74 Christopher King Beachy 



rocks in and along the edges of headwater streams and seepages. 

 These species of Desmognathus have larval periods less than one 

 year (Bruce 1989). Clutches of the D. ochrophaeus and D. santeetlah 

 were collected during July and August 1987. 



Desmognathus aeneus Brown and Bishop, 1947 females oviposit 

 under moist logs and moss. This species is direct-developing (Wake 

 1966), and desiccation risk should represent less constraint to it. 

 Clutches of D. aeneus were collected in the vicinity of Standing 

 Indian Campground in the Nantahala Mountains, Macon County, 

 North Carolina, in May 1988. 



Collection of Material 



As soon as I collected them, I placed egg clutches in individual 

 plastic containers. If I found a brooding female with the clutch, I 

 collected her and placed her with it. The plastic containers were 

 placed in a cooler and returned to the laboratory where the egg 

 clutches were assigned Harrison developmental stages (Duellman and 

 Trueb 1986). I used a dissecting microscope equipped with an ocular 

 micrometer to measure egg diameters to the nearest 0.1 mm. Late 

 developmental stages were assigned based primarily on gill ontogeny. 

 Embryos of clutches in which the embryos were in later stages of 

 development were adjusted for developmental increases in size with 

 the transformation formulas of Kaplan (1979). Clutches of A. 

 maculatum and D. aeneus were all collected very early in 

 development. The other three species were collected at various 

 developmental stages, some of them at late stages. Embryo 

 diameters for the plethodontid species in late development were 

 recorded as the length of the longest axis of the embryo (Fig. 1). 

 Using this measurement protocol, I observed that plethodontid 

 embryos do not begin to increase in size until after Harrison stage 

 30 (Beachy 1988). 



Analyses 



Intraclutch variation was quantified by calculating the 

 coefficient of variation (CV) for each clutch. The CV is a statistic 

 that expresses the standard deviation as a percentage of the mean so 

 that groups having very different means can be compared. These 

 intraclutch CVs were subjected to a one-way ANOVA by species. 

 Although a single female of A. maculatum may oviposit several 

 clusters of eggs, I assumed that differences among these clusters do 

 not contribute significantly to the variance in CVs, and all A. maculatum 

 clusters were treated as though they were from different females 

 (however, see Results). 



