ISSN 0038-3872 WOUDRMERN CALIFORNIA ACADEMY OF SCIENCES BULLETIN Velume 91 Number 3 BCAS-A91(3) 97-142 (1992) DECEMBER 1992 Southern California Academy of Sciences Founded 6 November 1891, incorporated 17 May 1907 © Southern California Academy of Sciences, 1992 tq OFFICERS June Lindstedt Siva, President Hans M. Bozler, Vice-President David L. Soltz, Secretary Allan D. Griesemer, Treasurer Jon E. Keeley, Technical Editor Gretchen Sibley, Managing Editor BOARD OF DIRECTORS 1990-1992 1991-1993 1992-1994 Jack W. Anderson Allan D. Griesemer Kristine Behrents Hartney Hans M. Bozler Daniel A. Guthrie Lillian Y. Kawasaki Theodore J. Crovello Margaret J. Hartman Gerald M. Scherba Peter L. Haaker Rodolfo Ruibal David L. Soltz June Lindstedt Siva Gloria J. Takahashi Susan E. Yoder Membership is open to scholars in the fields of natural and social sciences, and to any person interested in the advancement of science. Dues for membership, changes of address, and requests for missing numbers lost in shipment should be addressed to: Southern California Academy of Sciences, the Natural History Museum of Los Angeles County, Exposition Park, Los Angeles, California 90007. Annual Members =< .-.s.d0e. Son SS Ae SO Rees Se ee $ 25.00 Student: Members 2: .-. S32) ob ee ee eee 12.50 Lifé:Memibers : 0205.0 he cw ho a A 300.00 Fellows: Elected by the Board of Directors for meritorious services. The Bulletin is published three times each year by the Academy. Manuscripts for publication should be sent to the appropriate editor as explained in “Instructions for Authors” on the inside back cover of each number. All other communications should be addressed to the Southern California Academy of Sciences in care of the Natural History Museum of Los Angeles County, Exposition Park, Los Angeles, California 90007. Date of this issue 17 December 1992 | THIS PUBLICATION IS PRINTED ON ACID-FREE PAPER. Bull. Southern California Acad. Sci. 91(3), 1992, pp. 97-120 © Southern California Academy of Sciences, 1992 New Uintan and Duchesnean (Middle and Late Eocene) Rodents from the Sespe Formation, Simi Valley, California Thomas S. Kelly Vertebrate Paleontology Section, Natural History Museum of Los Angeles County, 900 Exposition Blvd, Los Angeles, California 90007 Abstract.—A paleontologic impact mitigation program being conducted at the Simi Valley Landfill in southern California is yielding new species and new geologic and geographic occurrences of middle and late Eocene rodents representing the families Eomyidae, Heliscomyidae, Simimyidae, and ?Zapodidae from the middle member of the continental Sespe Formation. These rodents include ““Namatomys”’ sp., cf. “N.” fantasma Lindsay, ““Namatomys” sp., Paradjidaumo reynoldsi new species, Heliscomys sp., Simimys landeri new species, and Simiacritomys whistleri new genus and species. Mason (1988) and Kelly (1990) described the mammalian paleontology and biostratigraphy of the continental Sespe Formation along the northern side of Simi Valley, Ventura County, California. Kelly (1990) recognized four superposed middle Eocene local faunas from the middle member of the Sespe Formation. In ascending stratigraphic order, these faunas include the Tapo Canyon Local Fauna (early late Uintan age), the Brea Canyon Local Fauna (late Uintan age), the Strath- ern Local Fauna (latest Uintan or earliest Duchesnean age), and the Pearson Ranch Local Fauna (early Duchesnean age). Kelly and others (1991) reported the preliminary results of a paleontologic resource impact mitigation program that is being conducted in the lower and middle members of the Sespe Formation at the Simi Valley Landfill. They rec- ognized a fifth local fauna, the Simi Valley Landfill Local Fauna of middle or late Duchesnean age, from the uppermost part of the middle member and an unnamed assemblage from the uppermost part of the lower member. The program has yielded many new taxa and geologic and geographic records from the Sespe For- mation that were only briefly discussed by Kelly and others (1991). These new taxa are biostratigraphically significant, especially those of the Simi Valley Landfill Local Fauna. This report describes the rodents of the families Eomyidae, Helis- comyidae, Simimyidae, and ?Zapodidae discovered during the program. Materials and Methods The specimens described herein were recovered from the middle member of the Sespe Formation by a process described by Kelly and others (1991) that included wet screening of bulk matrix samples and heavy liquid separation of fossils. All specimens have been deposited in the Natural History Museum of Los Angeles County. All measurements were made with an AO optical micrometer to the nearest 0.01 mm, and all teeth were measured at their greatest dimensions. Measurements 97 98 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES 7 anterior = (oe) _ —_ 4 10 13a 12 6 13b 15 16 ie 11 g 2 & 12 11 14 12 SESPE RODENTS 99 of the first two upper molars for some species were grouped, as were those for the first two lower molars because of the difficulty in unequivocally assigning isolated teeth to their correct positions in the dental arcade. All metric abbrevi- ations and dental formulae follow standard usage. Dental terminology used herein is presented in Figs. 1 and 2. Institutional acronyms are as follows: LACM, Natural History Museum of Los Angeles County; SBCM, San Bernardino County Museum. Abbreviations for anatomical terms are as follows: A-P, anteroposterior; Ant, anterior; L, left; Post, posterior; R, right; TR, transverse. Other abbreviations are as follows: CV, coef- ficient of variation; Loc., locality; N, number of specimens; OR, observed range; SD, standard deviation. Systematic Paleontology Class Mammalia Linnaeus, 1758 Order Rodential Bowdich, 1821 Family Eomyidae Deperet and Douxami, 1902 *“Namatomys” sp., cf. “N.”’ fantasma Lindsay, 1968 Figure 3, Table 1 Specimens. —RM!°7, LACM 130792; RM!?"?, LACM 130793; RM!°, LACM 130795; RM!°"?, LACM 134684; RM!°?, LACM 130800; LM!°?*7, LACM 130797; LM!°?, 130791; LM!°?7, LACM 130798; LM!°?, LACM 130799; RP,, LACM 130808; RM,,,,, LACM 130804; LM,?, LACM 130807; LM,,,., LACM 130805; RM,, LACM 130803; LM,, LACM 130806. Distribution and age. —LACM Locs. 5857 and 5859, Tapo Canyon Local Fauna, early late Uintan; LACM Loc. 5869, Brea Canyon Local Fauna, late Uintan. Description.—The occlusal surfaces of the upper molars are subquadrate in shape. The anterior cingulum (=anteroloph) is well-developed and extends from the labial aspect of the tooth to the lingual aspect. A distinct anterocone is con- nected by a small accessory crest to the protoloph near the protocone. The para- cone, protocone, metacone, and hypocone are distinct low-crowned cusps with protolophs and metalophs usually forming complete crests that are much lower than the cusps. The posterior arm of the protocone is usually developed labially as a distinct crest and it is commonly forked by additional small spurs. A small mure is present that projects anterolabially from the hypocone towards the pos- terior arm of the protoloph, but rarely connects with this arm. In five molars, the mesostyle is present as a small distinct cusp. A small spur on the mesocone is commonly present and extends lingually. The posterior cingulum (=posteroloph) is a well-developed, robust crest that extends from the posterolabial base of the metacone to the apex of the hypocone. — Fig. 1. Dental terminology in upper cheek teeth of A-““Namatomys” and Paradjidaumo, B-He- liscomys, C-Simimys M'?, D-Simimys M?, and E-Simiacritomys: 1, paracone; 2, metacone; 3, pro- tocone; 4, hypocone; 5, anterocone; 6, mesocone; 7, mesostyle; 8, protoloph; 9, metaloph; 10, mesoloph; li, anterior cingulum; 12, posterior cingulum; 13, mure (=entoloph in Simiacritomys), a—anterior, b—posterior; 14, posterior arm of protocone; 15, hypostyle; 16, protostyle; 17, parastyle; 18, prepro- tocrista; 19, protoconule; 20, anteroconal spur; 21, paralophule; 22, cingular-preprotoconular con- nection; 23, postprotoconular wing; 24, anterior metalophular spur; 25, postprotocrista; 26, postpro- toconular crest; 27, metalophule; 28, preprotoconular wing. SOUTHERN CALIFORNIA ACADEMY OF SCIENCES 100 ~~ “‘pruooejow JO uoTdofo1d [eIqe] ‘ZZ ‘plynsuro [erqejosojue ‘7 ‘prjnBur1o 101101SsO0d Jo uorsus}x9 [eIqe] ‘OZ :Prydojosow [erqe] 6] ‘prydojosow [ensury “gy ‘prynuocoddy ‘7 ] ‘prystio0j01djsod ‘9] ‘pnist100101daid ‘¢ | ‘pruosojoid Jo wie 10119}s0d “py ‘10119}s0d —q “1oLIoJUR — B “(SAWmOODIIG pue SAMIU1S Ut prydoyo}o9=) dN ‘E | “pI[NsuIO JOLI9}sod ‘T] SplynBur1o r0Ls9}ue ‘[{ ‘prydojosow ‘Oy ‘prydojodAy ‘6 ‘prydoyeyou “g ‘pr[Aisosour */ -pruooosoul “9 ‘pruososojue ‘¢ ‘pruosoddy “p ‘pruosojoid ‘¢ ‘pruogojua ‘7 {pruosejoU “| :SAUOJMIDIWIG-_- Pur ‘SAUIUIS-¢ ‘OWNDpIIpDAD_ PUL , SAUOJDUIDNY,,-W JO Y1991 YO9YO JoMoOT Ul ABOTOUTULIO] [RUS = 7 ‘BI JeIqe| = To) 10119} Ue SESPE RODENTS 101 Fig. 3. ““Namatomys” sp., cf. “N.” fantasma Lindsay (A—D) and ““Namatomys”’ sp. (E-H). “‘Na- matomys”’ sp., cf. ““N.” fantasma. A, RM'°?, LACM 130793. B, RM'??, LACM 130792. C, LM,,,., LACM 130805. D, LM;, LACM 130806. ““Namatomys” sp. E, RM!'°?, LACM 131044. F, LM,?, LACM 131057. G, LM,,,., LACM 131058. H, RP*, LACM 131040. All occlusal views; D reversed. Scale = 1 mm. The P, is transversely narrowed and anteroposteriorly elongated. A distinct anteroconid is present. The metaconid and protoconid are positioned close to- gether near the anterior midline of the tooth and are connected by a small complete metalophid. The posterior arm of the protoconid is a low, moderately developed crest that extends posterolabially into the central basin of the tooth. The entoconid and hypoconid are distinct cusps connected by a thin hypolophid (=entolophid). A small posterior cingulid is present that extends from the entoconid to the middle of the tooth, where it connects to the hypolophid. A thin crest is present that connects the metaconid with the entoconid. 102 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Table 1. Measurements (in mm) for teeth of ““Namatomys” sp., cf. N. fantasma Lindsay from Tapo Canyon and Brea Canyon Local Faunas. N Tooth Dimension OR Mean S.D. CV 9 Milor2 A-P .85-1.05 .98 .08 8.3 9 ANT-TR -97-1.15 1.05 .07 6.9 9 POST-TR .97—1.06 1.02 .04 3.4 1 IP. A-P .98 1 ANT-TR a 1 POST-TR 50 3 Mn oes A-P 1.02-1.08 1.05 3 ANT-TR .91-.95 .94 POST-TR .86—1.00 .94 2 M, A-P 1.04-1.15 1.10 D ANT-TR .90-.97 .94 2 POST-TR -90-.93 .92 The first two lower molars are slightly elongated anteroposteriorly and their occlusal outlines are subrectangular in shape. The anterior cingulid is a robust crest with a distinct anteroconid that is connected to the labial aspect of the metalophid by a small accessory crest. The metaconid, protoconid, entoconid, and hypoconid are distinct cusps. The metalophid and hypolophid are usually complete crests, especially in well worn teeth, and are much lower than the primary cusps. The mure is incomplete anteriorly. A short bifurcated mesolophid is present that extends from the mesoconid into the central basin of the tooth. The mesoconid is connected to the hypoconid by a posterior mure. A small mesostylid is usually present. The posterior cingulid is a well-developed crest extending from the en- toconid to the midline of the tooth, where it connects with the hypolophid. The M; is anteroposteriorly elongated and slightly narrower in width posteriorly. The anterior cingulid is a well-developed crest with a small anteroconid that is represented by a bulge at the middle of this crest. In unworn teeth, a small valley separates the anterior cingulid from the protoconid at the anterolabial aspect of the tooth. The metaconid is a sharp cusp that is taller than the moderately well- developed protoconid. The hypoconid is smaller and shorter than the metaconid and protoconid. The entoconid is the smallest of the primary cusps and Is posi- tioned towards the posterolingual corner of the tooth, along a crest that connects to the metaconid. The metalophid is a low crest that connects the metaconid with the protoconid. The mesolophid is a moderately distinct crest that extends lin- gually into the central basin of the tooth. The hypolophid is a thick crest that curves posteriorly from the hypoconid to the entoconid. A small low posterior cingulid is present in one M, at the posterolingual aspect of the tooth. Discussion. —Lindsay (1968) described a new species of eomyid rodent, Na- matomys fantasma of the Hartman Ranch Local Fauna from the Sespe Formation exposed along upper Sespe Creek, north of Simi Valley. Other investigators (Chi- ment 1977; Storer 1984, 1987; Korth 1989) concluded that this new species represented a different genus and should not be assigned to Namatomys Black (1965). Chiment (1977), in an unpublished Master’s thesis, proposed a new generic name, which he and W. W. Korth will publish in a forthcoming report (pers. comm.). Presently all investigators recognize this genus as ““Namatomys.” SESPE RODENTS 103 Table 2. Measurements (in mm) for teeth of ““Namatomys” sp. from Simi Valley Landfill Local Fauna. N Tooth Dimension OR Mean S.D. CV 4 P* A-P 1.04-1.15 1.08 4 ANT-TR 1.05-1.16 1.10 4 POST-TR 1.06-1.14 1.08 6 M! 22 A-P 1.03-1.13 1.08 .04 3.4 4 ANT-TR 1.17-1.28 1.23 4 POST-TR 1.10-1.15 3} 6 IM A-P 1.07—1.20 1.14 .05 4.8 6 ANT-TR 1.05-1.18 1.09 .O7 6.3 6 POST-TR 1.06-1.19 1D .04 4.0 ““Namatomys”’ sp., cf. “N.” fantasma is morphologically very similar to ““Na- matomys” fantasma, but differs in having much smaller teeth. It differs from “N.” fugitivus Storer (1984) of the Swift Current Creek Local Fauna from the Cypress Hills Formation in Saskatchewan by having the following characters: smaller teeth; posterior protoloph-paracone connection in the upper molars less common; and the mesolophid is not strongly forked. It is similar in size to “N.” lacus Storer (1987) of the LacPelletier Lower Fauna from the Cypress Hills For- mation in Saskatchewan, but differs from this species by having the following characters: the connection of the anterocone with the protoloph in the upper molars slightly more labially positioned, resulting in a longer lingual extension of the anterior cingulum from the anterocone; the mure in the upper molars is much less complete; and the ectolophids of the M, and the M, are more complete. It differs from “‘N.”’ sp. (see below) from LACM Loc. 5876, which occurs strati- graphically much higher in the section at the Simi Valley Landfill, by having the following characters: smaller teeth with lower crown height; the crests (lophs) are lower and less prominent; the posterior arm of the protocone is commonly bi- furcated; and the mesoloph is slightly more complex with additional small spurs present. It is similar in size to an unnamed species of ““Namatomys”’ described by Chiment (1977) from the Santiago Formation in the San Diego area of Cali- fornia. The “Namatomys’’ material from the Tapo Canyon and Brea Canyon Local Faunas appears to represent a new species most closely related to ““Namatomys” fantasma. However, until Chiment and Korth publish a formal description of the unnamed species from the San Diego area and comparisons with that species can be made, the Simi Valley species is herein assigned to “‘N.”’ sp., cf. “NN.” fantasma. ““Namatomys” sp. Figure 3, Table 2 Specimens. —RP*, LACM 131040; RP*, LACM 132438; LP*, LACM 131451; LP*, LACM 132448: RM!°?, LACM 131032; RM!°?, LACM 131044; RM!°?, LACM 130849; RM!°?, LACM 131033; LM!°*7, LACM 131055; partial RM'°”, LACM 132653; RM,?, LACM 131045; RM!27, LACM 131046; RM,,,., LACM 131060; LM,,,., LACM 131057; LM,,.5, LACM 131058; LM,..., LACM 131059. Distribution and age. —LACM Loc. 5876, Simi Valley Landfill Local Fauna, late Duchesnean. 104 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Fig. 4. Paradjidaumo reynoldsi new species. A, LP*, LACM 131036. B, LM'=*, LACM 131078. C, RM?, LACM 131449. D, LM,,.,, LACM 130846. E, LM,,.., LACM 131048. F, LM,?, LACM 131028. G, RP., LACM 131052. H, Holotype, RP., LACM 131042. L LP.. LACM 131064. All occlusal views; A, B, C, and F reversed. Scale = 1 mm. Description.—The occlusal outline of the P* is almost square. The anterior cingulum is robust with a small crest connecting it to the middle of the protoloph. The protoloph is a complete crest and usually well connected to the paracone. In one tooth, the protoloph is only weakly attached to the paracone. The metaloph is a complete crest connecting the metacone with the hypocone. The protoloph and the metaloph are much lower than the primary cusps. The paracone, pro- tocone, metacone, and hypocone are tall slender cusps that have increased crown height. This increase in crown height is the result of an increase in cusp height alone, whereas the base of the crown remains low. The mure is complete, con- necting the mesocone with the protocone and the hypocone. The posterior arm of the protocone is a distinct crest that connects to the posterolabial base of the paracone. The mesoloph varies from a very small spur to a moderately developed, thin, short crest. A distinct mesostyle is present. The posterior cingulum is a robust crest extending from the metacone to the hypocone. The first and second upper molars are similar in morphology to the P*, but exhibit the following differences. They are more transversely expanded. The an- SESPE RODENTS 105 terocone is slightly more robust and positioned further labially. The protoloph and metaloph are complete low crests that are slightly more developed. The mesoloph is usually absent, but, when present, is a very small short spur. The posterior arm of the protocone is usually not present, but, when present, it is a very short crest extending labially, but not joining the paracone. The mesostyle is present as a smaller cusp. The lower molars are represented by the M, and M, and exhibit the following characters. The anterior cingulid is well-developed and connected to the proto- conid. The metaconid, protoconid, entoconid, and hypoconid are distinct tall slender cusps. The protolophid and metalophid are complete and lower in height than the primary cusps. The mure is usually a complete crest connecting the mesoconid with the protoconid and the hypoconid. The protoconid of one tooth, presumably M,, has a well-developed posterior arm that extends to and connects with the posterolabial aspect of the metaconid. Apparently the posterior arm of the protoconid is not present in the M,. The mesolophid is a simple, thick crest that projects a short distance labially from the mesoconid and then turns poste- riorly towards the metalophid. The mesostylid is absent or sometimes present as an incipient cusp on the crest between the metaconid and entoconid. The posterior cingulid forms a thick crest that extends labially from the posterolingual corner of the entoconid to the midline of the tooth, where it connects with the metalophid. Discussion. —“‘Namatomys”’ sp. of the Simi Valley Landfill Local Fauna differs from “‘N.” sp., “N.”’ fantasma of the Tapo Canyon and Brea Canyon Local Faunas by having larger teeth, increased height of the primary cusps, greater development of the crests, and a simpler mesoloph. The increase in crown height and the more prominent crests indicate that “NV.” sp. is more derived than “N.”’ sp., cf. “N.” fantasma. The material of ““Namatomys” from LACM Loc. 5876 probably rep- resents a new species, but until a better sample is available from the Sespe For- mation and this sample can be adequately compared with the samples of *“Na- matomys”’ from the greater San Diego area to be described by Chiment and Korth, it is herein assigned to an unnamed species of ““Namatomys.” Paradjidaumo Burke, 1934 Paradjidaumo reynoldsi new species Figure 4, Table 3 Paradjidaumo n. sp. Kelly and others, 1991:7, 12. Holotype. —RP,, LACM 131042. Type locality. —LACM Loc. 5876. Diagnosis. —Difters from Paradjidaumo trilophus (Cope, 1873) (=P. nasutus Cope and P. minor Douglass), P. spokanensis White (1954), P. hansonorum Russell (1972), P. hypsodus Setoguchi (1978), and P. validus Korth (1980) by having smaller teeth (averaging 12%, 22%, 15%, 10%, 29% smaller, respectively). Further differs from P. trilophus by having P, less molariform and mesolophids with less tendency to join entoconids with wear. Further differs from P. spokanensis by having the following characters: P* smaller relative to molars; P, more antero- posteriorly elongated; and mesolophid of M, and M; not connected to entoconid. Further differs from P. a/berti Russell (1954) by having the following characters: P,, longer, wider posteriorly, and larger relative to molars; P, mesoconid connected by mure to hypoconid; and lower molars less elongated anteroposteriorly, resulting 106 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Table 3. Measurements (in mm) for teeth of Paradjidaumo reynoldsi new species from Simi Valley Landfill Local Fauna. ST ——————————————————————————————————— N Tooth Dimension OR Mean S.D. CV 5 jz A-P 1.10—1.25 1.19 .06 5.4 5 ANT-TR 1.10—1.24 1.18 .O7 5.9 5 POST-TR 1.08—-1.24 1.16 .06 4.9 11 M!° A-P 1.17—1.32 P25 06 4.9 10 ANT-TR 1.17—1.47 1.38 11 7.6 8 POST-TR 1.25-1.45 1.36 -06 4.4 5 M3 A-P -95-1.01 95 -05 BY) 5 TR .99-1.11 1.05 05 Sit 5 [2 A-P 1.16—-1.36 1.26 .09 as 5 ANT-TR .83—1.03 .88 -09 9.6 5 POST-TR 1.08—1.30 [p22 .O9 7.6 15 Mass A-P 1.07-1.37 1.24 .O7 5.9 11 ANT-TR 1.00—1.26 1.13 .09 8.0 13 POST-TR 1.07-1.29 1.20 .O7 6.0 3 M, A-P P= C35 77) 3 ANT-TR 1.10—1.28 [ Faj IS 3 POST-TR 1.16—1.17 1.16 in squarer occlusal outlines. Further differs from P. hansonorum by having P, not square in occlusal outline, but with anterior width much smaller relative to pos- terior width. Further differs from P. hypsodus by having the following characters: P., wider transversely relative to anteroposterior length: and cheek teeth with lower crowns, valleys between cusps and crests shallower, and cusps more prominent relative to crests. Further differs from P. validus by having the following characters: P, less molariform; greater development of cingulids in lower molars; and lack of accessory lophid joining mesolophid and hypolophid on P, through M,. Etymology. —Named in honor of Robert E. Reynolds, Curator of Earth Science, San Bernardino County Museum, in recognition of his extensive work in the recovery of fossil microvertebrates from southern California and his contributions to the Simi Valley Landfill Paleontologic Resource Impact Mitigation Program. Referred specimens. —RP*, LACM 131041: RP*, LACM 131036; LP*, LACM 131038; LP*, LACM 131037; LP*, LACM 131079; RM!°7, LACM 130843; RM!°”, LACM 130844; RM!°?, LACM 130847; RM!°?, LACM 131034; RM'e?, LACM 131035; RM!°, LACM 131068; RM!°, LACM 131078; RM!°7, LACM 131031; RM!°?, LACM 131456; RM!°?, LACM 132437; LM!°, 130842; LM'e?, LACM 130848; LM!°7, LACM 131027; LM!°*, LACM 132444; LM!°*, LACM 132457; RM?, LACM 130840; RM?, LACM 131077; RM3, LACM 131450; LM?, LACM 131449: LM?, LACM 131463; RP,, LACM 131052; RP,, LACM 132443; RP,, LACM 132446; LP,, LACM 131064; LP,, LACM 131051; LP,, LACM 131053; LP,, LACM 132435; RM,,,., LACM 131056; RM,,,., LACM 130845; RM,, LACM 130850; RM,, LACM 131039; RM,, LACM 131049; LM,, LACM 130846; RM,, LACM 131043; RM,,,,. LACM 130851; RM,,., LACM 131054; partial RM,,,,, LACM 131063; LM,,,,, LACM 131048; LM,,... LACM 131061; LM.o4, LACM 131453; LM,,,., LACM 131454; LM,,,,, LACM 131455; LM,, LACM 130841; LM;, LACM 131028; LM;, LACM 131050. SESPE RODENTS 107 Distribution and age.—LACM Loc. 5876, Simi Valley Landfill Local Fauna, late Duchesnean. Description.—The occlusal outline of the P* is almost square. The anterior cingulum is a low and weakly developed crest that usually connects with the lingual aspect of the protoloph. The anterior cingulum is not a distinct loph on the occlusal surface because of its small size and low height, and gives the tooth a four-crested pattern. The paracone, protocone, metacone, and hypocone are rounded cusps that have increased crown height through an increase in the height of the crown base. The protoloph and metaloph are complete, thick, high crests that connect the paracone with the protocone, and the metacone with the hypo- cone, respectively. The mesoloph is usually a long, high, well-defined crest ex- tending to the lingual aspect of the tooth. The mure is a complete high crest. The posterior cingulum is a well-developed, high crest that extends from the metacone to the hypocone. The first two upper molars are similar to the P* except for the following dif- ferences. They are expanded transversely and have a strong five-crested occlusal pattern. The anterior cingulum is well-developed forming a robust, high crest that extends lingually from anterolabial aspect of the paracone to the midline of the tooth, where it connects with the protoloph. The mesostyle is present as a small cusp in five molars. The occlusal outline of the M? is subtriangular, with the posterior aspect trans- versely narrowed. The anterior cingulum is a well-developed crest that extends lingually from the anterolabial aspect of the paracone to the midline of the tooth where it connects with the protoloph. The five-crested occlusal pattern is usually well-developed. The paracone is the tallest primary cusp. The protocone 1s a thick, rounded cusp that is positioned towards he center of the lingual aspect of the tooth. The protocone usually is connected to the hypocone. A small mesostyle is sometimes present. The protoloph and metaloph are complete, thick, high crests. The mesoloph is a high, distinct crest extending to the labial aspect of the tooth. The posterior cingulum is a distinct crest. The P, is transversely narrowed and anteroposteriorly elongated. A small an- terior cingulid is present that connects to the protoconid. The protoconid and metaconid are positioned medially and connected posteriorly by a complete meta- lophid. The hypolophid usually is a complete, high crest that connects the en- toconid with the hypoconid, except in one tooth, wherein the hypolophid is divided by a shallow cleft that is labial to the connection of the posterior cingulid and the hypolophid. The mesolophid usually is a long, well-developed crest that extends lingually to a distinct but small mesostylid. The posterior cingulid is a well-developed, thick crest that extends labially from the posterolingual corner of the entoconid to the midline of the tooth, where it connects with the center of the metalophid. The M, and M, are slightly anteroposteriorly elongated and their occlusal out- lines are subquadrate to subrectangular. The anterior cingulid extends labially from the anterolingual corner of the tooth and connects with the anterolabial corner of the protoconid. The metaconid, protoconid, entoconid, and hypoconid are conical cusps that have increased crown height through an increase in the height of the crown base. The metaconid and entoconid are taller than the pro- toconid and hypoconid. The metalophid and hypolophid are complete, high crests. 108 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES The mure is a high, complete crest connecting the mesoconid with the protoconid and hypoconid. The mesolophid is usually a long, high, well- defined | crest that extends to or near the lingual aspect of the tooth. The posterior cingulid isa moderately high crest that extends labially from the posterolingual aspect of the entoconid to the midline of the tooth where it connects with the center of the hypolophid. Three teeth are questionably considered M,s. These teeth are similar to the M, and M,, except for the following differences. The hypolophid is a more rounded crest and the posterior cingulid is smaller and lower in height. Discussion. —Burke (1934) named and described Paradjidaumo and docu- mented the unique attachment of the anterior cingulid to the anterolabial aspect of the protoconid in the lower molars of this genus. The teeth described above can be confidently assigned to Paradjidaumo because they exhibit the following characters: well-developed crests (=lophs) that are nearly as high as the primary cusps: the mesolophs and mesolophids of the molars are high and complete, and form a five-crested occlusal pattern: and the lower molars possess the unique attachment of the anterior cingulid to the protoconid. Paradjidaumo is primarily known from faunas of Chadronian to Orellan age. The only previous record of the genus from the Duchesnean was documented by Russell (1954). who described P. a/berti from the Kishenehn Formation of British Columbia. Paradjidaumo reynoldsi can be easily distinguished from P. alberti by its relatively larger P, that has the mesoconid connected by the mure to the hypoconid and its squarer lower molars. Paradjidaumo reynoldsi is restricted to the Simi Valley Landfill Local Fauna and this occurrence represents the first record of the genus from the Sespe Formation and the second record of the genus in the Duchesnean. Family Heliscomyidae Korth, Wahlert. and Emry, 1991 Heliscomys Cope. 1873 Heliscomys sp. Figure 5 Specimens. —RM', LACM 131452: LM*, LACM 132456. Distribution and age. —LACM Loc. 5876, Simi Valley Landfill Local Fauna, late Duchesnean. Description. —The two upper molars are the only teeth of Heliscomys thus far recovered during the impact mitigation program. They are well preserved and only slightly worn. The measurements of LACM 131452 are 0.70 mm A-P, 0.56 mm ANT-TR, and 0.59 mm POST-TR. and those of LACM 132456 are 0.82 mm A-P, 0.73 mm ANT-TR, and 0.81 mm POST-TR. The occlusal outlines of the upper molars are square. Each tooth has six cusps arranged in two rows of three cusps each. The paracone, metacone. protocone, hypocone are well-defined conical cusps. The paracone and the metacone are slightly taller than protocone and hypocone. Moderately deep valleys separate the cusps and the lingual cin- gulum from the hypocone and protocone. The median transverse valley is open lingually. The hypostyle (=entostyle of Black, 1965) is a distinct cusp on the lingual cingulum, lower than the hypocone, and separated from the protostyle by a small cleft that is an extension of the median transverse valley. The protostyle is weakly expressed as a small incipient cusp on the lingual cingulum. The anterior SESPE RODENTS 109 E F G Fig.5. Heliscomys sp. (A) and Simimys landeri new species (B—G). Heliscomys sp. A, RM', LACM 131452. Simimys landeri. B, Holotype, RM', LACM 131062. C, RM?, LACM 130759. D, RM3, LACM 131458. E, RM,, LACM 130771. F, RM,, LACM 130765. G, LM,, LACM 130767. All occlusal views; A—F reversed. Scale = 1 mm. cingulum is a thin crest separated from the paracone and the protocone by a shallower valley than the median transverse valley, and is continuous with the lingual cingulum. The posterior cingulum is moderately developed with a very small incipient cusp in the center of the cingulum, and it is separated from the metacone and the hypocone by a shallower valley than the median transverse valley. Discussion. —The upper molars from Simi Valley are assigned to Heliscomys because of the presence of six cusps that are positioned in two transverse rows of three cusps each, the square shape of the occlusal surface, and the conical shape of the primary cusps, which are separated by distinct valleys. The Heliscomys teeth from Simi Valley are very similar to those of H. sp., cf. H. vetus Cope (1873) from the Pipestone Springs Local Fauna (Black 1965) of the Renova Formation in Montana (Chadronian age) and the Pilgrim Creek Local 110 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Fauna (Sutton and Black 1975) of the Jackson Hole area in Wyoming (Chadronian age). It differs from H. sp., cf. H. vetus and all other species of Heliscomys by having the following characters: the M'* are slightly smaller; the primary cusps are subequal in height; the protostyle and the hypostyle are lower in height than the primary cusps; the posterior cingulum is slightly more developed with an incipient cusp present; the median transverse valley is slightly shallower; the valley between the anterior cingulum and the paracone and the protocone is slightly shallower; and the valley separating the posterior cingulum and the metacone and the hypocone is slightly shallower. Although the sample of H. sp. consists of only two teeth, all of the above characters indicate H. sp. is less derived than the other species of Heliscomys, as would be expected considering its earlier occurrence in the Duchesnean. Storer (1988) described a premolar, which he assigned to Helisco- mys sp., from the Lac Pelletier Lower Fauna (late Duchesnean age) of Saskatch- ewan. It is difficult to compare the Simi Valley teeth of Heliscomys with that from Saskatchewan because they represent different teeth. The Simi Valley teeth prob- ably represent a new species, but, until a larger sample is available, they are assigned to an unnamed species of Heliscomys. Heliscomys sp. is restricted to the Simi Valley Landfill Local Fauna and this occurrence is the first record of the genus and the family Heliscomyidae from the middle member of the Sespe Formation. Furthermore, the specimens of H. sp. from Simi Valley represent in the second record of the genus in the Duchesnean. Family Simimyidae Wood, 1980 Simimys Wilson, 1935a Simimys landeri new species Figure 5, Table 4 Simimys n. sp. Kelly and others, 1991:7, 12. Holotype.—RM!', LACM 131062. Type locality. —LACM Loc. 5876. Diagnosis. — Differs from other species of Simimys by having the following characters: larger (15% to 41% in tooth measurements); cheek teeth with less development of small accessory crests and stylids; M? hypocone more prominent; and M, lingual metalophid bifurcated at lingual aspect with one end connecting to metastylid and other to base of entoconid. Etymology.—Named in honor of E. Bruce Lander of Paleo Environmental Associates, Inc. and the Natural History Museum of Los Angeles County, in recognition of his efforts directing the Simi Valley Landfill Paleontologic Resource Impact Mitigation Program that resulted in the recovery of large samples of microvertebrates from the Sespe Formation. Referred specimens. —RM!, LACM 130760; RM!, LACM 130762; RM!, LACM 130769; LM!, 130772; LM?, LACM 130773; RM2, LACM 130759; RM?, LACM 130761; RM?, LACM 134685; RM?, LACM 131458; LM?, LACM 131459; LM?, LACM 132439; LM?, LACM 130768; LM,, LACM 131070; RM,, LACM 130771; RM,, LACM 132434; partial RM,, LACM 132454; RM,, LACM 130765; LM,, LACM 130766; LM,, LACM 132453; RM,, 130763; LM;, LACM 130767; LM,, LACM 132450; LM,, LACM 132455; RM,, LACM 130770. Distribution and age.—Type locality, Simi Valley Landfill Local Fauna, late Duchesnean. SESPE RODENTS 111 Table 4. Measurements (in mm) for teeth of Simimys landeri new species from Simi Valley Landfill Local Fauna. N Tooth Dimension OR Mean S.D. CV 5 M! A-P 1.70-1.94 1.81 11 5.8 5 ANT-TR 1.43-1.58 1.49 .06 4.2 5 POST-TR 1.47-1.66 1.56 .O7 4.8 4 M? A-P 1.76-1.81 1.79 4 ANT-TR 1.56—-1.64 1.60 4 POST-TR 1.49-1.58 1.52 4 M3 A-P 1.24-1.31 1.28 4 TR 1.15-1.24 1.21 3 M, A-P 1.76-1.85 1.80 3 ANT-TR 1.05-1.14 1.10 4 POST-TR 1.24—-1.40 1.32 4 M, A-P 1.75-1.91 1.85 4 ANT-TR 1.28-1.50 1.40 4 POST-TR 1.40-1.56 1.46 4 M, A-P 1.52-1.64 1.58 4 ANT-TR 1.23-1.34 1.29 4 POST-TR 1.04-1.18 1.13 Description. —The M' ts slightly elongated anteroposteriorly and the occlusal outline is subrectangular in shape. The paracone, metacone, protocone, and hy- pocone are well-developed cusps. The mure is complete and connected to the protocone and the hypocone. The protoloph 1s a well-developed crest joining the mesocone with the paracone. The mesoloph is a complete crest that extends from the mesocone to the mesostyle. The mesostyle is usually weakly expressed, but in some teeth, it is a small distinct cusp. The metalophule is absent or expressed as a thin crest. The parastyle is represented by a small bulge on the labial aspect of the anterior cingulum. The anterocone is a distinct cusp on lingual aspect of the anterior cingulum. In unworn teeth, the protoconule is a distinct and well- developed cusp that is positioned along the crest between the protocone and paracone. The anteroconal spur is absent to moderately developed. The paraloph- ule, postmesoconal spur, and enterostyle are absent. The M? is anteroposteriorly elongated and the occlusal outline is subrectangular. The anterocone is a small distinct cusp. An anteroconal spur is absent to well- developed. The parastyle is a moderately to well-developed cusp. In unworn teeth, a cleft is present that separates the parastyle and the anterior cingulum, except near the base of the crown. The protoconule is a distinct cusp. The protoloph is a distinct crest that joins the mesocone with the paracone. The mesoloph is a distinct crest that extends lingually to the mesostyle. The metalophule usually is well-developed. A small accessory cusp sometimes is present on the mesoloph near the mesostyle. The anterior and posterior cingulae are robust. The paraloph- ule, postmesoconule spur, and enterostyle are absent. The M3? has a subtriangular occlusal outline. The protocone 1s a well-developed cusp. The metacone is a moderately well-developed cusp. The hypocone is pos- teriorly positioned and moderately developed as a distinct expansion on the posterolingual aspect of the tooth. The protoconule is small. The paralophule is well-developed. The metalophule is a well-developed thick crest in moderately 112 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES worn teeth. The mesostyle is weakly developed. The cingular-preprotoconular connection is strongly developed. The anterocone is a distinct cusp witha | complete anteroconal spur that connects the anterocone with the preprotocrista. A small postprotoconular wing extends posteriorly a short distance towards the postpro- tocrista. The postprotocrista extends anterolingually into the central basin of tooth and a small anterolabial bifurcation is present about half-way along the postpro- tocrista that is directed towards, but does not connect with, the postprotoconular wing. The anterior metalophular spur connects the metalophule with the post- protocrista. Small depressions are present in the enamel of the central basin of the tooth. The M, is anteroposteriorly elongated and the occlusal outline is subrectangular in shape. The anteroconid (=preprotoconid of Lillegraven and Wilson 1975) is usually weakly developed, but in one tooth it is a distinct cusp. A preprotocristid is present that connects the protoconid with the anteroconid. The protoconid is connected to the metaconid by the postprotocristid. A small cusp-like swelling is present in the center of the postprotocristid. The entoconid and the hypoconulid are well-developed cusps. The mesoconid is robust with short. but distinct. labial and lingual mesolophids present. The hypolophid is a well-developed crest that connects the entoconid with the hypoconid. The mesostylid is well-developed and a labial extension of mesostylid is present. The posterior ectolophid connects the hypoconid with the mesoconid. The posterior cingulid is a well-developed crest that extends from the posterolingual aspect of the entoconid to the hypoconid. There is no labial extension of the posterior cingulid. The M, is anteroposteriorly elongated and the occlusal outline is subrectangular. The protoconid, metaconid. entoconid. and hypoconid are well-developed cusps. The mesoconid is large and robust. The hypoconulid is well-developed and at- tached to the prominent posterior cingulid. The lingual mesolophid is robust and extends to base of the mesostylid. The labial mesolophid is developed as a rounded projection extending from the mesoconid. The postprotocristid is a single crest with no bifurcation and joins the mesostylid with wear. The metalophid is a complete crest extending between the preprotocristid and the metaconid. A dis- tinct anterolabial cingulid is present. The premetastylid is absent. The M, is anteroposteriorly elongated, posteriorly narrowed, and the occlusal outline is subrectangular. The protoconid, metaconid, entoconid, and hypoconid are well-developed cusps. The labial mesolophid is short and does not extend below the hypoconid. In some teeth. a very small cross-crest is sometimes present on the labial mesolophid. The lingual metalophid is bifurcated at its lingual end, with one spur of the bifurcation connecting to the mesostyid and the other spur to the base of the entoconid. The ectostylid is absent. Discussion. — Wilson (1935a) described two rodents from the Sespe Formation as Eusmysops simplex and E. vetus. Wilson (1935b) then amended the generic name to Simimys for these two species. Wilson (1949) named a third species, S. murinus, from the Sespe Formation and described additional material, including a partial skull. Lillegraven and Wilson (1975) and Walsh (1987) proposed syn- onymizing S. vetus and S. murinus with S. simplex. The impact mitigation pro- gram at the Simi Valley Landfill has yielded many large superposed samples of S. simplex, and analysis of these samples will be described in a separate paper. Simimys landeri can be easily distinguished from S. simplex by its much larger SESPE RODENTS 113 teeth. The observed ranges of all tooth measurements for S. /anderi do not overlap those for S. simplex. Furthermore, most of the tooth measurements for S. /anderi are five or more standard deviations away from the means of the respective tooth measurements for S. simplex. The teeth of S. /anderi also appear to be less complex than those of S. simplex, with a reduction in the number of small crests and stylids. Simimys landeri further differs from S. simplex by having a better de- veloped M? hypocone and a bifurcated M, metalophid that connects lingually to the metastylid and entoconid. Simimys landeri is only known from the Simi Valley Landfill Local Fauna (late Duchesnean age) at LACM Loc. 5876, whereas S. simplex is known from the Tapo Canyon, Brea Canyon, Strathern, Pearson Ranch, and Simi Valley Landfill Local Faunas (late Uintan to late Duchesnean). Lillegraven and Wilson (1975) analyzed a large sample of S. simplex from the Camp San Onofre Local Fauna of the Santiago Formation in the Oceanside area, California, and found that almost all dental characters studied were highly variable and unreliable for separating species of Simimys. They found this to be particularly true for the following characters: the amount of metaconid development in the lower molars; the development of mesostylids in the lower molars; the presence of a mesolophid connection to the protoconid; the anterior ectolophid connection to the protoconid; and the morphology of M? and M,. The characters used herein to separate S. simplex from S. landeri were not discussed by Lillegraven and Wilson (1975) and presumably are diagnostic. However, even if future study indicates some of these characters are taxonomically unreliable for species diag- nosis, the much larger size of S. /anderi still warrants specific separation from S. simplex. The ambiguous nature of the morphological characters in Simimys has resulted in different familial assignments. Klingener (1964) and Lindsay (1968) suggested Simimys exhibited muroid characters and may belong to the Cricetidae, whereas Wood (1937), Stehlin and Schaub (1951), Klingener (1963), and Lillegraven and Wilson (1975) questionably assigned Simimys to the Zapodidae. Wood (1974, 1980) considered Simimys to be too derived to be ancestral to later zapodids and assigned Simimys to the superfamily Dipodoidea, family Simimyidae, whereas Emry (1981) referred Simimys to the superfamily Muroidea, family incertae sedis. Emry and Korth (1989) considered the loss of the P* in Simimys to possibly warrant its assignment to a separate family closely related to the Zapodidae and under the Dipodoidea. Wood (1980) presented the most convincing evidence that Simimys represents a distinct family of the Dipodoidea and this systematic as- signment is followed here. Family ? Zapodidae Coues, 1875 Simiacritomys new genus Type species. —Simiacritomys whistleri new species. Range. —Late Eocene (late Duchesnean) of southern California. Diagnosis. —Molars five-lophed with robust, rounded cusps; deep median trans- verse valley that divides tooth into anterior and posterior sections usually present in first two upper and lower molars; anterocone well-developed; M' and M7? protoloph usually divided, midway along its length, by cleft; M, and M, metaloph- id divided, midway along its length, by deep valley separating metaconid and protoconid; hypolophid thick, prominent crest connecting hypoconid and ento- 114 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES conid; mesoloph and mesolophid usually long, high, well-developed crests strongly connected to mesocone and mesoconid, respectively: anterior entoloph small anteriorly directed crest not connecting with protoloph; posterior entoloph thick, high, well-developed crest connecting hypocone with mesocone; ectolophids in- complete: distinct hypoconulid present; and cheek tooth enamel thick. Referred species. —Type species only. Etymology. — Simi, in reference to its occurrence in Simi Valley; acritos, Greek for mixed or confused: mys, Greek for mouse. Discussion. —The familial relations of Simiacritomys are uncertain. The teeth of Simiacritomys exhibit similarities to those of eomyid and zapodid rodents. The molars are similar to eomyids by having a five-crested occlusal pattern with a similar positioning of the crests and cusps. However, they differ from all eomyid genera by having the following characters: the first two upper and lower molars are divided into anterior and posterior portions by deep transverse valleys; the protoloph usually divided, midway along its length, by a cleft that separates the paracone from the protocone; and the metalophid is divided or not developed because of a deep valley between the metaconid and the protoconid. The cheek teeth of Simiacritomys are similar to those of the zapodids Plesiosminthus Viret (1926) and Megasminthus Klingener (1966) by having a five-crested occlusal pattern, the mesoloph and the mesolophid usually long and well-developed, and the presence of a hypoconulid on the lower molars. In particular, the upper molars of Plesiosminthus grangeri (Wood, 1935) figured by Green (1977, p. 1001, Figs. 3E, 3G) are strikingly similar to those of Simiacritomys, especially with regard to the presence of deep transverse valleys that separate these teeth into anterior and posterior portions. Plesiosminthus differs from Simiacritomys by having com- plete protolophs and metalophids, more complete entolophs and ectolophids, and more anteroposteriorly elongated lower molars. Megasminthus differs from Simi- acritomys by having cheek teeth with mesostyles and mesostylids, less anteriorly directed metalophs, less isolated metaconids, and more complete ectolophids. The molar occlusal patterns of Simiacritomys exhibit some similarities to those of the Bridgerian ?zapodid Elymys Emry and Korth (1989) and to those of the simimyid Simimys. The molars of Simiacritomys differ from those of Elymys by having squarer occlusal outlines, mesolophs and mesolophids, and incomplete protolophs, metalophids, entolophs and ectolophids. The teeth of Simiacritomys differ from those of Simimys by having molars with more quadrate (less antero- posteriorly elongated) occlusal outlines, a central transverse valley dividing the molars into anterior and posterior portions, upper molars with transversely ori- ented (less oblique) crests and lacking the W-shaped occlusal pattern, the anterior cingulum in the upper molars does not project lingually from the anterocone, no protoconules or mesostyles, and a valley present that separates the metaconid from the protoconid. Storer (1988, p. 98, Fig. 4) described and figured an upper molar of an unde- termined genus and species of rodent from the Lac Pelletier Lower Fauna of Saskatchewan (late Duchesnean age). This tooth is similar to those of Simiacri- fomys in having a five-crested occlusal pattern, a transverse valley dividing the tooth, the anterior cingulum connected with the protocone and extending to the anterolingual corner of the tooth, and the hypocone strongly connected to the mesoloph. The upper molars of Simiacritomys differ from this tooth by having SESPE RODENTS 115 a divided protoloph, a complete well-developed metaloph, and no mesostyle. Storer (1988) noted the similarity of the tooth from Saskatchewan with those of Plesiosminthus and with an upper molar described by Dawson (1966, p. 113) from Duchesne River Formation in the Uinta Basin of Utah (Randlett Fauna, Uintan age). Dawson (1966) assigned the tooth from Utah to “?sciuravid or myomorph sp.” and noted some similarities between this tooth and those of Sciuravus Marsh (1871) and Simimys. The teeth of Simiacritomys are similar to this tooth by having the protocone and the hypocone separated by a distinct valley, a long mesoloph, an incomplete entoloph that projects anteriorly from the mesocone but does not connect with the protoloph, a central transverse valley dividing the tooth into anterior and posterior portions, and a similar arrangement of the posterior cingulum, which extends from the posterolabial corner of the tooth to the hypocone. The teeth of Simiacritomys differ from this tooth by having no lingual projection of the anterior cingulum, an anterior cingulum that is con- nected with the protoloph near midline rather than labially, and no mesostyle. The indeterminate rodent teeth described by Storer (1988) and Dawson (1966) appear to represent taxa that are related to Simiacritomys. However, their generic assignments will remain uncertain until larger samples of these taxa are available. The teeth of Simiacritomys are morphologically most similar to those of the sicistine zapodid Plesiosminthus. This similarity suggests Simiacritomys 1s a mem- ber of the family Zapodidae. Definitive familial assignment of Simiacritomys requires knowledge of the morphology of the skull and the mandible, which are presently unknown. Simiacritomys whistleri new species Figure 6, Table 5 ?zapodid, new genus and species Kelly and others, 1991:6, 13. Holotype. —LM,, LACM 131462. Type locality. —LACM 5876. Diagnosis. —Same as for genus. Referred specimens. —RM'°", LACM 130854; partial RM'°?, LACM 130852; RM!°2, LACM 130853; LM!?7, LACM 130855; LM3, LACM 131075; LM?, LACM 131457; LM3?, LACM 132451; RM,, LACM 130857; RM,, LACM 131461; LM,, LACM 131065; LM,, LACM 130856; LM,, LACM 130858; LM,, LACM 130861; RM;, LACM 131460; LM3;, LACM 130860; LM;, LACM 131069. Etymology.—Named in honor of David P. Whistler of the Vertebrate Pale- ontology Section of the Natural History Museum of Los Angeles County for his work on the Paleontologic Resource Impact Mitigation Program on the Sespe Formation at the Simi Valley Landfill. Description. —The first two upper molars exhibit the following characters. The occlusal outline is subquadrate. The teeth have a five-crested or lophed occlusal pattern. The anterior cingulum is a thick, well-developed crest that extends from the anterolabial base of the paracone to the protoloph, and has no lingual exten- sion. The anterocone is usually represented by a small distinct cusp along the anterior cingulum near its connection with the protocone. The paracone, meta- cone, protocone, and hypocone are well-developed conical cusps. A central trans- verse valley completely divides the tooth into anterior and posterior portions, 116 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Sy G Fig. 6. Simiacritomys whistleri new genus and species. A, RM'!°?, LACM 130854. B, RM!°?, LACM 130853. C, LM'°?, LACM 130855. D, RM,, LACM 130857. E, Holotype, RM,, LACM 131462. F, RM;, LACM 130460. G, LM?, LACM 131075. All occlusal views; D reversed. Scale = 1 mm. and is open labially and lingually. The other valleys between the primary cusps and crests are deep. The protoloph is usually divided, midway along its length, by an obliquely directed valley that persists even in extreme wear. In one tooth (LACM 130854), the protoloph is a complete crest across this valley and connects the protocone with the metacone. A mesostyle is not present. The metaloph is a thick, well-developed crest connecting the metacone and hypocone. The mesoloph is a moderately long, thick, high crest. In one tooth (LACM 130854), the mesoloph has two accessory spurs, one directed anterolabially, the other, posterolabially. The posterior entoloph (=posterior mure) is a thick crest connecting the mesocone and hypocone. The anterior entoloph (=anterior mure) is a crest that extends anteriorly from the mesocone, but does not connect with the protoloph because SESPE RODENTS Table 5. Measurements (in mm) for teeth of Simiacritomys whistleri new genus and species from Simi Valley Landfill Local Fauna. N Tooth Dimension OR Mean S.D. CV 4 M'! er? A-P 1.68—-1.82 1.73 3 ANT-TR 1.79-1.90 1.83 4 POST-TR 1.65—-1.70 1.65 3 M: A-P 1.30-1.38 1.34 3 TR 1.45-1.55 1.50 6 M, A-P 1.72-1.90 1.82 .08 4.6 6 ANT-TR 1.50-1.72 1.67 .10 1 6 POST-TR 1.59-1.90 1.71 .09 5.0 3 M, A-P 1.55-1.92 1.74 3 ANT-TR 1.42-1.61 1.50 3 POST-TR 1.24-1.42 1.34 117 of the presence of the central transverse valley. The posterior cingulum is well- developed and a small incipient cusp (posterocone?) is present along this cingulum. The M? exhibits the following characters. The occlusal outline is subtriangular, with the posterior aspect narrowed transversely. The anterior cingulum extends lingually from the anterolabial corner of the tooth to the protoloph. The protoloph appears to be a complete crest with a small, lingually directed spur (?metalophid II of Green, 1977). The protocone is a well-developed cusp that is situated slightly posteriorly of the midline along the labial aspect of the tooth. The hypocone is a transversely elongated cusp that is positioned posterolabially. A complete ento- loph connects the protocone and hypocone with the mesocone. The mesoloph and metaloph are thick, moderately long crests that extend lingually from the mesocone and posterior entoloph, respectively. The metacone is an elongated cusp at the posterolabial corner of the tooth. There is no tooth in the sample that can be positively identified as an M,. The M, exhibits the following characters. The occlusal outline is slightly elongated anteroposteriorly, resulting in a subrectangular shape. The M, has a five-crested or lophed occlusal pattern with deep valleys between the cusps and the crests. The anterior cingulid is moderately well-developed. A distinct anteroconid is present along the anterior cingulid that sometimes connects with the base of the labial projection of the metaconid. A median transverse valley is usually present and divides the tooth into anterior and posterior portions. The metaconid, pro- toconid, entoconid, and hypoconid are large, rounded cusps. A deep valley divides the metalophid, midway along its length, and separates the metaconid from the protoconid. The mesoconid is a distinct cusp in the center of the ectolophid (=mure). The mesolophid is usually a long well-developed crest that is strongly connected to the mesoconid. The ectolophid is usually incomplete. A mesostylid is not present. The hypolophid is a thick, well-developed crest that connects the entoconid with the hypoconid. The posterior cingulid is a tall, well-developed crest that extends from the posterolingual aspect of the entoconid to the hypo- lophid, where it connects near the posterolingual aspect of the hypoconid. A hypoconulid is present as a distinct bulge or cusp along the posterior cingulid. The M, is anteroposteriorly elongated and the occlusal outline is subrectangular. The anteroconid is a small distinct cusp along the anterior cingulid that is con- 118 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES nected to the labial projection of the metaconid. The anterior cingulid is smaller than those of the M, and it extends labially a short distance beyond the antero- conid. The metaconid and protoconid are well-developed cusps. The valley that divides the metalophid in the M, is usually interrupted in the M,; by a more completely developed metalophid that usually forms a thin complete crest across this valley. The anterior ectolophid is a thick, complete crest that connects the protoconid to an indistinct mesoconid, whereas a posterior ectolophid is absent. The mesolophid is usually a long, thin crest that extends to the lingual aspect of the tooth. Short accessory spurs sometimes are present on the mesolophid. The hypolophid is a complete crest connecting the well-developed hypeconid with a very small entoconid. The posterior cingulid is a short distinct crest at the pos- terolingual aspect of the tooth. Discussion. —Two teeth, an upper molar (LACM 130854) and a lower molar (LACM 130857), differ in the development of the crests and their connections relative to those of other molars assigned to Simiacritomys. In LACM 130854 (Fig. 6A), the anterior entoloph extends across the deep transverse valley, almost connecting with the protoloph, and the protoloph is complete. In LACM 130857 (Fig. 6D), the posterior ectolophid extends across the central transverse valley and connects with the hypolophid, whereas the anterior ectolophid is divided, midway along its length, by a deep valley. It could be argued that these variations are taxonomically significant and these specimens should not be referred to S. whistleri. However, these differences are regarded as intraspecific variation be- cause, in other rodent species where large samples of teeth are available, a high degree of variability is observed (for example, see Lillegraven and Wilson 1975, and Green 1977). Furthermore, all the specimens were recovered from one locality and these molars, except for the characters noted above, have the same basic morphology as the other molars assigned to S. whistleri. Simiacritomys whistleri is of uncertain affinities and is restricted to the late Duchesnean Simi Valley Landfill Local Fauna. As noted above in the discussion of the genus, S. whistleri is morphologically similar to the middle Arikareean to early Hemingfordian sicistine zapodid Plesiosminthus grangeri from South Da- kota. Conclusions This report documents the discovery of new species and new geologic and geographic occurrences of middle and late Eocene rodents from the middle mem- ber of the Sespe Formation recovered during the Simi Valley Landfill Paleonto- logic Resource Impact Mitigation Program. The taxa discovered during the pro- gram are ““Namatomys’’ sp., cf. “N.” fantasma Lindsay, ““Namatomys”’ sp., Paradjidaumo reynoldsi new species, Heliscomys sp., Simimys landeri new species, and Simiacritomys whistleri new genus and species. New records from the middle member of the Sespe Formation in Simi Valley include ““Namatomys,” Parad- Jidaumo, Heliscomys, and Simiacritomys. The occurrence of Paradjidaumo in the Simi Valley Landfill Local Fauna represents the second record of this genus in the Duchesnean. The occurrence of Heliscomys in the Simi Valley Landfill Local Fauna represents the second record of the genus and the family Heliscomyidae in the Duchesnean. The results of this study and those summarized by Kelly and others (1991) SESPE RODENTS 119 allow reevaluation of the Simi Valley Landfill Local Fauna. A revised faunal list for this fauna includes: Sespedectes singularis Stock (1935); Proterixoides davisi Stock (1935); Leptotomus sp. undetermined; ““Namatomys”’ sp.; Paradjidaumo reynoldsi n. sp.; Heliscomys sp.; Simimys simplex; Simimys landeri n. sp.; Simi- acritomys whistlerin. gen. and sp.; Camelidae, gen. and sp. undetermined; Simime- ryx sp.; and Mammalia, gen. and sp. undetermined. The Simi Valley Landfill Local Fauna from LACM Loc. 5876 is characterized by the restricted stratigraphic ranges of the following taxa: ““Namatomys”’ sp.; Paradjidaumo reynoldsi; Helis- comys sp.; Simimys landeri; Simiacritomys whistleri; and Simimeryx sp. The shared occurrences of Sespedectes, Proterixoides, ““Namatomys,” Paradjidaumo, Heliscomys, Simimys, and Simimeryx indicate that the Simi Valley Landfill Local Fauna is late Duchesnean in age. Acknowledgments I am indebted to Everett H. Lindsay of the Department of Geosciences, Uni- versity of Arizona, E. Bruce Lander, David P. Whistler, Lawrence G. Barnes, and Samuel A. McLeod of the Natural History Museum of Los Angeles County (LACM), and Robert E. Reynolds of the San Bernardino County Museum (SBCM) for their support and advice in preparing this report. Collections at the University of California, Museum of Paleontology, LACM, and SBCM were made available by J. Howard Hutchison, Lawrence G. Barnes, and Robert E. Reynolds, respectively. The LACM and its Foundation supported the Simi Valley Landfill Paleontologic Resource Impact Mitigation Program. Literature Cited Black, C. C. 1965. Fossil mammals from Montana, part 2. Rodents from the early Oligocene Pipestone Springs Local Fauna. Ann. Carnegie Mus., 38(1):1-48. Burke, J. J. 1934. New Duchesne River rodents and a preliminary survey of the Adjidaumidae. Ann. Carnegie Mus., 13(8):391-398. Chiment, J. J. 1977. A new genus of eomyid rodents from the later Eocene (Uintan) of southern California. M.A. thesis, California State Univ., San Diego, v + 83 pp. Dawson, M.R. 1966. Additional late Eocene rodents (Mammalia) from the Uinta Basin, Utah. Ann. Carnegie Mus., 38(4):97-114. Emry, R. J. 1981. New material of the Oligocene muroid rodent Nonomys, and its bearing on muroid origins. Amer. Mus. Novitates, 2712:1-14. , and W. W. Korth. 1989. Rodents of the Bridgerian (middle Eocene) Elderberry Canyon Local Fauna of eastern Nevada. Smithsonian Contri. Paleobiol., 67:111 + 1-14. Green, M. 1977. Neogene Zapodidae (Mammalia: Rodentia) from South Dakota. J. Paleont., 51(5);996- 1015. Kelly, T. S. 1990. Biostratigraphy of Uintan and Duchesnean land mammal assemblages from the middle member of the Sespe Formation, Simi Valley, California. Nat. Hist. Mus. Los Angeles Co., Contri. in Science, 419:1—42. , E. B. Lander, D. P. Whistler, M. A. Roeder, and R. E. Reynolds. 1991. Preliminary report on a paleontologic investigation of the lower and middle members, Sespe Formation, Simi Valley Landfill, Ventura County, California. Univ. California, Mus. of Paleont., Paleo Bios, 13(50): 1-13. Klingener, D. 1963. Dental evolution of Zapus. J. Mamm., 44:248-260. 1964. The comparative myology of four dipodoid rodents (genera Zapus, Napaeozapus, Sicista, and Jaculus). Univ. Michigan, Mus. of Zool., Misc. Publ., 124:1-100. 1966. Dipodoid rodents from the Valetine Formation of Nebraska. Univ. Michigan, Mus. Zool. Occ. Pap., 644:1-9. Korth, W. W. 1989. Geomyid rodents (Mammalia) from the Orellan (middle Oligocene) of Nebraska. 120 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Pp. 31-46 in Papers on fossil rodents, in honor of Albert Elmer Wood. (C. C. Black and M. R. Dawson, eds.), Nat. Hist. Mus. Los Angeles Co., Science Series, 33:xxi + 192 pp. —, J. H. Wahlert, and R. J. Emry 1991. A new species of Heliscomys and recognition of the family Heliscomyidae (Geomyoidea: Rodentia). J. Vert. Paleont., 1 1(2):247—256. Lillegraven, J. A., and R. W. Wilson. 1975. Analysis of Simimys simplex, an Eocene rodent (?Za- podidae). J. Paleont., 49(5):856—-874. Lindsay, E. 1968. Rodents from the Hartman Ranch Local Fauna, California. Univ. California, Mus. of Paleont., Paleo Bios, 6:1—22. Mason, M.A. 1988. Mammalian paleontology and stratigraphy of the early to middle Tertiary Sespe and Titus Canyon Formations, southern California. Unpublished Ph.D. dissertation, Dept. of Paleont., Univ. California, Berkeley, 257 pp. Russell, L. S. 1954. Mammalian fauna of the Kishenehn Formation, southeastern British Columbia. Ann. Rep., Natl. Mus. Canada 1952-1953, Bull., 132:92-111. Setoguchi, T. 1978. Paleontology and geology of the Badwater Creek area, central Wyoming, part 16. The Cedar Ridge Local Fauna (late Oligocene). Bull. Carnegie Mus., 9:1-61. Stehlin, H. G., and S. Schaub. 1951. Die Trigonodontie der simplicidentaten Nager. Schweizerische Naturforschende Gesellschaft, Schweizerischen Palaeontologischen Abhandlungen, 67:1-385. Stock, C. 1935. Insectivora from the Sespe uppermost Eocene, California. Proc. Natl. Acad. of Sciences, 21:456—462. Storer, J.E. 1984. Mammals of the Swift Current Creek Local Fauna (Eocene: Uintan, Saskatchewan). Saskatchewan Cult. and Rec. Mus. Nat. Hist., Nat. Hist. Contri., 7:1-158. 1987. Dental evolution and radiation of Eocene and early Oligocene Eomyidae (Mammalia, Rodentia) of North America, with new material from the Duchesnean of Saskatchewan. Dak- oterra, 3:108-117. 1988. The rodents of the Lac Pelletier Lower Fauna, late Eocene (Duchesnean) of Saskatch- ewan. J. Vert. Paleont., 8(1):84-101. Sutton, J. F., and C. C. Black. 1975. Paleontology of the earliest Oligocene deposits in Jackson Hole, Wyoming, part 1. Rodents exclusive of the family Eomyidae. Ann. Carnegie Mus., 45(16):299-— 315. Walsh, S. L. 1987. Mammalian paleontology of the southern outcrops of the Mission Valley For- mation, San Diego County, California. Senior thesis, California State University, San Diego, vii + 171 pp. White, T. E. 1954. Preliminary analysis of the fossil vertebrates of the Canyon Ferry Reservoir area. Proc. U.S. Natl. Mus., 103(3326):395-438. Wilson, R. W. 1935a. Cricetine-like rodents from the Sespe Eocene of California. Proc. Natl. Acad. of Sciences, 21:26-32. . 1935b. Simimys, a new name to replace Eumysops Wilson, preoccupied. —a correction. Proc. Natl. Acad. of Sciences, 21:179-180. 1949. Additional Eocene rodent material from southern California. Carnegie Inst. of Wash- ington Publ., 584:1—25. Wood, A. E. 1937. The mammalian fauna of the White River Oligocene, part II. Rodentia. Trans. Amer. Philos. Soc., new ser., 28:155-269. 1974. Early Tertiary vertebrate faunas Vieja Group Trans-Pecos Texas: Rodentia. Texas Memorial Mus. Bull., 21:1—112. 1980. The Oligocene rodents of North America. Amer. Phil. Soc., 70(5):1-68. Accepted for publication 21 November 1991. Bull. Southern California Acad. Sci. 91(3), 1992, pp. 121-125 © Southern California Academy of Sciences, 1992 A Newly Adventive Ant of the Genus Pheidole in Southern California (Hymenoptera: Formicidae) Roy R. Snelling Entomology Section, Natural History Museum of Los Angeles County, 900 Exposition Boulevard, Los Angeles, California 90007 Abstract. —This paper records, for the first time, in the United States the presence of Pheidole teneriffana, a species originally described from the Canary Islands. Colonies are established in Long Beach, Los Angeles County. The species is briefly characterized and the worker subcastes illustrated. A key is included to facilitate the separation of P. teneriffana from native Pheidole species in the Los Angeles basin. Insects accidentally introduced into southern California from other parts of the world are a common occurrence. There are probably many more such introduc- tions than we realize because, more often than not, the insect does not become established. Some of those that do successfully colonize here may become con- spicuous as pests and others may be unobtrusive and, hence, undiscovered for years. Several non-native ant species are present and firmly established here, but none has yet proven to be anything other than a nuisance. The most well known of these species is, of course, the common Argentine Ant, /ridomyrmex humilis (Mayr) (from South America), a conspicuous nuisance species in urban areas. Other introduced ant species in California, all from various parts of the Old World, include Tetramorium caespitum (Linné) (“Pavement Ant’), Monomorium phara- onis (Linné) (“Pharaoh Ant’’), Cardiocondyla ectopia Snelling (no common name), and Paratrechina longicornis (Latreille) (“Crazy Ant’’). An additional species that must be placed among those exotic ants now estab- lished in California is the myrmicine species Pheidole teneriffana Forel. In an error-laden note, Martinez (1992) first reported the presence of this ant in Cali- fornia. There are no prior records of this species in the United States, although it was recorded from Cuba by Aguayo (1932). Pheidole teneriffana was originally described by Forel (1893) from Teneriffe in the Canary Islands, but is now known to occur across North Africa at least as far east as Egypt. At present, its known occurrence in California is limited to colonies found in Long Beach, Los Angeles County, by Mr. Mike Martinez, an employee of the Long Beach Parks Department. Several colonies are mature and have produced winged reproductives. Many native species of Pheidole occur in California, and P. teneriffana shares with these the characteristic presence of two distinctly different subcastes of work- ers (‘soldier and “‘worker’’). The “‘soldier’”’ subcaste is larger and more robust than the “‘workers” and has a disproportionately larger and more massive head. The most recent key for the separation of Pheidole species in the United States is that of Gregg (1958). In that key P. teneriffana goes to the set of couplets 121 122 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES 2 Figs. 1, 2. Pheidole teneriffana, minor worker (1) and major worker (2), lateral view. Figures by Tina Ross. separating the several forms of P. sitarches and keys most often to P. sitarches sitarches Wheeler, an ant that does not occur in California. The major worker (“soldier’’) (Figs. 2 and 4) of P. teneriffana differs from all known California species by the following combination of morphological features: head width 1.38-1.52 mm; antennal scape evenly curved and not flattened at base, extending about *%4 of distance between antennal sockets and top of vertexal lobe; margins of frontal lobes extending dorsad to base of vertexal lobes: frontal area Of head with conspicuous, widely spaced longitudinal rugae that extend to PHEIDOLE TENERIFFANA IN THE UNITED STATES 123 Figs. 3, 4. Pheidole teneriffana, minor worker (3) and major worker (4), frontal view of head. Figures by Tina Ross. vertex lobes, the interspaces shiny and mostly smooth; vertex strongly concave between prominent vertex lobes; pronotum without distinct humeral angles and strongly convex in profile; mesonotum in profile strongly convex; propodeal spines present and about as long as distance between their bases; pronotum anteriorly with a few transverse rugae; mesepisternum and side of propodeum contiguously punctate and with some longitudinal rugae (most conspicuous on propodeum); postpetiole mostly smooth and shiny, lateral margins angulate at about midlength; total length 3.8—4.0 mm. 124 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Minor workers (Figs. | and 3) are much more difficult to characterize, but the following combination of characters should separate them from those, of other species in southern California: head smooth and shiny, quadrate in frontal view, with dorsolateral angles broadly rounded and vertex convex across middle; an- tennal scape extending beyond vertex by about ' its length: pronotum almost entirely smooth and shiny: mesepisternum and propodeum contiguously punctate and dull; mesonotum distinctly angulate in profile; propodeal spines short, slender and acute; total length 2.25—2.60 mm. The following key will serve to separate the worker subcastes of P. teneriffana from those of other Pheidole known to occur in the Los Angeles basin. Key to Pheidole of Los Angeles Area la. Major (=*‘soldier’’): sculpture of front of head various, but not consisting of continuous longitudinal rugae that extend from base of clypeus to vertex— Minor: head shape various. but if top of head is convex, then scape and tibiae bear many long, suberect to erect hairs and longest hairs of pronotum are clearly longer than greatest eye diameter ........... 2 b. Major (Figs. 2. 4): sculpture of front of head consisting of coarse longi- tudinal rugae that extend continuously from level of base of clypeus to top of vertex. Minor (Figs. 1, 3): head broadly rounded in frontal view; scape and tibiae with sparse suberect to erect hairs, those of scape limited to frontal face: longest hairs of pronotum about as long as greatest eye diameter -i..5.242. So Ne ee eee eee P. teneriffana Forel 2a. Major: tops of vertexal lobes with transverse rugae or fine striae: antennal scape not extending more than one-half distance between antennal socket and top of vertex. Minor: head quadrate in frontal view, vertex transverse and abruptly rounded at sides and antennal scape extending beyond veriex by less: than one-fiithais lengily 35 -.. - eeeeeeeeeee 3 b. Major: tops of vertexal lobes dull to slightly shiny, devoid of striations or rugae, but with scattered punctures; apex of scape extending more than four-fifths of distance between antennal sockets and tops of vertexal lobes. Minor: vertex either rounded in frontal view or prolonged into short, narrow “neck”; antennal scape extending beyond vertex by at least one-fifth iis length . 2... 235 0. Seb be oe eee 4 3a. Major: top of vertex crossed by coarse transverse rugae; postpetiole node without lateral connules. Minor: head and pronotum smooth and shiny ee COME eee Sey Meee MP ere vie ee ORS ee og P. clementensis Gregg b. Major: top of vertex with fine, close striae which may be partially effaced: postpetiole node with sharp lateral connules. Minor: head and pronotum slightly shiny and distinctly shagreened, at least in large part ......... ; cyig 2S eee 1 I ee trees heels P. pacifica Wheeler 4a. \Vajor: base of antennal scape broad and flattened and apex not extending beyond vertexal lobes. Minor: vertex broadly rounded in frontal view Sw Dl ee ee oot beret re ae srr: etree P. hyatti Emery b. Vajor: base of antennal scape narrow and rounded and apex ex- tending well beyond vertexal lobes. Minor: vertex strongly and evenly Narrowed into shore acck * 4352) wee ke ae eee P. vistana Forel PHEIDOLE TENERIFFANA IN THE UNITED STATES 125 The potential importance of this ant as a pest species is unknown, but I suspect low. It is likely that this species is primarily a seed-harvester and general scavenger, as are most species of Pheidole in temperate regions. In urban areas, at least, it will have to compete against J. humilis, a notoriously successful urban pest species. An additional competitor will be the native fire ant, Solenopsis xyloni, one of the few native species that seems to be able to withstand the Argentine ant in disturbed habitats. Between them, these two ants utilize the same resources that P. tenerif- fana might be expected to exploit. _ Voucher specimens from the Long Beach population are deposited in the col- lections of the California Department of Food and Agriculture, Los Angeles Coun- ty Museum of Natural History, Museum of Comparative Zoology, and National Museum of Natural History. Acknowledgments My thanks go to Mike Martinez for bringing this ant to my attention and to E. O. Wilson for resolving its identity. The figures were prepared by Tina Ross. Literature Cited Aguayo, C. G. 1932. Notes on West Indian ants. Bull. Brook. Entomol. Soc., 27:215-227. Forel, A. 1893. Nouvelles fourmis d’Australie et des Canaries. Ann. Soc. Entomol. Belgique, 37: 454-466. Gregg, R. E. 1958. Key to the species of Pheidole (Hymenoptera: Formicidae) in the United States. J. N. Y. Entomol. Soc., 66:7-48. Martinez, M.S. 1992. A new ant introduction for North America: Pheidale teneriffana (Forel). Pan- Pacific Entomol. 68:153-154. Accepted for publication 21 November 1991. Bull. Southern California Acad. Sci. 91(3), 1992, pp. 126-136 © Southern California Academy of Sciences, 1992 4 Distinguishing the Endangered Stephens’ Kangaroo Rat (Dipodomys stephens) from the Pacific Kangaroo Rat (Dipodomys agilis) Mary V. Price, Patrick A. Kelly, and Ross L. Goldingay Department of Biology, University of California, Riverside, California 92521 Abstract. — Efforts to conserve the endangered Stephens’ Kangaroo Rat have been complicated by its morphological similarity to the sympatric Pacific Kangaroo Rat. We used discriminant analysis of a variety of morphometric characters to develop a protocol for distinguishing these two species in the field. The species differ significantly in skull, hair, and external metric characters measured on live specimens. Discriminant analysis indicated that the two species can be distin- guished more successfully on the basis of weight, ear length, and head shape than on the basis of hair characters alone. We outline a simple and effective protocol for identifying live-captured specimens using ear length and postorbital head width. Efforts to develop plans for conserving Stephens’ Kangaroo Rat [Dipodomys stephensi (Merriam)], listed as rare by the State of California in 1972 (Leach and Fisk 1972) and federally listed as endangered in 1988 (Kramer 1988), are com- plicated by its morphological similarity to the Pacific Kangaroo Rat (Dipodomys agilis Gambel), with which it often coexists on a fine spatial scale (Price et al. 1991). Indeed, Grinnell (1922:65) remarked that “‘stephensi has been caught in the same trap-line with agilis without any notion on the part of the trapper that two species were represented.” This similarity has led to uncertainty over the accuracy of species identifications based on live-trapping studies performed by relatively inexperienced workers and, hence, to uncertainty over the exact spatial distribution of D. stephensi within its range in southwestern San Bernardino, western Riverside, and northern San Diego counties, California. Published descriptions of Stephens’ Kangaroo Rat do exist for characters of the skull, pelage, and baculum, as well as for standard characters recorded for museum specimens such as ear length and weight (Grinnell 1922; Lackey 1967a, b; Bleich 1977; Homan and Genoways 1978). Although some of these papers compare D. stephensi with D. agilis, to our knowledge no paper has focused on devising a simple, quantitative method for discriminating the two species in the field. In this paper we use discriminant analysis of characters that are easily measured on live animals to suggest a simple protocol for distinguishing Stephens’ from Pacific Kangaroo Rats. We also analyse characters that might be of use for iden- tifying skull material. 126 DISTINGUISHING STEPHENS’ AND PACIFIC KANGAROO RATS 127 SKULL LENGTH MAXILLARY ARCH WIDTH BULLAR WIDTH BULLAR DEPTH DENTARY WIDTH / pha DENTARY LENGTH Fig. 1. Six skull characters measured on museum specimens. Methods Skull Characters Six skull characters (Fig. 1) were measured by a single observer, using Helios dial calipers, on a total of 173 specimens from the Museum of Vertebrate Zoology at University of California, Berkeley, and the Natural History Museum of Los Angeles County. Virtually all specimens were adults from western Riverside Coun- ty, California. 128 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES HAIR WIDTH CELL WIDTH Me y CELL LENGTH Fig. 2. Three characters of guard hairs. Hair Characters Five guard hairs were clipped from the back, approximately 2 cm above the tail, of each of 155 specimens from Riverside County. Some were from specimens housed in the Natural History Museum of Los Angeles County, and some were from specimens live-captured at two sites, the University of California Motte Rimrock Reserve (Steele Peak Quadrangle: T4S,R4W,S24) and the San Jacinto Wildlife Area (Lakeview Quadrangle: T3S,R2W,S32). All hair samples were from adult animals. The five hairs were mounted in permount on microscope slides and covered with glass cover slips, following standard methods described in Homan and Geno- ways (1978). Preparations were measured at d) AfpuRoyrUsIs JOYIP $1910] 1dLWOsiodns UdIOYIP YIM SUB SOTD9dg ‘soId0ds UIYIIM SOXOS W90MI9q (SO'O > qd) AfUROYIUBIS JayIP sues ISOYM SIIJIVIVYO JVOIPUT SYSIIOISW ‘soloods Aq (,,d109S “OSIC,,) SIONS JUBUTLUTIOSIP UBOW Pu ‘soIoods Aq puke xos Aq ‘S1dJOBILYO [[NYS 1OJ son[eA UBdIY “| IGP DISTINGUISHING STEPHENS’ AND PACIFIC KANGAROO RATS 131 Table 2. Mean values for four characters of guard hairs, and for the discriminant score. Hair data are in ocular micrometer units (one unit = 0.00556 mm). x = mean; S.D. = standard deviation; N = number of individuals; P = probability that the two species have the same mean value. Maximal Medullary Medullary Medullary Disc. Species hair width cell number cell length cell breadth score D. agilis x 14.81 3.13 5.59 2.97 10.76 S.D. 1.70 0.26 0.66 0.29 1.16 N 57 57 57 57 57 D. stephensi x 12.39 2.84 4.77 2.85 9.00 S.D. 1.29 0.26 0.66 0.30 0.90 N 98 98 98 98 98 P 0.0001 0.0001 0.0001 0.0124 Canonical Coefficient 0.51 0.78 0.46 —0.61 Hair Characters Males and females did not differ significantly in hair characters for either species; species comparisons therefore were performed with pooled data. Dipodomys agilis and D. stephensi differed significantly in all four characters tested individually (Table 2), as well as when all characters were considered together in a canonical discriminant analysis (Wilks’ lambda = 0.58, df = 4,150, P < 0.001). Species accounted for 42.3% of total variation in discriminant function scores. Dipodomys agilis had wider hairs, larger medullary cells, and more columns of cells than D. stephensi (Table 2). Maximal hair width was the variable with the highest correlation with species (total structure coefficient = 0.96). Indeed, the discrimination with only this vari- able was nearly as good as with all variables entered, as the squared canonical correlation only dropped to 39.3% (Wilks’ lambda = 0.61, df= 1,153, P < 0.0001). We used a split-sample discriminant analysis to assess the value of hair char- acters alone in classifying specimens. When hair from the 105 live-captured spec- imens was used as the basis for classifying the 47 museum specimens, 5/23 D. agilis and 3/24 D. stephensi were misclassified, for an overall misclassification rate of 17%. This misclassification rate was similar to that (19%) for live-captured specimens classified according to a discriminant function derived from the live- capture hair data. Characters of Live Specimens Males and females did not differ significantly for any individual variable in D. agilis, but did vary in body mass, preorbital breadth, and head length for D. stephensi (Table 3). For both species, sexes were significantly different when all variables were considered together: males were heavier and had longer heads than females. Although the sexes differed for both species, the significance of between-species differences was not substantially higher when discrimination was performed sep- arately for each sex. We therefore report results of an analysis with pooled sexes. The two species differed significantly for all individual variables except head length (Table 3), and they differed overall (Wilks’ lambda = 0.215, df= 5,81, P < 0.0001). SOUTHERN CALIFORNIA ACADEMY OF SCIENCES 132 060 cl 99°61 29 OV I Ot c8cl el 6€ CC VI x6 6 £¢ Ie C'6¢ vi 6C 0) 4 91098 ‘OSIC] yiau9| peoH 67 80 CSC CC 0) € SC €¢ cl (XG vi cl 9ET yipeo1q [e11G.10180q Ll v9 IC 6C 80 x 1C (K6 Vl *O OC €¢ 81 LIC vi Ll a YIpesiq [e11qQ4109.1g €C OC Lgl vl cl vol yysus] 1eq LC Ls «109 CC LL *9'SS (3G 6 61S vi Ll [EASIS YSIS AA JUDIOYJOOD [eoruouRD A+ W isudyda]s ‘q d+ W SIIB °C W A isuayda]s ‘qd W df $1130 °C X9S so1oods “WU UT SLOJORIVYO LOYO ‘sued UI poJ1Odd1 SI 1YBIOMA ‘polood soxos YM ‘soloods udNM10q UOT IedOS [PWUIXPW SOPIAOId Jey] UOTPOUNY JURUTLULIOSIP [VOTUOULO OY) UT S}UDIOYJOOS poZipsepuLysuN OF IOJOL SPUSTOYJOOS [ROTUOUR,D ‘vIeP 9}0]dCWIOd YIM SUdUUTOOdS A[UO OPNIOUT SOXOS POUTQUIOD OJ SOZIS o[duuVS ‘OZIS ojduies = NY ‘UONRIAOp psvpue]s = “C's -UvOW = X (S0'0 > d) Aturoyruais s9yIp $1919] 1d1sossodns JUdIOWIP YIM Suv Sso1dodg ‘so1oods UTYIIM SOXOS U99MI0q (S0') > d) APUBOYTUTIS J9JIP SUBOLU DSOYM S19] IBIVYS O}VOIPUI SYSLIOISY ‘soloods Aq ‘o109s JURUTLULIOSIp UBOWW pu ‘solsods Aq puv xos Aq poysodo. ‘suowuoods OAT] UO PpoNsKOLU SAOJOBILYO IO) SoN|BA UBIYT “E 9[QUL DISTINGUISHING STEPHENS’ AND PACIFIC KANGAROO RATS 133 Dipodomys agilis weighed less than D. stephensi and had longer ears and a narrower head. The discriminant function explained 78.5% of the total variation, much more than that explained by hair data alone, or skull data alone. Indeed, a dis- criminant analysis performed with characters taken from live specimens correctly classified 36 of 38 D. agilis and 48 of 49 D. stephensi, for a very low misclassi- fication rate of 3.4%. The misclassified individuals were not juveniles (individuals weighing less than 50 g). Judging from total structure coefficients, preorbital breadth (0.71), ear length (—0.63), and postorbital skull breadth (0.61) were most highly correlated with the score on the discriminant function, and stepwise discriminant analysis selected these three variables as the most useful for discriminating the two species, whether forward or backward selection protocols were used. Species differences with a discriminant function based on these three variables alone accounted for 77.8% of the total variation, nearly as much as with all five variables (78.5%). Analysis of Hair Plus Live-Capture Data When hair data were combined with overall morphological information for the live-captured individuals having complete data (15 D. agilis and 11 D. stephensi), the degree of discrimination was improved somewhat (percent of variation in discriminant scores explained by species differences increased to 87.1% from 78%, and misclassification rate from a discriminant analysis was virtually identical to the 3.6% misclassification rate based on live-capture data—1/15 D. agilis and 0/11 D. stephensi were misclassified, for an error rate of 3.8%). Total structure coefficients indicate that maximum hair width and ear length were the two variables having greatest correlation with the discriminant function (0.75 and 0.68, respectively). Stepwise discriminant analysis entered these two variables first; they, along with postorbital breadth, explained 82.6% of the total variation, not much less than that (87.1%) explained by all nine variables. Dis- criminant analysis based on these three variables correctly classified 25 of 26 individuals, for an error rate of 3.8%. Discussion Our data agree with previous reports that D. stephensi differs from D. agilis in a number of morphometric characters. Dipodomys stephensi weighs more and has thinner guard-hairs, a broader skull, deeper auditory bullae, longer dentary, and shorter ears than D. agilis (Grinnell 1922; Lackey 1967a; Bleich 1977; Homan and Genoways 1978). The two species can be discriminated with reasonable precision using subsets of these characters. Skull characters provided 85% correct classification of the museum specimens we measured and should be of some utility for identifying prepared museum material. Characters of the head have the greatest value for discriminating live-captured specimens. Including guard-hair characters improves the discrimination based on ear and head characters, but hair characters alone provide inferior discrimination of the two species. On the basis of our analyses, we recommend two methods for distinguishing live-captured D. stephensi from D. agilis, depending on the availability of hair data. If hair data are not available, then good discrimination can be obtained by measuring ear length and postorbital breath of the head. If hair samples and a 134 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Table 4. a) Unstandardized canonical discriminant coefficients for functions based of two vari- ables (ear length, postorbital breadth) or three variables (ear length, postorbital breadth, maximal hair width). Rsquare is the squared canonical correlation. b) Mean discriminant scores, by species, for two- and three-vanable discriminant functions. Two- Three- variable variable function function a) Discriminant Function Coefficients Hair Width _ —36.251 Ear Length — 1.049 —0.951 Postorbital Breadth 1.473 1.291 Rsquare 0.808 0.826 b) Mean Discriminant Scores D. agilis x 12.32 18.04 S.D. 1.09 1.10 N 15 15 D. stephensi Fe 16.55 22.03 S.D. 0.86 0.84 N 11 11 microscope are available. then slightly better discrimination can be obtained by including maximal hair width along with the other two vanables. Table 4 gives the unstandardized canonical coefficients for these two methods of discrimination. A “score” on the two- or three-variable discriminant axis can be calculated for each specimen by multiplying each measurement (in mm) by the canonical coefficient associated with it, and summing the products over the two or three variables. For example, if individual #F2485 has postorbital breadth of 25.3 mm and ear length of 15 mm, its two-variable discriminant score is 1-473 x (25:3) — £049 _ x 5) —7iss32 Figure 4 shows the distribution of scores on two- and three-variable discrim- inant axes for D. stephensi and D. agilis. The small region of overlap in the distributions of scores for the two species reflects the probability of misclassifying a specimen. which is 3.8—6.7% (depending on which data set is used) for the two- variable discriminant function and 3.8% for the three-variable discriminant func- tion. Once a discriminant score has been calculated for a specimen, as described above. a f-test can be used to assess the probability that the specimen belongs to a sample of D. stephensi or of D. agilis, and the specimen can be assigned to species on the basis of these probabilities. The f-statistic is calculated by sub- tracting the mean discriminant score of a species (given in Table 4 for our cali- bration sample) from the discriminant score of the specimen, and then dividing the difference by s[(m + 1)/n]'*. where s is the standard deviation for the species” sample and 7 is the number of individuals on which that sample is based. This value of ¢ is compared. using 1-tailed probabilities, with a ¢-distribution having (7 — 1) degrees of freedom, to determine the probability that the specimen was taken from the distribution for that species. Values for the mean discriminant scores. standard deviations. and sample sizes for our samples are given in Table 4. DISTINGUISHING STEPHENS’ AND PACIFIC KANGAROO RATS 135 INDIVIDUALS lOPliveml8e t 191) 20) se 2Nor2 2023 ee4 NUMBER OF DISCRIMINANT SCORE Fig. 4. Distributions of scores on discriminant axis for D. agilis (open bars) and D. stephensi (filled bars). Top = scores based on two characters (ear length and postorbital breadth). Bottom = scores based on three characters (ear length, postorbital breadth, and hair width). Arrows indicate mean scores for each species. When individual #F2485 described above is compared with our sample of D. stephensi, t = —0.564, df = 10, and 1-tailed probability that it belongs to that sample lies between 0.24 and 0.45. When the specimen is compared with our D. agilis sample, t = 3.079, df = 14, and 0.005 < P < 0.01. Because the specimen has a much higher probability of belonging to D. stephensi than to D. agilis, we would assign the specimen to D. stephensi. For specimens that have approximately equal probability of belonging to both species, additional qualitative pelage char- acters (Bleich 1977) could be used for identification. We used this protocol to classify 14 D. agilis and 13 D. stephensi that were captured at the Motte Reserve and San Jacinto Wildlife Area, but which were not included in the “‘calibration”’ sample used to calculate the discriminant func- tion presented in Table 4. The individuals were assigned to species prior to measuring ear length and postorbital breadth, using the suite of qualitative char- acters that is often used to distinguish these two species (e.g., darkness of the hairs on soles of the hind feet, dorsal tail stripe, and tail tuft; baculum shape in males; weight; calmness when handled; “‘richness”’ of the brown pelage coloration; width and distinctness of the lateral tail stripe; Bleich 1977). All 27 specimens were correctly classified by the two-variable discriminant function; this is less than the misclassification rate for the calibration data set. Field workers are urged to collect and report standard measurements on live- captured specimens to provide documentation for their identifications in cases where an endangered species is similar morphologically to a sympatric, nonen- dangered congener. Our discrimination procedure, based solely on ear length and postorbital head breadth, provides a standardized identification protocol for D. stephensi and D. agilis, with low probabilities of misclassification. 136 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Acknowledgments Yq We thank Chris Camarena for measuring hair samples; Dr. Sarah George (Nat- ural History Museum of Los Angeles County) and Drs. Jim Patton and Barbara Stein (Museum of Vertebrate Zoology) for permission to measure specimens in their collections; Barbara Carlson (Univ. California Natural Reserves System) for permission to work at the Motte Reserve; Alan Craig (California Department of Fish and Game) for permission to work at the San Jacinto Wildlife Area; and Vern Bleich for reviewing the manuscript. Work was performed under U.S. Fish and Wildlife Service permit #MVPRICE and a Memorandum of Understanding with the California Department of Fish and Game. Literature Cited Bleich, V. C. 1977. Dipodomys stephensi. Mammalian Species, 73:13. Grinnell, J. 1922. A geographical study of the kangaroo rats of California. Univ. Calif. Publ. Zool., 24:1-124. Homan, J. A., and H. H. Genoways. 1978. An analysis of hair structure and its phylogenetic implications among heteromyid rodents. J. Mammal., 59:740-760. Klecka, W. 1980. Discriminant analysis. Sage University Paper series on quantitative applications in the social sciences. Beverly Hills and London. Kramer, K. 1988. Endangered and threatened wildlife and plants; determination of endangered status for Stephens’ kangaroo rat. Final rule. Federal Register, 53(190):38,465-38,469. Lackey, J. A. 1967a. Biosystematics of heermanni group kangaroo rats in southern California. Trans. San Diego Soc. Nat. Hist., 14:313-344. 1967b. Growth and development of Dipodomys stephensi. J. Mammal., 48:624—632. Leach, H. R., and L. O. Fisk. 1972. At the crossroads: a report on California’s endangered and rare fish and wildlife. California Department of Fish and Game Report, January 1972. Price, M. V., W. S. Longland, and R. L. Goldingay. 1991. Niche relationships of Dipodomys agilis and D. stephensi: two sympatric kangaroo rats of similar size. Am. Midl. Nat., 126:172-186. SAS Institute, Inc. 1985. SAS user’s guide: statistics. SAS Inst., Inc., Cary, N.C. 584 pp. Accepted for publication 21 November 1991. Bull. Southern California Acad. Sci. 91(3), 1992, pp. 137-140 © Southern California Academy of Sciences, 1992 The Status of Neotoma anthonyi (Rodentia, Muridae, Cricetinae) of Todos Santos Islands, Baja California, Mexico Eric Mellink Centro de Investigacion Cientifica y de Educaci6n Superior de Ensenada, Apdo. Postal 2732, Ensenada, B.C., Mexico U.S. mailing address: CICESE, P.O. Box 434844, San Diego, California 92143, USA Abstract. —The present conservation status of the Todos Santos Islands packrat (Neotoma anthonyi) was determined through live trapping, search for evidence and interviews with local inhabitants. It is concluded that this species is either extinct or very close to extinction. This was likely caused by the introduction of domestic cats. The packrat Neotoma anthonyi is a species endemic to Todos Santos Islands, off the coast from Ensenada, Baja California. These are two islets located at the mouth of Todos Santos Bay (Fig. 1). The south island has a surface area of approximately 100 ha and maximum elevation of 95 m; the north island has 30 ha and rises 45 m (Secretaria de Marina 1974). The southern part of the south island is formed by rounded hills with steep cliffs; the northern part of this island and the north island are low plateaus; the shorelines are rocky and precipitous (Nelson 1921). The islands, which are an extension of the Punta Banda peninsula, lie 5 km to the northwest of it. The islands are separated from each other by a narrow channel. Vegetation is Californian coastal scrub (Pase and Brown 1982), and includes open herbaceous areas dotted with clumps (a few meters in diameter) of Encelia californica, Bergerocactus emoryi, and Opuntia prolifera. This vege- tation is typical for flat areas, including most of the north island, but is absent at the lighthouse and dwellings on the north island, and on hilltops, in the valleys and on low plateaus of the south island. These clumps of shrubs were probably important for the packrats, as they are to the mainland N. /epida and other species elsewhere (N. albigula; Rangel and Mellink in press). These islands also harbor the endemic subspecies of deer mouse Peromyscus maniculatus dubius. The Todos Santos Islands packrat was described by J. A. Allen from five specimens collected on 11 May 1897 by A. W. Anthony (Allen 1898). Between its first collection and 1910 at least 45 additional specimens were collected. Their identifications were verified by E. A. Goldman (1910). Neither Allen (1898) nor Goldman (1910) reported a measure of abundance for these packrats, but Nelson (1921) indicated that N. anthonyi was “‘extremely numerous” on the islands (im- plying it was found on both landmasses). None of these authors reported preferred habitats. Methods Five trips to the Todos Santos Islands to determine the status of this packrat were made. These visits were on 8 March, 29-31 March, 12-14 June and 3-5 137 138 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES BAHIA DE TODOS SANTOS [6°45 16°40" 16°35. Fig. 1. Location of Islas de Todos Santos, Baja California. An arrow marks the area where a packrat skull and midden-type material were found. October of 1989 to the south island, and 22—24 March 1990 to the north island. On these visits I searched for evidence of packrats in almost all cracks and shrub clumps of both islands, and trapped rodents during the last four trips (80, 160, 160 and 160 night traps, respectively). During four of the island visits, and at Ensenada, local fishermen and the lighthouse keeper were interviewed regarding the presence of the species or of its houses. The trapping regime utilized 80 Sherman traps, baited with rolled oats, and checked early in the morning and in the afternoon (to make them all available at night, since deer mice proved to be active during the day). Traps were placed favoring the best sites, and using different areas on each trip. When the cliffs were sampled, traps were set mainly next to cracks. Whenever the traps were set in flats (either on top of the hills or in the low areas), they were TODOS SANTOS ISLANDS PACKRAT 139 set in transects with 15, 20, 30, or 40 traps, with a 10 m spacing between traps. The transects ran through different habitat variations, but favored shrub clumps and rock outcrops (the most favorable habitat for packrats). Some traps were also set next to a housing development of the north island, since packrats have been known to use certain parts of abandoned buildings (E. Mellink pers. obs.). This survey was probably adequate since most of the suitable spots were in- spected, and trapping transects were set in all areas of the islands and most shrub clumps and rock outcrops were included. Similar surveying efforts on the mainland and Cedros island, which have low packrat populations, have rendered positive results. In Cedros island (E. Mellink accepted), trapping success for Neotoma bryanti varied from 2.5 to 7.5% in areas occupied by the species. Results and Discussion The only evidence of recent presence of packrat has been a skull of N. anthonyi that I found on the first visit in a crack of the eastern cliffs on the south island. The skull is undatable, but probably is not more than a decade old. I have not found any packrat nests or feces, but I found midden-type material in some small caves (less than | m deep). Trapping success was high, but restricted to deer mouse (P. maniculatus dubius). The oldest fisherman, who started living on the islands in the early 1950s, does not remember any packrats or evidence of them, but another one remembers having seen a “‘pile of debris” in the past (date unspeci- fied). Based on my information, it can be concluded that N. anthonyi is either extinct or very close to extinction. The most plausible explanation for the disappearance of the packrats is extir- pation by domestic cats, which were introduced prior to 1950. (I have been unable to find any collector’s field notes to check if cats had already been introduced.) Neotoma anthonyi would presumably be very susceptible to predation, since it evolved in the absence of mammalian predators (nocturnal birds of prey are not resident on the islands). Feral domestic rabbits also existed on the south island. It is unknown whether they competed with the packrats. The fishermen and lighthouse keeper stated that the populations of both cats and rabbits increased to “high” numbers about 15 years ago. Rabbits have almost been extirpated, allegedly by cats and humans. Only cats are now present, in low numbers. The large rabbit population might have stimulated a population increase in cats, but decline of the rabbits would have imposed a great predation pressure on the packrat population (probably an easier prey than the deer mouse, and providing a greater food return per capture), if still present. The deer mouse would be less susceptible than packrat because of its smaller size, larger populations, and higher reproductive rates (based on the life histories of other packrats and other subspecies of the deer mouse). Also deer mouse would probably be inadequate to sustain the high cat populations left over from the rabbit high. Cats may have caused other extinctions. The Todos Santos Island Rufous- crowned Sparrow (Aimophila ruficeps sanctorum) is en endemic bird that nests on the ground or close to it. Eduardo Palacios and myself searched the southern island for it, with the aid of recordings of the species, on 15 April 1991. We were unable to obtain any response or see any individuals. I had not recorded it on any of my former visits, although those trips were not aimed at this species in 140 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES particular. Similarly, Everett (1989) indicated that both the Xantus’ murrelet (Synthliboramphus hypoleucus) and Cassin’s auklets (Ptychoramphus aleucticus) historically nested on Todos Santos Islands, with only the prior species doing so currently. Current lack of nesting on the island by the latter species could be also an effect of the cats. Acknowledgments I thank Christine Schonewald-Cox for her major assistance in typescript prep- aration. Juan Sidon and Martin Diaz transported me to and from the island, Lucia Alfaro assisted me during part of the field work, Paul S. Martin discussed the subject with me and Amadeo Rea assisted in typescript preparation. Literature Cited Allen, J. A. 1898. Descriptions of new mammals from western Mexico in Lower California. Am. Mus. Nat. Hist., Bull., 10:143-158. Everett, W. T. 1989. Historic and present distribution of breeding marine birds of Baja California’s Pacific coast. Simp. Inter. Biol. Mar., 7:97—106. Goldman, E. A. 1910. Revision of the woodrats of the genus Neotoma. N. Am. Fauna, 31:3-124. Mellink, E. Accepted. Biological Conservation of Isle de Cedros, Baja California, Mexico: assessing multiple threats. Biodiversity and Conservation. Nelson, E. W. 1921. Lower California and its natural resources. Natl. Acad. Sci... Mem., 16:1—194. Pase, C. P., and D. E. Brown. 1982. 133.2 Californian coastalscrub. Jn Biotic communities of the American Southwest-United States and Mexico. (D. E. Brown, ed.), Desert Plants, 4:86—90. Rangel, M. G., and E. Mellink. Jn press. Historia natural de la rata magueyera (Neotoma albigula) en el Altiplano Potosino-Zacatecano. Jn Advances en la mastozoologia de México. (G. Ceballos and R. Medellin, eds.), Asociacion Mexicana de Mastozoologia— American Society of Mam- malogists. Secretaria de Marina. 1974. Estudio geografico de la region de Ensenada, B.C. Secretaria de Marina, México, D.F. Accepted for publication 9 March 1992. Bull. Southern California Acad. Sci. 91(3), 1992, pp. 141-142 © Southern California Academy of Sciences, 1992 INDEX TO VOLUME 91 Allen, Larry G., Lucienne S. Bouvier, and Robert E. Jensen: Abundance, Di- versity, and Seasonality of Cryptic Fishes and their Contribution to a Tem- perate Reef Fish Assemblage off Santa Cataline Island, California, 55 Bouvier, Lucienne S., see Larry G. Allen Balazs, George H., see Murray D. Dailey Collins, Charles T.: Metals in Eggs of the California Least Tern in Southern California, 49 Dailey, Murray D., Martha L. Fast, and George H. Balazs: A Survey of the Trematoda (Platyhelminthes: Digenea) Parasitic in Green Turtles Chelonia mydas (L.) from Hawaii, 84 Fast, Martha L., see Murray D. Dailey Fierstine, Harry L., see George T. Jefferson Gobalet, Kenneth H.: Colorado River Fishes of Lake Cahuilla, Salton Basin, Southern California: A Cautionary Tale for Zooarchaeologists, 70 Goldingay, Ross L., see Mary V. Price Hilbig, Brigitte, see Vivianne Solis Weiss Jefferson, George T., Harry L. Fierstine, John R. Wesling, and Teh-Lung Ku: Pleistocene Terrestrial Vertebrates from near Point San Luis, and Other Localities in San Luis Obispo County, California, 26 Jensen, Robert E., see Larry G. Allen Knight, Allen W., see Michael S. Parker Kroll, Gigi, see Timothy R. McAdams Ku, Teh-Lung, see George T. Jefferson Kelly, Thomas S.: New Uintan and Duchesnean (Middle and Late Eocene) Rodents from the Sespe Formation, Simi Valley, California, 97 Kelly, Patrick A., see Mary V. Price Loehr, Keith, see Timothy R. McAdams 141 142 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES McAdams, Timothy R., Gigi Kroll, and Keith Loehr: The Feeding Preferences of the Sculpins, Scorpaenichthys marmoratus and Leptocottus armatus, between Sand Crangon and Small-eyed Shrimp, Crangon nigricauda and Heptacarpus cf. carinatus, 44 4, Mellink, Eric: The Status of Neotoma anthonyi (Rodentia, Muridae, Cricetinae) of Todos Santos Island, Baja California, Mexico, 137 Parker, Michael S., and Allen W. Knight: Aquatic Invertebrates Inhabiting Saline Evaporation Ponds in the Southern San Joaquin Valley, California, 39 Price, Mary V., Patrick A. Kelly, and Ross L. Goldingay: Distinguishing the Endangered Stephens’ Kangaroo Rat (Dipodomys stephensi) from the Pacific Kangaroo Rat (Dipodomys agilis), 126 Snelling, Roy R.: A Newly Adventive Ant of the Genus Pheidole in Southern California (Hymenoptera: Formicidae), 121 Weiss, Vivianne Solis, and Brigitte Hilbig: Redescription of Ophryotrocha platykephale Blake (Polychaeta, Dorvilleidae) from the Guaymas Basin Hy- drothermal Vents, 92 Wells, Harrington, and Patrick H. Wells: The Monarch Butterfly: A Review, 1 Wells, Patrick H., see Harrington Wells Wesling, John R., see George T. Jefferson iii The BULLETIN is published three times each year (April, August, and December) and includes articles in English in any field of science with an emphasis on the southern California area. Manuscripts submitted for publication should contain results of original research, embrace sound principles of scientific investigation, and present data in a clear and concise manner. The current AIBS Style Manual for Biological Journals is recommended as a guide for contributors. Consult also recent issues of the BULLETIN. MANUSCRIPT PREPARATION The author should submit at least two additional copies with the original, on 82 x 11 opaque, nonerasable paper, double spacing the entire manuscript. Do not break words at right-hand margin anywhere in the manuscript. Foot- notes should be avoided. Manuscripts which do not conform to the style of the BULLETIN will be returned to the author. An abstract summarizing in concise terms the methods, findings, and implications discussed in the paper must accompany a feature article. Abstract should not exceed 100 words. A feature article comprises approximately five to thirty typewritten pages. Papers should usually be divided into the following sections: abstract, introduction, methods, results, discussion and conclusions, acknowledgments, literature cited, tables, figure legend page, and figures. Avoid using more than two levels of subheadings. A research note is usually one to six typewritten pages and rarely utilizes subheadings. Consult a recent issue of the BULLETIN for the format of notes. Abstracts are not used for notes. Abbreviations: Use of abbreviations and symbols can be determined by inspection of a recent issue of the BULLETIN. Omit periods after standard abbreviations: 1.2 mm, 2 km, 30 cm, but Figs. 1-2. Use numerals before units of measurements: 5 ml, but nine spines (10 or numbers above, such as 13 spines). The metric system of weights and measurements should be used wherever possible. Taxonomic procedures: Authors are advised to adhere to the taxonomic procedures as outlined in the International Code of Botanical Nomenclature (Lawjouw et al. 1956), the International Code of Nomenclature of Bacteria and Viruses (Buchanan et al. 1958), and the International Code of Zoological Nomenclature (Stoll et al. 1961). Special attention should be given to the description of new taxa, designation of holotype, etc. Reference to new taxa in titles and abstracts should be avoided. The literature cited: Entries for books and articles should take these forms. McWilliams, K. L. 1970. Insect mimicry. Academic Press, vii + 326 pp. Holmes, T. Jr., and S. Speak. 1971. Reproductive biology of Myotis lucifugus. J. Mamm., 54:452-458. Brattstrom, B. H. 1969. The Condor in California. Pp. 369-382 in Vertebrates of California. (S. E. Payne, ed.), Univ. California Press, xii + 635 pp. Tables should not repeat data in figures (line drawings, graphs, or black and white photographs) or contained in the text. The author must provide numbers and short legends for tables and figures and place reference to each of them in the text. Each table with legend must be on a separate sheet of paper. All figure legends should be placed together on a separate sheet. Illustrations and lettering thereon should be of sufficient size and clarity to permit reduction to standard page size; ordinarily they should not exceed 814 by 11 inches in size and after final reduction lettering must equal or exceed the size of the typeset. All half-tone illustrations will have light screen (grey) backgrounds. Special handling such as dropout half-tones, special screens, etc., must be requested by and will be charged to authors. As changes may be required after review, the authors should retain the original figures in their files until acceptance of the manuscript for publication. Assemble the manuscript as follows: cover page (with title, authors’ names and addresses), abstract, introduction, methods, results, discussion, acknowledgements, literature cited, appendices, tables, figure legends, and figures. A cover illustration pertaining to an article in the issue or one of general scientific interest will be printed on the cover of each issue. Such illustrations along with a brief caption should be sent to the Editor for review. PROCEDURE All manuscripts should be submitted to the Technical Editor, Jon E. Keeley, Biology Department, Occidental College, 1600 Campus Road, Los Angeles, California 90041. Authors are requested to submit the names, addresses and specialities of three persons who are capable of reviewing the manuscript. Evaluation of a paper submitted to the BULLETIN begins with a critical reading by the Editor; several referees also check the paper for scientific content, originality, and clarity of presentation. Judgments as to the acceptability of the paper and suggestions for enhancing it are sent to the author at which time he or she may be requested to rework portions of the paper considering these recommendations. The paper then is resubmitted and may be re-evaluated before final acceptance. Proof: The galley proof and manuscript, as well as reprint order blanks, will be sent to the author. He or she should promptly and carefully read the proof sheets for errors and omissions in text, tables, illustrations, legends, and bibliographical references. He or she marks corrections on the galley (copy editing and proof procedures in Style Manual) and promptly returns both galley and manuscript to the Editor. Manuscripts and original illustrations will not be returned unless requested at this time. All changes in galley proof attributable to the author (misspellings, inconsistent abbreviations, deviations from style, etc.) will be charged to the author. Reprint orders are placed with the printer, not the Editor. CONTENTS 97 Sespe Formation, Simi Valley, California. By Thomas S. Kelly A Newly Adventive Ant of the Genus Pheidole in Southern California (Hymenoptera: Formicidae). By Roy R. Snelling “mri Distinguishing the Endangered Stephens’ Kangaroo Rat (Dipodomys ste- phensi) from the Pacific Kangaroo Rat (Dipodomys agilis). By Mary | New Uintan and Duchesnean (Middle and Late Eocene) Rodents from the V. Price, Patrick A. Kelly and Ross L. Goldingay 126 The Status of Neotoma anthonyi (Rodentia, Muridae, Cricetinae) of Todos Santos Islands, Baja California, Mexico. By Eric Mellink _. =————Ss«137 Index to Volume 91 : uk 141 LIBRARY APR ~ 2 1996 NEW YORK BOTANICAL GARDEN COVER: Pheidole teneriffana, major worker, lateral view. (Page 122).