Demography of Green Snakes (Opheodrys Aestivus) Author(S): Michael V

Herpetologists' League

Demography of Green Snakes (Opheodrys aestivus) Author(s): Michael V. Plummer Source: Herpetologica, Vol. 41, No. 4 (Dec., 1985), pp. 373-381 Published by: Herpetologists' League Stable URL: http://www.jstor.org/stable/3892105 . Accessed: 21/03/2013 12:50

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 [email protected].

Herpetologists' League is collaborating with JSTOR to digitize, preserve and extend access to Herpetologica.

http://www.jstor.org

This content downloaded from 192.133.129.4 on Thu, 21 Mar 2013 12:50:31 PM All use subject to JSTOR Terms and Conditions VOL. 41 DECEMBER 1985 NO. 4

Herpetologica, 41(4), 1985, 373-381 ? 1985 by The Herpetologists' League, Inc.

DEMOGRAPHY OF GREEN SNAKES (OPHEODRYS AESTIVUS)

MICHAEL V. PLUMMER

ABSTRACT: Sex ratio, populationsize, age structure,and survivorshipof an Opheodrysaestivus populationwere studied by mark-recapturein Arkansas.Sex ratio varied monthly, favoring males in early season (60-80%) and females in midseason (60-80%). Females predominatedin one year but the sex ratio was 1:1 in two other years. Population density (approximately 430/ha) was statisticallyconstant over two years. Age structureof females varied little between years, but age structureof younger males changed from one year to the next. Age-specificsurvivorship, assumed to be constant in adults, was greater in females (49%)than in males (39%).Survivorship during the first year was low (21%).A life table yielded a net reproductiverate (R0= 0.84) insufficientto sustain the population. The calculated survival rate of first-year snakes, and possibly of adults, apparently underestimatedactual survival. Key words: Reptilia; Colubridae;Demography; Mark-recapture;Survivorship; Life history

REPTILIAN demographic studies have tivus, is an arboreal species which prefers been concerned primarily with lizards. the brushy vegetation of fence rows and The fact that demographic and other eco- forest edges. In contrast to many snake logical studies have provided much in- species, it has attributesthat would appear sight into ecological processes led Huey et to enhance demographic analysis. Al- al. (1983) to state: "Clearly,for many types though diurnally cryptic, green snakes are of ecological studies lizards are model or- abundant, have low vagility, and have a ganisms-moreover, they now challenge conspicuous sleeping posture at night (on birds as the paradigmatic organism of the distal ends of branches) which renders ecology." Demographic studies of their them easily and predictably collected close relatives, the snakes, have not been (Plummer, 1981). Some demographic at- as instructive. Several attributes of snakes tributes, such as reproductionand growth, generally contrast those of lizards and have been reported (Plummer, 1984, contribute to snakes being less tractable 1985). The purpose of this paper is to ana- study subjects.These attributesinclude in- lyze sex ratio, population size, age struc- conspicuousness, nocturnality, unpredict- ture, and survivorship.Field observations able and potentially extensive move- from August 1977 to September 1984 in- ments, long periods of inactivity, and dicated that my study population was re- apparent low population densities. Recent markablystable. If these observationswere evaluations of the suitability of snakes for correct, the analysis of the interaction of demographic studies (Lillywhite, 1982; reproduction and mortality should yield a Turner, 1977) have not been favorable. net reproductive rate that would sustain The rough green snake, Opheodrys aes- the population over the long term.

373

This content downloaded from 192.133.129.4 on Thu, 21 Mar 2013 12:50:31 PM All use subject to JSTOR Terms and Conditions 374 HERPETOLOGICA [Vol. 41, No. 4

TABLE 1.-Observed frequencies of single and mul- tiple captures of Opheodrys aestivus compared to a 80 . zero-truncatedPoisson distributionof expected fre- (f) quencies assuming catchability was constant. ~60 No. of monthly samples 1978 1979 w in which Q 40 // an individ- Observed Expected Observed Expected ual was no. of no. of no. of no. of captured snakes snakes snakes snakes Ql- 1978 * (n=390) / 20 1979 o (n=353) 1 126 131.88 114 114.89 1981 a (n=169) 2 65 57.24 44 51.53 1984 A (n=399) 23 19 20.88* 28 19.58* APR MAY JUN JUL AUG SEP-OCT X2 = 1.48 X2= 4.73 MONTH P > 0.10 P < 0.05

* FIG. 1.-Sex ratio variation in monthly samples of Cells with expected frequencies c5 were pooled. Opheodrys aestivus. (Plummer, 1981, 1984, 1985). I assigned snakes to age classes according to the size- METHODS age criteriaof Plummer (1985). Age classes The study area was the forest bordering were: age class E = egg (from oviposition a narrow (approximately50 m) 700 m long to hatching), H = hatchling (from hatch- channel where Overflow Creek entered ing to first hibernation), 0 = first-year Bald Knob Lake in White County, Arkan- snake (from May following the first hi- sas. The habitat has been described pre- bernation through September of the same viously (Plummer, 1981). Green snakes in- year), age class 1 = 1-yr-old (from May habited the dense forest edge vegetation following the second hibernation through near the shoreline and were collected by September of the same year), age class hand after searching the vegetation at 2 = 2-yr-old (from May following the third night with a spotlight while slowly cruis- hibernation through September of the ing the shoreline in a boat. Collecting oc- same year), and so forth. curred 2-3 times per week from late April through early October in 1978 and 1979 RESULTS and irregularly in 1977 and after 1979. In Sex Ratio 1984, snakes were sampled biweekly from From April 1978-October 1979, 743 Bald Knob Lake and nearby localities for captures were made on 333 snakes in the related studies. These data were included study area. In 1978, the overall sex ratio in the sex ratio analysis. On initial capture of 0.94 male: 1 female (210 snakes) was in 1977-1979, snakes were given a unique not significantly different from 1:1 (X2= mark by clipping ventral scutes (Brown 0.30; P > 0.50). In 1979, the overall sex and Parker, 1976). Date, snake number, ratio of 0.72 male: 1 female (186 snakes) sex, and snout-vent length (SVL) were re- was significantly different from 1:1 (X2= corded at each capture. Snakes were pro- 4.83; P < 0.05). Removal collecting of 169 cessed in the field and were released with- snakes in 1981 at the opposite end of the in 10 min at the point of capture. Statistical lake yielded a sex ratio of 1.01 male: 1 procedures follow SPSS Inc. (1983). female which was not significantly differ- At this locality, green snakes grow pri- ent from 1:1 (X2= 0.01; P > 0.90). Sex marily from May through September. ratio varied monthly in all years (Fig. 1). Oviposition occurs in June-July and Males tended to predominate in early sea- hatching in August-early September. son samples and again, but to a lesser ex- Snakes enter into hibernation by October tent, in late season samples. Females pre- and ernerge in late April-early May dominated in mid-season (late June-July)

This content downloaded from 192.133.129.4 on Thu, 21 Mar 2013 12:50:31 PM All use subject to JSTOR Terms and Conditions December 1985] HERPETOLOGICA 375

TABLE 2.-Statistics of population estimates of Opheodrys aestivus by the Schumacher-Eschmeyermethod. Confidence limits were calculated according to Caughley (1977). Correlationcoefficients and significance levels of regression equations are indicated. See text for an explanation of the meaning of the last four columns.

Estimated popula- No. tion size 95 a confidence snakes Year (N) limits of N captured Regressionof m,/n, on M, 1/N r P 1978 303 275-337 210 y = 0.001 + 0.0033x 0.0033 0.99 <0.001 1979 256 219-307 186 y = 0.007 + 0.0038x 0.0039 0.98 <0.001 samples when oviposition occurred (Plum- sumption of equal catchability. Unless the mer, 1984). assumption of equal catchability is violat- ed, regression of m,/n, on M, is lin- Population Size ear through the origin with slope 1/N An assumptionof population estimation (Caughley, 1977). Correlation coefficients methods by mark-recapture is the equal for both regressions were highly signifi- probability of capture of all individuals. cant and, corroboratingthe checks of equal To test this assumption on Opheodrys, I catchability (Table 1), the intercept was fitted the observed number of snakes cap- closer to the origin and there was closer tured 1, 2, and 3 or more times in monthly agreement between 1/N and slope in 1978 (May-September) samples to a zero-trun- than in 1979 (Table 2). For these reasons, cated Poisson distribution of frequencies I consider the populationestimate for 1978 to be expected if catchability was constant to be closer to the actual number of that (Caughley, 1977). The null hypothesis of year. Since the population estimate for equal catchability can be rejected for 1978 is contained within the confidence samples in 1979 but not in 1978 (Table 1). limits for 1979, it is likely that the esti- Population sizes in each year were es- mates for the 2 yr are not significantly timated by the method of Schumacherand different. Eschmeyer (1943) from monthly (May- Based on a population size of 300 (Ta- September) samples. The equation is ble 2), there were approximately 0.21 0. aestivus per linear meter of shoreline. This v Mi2ni density is similar to those reported (no./ N = - Mimi m) for other shoreline snakes (Regina sep- 2; temvittata 0.18-0.26, Bransonand Baker, where N is the estimated population size, 1974; three spp. Nerodia 0.03-0.18, He- M, is the number of individuals marked brard and Mushinsky, 1978). Plummer prior to the ith sampling period, and n, is (1981) found that 0. aestivus activity was the number of individuals cantured in the concentrated near the shoreline with 86- ith sample of which m, had been marked 88% of all captures made within 3 m of previously (Caughley, 1977). The method the shoreline and 96-97% made within 5 assumes no birth, mortality, emigration, m. Using the habitat boundaries of 3 m or immigration during the census. Young and 5 m, respectively, I calculated popu- of the year were not included in the cen- lation densities of 714 snakes/ha and 429/ sus. No estimate of emigration or immi- ha. While the exact extent of habitat uti- gration was available; the mean distance lization away from the shoreline may be between recaptures was only 26 m (Plum- argued, at a conservative estimate of 429 mer, 1981). Two advantages of the Schu- snakes/ha, this population of 0. aestivus macher-Eschmeyer method are allowing was, with few exceptions, among the most a standard error of N to be calculated and dense of reported snake populations (An- providing an additional check on the as- dren and Nilsen, 1983; Blaesing, 1979;

This content downloaded from 192.133.129.4 on Thu, 21 Mar 2013 12:50:31 PM All use subject to JSTOR Terms and Conditions 376 HERPETOLOGICA [Vol. 41, No. 4

fecundity of the 1978 population (Fig. 2). MALES FEMALES (78) _>5 (109) Using the function, number of eggs = -4.1 + 0.023 SVL (Plummer, 1984), the esti- 4 mated 1978 egg production was 482 eggs 3 of which 241 were males and 241 females 2 since the sex ratio at hatching, like most snakes (Shine and Bull, 1977), was 1:1

0 1979 (Plummer, 1984). Only 10 male and 11 female first-year snakes were captured in 1979. Based on MALES FEMALES egg productionin 1978, these capture data (101) >5 (109) provide survivorshipestimates of 4.1%and 4 4.6%, respectively. 3 For age classes -0, I determined sex- 2 specific survivorshipby the proportion of individuals in each 1978 age class n present in the 1979 age class n + 1. Survivor- 0 ~~~~~~978ship from age class 0 to age class 1 was 38.5% for males and 30.8% for females. 40 3020 10 10 20 3040 For age classes - 1, I regressed the natural PERCENTOF SNAKES logarithm of the number of survivors (In FIG. 2.-Age structure of an Opheodrys aestivus S) in an initial cohort of 100 on age x (in population in 2 yr. Data include only those snakes years). The regression equations were, for actually captured. Open bars indicate mature indi- males, In S = 5.43 - 0.947x (r = -0.995, viduals, closed bars represent immatures. Numbers in parentheses are sample sizes. P < 0.001) and, for females, In S = 5.48 - 0.722x (r = -0.995, P < 0.001). As with snakes in general (Brown and Parker, Brown, 1970; Clark, 1974; Clarkand Fleet, 1982; Parker and Brown, 1980; Turner, 1976; Fitch, 1982; Freedman and Catling, 1977), the correlation coefficients indicat- 1978; Godley, 1980; Madsen and Oster- ed only slight deviations from the hypoth- kamp, 1982; Spellerbergand Phelps, 1977; esis of linearity (age-constant mortality) Stickel et al., 1980; Turner, 1977). Perhaps (Turner, 1977). the greater extent of the vertical compo- The slopes for males and females were nent of the habitat of this arboreal snake significantly different (ANCOVA, P < permitted greater density than in non-ar- 0.01). Annual adult (-1 yr of age) surviv- boreal species. al (S = e", where b is the slope from the above regressionsand e is 2.718) for males Population Structure and Survivorship was 38.8% and for females 48.6%. The The resulting age structure (Fig. 2) greater adult survivorship of females ap- showed a year-to-year constancy in fe- parently influenced the sex ratio from males which suggested that a stable age younger to older snakes (Fig. 2). The x2 distribution had been achieved. The age values for the 0, 1, and 2-yr-olds were not distribution of males was more variable significantly different from a 1:1 sex ratio especially in the 0-2-yr-olds. in 1978 (X2= 3.46, P > 0.05) or in 1979 Reproduction in this population was of (x2 = 0.01, P > 0.90). In contrast, the x2 the birth-pulse type. Because of the small values for the 3, 4, and >5-yr-olds signif- size of hatchlings and, consequently, the icantly departed from 1:1 in favor of fe- difficulty of sampling them, and the re- males in both 1978 (X2= 12.86, P < 0.001) stricted period in which to sample them and 1979 (X2= 11.58, P < 0.001). before winter mortality occurred, I esti- A life table for 0. aestivus (Table 3) mated reproductiveoutput indirectly from was based on these assumptions:(1) age E

This content downloaded from 192.133.129.4 on Thu, 21 Mar 2013 12:50:31 PM All use subject to JSTOR Terms and Conditions December 1985] HERPETOLOGICA 377

began at oviposition; (2) fecundity was de- TABLE 3.-Life table for Opheodrys aestivus: x = termined from median-sized snakes in age (years); 1 = age-specific survival rate; m, = age- specific fecundity rate; Ro = net reproductive rate. each age class by the previous function See text for assumptions. and adjusted by assuming 50% of females reproduced at age 1 and 100% thereafter x m, l.m. (Plummer, 1985); (3) sex ratio at birth was E 1.000 0 0 1:1; (4) survivorshipin age classes - 1 was 1 0.212 1.03 0.218 0.486; (5) survivorshipfrom the egg to age 2 0.103 2.81 0.289 1 3 0.050 3.25 0.163 (= "first-year")was calculated indirect- 4 0.024 3.52 0.084 ly by dividing the mean number of 2-yr- 5 0.012 3.67 0.044 old females in 1978-1979 (25) by survi- 6 0.006 3.75 0.023 vorship in age class 1 (0.486) to determine 7 0.003 3.81 0.011 the mean number of snakes in age class 1 8 0.001 3.84 0.004 that should have been present in 1978- Ro = 0.836 1979 (number of 1-yr-olds = mean no. 2-yr-olds/survivorshipin age class 1 = 25/ 0.486 = 51). Survivorshipin the first year and Parker (1984) found greater adult (0.212) was calculated as the number in survival rates in snakes than reported in age class 1/number in age class E = 51/ other studies. They attributed their results 241 = 0.212). to techniques of enclosing virtually entire populations by fencing around communal DISCUSSION dens which resulted in capturing most in- Male preponderance in early spring dividuals each year. In most studies, such samples has been found in other snakes as the present, where such enclosure tech- (Clark and Fleet, 1976; Fitch, 1949, 1975; niques were not feasible, any estimate of Gregory, 1977; Turner, 1977; Voris and survivorship probably is lower than the Jayne, 1979), and probably is related to true rate because of less reliable recapture greater male activity when searching for proportions (Parker and Brown, 1980). In mates. Whether the same explanation 0. aestivus, several lines of evidence sug- holds true for the much less pronounced gested that actual survivorship,especially preponderance of male 0. aestivus in the in the first year, was much greater than fall is uncertain. Richmond (1956) ob- that estimated. Foremost was an unbeliev- served fall mating of 0. aestivus in Vir- ably low net reproductiverate (Ro= 0.058) ginia, but Plummer (1984) found no evi- which could be calculated using an esti- dence of fall mating in the Bald Knob mated first-year survivorship of 0.014 Lake population. The preponderance of (survivorshipfrom egg to age 0 x survi- females in midseason may be related to vorship from age 0 to age 1 = 0.046 x the greater conspicuousnessof the already 0.308 = 0.014), and assuming a constant larger females being swollen with eggs. adult survivorshipof 0.486 and the above An alternative explanation is that males fecundity schedule. This calculation in- decrease feeding and become relatively dicated that the population was rapidly inactive as occurs in Nerodia sipedon decreasingin 1978-1979. If continued, this (Feaver, 1977). Was the skewed sex ratio would quickly lead to local extinction, a in 1979 favoring females the result of conclusion inconsistent with field obser- greater collecting effort in June-July? It vationsfrom August 1977-September 1984 was not likely because 164 June-July cap- which indicated that the population was tures of 390 total captures (42.1%)in 1978 remarkably stable. As demographic attri- compared favorably to the 151 June-July butes in snakes are known to vary consid- captures of 353 total captures (42.8%) in erably from year to year (Brown and Par- 1979. ker, 1984; Feaver, 1977; Parker and Parker and Brown (1980) and Brown Brown, 1980), 1978 may have been an un-

This content downloaded from 192.133.129.4 on Thu, 21 Mar 2013 12:50:31 PM All use subject to JSTOR Terms and Conditions 378 HERPETOLOGICA [Vol. 41, No. 4 usually low year for survival. However, (Brown and Parker, 1982; Feaver, 1977; the constancy of the female age structure Turner, 1977). in 1978-1979 (Fig. 2) points toward the Implicit in demographic analysis by attainment of a stable age distribution mark-recaptureis the assumptionof equal which can result only from a relatively probability of capture of all individuals. long-term constant mortality schedule and This assumption held true because no rate of increase or decrease (Caughley, Opheodrys ever escaped capture once 1977). sighted. However, if the behavior of any Because of the discrepancy between snake had decreased the probability of its field observations and estimated survivor- being seen, the assumption would be vio- ship, I estimated first-year survivorshipof lated. Unequal catchability could not be females by another independent method. statistically demonstrated in 1978 but was Age class 2 is the first class to contain more likely in 1979 (Table 1). Possible sources individuals than the next older age class of unequal catchability included the ten- (Fig. 2), at least in 1978. In a population dency of males for greater movement with a stable age distribution, this pattern (Plummer, 1981), and seasonal differences probably exists because proportionally in activity between the sexes (Fig. 1). more 2-yr-old females were caught rela- Another possible source, with probably tive to the actual number in the class than greater impact, concerned juvenile snakes. in any younger age class. Therefore, I as- Numerousauthors have commented on the sumed that population size, mortality, and difficulty in sampling juveniles (Brownand fecundity had been constant for several Parker,1984; Clark, 1970; Clarkand Fleet, years, and I calculated a first-year survi- 1976; Fitch, 1960, 1963a,b, 1965; Grego- vorship (0.212, above) based on the ob- ry, 1977; Hirth and King, 1968; Jackson served number of 2-yr-olds. The life table and Franz, 1981; Lillywhite, 1982; Parker constructed from these assumptions re- and Brown, 1973, 1980; Platt, 1969; Sem- sulted in an Roof 0.836 (Table 3), a value litsch et al., 1981; Stickel et al., 1980). The still less than the Ro = 1.0 needed for each possibility of juveniles having behavior female to replace herself each generation, patterns different from adults such as rel- but probably much closer to reality than ative inactivity or different microhabitat the estimate based on actual recapture of preferences (Fitch, 1960, 1963a; Gibbons survivors in classes 0 and 1. et al., 1977; Jacksonand Franz, 1981; Lil- My best estimate of first-year survivor- lywhite, 1982; Semlitsch et al., 1981; Voris ship in 0. aestivus (0.212) is similar to and Jayne, 1979) would contribute to bias those found for other colubrids such as if the behavioral differences caused visible Coluber constrictor 0.156 (Fitch, 1963b); access of juveniles to be lower than that Rhabdophis tigrinus 0.199 and Elaphe of adults. Despite the lack of correlation quadrivirgata 0.177 (Fukada, 1969); Col- between perch height and SVL in 0. aes- uber constrictor 0.170, Masticophis tae- tivus (Plummer, 1981), the possibility ex- niatus 0.145, and Pituophis melanoleucus isted that juveniles spent relatively more 0.200 (Brown and Parker, 1982); and Ne- time on the ground or in the lowermost rodia sipedon 0.235 (Feaver, 1977). If in vegetation stratum where visible access fact actual adult survivorship was under- was minimal. Combined with their small estimated (cf. Brown and Parker, 1984; size, this behavior would render them less Parker and Brown, 1980), any increase in susceptible to capture and would result in this value also would increase Ro. Further, an unrealistically high mortality estimate annual adult (snakes -1-yr-old) survival and an underrepresentationof the group estimates in 0. aestivus (males 0.388, fe- in the age structure (cf. Fig. 2). males 0.486) are toward the low end of Maximum longevity also suggested that published values of approximately 0.30- survivorship was greater than measured. 0.86 (x = 0.57) for 12 species of colubrids Two 3-yr-old females first captured in

This content downloaded from 192.133.129.4 on Thu, 21 Mar 2013 12:50:31 PM All use subject to JSTOR Terms and Conditions December 1985] HERPETOLOGICA 379

September 1977 were recaptured in July- Feaver, 1977; Xenochrophis vittata, Col- August 1981, and thus survived to their uber constrictor, and Carphophis vermis, seventh year. Another female (505 mm Turner, 1977 (from data of Bergman, SVL) was first captured in April 1978 and 1950; Clark, 1970; Fitch, 1963b); 0. aes- was recaptured in April 1981. Although tivus, present study). Exceptions include estimating age at this size was not reliable only the viperid Agkistrodon contortrix (Plummer, 1985), conservatively(if it were (Vial et al., 1977) and those few colubrid a fast grower), an estimate of 5 yr would populations which could be completely be realistic and would indicate that this enclosed by fencing around communal snake survived to at least its eighth year. dens (Pituophis melanoleucus, Coluber If my inferred survivorship in the popu- constrictor, Masticophis taeniatus, Brown lation was correct, the chances of surviv- and Parker, 1984; Parker and Brown, ing to an age of 7 and 8 yr, would be three 1980). Despite the attributes of 0. aesti- in 1000 and one in 1000, respectively (Ta- vus (above), I was unable accurately to ble 3). In fact, these three snakes repre- assesssurvivorship, especially in juveniles. sented an approximatelyone in 100 chance This lack of success resulted in an Ro in- (three survivors in a population of ap- sufficient to sustain the population over proximately 300). An even greater num- the long term and was in direct contrast ber of snakes actually survived to at least to field observations. Because of these re- their fifth or sixth year despite the calcu- sults, I agree with Turner (1977) and Lil- lated probability being less than 12 in lywhite (1982) that snakes probably are 1000. poor candidates for studies of pure de- Feaver (1977) identified two general life mography. Why they are poor candidates, history tactics among 17 species of colu- however, is an interesting question in it- brid and viperid snakes. One, in which 0. self. While populationstudies in snakesare aestivus better fits, is characterized by proliferating, the life history stages that early maturity, small female size at ma- we know least about, the egg, hatchling, turity, large clutches, steep increase of and first-year stages, are receiving little clutch size with female size, annual attention. Few authors have ever found clutches, females the larger sex, and low snake eggs in the field, and the literature adult survivorship. Various measures of is replete with statements regarding the reproductive output were relatively in- unavailability of juveniles (above). Al- variable both within (Plummer, 1983; Sei- though juvenile snakes may be "miniature gel and Fitch, 1984) and between Opheo- adults" morphologically,the chances seem drys populations (summary in Plummer, quite good that their ecology is funda- 1984). In contrast, lipid storage in the Ar- mentally different from adults. Many be- kansas population was highly variable an- havioral and ecological characteristics of nually and presumably was affected by reptiles have underlying physiological food availability (Plummer, 1983). A re- bases, some of which change profoundly productive tactic of a relatively invariant through ontogeny (Pough, 1983). For ex- reproductive output compared to stored ample, juveniles of several species of lipids has been predicted to develop in snakes have limited endurance capacities relatively long-lived species in environ- compared to adults (Pough, 1983). Such ments where juvenile survivorshipis high- differences could fundamentally affect ly variable and unpredictable (Congdon foraging and other activity patterns. and Tinkle, 1982). Assessment of juvenile survivorshipap- Demographic studies of snakes in which pears to be the single greatest problem in the interaction of mortality and reproduc- snake demography. Approaches to the tion has been evaluated have yielded net study of such stages perhaps could include reproductiverates below that necessaryfor marking large numbers of hatchlings or long-term stability (Nerodia sipedon, neonates obtained from laboratory fe-

This content downloaded from 192.133.129.4 on Thu, 21 Mar 2013 12:50:31 PM All use subject to JSTOR Terms and Conditions 380 HERPETOLOGICA [Vol. 41, No. 4 males and releasing them in the field CAUGHLEY, G. 1977. Analysis of Vertebrate Pop- (Blanchard and Finster, 1933; Fukada, ulations. Wiley-Interscience,New York. CLARK, D. R., JR. 1970. Ecological study of the 1969) or by constructing artificial nesting worm snake Carphophisvermis (Kennicott).Univ. sites in the field. Whatever the approach, KansasPubl. Mus. Nat. Hist. 19:85-194. innovative work needs to be focused on . 1974. The western ribbon snake (Tham- these little known and apparently vulner- nophis proximus): ecology of a Texas population. able stages in the life histories of snakes. Herpetologica 30:372-379. CLARK, D. R., JR., AND R. R. FLEET. 1976. The Acknowledgments.-The assistanceof several stu- rough earth snake (Virginia striatula): ecology of dents, notably T. M. Baker, M. W. Patterson,D. E. a Texas population. Southwest.Nat. 20:467-478. Sanders,and M. White, is gratefully acknowledged. CONGDON, J. D., AND D. W. TINKLE. 1982. Re- Thanks are due to the Bald Knob City Government productiveenergetics of the painted turtle (Chrys- for permitting me gratis use of Bald Knob Lake. This emys picta). Herpetologica 38:228-237. paper was greatly improved by the constructivecrit- FEAVER, P. E. 1977. The Deography of a Michigan icisms of W. S. Parker and particularly of several Population of Natrix sipedon with Discussionsof anonymousreviewers. The numerous versionsof the Ophidian Growth and Reproduction. Ph.D. Dis- manuscript were typed by M. Groves, C. Howell, sertation,University of Michigan. and C. Lloyd. This study was funded in part by sev- FITCH, H. S. 1949. Study of snake populations in eral summer grants from Harding University. central California. Am. Midi. Nat. 41:513-579. 1960. Autecologyof the copperhead.Univ. KansasPubl. Mus. Nat. Hist. 13:85-288. LITERATURE CITED . 1963a. Natural history of the black rat snake (Elaphe o. obsoleta) in Kansas.Copeia 1963: ANDREN,C., AND G. NILSON. 1983. Reproductive 649-658. tactics in an island population of adders, Vipera . 1963b. Natural history of the racer Col- berus (L.) with a flunctuating food resource. Am- uber constrictor.Univ. KansasPubl. Mus.Nat. Hist. phibia-Reptilia 4:63-79. 15:351-468. BERGMAN,R. A. M. 1950. The life of Natrix vittata . 1965. An ecological study of the garter (L.). Zool. Mededel., Leiden 31:1-11. snake, Thamnophis sirtalis. Univ. Kansas Publ. BLAESING, M. E. 1979. Some aspects of the ecology Mus. Nat. Hist. 15:493-564. of the eastern garter snake (Thamnophis sirtalis . 1975. A demographic study of the ring- sirtalis) in a semi-disturbed habitat in west-central neck snake (Diadophispunctatus) in Kansas.Univ. Illinois. J. Herpetol. 13:177-181. KansasMus. Nat. Hist. Misc. Publ. 62:1-53. BLANCHARD, F. N., AND E. B. FINSTER. 1933. A 1982. Resourcesof a snake community in method of marking living snakes for future rec- prairie-woodlandhabitat of northeasternKansas. ognition, with a discussionof some problems and Pp. 83-97. In N. J. Scott, Jr. (Ed.), Herpetological results. Ecology 14:334-347. Communities.U.S. Fish Wildlife Ser. Wildlife Res. BRANSON, B. A., AND E. C. BAKER. 1974. An eco- Rep. 13. logical study of the queen snake, Regina septem- FREEDMAN, W., AND P. M. CATLING. 1978. Pop- vittata (Say) in Kentucky. Tulane Stud. Zool. Bot. ulation size and structureof four sympatricspecies 18: 153-171. of snakesat Amherstburg,Ontario. Can. Field-Nat. BROWN, T. 1970. Autecology of the Sidewinder 92:167-173. (Crotalis cerastes) at Kelso Dunes, MojaveDesert, FUKADA, H. 1969. Biological studies on the snakes California. Ph.D. Dissertation,University of Cali- XIII. Preliminary estimate of population size in fornia, Los Angeles. Tambabashi study area. Bull. Kyota Univ. Edu- BROWN, W. S., AND W. S. PARKER. 1976. A ventral cation Ser. B, 36:3-9. scale clipping system for permanently marking GIBBONS,J. W., J. W. COKER,AND T. M. MURPHY, snakes (Reptilia, Serpentes). J. Herpetol. 10:247- JR. 1977. Selected aspects of the life history of 249. the rainbowsnake (Faranciaerytrogramma). Her- 1982. Niche dimensionsand resourcepar- petologica 33:276-281. titioning in a Great Basin Desert snake commu- GODLEY, J. S. 1980. Foragingecology of the striped nity. Pp. 59-81. In N. J. Scott, Jr. (Ed.), Herpeto- swamp snake, Regina alleni, in Southern Florida. logical Communities. U.S. Fish Wildlife Ser. Ecol. Monogr.50:411-436. Wildlife Res. Rep. 13. GREGORY, P. T. 1977. Life-history parameters of . 1984. Growth, reproduction and demog- the red-sided garter snake (Thamnophis sirtalis raphy of the racer, Coluber constrictor mormon, parietalis) in an extreme environment, the Inter- in northernUtah. Pp. 13-40. In R. A. Seigel, L. E. lake Region of Manitoba. Natl. Mus. Can. Publ. Hunt, J. L. Knight, L. Malaret,and N. L. Zuschlag Zool. 13:1-44. (Eds.), Vertebrate Ecology and Systematics: A HEBRARD, J. J., AND H. R. MUSHINSKY. 1978. Hab- Tribute to Henry S. Fitch. Mus. Nat. Hist. Univ. itat use by five sympatric water snakes in a Loui- Kansas Spec. Publ. 10. siana swamp. Herpetologica 34:306-311.

This content downloaded from 192.133.129.4 on Thu, 21 Mar 2013 12:50:31 PM All use subject to JSTOR Terms and Conditions December 1985] HERPETOLOGICA 381

HIRTH, H. F., AND A. C. KING. 1968. Biomassden- low-energy systems. Pp. 141-188. In W. P. Aspey sities of snakes in the cold desert of Utah. Herpe- and S. I. Lustick (Eds.), BehavioralEnergetics: The tologica 24:333-335. Cost of Survival in Vertebrates. Ohio State Uni- HUEY, R. B., E. R. PIANKA, AND T. W. SCHOENER. versity Press, Columbus. 1983. Introduction.Pp. 1-6. In R. B. Huey, E. R. RICHMOND, N. D. 1956. Autumn mating of the Pianka, and T. W. Schoener (Eds.), Lizard Ecolo- rough green snake. Herpetologica 12:325. gy. Harvard University Press, Cambridge, Massa- SCHUMACHER, F. X., AND R. W. ESCHMEYER. 1943. chusetts. The estimate of fish population in lakes or ponds. JACKSON, D. R., AND R. FRANZ. 1981. Ecology of J. Tennessee Acad. Sci. 18:228-249. the eastern coral snake (Micrurus fulvius) in SEIGEL, R. A., AND H. S. FITCH. 1984. Ecological northernpeninsular Florida. Herpetologica 37:213- patterns of relative clutch mass in snakes. Oeco- 228. logia 61:293-301. LILLYWHITE, H. B. 1982. Tracking as an aid in SEMLITSCH, R. D., K. L. BROWN, AND J. P. CALD- ecological studies of snakes. Pp. 181-191. In N. J. WELL. 1981. Habitat utilization, seasonal activi- Scott, Jr. (Ed.), Herpetological Communities. U.S. ty, and population size structureof the southeast- Fish Wildlife Ser. Wildlife Res. Rep. 13. ern crowned snake Tantilla coronata. Her- MADSEN, T., AND M. OSTERKAMP. 1982. Notes on petologica 37:40-46. the biology of the fish-eatingsnake Lycodonomor- SHINE, R., AND J. J. BULL. 1977. Skewed sex ratios phus bicolor in Lake Tanganyika.J. Herpetol. 16: in snakes. Copeia 1977:228-234. 185-188. SPELLERBERG, I. F., AND T. E. PHELPS. 1977. Bi- PARKER, W. S., AND W. S. BROWN. 1973. Species ology, general ecology and behaviourof the snake, composition and population changes in two com- Coronella austriaca Laurenti. Biol. J. Linn. Soc. plexes of snake hibernaculain northernUtah. Her- (London) 9:133-164. petologica 29:319-326. SPSSINC. 1983. SPSSXUser's Guide. McGraw-Hill, . 1980. Comparative ecology of two colu- New York. brid snakes, Masticophis t. taeniatus and Pituo- STICKEL, L. F., W. H. STICKEL, AND F. C. SCHMID. phis melanoleucus deserticola, in northern Utah. 1980. Ecology of a Marylandpopulation of black Milwaukee Pub. Mus. Publ. Biol. Geol. 7:1-104. rat snakes (Elaphe o. obsoleta). Am. Midl. Nat. PLATT, D. R. 1969. Natural history of the hognose 103:1-14. snakes Heterodon platyrhinos and Heterodon na- TURNER, F. B. 1977. The dynamics of populations sicus. Univ. KansasPubl. Mus. Nat. Hist. 18:253- of squamates,crocodilians and rhynchocephalians. 420. Pp. 157-264. In C. Gans and D. W. Tinkle (Eds.), PLUMMER, M. V. 1981. Habitat utilization,diet and Biology of the Reptilia, Vol. 7. Academic Press, movements of a temperate arborealsnake (Opheo- New York. drys aestivus). J. Herpetol. 15:425-432. VIAL, J. L., T. J. BERGER, AND W. T. MCWILLIAMS, 1983. Annual variationin stored lipids and JR. 1977. Quantitative demography of copper- reproduction in green snakes (Opheodrys aesti- heads, Agkistrodon contortrix (Serpentes: Viperi- vus). Copeia 1983:741-745. dae). Res. Popul. Ecol. 18:223-234. 1984. Female reproductionin an Arkansas VORIS,H. K., AND B. C. JAYNE. 1979. Growth, re- populationof rough green snakes (Opheodrysaes- production and population structure of a marine tivus). Pp. 105-113. In R. A. Seigel, L. E. Hunt, J. snake,Enhydrina shistosa (Hydrophiidae).Copeia L. Knight, L. Malaret, and N. L. Zuschlag (Eds.), 1979:307-318. VertebrateEcology and Systematics:A Tribute to Henry S. Fitch. Mus. Nat. Hist. Univ. KansasSpec. Accepted: 10 April 1985 Publ. 10. Associate Editor: Kentwood Wells 1985. Growthand maturityin green snakes (Opheodrysaestivus). Herpetologica 41:28-33. Department of Biology, Harding Uni- POUGH, F. H. 1983. Amphibians and reptiles as versity, Searcy, AR 72143, USA

This content downloaded from 192.133.129.4 on Thu, 21 Mar 2013 12:50:31 PM All use subject to JSTOR Terms and Conditions