BULLETIN OF THE FLORIDA STATE MUSEUM BIOLOGICAL SCIENCES Volume I Number 2 SYSTEMATICS AND VARIATION OF THE ENDEMIC FLORIDA SNAKE GENUS STILOSOMA Richard Highton UNIVERSITY OF FLORIDA Gainesville March, 1956 The numbers of THE BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL SCIENCES, wil.1 be ptiblished at irregu- lay inter·uals. Volumes will contain about 250 to 300 pages, and will not „€cessayily be completed in any ofic calendar year. William J. Riemer, Editor John C. Briggs, Associate Editor All comm:ftnications concerning pit,chase or exchange of the publi- cation sliolild be addressed lo the Cltratow of Biological Sciences, Flor- ida State Museum, Sengle Building, Gaines·bille, Flofida. Manuscripts should be se,it to The Edito), Depmtment of Biology, University Of Floirida. Gainesville, Florida. Published Marci, 30, 1956 Pi·ice for this issue $.35 SYSTEMATICS AND VARIATION OF THE ENDEMIC FLORIDA SNAKE GENUS STILOSOMA RICHARD HIGHTON1 INTRODUCTION The genus Stilosorna, comprising a single fossorial species, S. ex- tenuatum, has the most restricted range of any snake genus in the United States, being known only from the northern half of the Florida peninsula. This distinctive snake was fif5t described in 1890 by A. E. Brown, who had a single specimen from Lake Kerr, Marion County, Florida. During the next two decades, most of the additional speci- mens collected were reported in articles by Loennberg (1894), Brown (1901) and Tucker (191,1). The species was then known only from Orange and Marion Counties in north-central Florida. Sinee 1911, very few papers have been published concerning this snake. In 1934, Carr described sallie aspects of the behavior of a captive specimen from Lake County. In 1939, Van Duyn reported a range entension based on an individual collected in Hernando County. A year later, Carr listed the counties from which museum specjmens were then available and also included notes on the ecological distri- bution of Stilosomci Allen and Neill ( 1953) discussed the distribution, habits and size of the short-tailed snake. None of these more recent contributions provided new data on the variation of this species and, for this reason, recent accounts of North American snakes have had to rely on Brown's 1901 paper, describing nine specimens from Marion County, for any information on its variation. An examination of the material which has accumula ted since Brown's time seems long over- due. A total of 89 specimens of S. extenuatum was available for study. The following characters were recorded for each specimen: sex, number of ventrals, caudals, supralabials, infralabials, body and tail blotches, arrangement and number of head scales: and the number of scale rows on the neck, at midbody and in front of the vent. Measurements of the body and tail length were also taken. Each character was analyzed for sexual dimorphism and geographic variation, and those that ex- hibited such variation are discussed below under separate headings. The measures cf validity used in this paper are standard errors. I wish to thank the following museum officials for the loan of inaterial in their care (initials folldwing the name of the institution 1Department of Biology, University of Florida. 74 BULL. OF THE FLA. STATE MUSEUM, BIOLOGICAL SCIENCES will be used as an aid iii reference to individual specimens): Mr. Charles M. Bogert, American Museum of Natural History (AMNH); Dr. Archie Carr, University of Florida (UF); Dr. Doris M. Cochran, United States National Museum (USNM); Drs. E. R. Dunn and Jay M. Savage, Academy of Natural Sciences of Philadelphia (ANSP); Dr. Norman Hartweg and Mr. William Duellman, University of Michigan, Museum of Zoology (UMMZ); Mr. Arthur Loveridge, Museum of Comparative Zoology (MCZ); Mi·. Wilfred T. Neill, Research Divi- sion, Ross Allen Reptile. Institute (ERA-WTN); Dr. O. T. Owre, University of Miami (UM); Mr. Neil D. Richmond, Carnegie Mu- seum (CM); and Drs. Karl P. Schmidt and Robert Inger, Chicago. Natural History M[iseum (CNHM) Mr. Wilfred T. Neill has kindly furnished me with much useful information, especially on the ecological distribution of Stilosoma in Marion County. Messrs. Ross Allen, Arthur Loveridge, and Guy Van Duyn, and Dr. Doris M. Cochran have also furnished valuable informa- tion. Drs. Lewis Berner, Pierce Brodkorb, Archie Carr, John Cren- shaw, Richard A. Edwards, Coleman J. Goin, Arnold B. Grobman and William Riemer have offered much helpful advice and criticism. Mr. Roger Conant furnished the color notes on the type specimen of S. e. multistictum. Dr. and Mrs. Robert F. Mason kindly permitted nle to examine a living specimen from Orange County. I wish to express my special appreciation and thanks to Mr. Walter Auffenberg for all the help he has given me during the course of this study. My wife, Anne, helped in numerous ways, and Miss Esther Coogle aided in the prepa- ration of figurd 3. SEXUAL DIMORPHKM Analyses of [he diaracters studied showed that sexual dimorphism was present in four characters: (1) the number of ventrals, (2) the number of caucials, (3) the nuniber of tail blotches, and (4) the ratio of tail length to body length. Fentrals.-The genus Stilosoma is characterized by a high number of ventrals. These scales were counted beginning with the first scale which was wider than it was long. Ventrals.in 43 males range from 289 to 261 (mean 249.1 =!= 0.65, standard deviation 4.3 + 0.46.). In 37 females, the number of ventrals varies from 256 to 27 (mean 265.0 + 0.78, standard deviation 4.8 ZE 0.55). The coefficient of sexual di- vergence (Klauber, 1940: 208).for this character is .062. No geographic SYSTEMATICS AND VARIATION OF STILOSOMA 15 variation can be detected in the number of ventrals. The anal plate is Single in all 5pecimens examined. Cauda/s:-Caudals were counted on the right side of the snake's tail. The enlarged terminal spine was not included. Of 87 snakes, 55 (63%) have one or more caudals undivided, but most of the urosteges are divided in all specimens. The caudals of 45 males range from 38 to 48 (mean 43.3 + 0.38, standard deviation 2.6 =t 0.27), while in 36 fe- males they vary from 33 to 45 (mean 37.2 + 0.38, standard deviation 2.3 + 0.29).'The coefficient of sexual divergence for this character is .151. Although there is some overlap between males and females in the range of both ventrals and caudals, when the number of caudals is subtracted from the number of ventrals, the resulting index shows no overlap between the sexes in the available sample. The range of varia- tion in the character "ventrals minus caudals" is 199 to 216 in males and 217 to 238 in females. This information could be useful in sexing living Stilosoma, since it is somewhat difficult to determine the sex of this snake except by dissection. Tail Blotches :-As will be shown below in the discussion of geographic variation in this character, specimens from the northern part of the range haye more tail blotches, on the average, than those from the southern part of the range. This character also shows sexual di- morphism. The mean number of tail blotches in males is higher than in females in each region (see figure 1), although in most cases there is not a sufficient number of specimens to comprise an adequate sample. Ratio of Tail Length to Body Length:-As in many snakes, the males have a significantly longer tail than the females. The ratio body length/tail length in males varies from 8.8 to 12.1 (mean 10.1 + 0.11. standard deviation 0.71 ZE 0.08). in females this ratio ranges from 9.5 to 14.1 (mean 12.2 + 0.18, standard deviation 0.98 + 0.13). The largest female examined (ANSP 11832, from Marion County) is 575 mm. in total length, and the longest male (UMMZ 76750, taken near Eustis, Lake County) is 567 mm. in total length. GEOGRAPHIC VARIATION In spite of the fact that only 125 miles separates the northernmost and southernmost localities from which specimens are known, pro- 76 BULL. OF THE FLA. STATE MUSEUM, BIOLOGICAL SCIENCES nounced geographic variation is evident in four of the characters studied. These are: (1) the number of dorsal body b16tches, (2) the number of dorsal tail blotches, (3) the number of infralabials, and (4) the fusion of in ternasals and prefrontals. One of these, the number of tail blotches, also exhibits sexual dimorphism. Thirty of the available specimens do not have exact locality data and could not be included in most of the analyses of geographic varia- t.ion. However, additional locality data can be offered for two series from Marion County. A series of 11 specimens was sent to two mu- seums (MCZ 41978-83 and CM 9407-11) by the Ross Allen Reptile Institute in 1936 and 1937. Mr. Allen (in a letter, September 24, 1954) has informed me that he believes most of these specimens were col- lected within four miles of Silver Springs, although a few may have been collected in other parts of Marion County. Another series of 11 specimens from Marion County (ANSP 11831-5, 16970-1, 26312-5), all from the A. E. Brown collection, may possibly have come from the vicinity of the town of Lake Kerr. Five specimens in the same col- lection are labelled Lake Kerr, and one other is from nearby Norwalk (actually in Putnam County). Four of these 11 without exact locality data (ANSP 26312-5) were collected at Norwalk and Lake Kerr ac- cording to Tucker (1911: 549). The entire ANSP series shows little variation and very likely' was collected at or near these same localities. If the Reptile Institute series and the ANSP series are assumed to have come from the vicinity of Silver Springs and Lake Kerr respectively, the conclusion6 presen ted below on geographic variation in each of the four characters would be further substantiated. Dorsal Blotches :-The number of blotches was counted on each side of the body and the average of the two counts taken. For example, two blotches on one side uniting with one blotch on the other side were counted as 1 42 blotches, while a blotch present on only one side of the body was counted as % a blotch. Specimens with exact locality data from the northern part of the range (Alachua and Levy Cotinties) have a much higher number of dorsal blotches than those from farther south (Putnam, eastern Marion, Lake, Seminole, Orange, Polk, Pinellas, Pasco, Hernando, Citrus, and Sumter Counties). Eight of nine specimens from the two n6rthern counties have 69 or more dorsal blotches (mean 71.8 zE 1.26), while all but two of the 29 southern snakes have fewer than 69 dorsal blotches (mean 62.5 + 1.05) (see figure 1). The difference between these two samples is statistically significant. If the separation indicated above were used as a key character, the geographic provenance of 92% of the SYSTEMATICS AND VARIATION OF STILOSOM.4 77 A = mir 45 SO) 55 60 65 70 75 80 8 12 8 I lyrr-•~~1 I n Fl 11 45 50 55 60 65 70 75 80 8 12 16 C .id~n. 45 50 55 60 65 70 75 80 8 12 16 n~ n Allin 45 -50 55 60 65 70 75 80 8 12 16 7 E :Ill ~ n . n 45 50 55 60 65 70 75 80 8 12 16 F B. W~ n Iwib 45 50 5 60 65 70 75 80 8 12 16 BODY BLOTCHES TAIL BLOTCHES Figure I .-The number of dot·sal body and tail 1) 16tchesin six samples of Stilosol ,za extenuatum. Solid symbols represent males, hollow= symbols females, and a ques- lion mark indicates a specimen of undetermined sex. A. Alachua and Levy Counties. B. Silver Springs and Dimnellon, Marion County. C. Marion County (MCZ 41978-83; CM 9407-11). D. Citrus, Sumter, Hernando, Pasco and Pinellas Comities. E. Marion County (Eureka and Lake Kerr), Putnam, Lake, Seminole, Orange, and Polk Counties: F. Marion County (ANSP 11831-5,.16970, 26312-5). available specimens could be correctly determined. A series of 17 specimens from the vicinity of Silver Springs, Marion County, is in- termediate. These snakes have ·a dorsal blotch count ranging from 58 to 79 (mean 67.1 =t 1.35). A specimen fro'm Dunnellon (ERA-WTN 15624), the Only one available from southwestern Marion County, has 79 dorsal blotches. The remaining specimens from Marion County have ·not been included in either the northern or southern samples because they lack accurate locality data. MCZ 13598 is listed as having come from the 78 BULL. OF THE FLA. STATE MUSEUM, BIOLOGICAL SCIENCES Oklawaha River. It has 53 dorsal blotches. Two other specimens from Marion County (USNM 56239 and CNHM 3389) also have a low number of dorsal blotches (61 and 62 respectively). In the 11 specimens from Marion County in the ANSP collection, the number of dorsal blotches ranges from, 57 to 68 (mean 62.4 + 0.97), typical of the southern sample. In the series of eleven specimens sent to MCZ and CM by the Ross Allen Rdptile Institute, the dorsal blotches range from 63 to 76 (mean 69.4 + 1.10), similar to the intermediate sample froni Silver Springs menti6ned above. Tail Blotches:-There appears to be little geographic variation in the number of tail blotches in females; however, in males, there is a de- cided decrease in the number of tail blotches from north to south. Five (83%) of the six males from Alachua and Levy Counties have 14 or more tail blotche5, while 11 (79%) of the 14 males with accurate 1gcality data from eastern Marion, southeastern Putnam, Lake, Semi- nole, Orange, Polk, Pinellas, Pasco, Hernando, Sumter and Citrus Coun ties have 13 or fewer tail blotches. The Silver Springs series is intermediate in this character as it is in the number of body blotches (see figure 1) A B Figure 2.-A. Chin of Solosoina e. exte),unium (ANSP 3351), showing fusion of an- terior infralabials. B. Chin of S. e. inultistictum (UF 8088), with seven in- fralabials. SYSTEMATICS AND VARIATION OF STILOSOMA 79 Infralabials:-lIn the eastern portion of its range, specimens of Stilo- soma extenuatum show a reduction in the number of anterior lower labials due to the fusion of two or more of the first five scales in the infralabial series. The two posterior infralabials are quite variable and have not been considered in this analysis. There is often a fusion or splitting of these tWO posterior infralabial scales which would partially obscure the actual observed geographic variation in the an- terior infralabials. Western specimens typically have seven lower labials on both sides of the head, while eastern specimens usually have less than seven lower labials on at least one side (see figure 2). The data on this character are summarized in table I. TABLE I. THE FREQUENCY OF OCCURRENCE OF FuSED ANTERIOR INFRALABIALS IN SIX SAMPLES OF 'STILOSOMA EXTENUATUM' FROM FLORIDA Fusion on one or No fusion on either Lacality both sides of head . side of head Alachua and Levy Counties 1 8 Marion County (Silver Springs and Dunnell6n) 4 14 Marion County (? Silver Springs) 0 11 (CM 9407-11; MCZ 41978-83) - Citrus, Sumter, Hernando, Pasco and 0 7 PineIJas Counties Marion County (Eureka and Lake Kerr); 17 5 Putnam, Lake, Seminole, Orange and Polk Counties Marion County (? Lake Kerr) 11 0. (ANSP 11831-5, 16970-1,26312-5) Fusion of Head Scales:-In the type specimen of the short- tailed snake, there are no preoculars and the internasals and prefrontals are fused. This unusual fusion of head scales, possibly related to the burrowing habits of the species, interested several early workers (Brown, 1890 and 1901; Stejneger in Loennberg, 1894; and Tucker, 1911). The pre- ocular is apparently rarely absent, for only five of the available speci- mens lack this scale. They are: ANSP 3351 (the type specjmen of S. extenuatum) and ANSP 12005, both from Lake Kerr, Marion County; ANSP 26314 from Marion County (absent on the left side only); UMMZ 76750, collected near Eustis, Lake County; and UF 8268 col- 80 BULI.. OF THE FLA. STATE MUSEUM, BIOLOGICAL SCIENCES lected near Gainesville, Alachua County (absent on the left side only). Only one (UF 8268) does not also possess fused internasals and pre- frontals. The geographic Variation iii the fusion of the internasals and pre- frontals is shown in table II. None of the specimens from the western portion of the range show this fusion, and only one of the 16 Silver Sprin'gs specimens has fused internasals and preffontals. In the eastern part of the range, 14 (58%) of 24 specimens have fused prefrontals and internasals on at least one side. In two specimens of the ANSP series from Marion County, these scales are fused on only one side (in ANSP 26313 they are fused on the right side, while in ANSP 11834 they are fused on the left side). In USNM 24353, from Eustis, Lake County, they are fused only on the right side. TABLE lI. THE FREQUENCY OF THE OCCURRENCE OF FUSED INTERNASALS AND PREFRO-N'TALS IN SIX SAMPLES OF "STILOSOMA EX'l'ENUATUM" FROM FLORIDA Fused prefrontals Prefrontals and Locality and internasals. internasals separate Alachwa and Levy Counties 0 10 Marion County (Silver Springs ancl 1)unnellon) 1 16 Marion County (? Silver Springs) 0 11 (CM 9407.-11; MCZ 41978-83) Citrus, Sumter, Hernando, Pasco and 0 Pinellas Counties Marion County (Eureka and Lake Kerr); 14* 10* Putnam, Lake, Seminole, Orange and Polk Counties Marion County (? Lake Kerr) 7 4 (ANSP 11831.5, 16970-1, 26312-5) *Including two specimens, one of each category, from Oakland, Orange County. not seen by the writer, but described by Stejneger in Loennberg (1894: 323) TAXONOMY AND NOMENCLATURE Brown described the short-tailed snake.in 1890, giving it the name Stilosoma· extenuata. Cope (1892: 595) changed the specific name to extenuatum. The species was known as St,losoma extenuatum until Berg (1901: 290) proposed the new generic name Stylophis to, replace Stilosoma, which he believed to be "preoccupied for a genus of Coleop- SYSTEMATICS AND VARIATION OF STILOSOMA 81 tera by Solier (1851)." Berg reasoned that since the word Stilosoma is derived from the Greek words 0- 1-vAo* and o- co p a, it should be spelled Stylosoma, making it a homonym 6f the earlier generic name of the beetle. Cope (1895: 205) had, indeed, already made this correction. In the first two editions of the Check List of Noyth American Am. phibians and Reptiles, Stejneger and Barbour (.1917: 90; 1923: 102) recognized Berg's Stylophts as the valid generic name of the short- tailed snake, but in the third edition (1933: 111), without stating a reason, they reverted to the use of the name Stitosbma. Dr. Doris M. C6chran has kindly checked through the files of the late Dr. Stejneger and has found no menti6n Of the reason for this change. For the last 20 years, most authors have followed the "Check List" by using the name Stilosoma Brown for this genus . Mr . Arthur Loveridge has suggested to me that the reason for the chahge back to the generic name Stilosoma in the 1933 edition of the "Check List" may have been because of the difference in spelling between Brown's St,losomn and Solier's Stylosoma. Since the names were originally spelled differently, Stejneger and Barbour may have reasoned that Stilosoma Brown was not preoccupied as claimed by Berg ( 1901: 290). How- ever, for purposes of synonomy, under the present International Rules of Zoological Nomenclature, it would not matter whether the name was spelled with an i or a y. Article 35 states that specific names of the same origin and meaning shall be considered homonyms if they are distinguished from each other only by an z.or a y. ln Opinion 147 (1943) the Commission extended to generic and subgeneric nanies the principles applied to specific names in Artiole 35. Fortunately, for the sake of stability, the name Stilosoma Brown does not have to be replaced by Stylophis Berg. Berg apparently did not examine the original description of the beetle "Stylosoma Solier (1851)." Actually this genus of Chilean carabid beetles was described by .Solier in Gay ( 1'849 : 241 ) under the name Systolosoma. Systolosoma is still recognized by entomologists as the valid generic name for this beetle (Blackwelder, 1944: 22). The first reference to the name Style,soma for the beetle described by Solier in 1849, appeared in a paper by- Schaum (1852: i47), in which he listed advances in entomology during the year 1851. A list of the species described by Solier appeared in Schaum's paper. (This is probably the source of the erroneous date, 1851, which Berg.listed as the date of the original name Stylosoma Solier. Actually, Solier's description of this beetle genus as Systolosoma appeared in 1849.) Schaum (1852: 146-7) merely printed a list of all the new genera and 82 BULL. OF THE FLA. STA'rE MUSEUM, BIOLOGICAL SCIENCES species described by Solier in 1849. All are spelled as they were orig- inally described, with the single exception of Systolosoma brguis, which is listed by Schaum as Stylosoma breve. Thus the origin 0£ the name Stylosoma for the carabid beetle is merely a misspelling of the generic name Systolosoma in Schaum's paper. This new name has no validity under the International Rules of Zoological Nomenclature, and is therefore not available for the beetle genus Systolosoma. Clearly, then, the name Stilosoma is not preoccupied and therefore remains the valid generic name of the short-tailed snake. Genus Stilosoma Brown, 1890 St. ilwoma Brown (1890: 199). Genotype: Stilosoma extenuata by monotypy. Stylosoma Cope (1895: 205). Berg (1901: 290) Stylophis Berg (1901: 290) Genotype: Stilosoma extenuatum by monotypy Stilosoma extenuatum Brown Stilosoma exte, litalit Brown ( 1890 : 199). Type locality : Lake Kerr . Marion Colin - ty, Florida St.i/650#a extenuntum Brown. Cope (1892: 595). Loennberg (1894: 323-324, figs. 1-3). Boulenger (1894: 325). Cope (1900: 925, fig. 232). Brown (1901: 80-81) Brimley (1910: 14). Tucker (1911: 549-550, fig.). Stejneger and Barbour (1933: 111). Carr (1934: 138). Ditmars (1939: 106-107, pl. 5). Van Duyn (1939: 51) Stejneger and Barbour (1939: 121). Carr (1940: 86). Schmidt and Davis (1941: 191-192, fig. 59, pl. 21). Allen and Neill (1953: 8-9, fig.) Stylosoma extenuatum Brown. Cope (1895:. pl. 18, fig. 12) Stylo·phis extenuatits (Brown). Stelneger and Barbour ( 191 .7 : 90 ; 1923 : 102). Blanchard (1924: 29, fig. 36). Ditmars (1933: pl. 39) Generic Description (modified ftom Brown , 1890):-Colubridae With very slender , cylindrical body, tail short; head rounded on frontal outline, not distinct from body; rostral prominent but not recurved; prefrontals and internasals may be fused or not; preocular usually present; two postoculars; loreal absent; nasal single; parietal in contact with fifth supralabial, supralabials 6; infralabials variable (5 to 8); two pairs of enlarged chin shields; ventrals 239-277; caudals 33-48, most divided; anal single; scales smooth, usually in 19 ·rows; no scale pits; 48-79 dorsal blotches on body, 8-16 on tail; teeth smooth; palatal teeth present. One species, S. extenuatum, occurs in the northern half of the Florida peninsula. Although S. extenuatum is known from a very restrictricted geo- graphic region, the present work has demonstrated marked geographic variation. Thus, in the available material, there is virtually no over- lap in the dorsal blotch counts of specimens from eastern Marion County (Lake Kerr and Eureka) and those from Alachua County, although the distance between these two areas is only approximately SYSTEMATICS AND VARIATION OF STILOSOMA 83 40 miles. Other differences between these two populations are also demonstrable. More than half of the specimens from the eastern por- tion of the range have fused internasals and prefrontals, and over 70% have a feduction in the number of infralabials. In the Alachua County sample the internasals and prefrontals are separate scales, and there are usually seven or more infralabials. The difference in the size and number of dorsal blotches between individuals from these localities is so striking that a person unfamiliar with this species can easily segregate specimens from the two areas without even counting the dorsal blotches. Only-one of 22 eastern specimens has neither fusion of anterior infralabials nor of internasals and prefrontals, while 15 of 16 western specimens lack both types of fusion. Clearly, we have populations which differ phenotypically to such an extent that accord- ing to current systematic procedure they should be given separate no- menclatorial status. I believe the relationship is subspecific because there is slight morphological overlap, the forms replace each other geographically, and there is eVidence of inteFgfadation. Upon further analysis, it is evident that there is a third race occurring to the south- ivest of the above mentioned subspecies. It has fewer dorsal body blotches than the Alachua County form and is distinguished from the eastern race by possessing separate internasals and prefrontals, and a greater number of infralabials. The type specimen of Stilosoma extenuatum was collected at Lake Kerr, Marion County, and is a typical example of the eastern subspecies, which, thereby, becomes the typical race. The other two forms are herein described as new sub- species. Stilosoma extenuatum extenuatuth Brown Diagnosis:-An eastern subspecies of Stilosoma extenuatum with 68 or fewer dorsal body blotches; 13 or fewer tail blotches in males; usually a fusion of two or more anterior infralabia}s; and fusion of the pre- frontal and internasal scales in over half the known specimens. Holotyl,e:.ANSP 3351 , a male, received at the Philadelphia Zoological Garden from Mr. N. P. Fry, collected at Lake Kerr, Marion County, Florida. Redescription of Holotype:-247 ventrals ; 44 caudals , all divided ; anal plate un - divided; supralabials 6-6; infralabials 6-6 (4th largest); no preoculars; 2 postoculari, internasal and prefrontal fused into one scale which is in contact with the anterior border of the orbit; nasal single, in contact with first and second supralabials; parietals very large, in contact with fifth supralabial; third and fourth supralabials entering orbit; tempdrals 1-1, no loreal; two pairs of elongated chin shields; scale 84 BULL. OF THE FLA. STATE MUSEUM. BIOLOGICAL SCIENCES rows 19-19-17; total length, 511 mm.; tail length, 47 mm.; ratio of body length to tail length, 9.9, 61 clorsal bjotches on right side, 57 on left; 11 dorsal blotches on tail; lateral blotches usually located below and between the dorsal blotches and extending downward to the, ventral scutes. Coior (Villalobos and Villalobos, 1947) (in alcohol), dorsal and lateral spots brown (SO 5-20); ground color light tan (OOY 15-20); light brown blotches on ventrals similar to dorsal spots anterionly but becoming lighter posteriorly (O 10-30); mottling of' tiny brown spots throughout ground color except for a clear area on median dorsal scales between blbiches. Brown (1890: 199) states that these three median dorsal scales are inottled with pale red, but 'this color has now faded. There is an injury on the left side of the back at the third dorsal blotch. The underSide of the tail is blotched with white. Discussion.·-Specimens of Sti/oso,na e. extenuatum from Lake, Orange, Seminole, and Polk. Counties are very similar to those from the vicinity 6f the type locality. Little geographic variation in the characters which are diagnostic of this race is noted in specimens from these counties. Two specimens (MCZ 7262 and USNM 123555), both from Orange County, are unusual in having a greater number of dorsal blotches (76 and 71 respectively). Two specimens from Polk County and one from Seminole County have separate prefrontals and inter- nasals, but in' Marion,· Lake, and Orange Counties, where larger series are available, over 60% of the snakes have fused prefrontals and inter- nasals. This condition is illustrated in Loennberg (1894: 323. fig. 2) and in Schmidt and Davis (1941: 192, fig. 59).2 Loennlierg (1894: 323) states, that he obtained three specimens of Stilosoma extenuatum in Orange C6unty: one at Lake Charm, near Oviedo (now in Seminole County), and the other two at Oakland. One of these specimens was presented to the United States National Museum (USNM 21327). R now bears the data "Oviedo, Orange County." The other. two, therefore. must have been the specimens taken at Oakland. These were retained by Dr. Loennberg and placed in the University of Upsala collection. The character of the, prefron- tals and internasals of these specimens is described by Dr. Stejneger in Loennberg's aper and thus may be included in the data on this character (table II). All but five specimens of this subspecies have a fusion of two or more anterior lower labials on one or both sides. Ditmars (1938: pl. 39) published a photograph of a Stilosoma ex- tenuatum with approximately 60 dorsal body blotches. It appears to have fused internasals and prefrontals and thus is.a typical specimen of 2-rhe,specimen 'figured by Schmidt and Davis is also typical of S. e. extenuatuth in having six lower labials. However, it is unusual in having only five supralabials. This condition is apparently of very infrequent occurrence since only two specimens examined had such a reduction (CNHM 3389, left side of head only; and CNHM 8557, right side of head only). SYSTEMATICS AND VARIATION OF STILOSOMA 85 S. e. extenuatum. In the caption accompanying ·the photograph, Dit- mars states tha t this form is a very rare species, found in Orange and Marion Counties, Florida, so it is quite probable that the specimen photographed is from the range of this subspecies. - Brimley (1910: 14) recorded a specimen from Ft. Meade, Hillsborough County (now in Polk County), which is the southernmost record for Stilosoma. The typical subspecies would be expected at this locality, but since the spedimen.has not been examined by the writer, the locality is indicated by a hollow circle on the distribution. map (figure 3): Stilosoma may eventually be collected in the Central Highlands of Highlands County, which extend for 50 miles to the south. Other Mate,·ial:-The series of 11 spediniens from Marion County in the Academy of Natural Sciences of Philadelphia collection all have a reduced number of infralabials and a dorsal blotch count typical of S. e. extenuatum. Seven of the 11 also have fused internasals and pre- frontals. They comprise a homogeneous sample and they may all have been Collected in the vicinity of Lake Kerr. There are three other specimens from Marion County without speeific locality data: MCZ 13598 from the "Oklawaha River," col- lected in 1920 by A. G. Reynolds; USNM.56239, collected in 1900 by J Hurter; and CNHM 3389, collected in 1922 by C. C. Tyler (Tyler collected AMNH 20359 and 22431 in 1921-22, at Eureka, Marion County, so this specimen may also have been collected _in the same part of the county). All three of these specimens have a low dorsal blotch count (53, 61 and 62 respectively), a reduced number of an- terior infralabials, and two of the three (USNM 56239. and CNHM 3389) have fused· internasals and prefrontals. Distribution.-From southeastern Putnam County and northeastern Marion County, east of the Oklawaha River, south through Lake, Seminole, Orange, and Polk Counties. There are no records of Stilo- sonia east of the St. Johns River (figure 3). Sp.ecimens ExamMed:-Florida: Putnam County, Norwalk (ANSP 26311). Mai·ioiz County, (USNM 56239, CNHM 3389, ANSP 11831-35,8 16970-71, 26312-15; Oklawaha River (MCZ 13598); Lake Kerr (ANSP 3351-holotype, 18920, 12005-6, 26310); Eureka (AMNH 20359, 22431). Lake County, Eustis (USNM 24353); near Eustis (USNM 76750); near Altoona (AMNH 63413); Tavares (CM 19859, USNM 86813); 2 mi. S. Tavarts (USNM 81099); Umatilla (UF 1570-headless skin). Seminote County, Oviedo MANSP 11835 is now MCZ 9309. 86 BULL. OF THE FLA. STATE MUSEUM, BIOLOGICAL SCIENCES (USNM 21327). 07·ange County, (MCZ 7262); 216 mi. W. Lockhart (specimen to be deposited in Rollins College collection); Orlando (CM 7153); vicinity of Orlando (USNM 123554,55). P6lk County, Auburndale (USNM 45569-headless skin, USNM 60500); Winter Haven (ERA-WTN 15617). Stilosomn extenuatu,n, are,Iicola, new subspecies Diagnosis.-A southwestern subspecies of Stitosonia extenuatum with 68 or fewer dorsal body blotches; no fusion of anterior infralabials (usually 7); and with prefrontals and internasals separate. Holotype:-UF 1574, a male, collected at Lecanto, Citrus County, Florida, in Sep- tember, 1927, by 0. F. Swed. Paratypes:-Florida: Citnts County, Floral City (ERA-WTN 15619); Hernando (ERA-WTN 15623). Sumter County , Bushnell (ERA-WTN 15620). HernandoCounty, 7 mi. N. Brooksville (specimen in WithIacoochee Wildlife Collection (no number) ). Pasco Co„nly, New Port Richey ( UF 2815). Pine//as Co.uniy, Tarpon Springs (USNM 23333). Description of Holotype.-254 ventrals ; 46 caudals , 10 undivided ; anal plate single ; supralabials 6-6, infralabials 7-7 (5th largest); 1 preocular; 2 postoculars; internasals and prefrontals separa te; nasal single, in contact with first and second supralabials; parietals very large, in contact with fifth upper labial; third and folirth supralabials entering orbit; temporals 1-1; no loreal; 2 pairs of elongated chin shields; scale rows 19-19-17; total length 494 mill.; tail length 46 mm.; ratio of body to tail length, 9.7. 66 dorsal body blotches on right side, 67 on left; 13 dorsal blotches on tail. Color (Villalobos and Villalobos, 1947) (in alcohol),· dorsal blotches SSO 2-30, ground color OOS 15-30; ventral dark bIotches variable, ranging from light brown (SO 5-30) to dark brown (OOS 11-40). The underside of the tail is b16tched with white. Discussion:-An insufficient number of specimens of this form are available to determine the nature of variation in dorsal blotches within the subspecies. T.he three specimens from Citrus County have 67 or 68 blotches; the Sumter County specimen has 67; the Hernando County specimen has 68; the Pasco County specimen has 65; and the ~ specimen froni Pinellas County has 48, the lowest bloteh count 6f all the· Stilosoma examined. The Pinellas County specimen was collected at the southern extreme of the known range of the subspecies a, enicola, but it is doubtful that it represents a different population occurring in that area, since the specimen from New Port Richey, less than ten miles to the north, has almost as many blotches as the Citrus County specimens. SYSTEMATICS AND VARIATION OF ST/LOSOMA 87 $ , 001 O P -e'- A L 0 Z n 0 9 4// 0 Figure,3.-Map showing the distribution of the subspecies of Stilosoma extenuatum. Circles represent speciniehs of S. e. extenuatum; squares, S. e. arenicola; and tri- angles, S. e. mu ttistictum. The cross repFesents the series of intermediate speci- mens from Silver Springs. The soilthernmost triangle represents one specimen from Dumiellon , Marion County , which fits the description of S. e . muitistictum, but is probably also a member of the sallie intetmediate population. The Wicomico shoreline (100 foot contour line) is also indicated. Solid symbols represent localities from which specimens have been examined. Hollow symbols represent litfrature records. 88 BULL. OF -I=HE FLA. STATE MUSEUM, BIOLOGICAL SCIENCES Distribution:-From Citrus County south to northern Pinellas County and east to central Sumter County. This subspecies may also be expected to occur in Hillsborough County, at least in the northwestern part. Specimens Examined.-All seven specimens of this subspecies examined have been designated above as holotype and paratypes. Stilosom.n. extenitatimi. multistictuin, new subspecies Diagnosis:-A northern subspecies of Stilosoma extenuatu'm with 69 or more dorsal body blotches and 14 or more blotches on the tail of males; no fusion of anterior infralabials (uslially 7); and prefrontals and internasals separate. Holotype:-UF 8088, a male collected at the Devil's Millhopper (3.7 mi. N.,4.2 mi. W. Gainesville), Alachua County, Florida, on July 4, 1953, by Richard Highton. Paratypes:-Florida: Alachua County, Gainesville (UMMZ 56986, €NHM 8557); 46 mi. S. Hogtown Sink (5.2 mi. W., 2.1 mi. S. Gainesville) (UF.1571); 5 mi. E. Gaines- ville (UF 396) Description # of Halotype:-245 ventrals ; 42 caudals , all divided except numbers 18 , 19 and 36. which are entire; anal plate undivided; supralabials 6-6, infralabials 7-7 (5th largest); 1 preocular; 2 postoculars; internasals and prefrontals separate; nasal single, in contact with first and second supralabials; parietals very large, in con- tact with fifth upper labial; third and fourth supralabials entering orbit; temporals 1-1; no loreal; 2 pairs of elongated chin shields; scale rows 19-19-19; total length, 318 min.; tail length, 29 min., ratio of body to tail length, 10.0; 70 dorsal blotches on right side, 71 on left; 15 dorsal blotches on tail: Color (Villalobos and Villalobos, 1947) (in alcohol), dorsal and lateral sp6ts dark brown (50 3-20); ground color (SO 11-20); ventral dark areas black (Gray-3); median dorsal scales (O 14-20). these scales were OOS 12-120 iii life. The under side of the tail is very dark: Other Material:-Florida: Alachua Coun/,3:4 mi. W. Gainesville (UF 1572-73 (both in poor condition), 8268) Leuy County, Chiefland , collected by A . Hyatt Verrill (CM 24918). UF 8268 is not designated as a paratype because it has only 64 dorsal blotches. One other specimen found DOR 2 miles north of Gainesville was also examined. The posterior third of this specimen is missing so the total number of dorsal, body blotches can not be counted. However, the closeness of the blotches indicates it is typical of multistictum. The internasals and prefrontals are separate, but the head is. too damaged to count the infralabials Discussion:-The relationships of this race with the other two sub- species will be discussed below in the section on the origin of sub- specific differentiation. SYSTEMATICS AND VARIATION OF STILOSOM,4 89 Dist,·ibution:-This forni is known only from the vicinity of Ga~ines- ville, Alachua County, and from Chiefland, Levy County, Fl rida. Intergradation:-Specimens from Marion County, west of the Okla- waha River, show, an intermixture of the characteristics of all three subspecies. Seventeen specimens from the vicinity of Silver Springs have been examined (ERA-WTN 15615-16, 15618, 15621-22; UF 81; CNHM 48433-422 UMMZ 79579) and one specimen from Dunnellon, Marion County (ERA-WTN 15624). In addition, there, are 11 speci- mens (MCZ 41978-83 and CM 9407-11) from Marion County that were sent to these museums by the Ross Allen Reptile Institute in. 1936 and 1937, some of which may have been collected in the vicinity of Silver Springs. The specimens froni western Marion County with exact locality data have an intermediate average number of blotches, ranging from 58 to 79 (mean 67.8 + 1.43) (see figure 1). Four specimens have a reduced infralabial count because of fusion of anterior lowef labials (see table I), and one specimen has fused internasals and prefrontals (see table II). This population seems to be Close to arenicola and multistictum on the basis of the character of the infralabials, pre- frontals and internasals, and with an intermediate average number of dorsal blotches. A photograph of a specimen from Ocala, Marion County, was published 1,y Allen and Neill (1953: 8). It appears to have about 68 dorsal body blotches. The eleven specimens from the Ross Allen Reptile Institute without exact locality data have dorsal blotches ranging from 63 to 76 (mean 69.4 + 1.10) Specimens with Doubtful Locality Data:-Of two specimens in the University of Miami collection, one (UM 55-572) is listed as having come from Dothan, Alabama, and the other (UM 55-573) is listed as "probably" from Dothan, Alabrima. The former has 67 dorsal blotches, the latter, 78. The specimen with 78 dorsal blotches has a reduced number of infralabials on one side of the head. Thus, they appear to show characters of all three subspecies. Stilosonia may actually occur 200 miles to the northwest of its present northernmost known station, but it seems unwise to accept these dubious records until further sub- stantiated by additional specimens with accurate 16cality data. The UM specimens are typical of those from Silver Springs which have been ICNHM 48433-42 are listed only as "Marion County, Florida," but Mr. Wilfred T. Neill has checked Ross Allen's notes on this series. They were all plowed u~ at Mt. Canaan, a small communitr about one mile sdiath of Silver Springs, Marion County, iii May, 1932: 90 BULL. OF THE FIX STATE MUSEUM, BIOLOGICAL SCIENCES sold to reptile fanciers by Mr. Ross Allen for many years. Three specimens (CNHM 38016-18) from "Florida" were sent to the Chicago Natural History Museum by the Ross Allen Reptile In- stitute in 1941. These specimens are also probably from the Silver Springs area. They have 77,57 and 63 dorsal·blotches, separate inter- - nasals and prefrontals, and lack any fusion in the anterior lower labials. KEY TO THE SUBSPECIES OF "STILOSOMA EXTENUATUM" The following key will correctly identify 95% of the specimens examined (other than intergrades). 1 A-Lower labials 7; internasals and prefrontals separate ________-__ _- 2 1 B-Lower labials 6 on one or both sides of head and/or internasals and prefrontals fused ___-_______----_--___- extenuatum 2 A-Dorsal body blotfhes 69 or more -_...-----_.---_-___-_-_---_ multistictum 2 B-Dorsal body blotches 68 or less arenicola ORIGIN OF SUBSPECIFIC DIFFERENTIATION The rise and fall of sea level during the Pleistocene has had a pro- found effect on land area in Florida. During the interglacial periods when the climate was warmer than at present, sea level rose, and much of the present peninsula of Florida was under water. Numerous islands were formed, isolating land animals from their continental relatives. There are many shore line features present today at various eleva- tions in Florida. Thefe is not complete agreement among geologists on the exact altitudes of the different terraces f6rmed by encroachments of the sea on the peninsula duringthe Pleistocene. However, there is general agreement among three recent workers (Cooke, 1945; Mac- Neil, 1950; and Vernon, 1942) that one of the well-defined terraces is present at or near the 100-foot elevation. They agree that the sea was at this level during Sangamon interglacial time, which according to Flint (1947: 242, 284) lasted about 120,000 years and ended with the beginning of the Wisconsin glaciation about 55,000 years ago. However, the sea level was not at the present i00-foot contour during the entire Sangamon interglacial period. Figure 3 shows the distribution of the races of Stiloso,na extenua- turn. The 100.foot contour line is also included on this map. It is clear that a portion of the area now occupied by the races extenuatum SYSTEMATICS AND VARIATION OF STILOSOMA W and arenicola was separated from the mainland at the time sea level stood at the present 100-foot elevation. (The surface has been only slightly modified by erosion between that time and the present, , judging from existing shore line features.) If Stilosoma were present in Florida at that time, it would have been divided into two or more island populations and a mainland population. If isolated for a sumciently long period, these populations might have differen- tiated into the three presently recognized subspecies. lf the sea level rose above the present 100:foot contour line, as believed by most geologists, then the area of the islands would have been further re- duced and isolati6n might have progressed from a still earlier period. Upon withdrawal of the sea, the region about Lake Kerr in eastern ,Marion County would first have been connected to the large south- eastern island. The short-tailed snakes from this area are typical of the 5ub5pecie5 extenuatum. Much of western Marion County was under water when the sea stood at 100 feet above its present level. This area later could have been colonized by snakes from either island or from the mainland to the north. If colonized from all three areas, inter- breeding might have occurred and the population could now be inter- mediate in nature. The present low Oklawaha River valley separates the western from the eastern part of the county. This valley, even now, is probably an effective barrier t6 a sand dwelling fossorial snake like Sti/osoma. The chances for gene exchange between exienuatum and the western Marion County population would have been much re- duced until the sea level receded below the present 40-foot contour line. Thus, the population in western Marion County would be ex- pected to shdw more resemblances to the mainland form (multi- stictum) and the southwestern island form (arenicola) than it would to extenautum. This is exactly the fituation in the Silver Springs sample. A_t present, the possibility for gene .How between the three sub- species is probably much reduced. The Oklawaha River separates the populatiods of extenuatum in eastern Marion County from the popu- lations to the north and west. The Withlacoochee River is the north- ern and eastern boundary of Citrus County and comprises most of the eastern border of Hernando County. A very short distance to the south, the Hillsborough River effectively isolates the western Hills- borough and Pinellas County populations from the Polk County popu- lations of extenuatum. Interbreeding between arenicola and extenua- tum could now be taking place in southeastern Pasco Caunty. If afenicola has established itself on the east side of the Withlacoochee 92 BULL. OF THE FLA. STATE MUSEUM, 1HOLOGICAL SCIENCES River (as indicated by the Sumter County specimen), then perhaps intergradatien between these two forms may occur in Sumter and southern Lake Counties. Specimens are not available from either of these areas of possible intergradation. As mentioned earlier, extenuatum differs quite mai.kedly from multistictunt although the distance between the Lake Kerr population of extenuatum and the Gainesville population of mu.ltistictum is only about 40 miles. The two races are now probably completely isolated from one another by the Oklawaha River. Their relationship to each other is obviou5ly through the subspecies arenicola, which is similar to extenuatum in blotch count and similar to multistictum in head scutellation. Without any knowledge of its fossil history, it is, of course, impossible to determine the ancestral condition of this very distinctive genus. Either arenico/a or mullisticium may be closer to the ancestral type, since extenuatum appears more specialized in the reduction in number of head scales from the more usual colubrid type. It may be argued that since the,Silver Springs population is in ter- mediate between the races arenicola«and multistictum in number of dorsal blotehes (the character used to distinguish the two), an in- ternal cline exists in this character, the number of blotches gradually decreasing from north to south: It has been shown that this is not the situation in the eastern races extenuatum, but this does not rule out the possibility that clinal variation of this nature could be present in the western portion of the range. If this were the case, then most systematists would recognize only two races, an eastern subspecies (extenuatum) and only one western subspecies. However, I do not believe this to be the true situation. The sample of multistictum has a low coefficient Of variation in the number of body blotches (5.3%), as do other homogeneous samples collected from a limited area. The intermediate population from Marion County, west 6f the Oklawaha River, has a higher coefficient of variation (8.9%). The high degree of variability of the Silver Springs population may be interpreted as an argument that this population is an intergradiant one, rather than merely an intermediate one in a smooth internal cline. Large series of specimens from each of the west coast counties must be obtained before determining whether or not this interpretation is correct. If the three subspecies were isolated on islands resulting froni changes in sea level during the Pleistocene (or perhaps,earlier), then it niight be expected that each subspecies would have developed other distingui5hing cliaracteristics during the period of isolation. In general, specimens of multistictum appear much darker than individuals of SYSTEMATICS AND VARIATION OF ST/LOSOMA 93 the other two forms. This may, however, be due entirely to artificial conditions, such as length or method of preservation, and actually not be a real difference. Careful color notes should be taken on living specimens with a standard color guide so that this character may be further analyzed by later workers. In addition, specimens of muitistic- tum have the brown dorsal coloration extending down onto the belly, and especially onto the ventral surface of the tail, to a greater extent than is the case in arenicola and extenuatum. It is quite possible that ' examination of living specimens may reveal other color differences. Subspecies of several other Florida reptiles and amphibians inter- grade in the north-central portion of the peninsula (Carr, 1940: 9). The detailed distrib,ution of some of these subspecies has not yet been studied, but it will be interesting to determine if their distribution is similar to that of the races of Stilosoma extenuatum. Rhineum Bori- dana has a range somewhat similar to Stilosoma and it might be profitably investigated for similar subspecific differentiati6n. Two other Florida endemics, Neoseps reynoldsi and Sceloporus woodi have ranges which apparently reach their northern limit in the same region as does the range of the subspecies S. e. extenuaturn. Apparently suitable habitat for Stiloso,na occurs to the north and south of its known distribution, and it is quite probable that this snake may be discovered beyond its present known range. ECOLOGICAL N0TES This species was recorded from three different ecological associa- tions by Carr (1940: 86). These are high pine, upland hammock and rosemary scrub. Dr. Carr informs me that his record of Stilosoma ex- tenuatum from rosemary scrub is based on a specimen from Lake Coun ty found DOR in a small patch of scrub completely surrounded by high pine (longleaf pine-turkey oak association of Laessle, 1942). There are no other records of this snake in rosemary scrub that are known to me. The specimens dug out of sphagnum beds reported by Carr (1940: 86) certainly were in an unusual habitat for this xeric species. The holotype of multistictum was collected at dusk, crawling on a sand road in the ecotone between high pine and upland hammock. Two specimens (UF 1572-3) were collected in early November, 1953, in high pine association, 1.3 miles west of the Gainesville city limits by Mr. Don Altmix, who informs me that both were found crawling on the surface of the ground during daylight hours. Another (UF 8268), collected by Mr. Altmix, was crawling on the surface of the 94 BULL. OF THE FLA. STATE MUSEUM, BIOLOGICAL SCIENCES ground at night at the same locality on October 7, 1955. Another (UF 1 571), was collected by Dr. J. C. Dickinson, Jr., at 5 p.m. on November 19, 1953, crawling on the surface of an upland hammock about 6 miles southwest of Gainesville. Van Duyn (1939: 51) also reported a specimen from dry upland hammock. The specimen from 242 miles west of Lockhart, Orange County, was dug up in a garden situated in high pine association. Mr. Wilfred T. Neill has kept careful habitat records of Stilosdma colletted in the Silver Springs area. Many specimens are plowed up by farmers in the spring of the year at Mt. Canaan, a small community south of Silvet Springs, in cut-over high pine association. Several other specimens have been taken at Silver Springs in high pine as- sociation, and· the specimen from Bushnell, Sumter County was als6 collected in this type of habitat: Mr. Neill has also supplied the information that the town of Lake Kerr (the type locality of S. extenuatum), although at present located in rosemary scrub, formerly was situated several miles from its present location. He has visited the remains of the former community and found it to be located in high pine. The great- number of records for this species in high pine seems to indicate that it is truly a characteristic inhabitant of this ecological association (Carr, 1940: 15). There is one record of predation on Stilosoma. UF 1570 was re- moved from the stomach of a DOR Micrunts f· fulvius found in high pine on September 30, 1949, at Umatilla, Lake County, by Mr. Walter Auffenberg. Ditmars (1939: 107) states that two captive specimens killed small brown snakes by constriction, but refused small lizards and newborn mice. However, Allen and Neill (1953: 8) list lizards as well as other snakes in their diet. A captive specimen ate a Tantil/a coronata two- thirds its own length '(Carr, 1934: 139) SUMMARY A study of variation in the available specimens of the short-tailed snake has indicated that there are three well-defined races of this species in the northern half of the Florida peninsula. Tw6 of these are described as new subspecies. Characters found to exhibit geogra- phic variation were number of dorsal body blotches, tail blotches in males, and fusion of certain head scales (anterior infralabials; prefrontals and internasals). Sexual dimorphism was pre5ent in num- ber of ventrals, caudals, tail blotches and in the body/tail length ratio. SYSTEMATICS AND VARIATION OF STILOSOM/1 95 The r¢duction in land area of the Florida peninsula due to the rise in sea level during the Pleistocene has been suggested as a possible ex- planation for the unusual microgeographic raciation of Stilosoma. S. e. extenuatum could have been isolated on the large island centering in Polk County; S. e. arenicola may have been isolated on a smaller island in Citrus, Hernando and Pasco Counties; while S. e. multistictum might have differentiated on the stubby peninsula to the north at the time when the sea level stood at 1 00 feet above its present elevation: Habitat records for this snake indicate that it is most Often found in xeric situations, especially in high pine association. LITERATURE CrnED ALLEN, Ross, ami Wilfred T. NEILL. 1 958. The Short-tailed snake. Florida Wild- life, 6 (11): 8-9,1 fig. BERG, Ch. 1901. Herpelological notes. Comun. Mus. Nac. Buenos Aires, 1 (8) 289-291. BLACKWELDER, Richard E. 1944. Checklist of the coleopterous insects of Mexico, Central America, the West. Indies, and South America. Part I. Bull. U. S. Nat. Mus. (185): i-xii, 1-188. BLANCHARD, Frank N. 1924. A key to the snakes of the United States, Canada and Lower California. Papers Mich. Acad. Sci. Arts and Letters, 4 (2): i-xiii, 1-65, figs. 1-78. BOULENGER, George A. 1894: Catalogue of the snakes in the British Museum (Natural History). Vol. 2., London: i-xi, 1-382, pls. 1-20. BRIMLEY, C. S. 1910. Records of some reptiles and batrachians from the south - eastern United States. Proc. Biol. Soc. Wash., 23: 9-18' BROWN, Arthur Erwin. 1890. On a new genus of Colubridae froin Florida. Proc. Acad. Nat. Sci. Phila., 42: 199,200. 1901. A review of the genera and species of American snakes, north of Mexico. Proc. Acad. Nat. Sci. Phila., 53: 10-110. CARR, Archie F., Jr. 1934. Noles on the habits of the short-tailed snake. Copeia (2): 138-139 ........_.----_. 1940. A contribution to the herpetology of Florida. U. Fla. I'lll}., Biol. Sci. Ser., 3 (1): 1-118: COOKE, C. Wythe. 1945. Geology of Florida. Bull. Fla. Geol. Survey (29): 1-339, COPE, E. D. 1892. A critical review of the characters and variations of the snakes of North America. Proc. U. S. Nat. Mus., 14 (882): 589 694. ... _ __ 1895. The classification of the Ophidia. Tran. Amen Phil. Soc., 18: 186-219, pls. 14-33. 1900. The crocodilians, lizards, and snakes of North America. Ann. Rept U. S. Nat. Mus., 1898: 153-1270, figs. 1-346, pls. 1-35. DITMARS, Raymond L. 1933. Reptiles of the world. The Macmillan Co., New York: i-xx, 1-821, pls. 1-89. 1939. A field book of North' American sliakes. Doubleday, Doran & Co., New York: i-xii, 1-305, pls. 1-48. 96 BULL. OF THE FLA. STATE MUSEUM, BIOLOGICAL SCIENCES FL] NT, Richard Foster. 194Z. Glacial geology and the Pleistocene Epoch. John Wiley & Sons, New York: i-xviii, 1 -589, figs. 1 -88, pls. 1-6. GAY, Claude. 1849. Historia fisica y politica de Chile. Vol. 4 (4): 1 -511. KLAUBER , Laurence M . 1940 . Two new subspecies of Phyllorhynchus, the leaf- nosed snake, with notes on the genus. Trans. San Diego Soc. Nat. Hist., 9 (20): 195-214, pl. 8, map. LAESSLE, Albert M. 1942. The plant communities of the Welaka area. Univ. Fla. Pub., Biol. Sci. Ser., 4 (1): 1 -143, figs. 1-25, pls. 1-14. LOENNBERG, Einar. 1894. Notes on reptiles and batrachians collected in Florida in 1892 and 1893. Proc. U. S. Nat. Mus., 17 (1003): 317-339, figs. 1-3. MACNEIL, F. Stearns. 1950. Plei5tocene shore lines in Florida atid Georgia. U. S. Geol. Survey Professional Paper, (221 F) 95-107, pls. 19-25. SCHAUM, Hermann Rudolph. 1852. Bericht Ober die wissenschaftlichen Leistun gen im Gebiete der Entomologie fur 1851. Archiv. Naturgesch., 19: 1-152: SCHMIDT, Karl P., and D. Dwight DAVIS. 1941. Field book of snakes of the United States and Canada. G. P. Putnam's Sons, New York: i-xiii, 1-322, figs. 1-103, pls. 1-34, frontispiece. STE~NEGER, Leonhard, and Thomas BARBOUR, 1917, A check list of North American amphibians. and reptiles. Harvard University Press, Cambridge: i-iv, 5-125. 1923. Ibid., 2 ed. i-x, 1-171. Ibid., 3 ed.: i-xiv, 1-185. Ibid., 4 ed.: i-xvi, 1-207. TUCKER, Henry. 1911. Scale variations in Stilosoma extenuati„ii (A. E. Brown) Proc. Acad. Nat. Sci. Phila., 63: 549-550. VAN DUYN, Guy. 1939. Extension in the range of Slilosuma certen'ual.tim. Copeia(1): hi-52 VERNON, Robert 0. 1942. Geology of Holmes and Washington Counties, Florida. Bull. Fla. Geol. Survey, (21): 1-161. VILLALOBOS, C., and Julio VILLALOBOS. 1947. Atlas de los colores. Buenos Aires: i-xv, 1 -86, pls. 1 -38. Bulletin of the Floridi Slate Museum, Biological Sciences Volume.I, Number 2, 1956