Allen Press and Herpetologists' League are collaborating with JSTOR to digitize, preserve and extend access to Herpetologica. http://www.jstor.org Herpetologists' League Allozyme Variation in Populations of Bothrops asper (Serpentes: Viperidae) in Costa Rica Author(s): Mahmood Sasa and Ramiro Barrantes Source: Herpetologica, Vol. 54, No. 4 (Dec., 1998), pp. 462-469 Published by: on behalf of the Allen Press Herpetologists' League Stable URL: http://www.jstor.org/stable/3893440 Accessed: 30-07-2015 17:04 UTC REFERENCES Linked references are available on JSTOR for this article: http://www.jstor.org/stable/3893440?seq=1&cid=pdf-reference#references_tab_contents You may need to log in to JSTOR to access the linked references. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. This content downloaded from 163.178.208.28 on Thu, 30 Jul 2015 17:04:39 UTC All use subject to JSTOR Terms and Conditions Herpetologica, 54(4), 1998, 462-469 ?) 1998 by The Herpetologists' League, Inc. ALLOZYME VARIATION IN POPULATIONS OF BOTHROPS ASPER (SERPENTES: VIPERIDAE) IN COSTA RICA MAHMOOD SASA' 3 AND RAMIRO BARRANTES2 'Instituto Clodomiro Picado and 2Instituto de Investigaciones en Salud, Universidad de Costa Rica, San Jose', Costa Rica ABSTRACT: We investigated allozyme variation from Costa Rican populations of the fer-de-lance, Bothrops asper Blood samples were. taken from a total of 100 specimens from six localities repre- senting three major physiographic regions: Atlantic versant, Central Pacific lowlands, and South- western Pacific lowlands. Five of 16 protein coding loci (phosphogluconate mutase, isocitrate de- hydrogenase, glucose 6-phosphate isomerase, malate dehydrogenase, and carbonic anhydrase-1) were polymorphic. Average heterozygosity for these loci varied between 2.9% and 5.5%. Genetic distances among samples from localities from the same physiographic region were not significantly different from zero. Pairwise comparisons between regions also resulted in low estimated distances. Analysis of population structure suggests high levels of gene flow among populations. Rates of evolution of venom and morphology previously reported for populations of this species seem to be uncoupled from rates of allozyme variation. Key words: Pitviper; Bothrops asper; Allozymes; Genetic variation; Population genetics; Costa Rica THE lancehead, Bothrops asper (Gar- man), is a common pitviper of Central America and is considered the most im- portant venomous snake in the region from the standpoint of human morbidity and mortality (Bolafios, 1984; Hardy, 1994). It has an extensive distribution ranging from the Atlantic versant of Ta- maulipas (Mexico), throughout the Carib- bean lowlands of Central America, to northeastern Colombia and Venezuela (Campbell and Lamar, 1989). Aside from a disjunct population occurring in western Chiapas and Guatemala, the distribution of this species is continuous from the south-central Pacific coast of Costa Rica to the Pacific versant of the Colombian and Ecuadorean Andes (Campbell and La- mar, 1989; Peters and Orejas-Miranda, 1970). In Central America, B. asper can be found from 0-1200 m altitude, al- though it is uncommon above 800 m. While the species is found in a great va- riety of habitats, it is particularly abundant in and well adapted to areas disturbed by humans (Greene and Campbell, 1992). At many localities, B. asper is the most com- PRESENT ADDRESS: Department of Biology, Uni- versity of Texas at Arlington, Arlington, TX 76019, USA. e-mail: SASA@UTA.edu mon species of snake (Campbell and La- mar, 1989). The geographic variation of morpholog- ical characters in this species is remarkable and undoubtly is the major factor respon- sible for the taxonomic chaos involving populations of B. asper, B. atrox, and other relatives from South America (Campbell and Lamar, 1992; Markezich and Taphorn, 1993). In Middle America, univariate and multivariate analyses of several morpho- logical characters may distinguish among populations of B. asper (Sasa, 1996). In Costa Rica, specimens from the Atlantic slopes differ from those of the Pacific low- lands in body size, litter size, and some scutellation characters (Solorzano and Cerdas, 1989; Tejeira-Rodriguez, 1993). Toxicological variation also exists in popu- lations of B. asper, with implications for snakebite therapy. Noticeable variation among populations in Costa Rica was re- ported in the expression patterns of my- otoxin isoforms (Lomonte and Carmona, 1992; Moreno et al., 1988), the specific en- zymatic activities of coagulant proteinases (Arag6n and Gubensek, 1981), and the bi- ological activities of the toxins (Gutierrez et al., 1980). Thus, adult venoms from At- lantic populations are more hemorrhagic and myonecrotic, whereas those of the 462 This content downloaded from 163.178.208.28 on Thu, 30 Jul 2015 17:04:39 UTC All use subject to JSTOR Terms and Conditions December 1998] HERPETOLOGICA 463 Central Pacific lowlands are more proteo- lytic. Venom of both regions have similar lethality, edema forming activity, and he- molytic effect (Gutierrez et al., 1980). Variation observed between snake pop- ulations might relate to overall genetic di- vergence (Arnold, 1988; Glenn et al., 1983; Murphy et al., 1987) or may be the result of the action of natural selection. For instance, venom variation probably re- sulted from natural selection for venom appropriate for feeding on local prey (Dal- try et al., 1996a). Aragon and Gubensek (1981) suggested that population isolation might account for the venom differences observed in B. as- per from Costa Rica. In this country, a central axis of mountain ranges runs north- west-southeast and divides the Atlantic and Pacific lowlands, apparently preclud- ing contact between populations located on each versant. In this study, we investi- gated whether the morphological and ven- om differences observed among popula- tions of B. asper in Costa Rica are corre- lated with genetic divergence among them. MATERIALS AND METHODS Specimens A sample of 100 specimens of Bothrops asper were collected for venom extraction between July 1991 and November 1992 and brought to the Serpentarium of the Instituto Clodomiro Picado, Universidad de Costa Rica. We sampled only adult snakes (snout-vent length -85 cm, both sexes), all from six localities of Costa Rican lowlands (Fig. 1). These localities are lo- cated in three physiographic regions. (1) Central Pacific Lowlands.-This area is located in the river basin of Grande de T arcoles and Pirn-is rivers, in the middle of the Pacific versant, and contains lands of moderate elevation (0-800 m) that are covered with Tropical Wet Forest and Tropical Moist Forest (Tosi, 1969). We sampled three localities in this area: Pur- iscal, Acosta, and Quebrada Ganado (Fig. 1). These localities are highly disturbed and all specimens collected came from grassland habitats. (2) Southeastern Pacific Lowlands.- The life zone of this region is Tropical Wet Forest (Tosi, 1969). The specimens col- lected from this region came from two lo- calities: Peninsula de Osa and Golfito. (3) Atlantic Versant. -Two life zones oc- cur in this region: Premontane Wet Forest and Tropical Moist Forest (Tosi, 1969). Snakes were collected at Penshurt and Si- quirres. Due to the proximity of these two localities, these specimens were consid- ered as a single geographical sample. Proteiii Electrophoresis We extracted a 1 ml blood sample from each individual by tail clipping or by car- diac puncture using a heparinized syringe. We centrifuged the blood samples at 4 C at 5000 rpm and then separated plasma and cellular fractions. The erythrocytes were washed three times in saline solution and were hemolyzed with 0.1% sodium ci- trate. After centrifugation, we stored blood plasma and the hemolyzed erythrocytes at -70 C; these were used later without modification. We performed horizontal starch gel electrophoresis (Murphy et al., 1990) for each locus using 12% (w/v) starch gels at 5 C. Transferrin and lactate dehydrogenase (Ldh) products were as- sayed using a 7% polyacrylamide gel at 5 C. The Ldh reaction mixture followed Schwantes (1973). Other histochemical staining procedures followed Harris and Hopkinson (1976) and Murphy et al. (1990). Enzymes and buffer systems are listed in Table 1. We considered electro- morphs homologous if they had the same mobility. Presumptive genotypes were as- signed to each individual based on the en- zyme banding pattern observed by eye and followed the criteria in Murphy and Crab- tree (1985). Data Analyses Allelic frequency data were analyzed us- ing BIOSYS-1 (Version 2.0: Swofford and Selander, 1981). The average locus hetero- zygosity (H) and percent of polymorphic loci (P) were used to measure genetic vari- ability estimates for each locality. We cal- culated the coefficients of genetic distance of Nei (1978) and Rogers (1972). Genetic This content downloaded from 163.178.208.28 on Thu, 30 Jul 2015 17:04:39 UTC All use subject to JSTOR Terms and Conditions 464 HERPETOLOGICA [Vol. 54, No. 4 Nicaragua Caribbean Sea Pacific Ocean 4 50km ! ~~~~~~85- 840 83, FIG. 1-Sampled localities of Bothrops asper in Costa Rica. Contour line represents elevations ?1000 m. Locality numbers as follow: (1) Puriscal, (2) Acosta, (3) Quebrada Ganado, (4) Peninsula de Osa, (5) Golfito, and (6) Penshurt-Siquirres. diversity within and among populations was analyzed using Wright's (1965) F-sta- tistics, according to the methods outlined by Weir and Cockerham (1984). We ex- amined two levels of subdivision: physio- graphic region and localities within each region. Estimates of the variance of F-sta- tistics were provided by a jackknife pro- cedure (Reynolds et al., 1983). Gene flow was estimated from F-statistics and by Slatkin's (1981) private alleles procedure. This method is based on the strong de- pendence between the total level of gene flow among several populations and the average frequency of an allele conditioned on the number of populations in which this allele occurs (Slatkin and Barton, 1989). Analyses of spatial correlation of genetic divergence among populations were per- formed using Mantel correlations (Mantel, 1967). We used Rogers (1972) distance for these analyses, because Nei distances de- pend upon mutation rates. The matrix of geographical distances was constructed from the linear distance following the al- titudinal isocline between localities. In this way, the geographical barrier between the Atlantic and Pacific populations in lower Central America is taken into account. Comparisons with the random distribution of coefficients obtained after 5000 per- This content downloaded from 163.178.208.28 on Thu, 30 Jul 2015 17:04:39 UTC All use subject to JSTOR Terms and Conditions December 1998] HERPETOLOGICA 465 TABLE 1.-Enzyme and electrophoretic conditions used in determining the allozyme characteristics. Prefixes "s" and "m" indicate supernatant and mitochondrial enzymes respectively. Enzyme nomenclature numbers are those of International Union of Biochemistry and Nomenclature Committee (1984). Locus assignments for enzymes follow Murphy and Crabtree (1985). Enzyme system Commission number Locus Assay* conditions Acid phosphatase 3.1.3.2 Acp-1 A, B Carbonic anhydrase 4.2.1.1 Ca-1 D Ca-2 D Esterase** 3.1.1.1 Est-A E, B Est-C E, F Est-D B, E, F Glucose-6-phosphate isomerase 5.3.1.9 Gpi-A B Isocitrate dehydrogenase 1.1.1.42 slcdh-A A, C L-lactate dehydrogenase 1.1.1.27 Ldh-A A, B, H Ldh-B B, H Malate dehydrogenase 1.1.1.37 sMdh-A A, F mMdh-A A, F Phosphoglucomutase 2.7.5.1 Pgm-A C, F Tripeptidase*** 3.4.11.13 Pep-5 E, G Superoxide dismutase 1.15.1.1 sSod-A C Tranferrin Tf H * Conditions: (A) 0.005 M DI-Histidine pH = 7 (gel), 0.41 M Sodium citrate (electrode) pH = 7.0, 5 v/cm, 17 h. (B) 0.1 M Tris, 0.028 M Citric acid, pH = 7.4, 5 v/cm, 12 h. (C) 0.25 M NaH2PO4, 0.15 M Citric acid pH = 5.9, 5 v/cm 12 h. (D) 0.9 M Tris, 0.5 M Boric acid, 0.02 M EDTA pH = 8.6, 5 v/cm, 18 h. (E) 0.1 M Tris, 0.1 M Maleic acid pH = 7.2, 5 v/cm 17 h. (F) 0.1 M Tris, 0.1 M Maleic acid, 0.01 M EDTA, 0.01 M MgCl2, pH = 7.4, 18 h at 35 mA/gel. (G) 0.1 M Tris, 0.1 M NaH2PO4, pH = 7.4, 5 v/cm 18 h. (H) 0.04 M Tris, 0.18 M Glycine pH = 8.5, gel 7% acrylamide 4 h at 40 mA/gel. ** Substrate used: Naphthyl acetate (Est-A, Est-C) and Methyl umbelliferyl acetate (Est-D). *** Substrate used: L-leucyl-glycyl-glycine. mutations of the genetic distance matrices determined the significance of the ob- served correlation coefficients. RESULTS Of 16 loci assayed for 100 specimens (Table 1), 11 were monomorphic in all in- dividuals studied: mMdh-A, Ldh-A, Ldh- B, Ca-2, Est-A, Est-C, Est-D, Pep-5, sSod- A, Acp-1, and Tf. Five loci were polymor- phic, with the frequency of the common allele '0.99 (Table 2). One sample (Que- brada Ganado) had a rare allele (c) in s- Icdh-A (frequency of 0.026). The percent- age of polymorphic loci varied between lo- calities, being highest in snakes from Que- brada Ganado (P = 30.8%) and lowest in the Peninsula de Osa sample (P = 7.7%) where the only variable locus was Gpi-A. Few alleles per locus (A ? SD = 1.21 ? 0.09) were found in these enzymatic sys- tems in B. asper Low levels of geographic variation were observed in allelic frequencies of polymor- phic loci. The Gpi-A common allele ap- peared fixed for the Acosta sample, per- haps due to the small sample size at that locality. Phosphoglucomutase (Pgm-A) was polymorphic only in samples from Golfito and Penshurt-Siquirres. In these localities, the isocitrate dehydrogenase fast electro- morph was fixed. Allelic frequencies were in Hardy-Weinberg equilibrium at all loci, except the mMdh-A in the Atlantic sample (X- = 23, P < 0.0001) which exhibited a deficiency of heterozygotes. The mean lo- cus heterozygosity (H) did not differ sig- nificantly between populations (F[, 72 0.05] = 0.19, P = 0.963), ranging from 0.029 in the Quebrada Ganado sample to 0.055 in snakes from Golfito (Table 2). Genetic identity (Nei, 1978) ranged from 0.995-1.000 among sampled locali- ties. Genetic distances within physiograph- ic regions were not significantly different from zero; the pairwise comparisons be- tween samples from those regions resulted in low estimated distances (Table 3). Nei's distance estimates the accumulated num- ber of codon substitutions per locus since the time of divergence between the pop- ulations. Based on our sample of 16 loci, on average the population of B. asper had only about one codon substitution per 250 loci. These values suggested a recent di- This content downloaded from 163.178.208.28 on Thu, 30 Jul 2015 17:04:39 UTC All use subject to JSTOR Terms and Conditions 466 HERPETOLOGICA [Vol. 54, No. 4 TABLE 2.-Summary of the allelic frequencies for five polymorphic loci in Bothrops asper for several localities of Costa Rica. Numerical designations for each population are as follow: 1 = Puriscal, 2 = Acosta, 3 = Quebrada Ganado, 4 = Peninsula de Osa, 5 = Golfito, 6 = Penshurt-Siquirres. Sample size for each locality is in parentheses. Locality Locus 1 2 3 4 5 6 (allele) (n = 15) (n = 10) (n = 20) (n = 10) (n = 22) (i 23) SMdh-A (a) 1.000 1.000 1.000 1.000 0.972 0.957 (b) 0.028 0.043 Ca-I (a) 0.962 0.929 0.952 1.000 1.000 1.000 (b) 0.038 0.071 0.048 Gpi-A (a) 0.955 1.000 0.944 0.688 0.800 0.900 (b) 0.045 0.056 0.313 0.200 0.100 sldh-A (a) 0.846 0.833 0.913 1.000 0.921 1.000 (b) 0.154 0.167 0.087 0.053 (c) 0.026 Pgm-1 (a) 1.000 1.000 1.000 1.000 0.947 0.881 (b) 0.053 0.119 % loci polymorphic* 23.1 15.4 23.1 7.7 30.8 23.1 Alleles per locus 1.2 1.2 1.2 1.1 1.4 1.2 (H)** 0.037 0.037 0.029 0.048 0.055 0.034 (+ SE) (0.024) (0.027) (0.016) (0.048) (0.032) (0.023) * 0.99 criterion. ** Mean locus heterozygosity by direct count. vergence of these populations and, or, a high degree of gene flow among them. Low F-values and their variance esti- mators were found for the five polymor- phic loci (Table 4). A slight negative value of the fixation index (Fst) was found for sMdh-A (Table 4), but Fst was so small that significant population subdivision could not be detected (Slatkin and Barton, 1989). The graphical analysis of rare alleles also showed a trend of high gene flow among the six sampled populations, as im- plied from the J-shaped relationship be- tween conditional average allelic frequen- cy and the occupancy number. DISCUSSION Measures of genetic variation presented here for populations of Bothrops asper are similar to those reported for other reptili- an species based on allozyme data (Nevo, 1978). The results from the analysis of rare alleles and the low values for F-statistics (Table 4) were consistent with the hypoth- TABLE 3.-Unbiased genetic distances of Nei (1978: above the diagonal) and Rogers (1972: below the diag- onal) genetic distance coefficients for all pairwise combinations of populations of Bothrops asper from Costa Rica. Numerical designations for each population are as in Table 2. Locality 1 2 3 4 5 6 Puriscal 0.000 0.000 0.006 0.001 0.002 Acosta 0.007 0.000 0.008 0.003 0.003 Quebrada Ganado 0.007 0.012 0.005 0.001 0.001 Peninsula de Osa 0.035 0.042 0.030 0.000 0.003 Golfito 0.028 0.035 0.023 0.020 0.000 Penshurt 0.031 0.039 0.026 0.029 0.019 This content downloaded from 163.178.208.28 on Thu, 30 Jul 2015 17:04:39 UTC All use subject to JSTOR Terms and Conditions December 1998] HERPETOLOGICA 467 TABLE 4.-Population F-statistics* for polymorphic loci in Costa Rican B. asper. Loci Fs, Fi, F,, Nm Ca-i 0.0007 -0.0181 -0.0188 342.2 clcdh-A 0.0308 -0.0529 -0.0865 8.0 Pgm-A 0.0446 -0.0251 -0.0738 5.3 sMdh-A -0.0201 0.3213 0.3347 12.1 Gpi-A 0.0714 -0.0930 -0.1770 3.3 Jackknife mean 0.0240 0.0263 -0.0042 Standard deviation 0.0160 0.0748 0.0884 * Estimates of Wright (1965) F-statistics follow Weir and Cockerham (1984). esis of high levels of gene flow among sam- pled geographic regions. These results are not totally unexpected when the extensive distribution of the species and its ability to adapt to human-disturbed areas are taken into consideration. The Continental Di- vide of Costa Rica is higher than the ele- vational limit of the species throughout most of the country, and, therefore, any region where substantial gene flow might be occurring between Atlantic and Pacific populations must be looked for extralimi- tally to Costa Rica. The dry-forest life zone on the Pacific versant of Nicaragua and northwestern Costa Rica is an effective barrier to the distribution of B. asper (Sasa, 1996). Thus, any gene flow between Atlantic and Pacific populations of this snake in Isthmian Central America must be occurring in Panam-a. The Cordillera de Talamanca in Costa Rica ranges southeast into the Chiriqui Province of Panama'. This mountain range is connected with Cordil- lera Central by highlands above 1000 m. The Cordillera Central extends eastward as far as Cocle Province in central Panama. Thus, contact between Atlantic and Pacific versant populations of B. asper in lower Central America is possible in central and eastern Panamai, where the lancehead is one of the most abundant snakes. Despite the great variation previously recorded for Bothrops asper in venom components, morphological characters, and natural history traits (Sasa, 1996), our results suggest that there might be little genetic heterogeneity among populations of this species in Costa Rica. Mantel tests (Manly, 1991) were used to evaluate the association between the linear geographic distance and the estimates of genetic dis- tance (Table 3). A significant correlation (r = 0.56, P = 0.012) was observed between Rogers' genetic distances and geographical distances. From these results, the slight genetic divergence among populations in- cluded in this study could be attributed to the effects of spatial correlation, or to an isolation by distance effect. In addition, morphological distances in samples of B. asper from different localities of Middle America are also correlated to spacial dis- tances (Sasa, 1996). Based upon the electrophoretic patterns and activities of L-aminoacid dehydroge- nase and other venom components, Jime- nez-Porras (1964) suggested that a distinct geographical pattern of venom variation occurs among samples from the Caribbean slope, Central, and South Pacific of Costa Rica, similar to those suggested by our allozyme data. The low values of genetic distances and the low number of loci sam- pled, however, make the concordance be- tween those data sets suspect. Our inter- pretation is that allozyme and venom vari- ation might not be correlated. The allo- zyme data do not reveal fixed differences among geographic localities, thus making it difficult to interpret the genetic basis for the observed phenotypic diversity. The independent evolution of tissue al- lozymes from venom characters was re- ported previously for Crotalus s. scutula- tus, where slight genetic differences, in- ferred from allozymes, existed between populations with different venom types (Wilkinson et al., 1991). In this species, the uncoupled rates of evolution of these characters may reflect geographically based selection on venom proteins, as op- posed to the neutral evolution of allozyme characters (Kimura, 1982). In a recent study, Daltry et al. (1996a,b) showed that This content downloaded from 163.178.208.28 on Thu, 30 Jul 2015 17:04:39 UTC All use subject to JSTOR Terms and Conditions 468 HERPETOLOGICA [Vol. 54, No. 4 venom variation among localities of the pitviper Calloselasna rhodostoma is asso- ciated with its diet and is unrelated to in- traspecific phylogenetic relationships and, or, current gene flow. Consequently, geo- graphical variation in venom composition might reflect directional selection for feed- ing in local prey (Daltry et al., 1996a). If strong selection is acting on venom components of Bothrops asper, then these characters could be expected to change faster than allozymes, consistent with our results. Nevertheless, the local selection hypothesis should be invoked with caution, because it is still unclear if geographical variation in diet occurs in this species. In summary, geographical variation in venom and in moxphological characters previously reported for Bothrops asper in three regions of Costa Rica may suggest that restricted gene flow occurs among them, or that their populations were in- dependently derived from a common source area. However, allozyme data showed high genetic similarity among pop- ulations of Bothrops asper in Costa Rica, and high gene flow apparently prevents further differentiation. Our results suggest that there is independent evolution of tis- sue protein characters from venom and morphological characters in this species that may be associated with local selective differences RESUMEN En Costa Rica, gran diversidad en car- acteres morfologicos, toxicologicos y de historia natural han sido previamente re- portados entre poblaciones de la serpiente Bothrops asper Con el fin de investigar la relacion genetica de esas diferencias, se evaluo la variacion isozimica entre pobla- ciones de Bothrops asper en ese pais. Se tomaron muestras de sangre de un total de 100 individuos provenientes de seis local- idades representando a tres regiones fisio- graficas: Vertiente Caribe, Pacifico Central y Pacffico Sur. De los 16 loci estudiados, cinco (Pgm-A, Icdh-A, Gpi-A, sMdh-A) re- sultaron polimorficos. El porcentaje de heterozigosis promedio por poblacion var- io entre 2.9 y 5.5%. Las distancias gene- ticas estimadas entre poblaciones fueron muy bajas. El andilisis de la estructura g6n- ica indica que poblaciones de Bothrops as- per en Costa Rica poseen alto flujo genico y gran homogeneidad genetica entre ellas. Nuestros anrlisis sugieren evolucion inde- pendiente entre los caracteres geneticos y fenotipicos observados en estas pobla- ciones. Se sugiere que la gran diversidad en veneno y morfologia en esta especie pueda estar asociada con diferencias de se- leccion a nivel local. Acknowledgments.-We are especially grateful to J. Lobo for his valuable advice with the electropho- retic techniques and to R. Aguilar and G. Vargas for their help in the laboratory. A. Solorzano, R. Aymer- ich, D. Chac6n, and G. Serrano helped us during collection of specimens and blood collection. For their critical remarks, we thank P. Phillips, N. John- son, P. Chippindale, and J. Campbell. M. Sasa is par- ticularly indebted to C. Amaya, F. Cohelo, and C. Code?o (University of Texas at Arlington) for their cooperation and helping in programming the Mantel procedure, and to J. M. Gutierrez and M. Springer (Universidad de Costa Rica) for their support during this research. This study was funded in part by a grant from Vicerrectorfa de Investigaci6n (Universidad de Costa Rica), and by Texas Advanced Research Pro- gram under grant 003656-001 to J. A. Campbell. LITERATURE CITED ARAGON, F., AND F. GUBENSEK. 1981. Bothrops asper venom from the Atlantic and Pacific zones of Costa Rica. Toxicon 19:797-805. ARNOLD, S. J. 1988. Quantitative genetics and selec- tion in natural populations: microevolution of ver- tebral numbers in the garter snake Thamnophis ele- gans. Pp. 619-636. In B. S. Weir, E. J. Eisen, M. J. Goodman, and G. Namkoong (Eds.), Proceed- ings of the Second Intemational Conference on Quantitative Genetics. Sinauer, Sunderland, Mas- sachusetts, U.S.A. BOLANOS, R. 1984. Serpientes, veneno y ofidismo en Centroame'rica. Editorial Universidad de Costa Rica. San Jose, Costa Rica. CAMPBELL, J. A., AND W W LAMAR. 1989. The Ven- omous Reptiles of Latin America. Cornell Univer- sity Press, Ithaca, New York, U.S.A. . 1992. Taxonomic status of miscellaneous neotropical viperids, with the description of a new genus. Occasional Papers, Museum of Texas Tech- nological University 153:1-31. DALTRY, J. C., G. PONNUDURAI, C. K. SHIN, N-H. TAN, R. S. THORPE, AND W WUSTER. 1996a. Elec- trophoretic profiles and biological activities in trhe venom of the Malayan pitviper (Calloselasma rho- dostoma). Toxicon 34:67-79. DALTRY, J. C., W WUSTER, AND R. S. THORPE. 1996b. Diet and snake venom evolution. Nature 379:537-540. GLENN, J. L., R. C. STRAIGHT, M. C. WOLFE, AND D. L. HARDY. 1983. Geographical variation in Cro- This content downloaded from 163.178.208.28 on Thu, 30 Jul 2015 17:04:39 UTC All use subject to JSTOR Terms and Conditions December 1998] HERPETOLOGICA 469 talus scutulatus scutulatus (Mojave rattlesnake) venom properties. Toxicon 21:119-130. GREENE, H., AND J. A. CAMPBELL. 1992. The future of pitvipers. Pp. 421-427. In J. A. Campbell, and E. D. Brodie, Jr. (Eds.), Biology of the Pitvipers. Selva Press, Tyler, Texas, U.S.A. GUTIERREZ, J. M., F. CHAVES, AND R. BoLANos. 1980. Estudio comparativo de ejemplares recien nacidos y adultos de Bothrops asper Revista de Biologfa Tropical 28:311-351. INTERNATIONAL UNION OF BIOCHEMISTRY. NOMEN- CLATURE COMMITEE. 1984. Enzyme Nomenclature 1984. Academic Press, Orlando, Florida, U.S.A. HARDY, D. L. 1994. Bothrops asper (Viperidae) snakebite and field researchers in Middle America. Biotropica 26:198-207. HARRIS, H., AND D. H. HOPKINSON. 1976. Handbook of enzyme electrophoresis in human genetics. North-Holland Publishing Company, Amsterdam, Holland. JIMENEZ-PORRAS, J. M. 1964. Venom proteins of the fer-de-lance Bothrops atrox from Costa Rica. Tox- icon 2:155-166. KIMURA, M. 1982. The neutral theory as a basis for understanding the mechanism of evolution and variation at the molecular level. Pp. 3-56. In M. Kimura (Ed.), Molecular Evolution, Protein Poly- morphism and the Neutral Theory. Springer-Ber- lang, Berlin, Germany. LOMONTE, B., AND E. CARMONA. 1992. Individual expression patterns of myotoxin isoforms in the venom of the snake Bothrops asper Comparative Biochemistry and Physiology 102B:325-329. MANLY, B. F. J. 1991. Randomization and Monte Carlo Methods in Biology. Chapman and Hall, London, U.K. MANTEL, N. 1967. The detection of desease cluster- ing and a generalized regression approach. Cancer Research 27:209-220. MARKEZICH, A. L., AND D. C. TAPHORN. 1993. A variational analysis of populations of Bothrops (Ser- pentes:Viperidae) from western Venezuela. Journal of Herpetology 27:248-254. MORENO, E., A. ALAPE, M. SANCHEZ, AND J. M. Gu- TIERREZ. 1988. A new method for the detection of phospholipase A2 variants: identification of iso- zymes in the venom of newborn and adult Both- rops asper (Terciopelo) snakes. Toxicon 26:363- 371. MURPHY, J. B., J. E. REHG, P. F. MADERSON, AND W B. MCCRADY. 1987. Scutellation and pigmentation defects in a laboratory colony of western diamond- back rattlesnakes (Crotalus atrox): mode of inher- itance. Herpetologica 43:292-300. MURPHY, R. W, AND C. B. CABTREE. 1985. Evolu- tionary aspects of isozyme patterns, number of loci and tissue-specific gene expression inthe praire rat- tlesnake Crotalus viridis viridis. Herpetologica 41: 451-470. MURPHY, R. W, J. W SITES, JR., D. G. BUTH, AND C. H. HAUFLER. 1990. Proteins I: Isozyme electro- phoresis. Pp. 45-126. In D. M. Hillis and C. Moritz (Eds.), Molecular Systematics. Sinauer Associates, Sunderland, Massachusetts, USA. NEI, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of in- dividuals. Genetics 89:225-233. NEVO, E. 1978. Genetic variation in natural popula- tions: patterns and theory. Theoretical Populational Biology 13:121-177. PETERS, J. A., AND B. OREJAS-MIRANDA. 1970. Cat- alogue of the Neotropical Squamata: Part I: Snakes. Bulletin of the U.S. National Museum 297:1-347. REYNOLDS, J., B. S. WEIR, AND C. C. COCKERHAM. 1983. Estimation of the coancestry coefficient: ba- sis for a short-term genetic distance. Genetics 105: 767-779. ROGERS, J. s. 1972. Measures of genetic similarity and genetic distance. Studies in Genetics Univer- sity of Texas Publications 7213:145-153. SASA, M. 1996. Morphological Variation in the Lance- head Snake Bothrops asper (Garman) from Middle America. M.S. Thesis, University of Texas at Ar- lington, Arlington, Texas, U.S.A. SCHWANTES, M. L. 1973. Lactate dehydrogenase iso- zyme patterns of thirteen species of snakes. Journal of Experimental Zoology 185:311-316. SLATKIN, M. 1981. Estimating levels of gene flow in natural populations. Genetics 99:323-335. SLATKIN, M., AND N. H. BARTON. 1989. A compari- son of three indirect methods for estimating aver- age levels of gene flow. Evolution 43:1349-1368. SOL6RZANO, A., AND L. CERDAS. 1989. Reproductive biology and distribution of the Terciopelo Bothrops asper Garman (Serpentes: Viperidae) in Costa Rica. Herpetologica 45:444-450. SWOFFORD, D. L., AND R. B. SELANDER. 1981. BIOSYS-1: a FORTRAN program for comprehen- sive analysis of electrophoretic data in population genetics and systematics. Journal of Heredity 72: 281-283. TEJEIRA-RODRIGUEZ, M. 1993. Comparaci6n de la morfologfa externa de Bothrops asper (Garman) Xenodon rabdocephalus (Wied) en el sector Pacf- fico de Costa Rica. M.S. Thesis, Universidad de Costa Rica, San Jos6, Costa Rica. TOSI, J. A. 1969. Mapa Ecologico de la Republica de Costa Rica. Centro Cientifico Tropical. San Jos6, Costa Rica. WEIR, B. S., AND C. C. COCKERHARM. 1984. Esti- mating F-statistics for the analysis of population structure. Evolution 38:1358-1370. WILKINSON, J. A., J. L. GLENN, R. C. STRAIGHT, AND J. W SITES, JR. 1991. Distribution and genetic vari- ation in venom A and B populations of the Mojave rattlesnake (Crotalus scutulatus scutulatus) in Ari- zona. Herpetologica 47:54-68. WRIGHT, S. 1965. The interpretation of population structure by F-statistics with special regard to sys- tems of mating. Evolution 19:395-420. Accepted: 27 November 1997 Associate Editor: David Green This content downloaded from 163.178.208.28 on Thu, 30 Jul 2015 17:04:39 UTC All use subject to JSTOR Terms and Conditions