124 



FISHERY BXJLLETm OF THE FISH AND WILDLIFE SERVICE 



Scott found that the number of eggs tended to 

 increase with length and weight of the female. 

 He pointed out, however, that the apparent 

 decrease in the number of eggs with increasing age 

 of the fish, as shown by his data, may be in error, 

 since all age group III fish were ripe when collected 

 and unknown numbers of eggs were lost in handling 

 them. Stone (1938) estimated (by the volumetric 

 method) the number of eggs of 104 Irondequoit 

 Bay lake herring to average 24,095; the mean 

 length of these fish was 13.4 inches. The average 

 number of eggs per fish in the different age groups 

 ranged from 13,723 in the 2-year-olds (average 

 length, 11.9 in.) to 48,999 in the 8-year-olds (aver- 

 age length, 16.7 in.). 



The number of eggs was estimated by the dry- 

 weight method for 72 Green Bay lake herring. 

 This method was developed by Paul H. Eschmeyer 

 and George F. Lunger, of the Service's Great 

 Lakes Fishery Investigations, in studies of the 

 fecundity of lake trout; they have not pub- 

 lished an account of the procedure. The general 

 procedure with lake herring ovaries (which differs 

 somewhat from that followed by Eschmeyer and 

 Lunger for lake trout) is as follows: 



The formalin-preserved ovaries are broken up thoroughly 

 and the larger pieces of connective tissue are removed; the 

 remaining materials are dried at 60° C. until there is no 

 further weight loss; a sample of 100 eggs is removed and 

 weighed (weighing is facilitated if the dried material is 

 allowed first to reach moisture equilibrium with the atmos- 

 phere) ; the total number of eggs is computed from ovary 

 weight and sample weight. 



The dependability of the method was tested by 

 making 38 estimates from 100-egg samples of 19 

 ovaries for which actual counts were made. The 

 advantage of the dry-weight method in reducing 

 error is clearly shown in the followmg comparisons: 



The number of eggs per fish in the Green Ba3'^ 

 lake herring (table 39; fig. 12) varied widely but 

 nevertheless exhibited a tendency to increase with 

 length of the fish. In the entire sample of 72 fish 

 with an average length of 11.2 inches, the average 

 number of eggs was 6,375. Fish of different age 

 but of the same length showed no diff'erence in egg 



10 I I 



TOTAL LENGTH 



Figure 12. — Relation between length of Green Bay lake 

 herring and number of eggs. The dots represent indi- 

 vidual fish and circles are averages for 0.3-inch length 

 groups; the line was fitted by least squares to the means 

 of the 0.3-inch groups. 



number (details of analysis are not presented here). 

 The relative number of eggs (expressed as number 

 of eggs per ounce of body weight), contrary to the 

 actual number, showed a downward trend with 

 increase in length (fig. 13). For the entire sample 

 (61 fish — 11 fish not weighed) the average number 

 of eggs per ounce of fish was 1,012 (table 39). 

 This value is below those for the ciscoes of Swains 



TOTAL LENGTH (INCHES) 



Figure 13. — Relation between length of Green Bay lake 

 herring and the number of eggs per ounce of body weight. 

 The dots represent individual fish and circles are aver- 

 ages for 0.3-inch length groups; the hne was fitted by 

 least squares to the means of the 0.3-inch groups. 



