery P. and Elsie M. Chace. Material Exam ined : Number 'specimens Locality Collector Date Vf^here Deposited 2 Avalon, Santa Cata- lina Id., Calif. S. S. Berry 1903 Berry Coll. 1424 2 Aich Beach, Orange Co., Calif. H. Grace Eaton " 9181 6 1 San Diego, Calif. San Pedro, Calif. F. W. Kelsey E. P. & E. M. Chace " 1484 11983 Holotype 10 Paratypes 10 •• Chace Coll. Paratypes 2 " San Diego Mus. Paratypes 1 ■' Stanford Coll. Paratypes 2 U.S.N.M. Paratypes Remarks: The determination of this pretty and altogether distinct little species has bothered me for some time. It clearly has nothing to do with the boreal group of L. soHd//la Loven and its allies, although I have several times seen it under the name solidula in collections. I have also seen it labeled aurantiaca Carpenter (1864: 656) despite that having been succinctly described as "keel obsolete, resembling the chinked Phasianel- lae" and hence evidently something quite different from the present species. A third name which I have seen applied to it is cariftata Gould (1848: 75), but that is a northern species and has a much more robust shell (Gould called it "ovato-globosa") and the extensive material I have seen does not lead me to the conviction that the two are in any way closely related. To my reading Carpenter's diagnosis of his L. com pacta (1864a: 429) is more suggestive of sue cine a than any of the three species afore- mentioned. However, his use of the word "compact," his failure to men- tion the characteristically drawn-out spire, and his description of the columella as "vix lacunata," together with the fact that compact a was a segregate from his variable Neah Bay series, whence he described a total of no less than five species and varieties of this genus, afford good evi- dence that he could not have been considering this southern species. The color matching I have given is from a shell taken with the ani- mal. Empty beach shells are often quite a clean bright amber in hue, hence the specific name, which is the L. siiccineiis, — of amber. 4. Turritella orthosymmetra new species Plate 28, fig. 5. Diagnosis: Shell of moderate size, rather thin; spire compactly coiled, elongate-conic, with straight smooth slopes; apical angle narrowly Berry — Notices of New West American Marine Mollusca 413 acute, showing no marked change with advancing growth; suture distinct, very narrowly channeled. Sculpture of early post-embryonic whorls com- prising a strong peripheral keel slightly below the middle, a weaker one a little below the upper suture, and very soon a third immediately above the lower suture, which last increases in strength until on the final whorl it forms a sharp carina which bounds the concavely flattened base; a smaller secondary rib also develops between the uppermost major rib and the suture, while cjn all the later whorls a fine somewhat wavy spiral threading covers all the surface between the major spirals and also extends over the base, the number of threads on the base varying from about 12 to 20 in addition to a few weak intercalaries. Aperture asymmetrically quadrate, slightly widest anteriorly, strongly angled and somewhat pinched where intersected by the basal keel; outer lip thin, sharp, usually more or less broken away; columella narrow, arcuate, slightly twisted, the basal lip very weakly everted where it adjoins it, sometimes appearing to form the bare rudiment of a canal; columellar callus rounding out over the base somewhat in advance of the lip. Color of shells always considerably obscured and dulled by a grayish deposit or bloom, but apparently fairly near Clay Color, more or less clouded with Army Brown and related tones. Measurements: Holotype, alt. 30. 0+, max. basal diam. 8.5, alt. aperture 5.0, diam. aperture 5.0 mm. Holotype: Cat. No. 11,979 Berry Collection. PcVcitypes: Cat. No. 502 Berry Collection, others to be deposited in the collections of Stanford University, United States National Museum, and San Diego Museum of Natural History. Type-Locality: Off Pebbly Beach, Santa Catalina Id., California; dredged in 50 fathoms by J. H. Paine and S. S. Berry, 31 Aug.-l Sept. 1903. Remarks: This beautiful Turr/tella has been known to collectors for a long time, but I can not find that any valid name has ever been applied to it. It has probably most frequently been labeled T. jewetti} Carpenter (1866: 276), but the description of that enigmatic and perhaps indeter- minable fossil species is no very close fit for the shells of this living race, especially the description of the base as "parum angulata," and I believe we are well advised to follow Merri^am (1941: 123) in regarding jeivettii either as a variant of cooperi Carpenter, or still better as a mere species inquirenda until its true identity is one day definitely established. Al- though apparently, from its plan of sculpture, allied to cooperi (Car- penter 1864: 612, 655), the present species is sharply distinguished by its less attenuate, evenly geometric, flat-sided, tightly coiled spire, acute basal keel, quadrate angled aperture, and definite spiral sculpture on the base. In form it is much more like T. pedroensis "Applin" Merriam (1941: 121, pi. 35, figs. 1-9), but is smaller and the shell is much less heavily sculptured. The species was a common one at the type-locality and Paine at one time had quite a number of the shells. I have seen few of them from other dredgings. 414 San Diego Society of Natural History The specific name is derived from the Gr. orthos, correctly, and symmetros. symmetrical, and is descriptive of the trimly aligned spire. 5. Ocenebra crispatissima new species PI. 28, fig. 6. Description: Shell rather small, thin, somewhat broadly fusiform, with a large, rounded body-whorl; spire short, turreted, acutely conical. Whorls (counting suturally) about seven (apex missing in holotype) ; larval shell small, high, straight-sided, sharply carinately tabulate and inward sloping above, nearly smooth, with about one and a quarter whorls; succeeding whorls more narrowly and flatly tabulate, abruptly bearing a small spiral cord on the shoulder, a slightly weaker one a little below it, and soon with traces of a third still weaker one appearing now and then in the suture, these cords at first smooth but soon becoming roughened by the appearance of fine incremental lirations, following which additional spiral cords gradually appear, the original two cords nevertheless maintaining their primacy throughout, until back of the aper- ture there may be as many as 14 or 15 cords of varying strength sepa- rated by considerably narrower, deeply squarish interspaces, beyond which anteriorly a few more similar spirals gradually fade away on the canal; simultaneously the incremental threads have likewise been rapidly increas- ing both in strength and number until on the last whorl they form a close, erect, crispate, laminate imbrication over-riding both the axial ridges, on the summit of which, most especially on the shoulder spiral, it attains its greatest and almost spinose emphasis. Aperture about three- tenths as long as the shell, ovate-pyriform, acute posteriorly; axial threads corresponding to the outer laminations continuous and closely crowded around the inner lip to separate it from the columellar wall and render the peritreme complete, eventually terminating on the roof of the canal; outer wall of aperture dentate with about five teeth, including one small tubercle situated a little above the middle, a second one so weak as to be hardly more than a trace, then two somewhat stronger ones, and finally a very heavy tubercle at the entrance of the canal. Canal a little longer than the aperture, narrow, nearly straight, gently recurved, and covered for all but the anterior third of its length. Color cream-white outwardly, the interior polished and warmly flesh-tinted. Operculum shortly pyriform, the columellar border weakly concave; striae low, rounded, rather heavy, not very crowded. Measure menls of holotype: Alt. 20.9, max. diam. 11.1, alt. aperture 6.1, length canal 7.8 mm. Holotype: To be deposited in the type-collection of the Department of Paleontology, Stanford University, No. 7855. Paratype: An immature shell with perfect apex. Cat. No. 18972 Berry Collection. Type-Locality: 33 fathoms, off Isthmus Cove, Santa Catalina Id., Berry — Notices of New West American Marine Mollusca 415 California; "Zaca" Sta. 23, Stanford-Crocker Expedition (Dr. George S. Myers), 16 Sept. 1938. Additional Matericd: A single adult shell from 15 fathoms, Avalon Harbor, Santa Catalina Id., California (Berry Coll. 8945). Commentary: For an Ocenebra this is a shell of unusual refinement and beauty, quite as lovely as some of the northern "trophons" and indeed recalling them in nearly every way except the more robust body-whorl. In its own genus nothing appears to demand particular comparison with it. It probably belongs in the group with O. harharensis (Gabb) and O. squam/dijera Carpenter, from both of which it differs in the smaller size, thin shell, relatively low spire, long and very narrow canal, very sharp sculpturing, and complete lack of any brown coloring. The specific name is the superlative of the L. crispatus. curled or crimped, and has reference to the incremental lamination. 6. Nassarius (Schizopyga) rhinetes new species Plate 28, fig. 7. Diagnosis: Shell quite large, thin, high-conic, with an acute apex, the slope of the spire distinctly arcuate; whorls about 8, the later ones quite inflated, with rounded profile and only the weakest suggestion of a very narrow shouldering; suture distinctly impressed and deep. Sculpture, subsequent to the abraded apical whorls, developed first as about 14 rather strong, nearly straight axial ridges, soon becoming somewhat re- tractively slanting, relatively finer and more crowded, and on the last whorl weakly arcuate or hypo-sigmoid, here increasing in number to about 30; spiral sculpture appearing soon after the axial; on the spire about 6, on the fully exposed last whorl about 12 cords, not all of them of equal strength, the first cord anterior to the suture being distinctly more slender and the second cord more emphasized than succeeding ones, while the 10th and 11th cords just posterior to the columellar fossa are again somewhat emphasized, while the 12th cord, which runs along the poste- rior wall of the fossa itself, is little more than a strong thread; cords 1 to 10 form distinctly squarish interspaces with the axial ribs, with the intersections rather sharply tuberculate; tubercles slightly emphasized at the shoulder along cord 2, and reduced along cord 11, while cord 12 and the 5 crowded ribs of the wall^of the canal anterior to the fossa are non-tuberculate. Aperture ample, about 51% of the total altitude, ovate-pyriform, truncate in front; outer lip thin, its inner surface marked by slender shallow spiral grooves corresponding in position to the cords of the exterior; inner lip strongly concave in profile, covered by a thin, smooth, whitish callus which extends well in front of the aperture, but barely passes the sutural line and is not very sharply bounded; columella strongly twisted and incurved, weakly biplicate near its termination; canal open, deep, its outer margin reflexed against the rather narrow sharply excavated fossa. Color of shell whitish, unhanded, but with traces of a thin, axially laminated, semi-dehiscent, pale yellowish brown periostracum between the costae. 416 San Diego Socm-ty of Natural History Measuvemeuh: Holotype — alt. 32.5, max. diam. 18.9, alt. aperture 16.6, diam. aperture (opening only) 8.8, diam. aperture to edge of callus, 11.2 mm. Holotype: Cat. No. 1182 Berry Collection. Pitratypes: Cat. No. 1163 Berry Collection. Type-Loccdity: Dredged in 40 fathoms, off Moss Landing, Mon- terey Bay, California; mud bottom; S. S. Berry, June 1906. Remarks: This beautiful species has been held in MS. for some time. It has long been known to Californian students as one of the strange congeries of forms which has passed under the name Nassa cali- fornicma (Conrad), the present type-material having been reported under this name by myself (1907: 40) shortly after its collection. And to avoid misunderstanding, it should be admitted at once that if Conrad's name can properly be associated with any living species, the subject in hand is the likeliest candidate for selection. Unfortunately the original descrip- tion of calijorniana is at once so brief and so generalized as to be nearly worthless, while the almost equally inadequate figure shows only two whorls, the spire represented by a restoration in outline. While the figure is about the same size as adult Recent shells and the reconstituted spire is rather startlingly close to the contours they exhibit, there exist otherwise what appear to me as quite troublesome discrepancies. There is shown no trace whatever of any parietal callus, no internal lirations, the sculpture is indicated as quite closely cancellate rather than ribbed, and while its delineation makes it appear nearly as much like an injury as a normal structure, the fossa is in profile both considerably more open and more "profound." Fortunately at least one fossil form has been reported and figured, which, in spite of the indefinite status of Conrad's type-horizon, is probably fairly near it in the time-scale, and is likewise more or less in accord with it in the characters just noted. This is the Nassa calif orniana recorded by Arnold (1908: pi. 36, fig. 6) from the Pliocene south-west of Capitola in Santa Cruz County, which I am accordingly inclined to accept as reasonably identified, enabling us to fix, far better than any other published evidence, Conrad's enigmatic name. Arnold gives no description, but the clean-cut photographic figure represents the shell of a species obviously of the same intimate group as the living one, but differing in its decidedly wider outline, more finely cancellate sculpture, the wide smooth fossa, and at least the appearance of a rather remarkable excavation of the columella, the importance of which is difficult to esti- mate without examination of actual specimens. Of other living species N. perpinguis (Hinds) seems about the closest but is much smaller, spiral cords are appreciably more numerous on the later whorls, and there are so many other differences of detail that only through the utmost carelessness could the two species readily be confused. The specific name chosen is from the Gr. rh'iuetes, one who rasps or files, and refers to the rasp-like sculpturing of the shell. Berry — Notices of New West American Marine Mollusca 417 7. Agaronia murrha new species PI. 29, % 1. Text-fig. 5. Descnptioir. Shell of moderate size, rather heavy at maturity, ovate- fusiform, widest near middle, the spire tapering rapidly and on the lip side somewhat convexly to the minutely mammillate apex; anterior ex- tremity truncate. Whorls 5 or a trifle more, rapidly increasing, the body- whorl convex and including about 90% of the total length of the shell. Suture sharp and strongly but not very deeply channeled. Aperture ample, sharply angled posteriorly, about three-fourths as high as the shell; outer lip sharp until maturity when it becomes reinforced inwardly by a whitish callus which becomes quite heavy approaching the suture; outer anterior lobe subangular, slightly surpassing the columella; canal open, moderately wide; parietal wall covered by a sharply delimited white callus which is thickened above the suture posteriorly and thence extends in most speci- mens completely across the whorls to the suture above so that the spire is completely calloused; anteriorly the callus passes over the columella to the canal. Columella with a single strong fold at the canal which is bounded by a rather sharp and deep channel; behind this appears a rather complex system of low folds, — first, two or three outer ones parallel to the callus-margin, second, entering against these at a low angle and sup- planting them toward the aperture are three or four more slender costae, and third, an apertural series of low less sharply ascending folds which are of rather irregular development and strength; fasciole hardly elevated. Surface smooth except for extremely weak traces of a microscopic spiral striation and the growth lines, the latter becoming stronger in nearing the aperture; parietal callus and fasciole microscopically wrinkly-punctate. Color of exterior of adolescent or mature shells everywhere a slightly grayish porcelain-white, the outer wall of the interior deep brownish except for the calloused margin of the lip which is white; very young shells light brownish gray with an indistinct yellowish zone below the suture, a rather sharp white band on the anterior part of the body-whorl, and a brown-bordered white zone marking the fasciole. Mecisurements of largest paratype, alt. 38.5 mm., max. diam. 16.6, alt. aperture 31.8 mm., no. whorls 5; of holotype, alt. 36.3, max. diam. 15.4, alt. aperture 29.3 mm., no. whorls 5+. Holotype: San Diego Museum of Natural History. Paratypes: Cat. No. 16431 Berry Collection; others to be deposited in the collections of the United States National Museum, the Department of Paleontology of Stanford University, the Museum of Comparative Zoology of Harvard College, and the private collection of Dr. Joshua L. Baily, Jr. Type-Locality: Cprinto, Nicaragua (Herbert N. Lowe, 1931). Commentary: In critically reviewing the available West American specimens of Agaronia with a view to correlating their distribution, it shortly became apparent that two quite different forms are involved 418 San Diego Society of Natural History whereas only a single species, the widespread A. testacea (Lamarck), has been hitherto recognized from the region. Since an exhaustive search of available literature has revealed no lost name which can be revived for the rarer form, a name is here supplied for it and tentatively given specific Fig. 5. Aga/onia miirrha n. sp. Camera outline of apex of paratype 16431a. rank. It differs from A. testacea in detail at almost all points, but adults may perhaps be most easily separated by 1 ) the low spire, slightly convex on the apertural side; 2) the long aperture; 3) the constricted and rela- tively shallow sutural channel; 4) the more evenly ovate outline; 5) the extension of the parietal callus across the whorl to the suture with a con- sequent smoothly calloused spire; 6) the relatively large and more mam- Fig. 6. Agaronia testacea (Lamarck). Camera outline of apex of shell from Cholla Cove, Sonera; no. 15122a, Berry Coll.; same scale as Fig. 5. millate nepionic shell; 7) the closely wrinkled inner lip, with its folds oblique to the thin major plaits; and 8) the barely raised fasciole. To these might be added the opaque whitish-porcelain coloring of the exte- rior and the inside margin of the outer lip, besides the warm deep brown of the interior without any trace of the usual bluish-gray of A. testacea. except that it must be borne in mind that the Olividae are a family in which extreme colored variants are of the commonest occurrence and these differences, conspicuous though they are, may turn out to be more Bhrrv— Notices of Ni-:w West American Marine Mollusca 419 characteristic of the particular colony whence the type-series was taken than of the species as a whole, although some describable difference in the respective color ranges of the two species is quite certainly to be anticipated. Until then such characters as those provided by the calloused spire and the nepionic shell should provide more certain criteria for sepa- ration than mere color, but this can be better told when larger series become available for comparison. Although the present species appar- ently lacks the lively coloring exhibited by A. teslcicea in its best forms and is not known to approach it in size, it is nevertheless attractive in a modest way. '■ The specific name chosen is the L. Du/rrhci. porcelain. 8. Antipianes (Rectiplanes) willetti new species PI. 29, fig. 2. Description: Shell elongate fusiform, with tall, sharply conic spire; whorls 8, convex, each appressed closely against the whorl above, the suture distinct; nepionic whorls two, mammillate, their inflation insuffi- cient to alter appreciably the symmetrical outline of the apex, polished, the first smooth, the second developing a few weak traces of spiral sculp- ture on its concluding portion, which develops into a system of about 10 fine spiral threads on the next succeeding whorl, and increases to about 20 gradually stronger but still very weak threads on the penultimate whorl and many more on the body-whorl, where they continue to the canal, those in the fasciolar area being excessively fine, crowded, and numerous; axial sculpture absent except for the numerous growth lines of rather uneven strength. Aperture elongate-pyriform, about 40% of the height of the shell; outer lip thin, sharp, strongly arcuately produced; inner lip and columella smooth, arcuate; canal open, short, slightly re- curved; anal sulcus wide and deep, passing smoothly into the produced segment of the lip; fasciole not sharply delimited, marked mainly by the fineness of the sculpture. Measurements: Holotype — alt. 18.4, maj. diam. 6.6, alt. aperture 7.0, diam. aperture 2.2 mm. Paratype — alt. 17.8, maj. diam. 6.7, alt. aper- ture 6.3, diam. aperture 2.0 mm. Holotype: Cat. No. 11,977 Berry Collection. Paratypes: Cat. No. 3843 Berry Collection, and the collection of the San Diego Museum of Natural History, No. 22856. Type-Locality: 50 fathoms, off Forrester Island, S.W. Alaska; dredged by George Willett, July 1917. Remarks: This very attractive little species has been previously rec- ognized under the name Crassispira rot/da (Willett 1919: 21). The latter species was originally described from fossil material under the name ^Too late for consideration except in a footnote, there was called to my atten- tion a large series of a small dark Agaronia in the San Diego Museum taken in 1931 by H. N. Lowe at San Juan del Sur, Nicaragua. These shells are mostly of purplish-gray coloring with a deep brown (rarely light yellowish-brown) apex and fasciole, and appear to represent a dark phase of the species here described. 420 San Diego Society of Natural History Ple/iyoloiud {Spirotropsis*) smith/ Arnold (1903: 216, pi. 6. %. 13), but the specific name having been preoccupied in Pleurotoma more than once, it was later renamed rotula by Dall (1921: 71). To this the north- ern living species is admittedly quite close, but differs from it according to the material seen 1 ) in being much smaller, with fewer whorls and a shorter canal; 2) in the distinct spiral sculpture, which in the fossil is either entirely absent as described by Arnold, or represented by some nearly obsolete traces on the anterior part of the body-whorl as shown in one of my Hilltop Quarry specimens; and, perhaps most importantly of all, 3) by the very much smaller and less bulbous nepionic shell. I have found A. rotula. or what I at present interpret as that species, only twice at Hilltop Quarry, so it would seem of relatively infrequent occurrence there, although the specimens are beautifully preserved. The holotype of rotula came from Arnold's supposed "Pliocene" of Dead Man's Island, now generally considered as the equivalent of the lower Pleistocene at Timm's Point. The nepionic shell in the fossils I have seen is similar in general form and structure to that of ivilletti, but is greatly larger and somewhat more inflated, rendering the apex distinctly obtuse. A. rotula is one of the extraordinary melange of species placed by Grant and Gale (1931: 553-4) in the synonymy of perversa (Gabb), but I am wholly at a loss to comprehend how anyone with both this and perversa in hand could possibly confound them, their characterization being sufficiently ample quite aside from the mere direction of coil, a prejudice against which as a taxonomic anchor appears to have blinded these authors to everything else. The species is dedicated to its disco\'erer, the late George Willett, of the Los Angeles Museum. 9. Knefastia princeps new species PI. 29, fig. 3. Text-fig. 7. Descr/pt/oir. Shell moderately large, elongate, fusiform; larval shell decollated in holotype but persisting whorls 81/^, turreted, strongly con- vex, sloping above to a high, subangulate shoulder; a flattened fold-like ridge strongly appressed against the suture is subtended by a shallow spiral furrow in the rather wide anal fasciole, the latter sculptured other- wise mainly by the very strong and coarse incremental striae and a few weak traces of low spirals; body of whorl marked by from 7 to 8 massive axial ribs (there are 7 on the last whorl) which are slightly knobbed on the shoulder, and 4 low but strong spiral ridges which over-ride both the axial ribs and their interspaces, the two central ridges being a trifle more widely separated than either is from its outer neighbor; base with about 13 or 14 gradually weakening spirals; the axial ribs pass onto the base but become obsolescent in the region of the canal. Aperture unarmored, about 45%' the length of the shell, elongate-pyriform, acute posteriorly; outer lip sharp-edged, hardly crenulated by the spiral ridges, its margin convex in front of the strong open anal notch M'hich nearly subtends the ^Apparently an inadvertence for Spirotropis Sars 1878. Berry — Notici;s of Nkw West American Marine Mollusca 421 end of the suture. Canal open and fairly long. Columellar wall some- what erose parietally, covered with a moderate wash of shining enamel. Periostracum shining, the spire and main portion of body-whorl lustrous Antique Brown, paling anteriorly and on the nodes to Raw Fig. 7. Knefastij prniceps n. sp. Camera sketch io show profile outline of lip. Umber, the spiral ridges sharply paler and the anterior portion of the shell as well, even to Chamois at the extreme portion and on the spirals, with one or two darker bands under the periphery. Interior polished and bright, near Cinnamon to Pinkish Cinnamon. Operculum a little smaller than the aperture, acute in front, more blunt posteriorly, with a shallow furrow running parallel to the inner margin. Measurements of holotype: Alt. 60. 3+, max. diam. 19.9, alt. aper- ture 28.0, max. diam. aperture ca. 9.2 mm. Holotype: Cat. No. 13949 Berry Collection. Type-Locality: 20 fathoms; pebble, mud and shell bottom; 2 miles N. of Cedros Village, Cedros Id., Baja California; Kenyon-Williams Mexican Expedition, Sta. 17B, 14 May 1946. Commentary: This magnificent species, one of the finest in a gen- erally striking genus, appears nearest to K. dalli Bartsch (1944: 28) among described forms, differing in its larger size, more slender form, the relatively longer and attenuate canal, the slender and well-spaced spiral cords, the very much fewer and less nodulous axial ribs, the more open sinus, and the warmer and brighter coloring, especially of the inte- rior. The single example taken was dredged up alive. The specific name is the L. princeps. the first in rank or most distin- guished. 422 San Dikgo Society of Natural History 10. WOODBRIDGEA genus Diagnosis: Shell minute, thin, transclucent, inflated, obliquely bulli- form; whorls nearly embracing, rather narrowly planate above, rapidly expanding to the greatly inflated body-whorl; aperture wide, ear-shaped, strongly and obliquely expanded in front; columella unarmed; surface strongly spirally striate. Type: the following species. new species pec Woodbridgea williamsi Text-fig. 8. Description: Shell minute, very thin, subvitreous, whitish, translu- cent; somewhat suggesting a lop-sided Hcvuinoea in form; whorls closely embracing and very rapidly expanding from the minute, planate, barely exposed apex to the greatly expanded body-whorl. Aperture ample, ear- shaped, strongly oblique to the shell-axis, narrow and sub-acute poste- riorly, widely produced in front; outer lip thin, sharp, forming a shallow but distinct sinus posteriorly, whence it descends slantingly to its inser- tion; inner lip weakly calloused, sigmoid, the columellar portion nearly straight, slender, in front very narrowly flattened or possibly even sulcate Fig. 8. Woodbridgea williamsi n.g. spiral sculpture merely indicated. & sp. Camera sketch of holotype; between the rather solid appearing pillar and the thin margin of the lip, the precise formation being difficult to see and interpret clearly in so tiny and transparent a shell. Surface covered with numerous conspicuous sharp spiral striations fanning out from the aperture in two series, those above the periphery trending parallel to the suture and those below the peri- phery descending very much more rapidly, especially just in front of the columella. Measuremenls of the unique holotype: Alt. 1.33, max. diam. 0.96, alt. aperture 1.22, max. diam. aperture 0.71 mm. Holotype: Cat. No. 13907 Berry Collection. Type-Locality: 25 fathoms, fine brown sand and mud, % mile off Cedros Village, Cedros Id., Baja California; one example, Kenyon-Wil- liams Mexican Expedition, Sta. 17 A, 14 May 1946. Berry — Notices of New West American Marine Mollusca 423 Commentary: I know nothing sufficiently similar to this minute but very beautiful little sea-snail to render fruitful any precise comparisons. Until something can become known of the animal its nearer affinities can be hardly more than surmised, but from the form and texture of the shell one can scarcely doubt that its general relationships are tectibranch. In some respects the Scaphandridae are suggested, in others the Diaphaui- dae or the Philinidae. Both the species and its genus are dedicated to its assiduous col- lector, Mr. M. Woodbridge Williams of Inverness, California. 11. MICRAENIGMA new genus Diagnosis: Shell minute, white, translucent, almost exactly like a very minute Bulla in form, but imperforate both apically and basally, and unique in that the columella is provided with a strong downward-slanting tooth-like plate. Type: the following species. The name is derived from the Gr. mikros. small, + ainigma, diffi- culty or riddle, and refers to the development of the large columellar tooth, anomalous in the group. Micraenigma oxystoma new species Text figs. 9, 10. Description: Shell minute, milky white, moderately translucent, evenly ovoid, buUiform, widest near middle, about 63% 3.s wide as long; whorls embracing, but merely flattened at summit without the formation of an apical pit, imperforate basally; aperture elongate-ear-shaped, widest in front, narrowing posteriorly where it appreciably exceeds the spire; outer lip simple, sharp, thin, evenly curved; parietal wall less strongly Fig. 9. Micraenigma oxystoma n.g. & sp. Camera outline of holotype 424 San Diego Society of Natural History convex than the outer, covered with a restricted Hght callus, and curving back rather sharply into the short straight pillar-like columella, which bears a heavy and very long, downward and slightly backward slanting, triangular, plate-like, somewhat penetrating and ascending tooth, which is narrow and sharply conic in front view with a slight twist. Surface smooth except for the very fine microscopic incremental striation, some traces of a very delicate axial striation (perhaps only an emphasis of the Fig. 10. Micraenigfna oxy stoma n.g. & sp. Camera outline of holotype, looking obliquely into the aperture to show the full length of the columellar tooth ; same scale as Fig. 9. first) on the upper moiety of the whorl, and four or five fine spiral striae on the otherwise ill-defined columellar fascicle. Measurements of holotype: Alt. 1.78, max. diam. 1.13, max. diam. aperture 0.51, max. long, tooth (upper side) 0.28 mm. Holotype: Cat. No. 13907 Berry Collection. Type-Locality: 25 fathoms, fine brown sand and mud, % mile off Cedros Village, Cedros Id., Baja California; one example, Kenyon-Wil- liams Mexican Expedition, Sta. 17A, 14 May 1946. Commentary: To judge from the texture, form, and general appear- ance of the shell there seems little doubt that we have in this pleasant little species another tectibranch, yet, so far as I can discover, the heavy basal tooth is a feature unique in the entire order, although in some groups (such as the Reti/sidae. to which family perhaps M/craen/gma may be tentatively referred until the animal is known) a low columellar fold may at times appear which from its position would seem clearly homologous with the tooth. The specific name chosen is derived from the Gr. oxystomos. fanged, and appropriately refers to the apertural tooth. Berry— Notices of New West American Marine Mollusca 425 References Arnold, R. 1903. The paleontology and stratigraphy of the marine Pliocene and Pleistocene of San Pedro, California. Memoirs Califor- nia Academy Sciences, 3: 1-420, pis. 1-37, June 1903. 1908. Descriptions of new Cretaceous and Tertiary fossils from the Santa Cruz Mountains, California. Proceedings United States National Museum, 34: 345-390, pis. 31-37, 8 Aug. 1908. Bartsch, p. 1944. Some notes upon west American turrid moUusks. Proceed- ings Biological Society Washington, 57: 25-29, 28 June 1944.. Berry, S. S. 1907. Molluscan fauna of Monterey Bay, California. Nautilus, 21: 17-22, 34-35, 39-47, 51-52, June-Sept. 1907. Carpenter, P. P. 1864. Supplementary report on the present state of our knowledge with regard to the Mollusca of the west coast of North America Report British Association Advancement Science 1863: 517-686, Aug. 1864. [Repr. in Smiths. Misc. Coll., pp. 3-172, Dec. 1872.] 1864a. Diagnoses of new forms of Mollusca from the Vancouver District. Annals and Magazine of Natural History, ser, 3, 14: 423-429, Dec. 1864. [Repr. in Smiths. Misc. Coll., pp. 233-240, Dec. 1872.] 1866. On the Pleistocene fossils collected by Col. E. Jewett, at Santa Barbara, California; with descriptions of new species. Annals and Magazine of Natural History, ser. 3, 17: 274- 278, Apr. 1866. [Repr. in Smiths. Misc. Coll., pp. 319-325, Dec. 1872.] Dale, W. H. 1901. Synopsis of the Lucinacea and of the American species. Proceedings United States National Museum, 23: 779-833, pis. 39-42, 1901. 1921. Summary of the marine shellbearing mollusks of the north- west coast of America, from San Diego, California, to the Polar Sea, mostly contained in the collection of the United States National Museum, with illustrations of hitherto un- figured species. Bulletin United States National Museum, no. 112, i-iii, 1-217, pis. 1-22, 1921. Gould, A. A. 1848. . . . shells from the collection of the Exploring Expedition. Proceedings Boston Society Natural History, 3: 73-75, Nov. 1848. 426 San Diego Society of Natural History Grant, U. S., IV, and Gale, H. R. 1931. Catalogue of the marine Pliocene and Pleistocene MoUusca of California and adjacent regions, etc. Memoirs San Diego Society Natural History, 1: 1-1036, diag. A-D, Tab. 1-3, text figs. 1-15, pis. 1-32, Nov. 1931. Merriam, C. W. 1941. Fossil Turritellas from the Pacific coast region of North America. University California Publications, Bulletin De- partment Geological Sciences, 26 (1): 1-214, text figs. 1-19, map, pis. 1-41, 8 March 1941. WiLLETT, G. 1918. Notes on the Mollusca of Forrester Island, Alaska. Nau- tilus, 32 (2): 65-69, Oct. 1918. 1919. Mollusca of Forrester Island, Alaska. Univalves. Nautilus, 33 (1): 21-28, July 1919. Explanation of Plates PLATE 28 Fig. 1. Volsella sacculifer n. sp. Exterior view of right valve of holotype, San Pedro Harbor. California; x ca. 1 2/3. Fig. 2. Volsella sacci/l/fer n. sp. Interior view of left valve of holotype; same scale. Fig. 3. Diplodonta itupolita n. sp. Exterior view of right valve, the holotype, from 15 fms., off Forrester Id., Alaska; x ca. 1%. Fig. 4. Diplodoiitci nupolitd n. sp. Interior view of same valve; same scale. Fig. 5. Turritella orthosynnnetra n. sp. Holotype, from 50 fms., off Pebbly Beach, Santa Catalina Id., California; x 2.1. Fig. 6. Ocenehra crispatisshna n. sp. Holotype, from 33 fms., off Isthmus Cove, Santa Catalina Id., California; x 2.9. Fig. 7. Nassarius rhtnetes n. sp. Holotype, from 40 fms., off Moss Landing, Monterey Bay, California; x ca. 1 2/3. PLATE 29 Fig. 1. Agarou'id nmrrhd n. sp. Holotype, Corinto, Nicarague; x ca. 1.6. Fig. 2. Antiplciiies {RectipLines) ivillelt} n. sp. Holotype, from 50 fms., off Forrester Id., Alaska; x 2%. Fig. 3. Knefastia princeps n. sp. Holotype, from 20 fms., off Cedros Id., Baja California; x ca. 1.1. Fig. 4. Ensis calijorniciis Dall. Exterior \ iew of left valve, Mira- mar, Sonora; x 1. Fig. 5. Ens'is uiyrae n. sp. Exterior view of left valve of holotype, San Pedro, California; same scale as preceding. Fig. 6. Eus'is myrae n. sp. Interior of right valve of same; same scale. Berry — Notices of New West American Marine Mollusca Pl. 28 ^ w: ■^- p^^ San Diego Society of Natural History Plate 29 -4 * 2 ■^•S§0^ TRANSACTIONS OF THE SAN DIEGO SOCIETY OF NATURAL HISTORY Volume XI, No. 17, pp. 429-444, map A PLEISTOCENE INVERTEBRATE FAUNA FROM THE SOUTHWEST CORNER OF SAN DIEGO COUNTY, CALIFORNIA^ BY William K. Emerson Museum of Paleontology University of California and Warren O. Addicott Department of Paleontology University of California MUS. CO^P. ZOOL LIBRARY NOV 2 3 1953 HARVARD URiVEUSITY SAN DIEGO, CALIFORNIA Printed for the Society November 10, 1953 ^Contribution from the Museum of Paleontology, University of California, Berkeley, California. COMMITTEE ON PUBLICATION Joshua L. Baily Carl L. Hubbs Laurence M. Klauber Charles C. Haines mz. ccMP. zooL LIBRARY I^OV 2 3 1953 HARVARD UmVEHSITY INTRODUCTION The purpose of this paper is to enumerate the mega-invertebrate fauna collected from a Pleistocene terrace deposit situated in the extreme southwest corner of San Diego County, California. Paleoecologic inter- pretations of the fauna are undertaken and an attempt is made to corre- late this terrace with local terrace deposits. The study is based on a collected fauna totaling 91 identified species and subspecies. Acknoivledg})ients. The writers are indebted to Messrs. Ralph O. Fox and James W. Emerson for aid in field collecting and to Drs. J. Wyatt Durham, L. G. Hertlein, W. P. Woodring, and Mr. James W. Valentine for offering helpful suggestions and pertinent data pertaining to Pacific Coast Pleistocene problems. Mr. Owen J. Poe prepared the illustration. REVIEW OF LITERATURE Though the Pleistocene terrace fossils from the San Diego area have been recorded and discussed by a number of writers, including Dall (1878), Arnold (1903), Berry (1922), Stephens (1929), Webb (1937), and Hertlein and Grant (1939, 1944), the locality herein reported upon is specifically mentioned only by Stephens. In a paper pertaining to the fossiliferous Pleistocene deposits of the San Diego area, Stephens (1929: 250) briefly described this locality and listed the most conspicuous faunal elements. His description and remarks follow: "International Boundary Monument No. 258 stands on a mesa a few yards from its edge, which is at the Pacific Ocean. This mesa is a terrace about seventy feet in altitude at the beach and about one hundred and fifty feet where it abuts against the hills to the eastward. It is about half a mile in width. North- ward, it is bounded a quarter of a mile from the international boundary by the Tijuana River bottom. The sand beach on the ocean side of the terrace is a hundred yards in width. The bank at the edge of the terrace is a steep, soil-covered slope carrying a considerable growth of brush. The soil covers the edges of the more or less consolidated strata of sand and clay underlying the terrace and rock exposures are very few. Seventy-five yards north of Monument No. 258, a small ravine cuts through the talus and exposes a fossiliferous stratum about forty feet below the upper edge of the slope. The stratum is about a foot thick and contains an abundance of fragments of large bivalves. These fragments are so thoroughly broken that few can be rec- ognized specifically. By sifting, a number of small univalves were obtained, mostly small species. The material underlying the fossiliferous stratum appears to be a soft sandstone. That 432 San Diizgo Society of Natural History over it is sand and clay, poorly consolidated except a hard stratum near the top of the slope. The species most abundant here are: Alectrion [=Nassar/us'] perpinguis. Alectrion jossa- tus, Columbellci \_—M}trella'] cariiiata. Conns calijornicus, Cre- pidula adunca, Dentalium neohexagonnm, Ol'ivella boetica, Tellina meropsis, Tnrbonilla sp.," (S. D. S. N. H. Loc. no. 118, and =U.C.M.P. Loc. no. A-9003 of this report). In addition to the species recorded above Stephens (1929:251) listed Glycynieris septentrionalis [=G. (?) profunda Dall, 1897], Kellia laperousii, Odostomia sp., Olivella pedroana, and Tellina bode gen sis from outcrops in a ravine about 200 yards north of the international boundary and approximately a quarter of a mile from the ocean, (S. D. S. N. H. loc. 119 and, probably, the same as U. C. M. P. loc. A-9005).- On the basis of similar lithology and altitude of the beds, he concluded that localities 118 and 119 represent outcrops of the same stratum which ". . . underlies much of the terrace, possibly as far as it extends . . .," (Stephens, 1929:250). For the purposes of this report, Stephens' locali- ties were re-examined and a reconnaissance of the area concerned was undertaken. ^Hereafter locality citations refer to those of the Museum of Paleontology of the University of California, Berkeley, unless otherwise stated. Emerson-Addicott — Pleistocene Invertebrate Fauna 433 ^ MILES CO/VrOO/? /NTERVAL 25 FT. A-9007s A- 9004- ':U-- oSanDiego Indian Point Index map of area showing location of Border, Spanish Bight and Indian Point localities. 434 San Diego Society of Natural History DESCRIPTION OF AREA General Aspects. In the extreme southwest corner of San Diego County, Cahfornia, a low terrace forms a bank facing the ocean for a short distance north of the United States-Mexican Boundary (see map). Though the terrace extends south for some distance along the Baja Cali- fornia coast, the erosional action of the Tijuana River has all but de- stroyed that part which extends immediately north of the border. On the American side, the terrace attains a maximum elevation of 53 feet, gain- ing very slightly in altitude as it runs from the beach front to the foot of higher Pliocene terraces to the east. Hereafter, for expediency, this deposit is referred to as the "Border locality." A fossiliferous stratum outcrops in several small gullies along the truncated west face of the terrace (loc. A-9003) and is well exposed along the northwest river-cut face of the terrace (loc. A-9004). Approxi- mately one-fourth of a mile east of the ocean front a small ravine runs north through the terrace from the Mexican side of the border. A highly fossiliferous exposure occurs in a road cut leading to International Monu- ment No. 258 along the west face of this ravine (loc. A-9005). The stratum also outcrops for a short distance immediately east of the ravine in the north face of the terrace (loc. A-9006). Beyond this point the face of the terrace is obscured by vegetation and soil creep. A second sparsely ""fossiliferous" stratum caps the westernmost portion of the ter- race (loc. A-9007). Inasmuch as moUusks in this deposit may represent scattered kitchen midden remnants, this assemblage is not included in subsequent discussions relating to the fauna of the Border locality. Stratigraphic Aspects. The principal fossiliferous stratum is essen- tially horizontal. It immediately overlies a grey weathering, micaceous, sandy siltstone of probable Pliocene age.^ The contact is best exposed in small gullies along the seaward face of the terrace (loc. A-9003). Here the Pliocene (?) siltstones show evidence of channeling and contain specimens of the burrowing bivalve, Zirfaea pilsbry/. in situ. At locality A-9004, the basal sand is overlain by 30 feet of poorly consolidated sediments. These beds appear to be unfossiliferous throughout and range from a massive, poorly sorted, gravelly sand to thinly bedded, very fine grained sands. The sands are capped by 1 to 2 feet of red weathering, sandy soil containing occasional kitchen midden (?) mollusks, (loc. A-9007). The highly fossiliferous basal bed ranges in thickness from 1 to 12 inches. It is composed of grey, fine grained sands containing occasional cobbles and boulders. The sands are predominately well-sorted and un- consolidated except for local calcareous cemented pockets. spor a discussion of the evidence for Pliocene dating for these sediments see page 438. Emerson-Addicott — Pleistocene Invertebrate Fauna 435 RELATION TO OTHER TERRACES For the purposes of this paper only the principal terrace deposits in the San Diego region are considered. EHis in Ellis and Lee (1919:26, pi. 6) described five marine terraces from the San Diego Bay region, namely: Tijuana terrace, 20 to 50 ft.; Nestor terrace, 25 to 100 ft.; Chula Vista terrace, 100 to 130 ft.; Avondale terrace, 200 to 250 ft.; and Otay terrace, 430 to 525 ft. Hertlein and Grant (1944:20) in dis- cussing the younger terraces of the San Diego Quadrangle record the presence of still other smaller, though poorly developed and preserved, intermediate terrace deposits in addition to those named by Ellis. Hanna (1926:194) proposed the name La Jolla terrace for the prominent, lowermost terrace that ranges from 25 to 200 feet and extends from La Jolla to Pacific Beach; he concluded that it corresponds, at least in part, with the Chula Vista terrace of Ellis. Gale in Grant and Gale (1931: 39,49) considered the La Jolla terrace to be a northern extension of the Nestor terrace. Webb (1937) described a rock-cliff, tide-pool fauna from the lowermost terrace, ranging from 25 to 100 feet in elevation, on the west side of Point Loma, a peninsula extending into the Pacific Ocean to form a protecting promontory to the northwestern portion of San Diego Bay (see map). He assigned a late Pleistocene age to the Point Loma terrace on physiographic and faunistic evidence. Ellis (1919:pls. 3,6) assigned the terrace deposit containing the fauna enumerated in this paper to the Nestor which he believed to be correlative with the "San Pedro formation." Though no fossils were recorded from this locality by Ellis, "Quarternary fossils" are reported by him from the Nestor deposits occurring north of the Tijuana River near Oneonta (south San Diego) at an elevation of about 50 feet. In proposing the name. Bay Point formation, for late Pleistocene terrace deposits in the San Diego area, Hertlein and Grant (1939:72) specifically included exposures occurring at Bay Point in Mission Bay (type locality), Spanish Bight and Indian Point in San Diego Bay, and Pacific Beach; correlation, at least in part, with the Palos Verdes forma- tion in the Los Angeles region was suggested. Available physiographic evidence does not permit precise correlation of the Border terrace remnant with other terrace deposits in the region. Its relationship to the Bay Point formation is uncertain at present. DEPOSITIONAL ENVIRONMENT Suggested Environment. The composition of the Border locality fauna, enumerated in the check list on page 439, indicates relatively shal- low water conditions, possibly a protected coast facies. This inner neritic interpretation is corroborated by depositional evidence, the fossiliferous sediments being composed of sands containing gravel, cobbles, and broken shell fragments. With the exception of a few deeper water elements (Fus/tr/ton, Barbarofusus, Glycymeris, Dental'tum, etc.),* a similar Recent *In the vicinity of San Diego Fusitrilon oregonensis is recorded from deep water (Burch, 1945, no. 54,5) and "dredged" (Kelsey, 1907:48); and Dentalium neohexagonum from 5 to 15 fms. (Burch, 1945: no. 46,9). Off Redondo Beach, Calif., Glycymeris profunda is recorded from 25 fathoms (Willett, 1945:110). 436 San Diego Society of Natural History fauna is recorded from Mission Bay, San Diego, '■ where a number of diversified ecologic niches, including sand spits and rocky bottom, are present in a shallow water embayment which is protected from severe wave and current action. However, the absence of diagnostic tidal flat dwellers such as Melampus and Cerithidea from the fauna precludes definite assignment to a protected bay facies. It is probable that the Pleistocene Point Loma island^ constituted a protecting barrier against southerly directed currents for the coastal region immediately to the south, including the present localities. A protected coast facies need not be postulated for the fauna as rep- resentatives of it, with very few exceptions, are known to be living today off Redondo Beach, California, in waters ranging from 10 to 50 fathoms; most of the species are recorded in 10 to 25 fathoms from bottoms com- posed of fine to coarse sands and gravels. '^ The fauna may not conclu- sively be referred to depths greater than the eulittoral zone and the shal- lower limits of the sublittoral zone as only a small part of the fauna con- sists of deeper water species. The deeper water faunal element is com- posed of very few individual specimens all of which are badly worn or fragmented indicating probable transport. Climatic inferences. An analysis of the geographical ranges of extant representatives of the fauna indicates that all are living today in the lati- tude of San Diego with the exception of one species which is recorded from the San Pedro area. Thus temperate marine conditions are suggested with water temperatures at the time of deposition being essentially the same as today. ^ Ecologic requirements. Ecologic evaluations of Pleistocene inverte- brate faunas can be drawn through comparisons with present distribution of component species. On the basis of ecologic data available for 84 per- cent of the fauna, two primary substrate requirements must be satisfied. First, a few gastropods and most of the pelecypods occur only in sandy bottoms. Secondly, rock areas are required to support a majority of the gastropods. Ecologic niches in which rocks may be found in association with sand occur in Mission Bay and inner Newport Bay, California, as well as subtidally along the outer, southern California coast, and would suggest that the fossil fauna lived in an environment similar to one of these. •"^Morrison, R. L., (1943: no. 19, 5-6); Wilson, E. N., (1943: no. 19, 6-8). "Stephens, F., (1929:245). ^Burch, Tom, (1942: no. 17, 5-11). sThe mean annual surface temperature for San Diego is 16.7°C. Emerson-Addicott — Pleistocene Invertebrate Fauna 437 FAUNAL COMPARISONS Schenck and Keen (1937) have proposed an index method for com- paring molluscan faunas. By this method, the medians of latitudinal mid- points of present geographic ranges are compiled for extant constituents of the fossil assemblages. With these findings it is possible to evaluate temperature requirements of a fossil fauna by reference to present day average ocean temperatures for the calculated latitudinal median. By comparing this average temperature with that prevailing at the actual latitude of the fossil locality, precise differences in water temperatures have been postulated. Inasmuch as the reliability and limitations of this comparative system have been discussed by Schenck (1945:11), Wood- ring et al. (1946:103), and Newell (1948:165), these problems are not treated here. Notwithstanding the inherent difficulties involved in an analysis of this type, it would appear that such evaluations may provide a useful method for comparing Pleistocene faunas provided that the assem- blages are based upon adequate, accurately identified samples and repre- sent comparable bathymetric and ecologic facies. For the purpose of faunal comparisons, medians of mid-points for the best known Pleistocene faunas of the San Diego region are tabulated in table 1. From these data it must be inferred that the faunas, taken as a whole, indicate a latitudinal occurrence in the region to the north of Point Conception, 34° 26' N. latitude and, consequently, average water temperatures definitely cooler than those of the San Diego region today. These findings are not in harmony with interpretations based upon pri- marily qualitative evaluations of the same faunas. None of the warm water indicators such as Ch'ione gn'id'ia. Card'tum procerum. Dosinid pon- derosch and Tunitellcj gonostoma. appear in the Border or Point Loma (Webb, 1937) faunas. Since these species occur in other late Pleistocene deposits in the San Diego area, previous workers have postulated, on the basis of present geographic ranges of the species," that warmer water con- ditions than those now prevailing existed during deposition of those sediments. Thus an extension of marine surface water isotherms of sev- eral hundred miles northward is indicated (Grant and Gale, 1931:49; Hertlein and Grant, 1944:71). It is also difficult to explain the identical medians of mid-points for the faunas of the Point Loma terrace and the Border locality as the two deposits represent diverse ecologic types, the former being a typical rock-cliff and tide-pool facies and the latter being a probable protected coast, rock and sand facies. Hertlein and Grant (1944:68) interpreted the absence of southern indicators and the totally modern aspect of the Point Loma fauna as indicative of a younger age than the Bay Point faunas. Conversely, Woodring, et id., (1946:106) ''Present northern limits of distribution based upon living material and mean annual surface temperatures are: Ch'tone gnid'ia Cedros Island (17.0°C.) Dosinia ponderosa Scammon Lagoon (18.3°C.) Card'tum procerum Scammon Lagoon (18.3°C.) Turritella gonostoma Scammon Lagoon (18.3°C.) 438 San Diego Society of Natural History attribute the absence of southern species to facies differences and presume the deposits to be of the same age. Warm water species which characterize the late Pleistocene Bay Point and Palos Verdes faunas are absent from the Border fauna. Therefore, it can be postulated that water temperatures during the deposition of the Border terrace were possibly as cold as those of San Diego at present. Results of the median of mid-points analysis corroborate this assumption and suggest even colder water temperatures. However, the evidence at hand is not entirely conclusive. This apparent lowered water temperature requirement may represent a local facies difference, possibly the influence of upwelling, which is not reflected by the faunas of the Bay Point for- mation complex, and may, or may not, indicate an isothermal shift of temporal magnitude. In the final analysis, physiographic reality rather than faunal inter- pretation appears to be the more promising tool for determining contem- poraneity of the terrace remnants in this area. Table 1 Upper Pleistocene Deposits of the San Diego Region With Medians of Mid-points of Mega-Inveftebrate Faunas Locality Latitude 17 ,„»• Median of roimation n,-, . ,, Mid-point^' No. of Air Species Authonty for Utilized P=^""^ "Border Locality" Point Loma Spanish Bight Pacific Beach 32° 32' 32° 40' 32° 42' 32° 48' Bay Point? Bay Point? Bay Point Bay Point 35° 06' 35° 06' 35° 48' 34° 24' This paper Webb, 1937 Arnold, 1903 Arnold, 1903 AGE Previous mention is made in this paper of the superposition of the Border locality terrace deposits over apparent Pliocene sediments. Hert- lein and Grant (1944:59) record the occurrence of Pliocene mollusks, including Patinopecte)7 healeyi. from a well just north of the Mexican Boundary and from the bottom of Matadero Canyon which is situated four-tenths of a mile west of the United States-Mexican Boundary Monu- ment No. 256. The writers collected Pliocene fossils, including Patino- pecten healeyi, from a bed of fine grained sandstone outcropping in the mouth of a ravine one-half mile east of loc. A-9006. If the deposit is correctly assigned to the Nestor terrace, an unques- tioned post-San Diego age is indicated on the basis of physiographic data since the Nestor bevels the Pliocene San Diego formation. The fauna, totaling 91 identified invertebrate species, is composed of extant species with one exception, a fragment of a nassarid gastropod which may be referred to Nassarius delosi (Woodring). This species was loQeographic range data and mid-points were compiled from Keen (1937) with the exception of "extra-limital species." Emerson-Addicott — Plkistoci-ne Inverti-brate Fauna 439 originally described from the Pales Verdes sands and is not known from earlier Pleistocene deposits or from Recent faunas. Woodring (/;z Uteris) reports the presence of this species in the late Pleistocene faunas of Pa- cific Beach and Spanish Bight, San Diego. The lone specimen obtained shows evidence of transport and may have been re-worked from previ- ously deposited sediments. The data, both faunistic and physiographic, indicate a late Pleisto- cene age for the deposit. Though faunal evidence points to a possible younger age than the "so-called" Palos Verdes sands equivalents in the San Diego area, this evidence is not conclusive and the deposit may only be referred to the late Pleistocene on the basis of the available data. ALPHABETICAL CHECK LIST OF FOSSILS The nomenclature used in the check list of the fauna generally fol- lows that of Keen (1937). Though no screening was undertaken to obtain minute organisms, a fauna was collected from the primary fossili- ferous bed totaling 88 species of mollusks (48 gastropods, 2 scaphopods, and 38 pelecypods), 1 echinoid, and 2 barnacles. From the soil capping the terrace (loc. A-9007) a total of 7 species of mollusks (1 gastropod and 6 pelecypods possibly of kitchen midden origin) were collected; only three species are common to both beds. Locality Nos. GASTROPODA A-9004 A-9005 A-9006 A-9003 A-9007 Acanthina spirata (Blainville, 1832) X X Aletes squamigerus Carpenter, 1856 X Amphissa versicolor Dall, 1871 X Astraea undosa (Wood, 1828) X X Barbarofusus aff. B. barbarensis arnoldi (Cossmann, 1903) X Bursa calif ornica Hinds, 1843 X Calliostoma ligatum Gould, 1852 = C. costatum Martyn,1784 X Calliostoma decarinatum (Perry, 1811) = C. canaliculatum (Martyn, 1784) X Calliostoma tricolor Gabb, 1865 X Clathrodrilla incisa ophioderma (Dall, 1908) X Conus californicus Hinds, 1844 X X X X Crepidula adunca Sowerby, 1825 X X X Crepidula lingulata Gould, 1846 X X X X Crepidula nivea C. B. Adams, 1852 X Crepidula norrisianum Williamson, 1905 X Crepidula onyx Sowerby, 1824 XXX Cytharella hexagona (Gabb, 1865) X Diodora aspera (Eschscholtz, 1833) XX X Diodora inaequalis (Sowerby, 1835) X Epitonium iiidianorum (Carpenter, 1864) X Fartulum cf. F. hemphilli Bartsch, 1920 X Fusitriton oregonensis (Redfield, 1846) X Halistylus subpupoideus (Tryon, 1887) XXX Homalopoma carpenteri (Pilsbry, 1888) X Jalon jestiva (Hinds, 1844) X 440 San Diego Society of Natural History Locality Nos. A-9004 A-9005 A-9006 A-9003 A-9007 Kellettia kelletti (Forbes, 1850) X X Littorina scutulata Gould, 1849 X Maxwellia gemma (Soweiby, 1879) X Megasurcula carpenteriana (Gabb, 1865) X Megatebennu.s bimaculata (Dall, 1871) X Megathura crenulata (Soweiby, 1825) X Mitra idae Melvill, 1893 X Mitrella carinata (Hinds, 1844) X X X Mitrella carinata gausapata (Gould, 1851) X X N as sari us delosi (Woodring, 1946) X Nassarius jossatus (Gould, 1849) XXX Nassarius mendicus cooperi (Forbes, 1850) X X X X Nassarius perpinguis (Hinds, 1844) X X X X Norrisia norrissi (Sowerby, 1838) X Ocenebra inter jossa Carpenter, 1864 X Ocenebra lurida aspera (Baird, 1863) X X Olivella biplicata (Sowerby, 1825) XXX Olivella pedroana "Conrad" of Woodring, 1946 XXX Polinices lewisi (Gould, 1847) X X ? Pusula calif ornica (Gray, 1863) X Seila montereyensis Bartsch, 1907 X Tegula aureotincta (Forbes, 1850) X Triphora cf. T. pedroana (Bartsch, 1920) X Zonaria spadicea (Swainson, 1823) X SCAPHOPODA Dentalium neohexagonum Pilsbry and Sharp, 1897 Dentalium pretiosum Sowerby, I860 PELECYPODA Aequipecten aequisulcatus Carpenter, 1864 Aloidis luteola Carpenter, 1864 Anomia peruviana d'Orbigny, 1846 Apolymetis biangulata (Carpenter, 1855) Botula calif ornien sis (Philippi, 1847) Chama pellucida Broderip, 1835 Chione californiensis (Broderip, 1835) Crassinella branneri (Arnold, 1903) Cryptomya calif ornica (Conrad, 1837) Cumingia californica Conrad, 1837 Donax gouldi Dall, 1921 Gari californica (Conrad, 1849) Glans carpenteri (Lamy, 1922) Glycymeris (?) profunda Ball, 1879" Leptopecten latiauratus monotimeris (Conrad, 1837) Lucina nuttalli Conrad, 1837 Lucinisca californica (Conrad, 1837) Macoma nasuta (Conrad, 1837) Mactra californica Conrad, 1837 Mytilus californianus Conrad, 1837 Nuculana taphria (Dall, 1896) Pandora punctata Conrad, 1837 Petricola carditoides (Conrad, 1837) Platydon cancellatus Conrad, 1837 X X X X X X X X X v X X X X X X X X X X X X X X X X X X X X X X X X X -. > X X X X X X X X X X X X X X X X X i^Specimens compare favorably with Arnold's (1903) figure of G. septen- trionalis (= 'i G. corteziana Dall, 1918), but are dissimilar with Willett"s (1944) figures of G. corteziana. Locality N OS. .-9004 A-9005 A-9006 A-9003 A-9007 X X X X X X X X X X > X X X •X X X X X X X X X X X X X X X X X X X X X X X X X X ? X EMnRSON-ADDicoTT — Pleistocenh Invertebrate Fauna 441 Pododesmus macroschisma (Deshayes, 1839) Protothaca jluctifraga (Sowerby, 1853) Protothaca staminea (Conrad, 1837) Saxidomus nuttalli (Conrad, 1837) Schizothaerus nuttalli (Conrad, 1837) Semele decisa (Conrad, 1837) Semele pulchra (Sowerby, 1832) Spisula catilliformis (Conrad, 1867) Spisula falcata Gould, 1850 Taras orbellus (Gould, 1852) Tellina bodegensis Hinds, 1844 Tellina meropsis Dall, 1900 Tivela stultorum (Mawe, 1823) Trachycardium quadragenariujn (Conrad, 1837) Transennella tantilla (Gould, 1853) Vol sella fornkata (Carpenter, 1864) Yoldia coo peri Gabb, 1865 Zirjaea pilsbryi Lowe, 1931 ECHINOIDEA Dendraster excentricus (Eschscholtz, 1831) XXX CIRRIPEDIA Balanus cf. B. niibilus Darwin, 1854 X X Balanus cf. B. tintinnabulum calijornicus Pilsbry, 1916 X X FOSSIL LOCALITIES Collecting localities were located on the U.S.G.S. San Ysidro, Cali- fornia quadrangle (1:31680), edition of 1943. AH localities are in the NE14 of Sect. 7, T.19S., R.2W.; the localities are plotted on the index map, page 433. A-9003. (32° 32' 5.7" N. - 117° 07' 19.5" W.). Small gulley in ter- race, 250 feet along shore line north of U.S.-Mexico border. From 1-5 inch stratum immediately overlying channeled Pliocene (.^) sediments. Same locality as S.D.S.N.H. 118 (Stephens, 1929). Rock type: soft, poorly sorted sands with occasional cobbles and boulders. Elevation: 35 feet. Collected by R. O. Fox, J. W. Emerson, and the authors. A-9004. (32° 32' 12.2" N. - 117° 07' 18.6" W.). North face of ter- race, 700 feet north of International Boundary Monument No. 258. From 1-3 inch stratum overlying Pliocene (?) sandy siltstone. Rock type: grey, unconsolidated, fine grained sand. Elevation: 20-25 feet. Collected by R. O. Fox, J. W. Emerson, and W. K. Emerson. A-9005. (32° 32' 10.6" N. - 117° 07' 11.5" W.). Northeast face of terrace, 600 feet east of locality A-9004. From highly fossiliferous, 6-12 inch stratum exposed in cut along road to International Boundary Monu- 442 San Diego Society of Natural History ment No. 258. Apparently same locality as S.D.S.N.H. 119 (Stephens, 1929). Rock type: grey, well-sorted, fine grained sand. Elevation: about 30 feet. Collected by R. O. Fox, J. W. Emerson, and the authors. A-9006. (32° 32' 9.4" N. - 117° 07' 6.3" W.). North face of ter- race immediately east of 250 foot wide, north trending ravine and 1,750 feet inland from ocean. From 4-8 inch stratum underlying a boulder conglomerate. Rock type: poorly sorted sand with occasional cobbles. Collected by R. O. Fox, J. W. Emerson, and W. K. Emerson. A-9007. Same as locality A-9004 only 25 feet higher stratigraphically. From irregular 2 foot stratum capping terrace. Rock type: red weather- ing, poorly sorted sand. Elevation: 50 feet. Collected by J. W. Emerson and the authors. Literature Cited Arnold, Ralph 1903. The paleontology and stratigraphy of the marine Pliocene and Pleistocene of San Pedro, Calif. Calif. Acad. Sci. Mem., vol. 3, 420 pp., 37 pis. Berry, S. S. 1922. Fossil chitons of western North America. Calif. Acad. Sci. Proc, 4th ser., vol. 11, pp. 399-526, pis. 1-16, figs. 1-11. BuRCH, J. Q. (editor) 1945. Distributional list of the West American Marine mollusks from San Diego, Calif, to the Polar Sea. Minutes Concho- logical Club Southern Calif., no. 46, p. 9, no. 54, p. 5. BuRCH, Tom, /'/; Burch, J. Q. (editor) 1942. Dredging off Redondo Beach, California. Minutes Concho- logical Club Southern Calif., no. 17, pp. 5-12. Dall, W. H. 1878. Distribution of Calif. Tertiary fossils. Proc. U. S. National Museum, vol. 1, pp. 26-30. Ellis, A. J., /';; A. J. Ellis and C. H. Lee 1919. Geology and ground waters of the Western part of San Diego County, Calif. U. S. Geol. Surv., Water Supply Pap. 446, pp. 20-50, pis. 6, 7. Gale, H. R., /;; Grant, U. S., IV and H. R. Gale 1931. Catalogue of the marine Pliocene and Pleistocene Mollusca of Calif, and adjacent regions . . . Mem. San Diego Soc. Nat. Hist., vol. 1, pp. 1-78, tables 1-3. Hanna, M. a. 1926. Geology of the La Jolla Quadrangle, Calif. Univ. Calif. Publ., Bull. Dept. Geol. Sci., vol. 'l6, no. 7, pp. 187-246, pis. 17-23, 1 map. Hertlein, L. G., and U. S. Grant, IV 1939. Geology and Oil possibilities of Southwestern San Diego County. Calif. Journ. Mines and Geology, vol. 35, no. 1, pp. 57-78, 8 figs. Emerson-Addicott — Plhistocene Invertebrate Fauna 443 1944. The geology and paleontology of the marine Pliocene of San Diego, Calif. Mem. San Diego Soc. Nat. Hist., vol. 2, pt. 1, Geology, pp. 1-72, pis. 1-18, 1 map. Keen, A. M. 1937. An abridged check list and bibliography of West North American marine mollusca. Stanford Univ. Press, 87 pp., 3 figs. Kelsev, F. W. 1907. Mollusks and brachiopods collected in San Diego, Calif. San Diego Soc. Nat. Hist. Trans., vol. 1, no. 2, pp. 31-55. Morrison, R. L., /;; Burch, J. Q. (editor) 1943. Partial list of mollusks found in Mission Bay, San Diego, Calif. Minutes Conchological Club Southern Calif., no. 19, pp. 5, 6. Newell, I. M. 1948. Marine molluscan provinces of Western North America: a critique and a new analysis. Amer. Philos. Soc. Proc, vol. 92, no. 3, pp. 155-166, 7 figs. SCHENCK, H. G. 1945. Geologic application of biometrical analysis of molluscan as- semblages. Journ. of Paleont., vol. 19, no. 5, pp. 504-521, 3 figs. , and A. M. Keen 1937. An index method for comparing molluscan faunules. Amer. Philos. Soc. Proc, vol. 77, no. 2, pp. 161-182, 4 figs. Stephens, Frank 1929. Notes on marine Pleistocene deposits of San Diego County, Calif. San Diego Soc. Nat. Hist. Trans., vol. 5, no. .16, pp. 245-256, 1 fig. Webb, R. W. 1937. Paleontology of the Pleistocene of Point Loma, San Diego County, Calif. San Diego Soc. Nat. Hist. Trans., vol. 8, no. 24, pp. 337-348. Willett, George 1945. Northwest American species of Glycimeris. Southern Calif. Acad. Sci. Bull., vol. 42, pt. 3, pp. 107-114, pis. 11, 12. Wilson, E. N., 'm Burch, J. Q. (editor) 1943. Shells (alive and dead) . . . found in Mission Bay [San Diego, Calif.}. Minutes Conchological Club Southern Calif., no. 19, pp. 6-8. 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, 145 pp., 37 pis. INDEX Transactions of the San Diego Society of Natural History, Volume XI Titles of papers and new systematic names are in heavy-faced type A New Fossil Chilopod from the Late Cenozoic, 117-120 A New Gopher Snake (Pituophis) from Santa Cruz Island, California, 41-48 A Pleistocene Invertebrate Fauna from the Southwest Corner of San Diego County, California, 429-444 Abbott, Clinton G., 32, 117, 261 Abbott, Waldo G., 45, 47, 107 Acanthina spirata, 439 Acar, 350, 352 Achatella carleyi, 287, 312 Acidaspis, 270 anchoralls, 277 brightii, 304 cincinnatiensis, 277-278 crenata, 296 crenatus, 296 dormitzeri, 304 leonhardi, 304 longispina, 305 longispinis, 305 mira, 305 prevosti, 304-305 verneuili, 305 vesiculosa, 305 Addicott, Warren O., 429-444 Adenostoma, 232, 239 fasciculatum, 234-235 sparsifolium, 234 Aequipecten aequisulcatus, 440 Agaronia, 407-428 murrha, 417-426 testacea, 418 Agave Shawi, 232 Agkistrodon, 135 Alectrion fossatus, 432 perpinguis, 432 Aletes squamigerus, 439 Aloides luteola, 440 Amaral, Afranio do, 68, 83, 107 Amphion beaumonti, 297 Amphissa versicolor, 439 An Undescribed Race of the Mangrove Warbler from Baja California, Mexico, 49-52 Anchiopsis anchiops, 286, 312 Anniella, 192 Anniella pulchra, 380 Anomia peruviana, 350, 440 Anota, 383 Antiplanes perversa, 420 rotula, 420 willetti, 419-426 Apolymetis biangulata, 440 Arctostaphylos, 235, 378 Arethusa koninckii, 291 Arizona, 149, 190-191, 195 elegans, 184 elegans Candida, 190 elegans eburnata, 191 Arnold, J. R., 368, 383 Arnold, R., 416, 425, 431, 440, 442 Asaphus armadillo, 292 aspectans, 287 canadensis, 265 expansus, 291-292 extans, 264 hausmanni, 298 limulurus, 285 micrurus, 286 vigilans, 267 Astraea undosa, 439 Aulacopleura angusticeps, 291 konincki, 291 Axelrod, Dan I., 367 446 San Diego Society of Natural History Baily, Joshua L., Jr., 107, 197, 261, 395, 407 Baird, Spencer F., 33, 108, 147, 198 Balanus nubilus, 441 tintinnabulum californicus, 441 Barbarofusus, 435 arnoldi, 439 barbarensis, 439 Barber, Charles M., 131 Barbour, T., 35, 137 Barlow, Alice W., 261 Bartholomew, George A., 367 Bartsch, P., 340, 350-352, 356, 425 Bathyurus extans, 264, 308 Belcher, E., 348, 356 Belding, Lyman, 33, 51 Berry, S. Stillman, 329-344, 393-428, 431, 442 Bibron, G., 22, 34 Bilderback, Norman, 41, 136 Blackwelder, R. E., 69 Blainville, H. D. de, 7, 22, 33 Blanchard, Frank N., 18, 33 Bocourt, F., 22, 33 Bogert, Charles M., 32-33, 107, 136, 188, 197, 199, 380-381, 383 Bothrops, 135 Botula californiensis, 440 Boulenger, George A., 108, 198 Brachymetopus armatus, 275 cuyahogae, 274, 311 lodiensis, 274 spinosus, 275 Bramlette, M. N., 368, 385, 443 Brand, Donald D., 135-136 Brattstrom, Bayard, 197, 365-392 Brissus latecarinatus, 352 Bronteus bowingensis, 301 jenkinsi, 302 longispinifex, 302 partschi, 293, 302 umbellifer, 293 Brooks, Baylor, 26 Brown, Arthur E., 108, 198 Brown, John S., 198 Bryan, Kirk, 198 Bryant, H. C, 383 Bufo, 369-391 boreas, 369-391 boreas boreas, 371, 391 boreas halophilus, 370-371 nestor, 369-391 Bulla, 423 Bumastus armatus, 268, 309 barriensis, 268 chicagoensis, 267, 309 cuniculus, 269, 310 imperator, 269-270 insignis, 267 ioxus, 268 vogdesi, 269, 309 Burch, John Q., 401-402, 435, 442 Burch, Tom, 436, 442 Burger, W. L., 197 Bursa californica, 439 granulans, 350, 363 pustulosa jabick, 350 Calciphilus abboti, 118 Calcitro fisheri, 117 Calliops callicephalus, 287 Galliostoma canaliculatum, 439 costatum, 439 decarinatum, 439 ligatum, 439 tricolor, 439 Callisaurus, 380 Calymene anchiops, 286 beaumonti, 297 blumcnbachii, 298 blumenbachii niagarensis, 280 bufo rana, 288 Index to Volume XI 447 celebra, 280-281 clintoni, 278, 280, 282-283 concinna, 293 crassiniarginatus, 271 declinata, 297 downingiae, 300 latifrons, 300 meeki, 282 niagarensis, 280-281 platys, 281 punctata, 296 rostrata, 278, 283-284 senaria, 281-282 variolaris, 296 vogdesi, 278, 280, 283-284, 311- 312 Calymenella rostrata, 270, 278, 283-284, 312 Camelops, 368 Camp, Charles L., 109, 198-199, 367, 370, 379, 381, 383 Caphyra radians, 291 Cardium procerum, 437 Carpenter, P. P., 412-413, 425 Carr, Archie, 381, 383 Catazyga headi schuchertana, 266 Caudisona confluenta, 83 confluenta confluenta, 83 lucifer, 67, 83-84 lucifer cerberus, 83 Celtis mississippiensis reticulata, 378 Ceratocephala anchoralis, 277-278, 311 ceralepta, 278 cincinnatiensis, 278 longispina, 305, 314 vogdesi, 305 Ceraurus cerenatus, 296 pleurexanthemus, 284-285, 312 Cerithidea, 436 Chace, E. M., 331 Chace, E. P., 331, 336 Chama pellucida, 440 squamuligera, 350 squamuligera rubropicta, 350 Chamberlin, Ralph V., 117-120 Charodotcs, 337-339 arrosa, 337 californiensis, 337 dupetithouarsii, 337 pctricola, 338 traskii, 338 tudiculata, 337 Chasmops bucculenta, 299 macrourus, 299 Cheirurus sternbergi, 298 Chilomeniscus, 153, 192 cinctus, 181 Chionactis, 141-204 occipitale, 151 occipitalis, 146-147, 153, 162, 184 occipitalis annulata, 143-202 occipitalis klauberi, 143-197 occipitalis occipitalis, 143-202 occipitalis palarostris, 175, 178 occipitalis talpina, 143-203 palarostris organica, 143-203 palarostris palarostris, 143-197 Chione californiensis, 440 gnidia, 437 Clanculus clippertonensis, 350, 354, 355, 363 pharaonicus, 355 rubidus, 355 Clathodrilla incisa ophioderma, 439 Clemmys, 369-391 hesperia, 377 marmorata, 369-391 saxea, 377, 389 Clipperton Island, 345-364 Cnemidophorus, 369-391 estebanensis, 106 hyperythrus, 375 tigris, 369-391 Cochran, Doris M., 32, 107, 136, 162, 197 Cole, La Mont C, 383 Coleonyx, 25 Coluber, 369-391 Coluber catenifer, 7 constrictor mormon, 90 vertebralis, 22 Columbella carinata, 432 448 San Diego Society of Natural History Comstock, Eric, 173 Contia, 146 occipitale, 153, 162 occipitalis, 153, 162 occipitalis annulata, 162 occipitalis occipitalis, 153 Conus, 350-354 brunneus, 351 californicus, 432, 439 ebraeus, 348, 351-353, 363 ebraeus vermiculatus, 351 edaphus, 352 gradatus, 351 miliaris, 351 purpurascens regalitatus, 351 roosevelti, 351 tessulatus, 352 tiaratus, 351 Cooper, J. G., 108, 198 Cope, E. D., 34, 68, 108, 146, 198 Coronura aspectans, 287 Corycephalus dentatus, 286, 312 Coues, Elliott, 108 Cowles, Raymond B., 32, 107, 185, 188, 191-194, 197, 199, 367, 375, 377, 381, 383 Craig, Willis G., 331, 336 Crassinella branneri, 440 Crassispira rotula, 419 Crepidula adunca, 432, 439 lingulata, 439 nivea, 439 norrisiarum, 439 onyx, 439 Cromus beaumonti, 297 bohemicus, 297 murchisoni, 305 Crotalus, 25, 57-59, 63-113, 191, 369-391 adamanteus, 94 adamanteus atrox, 90 adamanteus lucifer, 77 atrox, 59, 90, 94, 105, 109 cerastes, 135, 184-185, 190, 192 cerastes cerastes, 67, 76, 154, 190 cerastes laterorepens, 154, 191 confluentus, 68, 83, 90, 107, 109 confluentus lucifer, 68, 77 confluentus oreganus, 68, 77, 83, 90 exsul, 59, 82, 85 Hallowelli, 67, 77-78 helleri, 77 lecontei, 67, 77-78 lepidus, 128 lepidus klauberi, 135 lucasensis, 57-59 lucifer, 67-68, 77-78, 83, 90 mitchellii, 64, 85, 96-102, 109 mitchellii mitchellii, 97-102 mitchellii muertensis, 64-113 mitchellii pyrrhus, 97-102 mitchellii Stephens!, 76, 97-101 molossus, 70, 84, 103-106 molossus estebanensis, 64-113 molossus molossus, 104-106 molossus nigrescens, 106 oreganus, 67, 77, 83, 90 oregonus, 67, 77, 83, 90 oregonus Cerberus, 83 oregonus lucifer, 68, 77 oregonus oregonus, 68 ruber, 57-59 ruber lucasensis, 59, 85 ruber ruber, 59, 85, 105 scutulatus, 70, 184 scutulatus scutulatus, 76, 191 viridis, 369-391 viridis caliginis, 64-112 viridis cerberus, 63-112 viridis decolor, 68 viridis helleri, 63-111, 377 viridis lutosus, 66, 73-77, 88 viridis nuntius, 68, 88 viridis oreganus, 63-111, 377 viridis viridis, 66, 73, 76 willardi, 121-138 willardi meridionalis, 123-136 willardi silus, 123-136 willardi willardi, 123-138 Cryptolithus goldfussi, 290 tessellatus, 262 Cryptomya californica, 440 Ctena bella, 350 clarionensis, 350 clippertonensis, 350 Ctenopyge flagellifer, 291 Cubas, A. G., 348, 357 Cumingia californica, 440 Cupressus sp., 378, 381-382 Cybele punctata, 296 variolaris, 296 Index to Volume XI 449 Cymatium pileare, 351 vestitum, 351, 362 vestitum insulare, 351 Cyphaspis christyi, 274, 310 elegantula, 293, 295 halli, 303 megalops, 295 yassensis, 303, 314 Cypraea, 350-354 depressa, 351 gillei, 351 Isabella, 351, 353, 362 Isabella mexicana, 351, 362 scurra, 351, 362 teres, 351, 362-363 Cytharella hexagona, 439 D Ball, W. H., 350, 357, 402, 410, 420, 425, 431, 442 Dalmania hausmanni, 299 micrurus, 286 socialis, 299 Dalmanites, 275 anchiops, 286 aspectans, 287 callicephalus, 287 carleyi, 287 concinnus, 285 cuyahogae, 274 dentata, 286 dentatus, 286 limulurus, 285-286 Dalmanitina socialis, 299 Davis, D. Dwight, 68, 110 Davis, W. M., 349, 357 Dechenella verneuili, 295 Delectopecten zacae, 349-350, 362-363 De May, I. S., 378, 384 Dendraster excentricus, 441 Dendroica bryanti castaneiceps, 51 erithachorides castaneiceps, 49-51 erithachorides hueyi, 50-51 erithachorides rhizophorae, 51 Dentalium neohexagonum, 432, 435, 440 pretiosum, 440 Desmanthus fruticosus, 248 Diaphanidae, 423 Diadophis, 33 amabilis similis, 18 Dickey, Donald R., 225 Diodora aspera, 439 inaequalis, 439 Dipleura dekayi, 284 Diplodonta, 407-428 aleutica, 410 orbella, 410 orbellus, 410 Dipodomys, 205-256 agilis, 207-255 agilis cabezonae, 207-238 agilis latimaxillaris, 234 agilis martirensis, 207-247 agilis plectilis, 207-247 agilis simulans, 207-241 arenivagus, 221 deserti, 208-253 deserti deserti, 253 gravipes, 208-255 heermanni, 208-255 insularis, 207-231 margaritae, 207-231 merriami, 207-252 merriami annulus, 207-231 merriami arenivagus, 207-231 merriami brunensis, 207-231 merriami llanoensis, 207-231 merriami melanurus, 207-231 merriami parvus, 207-231 merriami platycephalus, 207-231 merriami quintlnensis, 207-231 merriami semipallidus, 207-231 merriami simiolus, 207-231 merriami trinidadensis, 207-231 paralius, 207-241 parvus, 220 peninsularis, 208-250 peninsularis australis, 208-250 peninsularis eremoecus, 208-250 peninsularis pedionomus, 208-250 peninsularis peninsularis, 208-250 platycephalus, 223 simiolus, 220 Stephens!, 254-255 Dipsosaurus, 380 Ditomopyge scitula, 277 Donax gouldi, 440 Dosinia ponderosa, 437 450 San Diego Society of Natural History Dranga, Ted, 348, 351 Drupa morus, 351-352 ricinus, 351, 363 vicina, 351 Du Bois, E. P., 375, 383 Dulcerana, 350 Dumeril, A. M. C, 22, 33-34 Dunn, Emmett R., 32, 107, 136, 197 Durham, J. W., 347, 431 E Eaton, J. E., 368, 383 Echinometra oblonga, 352 Edwords, C. E., 349, 357 Ekman, S., 353, 357 Elaphrus, 385 Elgaria coerulens, 374 kingii, 135 multicarinata, 369-391 multicarinata webbi, 374 Ellis, A. J., 435, 442 Elschner, C, 349, 357 Emerson, J. W., 431, 441-442 Emerson, W. K., 345-364, 429-444 Encrinurus americanus, 278-279, 311 beaumonti, 297 bohemicus, 297 bowningcnsis, 305 browningi, 306 elegantulus, 279 mitchelli, 305-306 ornatus, 297 punctatus, 296-297 variolaris, 296 Ensis, 393-404 californicus, 395-404, 426 ensis, 401 mexicanus, 398 myrae, 396-404, 426 Entomolithus paradoxus expansus, 291 Entromostracites expansus, 291 punctatus, 296 Eophacops catoosaensis, 289, 313 handwerki, 289 trisulcatus, 289 Epiphragmophora cuyamacensis lowei, 331 Epitonium indianorum, 439 Eshnaur, W. H., 400-401 Eumeces, 369-391 gilberti, 375 lagunensis, 381 skiltonianus, 96, 369, 391 Euphorbia xanti, 248 Fartulum hemphilli, 439 Ficus palmeri, 248 Fischer, Arthur F., 261, 407 Fisher, Edgar R., 395, 407 Fisher, J. W., 117 Fitch, H. S., 18, 34, 107, 381, 385 Fitzinger, Leopold J., 146, 199 Fleming, R. H., 360 Flexicalymene meeki, 282 senaria, 281-282 Foerste, A. F., 261 Forbesia concinna, 293 Fox, R. O., 431, 441-442 Fox, Wade, Jr., 136, 192, 197 Fusitriton oregonensis, 435, 439 G Gaige, Helen T., 32, 107 Gale, H. R., 420, 426, 435, 437, 442 Gardiner, J. S., 357 Gari californica, 440 Garman, S., 34, 68, 108, 146, 199 Geophilida, 118 Geophilus, 118 Index to Volumh XI 451 Gerrhonotus, 34, 374 multicarinatus ignavus, 18 multicarinatus nanus, 94, 96 Gibbs, J. R., Jr., 348 Gifford, E. W., 349, 358 Gill, Theodore, 68, 109 Gilmore, C. W., 376, 383 Girard, Charles, 33, 108 Glans carpenteri, 440 Glasser, M., 349, 358 Gloyd, Howard K., 107, 109, 136, 197, 199 Glycymeris corteziana, 440 profunda, 432, 435, 440 septentrionalis, 432, 440 Goldman, E. A., 219, 246 Goniopleura elegantula, 295 Gopher, Pocket, 53-55 Los Angeles Bay, 54 Rancho Lagunitas, 55 Santa Catarina, 53 Gordon, M., Jr., 402 Gould, A. A., 425 Grant, U. S., IV, 368, 383, 420, 426, 431, 435, 437-438, 442 Gregg, W. O., 333-336, 340 Griffithides bufo, 277, 311 scitula, 277 Grinnell, Joseph, 49, 109, 199, 210, 217, 245, 254, 381, 383 H Hadley, E. E., 368 Halistylus subpupoideus, 439 Hallowell, E., 34, 109, 146, 154, 199 Hamilton, E. L., 347 Haminoea, 422 Hanna, G. D., 347, 350-351, 358, 442 Hancock, G. Allen, 104, 107 Harpcs megalops, 295 trinucleoides, 301 vcnulosus, 290 Harris, G. D., 407 Hartman, Frank A., 137 Hartweg, Norman, 197 Hausmannia meridianus, 307 Hay, O. P., 377, 384 Hedley, C., 353, 358 Heermann, A. L., 146 Heller, Edmund, 131 Helminthoglypta, 329-344 ayresiana, 338 cuyamacensis lowei, 331 lowei, 331-342 phlyctaena, 338 thermimontis, 331-342 walkeriana, 338 Hemicrypturus clintoni, 282 Hemphill, Henry, 363 Hensley, M. Max, 176, 194, 197, 199 Herron, R. B., 220 Hertlein, L. G., 345-364, 402, 431, 435, 437-438, 442 Hill, Howard R., 32, 107, 197, 380, 384, 409 Hipponixbarbata, 351 barbatus, 351-352 fimbriatus, 351 pilosus, 351-353 serratus, 351 Hoard, Robert S., 7, 38, 57 Holbrook, J. E., 65, 68, 109 Homalonotus dekayi, 284 delphinocephalus, 284 Homalopoma carpenteri, 439; Housholder, Victor H., 197 Howard, Hildegarde, 367-368, 379, 384 Howell, B. F., 257-328 Hubbs, Carl L., 38, 90-107, 136, 186, 197, 200, 219 452 San Diego Society of Natural History Huey, Laurence M., 19, 49-51, 53-55, 95-96, 205-256 Hungaia magnifica, 263-264 Hydrographic office, U. S. Navy, 34 Hyla, 369-391 arenicolor, 371 regilla, 371 Hypsiglena, 149, 379 Illaenus armatus, 268 chicagoensis, 267 cuniculus, 269 esmarki, 292-293 insignis, 267 ioxus, 268 katzeri, 292 transvarsalis, 268 transversalis, 268 Ingram, W. M., 350, 358 Iredale, Tom, 250, 254, 359 Irwin, J. B. D., 90 Isotelus gigas, 265-266, 309-310 maximus, 266-267 platycephalus, 318 J Jacquemontia abutiloides, 248 Jan, Georg, 22, 34, 109 Jaton festiva, 439 Johnson, M. W., 360 Jordan, E. K., 401-402 Juniperus californica, 382 K Kanakoff, G. A., 368 Kapp, Kay, 197 Kauffeld, C. E., 126, 135-137 Keen, A. Myra, 348, 395, 401, 407, 437- 439, 443 Kellia laperousii, 432 Kellettia kelletii, 440 Kelsey, F. W., 402, 435, 443 Kew, W. S. W., 368, 385 Klauber, Laurence M., 1-48, 57-116, 121- 204, 367 Klauber, Philip M., 19 Knefastia, 407-428 dalli, 421 princeps, 420-426 Knight, J. B., 69 Kobler, Leslie C., 32, 37-38, 107, 136, 197 Lacroix, A., 349, 359 Lacuna aurantiaca, 412 carinata, 412 compacta, 412 solidula, 412 succinea, 411-412 Lampropeltis getulus, 184, 195, 369-391 californiae, 376 zonatus, 376, 380 Lamprosoma, 146 annulatum, 147, 162 occipitale, 151, 153, 161-162 Larson, Verne, 173 Lee, C. H., 435, 442 Leptopecten latiauratus monotimeris, 440 Leptotyphlops humilis, 191 humilis humilis, 191 Lewis, Fred, 37 Lichanura, 25 roseofusca, 184 Linotaenia laevipes, 118 Linsdale, J. M., 35 Liocalymene clintoni, 282 Littorina pintado, 351 schmitti, 351 scutulata, 440 Lloydia brevis, 264 parva, 263-264, 308 seelyi, 264 Loganellus macropleurus, 263-264, 308 Lovenia cordiformis, 352 Index to Volume XI 453 Loveridge, Arthur, 32, i7, 107, 136, 175, 197 Lowe, Charles H., Jr., 96-97, 102, 107, 197 Lowe, H. N., 331, 350, 359, 398 Lucina nutalli, 440 Lucinisca californica, 440 Lysiloma Candida, 248 M Macoma nasuta, 440 Mactra californica, 440 Magilus robillardi, 352 Marrolithus ornatus, 290 Mason, H. L., 378, 384 Masticophis, 376, 385 flagellum piceus, 184, 195 lateralis, 90 Maxwellia gemma, 440 McCabe, R., 128 McKee, E. D., 117 Mearns, E. A., 35 Meek, Seth E., 35, 109, 137 Megasurcula carpenteriana, 440 Megatebennus bimaculata, 440 Megathura crenulata, 440 Melampus, 436 Menard, H. W., Jr., 354, 359 Mendenhall, Walter C, 200 Merriam, C. H., 210 Merriam, C. W., 413, 426 Merriam, J. C, 368, 379, 384 Metalia spatagus, 352 Micraenigma, 423-424 oxystoma, 423-424 Micruroides euryxanthus, 180 Miller, Alden H., 219, 368, 378-379, 384 Miller, Alice K., 32 Miller, L. H., 368, 378, 384 Miller, Robert P., 200 Mitchell, John, 261 Mitra, 354 idae, 440 papalis, 352 Mitrella carinata, 440 carinata gausapata, 440 Mittleman, M. B., 35 Mocquard, F., 10, 33, 35 MoUusks from Clipperton Island (East- ern Pacific) with the Description of a New Species of Gastropod, 345-364 Morrell, B., 348, 359 Morrison, J. E. P., 359 Morrison, R. L., 436, 443 Morula nodus, 352 papulosa, 352 uva, 352 Mosauer, W., 35, 192-193, 200 Myers, George S., 107, 197 Mytilus californianus, 440 N Nassa californiana, 416 perpinguis, 416 Nassarius, 407-428 catallus, 352, 362 delosi, 438, 440 fossatus, 440 mendicus cooperi, 440 perpinguis, 440 rhinetes, 415-426 Nelson, E. W., 35, 219, 246 Neotoma martinensis, 19 Nerita plicata, 352 Netting, M. Graham, 136, 197 Newell, L M., 437, 443 Nileus armadillo, 292 vigilans, 267 454 San Diego Society of Natural History Norris, Kenneth S., 176, 197 Norrisia norrisii, 440 Nothrotherium, 368 Notices of New West American Marine Mollusca, 405-428 Nuculana taphria, 440 O Ocenebra, 407-428 barbarensis, 415 crispatissima, 414-428 interfossa, 440 lurida aspera, 440 squamulifera, 415 Ochsner, W. H., 349-350, 363 Odontochile applanata, 298 concinnus, 285 hausmanni, 298 meridianus, 307 micrurus, 286 Odontoplrura bowningensis, 303 crcnata, 296 jenkinsi, 304, 314 parvissima, 304 rattei, 304, 314 Odostomia, 432 Ogygites canadensis, 263, 265-266, 308- 309 Oldroyd, Ida S., 401-402 Oldroyd, T. S., 400 Olivella biplicata, 440 boetica, 432 pedroana, 432, 440 Olividae, 418 Opeas oparanum, 352 Ortenburger, A. T., 376, 385 Ostergaard, J. M., 359 Ottley, Allan R., 348 Palmer, Katherine V. W., 407 Pandora punctata, 440 Patinopccten healeyi, 438 Pease, W. H., 359 Peet, Max M., 51 Peltura praecursor, 291 Perkins, C. B., 32, 46, 47, 107, 136, II 197, 201 Perodipus cabezonae, 234 simulans peninsularis, 210, 246 streatori simulans, 234 Peromyscus, 103 maniculatus assimilis, 95 maniculatus exiguus, 19 Perrill, C. K., 349, 359 Peters, James A., 136 Petricola carditoides, 440 Petrunkevitch, A., 117 Phacopidella downingiae, 300-301 downingiae constricta, 301 Phacops bronni, 300 cacapona, 288, 312-313 callicephalus, 287 catoosaensis, 289 caudatus, 307 cristata, 289 cristata pipa, 289 crossleii, 307-308, 314 downingiae, 300 downingiae constrictus, 301 fecundus major, 299 hudsonica, 288, 313 latifroms, 300 latifrons, 300 macroura, 299 musheni, 300 protractus, 300 rana, 288, 312 serratus, 307, 314 socialis, 299 Phasianellae, 412 Philindiae, 423 Phillipsia, 272 bufo, 277 doris, 271 insignis, 275, 311 lodiensis, 274 major, 275. 311 sampsoni, 276, 311 stevensoni, 276, 311 verneuili, 295 Index to Volume XI 455 41, 369-391 1-38, 46, 154, 1-38, 41-47, Phrynosoma, 369-391 coronatum, 369-391 coronatum blainvilli, 373-374 coronatum frontale, 373-374 phrynosoma platyrhinos, 374 Phyllorhynchus decurtatus, 184-185 decurtatus nubilus, 184 decurtatus perkinsi, 184, 191 Pierce, W. D., 368-369, 378-379, 385 Pilsbry, H. A., 340 Pinus, 382 linguiformis, 378 muricata, 378-379, 381 sabiniana, 378 Pituophis, 1-38, 41-48 affinis, 28 annectens, 6, 25 catenifer, 14, 31, catenifer affinis, 184, 376 catenifer annectens, 376 catenifer bimaris, 1-33, 46 catenifer catenifer, 5-28, 41-47, 376 catenifer coronalis, 1-33, 42, 46, 94 catenifer deserticola, 5-31, 154, 376 catenifer fuliginatus, 1-33, 42, 46 catenifer insulanus, 1-37, 42, 46 catenifer pumilus, 41-48 catenifer rutilus, 28 catenifer sayi, 6, 46 catenifer vertebralis, 1-37 deserticola, 5 ruthveni, 5 sayi affinis, 28 Pityophis annectens, 25 catenifer vertebralis, 22 haematois, 22-23 sayi bellona, 14 vertebralis, 7, 22 Platydon cancellatus, 440 Pleurotoma rotula, 420 smithi, 420 Pododesmus macroschima, 441 Poe, Owen J., 431 Polinices lewisi, 440 Pope, Clifford H., 32, 107, 136, 197 Powell, A. W. B., 359 Proctus, 270-273, 283, 310 ascanius, 303 bairdcnsis, 272-274, 310 bohemicus, 294 bowningensis, 302-303 clarus, 271 concinnus, 293-294 crassimarginatus, 271, 310 doris, 27r, 310 elegantulus, 293, 295 ellipticus, 272-274 eremita, 294 haldemani, 272, 310 lejurus, 295 myops, 294 orbitatus, 294 parviusculus, 270, 309 rattei, 303, 314 venustus, 295 Protopeltura praecursor, 291 Protothaca fluctifraga, 441 staminea, 441 Provin, Bruce, 107 Ptcrygometopus carleyi, 287 Puffer, Elton L., 348 Purpura nuttalli, 352 Pusula californica, 440 Quayle, E. H., 90 Quercus agrifolia, 378, 382 Quoyula madreporarum, 352 Rana, 369-391 aurora, 369-391 aurora draytoni, 371-372 Ranella granifera, 350 Rat, Kangaroo, 205-256 Allied, 220 Big Desert, 253 Cabezon, 234 Calmalli, 273 Cape San Lucas, 226 Chapala, 247 Dulzura, 234 El Rosario, 240 Gulf-Coast, 248 Magdalena Plain, 226, 249 Margarita Island, 228 456 San Diego Society of Natural History Mid-Peninsula, 223 San Bernardino, 220 San Bruno, 225 San Felipe, 221 San Francisquito, 224 San Jose Island, 227 San Quintin, 222 San Quintin Broad-faced, 255 Santa Catarina, 24 1 Sierra, 239 Trinidad, 220 Viscaino Desert, 246 Rattlesnake, 61-113, 121-140 Arizona Black, 83 Arizona Ridge-nosed, 125 Chihuahua Ridge-nosed, 128 Coronado Island, 90 El Muerto Island Speckled, 97 Northern Pacific, 67 Pacific, 65 Ridge-nosed, 121-138 San Esteban Island, 104 Southern Pacific, 77 Southern Ridge-nosed, 131 Reedia bronni, 300 Reedops bronni, 300 Reeve, W. L., 385 Regnery, David, 194 Rehder, Harald A., 350, 356, 395, 107 Remopleurides radians, 291 Rett, E. Z., 47 Retusidae, 424 Rhinocheilus, 190-191, 195 lecontei, 184 lecontei clarus, 191 lecontei lecontei, 190 Rhinostoma occipitale, 146, 153 rufofusca, 146 Rhizophora, 49 Richardson, Charles H., Jr., 201 Richardsonella unisulcata, 263-264, 308 Richter, E., 301 Richter, R., 301 Ricinula morus, 352 nodus, 352 Ridgway, Robert, 105, 156, 167, 177, 180, 201, 214 Rodgers, Thomas L., 32, 47, 381, 385 Rogers, Frank L., 348 Rosa minutifolia, 232, 235 RoLs, Clyde P., 201 S Sabrosky, C. W., 69 Salvadora, 195 Sambucus glauca, 378 Sampson, F. A., 276 Sauromalus varius, 106 Saxidomus nuttalli, 441 Scaphandridae, 423 Scaphiopus, 370-371, 379 Sceloporus, 135, 375, 380 graciosus, 373, 380 jarrovii jarrovii, 135 magister, 369-391 occidentalis, 369-391 occidentalis biseriatus, 373 orcutti, 372-373 Schenk, H. G., 437, 443 Schizothaerus nuttalli, 441 Schmidt, Karl P., 32, 35, 68, 103, 107, 110, 136, 197 Schmitt, Waldo L., 32, 107, 136, 197 Schott, A., 146 Schumacher, C. F., 402 Schwenkmeyer, Richard, 107, 136, 194, 197 Scutellum bowningense, 301, 313 jenkinsi, 302, 313-314 longispinifex, 302, 314 partschi, 293, 313 umbellifer, 293 Seeliger, L. M., 381, 385 Seila montereyensis, 440 Semele decisa, 441 pulchra, 441 Index to Volume XI 457 Setzer, Henry W., 210 Shaw, Charles E., 32, 107, 136, 188, 190, 194, 197 Sheppard, W. E., 383 Shreve, Benjamin, 136 Sinusigera, 354 Sistrurus, 109 Skiltonianus, 385 Slevin, Joseph R., 5, 11, 14, 19, 32, 35- 37, 47-48, 97, 103, 107, 110, 136, 146, 173, 197, 201, 347, 350, 360 Smith, A. G., 336, 402 Smith, Gordon M., Jr., 402 Smith, Hobart M., 35, 197, 385 Snails, Mountain, 329-344 Snake, Baja California Gopher, 7 Cedros Island Gopher, 1 1 Colorado Desert Shovel-nosed, 161 Coronado Island Gopher, 19 Mojave Desert Shovel-nosed, 153 Northern Shovel-nosed, 178 Organ Pipe Shovel-nosed, 178 San Diegan Gopher, 25 San Lucan Gopher, 22 San Martin Island Gopher, 14 Santa Cruz Island Gopher, 41 Shovel-nosed, 141-204 Sonoran Gopher, 28 Sonoran Shovel-nosed, 175 Tucson Shovel-nosed, 170 Solen, 395-404 ensis minor, 398 Solenidae, 395-404 Some New and Revived Subspecies of Rattlesnakes, 61-116 Sonora, 145-201 occipitalis, 153, 162 occipitalis annulata, 162 occipitalis klauberi, 170 occipitalis occipitalis, 153 occipitalis palarostris, 175 palarostris, 175 Sordelli, F., 34 Sphaeralcea ambigua, 55 Sphaerrexochus mirus, 285, 298 romingeri, 285 Spisula catilliformis, 441 falcata, 441 Staurocephalus murchisoni, 306 Stebbins, Robert C, 107, 136, 197, 381, 385 Steele, Grover E., 178 Stejneger, L., 35, 110, 137 Stephens, Frank, 431-432, 436, 443 Stickcl, William H., 145-147, 150, 201 Stock, C, 368, 378-379, 385 Storey, Margaret, 32, 47, 107, 197 Streets, T. H., 36, 110 Strong, A. M., 350, 358, 402 Stull, Olive, 5, 9, 36 Sullivan, M., 197 Supernaugh, William R., 176, 178, 190, 197 Sverdrup, H. V., 353, 360 Swarth, H. S., 137 Sykes, Godfrey, 201 T Tantilla, 35, 379 Taras orbellus, 441 Taylor, Edward H., 35, 107, 136, 197, 375, 385 Teall, J. J. H., 349, 360 Tegula aureotincta, 440 Tellina bodegensis, 432, 441 meropsis, 432, 441 Tenebrio molitor, 194 Terron, C. C, 36 Thamnophis, 376, 379 The Amphibians and Reptiles from Rancho La Brea, 365-392 458 San Diego Society of Natural History The Gopher Snakes of Baja California With Descriptions of New Subspecies of Pituophis Catenifer, 1-40 The Kangaroo Rats (Dipodomys) of Baja California, Mexico, 205-256 The Relationship of Crotalus Ruber and Crotalus Lucasensis, 57-60 The Shovel-nosed Snake, Chionactis, with Descriptions of Two New Subspecies, 141-204 The Subspecies of the Ridge-nosed Rat- tlesnake, Crotalus Willardi, 121-140 The Vogdes Collection of Trilobites, 257- 328 Thomomys, 53-55 bottae abbotti, 53-54 bottae borjasensis, 54-55 bottae catavinensis, 53-54 bottae honiorus, 55 bottae rhizophagus, 54 bottae ruricola, 53 bottae russeolus, 54-55 Thompson, Davis G., 201 Three New Races of Pocket Gophers (Thomomys) from Baja California, Mexico, 53-56 Tihen, J. A., 374, 385, 441 Townsend, C. H., 104 Trachycardium quadragenarium, 441 Transennella tantilla, 441 Tremper, R. H., 400 Triarthrus becki, 262 Trifora pedroana, 440 Trilobites, 257-328 esmarkii, 292 orbitatus, 294 ornatus, 290 sternbergi, 298 Trimerus delphinoccphalus, 284 Trinucleus goldfussi, 290 ornatus, 290 Triphora pedroana, 440 Tritonium granifera, 350 granularis, 350 jabick, 350 Trochus pharaonicus, 355 Turbonilla, 432 Turritella, 412-428 cooperi, 413 gonostoma, 437 jewettii, 413 orthosynimetra, 412-426 pedrocnsis, 413 Two CalifornJan Mountain Snails of the Genus Helmlnthoglypta — A Problem in the Relationship of Species, 329-344 U Umbellularia, 381 Uta, 194, 369-391 martinensis, 18 mearnsi, 102-103 stansburiana, 102-103, 369-391 stansburiana hesperis, 96 Valentine, James W., 431 Van Denburgh, John, 5, 18-19, 36, 68, 110, 146, 201 Van Rosscm, A. J., 49-52, 219, 246 Vogdes, General A. W., 257-328 Vogdesia vigilans, 267, 309 Volsella, 407 fornicata, 408, 441 sacculifer, 407-426 Voluta ancilla, 352 deshayesii, 352 Vorhies, Charles T., 136, 197 w Walker, Boyd W., 186, 200 Walker, Lewis W., 15, 96-97, 107 Warbler, La Paz Mangrove, 51 Mangrove, 49-51 San Ignacio Mangrove, 50-51 Sonora Mangrove, 51 Warburg, Elsa, 292 Webb, R. W., 431, 435, 437, 443 Wedekind, R., 301 Index to Volume XI 459 West American Razor-Clanis of the Genus Ensis, 393-404 Wetmore, Alexander, 51 Wharton, W. J., 349, 360 Wheeler, Harry E., 273 Wichterman, Ralph, 69 Willard, Frank C, 125 Willetts, G., 401-402, 409-410, 420, 426, 435, 440 Williams, M. Woodbridge, 423-424 Williams, R. M., 107 Williamson, Mrs. M. B., 400, 402 Wilson, E. N., 443 Wood, Wallace F., 97, 173 Woodbridgea, 422 williamsi, 422 Woodin, William H., Ill, 171, 197 Woodring, W. P., 368, 385, 431, 435, 437, 439, 443 Wossekia katzeri, 292 Wright, A. H., 197 X Xanthium calvum, 378 Xantusia, 369-391 hcnshawi, 374 riversiana, 374 vigilis, 369-391 Y Yarrow, H. C., 110 Yoldia cooperi, 441 Z Zethus bellatula, 279 verrucosus, 279 Zimmerman, Carma R., 348 Zirfaea pilsbryi, 434, 441 Zonaria spadicea, 440 Zweifel, Richard G., 176, 197 Zygospira recurvirostris, 262 p. 210, line p. 224, hne p. 225, line p. 235, line p. 236, line p. 239, line p. 270, line p. 276, line p. 296, line p. 302, line p. 311, line p. 385, line p. 440, line p. 440, line Article no. 1 ERRATA, VOLUME XI 7 Merrian = Merriam 22 Subsp. = subsp. 18 Subsp. ^= subsp. 7 (Rosa minuifoldia) ^ {Rosa minutijolia) 1 1 peralius ^ paralius 30 Adenostroma = Adenostoma 13 hs. = has 35 stevensoni = stevensoni 18 {Brongniart) = (Biongniart) 10 Jenkins =: Jenkins 20 bufo = bufo '■ 38 N.N.Bramlette = M.N.Bramlette 3 Kellettia kelletti = Kelletia kellelii 16 Norrisia nonissi ^ Norrisia norr'tsii 1, plates 1-13 = plates no. 11-23 ^.f Date Due F€#^:«:^|