a large difference in size between immature ova, 

 developing intermediate ova, and maturing ova, 

 the spawning period will be short and definite. If 

 there is only a gradual change in size between 

 these groups of ova, individual fish may spawn 

 several times over an extended period (Hickling 

 and Rutenberg 1936; de Vlaming 1974). 



Since there was a gradual change in size be- 

 tween groups of ova, and since the number and 

 diameter of maturing or ripe ova for fish of the 

 same length varied considerably, we inferred that 

 Gulf menhaden were intermittent, or fractional 

 spawn ers. The number of maturing ova did not 

 change markedly from month to month, or even 

 within the same month. The ripe ova, after being 

 spawned, probably are replaced by a group of the 

 largest maturing ova which in turn are replaced 

 by a group of intermediate ova. Perhaps four or 

 five different groups of ova ripen and are spawned 

 during a single spawning season, although the 

 exact number cannot be estimated. 



Higham and Nicholson (1964) stated that, from 

 available evidence, it is impossible to decide con- 

 clusively the frequency of spawning of individual 

 Atlantic menhaden but they favored the 

 hypothesis of maturation and fractional spawning 

 of more than one group of ova during the season. 



Combs (1969) concluded that B. patronus 

 spawns several times from October to February. 

 He found that over a period of months spawnable 

 oocytes occurred together with advanced stages 

 of ova that were potentially spawnable. He de- 

 scribed the histological events in the development 

 of Gulf menhaden ova from formation to maturity 

 and found that once the provisional yolk had 

 formed, ova lost all potential to remain in the 

 ovary and had to complete their development prior 

 to spawning or be aborted. 



Number of Ova Spawned 



Since analysis of variance tests showed no sig- 

 nificant difference in the size or number of matur- 

 ing ova in gram samples from the left and right 

 ovaries, we used either ovary for measurements 

 and counts. (Counts were made of the number of 

 maturing ova in a sample of 0.1 g or less from an 

 ovary. ) The number of maturing or ripe ova in each 

 female was estimated by dividing the combined 

 weight of the left and right ovaries by the sample 

 weight and multiplying this number by the 

 number of maturing or ripe ova in the sample. 



If fractional spawning occurs, the number of ova 



estimated to have been spawned by fish of any 

 given age or size would necessarily be minimal, 

 since some ova probably would have been spawned 

 by the time some ovaries were collected. Fractional 

 spawning also would increase the variability in 

 the number of ova estimated for fish of the same 

 age or size (Bagenal and Braum 1971). Of the 70 

 maturing females that we could age, 44 were age 1, 

 20 were age 2, 5 were age 3, and 1 was age 4. The 

 mean number of ova and its standard error for 

 each age-group respectively were 37,100 ±3,467; 

 47,900±5,038; 61,800±9,486, and 151,000. 



Three relationships that are most useftil in ex- 

 plaining and understanding population dynamics 

 of a species are those of fecundity with age, length, 

 and weight. To determine what mathematical 

 models would be most appropriate in describing 

 these relations, we used Statistical Analysis Sys- 

 tem"* to test various statistical regression models. 

 We chose those which had the greater r^ values 

 and the minimum deviations from the regression 

 line. In the following models F - fecundity, A - 

 age, L = fork length, and W = body weight. For 

 fecundity on age (Figure 3): 



■•Statistical Analysis System; Barr, Goodnight, Sail and Hel- 

 wig, SAS Institute Inc., PO. Box 10066, Raleigh, NC 27506. 



c 

 o 



< 

 > 

 O 



Z 



1601- 



150- 



140- 



130- 



120- 



110- 



100- 



90- 



80- 



70- 



60- 



50- 



40- 



30- 



20- 



10- 



— 



F = 21402(1.44156)*^^ 



AGE 



Figure 3. — Relation between number of maturing ova and i 

 for Gulf menhaden. 



950 



