BULLETIN OF THE FLORIDA STATE MUSEUM BIOLOGICAL SCIENCES Volume 11 Number 2 THE PINE WOODS SNAKE, RHADINAEA FLAVI LATA (COPE) Charles W. Myers :t of '' '- /853 ' UNIVERSITY OF FLORIDA Gainesville 1967 Numbers of the BULLETIN OF'THE FLORIDA STATE MUSEUM are pub- lished at irregular intet*als. Volume* contain about 300 pages and are not nec- essarily completed in any one Balundar' year. WALTER AUFFENBERG, Managing Editor OLIVER L. AUSTIN, JR., Editor Consultant for this 1*sue: Charles W. Myefs Communications concerning purchase 6r achange of the publication and all manuscripts should be addressed to the Managing Editor of the Bullatin, Florida State Museum, Seagle Building, Gainesville, Florida. 82601 Published February 3, 1967 Price for this issue $.90 THE PINE WOODS SNAKE, RHADINAEA FLAVILATA (COPE) CHARLES W. MYERS 1 SYNopsIs: Rhadinaea flavitata (Cope, 1871) is a small colubrid snake inhabiting a narrow coastal range in the southeastern United States. Its distribution closely apprOximates that of the ldw pine fiatwoods, which constitute its principal habitat. A few individuals have been found ih Other situations, especially in Florida ham- mocks and on coastal islands. The species is mb5t commonly encountered in pine logs and stumps during March and April; with the approach of warmer weather and drier conditions in May and June, Rhadinaea and other small snakes become more difficult to find. R. flat;ilata feeds on small amphibians, lizards, and perhaps snakes; literature records referring to predation on .insects, are ques- tioned. The pine woods snake kills or incapacitates some 6f its prey with a venom introduced into wounds made by enlarged rear maxillary teeth. Other snakes are likely its main predators and food competitors. R. #auilata lays two to four elongate eggs; the period Of egg deposition is speculated.to extend from May into August, although it probably is not so extensive in any given season; the natural nest is unknown. There is slight statistical evidence of unbalanced sex ratios in some populations, Structural variation was observed in many characters, and for discussion is classified as intrapopulational variation (including 6ntogenetic, sexual, and un- correlated or "individzial" variation) and interpopulational (geographic) variation, the latter being compounded from the former. No geographic races are recog- nized in this species. Structurally the pine woods snake is a rather generalized Rhadinaea. Its closest living relative is R. laureata of the highlands of western Mexico. Fossils show that R. flavilata, or a genetic predecessor, had arrived in the southeastern United States at least by the third (Illinoian) glacial stage of the Pleistocene. The lack of marked geographic variation in this species can be explained by the relatively recent (postglacial) dispersal of a homogeneous Floridian stock through- out a relatively uniform pine-flatwoods habitat. Environment, nonselective gen- etic mechanisms, and natural selection are considered briefly in terms of the evolution of geographic variation. A seemingly primitive color pattern is re- tained at the northern limits of the range; the southern populations are charac- terized by loss of pattern and. by increased variability (often anomalous) in scutellation., It is inferred that southern populations, being most inf}uenced by a warming post-glacial climate, are evolving faster than northern populations. 1 The study reported on in this p.aper was initiated when the author was an undergraduate student at the University of Florida and a research- assistant in the Florida State Museum (1958-1960). Tha paper was completed at the C6- operative Wildlife Research Laboratory, Southern Illinois University, wbere it was accepted as a thesis in partial fulfillment of the requirements for the Master of Arts degree (1962). The author is currently a visiting scientist at Gorgas Me- morial Laboratory, Panama City, Panama, and research associate of the Museum of Natural History University of Kansas. An earlier contribution to this Bulle- tin deals with the 6iology of the ringneck snake in Florida. Manuscript received 1 December 1965.-ED. Myers, Charles W. 1967. The Pine Woods Snake, Rhadinaea favilata (Cope) Bull. Florida State Museum, voI. 11, no. 2, pA 47-97. 48 BULLETIN FLORIDA STATE MUSEUM Vol. 11 TABLE OF CONTENTS INTRODUCTION - 48 VARIATION 68 ' TAXONOMY - 51 Ontogenetic variation 70 DISTRIBUTION 54 Sexual dimorphism_,__ 70 ECOLOGY 57 Uncorrelated variation 75 Habitats --- 57 Geographic variation 78 Animal associates 61 AFFINITIES 88 Habitats and behavior 68 EVOLUTION - ---· -- -- 88 Sex ratios 67 LITERATURE CrrED 92 INTRODUCTION Rhadinaea #auilata ( Cope, 1871) is a small snake of the extensive pine flatwoods on the coastal plain of·southeastern United States. As with many small and secretive animals, especially those of limited distribution, knowledge of this species is slight. In this paper I have attempted to provide a more complete account of R. /lauilata.than heretofore available. The principal aim was to determine its struc- tural variation, distribution, affinities, and ecological relationships. On the data assembled is based a hyp6thetical history of the species' evo- lution. The written history of Rhadinaea #aoilata is not extensive. Post- surgeon H. C. Yarrow found the type specimen near Fort Macon on Bogue Banks, North Carolina, in November 1871 ( Coues and Yarrow, 1878). The novelty was forwarded alive to E. D. Cope, who prompt- ly named it Dromicus flavilatus, believing it to have affinities with D. callilawnus of Jamaica, and a possible origin in the United States via floating drift in the Gulf Stream ( Cope, 1871). Specimens next be- came available from Florida (Cope, 1877, 1878, 1888), and after an examination of the hemipenis Cope ( 1894, 1895)- decided that #avila- tim was most closely allied with the species of Rhadinaea, a genus he had described in 1863 from the tropical mainland. Boulenger ( 1894) concurrently placed the species in Liophis, a designation no other worker followed, and Boulenger is said ( Malnate, 1939) to have re- alized his error later. Steneger and Barbour ( 1917) allocated #auila- tus to Leimadophis, and Amaral ( 1929) placed the remaining species of Rhadinaea in Liophis. -Dunn resurrected-the .genus Rhadinaea in 1982, and since that time the affinities of Rhadinaea ~auilata have not been questioned. Here ignored are suggestions ( Dunn 1944, 1957; 1967 MYERS: THE PINE WOODS SNAKE 49 Roze 1958, 1959) that Urotheca Bibron, 1843 is the proper generic name for species currently placed in Rhadinaea Cope, -1863. I have examined the holotype of Calamaria dumerilli, type species of Uro- theca, and intend to take up this strictly nomenclatorial problem else- where. Rhadinaea #at)ilata has remained scarce in collections, although Brown ( 190I) early extended its known range to include the eastern Gulf Coast. Malnate ( 1939) made the first substantial contribution to a knowledge of this snake; he summarized previous literature, gave original observations on habits and habitats, and an~lyzed variation on the basis of 55 specimens. E. Ross Allen ( 1939 ) found the species abundant at Burbank, Florida, and cited a catch nearly twice.as large as Malnate's sample from the entire range. Other interesting 06ntri- butions were the discoveries that R. #avilata possesses a weak venom ( Neill, 1954a), and that· fossil vertebrae from the Florida Pleistodne can be assigned to it (Auffenberg, 1968; Holman, 1958, 1959). Several other writers cited at appropriate places in the text have supplied additional information. PR0CEDURE OF STUDY. Much of this paper is based on data from museum specimens. My field experience with the pine woods snake is limited to northern Florida. Available for this study were 192 pre- served specimens and a small series of skeletons in the following col- lections: American Museum of Natural History ( AMNH); Academy of Natural Sciences of Philadelphia ( ANSP); collection of Barry Man- sell ( BM); Chicago Academy of Sciences ( CAS); Carnegie Museum (CM); Chicago Natural History Museum (CNHM); Cornell Univer- sity, ( CU); collection of the writer ( CWM); Duke University ( DU); Illinois Natural History Survey ( INHS); collection of J. Alan Holman ( JAH); collection of Joseph Pylka ( JP); Louisiana State University Museum of Zoology ( LSU); Loyola University, New Orleans ( LU ); Museum of Comparative Zoology ( MCZ); Museum of Vertebrate Zoology ( MVZ); collection of Sam R. Telford ( SRT); Texas Coopera- tive Wildlife Museum ( TCWM) ; Tulane University ( TU); University of Florida ( UF) ; University of Illinois Museum of Natural History ( UIMNH); University of Kansas' Museum of Natural History ( KU); University of Michigan Museum of Zoology ( UMMZ); United States National Museum ( USNM); collection of W. E. Brode ( WEB); col- lection of William L. Engels ( WLE). Data on scutellation, color pattern, size, and sex were recorded when possible ( s6me specimens were poorly preserved or mutilated); observations on certain aspects of the internal anat6my were Made from selected specimens. All measurements and ratios' werd' based on 50 BULLETIN FLORIDA STATE MUSEUM Vol. 11 preserved specimens. Tooth counts were obtained from the skeleton- ized specimens; no attempt was made to record detailed osteological descriptions. The terminology used te describe the hemipenis is that of Dow- ling and Savage (1960). Ventral plates were counted in the manner suggested by Dowling (1951); neither the anal plate nor terminal caudal spine were included in the ventral or subcaudal counts. The Brst ventral plate (uide Dowling, 1951, Bg. 1) was used as the ref- erence point for the Brst dorsal scale count; this count started with the dorsal scale bordering the posterior corner of the first Ventral, and ended at the starting place for the next_ count (approximately a head's length behind the head). Other dorsal scale counts were made at midbody and immediately anterior to the anal plate. The temporals of Rhadinaea #avilata are defined as those plates lying in vertical rows between the parietals and supralabials. (A row of scales bordering the p6sterior margins of the parietals and labials has been considered temporals by some authors.) Numbers on the left and right sides of a slant line (/) show varia- ti6n in counts made on the left and right sides of the body, respec- tively (e.g. 9/10 infralabials). Numbers above and below a horizontal line show that a vertical division has occurred in one or' more plates 2 of a vertical row. Thus, the temporal formula 1 + - means that 1 2 the top plate in the sedond row is divided; the formula 1+- shows 2 that both plates are divided in the second row. The + sign indicates that counts have been made in vertically arranged rows,.rather than horizontal ones. Statistics found useful are range and standard deviation as meas- ures of absolute dispersion, arithmetic mean as d measure of central tendency, standard errors of means and chi-square (xe) as clues to the probability of difference between samples, coefficient of diver- gence (Klauber, 1940, 1948) as a relative measure of difference be- tween characters, and tail length divided by total length as a meas- ure of proportion. Except where sex ratios are tested, chi-square was obtained by use of a2x2 table and a formula that contains a Yates' correction factor for small numbers (Croxton, 1959, p. 276). Botanical names used are those given by Small (1988), except for the species of Pinus, where the nomenclature is that of Critchfield and Little (1966). 1967 MYERS: THE PINE WOODS SNAKE 51 ACKNOWLEDGMENTS. For making available facilities at the Flor- ida State Museum where this study was initiated, and for much en- couragement, I am grateful to William J. Riemer. The manuscript was prepared at Southern Illinois University, where it was improved by comments and helpful criticisms given by W. D. Klimstra and Richard E. Blackwelder. The paper also prOfited from a feading by William E. Duellman, University of Kansas. For Beld notes and other information I 'am indebted to Charles M. Bogert, William L. Engels, Sam R. Telford, and Wilfred T. Neill. Isabelle. Hunt Conant photographed the pine woods snake in fig. 1 and Robert McFarlane took the photographs in figs. 8,4, and 10 for me. For typing the man- uscript and other favors,' I am especially grateful to. my wife, Joan W. Myers. TAXONOMY - Rhadinaea flavilata (Cope, 1871) 1871. Dromicus jiauilatus Cope, Proc. Acad. Nat. Sci. Philadelphia, vol. 28, pp. 222-228. 1894. Liophis 8avilatus. (Cope)., Boulenger, Cat. Snakes British Mus. (Nat. Hist.),9 vol. 2, p 148. 1894. Rhadinaed #auitata (Cope). Cope, Proc. Acad. Nat. Sci. Philadelphia, vol. 46; p. 428. 1901. Rhadinea Kauitata (Cope) Brown, Proc. Acad. Nat. Sci. Philadelphia, -vol. 53, p. 88. [Probably an intended emenddtion as this,generic spelling was used elsewhere by Carman (1884, p. 29) and Brown (r904, p. 467; 1908, p. 123). Several recent authors, apparently following Schmidt's check- list (1953), have. used this spelling, but I have been unable to. find a valid basis for it.] 1917. Leimadophis flauitatus (Cope) Steineger and . Barbour, Check List N. Amer. Amphibians and Reptiles, Ist. ed., p. 86. 1958. Urotheca [flavilata (Cope) included by inferenbe]*' Roze, Breviora, Mus. Comp. Zool., no. 88, p. 5. [Followed in this usage by Neill (1968, p. 205; 1964, pp. 287-288).] HOLOTYPE. ANSP 5583, collected by Dr. H. C. Yarrow, in No- vember, 1871. Now. lost (Malnate, 1989; James E. B6hlke, in litt.). TYPE LOCALITY. Approximately 8 miles westward from Fort Ma- con, on Bogue Banks, Carteret County; North Carolina. In the original description, Cope (1871) says that the type speci- men came from "near Fort Macon, on the coast of North Carolina." Coues and Yarrow (1878) relate that it was found "on Bogue Banks some eight miles south of Fort Macon, near marshy grozind." Rob- ertson and Tyson (1950), however, poiht out that the last is an un- likely locality (being in the Atlantic Ocean), and that Coues and Yarrow probably intended 8 miles west or southwest of Fort Macon. 52 BULLETIN FLORIDA STATE MUSEUM Vol. 11 ETYMOLOGY. The name Rhadinaea Bacilata is presumably de- rived from the Greek rhadinos (slender or lithe) and the Latin flauus (gold colored, or yellow) + latus (broad or extensive, full or rich). It seems probable, at least, that the specific epithet was intended to refer to the rich (or extensive) golden-brown coloring of the body. Latus is also the past participle of the verb fero, and so #auitata conceivably could mean "golden borne, but this seems less likely. Latus is furthermore a noun meaning side or Hank, but it is unlikely " that the name is based on this. Cope added the feminine ending a" when he transferred Bavilatus from Dromicus to Rhadinaca, and so apparently did not intend a noun in apposition; and had "golden sided" been meant the name should have been -Bauilateralis." There seem to be no true vernacular names for this species, as humans living within the range of R. flauilata are usually unaware of its existence. The common name in current usage, yellow-lipped snake (A.I.S.H., 1956), is not suitable for most populations; it was apparently coined by Ditmars (1907), who described the upper lip as «bright yellow." The name "brown-headed snake" has been used, but is equally applicable to any of several other small snakes in the "eastern United States. Yarrow's Dromicus is the only other conn- " mon name that has been used in print. I suggest pine woods snake for those in need of an English name. The species is partial to the pine flatwoods, probably even more so than the pine woods treefrog, Hyla femoralis. 6 4- 1r Figure 1 . A pine woods snake, Rhadinciea flavilata, from Burbank, Marion County, Florida. Isabelle Hunt Commt DESCRIPTION. A moderately slender snake with head slightly wider than neck. Largest specimen examined 887 mm. Tail/total length ratio 27.0-85.9 percent. 1967 MYERS: THE PINE WOODS SNAKE 58 Color above golden brow-n, lightest on frst two scale rows. Mid- dorsal stripe of diffused chr6matophores sometimes present on verte- bral scale row, with the pigment often confined to apexes of the vertebral scales. Diffused lateral stripe usually present on scale rows 1-4, mainly on 2-8 anteriorly and on S posteriorly (Sometimes not evi' dent until the stratum corneum falls away in preservative). Some- times a yellowish occipital spot on each side of neck behind head. Top of head usually darker than body and often marked with pale vermiculations. A brown stripe, light-bordered above and dark- bordered below, extends from the snout through the eye and to the angle of the jaws. Labials white to pale yellowish, variably spotted. with black, and sometimes with faint splotches of brown. Underside of head and neck white; rest of venter white or pale yellowish green. Dorsal scales smooth, except for anal ridges on some specimens, and without apical pits; in 17 rows except immediately behind the head, where the count is usually 18 but occasionally 17 Or 19. Ven- trals 112-189; anal plate diyided; subcaudals in 59-88 pairs; tail with terminal spine. Supralabials normally 7, with third and fourth en- tering orbit and the sixth the largest; occasionally 8, with fourth and fifth entering the orbit and the seventh the largest. Infralabials usually 9, with the first Eve bordering the genials and the fifth the largest; infralabials sometimes 7, 8, or 10, often with corresponding changes in position of the latgest plate and the number bordering the genials. Posterior genials slightly longer than anterior ones; first pair of infralabials meet behind mental. Temporals typically 1 + 2, but these plates, especially in row 2, are frequently divided or fused. One preocular, two postoculars; pseudo-oculars some- times present as the result of transverse labial divisions. Loreal as high or higher than long, rarely absent. Nasal single, but grooved and may appear divided on casual examination. Rostral about twice as wide as high and barely visible from above. Internasals nearly quadrate; prefrontals nearly as long as wide; supraoculars narrow, longer than wide; frontal about one and one-half to two times longer than wide, pentagonal with apex caudad; parietals elongate; truncate or slightly pointed posteriorly. Everted hemipenis a single, clavate organ, extending to between the sixth and ninth subcaudals. Basal part smooth; middle part spinose with greatest enIargement of spines towards sutface opposite sulcus spermaticus; distal end capitate and calyculate (calyces papillate apically and spinulate basally). Sulcus spermaticus bifurcate, di- viding on the capitate portion of the organ at about the Sth or 6th 54 BULLETIN FLORIDA STATE MUSEUM Vol. 11 subcaudal level and not extending to the apex. The m. retractor penis magnus originates at the level of the 21st to 24th subcaudal. Anal sac extending to between the 9th and 14th subcaudals. Anterior maxillary teeth 14-15, slightly increasing in length from front to rear, followed by a short diastema and two enlarged, un- grooved teeth. Palatine teeth 11-18; pterygoid teeth 19-28; dentary teeth 19-22. The middle, precaudal vertebrae are described by Auf- fenberg (1968). DISTRIBUTION Rhadinaea Ravilata occupies a narrow coastal range (fig. 2) from the vicinity of Cape Hatteras, North Carolina (Carteret County), south oo 0 Figure 2. GeograRhic distribution of Rhadinaea #auilata. Only one sym- bol is plotted per county (or parish) but, with few exceptions, this covers all known lacalities within the county. Open symbols indicate literature records. 1967 MYERS. THE PINE WOODS SNAKE 55 through the northern four-fifths of the Florida peninsula (to Palm Beach County), and west to extreme eastern Louisiana (Livingston Parish). Most localities are less than 100 feet above sea level, although specimens have been found at approximately 180 feet in Gainesville, Florida. At no place has the species been found more than about 70 miles from the coast. The present-day range of Rhadinaea Bat)ilata appears to lie en- tirely east of the Mississippi River. A Texas record (Netting, 1986), based on a specimen (CM 8987) reputedly from the vicinity of Clif- ton, in Bosque County, is almost certainly erroneous. Rhadinaea jiat)ilata is a coastal form, whereas the Clifton area lies farther inland (by about 175 miles) and at a higher elevation (by over 400 feet) than other known localities for the specres. Correlated with this are con- spicuous differences in habitat. Rhadinaea ~avitata is found mainly in low, poorly drained pine woods, whereas Clifton is in a region of cedar-covered ridges and limestone outcroppings. I Visited this area in April 1961 with Sam R. Telford and Robert Mount; we col- lected or observed such animals as the plains narrow-mouthed toad (Gastrophryne olivacea), spiny lizard (Sceloporous oliuaceus), blind snake (Leptotyphlops dulcish ground snake (Sonora episcopa), and Rat-headed snake (Tantilla gracilis). These species occupy habitats more arid than the coastal pine Hatwoods, and are members of a faunal unit different fr6m that associated with R. flavilata. Grobman (1941, 1944, 1950) advised caution. in the acceptance of this and cer- tain other records based on specimens not individually tagged and for which the place of shipment may be given as the collecting locality. SPECIMENS EXAMINED.-ALABAMA: Baldwin County-10 mi. S Foley (CM 9879). Mobile Countg--no other locality data (CU 1789,· USNM 51888, 56445- 56446); Mobile (CAS 12180-12181). FL6RIDA: No other locality data (ANSP 10800, 26075; CNHM 38022-88028). Alachua County-Air Base (UIMNH 25700); Gainesville (CWM 1855; TCWM 10418; UF 460, 620, 2740, 2977, 7087, 7278, 8021, 8860); ·near Gainesville (AMNH 86589); approximately 6 mi. NE Gainesville (CWM 1863); 7.1 mi. NNE Gainesville (UF 10028); 7 mi. E Gaines- ville, ik mi. W Hatchet Creek (SRT 628); 7 mi. E, 242 mi. N Gainesville (CWM 1560, 1580-1582, 1800, 1860-1862, 1864); 2 mi. E Paradise (UIMNH 25701- 25706). Breuard County-Georgiana (USNM 11989, 13642, 18649, 13661,13708). Duual County-Jacksonville (BM [8]); Jacksonville, north section along St. Johns River (UF 3271): Glades County-Indian Prairie Canal, 18 mi. SW Okeechobee (towh) (UF 8864). Hamilton County-15 mi. NNE, 7 mi. W White Springs (town) (KU 68940). Indian River County-Sebastian (MCZ 12792; UMMZ 56987). Levy County-4 mi. S, 1 mi. W Otter Creek (town) (UF 14904). Marion Count~-no other locality data (CNHM 48287, 95841 [2]), Burbank (CNHM 56 BULLETIN FLORIDA STATE MUSEUM Vol. 11 48288-48296; MVZ 58907; TCWM 10416-10417; UF 2714, 2750 [2], 7504); Burbank, 7 mi. N Silver Springs (CNHM 48297-48306); 4 mi. NE Burbank (UF 10026); Ft. McCoy (UF 10028); near Silver Springs (CM 9636-9646); '10 mi. from Silver Springs (CU 2211 [2]). Okeechobee Countu-Opal, NE of Okee- chobee (AMNH 63891-63892); near Opal, N of Okeeshobee (AMNH 50491); S of Okeechobee (AMNH 68864, 63486:63438). Orange County-Orlando (MCZ 6978). Palm Beach County-Palm Beach (UMMZ 85110). .Polk County-5 nii. N Lakeland (JAH [skeleton orily]); JW mi. SE Pasco*Polk County line, on U.S. Highway 98 (JP 58-20); Winter Haven, Lake Shipp (SRT 49); 7 mi. SW Winter Haven, Lake Hancock (SRT 119, 1136) Putnam County-Univ. Florida Con- servation Reserve, Welaka Mud Sp¥ings (CM 21489). St. Johns County-An- astasia. Islahd, near St. Augustine (AMNH 68362-63863,63484-68435); Anastasia Island, 5 mi. SE St. Augustine (CU 4980). Seminole County-4 mi. E Sanford (UF 10024). Taylor County--7% mi. 815' from Perry (UP 14908). V.o~usia County-Daytona Beach (UF 10080); 8 mi. N DeLand (UF 10029); 7 mi. E DeLdnd (UF 10027). "Warren" [Walton ?] Count!,-no other locality data (ANSP 11780). GEORGIA: Charlton Cmmty-Okefenokee Swamp, Chesser's Island (CM 19869). LOUISIANA: Livingston Parish-5 mi. NW Springfield (LSU 7483-7435). St. Tammany Parish-Bayou Lacombe (TU 11849); CovingtOn (TU 3237-8288); Mandeville (*U 8285); Oaklawn (TU 3289); Pearl River (town) (TU 869, 3889); 5 mi. W Slidell (TU 7081, 14991, 15044, 15073); Sun (CAS 12182); 0.8 mi. N Talisheek (TU 16098). MISSISSIPPI: No 6ther. locality data (USNM 56443-56444) Jackson, Hancock, Ham'son, and Pearl River counties. [The tags on the following specimens were attached by slip-knots and came loose during shipment.]-(WEB Bla-8lb, 87, 52, 68,78,87,91, 109, 111-113, 115). Hancock Count~-Bay St. Louis (ANSP 12061-12062; LU 456 [2]; USNM 24452-24454); 10. mi. W Bay St. Louis (TU 13770 [2]), 4 mi. NE Logtown (TU 14264); Pearling- ton (LU 295 [8]); 7.5 mi. NNW Pearlijgton, just W of Westonia(TU 17618); 5 mi. N, 8 mi. W Pearlington (TU 17851-17852); 4 mi. W Waveland on U. S. Highway 90 (UF 10025 [2]). Harrison Countu-Biloxi (CNHM 21538, CU 1867 [2]; UMMZ 76827; USNM 125546); near. Biloxi (AMNH 46745; CM 5240), 3 mi. N Biloxi (CNHM 12000), 6 mi. N Biloxi (INHS 6886); 13 mi. NW Biloxi on State Highway 55 (TU 17358); Gulfport; AAF (UMMZ 98998); 4 mi, N d'Iber- ville (UIMNH 29110). NORTH CAROLINA: Biaden County-North River, Cam- deh Pt., Councils ["Gouncil" on recent maps] (CU 1342). Carteret County- 6 mi. E Beaufort, on Highway 70 (DU [1]); Harkers Island (WLE 1142-1148); 24 mi. SW Morehead City,. 8 mi. from Swansboro (DU [1]); Shackleford Banks [an island] (WLE 778); South end of Shackleford Banks (UIMNH 5283). Sour:~ CAROLINA: Berkeley County-Alvih (CM 21791* 2.1 mi. N, 0.8 mi. E dainhoy (UMMZ 109235); 7 mi. W Moncks Comer (CM 25190-25191); 8 mi. W Moncks Corner (CM 28892-28898). Charleston County-Mt. Pleasant (CNHM 4076). Horry County-St. Patk [presumably Myrtle Beach State Park] (UMMZ 94166). TEXAS: Bosque County-Clifton (CM. 8987). [Locality data not acceptable; see text.] LITERATURE RECORDS. Specimens were not available from the followihg counties, indicated on the distribution .map (fig. 2) by unshaded symbols. FLORIDA: Sarasota County (Allen, 1989). MississIppI: Forest Countv (Cli' burn, 1959). NORTH CAROLINA. B,unswick County-near Shallotte (White, 1960); New Hanouer County (Funderberg, 1958), 1967 MYERS: THE PINE WOODS SNAKE 57 ECOLOGY The pine woods snake is too secretive and difficult to find to be studied easily in the field, but the study of specimens, habitats, and distribution provides information of ecological import, and observa- tions on a dozen or so individuals in captivity give additional chies to the species' nature. HABITATS PINE FLATWOODS. Several types of flat pine forests comprise the high flatwoods of the upper coastal plain and the low Ratwoods of the lower coastal plain. The pine flatwoods are vegetational associ- ations distinct from the pine or pine-oak forests of the hills and ridges, nor are they to be confused with the New Jersey pine barrens, which are of different origin and greater antiquity (uide Harshberger, 1916). A generalized profile of coastal plain pine forests is given by Wahlen- berg (1946, fig. 14). Figure 8. Low pine flatwoods near Hatchet Creek, Alachua County, Florida. Robert McF 0.20 Alachua Co. 36:64 9 16 Marion Co. 37:68 16 27 Alachua-Marion Co. (combined) 87:68 25 48 4.764 <0,05> 0.025 All other counties 58:42 22 16 GEORGIA - 0 1 LOUISIANA 62:88 8 5 MISSISSIPPI 63:87 27 16 LOUISIANA- - MISSISSIPPI (combined) 63:87 35 21 8.500 <0.10> 0.05 NORTH CAROLINA 57:48 4 3 SOUTH CAROLINA 50:50 4 4 ALL STATES COMBINED 50:50 98 92 VARIATION All observed structural variations are treated in this section. Most variation in this and other species of 5nakes presumably has a genetic basis, but studies by Fox (1948) and Fox et al (1961) show that environment may have major morphological influence in some cases. Until such problems can be studied and discussed in more detail, it seems best to classify variations of snake species without regard to possible genetic or environmental causes. The following scheme seems useful. A) Intrapopulational variation 1) Ontogenetic variation 2) Sexual dimorphism 8) Uncorrelated variation B) Interpopulational variation 4) Geographic variation Overlap is frequent between classes 1 and 2. Class 8, uncorre- lated variation, is equivalent to the "individual variation" of most f967 MYERS: THE PINE WOODS SNAKE 69 authors, which seems to me a poor term because all variation relates ultimately to the individual. , Uncorrelated, variation is that which· is not basically related to age or sex, although differential mortality rates may sometimes superimpose an age or sex correlation. Interpopulational, or geographic variation can be variously sub- divided, but, as R. flavilata provides few good examples, this is not attempted here. The study of geographic variation is essentiblly the study of shifting frequencies of a character or attribute between populations. Whatever pattern such shifting involves, be it mosaic or smooth or stepped clines, it must be remembered that geographic variation is not something completely apart, but rather is compounded from any or all of the three classes of intrapopulational variation. The only. exception is the relatively rare intraspecific situation (absent in R. flavilata) where a geographically variable character shows no intrapopulational variation, its frequency being either 0 or 100. In the pine woads snake geographic variation is compounded mainly from uncorrelated variation, partly from sexual variation, and, so far as known, not at all from ontogenetit variation. Table 4 lists the variable characters determined for Rhadinaea Bat)ilata and the classes of variation to which they are assigned. TABLE 4. CLASSIFICATION OF OBSERV-ED.VARIATION IN Rhadinaea flavilata. Type of variation Character studied~ Ontogenetic Sexual Uncorrelated Geographic Color brightness and pattern X Anal ridges X X Sexual segments of kidney X X Tail length X X Total and body lengths X Anal sacs X Tail breakage X Number of subcaudals X Number of ventrals X Color pattern X Labials X X X. K X X Temporals X Dor5al scales On neck X Labial and ventral color X Loreals 2/ X Oculars X Divided or half-ventrals X Fused subcaudals X Number of teeth X 70 BULLETIN,FLORIDA STATE MUSEUM Vol. 11 Characters that vary in. more than one way are so indicated in the text. In the following accounts, individuals with total length% (pre- served)' over 250 mIn are considered adult, and those under 200 mm juvenile, as explained in the discussion of anal. ridges. ONTOGENETIC VARIATION COLOR AND PATTERN. Most snakes probably undergo Some onto- genetic changes in color brightness (= value or brilliance) and satu- ration. Young.R., flauilata seem somewhat brighter than adults. Mal- nate (1989) mentions that some individuals have an indistinct light yellow patch on each side of the neck. The prominence of, these occipital spots seems determined partly by age, as they Were. conspic- uous on three juveniles from Okeechobee County, Florida and on one juvenile from Berkely County, South Carolina, but absent or faint on other juveniles and adults. ANAL RIDGES. These structures are discussed more fully on pages 72-78. Anal ridges are absent on most juveniles, but present on some adult females and most adult males (fig. 9). KIDNEys. The kidney in' juvenile males is similar to that in fe- "males lacking noticeably enlarged sexual segments of the urinary tubules. For more details see below. TAIL LENGTH. This is also a sexually dimorphic feature (fig. 8); tails average proportionally shorter in juveniles of both sexes than in adults. Nine juvenile males have a mean tail/total length ratio of 30.0 per cent (range 28.7-30.7), whereas 47 adult males have a mean of 82.6 per cent (range 80.2-35.9). Corresponding percentages for nine juvenile females and 42 adult females are 28.6 (27.0-80.6), ahd 29.8 (27.5-32.0), respectively. SEXUAL DIMORPHISM Aside from differences in the reproductive organs, the sexes of Rhadinaea fauilata were found to vary in the following ways. KIDNEYS. The adult male kidney is relatively larger than that of the female, and is white with a convoluted surface caused by en- larged segments ("sexual segments") of the urinary tubules. The juvenile and female kidneys are smoother in texture and darker in color. Kidney texture in Rhadinaea #auilata is similar in general ap- pearance to that in Diadophis punctatus (see Myers, 1965). The func- tion of this dimorphism in the kidneys of some snakes and lizards is unknown. ANAL SACS. -The anal sacs (also called musk or scent glands) of snakes are paired structures lying in the base of the tail and opening 1967 MYERS: THE PINE WOODS SNAKE 71 into the cloaca. In the males the sacs lie dorsad from the hemipenes, and for this reason occupy a lesser circumference than in females. The function of these structures is not known with certainty, but as the contents are discharged when R. /lavilata and most other snakes are handled, they may have some protective value. The sacs are well developed in juveniles. The length of the anal sacs was determined in only 26 specimens, but the data suggest that the sacs tend to extend more caudad in males .than in females (table 5). This is perhaps a compensation for the smaller diameter of the sacs in males. TABLES. LENGTH OF THE LEFT ANAL SAC IN Rhadinaea #auilata. Number of subcaudals spanned Sex 9 10 11 12 18 14 .No. Mean Males 0 1- 7 4 8 1 16 11.8 Females 28221 0 10 10.7 Both sexes 24964126 11.3 SIZE AND PROPORTIONS. Females attain greater total length than do males (fig. 7). Th6 seven largest specimens (365-887 mm total length) are females. Assuming individuals over 250 mm total length to be adults, 46 adult mdles with cogiplete tails average 801.88 + 4.15 mm in total length, and 45 adult females 825.40 + 5.20 min. These differences are due to the greater body (snout-vent) length attained by females (fig. 9), as males usually have longer tails than do females of equivalent length (fig. 8). The largest male (St. Johns County, Florida) is 868 mm in total length and has a tail length of 125 mm; 62 MALES 6- 4- ...1/1.1 11.11.111111.. FR EQ UE NC Y 140 160 180 200 220 240 260 280 300 320 340 360 380 400 , IIIII.IIII. 6- . 60 FEMALES TOTAL LENGTH (mm.) Figure 7. Observed frequency of total lengths in Rhadina€a #at>ilata. In- dividuals with incomplete tails are exclbded. 72 BULLETIN FLORIDA STATE MUSEUM Vol 11 in contrast, the largest female (Alachua County, Florida) is 887 mm in total length but has a tail measurement of only 114 mm. Mean proportional tail to total length for 60 males is 82.0 per cent (range 28.7-85.9), and for 57 females is 29.5 per cent (range 27.0-82.0). The presence of the hemipenes and associated muscles is un- doubtedly the basis for longer tails in male snakes generally, while the value of additional space for egg or embryo development is logi- cally the selective agent responsible for larger body size in the fe- males. Because of the hemipenes, the base of the tail in male R. #avilata is noticeably wider than in females. . 120 MALES C . ..0 110 FEMALES O ..%0& I. , 0%% loo ,/%0099 TA IL LE N G TH (m m .) ,O 1 60 . 4. 00 . '0 :a 5 JOI //0 .' , flo 50 .go 40 ~ 30 20 10 0 0 40 00 120 160 200 240 280 320 360 TOTAL LENGTH (mm.) Figure 8 . Tail length to total length relationships in Rhadinaea #auitata. Individuals with incomplete thils are excluded. ANAL RIDGES. This term was coined by Blanchard (1981) for the keel-like. ridges on dorsal scales in the anal region of certain other- wise smooth-scaled snakes, including Rhadinaea flauilata. In snakes of the genera Carphophis and Diadophis Blanchard found such struc- tures characteristic of males over certain lengths, and seldom present in females and yoong males. Blanchard and other workers have logically assumed that the ridges indicate the approximate size at which sexual maturity is reached. This belief has been objectively documented for Diadophis· punctatus in Florida (Myers, 1965), but the function of tha structures remains uncertain. 1967 MYERS: THE PINE WOODS SNAKE 73 In Rhadinaea #avilata, anal ridges are present on all male speci- mens over 164 mm snout-vent length (fig. 9) or 248 mm total length. Therefore all specimens over 250 mm total length are arbitrarily considered to be adults, and all specimens under 200 mm to ba juve- niles. Malnate (1989) mentioned two males of 275 mm and 805 mm total length that lacked anal ridges. Possibly Malnate overlooked poorly developed ridges on these specimens bdcause.of poor magni- ficati6n or lighting. Nevertheless occasional large Diadophis males lack anal ridges (Blanchard, 1'981, 1942; Myers, 1965), and so might some large Rhadinaea males. Malnate (1939) also said that anal ridges were not present on any females examined, but they appear on 84 of 87 females I examined (fig. 9). 18 - 82 MALES 16 - 14 - I2 to- 8 -i- „41.1~| ~" .. FR EQ UE NC Y 100 120 140 160 180 200 220 240 260 280 4- 2- Im.0- lilli. 1 41141112- 4- e- 87 FEMALES SNOUT-VENT LENGTH (mm.) Figure 9. Distribution of anal ridges by size and sex in Rbdinaea #auttata Specimens plotted above the horizontal lines have anal ridges, those below lack them. The degree of development of anal ridges varies considerably in favilata, but females tend to Kave them less strongly developed than males. The most conspicuous anal ridges are found in the larger males, but some of'the largest specimens have them poorly developed. 74 BULLETIN FLORIDA STATE MUSEUM Vol. 11 TAIL BREAKAGE. Incomplete tails are present in 27 males and 28 females, or 29.4 per cent of the total sample. There is no sig- nignificant sexual dimorphism in the piace of breakage, although the break is closer to the· anal plate in some females than in males. The hemipenial muscles normally extend to the 21st subcaudal (to the 24th in an Alabama specimen), and only one male, with 18 pairs of caudals, has less than this number. Seven females, however, have 20 or fewer pairs of caudals. Females with broken tails have 12-70 pairs of subcaudals remaining, with a mean of 38.75 + 8.30; males have 18-69 pairs with a mean of 42.87 + 2.59. VENTRALS. The number of ventral plates varies both geographi- cally and sexually. For the entire sample the number of ventrals in males ranges from 112 to 184, and for females 118 to 189. For given populations the overlap is not nearly so great, and the means are well separated (fig. 11); females usually average four or Bve more ventrals than males (table 6). The greater number of ventrals in females is correlated with larger body size. TABLE 6. SEXUAL DIMORPHISM OF Rhadinaea /lauilata- IN NUMBER 6F VENTRAL PLATES. Males Differences Females Series N M SE DM CD N M SE North and South Carolina 8 119.88 - 8.40 6.77 7 128.28 Alabama, Louisiana, and Mississippi 87 128.80 .31 4.70 8.74 26 128.00 .58 Alachua and Marion counties, F16rida 25 125.82 .28 4.70 3.68 48 180.02 ·.87 Entire range 91 124.20 .28 5.50 4.88 98 129.70 .80 N = numbet; M = mean; SE = standard error Of mean; DM difference between means; CD = coefficient of divergence (in per cent). SUBCAUDALS. Males have longer tails (Bg. 8) and hence more subcaudal plates than females. For the entire sample males have 68-88 pairs of subcaudals, females 59-75 pairs; there is little or no geographic variation (fig. 12). In given populations males average five or six more subcaudals than females (table 7). As expressed by the coefficients of divergence (tablds 6, 7), sexual dimorphism is more marked in subcaudals than in ventrals. 1967 MYERS : THE PINE WOODS SNAKE 75 TABLE 7, SEXUAL DIMORPHISM OF Rhadinaea flavilata IN NUMBER OF ·SUBCAUDAL PLATES. Males Differences Females Series N M SE DM CD N M SE North and South Carolina 8 71.62 - 4.62 6.67 4 67.00 - Alabama, Louisiana, and Mississippi 22 72.64 .55 5.95 8.54 18 66.69 .99 Alachua and Marion , counties, Florida 18 78.61 .51 5.58 7.88 88 68.08 .51 Entire range 60 72.87 .81 5.24 7.46 60 67.68 .40 N = number; M = mean; SE = standard error of mean; DM = difference between means; CD = coefficient of divergence (in per cent). UNCORRELATED VARIATION Several uncorrelated variations of Rhadinaea #avilata seem to have distributional significance and are discu5sed under geographic variation. These are ,certain aspects of color pattern and of labial, temporal, and dorsal scale scutellation. It seems likely that addi- tional data would give geographical significance to many of the re- maining uncorrelated variations. Only in the case of an aberration of the last ventral plate are there sufEcient data to indicate an un- ~ correlated variation that may not undergo interpopulational shifts in frequency. C6L0R AND PATIERN. Color pattern variation is discussed under geographic variation. Also of pos5ible geographic significance is the coloration of labials and venter, but too few color descriptions 6f living individuals are available to document this. Cope (1871) says that the labials and venter,of the type specimen (North Carolina) were white; this was in life, for Cope received the type alive (Coues and Yarrow, 1878), kept it alive for several months (Cope, 1900), and published his description of the specimen less than two months after receiving it. Ditmars (1907) states without elaboration that the upper lip is "bright-yellow." In a combined description of two living speci- mens from Florida and Mississippi Brown (1901) states that the ventrals and labials were light yellow. Malnate (1939) gives the fol- lowing description, probably from S6uth Carolina Specimens: -(color nomenclature from Ridgway) ... Ventral surface pale martius or marguerite yellow, fading to whitish on the chin and throat.... Labials light maize yellow or sulphur yellow." Ventral coloration 76 BULLETIN FLORIDA STATE MUSEUM· Vol. 11 of living specimens from northern Florida ranges from white to chartreuse (yellow-green), and the labials are white, sometimes with a trace of the brown head coloring as well as the usual black spot- ting. A specimen in the Duke University collections has bright yellow labials-from a slip of yellow paper that faded in the jar of preservative. LoREALS. A male from Mississippi lacks a loreal on the left and has the right one very much reduced. OCULARS. A female from Hancock County, Mississippi has the right preocular fused with the supraocular. Netting (1986) notes that a supposed Texas female has the same condition on the left side Of the head. Pseudo-oculars rarely appear anterior to the lower edge of the eye, but these arise from labial divisions as discussed under geographic variation. LABIALS AND TEMPORALS. All variation in these plates is for con- venience considered under geographic variation, although it is pos- sible that a few rare conditions of the supralabials and temporals might be iminfluenced by geography. DORSAL SCAT.ER. See under geographic variation. VEN·rRALS. The last ventral is abnormal in 11 per cent of all specimens examined. In most cases (17 of 21) only a half-ventral is present; in the others two h«lf-ventrals lie side by side, giving the appearance of a full plate divided sagittally. The percentage of ~ aberrant specimens is nearly the same in the Florida and western samples, but higher in a series from the Carolinas (table 8). The ~ Carolinas sample was tested for statistical difference from tha com- bined Florida and western sample. The results (x? = 0.509; P < 0.50> 0.80) suggest that this anomaly is not characteristic of any one part of the country; also no evidence suggests that local popula- TABLE 8. ABERRANCY® OF THE LAST VENTRAL PLATE IN Rhadinaea #auitata. Series No. aberrant Total sample % aberrant FLORIDA 12 110 10.9 WESTERN 6 64 9.8 (Alabama, Louisiana, and Missishippi) NORTHERN 3 15 20.0 (North and South Carolina) Total 21 189 11.1 *"Divided" or "half-ventral," probably indicative of vertebra duplication. 1967 MYERS: THE PINE WOODS SNAKE 77 tions vary significantly from one another in this respect. There is no sexual dimorphism, as 11 variants are males and 10 females. Rarely do aberrations occur other than in the -last ventral. Half- ventrals are present on the anterior part of the body in two males (Mississippi and Florida). A female from.Flodda has several anterior ventrals "divided." King (1959) demonstrated that half-ventrals correspond to verte- bral duplication (duplicate accessory processes and rib) on one side of the body. Consequently a ventral scute that seems to be divided at the midline probably represents two half-ventrals corresponding to vertebral duplication on both sides. The half-v.entral is illustrated diagrammatically by King (1959, fig. 1-E) and Peters (1956, fig. 8-1; 1960, fig. 2-1); the "divided" type of ventral is illustrated by Peters (1956, fig. 8-2;' 1960, fig. 2-2). The conditions observed in Rhadinaea Bat:ilata are similar to these illustrations; the other types of anoma- lous ventrals Egured by King and by Peters were not observed. King (1959) mentions that half-ventrals were found in Rhadinaea; this statement was based on R. #auilata (King, verbal communica- tion), although I can state that it also occurs in other members of the genus. SUBCAUDALS. Two males-and two females have one or more sin- gle subcaudal plates. A specimdn from Okeechobee County, Florida has the first caudal undivided, and one from Alachua County, Florida has the last undivided. One oftwo snakes collected 5 miles west of Slidell, St. Tammany Parish, Louisiana has the first two plates un- divided, the other has the second, third, and fourth plates entire: DENATION. No attempt was made to ascertain possible geo- graphic variation in the number of teeth, which does occur in some TABLE 9. VARIATION IN THE NUMBER OF TEETH IN FLORIDA SPECIMENS of Rhadinaea #auilata. No. Number of teeth Bone bones 11 12 18 14 15 16 17 18 19 20 21 22 28 MAXILLARY 16 14° 2 (anterior to diastema) PALATINE 18 80 3 2 PTERYGOID 12 12152 DENTARY 12 5* 6 1 *Location of counts for P6lk County specimen (1 maxilla, 1 palatine, and 1 dentary were available). All other specimens from Alachua County. 78 BULLETIN FLORIDA STATE MUSEUM Vol. 11 snakes (e.g. Coluber constrictor, Auffenberg, 1955). A small series of skeletons from Alachua County, Florida and one from Polk County, Florida indicate the extent of variation in one geographic area (ta- ble 9). Malnate (1989) apparently attempted to count the teeth of R. #avilata in- situ, and perhaps for this reason obtained lower eounts than given in table 9. GEOGRAPHIC VARIATION COLOR PATFERN. The most conspicuous geographic variation in R. #aviata is the dark pigmentation on the labials ·(fig. 10), which increases from south to north. In Florida the labials are usually lin- marked except for a few plates anterior to the eyes, although some Florida individuals show scattered spots of pigment on all the supra- labials and most of the infralabials. In specimens from Alabama, Louisiana, and Mississippi most of the labials are usually spotted and generally more pigmented than in Florida individuals. Some South Carolina specimens have still more pigment, and in North Carolina all the labials and even the genials may be profusely pig- mented. Some snakes from the'Carolinas resemble southern speci- mens in labial pigmentation, but no southern individual (Florida, Georgia; Alabama, Mississippi, or Louisiana) remotely approaches the extreme peppered condition shown in fig. 10. The dorsal stripe also varies geographically. When present this stripe consists of a band of diffused pigment on the vertebral scale row, the pigment normally being restricted to the distal ends of the vertebral scales. The stripe is absent in about 45 percent of the Florida sample and usually only faintly indicated in the rest. Only about 10 percent of the Alabama,. Louisiana; and Mississippi speci- mens lack a middorsal stripe, and it is usually more distinct than on Florida specimens. One of eight specimens from South Carolina does not have a vertebral stripe, but 6n most it is distinct. The stripe is best deyeloped in the North Carolina sample (seven specimens); vertebral scale row pigmentation is more or less continuous on four specimens, a rare condition anywhere else in the range. I fail to detect geographic variation in the lateral stripe, which is rarely ab- sent but is often weak. Neill (1968: 205) mistakenly attributes pres- ence or absence of striping in favilata to polychromatism, but varia- tion in this character is cclntinuous. VENTRALS AND SUBCAUDALS. The number of ventrals decreases slightly from south to north (fig. 11). Geogtaphic variation in the 1967 MYERS: THE PINE WOODS SNAKE 79 number of subcaudal plates is either nonexistant or too slight to demonstrate in the available sample (fig. 12.) NORTH .. CAROLINA 4<; A. jkj. 4.4>«':..:.. -- ., -- SOUTH -572-'--*< 4 3 3 7 CAROLINA FLORIDA : Figure 10. Geographic trends iii labial pigmentation of Rhadinaea flacilata. Top-DU specimen from 24 miles SW Morehead City, Carteret County, North Carolina; Center-CM 25190, Berkely County, South Carolina; Bottom-speci- men from Alachua County. Florida . Robert Al <:Farlane 80 BULLETIN FLORIDA STATE MUSEUM Vol. 11 NORTH CAROLINA 14 ' '9 I , 1 SOUTH CAROUNA ' -4.-4 , Id' , ALABAMA, i i , , I , MISSISSIPPI, 8 ~' ' ' Fl/IlLOUISIANA ~ \ r*m ,;. NORTH FLORIDA \ I. r----I . . . CENTRAL FLORIDA •~ , 1 , 1 : , /9 SOUTH FLORIDA , 112 114 116 118 120 122 124 126 128 130 132 134 136 138 NUMBER OF VENTRALS Figure 11. Geographic and sexual variation in the number of ventral plates in Rhadinaea Kavilata. Horizontal lines represent the ranges of variation, and small triangles the means. A solid rectangle represents two standard errors on each side of the mean, this plus the open rectangle, one standard deviation on each side of the mean. Sample sizes: North Caroliria--4'males, 8 females; South Carolina-4 males, 4 females; Alabama, Mississippi, and Louisiana--87 males, 26 females; North Florida (Alachua and Marion counties>-25 males, 48 females; Central Florida (Brevard, Orange, Polk, and Seminole counties)-5 males, 5 females; South Florida (Glades, Okeechobee, and Palm Beach countids)-4 males, 5 females. 1967 MYERS: THE .PINE WOODS SNAKE 81 NORTH CAROLINA + SOUTH CAROLINA ' 1.J ALABAMA, MISSISSIPPI, a , .IZZI~IM'--1LOUISIANA . I 5--1//~//r--3 NORTH FLORIDA , id'19 SOUTH FLORIDA ' 60 62 64 66 68 70 72 74 76 78 80 82 NUMBER OF SUBCAUDALS Figure 12. Geographic and sexual variation in the number of subcaudal plates in Rhadincea #auilata. See- fig. 11 for interpretation. Sample sizes: North Carolifia-4.males, 1 female; South Carolin6-4 males, 3 females; Alabama, Mississippi, and Louisiana, 22 males, 18 females; North Florida (Alachua and Marion counties)-18 males , 38 females; South Florida (Glades, Okeechobee, and Palm Beach counties)-4 males, 8 females. 82 BULLETIN FLORIDA STATE MUSEUM Vol. 11 INFRALABIALS. The lower labials usually number 9/9, but there are occasionally 7, 8, or 10. Observed variation is listed in table 10; several formulae (e. g. 7/7) that may exist in nature were not observed. Sexual dimorphism appears not to be involved in deviations from the normal (13 males to 18 females). Deviations in upper and lower labial counts occur independently more often than not; only four specimens (from Florida) showed deviations in both the supra- and infralabial formulae. Reduction or increment of the infralabials occurs most d6mmonly in Florida (table 10); elsewhere such cases are known only from Alabama (1) and Mississippi (2). The Florida sample diRers signill- cantly even when compared only to the western p6pulation (7(2 - 13.896; P<0.001). A greater percentage of variant specimens occurs in the Alachua-Marion counties sample than all other Florida locali- TABLE 10. VARIATION IN THE NUMBER OF INFRALABIALS OF Rhadinaea #avilata.* Number of labials (left/right) Series 7/8 7/9 8/8 8/9 9/8 9/9 9/10 10/9 8/10 10/10 Total FLORIDA 117 5 5 70 4 4 1 1 99 Alachua Co. 8 8 16 1 1 24 Marion Co. 1 3. 2 25 2 8 1 1 88 Brevard Co. 2 3 ' 5 Duval Co. 1 1 Okeechobee Co. 115 7 Orange Co. 1 - 1 St. Johns Co. 81 4 Other counties (8) 14 ' 14 Unknown 1°* 4 5 GEORGIA 1 1 WESTERN (Alabama, 1 61 1 1" 64 Louisiana, and Mississippi) NORTHERN. (North 15 15 and South Carolina) UNKNOWN ("Texas") 1 1 Total 1 1 8 5 5 148 4 512 180 *A mutilated male with the formula 10/? (St. Johns Co., Fla.) was not in- eluded, nor were a few similar specimens with formulae 9/? or ?79. ' *Specimens mentioned by Malnate (1989, p. 868) that were not available for the present study. 1967 MYERS: THE PINE WOODS SNAKE 88 ties combined (table 10), but the difference is not statistically signifi- cant (*2 - 1.189; P <0.25> 0.20). Infralabidl variation is best explained by assuming the division or fusion of plates during embryonic development. The reduction to 8 plates involves the fusion of the eighth and ninth or the second and third plates; a reduction to 7 plates involves both of the above com- binations. An increase to 10 plates is the result of a division in one of the frst several plates (second to fourth, apparently) or in one of the last. Fox et al. (1961) report a significant correlation between low supra- and infralabial counts and low environmental tempera- tures during embryonic development in the snake Thamnophis ele- gans. SUPRALABIALS, The upper labials usually number 7/7, but oc- casional combinations of 7/8,8/7, and 8/8 appear with equal fre- quency' to one another. No apparent sexual dimorphism is involved, as six males and nine females were among the v-ariants studied. Deviations from the normal count may appear anywhere in the range, but seem most common in Florida (table 11) where 81 percent of the observed variation occurred. Statistidally the Florida sample does not differ significantly from the rest of the range 6(2 - 8.898; TABLE 11. VARIATION IN THE NUMBER OF SUPRALABIALS OF Rhadinaea flavilata. Number of labials (left/right) Series 7/7 1 /8 8/7 8/8 Total FLORIDA 91 4 5 4 104 Alachua Co. 19 1 3 2 25 Marion Co. 38 1 1 1 41 Polk Co. 8 1 4 Seminole Co. 1 1 Unknown 8 1 1 5 Other 28 28 GEORGIA 1 1 ALABAMA 7 1* 8 LOUISIANA 18 18 MISSISSIPPI 42 1 43 NORTH CAROLINA 7 7 SOUTH CAROLINA 7 1 8 UNKNOWN ("Texas") 1 1 Total 169 5 6 5 185 *A specimen mentioned by Malnate (1939, p. 868) that was not available during the present study. 84 BULLETIN FLORIDA STATE MUSEUM Vol. 11 P < 0.10 > 0.05), although the probability obtained is not much above the level (0.05) here accorded significance. While the possibility of chance collecting cannot be ruled out, I suspect that supralabial num- bers do vary more frequently iIi Florida. Most supralabial variants in Florida occurred in the large Alachua-Marion counties sample (table 11), but no significance can be attached to this in comparison with the rest of the state (2 - 0.028% P < 0.90 > 0.80). Several anomalous supralabial conditions were observed. A fe- male from Alachua County, Florida and another from adjoining Marion County each have.a small extra plate (pseudopteocular) added at the anterodorsal corner of the third labial on the left. In a Geor« gia female a small plate present above the third labial on both sides of the head gives an apparent preocular formula of 2/2. A female from Putnam County, Florida has the first labial on the left excessive- ly elongated along the horizontal axis. Some specimens have an extra plate between the fifth labial (sixth labial in a specimen with 8 supralabials) and the first temporal. In the sample from Alabama, Louisiana, and Mississippi, 19 have such a plate on 6ne or both sides of the head and 48 do not. The anomaly is absent in the small series from Georgia and the Carolinas but present in four of I04 specimens from Florida. By comparing the Florida sample with the western one it is seen that geographic dif- ferences are probably not due to chance alone (x2 - 21.184; P < 0.001). This is the only head plate anomaly found to be character- istic of the western population. No sexual dimorphism seems to be involved, for the ratio of 15 aberrant males to 8 females can be explained on the basis of an unbalanced sex ratio in the western sample (table 8). As with the infralabials, supralabial variation is best explained by assuming the division or fusion. of plates during embryonic de- velopment. The addition of extra plates at the dorsal borders of the third and fifth labials is caused by a division of those plates, as,is evident from their reduced area and altered shape. Increase from a normal number of seven to one of eight seems to be the result of a vertical split of the second or third labial, most often the latter. The single instance where the first labial was elongate probably was caused by a fusion of the first two plates and a division of the third. TEMPORALS. The normal temporal fbrmula *1+2, but these plates are frequently altered by division, fusion, enlargement, or reduction; 88 percent of all specimens examined have one or more aberrant temporals, but usually on only one side of the head. The following kinds of variation were observed. 1967 MYERS: THE PINE WOODS SNAKE 85 1) Vertical division of one or rarely both plates in row 2. This 12 2 gives the formula 1 + -, 1 + -, or rarely 1 + -. (52 specimens: 21 2 Florida 86, Louisiana 5, Mississippi 9, North Carolina 1, South Carolina 1). 2) Reduction in size of one or both plates in row 2, usually with a noticeable enlargement of an adjoining post-temporal (7 specimens: Florida 6, Mississippi 1). 8) Fusion of the top plate in row 2 with the adjoining post- temporal, thereby forming an elongated plate. (7 specimens: Florida 6, South Carolina 1). 4) Enlargement of the bottom plate in row 2 and a correspond- ing reduction of the top plate. (2 specimens: Florida .1, Missis- sippi 1). 5) Fusion of both plates in row 2 on the left side of the head. (1 specimen: Mississippi). 6) Fusion of the single plate ·in row 1 with the top plate in row 2 on the right side of the head. This was the only observed instance in which the frst temp6ral was affected. (1 specimen: Mississippi). The various temporal conditions are so mixed in some samples that they are best considered all together. Only 11 specimens had the same temporal aberration on both sides of the head; only four specimens had more than one kind of temporal defect; aberrations occurred as frequently on one side of the head as on the other. As they occurred in 84 males and 28 females, no sexual dimorphism is evident. Although the observed sex ratio in the Alachua-Marion counties, Florida population was 37 per cent males to 63 per cent females, temporal variants occurred in nearly a 1:1 sex ratio (16 males: 17 females); however, no statistical significance can be at- tached to these differences (xz - 2.360; P < 0.20 > 0.10). Temporal plate anomalies may be slightly more common in Flor- ida than elsewhere (table 12), but differences based on the available sample are not really signiftcant (xz = 8.442; P < 0.10.> 0.05). The issue is further confused by sample bias: The number of anomalous specimens in the Alachua-Marion counties series differs significantly from the rest of the Florida specimens combined (xe - 7.085; P < 0.01 > 0.005). Specimens with aberrant tempbrals are also 86 BULLETIN FLORIDA STATE MUSEUM Vol. 11 known from Brevard, Duval, and Volusia counties on the east Florida coast; 6 of 12 individuals from these counties have anomalous tem- porals compared with 25 normal specimens from 11 other Florida counties (table 12). This distribution is not random (7(2 - 10.444; P < 0.005 > 0.001), and the am6unt of temporal plate aberrancy in Florida differs significantly from one local population to the next. TABLE 12. GEOGRAPHIC DISTRIBUTION AMID FREQUENCY OF SPECIMENS OF Rhadinaea flavilata HAVING ANOMALOUS TEMPORAL PLATES. Per cent Series Anomalous Normal Total anomalous FLORIDA 42 67 109 88.5 Alachua Co. 12 13 25 48.0 Marion Co. 21 21 42 50.0 Brevard Co. 8 2 5 60.0 Duval Co. 1 8 4 25.0 Volusia Co. 2 1 3 66.7 Counties unknown 8 2 5 60.0 Other counties (11) 0 25 25 0.0 GEORGIA 0 1 1 0.0 WESTERN 17 46 68 .27.0 Alabama 0 7 7 0.0 Louisiana 5 8 18 88.5 Mississippi 12 31 48 28,0 NORTHERN 8 12 15 20.0 North Carolina 1 6 7 14.8 South Carolina 2 6 8 25.0 UNKNOWN ("Texas") 0 1 1 0,0 COMBINED 62 127 189 82:8 DORSAL SCALES. The dorsal scale row formula is normally 18 + 17 + 17 + 17, but Occasionally the frst count (immediately behind the head) is reduced to 17 or very rarely increased to 19. Deviations were observed in Florida and Mississippi specimens (table 18), A significantly higher number of deviations occurred in Florida than in the western sample (9 - 5.822; P < 0.02 > 0.01). The Alachua- Marion series showed no significant difference from other Florida localities combined 6(2 = 0.075; P < 0.80 > 0.75). 1967 MYERS: THE PINE WOODS SNAKE 87 TABLE 13. GEOGRAPHIC VARIATION IN NUMBERS OF DORSAL SCALES ON THE NECK of Rhadinaea flavilata. Per cent- deviationNumber of scales on neck from Sefies 18 17 19 Total normal (18) FLORIDA , 77 23 100 28.0 Alachua Co. 17 8 25 82.0 Marion Co. 82 8 40 20.0 Brevard Co. 2 2 4 Duval Co. 1 2 3 Clades Co. 1 1 Levy Co. 1 1 Taylor Co. 1 1 Other counties (9) 22 22 Counties unknown 8 8 GEORGIA 1 1 0.0 WESTERN 55 30 1® 59 6.8 NORTHERN 14 14 0.0 UNKNOWN ("Texas") 1 1 0.0 Total 148 26 1 175 18.2 *From Mississippi. SUMMARY. Interpopulational variation in Rhadinaea #avilata is compounded from intrapopulational uncorrelated variation, except for the geographic changes in numbers of ventral plates, which first vary according to'sex. Clinal variation is seen in features of color pattern and ventral numbers, pigmentation increasing from south to north and Vintrals decreasing. Other geographically variable characters vary regionally and/or microgeographically: Variation in labial num- bers, ·presence of anomalous supralabials, and deviation from an an- terior count of 17 dorsal scale rows, all occur'in highest frequencies in Florida. An extra plate between the fifth supralabial and.anterior temporal occurs most frequently in the western portion of'the range in Alabama, Mississippi and Louisiana. Frequency of occurrence of anomalous temporal plates seems to vary from population to population on a distinctly microgeographic basis. Except for clinal changes in color pattern, geographic variation in the pine woods shake is not striking. The present study conErms Malnate's (1989) view that no subspecies are recognizable. 88 BULLETIN FLORIDA STATE MUSEUM Vol. 11 AFFINITIES Rhadinaea #avilata is a relatively generalized member of its genus, as shown by the following characters: 1) Dorsal body scales in 17 rows and not reduced posteriorly. 2) Two enlarged and ungrooved rear maxillary teeth, set off by a diastema. 8) Hemipenis clearly single, capitate, and calyculate. 4) Oculars 1 + 2, temporals 1 + 2. The species is unusual in normally having only 7 supralabials, and is unique in its strong tendency toward a uniform golden-brown dorsal color (most species are conspicuously striped) and,in its north- ern distribution, widely isolated frorri other species of the genus. Concerning a more specific relationship for R. flauilata within the genus, I concur with an idea advanced independently by Malnate and by Bailey 27 years ago. Malnate (1939) suspected that the dorsal striping on some Rhadinaea ~avilata indicated relati6nship with R. laureata (Gunther) of the highlands of western Mexico. Bailey (1940, p. 16) expresses the same conclusion as follows (remarks in brackets mine): "A third species, laureata, offers a clue to the relationships 'of the isolated favilata of the southeastern United States. In' these species. the upper labials are ·normally 7, a reduction from the usual 8 in other forms. Recent examination of a living specimen of #avilata revealed further similarities to laureata which were unsuspected from preserved material. The head is uniform above for 2 or 8 scales on the neck, as in kiureata, and a very faint trace of striped body pattern is present in flavilata for which the same scale-Fow relationship is found as in laureata. [In #auilata the middorsal stripe, when present, occupies the vertebraI scale row, whereas in taureata this stripe includes the paravertebral .rows. The weak lateral stripes in #auilata and bureata bear exactly the same scale-row re- lationship.] The stripes of laureata show unmistakable signs of diffusion (see Pl. I) which is simply further developed in flavilata, resulting in an almost uni- · color pattern [except that body stfipes are fairly well developed in North Caro- lina]. In keeping with this general pattern weakness, the white collar of laureata is lacking and the light temporal stripes are less distinct in #auilata," An additional characteristic shared by flauilata and laureata is the color pattern of the chin and lips. R. laureata throughout its range has the labials and genials profusely peppered with dark pig- ment as in some R. flauilata from North Carolina (fig. 10). The fact that this pattern occasionally occurs in other sections of the «genus (e.g. in R. pachyura fuluiceps) does not lessen its significance as an indicator of relati6nship between #avitata and laureata. EVOLUTION The following hypothesis presumes that Rhadinaea ~auilata or an ancestral predecessor reached -the southeastern United States 1967 MYERS: THE PINE WOODS SNAKE 89 via a coastal route from the west. This assumption is based on the present coastal distribution of favilata and its apparent relationship to Mexican Rhadinaea (esDecially laureata). - A number of other ani- mals seem to haye had similar origins (Blair, 1958; Coin, 1958).. Fossil vertebrae assigned to R. #avilata are known from Pleisto- cene localities in Florida as .follows: Reddick I B, Marion County (Auffenberg, 1968% Williston, Levy County (Holman, 1959); and Saber-tooth Cave, Citrus County (Holman, 1958). The Reddick and Williston localities seemingly represent the third (Illinoian) glacial stage (Holman, 1959); the Saber-tooth Cave deposit is assigned to th6 last (fourth or Wisconsin) glacial itage (Holman, 1958). The present-day habitat preferences of R. flauilata may have been devel- oped early, f6r Holman (1958, 1959) interprets the Williston and Saber-tooth Cave faunas as indicative of pine flatwood regions with associated ponds. Thus #auilata possibly migrated into the south- eastern states during the second glacial period, when the Gulf Coast probably afforded a low, dan* corridor from the west and south- west. An earlier entry is possible, even into Florida, where land may have persisted since Miocene time (Vernon, 1951 and Vernon in Coin, 1958). Auffenberg' (1968) shows Recent Florida snake genera present in the Pliocene and visualizes little change in distribution of snake genera since then. He further states that, "On the basis of data obtained fr6m the amphibians of the Florida Middle Pliocene . . . there is every reason to believe that it was between Lower Miocene and Middle Pliocene that Florida felt the effects of a west- ern herpetofaunal immigration. This is also suggested by the pres- ence of certain genera of snakes in the Middle Pliocene which are thought to have their ancestral home in southwestern North America (Crotalus, Micrurus, etc .). He also suggests an eastward coastal immigration "during the time represented by Stratum 2 at Ver6, which is subsequent to the entry of Rhadinaea. Regardless of exact time of arrival, the east-west distribution of this species was perhaps reduced by climatic changes during glacia- tion, quite possibly in the third (Illinoian) glacial period. The third great glacier extended farther south than any of the others, and climatic changes are thought to have dfiven many warmth-requiring animals into separate refuges in Mexico and Florida. At that time\ #auilata was presumably isolated in Florida where it has survived to the present. During the late Pleistocene the Florida land mass was occupied by such now-extinct animals as lions, saber-tooth tigers, camels, horses, mammoths, mastodons, ground sloths, giant armadillos, dire wolves, and peccaries (Simpson, 1929). But where- 90 BULLETIN FLORIDA STATE MUSEUM Vol. 11 as the mammals (and birds) underwent considerable extinction, the snake fauna of Florida differs little between late Pleistocene and modern times (Auffenberg, 1968). With the retreat of the Pleistocene ice sheets and subsequent climatic changes, R. jiavilata spread out of Florida and followed an expanding habitat westward and northward. The hypothesis of a relatively recent dispersal of a homogeneous stock into relatively uniform habitat (pine flatwoods) explains nicely the absence of any marked geographic variation in the species. Resemblance in color pattern of North Carolina /lauilata to Mexican laureata suggests that the pine woods snake retains primitive characteristics at. the northern limits of its range. Because a northward shift in climate since glaciation is very probable, possibly , the ~ tendency toward loss of color pattern and toward increased variability in scutellation in the southern populations represent adaptation to a changing en- vironment. Much variability in scutellation is anomalous, indicating that southern populations are in some ways less adjusted than north- them ones, which exist in a climate cloker to that for which #auilata has been longest adapted.° Thus it seems that the southern po4ulations of Rhadinaea #avilata are evolving faster than the northern ones, presumably in response to a warming postglacial climate. Though laboratory studies have shown tempetature increase'to be an effective mutagen in Drosophila, it is not wise to demand a genetic basis for all cases of intraspecific variation in snakes. Scutellation variations similar to some in Rhadi- naea have been induced experimentally in Thamnophis elegans by lowering environmental temperatures during embryonic growth (Fox et al., 1961). Possibly R. #avilata is also showing phenotypic suscept- ibility to environmental change, but the implications of the Thamno- phis study are not easily applied. The two main reasons for this diffi- culty are: 1) In the absence of adequate breeding data, nearly im- possible to obtain for many or most snakes, one can seldom diStin- guish between heritable and non-heritable variation; comparing en- vironmentally-induced variation in Thamnophis with evolutionary trends in this and other genera, it seems certain that a single kind * Telford (1966) presents an ekcellent analysis of. variation in southeastern snakes of the genus Tantilla, in which the peninsular Florida species, T. oolitica and relicta. "clearly show a trend toward more aberrant individuals than does T. coronata" which occurs n6rth of the peninsula. The fact that the peninsular Tantitta are comprised 6f rather isolated demes does not seem sufficient to ex- plain this particular case of regional variation; one would expect a low frequency of aberrations in some demes to balance out a high frequency in others, unless a non:random factor were involved. I suggest that these differences in Tantind may be related ta the phenomenon discussed here in Rhadinaea #avilata 1967 MYERS: THE PINE WOODS SNAKE 91 of variation is sometimes strictly phenotypic and 6thertimes genetic, even possibly within single species. 2) We lack comparative data on the potential effects of temperature on the deyeloping embryo. Can, for example, the same sort of phenotype be induced by lowered tem- peratures in some species of snakes and by increased temperatures in others, or can temperature shifts to either, side of the physiological optimum induce similar results in the same species? These are a few questions that come to mind when assuming environmental causes of variation in the scutellation of R. #aoilata. Ventral plate numbers in the pine woods snake are lower in the north in accordance with the findings in Thamnophis, but .what about scales and plates that are most variable in the warmer parts of the range? If some of this variability is indeed nonheritable and if it is induced by temperatures below physiological optimum (as in Thamnophis), we must assume that, because of genetic adaptation to overall warming trends, the southern populations have become more responsivethan northern ones to local peri6ds of cool weather during ontogeny. Such speculation as above is warranted by the extreme impor- tance of the work of Fox et at., but should n6t be extended to oyer- shadow suggestive, albeit indirect, evidenc6 that some interpopula- tional variation in Bavilata is attributable to non-selective genetic mechanisms. Differences between local populations in frequendy of anomalous temporal plates, for example, are easily explained by genetic drift or the founder principle (see Mayr, 1968 and- Ford, 1964, for recent discussions of such mechanisms). Regional differ- ences may also have arisen by n60-selective g6netic mechanisms under certain circumstances; fragmentation of the pine flatwoods, as in northwestern Florida, provides suiRcient isolation to maintain regional differences developed from pi6neer stock. An extra plate between a supralabial and the first temporal occurs in 80.6 percent of the western population, but in only 8.8 percent of Florida snakes; that this is a genetic trait responsive to drift in pioneer populations is strongly suggested by the fact that of the four (of 104) Florida individuals having the extra plate, three are from an island popula- ti6n (Anastasia Island, St: Johns County) represented by only five specimens. In addition to mechanisms discussed above, natural selection probably accounts for some geographic variation. The basis for selection may be that R. #auilata has shifted habits in response to climatic change directly, or to some ecological situation for which higher temperatures increase- the .number of favorable. mutations. As inferred from study of variation and relationships, the principle 92 BULLETIN FLORIDA STATE MUSEUM Vol. 11. evolutionary trend is reduction-in color patterii, in supralabials, and perhaps in certain other head plates. Such traits are most developed in truly fossorial snakes and lead me to suggest that #auilata is be- coming more secretive in habits. Certainly tbe pine woods snake is not often found away from cover, in contrast to several species of tropical Rhadinaea with which I have had field experience. LITERATURE CITED Allen, E. Ross 1939. Habits Of Rhadinaea #auilata. Copeia, 1989, no. 8, p. 175. Allen, Morrow J. 1932. A survey of the amphibians and reptiles of Harrison, County, Mississippi. Amer. Mus. Novitates, no. 542, pp. 1-20. Amaral, Afranio do 1929. Estudos sobre ophidios Neotropicos, XVIII. Lista remissiva dos ophidios da regiao Neotropica. Mem. Inst. Butantan, vol. 4, pp. viii + , 129-271. Anderson, Paul K. 1954. Studies in the ecology of the narrow-mouthed toad, Microh!/la caro- Zinensis carolinensis. Tulane Studies in Zool,, vol. 2, no. 2, pA 15-46. American Society of Ichthyologists and Herpetologists 1956. Common names for North American amphibians and reptiles. Copeia, 1956, no. 3, pp 172-185. Auffenberg, Walter 1955. A reconsideration of the racer, Coluber constrictor, in eastern United States. Tulane Studies in Zool., vol. 2, no. 6, pp. 89-155. 1968. The fossil snakes of Florida. Ibid., vol. 10, no. 8, pp 181-216. Bailey, Joseph R. 1940. The Mexican snakes of the genus Rhadinaea. Occ. Papers Mus. Zool., Univ. Michigan, no. 412, pp. 1-19 + pls. I-II. Blair, W. Frank 1958. Distributional patterns 6f vertebrates in the southern United States in relation to past and present environments. Pp. 488-468 in Zo6geog- raphy. Amer. Assoc. Advancement Sci.. Blanchard, Frank Nelson 1981. Secondary sex characters of certain snakes. Bull. Antivenin Inst. Amen, vol. 4, no. 4, pp. 95-104. 1942. The ring-neck snakes, genus Diadophis: Bull. Chicago Acad. Sci; val. 7, no. i, pp. 1-144. Boulenger, George A. 1894. Catalogue 6f. the snakes in the British Museum (Natural History). Vol. 2. London: pp. i-xi + 1-882 + pls. 1-20. 1967 MYERS: THE PINE WOODS SNAKE 98 Brimley, C. S. 1925. The seasonal catch of snakes at Raleigh, North Carolina. J. Elisha Mitchell Sci. Soc., vol. 41, pp. 100-I03. Brode, W. E., and Phillis Allison 1958. Burrowing snake5 of the panhandle counties of Mississippi. Herpeto- logica, vol. 14, pt. 1, pp. 37-40. Brown, Arthur Erwin 1901. A review of the genera and species of American snakes noFth of Mex- ico. Proc. Acad. Nat. Sci. Philadelphia, vol. 53, pp 10-110. 1903. Texas reptiles and their faunal rehtkns. Ibid., vol. 55, pp 548-558. 1904. Post-glacial Nearctic centres of dispersal for reptiles. Ibid., vol. 56, pp,, 464-474. 1908. Generic types of Nearctic reptilia and amphibia. Ibid., vol. 60, pp. 112-127. Campbell, George R., and William H. Stickel 1989. Notes on the yellow-lipped snake. Copeia, 1989, no. 2, p. 105. Carr, Archie Fairly, Jr. 1940. A contribution to the herpetdlogy of Florida. Univ. Florida Publ., Biol. Sci. Ser., vol. 8, no. 1, pp. 1-118. Cliburn, J. William 1959. The distribution df some snakes in Mississippi. Amer. Midland Nat., vol. 62, no. 1, pp. 218-221. Conant, Roger 1958. A Eeld guide to reptiles and amphibians. Boston: Houghton Mimin Co., xvii -+ 866 pp Cope, Edward Drinker 1871. Ninth contribution to the herpetolbgy of tropical America. Proc. Acad. Nat. Sci. Philadelphia, vol. 23, pp. 206-224. 1877. Rare snakes from Florida. Amer. Nat., vol. 11, no. 9, p. 565. 1878. On some new and little known.reptiles and fishes from the Austroiparian region. Proc. Amer. Philos. Soc:, val. 17, pA 68-68. 1888. On the snakes of Florida. Proc. U. S. Natl. Mus., vol. 11, pp. 381-894. 1894. On a collection of batrachia and reptilia from the island of Hainan. Proc. Acad. Nat. Sci. Philadelphia, vol. 46, pp. 428-428 + pl. no. X. 1895. The classification Of the .ophidia. Trans. Amer. Philos. Soc., vol. 18 (n. ser.), art. 3, pp 186-220 + pls. XIV-XXXIII. 1900. The crocodilians, lizards, and snakes of North America. Rept. U. S. Natl. Mus. for 1898, pp. 155-1270 + pls. 1-86. Coues, Elliott, and H. C. Yarrow 1878. Notes on the natural history of Fort Macon, N. C., and vicinity '(no. 4). Proc. Acad. Nat. Sci. Philadelphia, vol. 30, pp. 21-28. Critchfield, William B., and Elbert L. Little, Jr. 1966. Ge6graphic distribution of the pines of the. world. U. S. Dept. Agric. (Forest Service), Misc. Publ. no. '991, v + 97 pp 94 BULLETIN FLORIDA STATE MUSEUM Vol. 11 Croxton, Frederick E. 1959. Elementary statistics, with applications in medicine and the biological sciences. New York: Dover Publ., vii + 876 pp Ditmars, R. L. 1907. The reptile book. New York: Doubleday, Page and Co., xxxii + 472 pA + 136 pls. Dowling, Herndon G. 1951. A proposed standard system of counting ventrals in snakes. Briti5h J. Herpetology, vol. 1, no. 5, pp. 97-99. Dowling, Her*don G., and Jay M. Savage 1960. A guide to the snake hemipenis: a survey of basic structure and sys- tematic characteristics. Zoologica, vol. 45, pt. 1, pp. 17-28 + pls. I-III. Dunn, Emmett Reid 1982. The status of the snake genus Rhadinaea Cope. Oce. Papers Mus. Zool., Univ. Michigan, no. 251, pp. 1-2. 1944. A revisi6n of the Colombian snakes 6f the genera Leimadophis, Lugo- phis, Liophis, Rhadinaea, and Ptiocercus, with a note on Colombian Coniophanes. Caldasia, vol, 2, no. 10, pp. 479-495. [Reprinted in Dunn, E. R. 1957. Contributions to tHE herpetology of Colombia, 1948-1946, Privately printed. ix + 296 pp.] 1957. Neotropical frog genera: Prostherapis versus Hyloxalus, with remarks on Phunobates Copeia, 1957, no. 2, pp.. 77-78. Enge15, William L. 1952. Vertebrate fauna of North Carolina coastal islands II. Shackleford Banks. Amer. Midland Nat., vol. 47, no. 8, pp. 702-742. Ford, E. B. 1964. Ecological genetics. London: Methuen and Co., xv + 835 pp Fox, Wade 1948. Effect. of temperature' on development 4.scutellation in the garter snake, Thamnophis elegans atratus. Copeia, 1948, no. 4, pp. 252-262. Fox, Wade, Charles Gordon, and Marjorie H. Fox 1961. Morphological effects of 104 temperatures during the embryonic de- velopment of the garter snake, Thamnophis elegans. Zoologica, vol. 46, pt. 2, pp. 57-71. Funderberg, John B., Jr. 1958. The yellow-tipped snake, Rhadinaea flavitata Cope, in North Carolina. J. Elisha Mitchell Sci. Soc., vol. 74, no. 2, PA 185-136. Garman, Samuel 1884. The North American reptiles and batrachians. ' A list of the species occurring north of the Isthmus of Tehuantepec, with references. Bull. Essex Inst., vol. 16, pp. 1-46. Goin, Coleman J. 1958.. Comments upon. the origin,of the herpetofauna of Florida. Quart. J. Florida Acad. Sci., vol. 21, no. 1, pp. 61-70. 1967 MYERS: THE PINE WOODS SNAKE 95 Coin,' Coleman-J., and Olive B. Coin 1958. Temporal variation in a small community of amphibians and reptiles. Ecology, vol. 34, no. 2, pp. 406-408. Grobman, Arnold B. 1941. A contribution t6 the knowledge of variation in Opheodrus vernalis (Harlan), with the description of a new subspecies. Misc: Publ., Mus. Zool., Chiv. Michigan, no. 50, pp. 1-88. 1944. The distribution 6f the salamanders of the genus Plethodon in.eastern United States and Canada. Annals New York Adad. Sci., vol. 45, art. 7, pp. 261-316. 1950. The problem of the natural range of a species. Copeia, 1950,.no, 8, pA 231-282. Haltom, William L. 1981. Alabama reptiles. Alabama Mus. Nat, Hist,, no. 11, pp. vi + 145 pA Harshberger, John W. 1916. The vegetation 6f the New Jersey pine-barrens. Philadelphia: Christo- pher Sower Co., xi + 329 pp. + f61dout fig. and map. Holman, J. Alan 1958. The Pleistocene herpetofauna of Saber-tooth Cave, Citrus County, ,Flor- ida. Copeia, 1958, no. 4, pp 276-280. 1959. Amphibians and reptiles from the Pleistocene (Illinoian) of Williston, F.lorida. Copeia, 1959, no. 2, pp. 96-102. King, Wayne 1959. Vertebra duplication, an osteological anomily widespread in snakes. Herpetologica, vol. 15, pt. 2, pp. 87-88. Klauber, Laurence M. 1940. Two new subspecies of Phyllorhvnchus. the leaf-nosed snake, with notes on the genus. Trans. San. Diego Soc. Nat: Hist., vol. 9, no. 20, pp. 197-212 + pl. 8 and map. , 1948. Tail-length differences in snakes, with notes on sexual dimorphism and the coeBcient of divergence. Bull. Zool. Soc. San Diego, no. 18, Pp. 5-60. Laessle, Albert Middleton 1942. The plant communities of the Welaka Area. Univ. Florida Publ., Biol. Sci. Ser., vol. 4, no. 1, pA 1- 148. L6ding, H. P. 1922. A preliminary catalogue of Alabama amphibians and reptiles. Geol. Survey Alabama Mus. Paper no. 5, pp. 1-58. Malnate, Edmond 1939. A study of the yellow-lipped snake, Rhadinaea Bavilata (Cope). Zoolog- ica, vol. 24, no. 8, pp. 859-866 + pl. 1. Mayr, Ernest 1963. Animal species and evolution. Cambridge: Harvard Univ. Press, xiv + 797 pp. 96 BULLETIN FLORIDA STATE MUSEUM Vol. 11 Myers, Charles W. 1965. Biology of the ringneck snake, Diadophts punctatus, in Florida. Bull. Florida State Mus., vol. 10, no. 2, pp. 43-90. Neill, Wilfred T. 1954a. Evidence of venom in snakes of *e genera Atgophis and Rhadinaea. C6peia, 1954, no. 1, pp. 59-60. 1954b. Ranges and taxonomic allocations of amphibians and reptiles in the southeastern United States. Publ. Research Div. Ross Allen's Reptile Inst; 961.1, no. 7, pp. 75-96. 1960. The caudal lure of various juvenile snakes. Quart. J. Florida Acad. Sci., vol. 28,.no. 3, pp. 178-200. 1968. Polychromatism in snakes. Ibid., vol. 26, no. 2, pp I94-216. 1964. Taxonomy, natural historx, and zochgeography of the rainbow snake, Farancia erytrogramma (Palisot de Beauvois). Amer. Midland Nat., v61. 71, no. 2, pp. 257-295. Neill, Wilfred T., and E. Ross Allen 1956. Secondarily ingested food items in, snakes. Herpetologica, vol. 12, pt. 8, pp. 172-174. Netting, M. Graham 1936. Rhadinaea flavilata (Cope) in Texas. Copeia, 1986, no. 2, p. 114. Peters, James A. 1956. An analysis of variation in a South American snake, Catesby's snail- sucker (Dipsas catesbvi Sentzen). Amer. Mus. Novitates, no. 1783, pp. 1-41. 1960. The snakes of the subfamily Dipsadinae. Misc. Publ. Mus. Zool., Univ. Michigan, no. 114, pp. 1-224 + pls. I-VIIL Robertson, William B., and Edwin L. Tyson 1950. Herpetological notes from eastern North Carolina. J. Elisha Mitchell Sci. Soc., vol. 66, no. 2, Dp· 130-147. Roze„ Jdnis A. 1958. A new species of the genus Urotheca (Serpentes: Colubridae) from Ven- ezuela. Breviora, Mus, Comp. Zool:, no. 88, pp. 1-5. 1959. Taxonomic n6tes on a collection of Venezuelan reptiles in the American Museum of Natural History. Amer. Mus. Novitates, no. 1934, pp. 1-14, Schmidt, Karl P. 1916. Notes on the herpetology of North Carolina. J. Elisha Mitchell Sci. Soc., vol. 82, pp. 83-87. 1953. A check list of North American amphibians and reptiles. 6th ed. Univ. Chicago Press: Amer. Soc. Ichthyologists and Herpetologists, viii + 280 pp. Simpson, George Gaylord 1929. The extinct land mammals of Florida. Ann. Rept. Florida Geol. Surv., vol. 20, pp. 229-279. 1967 MYERS: THE PINE WOODS SNAKE 97 Small, John Kunkel 1938. Manual of the southeastern Rora. Chapel Hill: Univ. North Carolina Press, xxii + 1554 pp. Smith, Philip W., and James C. List 1955. Notes on Mississippi amphibians and reptiles. Amer. Midland Nat., vol. 58, no. 1, pp 115-125. Stejneger, Leonhard, and Thomas Barbour 1917. A check list of North American amphibians and reptiles. 1st ed. Cam- bridge: Harvard Uhiv: Press, iv + 125 pp. Telford, Sam Rountree, Jr. 1952. A herpetological survey in the vicinity of Lake Shipp, Polk County, F16rida. Quart. J. Florida Acad. Sci., vol. 15, no. 8. pp. 175-185. 1961. Studies on the incidence of intestinal proto-zoan inquilines in snakes and lizards of southeastern United States. Uhpubl. Master's Thesis, Univ. Florida, 102 pp. 1966. Variation among the southeastern crowned snakes, genus Tantilla. Bull. Florida State Mus., Biol. Sci., v61. 10, no. 7, pp. 261-804. Vernon, Robert 0. 1951. Geology of Citrus and Levy counties, Florida. Geol. Bull., Florida Geol. Surv., no. 83, ix + 256 pp Wahlenberg, W. G. 1946. Longleaf pine. Washington, D. C.: Charles Lathrop Pack Forestry Foundation, xxii + 429 pp. Wells, B. W. 1924. Major plant communities of North Carolina. North Carblina Agri. Exii. Sta. Tech. Bull., no. 25, pp. 1-20. White, Laurence B. 1960. .Notes on collecting in Ncrth Carolina pine plantations. Copeia, 1960, no. 1, pp. 49-50. Contributions to the BULLETIN OF THE FLORIDA STAfE MUSEUM may- be in any field of biology. Manuscripts dealing with natural history or systematic problems involving the southeastern United States or the Caribbean area are solicited especially. Manuscripts should be of medium length-50 to 200 pages. Examination for suitability is made by an Editurial Board. The Bul.~ETIN is distributed worldwide thr6ugh institutiofial subscripti6ns and exchanges only. It is considered the responsibility of the author t6 distribute his paper to all interested individuals. To aid in this, fifty copies are furnished the author without cost. PREPARATION OF MANUSCRIPT Highly recommended as a guide is the volume: Conference of Biological Editors, Committee on Form and Style. 1960. Style manual for biological journals. Amer. In5t. Biol. Sci., Washington. 92 p. Manuscripts should be typewritten with double spacing throughout, with ample margins, and on only one side of the paper. The author should keep a copy; the original ind a carb6n must be submitted. Tables afd legends of figures should be typed on sheets separate from the text. Several legends or tables may be placed on a single sheet. Illustrations, including maDs ahd ph6tograph5, should be referred to .as "figures." All illustrations .are reduced to a maximum of 4-1/4 by 7-1/8 inches. Size 5cales, wherever they are: necessary, should be incorporated· into the figure. References to literature should conform with the bibliographic style used in recent numbers of the BULLETIN. Spell out in full the titles 6f non-English serials and places of publication. Footnote material should be kept to a minimum. However, provide copy for a footnote detailing the title, afRhations, and address of the author (see recent numbers of the BULLETIN). Manuscripts must be accompanied by a synopsis-a brief and factual summary (not a mere descripti6n) of the contents and conclusions, which points out the presence of any new information .and indicates its relevance. In it list all new organisms described and give their ranges; iddicate all. taxonomic changes pro- posed. The synopsis, written in full sentences, should be concise, but completely intelligible in itself without references to the paper, thereby enabling the busy reader t6 decide more surely than he can from the title alone whether the paper merits his reading. The synOpsis will be published with the paper, It dkles not replace the usual conclusions or summary sections. It may also serye as copy f6f the abstracting services. Manuscripts and all editorial matters should be addressed to: Managing Editor of the BULLETIN F16rida State Museum Seagle Building Gainesville, Florida