B ULLETIN OF THE FLORIDA STATE MUSEUM BIOLOGICAL SCIENCES Volume 12 Number 4 GEOGRAPHIC VARIATION IN RHINEURA FLORIDANA (REPTILIA: AMPHISBAENIDAE} George R. Zug .tu Of - *-Me- 19* /853 UNIVERSITY OF FLORIDA Gainesville 1968 Numbers of the BULLETIN OF THE FLORIDA STATE MUSEUM are pub- lished at irregular intervals. Volumes contain about 800 pages and are not necessarily completed in any one calendar year. WALTER AUFFENBERG, Managing Editor OLIvER L. AUSTIN, JR., Editor Consultant for this issue: WALTER AUFFENBERG Communications concerning purchase or exchange of the publication and all manuscripts should be addressed to the Managing Editor of the Bulletin, Florida State Museum, Seagle Building, Gainesville, Florida. 82601. Published June 28, 1968 Price for this issue .60¢ 1968 ZUG: VARIATION IN RHINEURA FLORIDANA 185 GEOGRAPHIC VARIATION IN RHINEURA FLORIDANA ( REPTILIA: AMPHISBAENIDAE) GEORGE R. ZUG 1 SYNOPSIS: Examination of 11 characters in Rhineura #oridana reveals no sexual dimorphism and a geographic pattern of variation in only 6 characters: Lateral head scales, supralabials, infralabials, body annuli, caudal annuli having dorsal granular segments, and segments around the midbody. Two populations are. identified but not separated nomenclaturally, 6ne on the Lake Wales ridge, the other occupying the remainder of north-central Florida. It is suggested that other Florida scrub or sandhill associated reptiles such as Ophisaurus compressus, Neoseps Teynoldsi, Eumeces egregius, Sceloporus woodi, Stilosoma extenuatum, and Tantilla relicta share a history of isolation and differentiation similar to that proposed for R. Bondana. INTRODUCTION The herpetofauna of peninsular Florida contains a relatively high frequency of endemics that are generally thought to be the result of ecological changes associated with the Pleistocene eustatic sea level fluctuations. Several levels of endemism are recognized: subspecific, specific, and generic ( Neill, 1957) ; these levels are partly correlated with the dispersal abilities of the different faunal elements and the survival of the mainland populations. Of particular interest is the fact that the three monotypic endemic squamate genera (Neoseps, Rhineura, and Stilosoma) are fossorial and thus would be expected to have relatively low rates of dispersal. Of the three, Rhineura has the largest geographic range and great- est ecological tolerance. Neoseps reynoldsi is found only in the mar- ginal areas of rosemary scrub and in the longleaf pine-turkey oak habitat of the central Florida highlands ( Telford, 1959; Myers and Telford, 1965). Stilosoma extenuatum appears to be restricted to high pine associations in the northern half of peninsular Florida ( Highton, 1956). In contrast R. Boridana occupies the northern two-thirds of the peninsula ( Fig. 1) and is found in a variety of mesic and xeric habitats, scrub, sand-hill, flatwood, hammocks, ( Neill, 1951). i The author is a paduate student at the University of Michigan. His current research emphasizes iunctional anatomy of turtles. Manuscript submitted 8 Sep- tember 1967.-ED. Zug, Gedrge R. 1968. Geographic Variation in Rhineura floridana ( Reptilia, Amphisbaenidae). Bull. Florida State Mus., vol. 12, No. 4, pp 185-211. 186 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Highton ( 1956) and Telford ( 1959) have shown that the distribu- tions and variations of Stilosoma and Neoseps are positively correlated with marine terraces. Sea level fluctuation must have affected the distribution and variation of other organisms as well. This paper examines the variation of certain morphometric and meristic charac- ters of R. Boridana to determine its extent, type, and geographic pat- tern and, if possible, the geologic period of isolation in which it ap- peared. ACKNOWLEDGMENTS I wish to thank the following individuals and their respective institutions for allowing me to examine their specimens: Charles M. Bogert, American Museum of Natural History ( AMNH); James E. B6hlke and Edmond V. Malnate, Academy of Natural Sciences of Philadelphia ( ANSP); Clarence J. McCoy, Carnegie Mus- eum ( CM); Hymen Marx, Field Mu5eum of Natural History ( FMNH); Ernest E. Williams, Museum of Comparative Zoology ( MCZ); William J. Riemer and Walter Auffenberg, University of Florida/Florida State Museum ( UF); James A. Peters, United States National Museum ( USNM). Carl Gans gave me a list of museums having specimens of Rhineura. Kraig Adler, Arnold G. Kluge, and Charles F. Walker were extremely helpful in improving and clarifying the text. My wife, Patricia, aided in the preparation of the manuscript. This study was supported in part by a National Science Foundation grant ( GB-3366) to the University of Michigan Museum of Zoology. TAXONOMY AND DISTRIBUTION The Florida worm lizard was described by S. F. Baird in March 1859 as Lepidostemon Boridanum; the holotype was collected at Micanopy, Alachua County, Florida, by J. B. Barratt. In June 1861 E. D. Cope proposed on the basis of additional material that this amphisbaenid was generically distinct from Lepidosternon and erected for it the genus Rhineura, in which it has remained ever since. As shown by museum specimens and literature references, the geo- graphic range of R. #oridana ( Fig. 1) occupies the northern two- thirds of the Florida penimisla except for the area east of the Kissim- mee drainage system. Its absence there may represent inadequate collecting, but should be considered real until its presence is con- firmed unquestionably. The northernmost locality is Lake City, Colum- bia County, and the southernmost is the Archbold Biological Station, Highlands County. Burt's ( 1937) record of a Desoto County speci- men is actually from Highlands County, collected before Desoto coun- ty was partitioned. A number of records need verification. One speci- men ( MCZ 4655) reportedly collected in "Georgia" is not improbable, 1968 ZUG: VARIATION IN RHINEURA FLORIDANA 187 as Lake City is only 30 miles south of the Georgia line, but the record needs confirming. Telford ( 1959) reports Rhineura occurring on the east coast of Florida "as far south as Miami," but unfortunately pro- 9 r ----L L- L._.J0 .• O . r 1 1.0. f T- r 'i_ L Figure 1. Present distribution of Rhineura floridana based on museum specimens ( solid circles) and literature records ( open circles). The question mark represents an unconfirmed Georgia locality. 188 BULLETIN FLORIDA STATE MUSEUM Vol. 12 vides no positive evidence for the statement. Wilfred Neill ( pers. comm., 1963) suggested that R. floridana occurs as far south as Im- mokalee, Collier County, but I have been unable to locate any speci- mens from this region. CHARACTER DESCRIPTIONS AND ANALYSES The characters used in this study are those of scutellation, snout- vent length, and tail length. The scale characters are divided into two Table 1. COMPARISON oF MALE AND FEMALE Rhineura foridana Coem- Stand- cient ard of Mean Devia- Vari- t F Number Range X tion ance value value 10 245-256 251.7 8.77 1.5 Body 2.04 1.68 9 10 244-234 248.6 2.91 1.2 10 6-10 7.9 1.20 15.1 Total 1.06 1.44 A N N U LI Cloacal 9 10 6-10 8.5 1.48 16.9 10 16-17 16.7 0.48 2.9 Caudal 2.48 1.80 9 10 16-17 16.2 0.18 0.8 9 12-18 12.4 0.52 4.2 Granular 2.42 0 9 10 12 12.0 0.00 0 Segments & 10 27-82 29.2 1.31 4.5 around 1.91 1.10 Midbody 9 10 28-82 80.3 1.25 4.1 r. Total lateral 8 10 19-29 24.9 3.12 12.5 1.44 1.24 head scales 9 10 19-29 28.0 2.79 12.1 10 13.70-16.14 15.4 0.71 4.6 SV/TL 1.47 1.44 9 10 14.00-16.67 15.8 0.85 5.4 1968 ZUG: VARIATION IN RHINEURA FLORIDANA 189 groups, those associated with body annuli and those of the head region. These are described below in detail and their variation ana- lyzed, but only those characters showing geographic variation are discussed fully. In most cases I have followed the procedures and terminology used by Cans and Alexander ( 1962) for Antillean amphis- baenids except for their head scale terminology, which is unsuitable for R. foridana Instead Smith's ( 1946) terminology erected specifi- cally for R. #oridana is the main framework used, though I define the loreals, oculars, and temporals somewhat differently than Smith does. SEXUAL DIMORPHISM A sample of 20 R. Boridana from Gainesville was used to deter- mine the extent of sexual dimorphism. Sex was determined by dis- section, and seven characters were compared between the female and male samples ( Table 1). The method by which these characters were recorded is discussed below. Comparisons by the t-test shows no significant difference between the means of the sexes in any charac- ter at the 0.01 level. The variability of the two sexes is also similar. Thus in these characters Rhineura shows no sexual dimorphism. HEAD SCUTELLATION The head scales can be divided into four series: d6rsal, lateral, labial, and genial. Anteriorly to posteriorly the dorsal series ( Fig. 2A) is comprised of a large frontonasal, a small prefrontal, a large penta- gonal frontal, and a pair of parietal scales. The frontonasal covers the snout and extends onto the ventral surface; the parietals are usual- ly borded posteriorly by the first body annulus, although in some specimens a wedge of temporal scales extends upward between the parietals and the first body annulus. Variation in the dorsal series includes both the fusion and fragmentation of scales. The fusion may include all members of the series or any fraction thereof. In some instances fusion appears to be a result of injury, for the texture of the scales is different. New scales seem to be formed by the development of a new suture plane in one of the regular members of the dorsal series. As no pattern or regularity could be determined for this varia- tion, no method was devised for recording it. The lateral series of scales ( Fig. 2C) is subdivided into a single anterior nasal, an anterior loreal group, a central ocular group, and a posterior temporal group. This series of scales is extremely variable, making it necessary to delineate boundaries and landmarks. As the anterior most scale of the series is perforated by the external naris, 190 BULLETIN FLORIDA STATE MUSEUM Vol. 12 fn 'r' pf m d Pm P 1I BA I C 1 , D Figure 2. Scutellation of the head and tail of Rhineura floridana. A, dorsal view of head; B, ventral view of head; C, lateral view of head; D, ventral view of tail ( arrows indicate first caudal annulus and last body annu- lus). f, frontal; fn, frontonasal; il, infralabial; m, mental; ma, malar; n, nasal; 0, ocular; p, parietal; pf, prefrontal; pm, postmental; po, preocular; pt, postocular; r, rostral; sl, supralabial, sc, subocular; t, temporal. 1968 ZUG: VARIATION IN RHINEURA FLORIDANA 191 it is referred to as the nasal. Posteriorly the lateral series is bounded by the first body annulus. Establishment of an anterior and a posterior boundary is useful, but does not distinguish the different elements lying between these borders. For this purpose the site of the eye was selected as a landmark; the scale over the eye is considered to be the ocular and the scales immediately surrounding it are referred to the ocular group. The eye has certain limitations as a landmark, as the position of the overlying scales varies and the eye is visible in only a minority of specimens. The eye most often lies beneath the posteroventral corner of a large scale bordering both the parietal and frontal dorsally ( Fig. 2C). This scale is designated the ocular, because.the eye is most frequently beneath it and the scale maintains a regularity in shape and position. The eyes may also lie beneath the ocular and postocular, the ocular and subocular, the ocular, sub- ocular, and postocular, the postocular, and the subocular. Although its position varies, the use of the eye as a landmark is defensible, for once a terminology has been established arbitrarily and clearly defined, it is possible to assign a scale a definite name with a minimum of doubt. The eye is visible in less than 10 per cent of the specimens examined and only in individuals less than 260 mm. Neill ( 1951) noted the eye in a 311-mm-specimen and stated that no correlation exists between its development and body length, body scutellation, or geography. Three other authors have commented on the eye of Rhineura; Carman ( 1883) stated that it is present in young individ- uals, Carr ( 1949) supported this with observations on hatchling specimens, and Eigenmann ( 1902) discussed eye anatomy. The ocular is bordered dorsally by the parietal and frontal. Usually a pair of suboculars are present between the ocular and the fourth supralabial. A single preocular lies anterior to the ocular and con- tacts the frontal dorsally and usually contacts the anterodorsal edge of the fourth supralabial ventrally. The anterior scale of the lateral series is the nasal, wedged between the frontonasal and rostral. Be- tween the nasal and the preocular lie the loreals; these are bordered dorsally by the frontal and prefrontal, anterodorsally by the fronto- nasal, and ventrally by the second, third, and extreme anterodorsal edge of the fourth supralabial. Characteristically there is a large dor- sal loreal, and seldom to the loreals contact the ocular. Posteriorly the ocular is in contact with the postocular. The postocular has a broad contact with the fourth supralabial and rarely touches the parietal dorsally. The temporals are enclosed by the first body annulus poster- 192 BULLETIN FLORIDA STATE MUSEUM Vol. 12 iorly, parietal dorsally, preocular and ocular anteriorly, and postero- dorsal edge of fourth supralabial ventrally. Not infrequently one or two temporals extend into or completely through the first body an- nulus. The two definite trends observable in the lateral series are reduc- tion and increase in number of scales. The most frequent pattern, that most closely resembling the mean number of scales, is apI)roxi- mated by figure 3B and C. Figure 3A is the reduced pattern; probably A B C D E Figure 3. Patterns of variation in the scales of the lateral series of the head. The sequence of A through E illustrates a progressive increase in number; A shows the typical Lake Wales Ridge pattern. best interpreted as the fusion of elements: the ocular and preocular, a loreal and subocular, and the postocular and some temporals. As a result of fusion the preocular and frequently the subocular are absent. Although this reduced pattern is found throughout the range, it oc- curs in all the specimens from Highlands County and in 54 per cent of the specimens from Polk County. The increase in the number of scales shown in figures 3D and E can be accounted for mainly by the appearance of new suture planes that create more and smaller elements. Another cause of a high total number of lateral head scales is the presence of an additional row of temporals extending upward posterior to the parietals. This row might be interpreted as the dorsal portion of the first body annulus, but if so interpreted the following or second annulus would be incomplete ventrolaterally and ventrally. Specimens with high scale counts occur sporadically throughout the entire geographic range except in Highlands County. 1968 ZUG: VARIATION IN RHINEURA FLORIDANA 198 The labial series includes those scales entlosing the mouth and is composed of a supralabial row and an infralabial row ( Fig. 2B). The supralabial row has a single anteromedial rostral scale and usually four supralabials extending posteriorly on each side of the mouth. The in- dividual supralabials become progressively larger in size posteriorly; the length of the fourth supralabial almost equals the combined lengths of the first three. In 92% of the samples there are 4 suprala- bials on each side of the head; other combinations are 3/3,3/4, and 4/5 ( either left or right). The absence of a 5/5 condition is not sur- prising, as only two specimens have 5 supralabials. In contrast the Highlands County sample all have a 3/3 condition, as do a few speci- mens from Polk County. Five supralabials result from the division of the first supralabial; three supralabials result from the fusion of the first and second supralabials. The infralabial row has a large anteromedial mental scale and usually 4 infralabials extending posteriorly on each side. The infrala- bials are more variable than the supralabials and show the following combinations: 3/3,3/4,4/4,4/5,5/5, and 6/6. The last combination was present in only one specimen. The frequency of occurrence of the other combinations was 3/3 - 12%, 3/4 - 8%, 4/4 - 59%, 4/5 -9%, and 5/5 - 12 %. Although these combinations occur throughout the range of Rhineura, the frequency of some combinations is often higher in certain localities, e.g., 3/3 in 87% of the specimens from Highlands and Polk Counties; 5/5 in 75 % of the individuals from Pasco and Pinellas Counties; 4/5 in 31 % and 5/5 in 54% of the specimens from Volusia County. Three infralabials result from the fusion of the first infralabial either to the mental or to the second infralabial; five in- fralabials arise from a division of the fourth infralabial. The infralabial series usually encloses four scales anteriorly and laterally, the postmentals and malars of the genial series ( Fig. 2B). The scales on each side of the midline are the postmentals, and between the postmental and infralabials of each side lies the malar, The variation in the number of genials ( Table 2) is not the result of their progressive subdivision into smaller and smaller units, but the development of sutures in their posterior corners. The presence or absence of these small posterior genial scales has no meaningful geographic pattern. As previously noted, only certain scales or scale series show geo- graphic variation, and the discussion is limited to these characters, which are three in number: the total number of lateral head 194 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Table 2. SCALE AND PROPORTION VARIATION IN Rhineura floridana Coeffi- Prob. cient error of Mean Stan. Dev. rrlean Vari- Number Range X s Sx ance HEAD SCALES Lt Loreals 169 1-6 8.2 0.84 0.06 26.19 Rt. Loreals 171 1-6 8.8 0.80 0.06 24.27 Lt. Temporals 170 1-11 4.5 1.86 0.14 41.38 Rt. Temporals 172 1-9 4.4 1.66 0.13 37.78 Lt. Oculars 160 2-7 4.4 1.79 0.14 40.68 Rt. Oculars 169 2-7 4.5 0.94 0.07 20.89 Tot. Lat. Head Scales 159 12.89 24.2 4.89 0.89 20.21 Tot. Supralabials 167 6-9 7.9 0.51 0.04 6.89 Tot. Infralabials 168 6-12 8.0 2.15 0.17 26.74 Total Genials 176 4-18 5.4 1.59 0.12 29.44 BODY SCALATION Annuli Body 187 226-267 250.2 5.46 0.40 2.18 Lt. Cloacal 189 8-6 4.0 0.72 0.05 18.88 Rt. Cloacal 187 8-6 8.9 0.75 0.05 19.04 Caudal 187 14-19 16.0 0.85 0.06 5.81 Granular 185 8-15 12.4 1.19 0.09 9.64 Segments around Midbody 189 28-87 29.8 2.58 0.18 8.64 SV/TL 179 11.8-20.8 16.41 1.31 0.10 7.98 scales excluding the nasal, total number of supralabials, and total number of infralabials. It might be better to consider only the ocular group of the lateral series, which demonstrates the variation most clearly, but I have used the entire lateral series even though the reduction trend was not always as distinct in the other members. To demonstrate geographic variation the sample is divided into units encompassing as discrete localities as possible ( Table 3). Ideally each sample should be of equal size and from localities of nearly equal size and uniform habitat; unfortunately the fortuitous nature of Rhineura collections prevents this. The means for the total number of supralabials, total number of infralabials, and total number of lateral head scales excluding the nasals were determined for each sample; classes for the means of the three characters were established and as- signed numerical values, zero representing the class with the lowest mean ( Table 4). The assigned values are plotted on maps ( Fig. 4), Table 8. GEOGRAPHIC vARIATION IN Rhineura 8oridana Lateral Granular Midbodv Head Scales Supralabials Infralabials Body Annuli Annuh Segments N Mean N Mean N Mean N Mean N Mean N Mean 1968 ZU G : VA R IAT IO N IN R H IN E U R A FLO R ID A N A 195 Sample locality Alachua County Gainesville ( G) 20 28.6 21 8.0 21 8.0 22 250.8 21 12.2 21 29.7 Micanopy (M) 11 29.8 11 8.0 11 7.8 11 249.7 9 12.4 11 29.9 Waldo-Lake Santa Fe (Wa) 8 81.0 8 8.0 3 8.0 8 245.7 8 14.3 8 29.8 Marion County Ocala-Silver Springs ( Oc) 5 24.4 5 8.0 5 8.0 5 252.6 5 11.4 5 26.6 Eureka (E) 1 25.0 1 8.0 1 7.0 8 258.8 8 11.8 3 29.7 Marion County 84 26.8 34 7.9 84 8.0 84 252.5 84 11.9 84 28.6 Lake County Leesburg-Fruitland Park ( Le) 4 22.5 5 8.0 5 7.4 5 252.2 5 13.8 5 28.0 Travares-Mt. D6ra- Eustis-Umatilla (U) 5 21.6 5 8.0 5 8.0 6 253.0 6 12.8 6 29.0 Orange County Orlando ( Or) 9 28.8 11 8.0 11 8.4 11 253.7 12 12.9 12 29.6 Hillsborough County Tampa area ( Tm) 3 25.8 8 8.3 8 9.7 4 251.5 4 13.8 4 80.5 Pasco County (P) 8 24.0 3 8.0 8 10.0 3 252.8 8 13.8 3 88.7 Pinellas County Clearwater (C) 2 28.5 2 8.0 2 9.5 2 253.5 2 14.0 2 82,0 Tarpon Springs ( Ts) 7 28.4 8 8.0 6 9.5 8 252.1 8 13.8 8 29.5 Volusia County Volusia (V) 9 26.1 9 8.0 9 9.6 12 248.7 12 12.8 12 29.3 Polk County Winter Haven area ( Wh) 6 17.2 7 6.8 7 6.8 8 241.7 8 10.5 8 26.3 Lakeland (L) 8 28.8 4 7.5 4 7.7 5 248.4 5 12.4 5 83.6 Aubumdale (A) 8 28.7 8 8.0 8 8.0 4 248.0 4 12.5 4 30.3 Highlands County Sebring-Avon Park (S) 1 12.0 1 6.0 1 6.0 2 245.0 2 12.0 2 25.0 Hicoria-Lake Placid ( H) 5 12.2 5 6.0 5 6.0 5 2412 5 10.4 5 24.8 196 BULLETIN FLORIDA STATE MUSEUM Vol. 12 which show a trend towards the reduction of scale number in all three characters at the southern extremity of the range. This area, the Lake Wales Ridge, is the southern spine of the Central Highlands and is a region of high sandy hills largely covered with scrub. Although the three characters are similar at the southern end of the range, the reduction pattern ends differently in central Polk County. The total lateral head scale mean for the Winter Haven area is inter- mediate between the extreme low to the south and the areas to the north and west. The supralabial and infralabial means for Winter Haven and Highlands are nearly identical, and may suggest the pre- sence of a continuous population between these two areas. Winter Haven is less than 20 miles from Auburndale and Lakeland, yet the means for the lateral ,head scales of these latter areas are similar to those of north-central Florida. The situation is the same for the in- fralabial means of these two areas; however the Lakeland supralabial mean is appr6ximately intermediate between the conditions to the north and the Lake Wales Ridge. As Auburndale is closer to Winter Haven than Lakeland, it would seem more likely for the intermediate condition to be present at Auburndale. A soil map of this region ( Fowler, 1927) shows high sand ridges in all three areas, each of which is isolated by low marsh soils. Thus the Lake Wales Ridge 1/f / 1-1i 2 A: .Fh 1 711.2A 4 2 'i tz 9 5[72 1% \36* 2 20 U 1 ~- o ~ ~C, 1 ~ 4 for C-J Figure 4. Geographical variation in head scales.1 A, the sample localities, ( see Table 8 for explanation of abbreviations); B, number of lateral head scales; C, number of supralabials; D, number of infralabials: 1The integers denote classes of means (Table 4). 1968 ZUG: VARIATION IN RHINEURA FLORIDANA 197 population appears to be characterized by a reduction in three scale groups. In north-central Polk County the morphological evidence suggests that gene flow may be occurring between the populations of the Lake Wales Ridge and north-central Florida, though interrupted or reduced by the marshy lowlands between the high sand ridges. The Lake Wales Ridge samples are not the only ones that fall out- side the north-central Florida means. In southwestern Pasco County, eastern Hillsborough County and Pinellas County, the means for the infralabials are appreciably higher than for the rest of the range. This is the only character that shows any differentiation in this region, and it may be explained in two ways: the swamping or elimination Of a previously differentiated population, or the beginning of differentia- tion here. The former hypothesis seems the more reasonable, because R. ~loridana is present in most of the Floridian habitats with the ex- ception of marshes, no barriers isolate this region today. Other local differences are illustrated by the maps. Throughout most of the range of Rhineura the means of total lateral head scales fall between 20.6 and 25.5, except in Volusia, Micanopy, and Waldo sub-samples. The Micanopy and Waldo means, 29.8 and 31.0, respec- tively are similar to one another and differ from the remainder of the north-central samples, yet lying between them is the Gainesville sample with an appreciably lower mean of 23.6. Although the Volusia mean ( 26.1) is higher than that for most of the population, it can hardly be considered divergent. The similarity of the mean number of infralabials for Volusia and the Tampa Bay area may indicate parallelism in this character. BODY SCUTELLATION The body and tail of R. foridana are encircled by a solid row of annuli, and the only incomplete annuli are those interrupted by the vent. Each annulus is composed of rectangular or square segments or scales. Middorsally the annuli are bisected by a groove, which can always be recognized by the presence of tiny scales wedged between the right and left halves. The segments forming the annuli are very narrow (length>width) dorsally; they become progressively wider ventrad so that laterally they are square (length=width) and ventr- ally rectangular (length80.5 4 10.1-10.5 0 10.6-11.0 1 5.5-6.5(3/8) 0 Number of 11.1-11.5 2 Total number of 6.6-7.5(3/4) 1 caudal annuli 11.6-12.0 8 supralabials 7.6-8.5(4/4) 2 with dorsal 12.1-12.5 4 8.6-9.5(4/5) 8 granular 12.6-18.0 5 5.6-6.5(8/3) 0segments 18.1-18.5 6 Total number of 6.6-7.5(8/4) 118.6-14.0 7 infralabials 7.6-8.504/4) 2 14.1-14.5 8 8.6-9.5(4/5) 3 24.0-25.9 0 9.6-10.5(5/5) 4 Number of 26.0-27.9 1 segments around 28.0-29.9 2 midbody 80.0-81.9 8 82.0-38.9 4 200 BULLETIN FLORIDA STATE MUSEOM Vol. 12 gin anteriorly as a pair of middorsal granular segments. Posteriorly the number of granular segments rapidly increases so that the dorsal surface of the posterior half of the flattened tail is completely covered. This character also demonstrates a reduction trend on the Lake Wales Ridge ( Fig. 5B), but the Sebring mean is higher than the two Marion County nneans. 42 2 3 6 2 33 75 3 3 5 26 4 6 4% A B C Figure 5. Geographical variation in body scutellation. A, number of body an- nuli; B, number of caudal annuli bearing dorsal granular segments; C, number of segments around midbody. Sample localities as in Table 3, classes of means from Table 4. The number of segments around midbody is variable both on the individual lizard and within the population. When the number of segments of consecutive annuli is counted on a single specimen, the following form of variation was seen: 28, 29, 28, 29, 29, 29, 30, 28, 28, 29. As I have counted only a single midbody annulus per specimen, the record count is an approximation and the mean condition may be one and, at the most, two units above or below the recorded count. Even with this individual variation, a reduction trend is evident in the south ( Fig. 5C). Three characters, number of body annuli, number of eaudal an- nuli with dorsal granular segments, and number of segments around midbody, show a distribution pattern similar to that of the head scutel- lation. The specimens from the Lake Wales Ridge have the lowest number of body annuli, but the Sebring sample is more similar to the 1968 ZUG: VARIATION IN RHINEURA FLORIDANA 201 Waldo sample than either is to its nearest geographic sample. Al- though the similarity of these samples may be due to their small size, it does indicate the existence of continuity between the two popula- tions. In northern Polk County the Auburndale sample is typical of the Lake Wales Ridge samples and the Lakeland sample iS inter- mediate. The remainder of the sample localities with the exception of Micanapy and Volusia have means above 250, although the Micanopy mean is barely below 250. The Volusia mean is similar to the Lakeland mean. Two areas in the Central Highlands have low mean numbers of dorsal granular annuli. The Lake Wales Ridge has the lowest means with the exception of the Sebring sample; the next lowest mean is in northeastern Marion County. The Lakeland and Auburndale samples again show a more or less intermediate condition; this is also true Of the Gainesville and Micanopy samples. All the other sample means are above 12.5 with the highest mean in northeastern Alachua County. The lowest number of segments around midbody occur in the samples from the Lake Wales Ridge and Ocala. The highest means are from the Tampa Bay area, north-central Polk County and Waldo. All the other sample means are intermediate between these two condi- tions. Thus the characters of body scutellation that exhibit geographic variation show the same pattern as the head scutellation. This further suggests the presence of a distinct population on the Lake Wales Ridge, with northern Polk County as a zone of overlap between this population and that to the north. The distinctiveness of the southern population is somewhat lessened by the means of the Sebring sample and its partial similarity to the northern population. The Tampa Bay area samples show a slight trend towards higher means in the three annuli characters, but hardly sumcient to suggest the presence of a distinct populati6n. SNOUT-VENT LENGTH AND TAIL LENGHT Neither of these characters nor their ratio (s-v/t) show any cor- relation with geographic region. When snout-vent length and tail length are plotted, the regression line is straight with a low slope, which indicates isometric growth and a low rate of tail growth. The smallest specimen examined had a 111mm snout-vent length and a 9mm tail length, the largest specimen 379mm and 26mm. The latter specimen is larger than the one reported by Telford ( 1955). Carr ( 1949) reported smaller specimens with yolk sac scars; the 111mm 202 BULLETIN FLORIDA STATE MUSEUM Vol. 12 and a 113mm specimen have scars, whereas the scar was absent in a 115mm individual. DISCUSSION It is evident from the preceding descriptions of variation that Rhineura Boridana populations are not geographically uniform, but consist of two differentiated populations, one in north-central Florida, and the other on the Lake Wales Ridge. The characters demonstra- ting this pattern are the total number of lateral head scales, total num- ber of supralabials, total number of infralabials, number of body an- nuli, number of dorsal granular annuli of the tail, and number of seg- ments around midbody. As the variation is illustrated by the use of small local samples, it is necessary to consider those specimens not in- cluded in the samples and to determine their agreement with the observed trend of reduction in scutellation at the southern extremity of the range. The three specimens from Clay County, Columbia Coun- ty, and Georgia fall within the range of variation of the Alachua County samples except.for the number of body annuli of the Columbia County specimen ( 242) which is only 2 annuli below the range of the Alachua samples. The holotype from Micanopy and the speci- mens from Flagler, Putnam, and Levy Counties agree closely with the Alachua and other north-central Florida samples. If the 34 speci- mens labeled only Marion County were included in the Ocala and Eureka samples, they would change the means and ranges only slightly ( see Table 3 and 4). The Tampa Bay area samples are characterized by the highest mean number of infralabials, high mean numbers of caudal granular annuli, and segments around midbody. the other characters are in- dentical with the north-central population. The two Hernando Coun- ty specimens are similar to the Tampa Bay area sample only in a high number of granular annuli ( 13 and 15). The affinities of two Man- atee County specimens appear to be with the Lakeland sample in that the counts are intermediate, and they have a low number of infralabials ( 3/4,3/3). The only specimen from Polk County not included in the samples is from Phantom Grove, a locality I have been unable to find; its characteristics indicate that this specimen belongs to the Lake Wales Ridge population: lateral head scales 12, supralabials 6, infralabials 6, body annuli 226, granular annuli 10, and midbody segments 25. Variations in sample sizes and in size and habitat uniformity of the local areas sampled have doubtless introduced some bias into 1968 ZUG: VARIATION IN RHINEURA FLORIDANA 208 the analysis, but the data nevertheless indicate strongly the existence of two populations, one in the Lake Wales Ridge, the other in north-central Florida. The unique characteristic of the Lake Wales Ridge population is the consistent reduction in scutellation. One would not expect a series of unrelated characters all to show the same trend. The number of head scales might be reduced if the head were shortened. There is no evidence for this, as individuals of the same snout-vent length from the two populations have similar facial lengths (length from tip of snout to posterior edge of last supralabial). Alachua County, s-v 192.5, mean, 137-239, range; facial 4.76,3.8-5.4; High- lands County, s-v 189.5, 115-240; facial 4.85,3.9-5.4. Thus no cor- relation between facial or head length and number of head scales is apparent. Nor does there appear to be any reduction in body length that would account for a reduced number of body annuli; although the maximum length of specimens examined from the Lake Wales Ridge was 248 mm, Telford ( 1955) reported a 356 mm ( snout-vent) specimen from Winter Haven. The Lake Wales Ridge population is distinct, but it is extremely doubtful that it should be recognized taxonomically. The obvious area of overlap between the two populations is between Winter Haven and the Auburndale and Lakeland area; this region is probably a zone of intergradation. Before a name can be applied validly to the Lake Wales Ridge population, I believe that larger samples should be procured from Highlands and Polk Counties to delineate more accurately the statistical parameters, range, and size of the intergradation zone. At this time it seems sumcient to recognize that sea level fluctuation probably isolated this population and that different selective pressures caused divergence. The two popula- tions, still appear to be subject to differential selective pressures, for if these were the same and gene flow were possible between them, one would expect a genetic swamping of the Lake Wales Ridge population. The important question is what has caused a reduction in scutellation? Is the reduction related to difference in soil friability, microclimatic factors, founders principle, or genetic drift? Fosm HISTORY AND PAST DISTRIBUTION To date seven fossil species of Rhineura have been described: R. coloradoensis Cope, 1873; R. haeheri Baur, 1893; R. stembergi Walker, 1932; R. minuta Gilmore, 1938; and R. hibbardi, R. ambly- ceps, and R. wilsoni, all Taylor, 1951. These were all recovered from 204 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Oligocene deposits and distinguished on the basis of skull character- istics, except for R. coloradoensis which is known only from several dorsal vertebrae. Their combined range is northward from north- eastern Colorado through northwestern Nebraska to southwestern South Dakota and east-central Wyoming. Rhineura coloradoensis is Early Oligocene; all the others are Middle Oligocene except R. hateheri which extends into the Late Oligocene. All the species are currently recognized, although their status has changed. Vanzolini ( 1951) erected the genus Pseudorhineura for minuta. Gilmore ( 1938) suggested that coloradoensis is con- specific with hatcheri, and Brattstrom ( 1958) implied that amblyceps is also conspecific with hatcheri. As all the nominate species are temporally and geographically close, some could well be conspecific, but the status of the fossil species must remain questionable until we have more detailed knowledge of the skeletal variation in recent Rhineura. No fossil remains of Rhineura have been discovered in any Pliocene or Early Pleistocene sediments. Kluge (pers. comm., 1967) has examined a Rhineura skull from a Miocene deposit of the High Plains. Holman ( 1958, 1959, 1962) reported R. #oridana remains from Florida Rancholabrean deposits in Arredondo, Alachua County; Williston, Levy County; Reddick, Marion County; Saber-tooth Cave, Citrus County. These localities are all within the present range of R. Boridana and add no information on Pleistocene dispersal. Before erecting a hypothetical history for Rhineura, it seems advisable to review the geologic and paleoclimatic history of the plains and the Southeast. A gradual cooling trend was evident throughout the entire Tertiary and was climaxed by the Pleistocene glaciation. On the plains the subtropical conditions of the Oligocene slowly shifted to a warm temperate climate in the Pliocene; within this trend other shorter climatic cycles existed, hot-dry and cool- moist periods. The climate of the Southeast and peninsular Florida ranged from tropical in the Oligocene to subtropical in the Late Pliocene. The Pleistocene climate fluctuated markedly; the Southeast was temperate during the glacial epochs and subtropical-tropical in the interglacials, and the plains were subarctic-cool temperate and warm temperature, respectively ( Dorf, 1959; Hibbard, 1960). From this climatic data we can conclude that Rhineura could have lived in the Southeast since the Oligocene and that it probably remained on the High Plains until the end of the Pliocene. 1968 ZUG: VARIATION IN RHINEURA FLORIDANA 205 The plains have been stable since the Oligocene with no uplift or subsidence, and they, particularly the western part, have acted as a depositional site for the erosion of the developing Cordilleras. The history of the Southeast and Florida cannot be as easily sum- marized because of the various ages assigned to the marine terraces. Cooke ( 1945) and MacNeill ( 1950) correlate them with the different Pleistocene interglacial sea levels, whereas Alt and Brooks ( 1965) suggest that the terraces range from Miocene to Recent in age. Basically the issue rests on the maximum elevation of sea level during the Pleistocene. Cooke ( 1945) and MacNeill ( 1950) obviously adhere to a 250-300 feet elevation, Alt and Brooks ( 1965) to a 25-30 feet elevation. The only point of agreement is that the highest tel'race is the oldest and the successively lower terrances are younger or more recent. Cooke ( 1945) recognized seven terraces ( 270 ft., Aftonian; 215 and 170 ft., Yarmouth; 100, 70, and 42 ft., Sangamon; 25 ft., Mid-Wisconsin), Alt and Brooks ( 1965) Eve terraces ( 215- 250 ft., Upper Miocene; 90-100 ft., Pliocene; 45-55 ft., and 70-80 ft., Upper Pliocene or Lower Pleistocene; 25-30 ft., Pleistocene). As I have no means for dating the terraces on biological evidence and believe in higher sea levels, I am inclined to follow the general land patterns proposed by Cooke or MacNeill. The basic problem with their terraces is their reliance on topographic maps for locating marine terraces ( White, 1958). Also, although I am not following the terrace ages of Alt and Brooks, their article convinces me of the existence of five terraces and especially of the highest terrace which has been questioned. Rhineura probably existed in the High Plains until the Pliocene. If is was not extirpated from the plains by the recurring arid condi- tions of the Middle and Late Pliocene, it would have certainly been eliminated by the subarctic conditions of the Nebraskan glaciation. Migration of the worm lizard may have occurred in the Miocene. The development of grasslands and savannahs would have made the Gulf Coast Corridor a suitable route, and even if this area contained a more mesic flora, no reason exists why it would not have been passable as R. #oridana lives today in a humid or moist sub- humid environment. It could not have extended into the peninsular at this time, for the Trail-Lake Wales Ridge was probably a sand bar during the Late Miocene. Penetration of Rhineura into the Southeast during the Miocene or even in the Pliocene ( Auffenberg and Milstead, 1965) would have 206 BULLETIN FLORIDA STATE MUSEUM Vol. 12 \0 M : A . . B M M C s D M M 5,4 a j 814;EP4- 64*.S E . F ''~]~'654 1968 ZUG: VARIATION IN RHINEURA FLORIDANA 207 permitted it to extend into the emerging peninsula of the Pliocene. It is likely that by Middle Pliocene the sea level was similar to that of today ( Webb and Tessman, 1967), thus the peninsula probably had numerous habitats as today and Rhineura could have occupied a number of them. By the end of the Pliocene Rhineura should have been well established in Florida. The Aftonian inundation split Rhineura into an insular and a mainland population ( Fig.6).I assume this island was part of what is now the Lake Wales Ridge and was covered mainly with scrub. On the other hand the mainland probably offered a variety of suit- able habitats. The initial isolation in the scrub presumably lead the insular population to adapt to the scrub environment, which the mainland population was less likely to do with the several different habitats available. In the succeeding glacial period, the Kansan, the sea level dropped and both populations probably spread into the recently flooded areas ( Fig. 6). If the insular population had become so adapted to a scrub situation that it was unable to occupy other habitats, its ex- pansi6n would have been drastically limited in relation to that of a more tolerant mainland population. By the end of the glacial period the mainland population probably occupied much of north-central Florida and the insular population the south-central portion. Inter- gradation of the two populations may have occurred in the region of northern Lake County. The Yarmouth interglacial rise in sea level submerged the area of intergradation and again split the peninsular population. ( Fig. 6). A group of southern populations remained on the three parallel island chains. The small size and sandy nature of this area suggest that scrub may have been the only available habitat and that selec- tive pressures on all these populations would have been similar to those on the preceding interglacial population of this region. The northern population was probably restricted to the mainland and Figure 6. Hypothetical distributions and dispersal of Rhineura in Florida during the Pliocene and Pleistocene. Solid line = actual coast line; dotted line = present coast line; diagonal hatching = initial peninsular range of Rhineura; M, mainland population; S, insular population; arrows, direction of dispersal. Land forms based on Cooke ( 1945) and Mac- Neill ( 1950). A. Late Pliocene, B. Aftonian Interglacial, C. Nebraskan Glacial, D. Yarmouth Interglacial, E. Kansan Glacial, F. Sangamon Interglacial. 208 BULLETIN FLORIDA STATE MUSEUM Vol. 12 the Marion County islands. The Hernando-Pasco island harbored either intergrades or most likely representatives of the northern mainland population. A similar regional expansion of Rhineura occurred during the Illinoian glacial period. The spread of the combined southern popu- lation was probably no greater than in the previous glacial period and may have been less. If this were the case, the northern popu- lation would have extended farther south and consequently also have pushed or shifted the area of intergradation southward. With the Yarmouth rise of sea level, a portion of the northern population was probably left on the large Polk County Island and restricted to the scrub ridges of the southern part of the island, thereby preventing a northern expansion of the southern population, On the other hand the adaptation of the southern population to the scrub habitat and the barriers of marshy lowlands in northern Polk County would have prevented a southern expansion of the northern population. Thus the present area of intergradation was probably established at this time. The other islands would have had representatives of the northern population which joined with one another as the land emerged during the Wisconsin glaciation. Thus from the Wisconsin onward the range of R. ~oridana was very similar to its range at the present. Two main assumptions are made in the postulated history of Rhineura to explain the present pattern of geographic variation. First a peninsular population was divided into a mainland and an insular population by the Aftonian rise in sea level. Second the insular population lived in a scrub or scrublike habitat, and adapta- tion of this population restricted it to this type of habitat. Similar histories have been postulated for other endemic reptiles in Florida: Ophisaurus compressus C McConkey, 1954), Stilosoma extenuatum ( Highton, 1956), Neoseps re!/noldsi ( Telford, 1959), Eumeces egregius C Mount, 1965), and TantiNa relicta ( Telford, 1966). Sceloporus woodi probably shares a similar history ( Carr, 1940; Neill, 1957; Telford, 1965). The present ranges of these species and their restriction to scrub, sand hill, and/or xeric hammock associations strongly supports the second of my two basic assumptions, that the initial insular population was isolated in a scrub or scrub- like habitat to which it became restricted by adaptation. The first assumption, that Rhineura was divided into an insular and a mainland population by the Aftonian Sea, is somewhat sup- 1968 ZUG: VARIATION IN RHINEURA FLORIDANA 209 ported by distributional patterns of 0. compressus, E. egregius, T. relicta, and S. woodi. Increased knowledge of Floridian Pliocene faunas suggests that many reptile species had reached Florida by the Middle Pliocene ( Auffenberg, 1963; Auffenberg and Milstead, 1956). Two sets of facts can be marshalled to support the assumption that the initial separations and isolation occurred during the Aftonian interglacial rise in sea level The first and best evidence is that three species have differentiated populations (R. #oridana, E. e. lividus, and T. r. relicta) on the Lake Wales Ridge. Secondly and a bit cir- cuitous, all the endemic squamate genera except Stilosoma are found on the Lake Wales Ridge, and only in Stilosoma extenuatum is a different pattern of geographic variation discerned. On these grounds I believe that the ancestors of N. reynoldsi, E. egregius, 0. compres- sus, S. woodi, and T. relicta were also present on the Polk County island in the Aftonian interglacial sea. Only for S. extenuatum is isolation in. a later interglacial period required to explain its present subspecific distribution: S. e. extenuatum on the Polk County archi- pelago, S. e. arenicola on the Hernando-Pasco island, and S. e. mul- tistictum on the Alachua County islands. The succeeding inundation of Florida would probably have reinforced the initial differentiation of the populations of these species, and if their present distributions match the Sangamon marine terraces most closely, it is because this is the area of continuous scrub and sand-hill associations, the only area in which these species can survive. SPECIMENS EXAMINED FLORIDA: Alachua Co., UMMZ 56601-02, 57876; Gainesville, UF 268,682,2074, 2193, 2725, 2865, 2902, 8298-94, 5999, 7077, 7997, 8482, 9060, 11134, 14416, 21622; UMMZ 64170, 98747, 100684-5; 8.8 mi S of Gaines- ville, UFT 1094; 2 mi N of Gainesville, UF 3670; SE of Gainesville, UF 8895; 8.5 mi W of Gainesville, UF 14201; 8 mi SW of Gainesville FMNH 51989-48; west side of Lake Santa Fe, UMMZ 100688; Micanopy, USNM 8202,4983 ( 11); Waldo, USNM 18964, 26182. Clay Co., Gold Head Branch State Park, UF 8666. Columbia Co., Lake City, USNM 22818. Flagler Co., Bunnell, UF 12084. Hernando Co., Bayport, UF 2770; Annutheliga Hammock nr Brooksville, USNM 142478. Highlands Co., Avon Park, UF 7218; nr Archbold Biological Station, Hicoria, AMNH 65659-60, 66278; Lake Placid, AMNH 64496, FMNH 59198; Sebring, USNM 50960. Hillsborough Co., USNM 58742-44; Tampa, AMNH 70872. 210 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Lake Co., UMMZ 56538; Eustis, USNM 19998; Fruitland Park, USNM 64018, 69655; nr Lake Joanna, USNM 69656; Leesburg, CM 16808, UMMZ 77490, 97448; Mt. Dora, UF 14686; Tavares, UF 8081, USNM 75238; Umatilla, USNM 80299. Levy Co., 6 mi S of Bronson, UF 14687; Chiefland, USNM 185472; Sea- horse Key UF/SKRC 7,22. Manatee Co., 5 mi W of Bradenton, CM 26707; Manatee, USNM 8371. Marion Co., CM 1117-28, FMNH 48885-400, 48402-4, 48407-9; Candler, UMMZ 46938; Dunnellon, UF 11135;. Eureka, AMNH 8811-12, UMMZ 48608; Ocala, FMNH 28454-56; Silver Spring, FMNH 85136-37. Orange Co., MCZ 6808, 6851, 8331; Fairville, CM 23448; Orlando, ANSP 18479-81, FMNH 2016(5), UF 7954, USNM 51075, 88819. Pasco Co., Argo, ANSP 13478; Odessa, AMNH 64490-91. Pinellas Co., Clearwater, ANSP uncataloged, USNM 10787; west coast of Tampa Bay, Wieden Mound, FMNH 55498; Tarpon Springs, ANSP 4557, 15237-43. Polk Co., Auburndale, USNM 59412, 59417, 60501-2; Lakeland, FMNH 3078, 72889-92; Phantom Grove, FMNH 27701; Winter Haven, AMNH 5724-25, 69705, UF 2994; Blue Lake nr Winter Haven, UF 8000; 4 mi NW of Winter Haven, UF 14688, 14689-1 & 2. Putnam Co., UF Conservation Reserve nr Welaka, UMMZ 100726. Volusia Co., MCZ 14089-91; Port Orange, MCZ 5125; Volusia, ANSP 18466-76. GEORGIA: MCZ 4655. LITERATURE CITED Alt, D. and H. K, Brooks. 1965. Age of the Florida marine terraces. Jour. Geology, 73:406-411. Auffenberg, W. 1968. The fossil snakes of Florida. Tulane Stud. ZooL, 10: 131- 216. --- and W. W. Milstead. 1965. Reptiles in the Quaternary of North America, pp. 557-568, in Wright, Jr., H. E. and D. G. Frey, ed., The Quarternary of the United States, Princeton Univ. Press. Baird, S. F. 1859. Description of new genera and species of North American lizards in the Museum of the Smithsonian Institution. Proc. Acad. Nat. Sci. Philad., 1858:253-256. Baur, G. 1898. The discovery of Miocene amphisbaenians. Amen Nat., 27:998- 999. Brattstrom, B. H. 1958. Two Oligocene lizards. Herpetologica, 14:48-44. Burt, C. E. 1987. The lizards of southeastern United States. Trans. Kansas Acad. Sci., 40:349-366. Carr, A. 1949. Notes on eggs and young of the lizard Rhineura floridana, Copeia, 8:1-118. ---. 1950. The herpetology of Florida. Univ. Florida Publ., Biol, Sci. Ser. 1949, 1:77. Cooke, C. W. 1945. Geology of Florida. Florida Geol. Surv. Geol. Bull., 29: 1- 899. Cope, E. D. 1861. Remarks on reptiles. Proc. Acad. Nat. Sci. Philad., 1861:75. 1968 ZUG: VARIATION IN RHINEURA FLORIDANA 211 -. 1878. Synopsis of new Vertebrata from the Tertiary Of Colorado. U. S. Geol, Surv. Territories, 1878: 16-19. Dorf, E. 1959. Climatic changes of the past and present. Contrib. Mus. Paleont. Univ. Mich.; 13: 181-210. Eigenman, C. H. 1902. The eyes of Rhineura #oridand. Proc. Wash. Acad. Sci., vol 4: 583-548. Fowler, E. D. et al. 1927. Soil survey of Polk County, Florida. U. S. Dept. Agric. Bur. Chem. Soils, ser. 1927, No. 89, 1-89. Gans, C. and A. A. Alexander. 1962. Studies on amphisbaenids ( Amphisbaenia, Reptilia). 2. On the amphisbaenids of the Antilles. Bull. Mus. Comp. Zool., 128:67-158. Carman, S. 1883. On certain reptiles from Brazil and Florida. Sci. Obs., 4:47-48. Gilmore, C. W. 1928. Fossil lizards of North America. Mem. Nat. Acad. Sci., 22: 1-201. --. 1938. Descriptions of new and little-lmown fossil lizards from North America. Proc. U. S. Nat. Mus., 86: 11-26. Hibbard, C. W. 1960. An interpretation of Pliocene and Pleistocene climates in North America. Ann. Report Mich. Acad. Sci. Arts Letters, 62:5-30. Highton, R. 1956. Systematic and variation of the endemic Florida snake genus Stilosoma. Bull. Florida State Mus., 1:73-96. Holman, J. A. 1958. The Pleistocene herpetofauna of Saber-tooth Cave, Citrus County, Florida. Copeia, 1958:276-280. --. 1959. Amphibians and reptiles from the Pleistocene ( Illinoian) of Wil- liston, Florida. Copeia, 1959:96-102. -- 1962. Additional records of Florida Pleistocene amphibians and reptiles. Herpetologica, 18:115-119. MacNeill, F. S. 1950. Pleistocene shore lines in Florida and Georgia. Geol. Surv. Prof. Paper, 221-F:95-107. McConkey, E. H. 1954. A systematic study of the North American lizards of the genus Ophisaurus. Amen Midl. Nat., 51:183-171. Mount, R. H. 1965. Variation and systematics of the scincoid lizard, Eumeces egregius C Baird). Bull. Florida State Mus., 9: 183-218. Myers, C. W. and S. R. Telford, Jr. 1965. Foods of Neoseps, the Florida sand skink. Quart. J. Florida Acad. Sci., 28: 190-194. Neill, W. T. 1951. The eyes of the worm lizard, and notes on the habits of the species. Copeia, 1951: 177-178. ---. 1957. Historical biogeographical of present-day Florida. Bull. Florida State Mus., 2: 175-220. Smith, H. M. 1946. Handbook of Lizards. Lizards of the United States and of Canada. Comstock Publ. Co., Ithaca, i-xi, 557 p. Taylor, E. H. 1951. C6ncerning Oligocene amphisbaenid reptiles. Univ. Kansas Sci. Bull., 84:521-579. Telford, Jr., S. It. 1955. Notes on an expectionally large worm lizard, Rhineura floridana, Copeia, 1955:258-259. --. 1959. A study of the sand skink, Neoseps regnoldsi Stejneger. Copeia, 1959:110-119. ---. 1965. Some biogeographical aspects of the Florida herpetofauna. Acta Herpt. Japonica, 2: 16-21. --. 1966. Variati6n among the southeastern crowned snakes, genus Tantilla. 212 BULLETIN FLORIDA STATE MUSEUM Vol. 12 Bull. Florida State Museum, 10:261-804. Vanzolini, P. E. 1951. A systematic arrangement of the family Amphisbaenidae ( Sauria). Herpetologica, 7: 118-123. Walker, M. V. 1932. A new burrowing lizard for the Oligocene of Central Wyoming. Trans. Kansas Acad. Sci., 85:224-281. Webb, S. D. and N. Tessman. 1967. Vertebrate evidence of a low sea level in the Middle Pliocene. Science, 156: 379: White, W. A. 1958. Some geomorphic features of central peninsular Florida. Florida Geol. Surv. Geol. Bull., 41: 1-92. Contributions to the BULLETIN OF THE FLORIDA STATE 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 Editorial Board. The BULLETIN is distributed worldwide through institutional subscriptions and _ exchanges only. It is considered the responsibility of the author to 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. Inst. 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 and a carbon must be submitted. Tables and 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 maps and photographs, should be referred to as "figures." All illustrations are reduced to a maximum of 4-1/4 by 7-1/8 inches. Size scales, 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 BULLET[N. Spell out in full the titles of non-English serials and places of publication. Footnote material should be kept to a minimum. However, provide copy for a footnote detailing the title, affiliations, 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 description) 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; indicate 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 to 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 does not replace the usual conclusions or summary sections. It may also serve as copy for the abstracting services. Manuscripts and all editorial matters should be addressed to: Managing Editor of the BULLERN Florida State Museum Seagle Building Gainesville. Florida