AN EASTERN PENNSYLVANIA TROUT STREAM 775 



Company, 1969), and the resulting color was read on the Spectronic 

 20. Calcium hardness was determined using titration with disodium 

 dihydrogen-1, 2-cyclohexanediamine-tetra-acetate (CDTA) (Hatch 

 Chemical Company, 1969). 



The benthic macro invertebrate fauna was sampled in riffle areas 

 with a Surber square-foot sampler (Surber, 1936). The mesh size was 

 9 threads/cm. Two Surber samples were taken at each station, 

 sampling 0.18 m^ of stream bottom. The mean of the two samples 

 was used in calculating the macro invertebrate parameters. The 

 sampling stations (28, 12, 10, and 2) were similar in depth (15 to 

 25 cm) and substrate. The benthic macro invertebrates were removed 

 from the rocks with a brush, partially sorted from rubble, and 

 preserved in 70% ethyl alcohol. Organisms were identified to genus 

 when possible (Pennak, 1953; Usinger, 1956). Members of the 

 families Elmidae (Coleoptera) and Chironomidae (Diptera) were 

 identified by family. Because Elmidae and Chironomidae were not 

 identified to genus, the diversity index may be biased and slightly 

 low. The same level of resolution was used for all microinvertebrate 

 samples. 



The diversity index was derived from the formula of Shannon 

 (Shannon and Weaver, 1963): 



n 2n 



where d is diversity, nj is the number of individuals in the ith taxa, 

 and n is the total number of individuals. Wilhm and Dorris (1966; 

 1968) proposed the use of a diversity index based on information 

 theory to estimate the diversity of aquatic macro invertebrates. The 

 reliability of a mean diversity index should increase with several 

 collections at each station (Denoncourt and Polk, 1975). 



Water quality and macroinvertebrate data were analyzed statis- 

 tically using the Lehigh Amalgamated Packages for Statistics 

 (LEAPS) Program (Palchak and Koko, 1973). Mean, standard 

 deviation, and correlation matrix were calculated. Significant levels 

 of correlations were calculated from the tables of Edwards (1950). 



RESULTS 



Chemical and Physical Parameters 



The mean temperature at the eight stations on Bushkill Creek did 

 not increase significantly downstream. The mean temperature at 



