492 MANN 



genesis in three species of bivalves when all other environmental 

 factors (salinity, v^ater flow, and phytoplankton concentration) were 

 constant. Four populations of each of three species (Crassostrea 

 gigas, Ostrea edulis, and Tapes japonica), matched by numbers of 

 animals and mean individual live weight, were selected from common 

 parental stocks. Populations of each species were maintained at 

 sustained temperatures of 12, 15, 18, and 21°C for 19 weeks. 

 Samples of 15 to 20 animals of each species were sacrificed at 

 biweekly intervals for assay of live weight, freeze-dried meat weight, 

 dry-shell weight (100°C for 24 hr), and the following biochemical 

 components: 



1. Carbon and nitrogen assayed by a Perkin— Elmer CHN analyzer 



2. Carbohydrate assayed by the anthrone method of Strickland 

 and Parsons (1968) after extraction in cold 5% wt/vol trichloroacetic 

 acid 



3. Ash, defined as the material remaining after ignition for 12 hr 

 at 500°C 



Within each species— temperature combination, data for each of 

 the biochemical indexes (% carbohydrate, carbohydrate/N ratio, C/N 

 ratio, and % organic content) were compared to the gravimetric 

 dry-meat/dry-shell condition index by linear regression analysis. 

 Since no significant differences in values of both slope and intercept 

 were found between temperatures within each species, data for all 

 temperatures were pooled and recalculated to obtain one regression 

 line for each biochemical— condition index combination for each 

 species. 



RESULTS AND DISCUSSiOlM 



Linear regressions of biochemical and gravimetric indexes for 

 natural and laboratory populations are summarized in Tables 1 and 

 2, respectively. A good correlation is evident in all regressions for all 

 species tested. It is notable, however, that, although slope and 

 intercept values are similar for C. gigas in both natural and laboratory 

 populations for percentage carbohydrate and carbohydrate/N ratio 

 computations, only slope values are similar for natural and labora- 

 tory populations of O. edulis. This indicates a consistently higher 

 carbohydrate content in laboratory animals. A complete explanation 

 of this phenomenon must await further study, however. 



Previously a number of investigators examined the use of 

 glycogen content as an index of condition in bivalve molluscs, 

 usually by direct comparison of dry-meat/shell-cavity ratios (index 2) 

 with percentages of glycogen. Ingle (1949), comparing 34 samples of 



