112 Pamela S. Dolin and Donald C. Tarter 



approximately 8050 m east of Homestead, West Virginia, on State 

 Road n2. The study area is located 550 m from the Ohio River's south 

 edge. The swamp proper is 1450 m long and has a contiguous marsh on 

 the north end that continues another 1200 m, forming about 14 ha of 

 swamp forest. The mean water depth varies from 0.5 m during dry peri- 

 ods to 1.5 m during high waters. Elevation is 168 m above sea level. In 

 the permanently inundated areas, the only tree species is black willow, 

 Salix nigra Marsh. Button-bush saplings, Cephalanthus occidentalis L., 

 have the greatest density of any woody plant seedling in the swamp. 



The study began in March 1978 and ended in September 1980 

 (Dolin 1980). Larval collections of both species were made on a 

 monthly basis, by searching under moss, duckweed, and loose bark of 

 small stems and decaying logs. 



Monthly water chemistry was assessed with a Hach chemical kit, 

 Model AL-36B. The measurements taken included: hydrogen ion con- 

 centration (pH) determined colorimetrically; dissolved oxygen and car- 

 bon dioxide (mg/ 1); and total hardness and alkalinity (mg/ 1 of CaCO^). 

 Water temperature ranges were measured with a Taylor maximum- 

 minimum thermometer secured 0.3 m below the surface. 



Head width was measured to obtain growth records. Total width 

 was taken at the widest section of the head capsule, usually at the eyes. 

 Using a compound microscope and a caliper, the measurements were 

 taken to the nearest 0.1 mm. Monthly differences in head widths of 246 

 C. rastricornis and 1 12 C pectinicornis larvae were determined to show 

 the mean, range, standard deviation, and two standard errors of the 

 mean in a Hubbs and Perlmutter (1942) population range diagram. Per- 

 cent growth was calculated monthly by comparing the mean head width 

 values. Size classes were assessed by frequency histograms arranged in 

 0.5 mm head width categories. 



Generally, foreguts were examined from five larvae of each species 

 per month. After the head was removed with microdissecting scissors, 

 the abdomen was cut and the foregut extracted. The gut was then slit 

 ventrally and the contents placed on a clean slide for identification. The 

 number of foreguts containing the various taxa was recorded, and the 

 mean and percent frequency of occurrence were calculated. 



Chauliodes pupae were obtained by digging into the grooves of logs 

 that had started to desiccate due to receding spring waters. For both 

 species, a few larvae in the last instars were reared to the pupal and 

 adult stages. Rearing was accomplished by leaving the larvae in place 

 and quickly transporting the log section to aerated pint jars containing 

 water from the swamp. Larvae were placed one to a jar to help insure 

 pupation by reducing competition for foodstuffs, to prevent cannibal- 

 ism, and to determine when the individuals pupated. Pupae were placed 



