Cave Isopod Ecology 
105 
For laboratory stream studies the isopods were collected alive and 
measured using an ocular micrometer on a dissecting microscope. Collec- 
tions were sorted into two size classes: small (less than 4.5 mm long) and 
large (greater than 8 mm long). All individuals, including intermediate 
sizes, were eventually preserved, measured, and identified. All small in- 
dividuals were C. holsingeri and all large individuals were C. cannulus. 
Two experiments were done in an artificial stream in the laboratory 
(see Culver 1 97 1 for design of the stream). To measure the effect of gravel 
size on washout rate, two 10 cm by 10 cm areas of rocks were used in the 
stream. One consisted of gravels between 0.2 cm and 1 cm in diameter; 
the other consisted of gravels between 2.2 cm and 3.8 cm in diameter. The 
washout rate of each size class of isopods was measured for each size 
class of gravels. Five individuals were used in each run, and each run was 
repeated at least four times. Animals that washed out were collected at 
the end of the artificial stream bed. Individuals washing out in the first 30 
minutes were placed back in the stream, and then the number washed out 
after 12 hours was recorded. 
The second experiment measured interspecific effects on washout 
rate, as reflected by size differences of the two species. To measure in- 
trasize washout rates, 10 isopods of a single size class were placed in a 15 
cm X 15 cm section of gravels of various sizes, patterned after a cave 
stream. Each of these runs lasted 24 hours. Intersize class competition 
was measured by following the procedure outlined above, but using five 
small isopods and five large isopods. Thus, the total number of isopods 
present at the beginning of a run was always 10. Each run was repeated 
five times. 
All isopods collected were measured and identified. Identification 
posed some problems. Species can be separated only on the basis of the 
second male pleopod. Most males can be identified since the second 
pleopod sclerotizes at an early age. We found the following two charac- 
ters to be reliable; the comparative length of the endopodial groove, and 
the angle between the endopodite tip and the cannula. Since males are 
usually scarcer than females, and none of the populations are large (we 
collected less than 50, usually less than 25, in any one cave), it was 
necessary to try to include females. There is little if any size dimorphism, 
so we called all females C. holsingeri that were smaller than the largest C. 
holsingeri male, and all females C. cannulus that were larger than the 
smallest C. cannulus male. Ambiguous cases were randomly assigned to 
the two species. In practice, such random assignments were only 
necessary for Glady Cave. We also present size histograms to obviate 
identification problems. 
RESULTS 
Due to the relatively small number of isopods available, it was often 
necessary to use the same individual for more than one run in the arti- 
