Cave Isopod Ecology 
107 
artificial stream with five intermediate-size C. holsingeri or C. cannulus. If 
species differences are important, then total number of isopods washing 
out when all individuals are the same species should be different than 
when two species are involved. In fact, total washout of conspecifics (x = 
0.28, S.E. = 0.03, N = 6) was nearly identical to total washout of in- 
terspecifics (x = 0.32, S.E. = 0.03, N = 4). 
Table 3. Fraction of isopods washing out in artificial stream with various sizes 
of gravel. Statistical analysis used washout frequency (x) transformed to 
sin'' /x. Each run was started with 10 individuals (10 of one size, or 5 
each of two sizes). 
x sin'' /x = y 
N 
X 
S.E. 
y 
S.E. 
t 
P 
C. holsingeri alone 
5 
.42 
0.04 
40.3 
2.2 
1.08 
N.S. 
C. holsingeri with C. cannulus 
5 
.36 
0.04 
36.7 
2.5 
C. cannulus alone 
5 
.44 
0.04 
41.5 
2.4 
4.46 
CO.Ol 
C. cannulus with C. holsingeri 
5 
.08 
0.05 
10.6 
6.5 
Very limited information is available concerning Caecidotea in three 
of the caves listed in Table 1 . Nelson and Bazzle caves have large popula- 
tions of the amphipod Gammarus minus. Only two C. holsingeri were 
found in Bazzle Cave after extensive searching. A situation is similar in 
Nelson Cave. No isopods could be found in this cave in February 1980. 
Mill Run Cave has a large isopod population, but is closed by the 
owner. Specimens in National Museum of Natural Flistory collections 
were measured, and had a bimodal size distribution. Male second 
pleopods were not removed for examination, but the bimodal size dis- 
tribution is strong presumptive evidence that both species were present. 
Length measurements for C. cannulus and C. holsingeri are sum- 
marized in Table 4. In all cases of syntopy, C. cannulus is at least 1.9 times 
as large as C. holsingeri, all the differences being statistically significant. 
However, consideration of allotopic populations complicates the issue. 
The allotopic population of C. cannulus in Cave Hollow Cave is slightly 
smaller (x = 7.05 mm) than syntopic populations of the same species (x = 
7.75 mm, S.E. = 0.25), but the difference is not statistically significant 
(t =-0.83). The allotopic population of C. holsingeri in Harman Cave is 
smaller (x = 2.8 mm) than syntopic populations (x = 3.1 mm, S.E. = 
0.1), but the difference is not statistically significant (t = 1.42). On the 
other hand, the allotopic population of C. holsingeri in Linwood Cave 
(which is in a different drainage basin) is larger (x = 5.7 mm) than syn- 
topic populations, and the difference is statistically significant (t = 1 1.34, 
