--Studies of snake demography usually involve the estimation of litter size and neonate size, but potential sources of variation resulting from the methods used to obtain this information are rarely considered. Using data from four populations of Thamnophis elegans, we found that estimates of litter size obtained by holding gravid snakes in captivity were less variable when dead young or undeveloped eggs were included rather than excluded; we recommend that all components of litters be reported in future studies. Litter sizes obtained by palpation of gravid snakes in the field were similar to estimates from captive births in two of three populations we examined, but not the third. There was some evidence that litters from longer-term captives were more likely to contain dead young, and, in one population, smaller young. Access to substrate heat by gravid snakes was correlated with smaller progeny in another population. In general, captive-born snakes were smaller than neonates born in the field, but some growth may have occurred in the latter. Factors that might influence the size of snakes born in captivity are temperature, feeding level of mother, and stress; in order to minimize their potential effects, we recommend that gravid snakes be held for as short a time as possible. The study of life history variation in snakes has lagged behind that of other kinds of organisms (Parker and Plummer, 1987), mainly because of the difficulties of sampling snake populations (Turner, 1977). Although field measurements of important parameters such as survivorship have rarely been measured adequately (Parker and Plummer, 1987), reports of clutch or litter size are abundant (Seigel and Ford, 1987). These data can be obtained by dissection of preserved specimens, gentle hand palpation of gravid snakes in the field, or holding gravid snakes in captivity until they have laid eggs or produced young. The accuracy of clutch or litter size estimates obtained by palpation is unclear, but with practice, embryos can usually be distinguished from one another and from lumps of fecal material within the large intestine (Fitch, 1987; Farr and Gregory, pers. obs.). Sometimes, especially in the later stages of gestation, embryos may be more difficult to distinguish from one another; an embryo and its yolk sac may also be felt as two separate lumps (Fitch, 1987). Difficulties such as these will influence estimates of litter