40 Mark S. Davis and George W. Folkerts 



was in a small woodland pond in northern Tallapoosa County. In both 

 ponds all egg masses were restricted to the COS. The third largest popu- 

 lation (60 egg masses) was in a pasture breeding pond, where communal 

 oviposition occurred, but to a lesser extent. In this pond 31 egg masses 

 were in a communal site, and the rest were deposited in small clumps 

 separate from the COS. The 1 1 other wood frog breeding ponds discov- 

 ered during this study were characterized by extremely small popula- 

 tions (compare Howard 1980, Berven 1981, Seale 1982, Waldman 1982). 

 The number of egg masses found in each of these ponds varied from 4 

 to 28, and the tendency toward communal oviposition was less 

 pronounced. 



Clutch size varied from 350 to 709 eggs per mass (x = 496). Ovarian 

 and /or ovisac counts indicated that oviposition may occur once or twice 

 during the breeding season. The number of ovarian eggs per female 

 ranged from 618 to 966. When all eggs are deposited at one time, the 

 resultant egg mass appears as two fused masses, indicating that females 

 empty each ovisac separately. If a female moves to another site after 

 emptying one ovisac, the resultant egg mass represents approximately 

 one-half the ovarian complement. This probably accounts for much of 

 the apparent variability in clutch sizes observed in the field. Even so, 

 there is some variability in reproductive potential, as evidenced by the 

 range in egg complements seen in gravid females. This is probably 

 attributable to a combination of individual and ontogenetic variation. 

 Seale (1982) found no significant difference between clutch size and 

 ovarian egg counts in Pennsylvania wood frogs (clutch: x~ = 895; ovarian 

 eggs: x = 840). Although there are few data available concerning ovarian 

 egg counts, several authors have presented information on clutch size 

 (Table 2). There is some evidence for smaller clutch size in the southern 

 parts of the range, although this trend may be obscured by altitudinal 

 differences (Berven 1982a). Clutch size probably varies in response to 

 different selection pressures throughout the geographic range, creating a 

 chaotic pattern of variation. Furthermore, variation in clutch size 

 should be viewed with respect to differences in adult body size and egg 

 size. At the southern terminus of the frog's range, the probability of egg 

 mortality resulting from freezing is reduced and may be a factor in 

 decreased clutch size. Moore (1949) pointed out that the submerged egg 

 masses of northern species of Rana were poorly adapted for higher 

 pond temperatures because diffusion of oxygen would not be rapid 

 enough to supply the metabolic needs of embryos in the center of the 

 egg mass. Thus, the smaller egg masses characteristic of southern popu- 

 lations of R. sylvatica would allow a more rapid diffusion of oxygen to 

 these inner embryos. However, Savage (1961) claimed that egg masses 

 possess intercapsular channels and that gaseous diffusion need not take 



