DURBIN and DURBIN: ASSIMILATION EFFICIENCY OF ATLANTIC MENHADEN 



and DON concentrations in the tank was approx- 

 imately linear, and linear regressions were calcu- 

 lated to estimate the excretion rates. The ratio of 

 the DON:NH;i regression slopes was similar in all 

 experiments (overall means ADON;ANH;j = 

 0.437, (T = 0.088). We assumed that this average 

 value would provide the best estimate of DON 

 excretion in both feeding and nonfeeding fish, and 

 therefore multiplied the observed NH3-N excre- 

 tion rates by 0.437 to calculate DON-N excretion. 



The Atlantic menhaden did not excrete any 

 measurable quantities of nitrate, nitrite, or dis- 

 solved silicon. The latter was measured because of 

 the large amounts of particulate silicon in the 

 phytoplankton used as food. Nitrate, nitrite, and 

 silicate were determined in triplicate on selected 

 samples using methods described in Strickland 

 and Parsons (1972). 



Assimilation efficiency was calculated in two 

 ways: by calculating the overall assimilation effi- 

 ciency of C, N, and calories, and by following the 

 time course of changes in the assimilation effi- 

 ciency during each experiment. Both methods 

 were based on the assumption that all of the feces 

 were quantitatively collected. 



The overall assimilation of C, N, or calories was 

 determined by subtracting the total amount of 

 each of these constituents that remained in the 

 feces, from the amount in the food for each experi- 

 ment. The C assimilation efficiency (percent), for 

 example, was calculated: 



C assimilation = ^^^^ ^<££^(100). 



^food 



(1) 



In the second method, assimilation was calcu- 

 lated separately for each sample of feces. Since 

 silicon (Si) was not absorbed by the fish, the Si 

 content of each sample of feces was used to calcu- 

 late the quantity of food (C, N, calories) which 

 corresponded to each fecal sample. This was done 

 by using the C:Si, N:Si, and calories:Si ratios of the 

 food for each experiment, which were calculated 

 from the measurements of the C, N, and calories in 

 the food, and the total Si content of the feces. The 

 assimilation efficiency was then determined in a 

 manner analagous to Equation (1), using the 

 back-calculated amount of C, N, and calories in 

 the food which corresponded to each fecal sample, 

 and the actual amount of C, N, and calories re- 

 maining in the sample. 



The assumptions that all of the feces were col- 

 lected quantitatively, and that Si was not absorbed 



by the fish, were verified in a phytoplankton exper- 

 iment in which Si was used as a tracer. Subsam- 

 ples of the phytoplankton were centrifuged and 

 the Si content of the freeze-dried pellets then de- 

 termined. This gave an estimate of 1.87 g for the 

 total Si in the ration, which was very similar to 

 that actually collected in the feces, 1.98 g. 



A small but unknown amount of seawater re- 

 mained interstitially within the pellets of both the 

 phytoplankton and the feces, the salt content of 

 which may have contributed to the estimates of ash 

 content (Table 1). Because of this it was not possi- 

 ble to use ash content as a tracer to calculate 

 assimilation efficiency. However, it should be noted 

 that since the calculations of assimilation were 

 based on direct measurements of C, N, and Si 

 in the food and feces, they were not affected by 

 the presence of trace amounts of seawater in the 

 samples. 



Nitrogen excretion and feces elimination by At- 

 lantic menhaden are reported per gram dry 

 weight, as micrograms N per gram dry weight per 

 hour and milligrams per gram dry weight hour, 

 respectively. It should be noted that mea- 

 surements of oxygen consumption by the Atlantic 

 menhaden during these experiments were re- 

 ported in Durbin et al. (1981) as milligrams O2 

 per gram wet weight per hour, to conform to the 

 usual manner of reporting respiration rates in the 

 literature. 



RESULTS 



The chemical composition of Atlantic menha- 

 den, zooplankton, and D. brightwelli differed con- 

 siderably (Table 1). Atlantic menhaden contained 

 more C and calories, and less ash, per gram dry 

 weight than the plankton, and were intermediate 

 in N content between D. brightwelli and zooplank- 

 ton. The composition of the fecal pellets was con- 

 siderably altered from that of the food (Table 1). 

 Atlantic menhaden assimilated N more efficiently 

 than C, causing the C:N ratios of the fecal pellets 

 to be higher than in the food. 



The total dry weight of plankton fed to the 12 

 fish in each experiment ranged from 7.20 to 17.67 g 

 of zooplankton, and 9.60 to 94.79 g of phyto- 

 plankton (Table 2). However, since the chemical 

 composition of Atlantic menhaden differs from 

 that of the plankton, the food rations are equiva- 

 lent to different percentages of Atlantic menhaden 

 dry weight, C, N, and calories. For example, the 

 phytoplankton rations represented from 0.79% to 



603 



