FISHERY BULLETIN: VOL. 83, NO. 2 



literature exists on follicular atresia in fishes (re- 

 viewed by Saidapur 1978) and stages of atretic 

 oocytes and follicles have been defined 

 (Bretschneider and Duyvene de Wit 1947; Lambert 

 1970a), but only the atretic rates in the guppy have 

 been measured (Lambert 1970a). Considerable at- 

 tention has been devoted to stages of atretic folli- 

 cles (corpora atretica or "preovulatory corpora 

 lutea") because of a presumed endocrine function 

 (see reviews by Hoar 1965; Byskov 1978). The sea- 

 sonal occurrence of atretic oocytes and follicles is 

 often discussed as part of a general description of 

 seasonal changes in the ovary of marine fish; see, 

 for example, cycles described for the gobiid, Gil- 

 lichthys mirabilis (de Vlaming 1972); plaice, 

 Pleuronectes platessa (Barr 1963); Paracentropris- 

 tis cabrilla (Zanuy 1977); and three species of 

 Epinephelus (Bouain and Siau 1983). The propor- 

 tion of females with atretic ovaries or the numbers 

 of atretic oocytes within the ovary is given less 

 often, but a few reports exist. For example, atresia 

 ranged from to 69c of the oocytes in female had- 

 dock, Melanogrammus aeglefinus (L.) (Robb 

 1982); corpora atretica increased to about 39c of the 

 oocytes during the postspawning period of the dab, 

 Limanda limanda (L.) (Htun-Han 1978); and atre- 

 tic oocytes varied from 13% of yolked oocytes dur- 

 ing the prespawning period to 100% during the 

 postspawning period of the snapper, Chrysophrys 

 auratus (F.) (Crossland 1977). Some attention has 

 been given to the issue of whether or not atretic 

 rates can account for differences in fecundity 

 among females fed high and low rations. It ap- 

 pears that ration-related differences in fecundity 

 are more closely tied to production rates of oocytes 

 rather than atretic rates (Tyler and Dunn 1976; 

 Wootton 1979). In summary, our literature review 

 indicates that ovarian atresia has yet to be used for 

 quantitative estimation of any reproductive pro- 

 cesses in marine fish populations, although it has 

 been used in general descriptions of the seasonal- 

 ity of reproduction for many years. 



METHODS 

 Laboratory Experiment 



Adult northern anchovy captured by commer- 

 cial bait fishermen on 23 February 1982 were kept 

 in a live car in San Diego Bay. Three days later 

 about 1,000 fish averaging 104 mm SL (9.50 g) 

 were taken to the laboratory and held in a 4.6 m 

 diameter pool (1 m deep ) at which time the first fish 

 sample was taken. Over the first 34 d in captivity, 



120 



samples of 18-24 females were taken at 3-4 d inter- 

 vals with the final sample taken after 62 d in 

 captivity. The temperature of the seawater ranged 

 from 15.5° to 16.5°C. 



The fish were not fed during the first 27 d in 

 captivity because starvation was used to trigger 

 the resorption of the ovary; thereafter they were 

 fed daily. On the 27th day of starvation the ovaries 

 had regressed from A9c of female body weight to 

 0.8% and feeding was resumed because we wished 

 to learn how long the atretic characters would last 

 once the fish began to feed. 



In our calculations of atretic rates of laboratory 

 females, we assumed that all the females at the 

 time of capture had active ovaries without atresia, 

 although no samples were taken until 3 d after the 

 fish were captured. Only 3% of the 1,680 females 

 taken in a survey conducted at the same time (28 

 January-8 March 1982) had atretic ovaries, and it 

 was prominent in only 0.1% of the females (50% or 

 more of yolked oocytes were affected). Ninety-six 

 percent of the females in our first sample (taken 3 

 d after capture) had yolked eggs, and half of them 

 had no atresia. 



All females sampled during the course of the 

 laboratory experiment were weighed and mea- 

 sured, and the ovary removed, weighed, and a sec- 

 tion removed for histological analysis. Ovaries 

 were fixed in 10% neutral buffered Formalin^ and 

 embedded in Paraplast. Histological sections were 

 cut at 6 /um and stained with Harris hematoxylin 

 followed by eosin counterstain. 



Sea Data 



The ovaries of northern anchovy taken in trawl 

 surveys used for biomass estimation (Stauffer and 

 Picquelle^) and various other collections from 

 commercial seiners and midwater trawls were his- 

 tologically examined. The number of females 

 examined per catch (trawl, purse seine, or lam- 

 para net) has varied from 10 to 20. Some collec- 

 tions were quite small, especially those taken out- 

 side the main spawning season in the Southern 

 California Bight; these small collections may con- 

 sist of only two catches, whereas those taken dur- 

 ing the main spawning months (February-March) 



^Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



■■'Stauffer, G., and S. Picquelle. The 1980 and 1981 egg produc- 

 tion estimates of anchovy spawning biomass. Unpubl. manu- 

 scr. Southwest Fisheries Center La Jolla Laboratory, National 

 Marine Fisheries Service, NOAA, P.O. Box 271, La Jolla, CA 

 92038. 



