WILLIAMS and CLARKE: BIOLOGY OF THE GOLD SPOT HERRING 



DISCUSSION 



The gold spot herring is an "annual" species which 

 grows rapidly, but matures at an early age while 

 small, and rarely lives more than a few months there- 

 after beyond maturity. Reproduction probably 

 occurs all year, but there is a definite peak in the sum- 

 mer. Less extensive data on what is probably the 

 same species 4 in the Marshall Islands (Hida and 

 Uchiyama 1977) indicate the same life history pat- 

 tern. Although seasonal changes in abundance and 

 size composition observed in Kaneohe Bay were in 

 part due to movements in and out of the bay, they cor- 

 relate generally with the reproductive season and the 

 1-yr life cycle and are probably indicative of changes 

 in the general population. 



Few comparable data exist on other Herklotsichthys 

 spp. or the closely related Sordinella spp. Maximum 

 size in many, e.g., H. punctatus (Marichamy 1974) 

 andS. marquesensis (Nakamura and Wilson 1970), is 

 about the same as inH. quadrimaculatus . Sardinella 

 jussieui (= 5. gibboso) and S. brachysoma (= S. 

 albella) also appear to have a maximum age of about 1 

 yr (Nair 1960; Okera 1974). Other aspects of the life 

 history of these small species may prove to be similar 

 to those of H. quadrimaculatus. Within the tropics 

 (and often cooccurring geographically with small 

 clupeid species) some Sardinella spp., e.g., 5. lon- 

 giceps (Nair 1960; Ritterbush 1974) and S. aurita 

 (Postel 1960), grow more than twice the length of H. 

 quadrimaculatus, live for several years, and may re- 

 produce more than one season. Such species are 

 qualitatively more similar to the Sardinops spp. from 

 higher latitudes than to the small tropical species. 

 Following the arguments of Murphy (1968) and 

 Leggett and Carscadden (1978), this indicates that 

 reproductive success is more consistent in the 

 smaller species and that even within the tropics, the 

 basic population regulatory mechanisms of closely 

 related species may be qualitatively different. 



Data on ova development and fecundity of other 

 species are also limited. The gold spot herring is very 

 similar in these respects to S. marquesensis (Naka- 

 mura and Wilson 1970). Relative fecundity of S. 

 aurita and 5. maderensis (calculated from the respec- 

 tive data of Kwan-Ming 1960 and Ben-Tuvia 1960) is 

 comparable with that of H. quadrimaculatus, but 

 because of their larger size, batch fecundity in these 

 species is much higher. Sardinella longiceps, how- 





4 Hida and Uchiyama identified their specimens as H. punctatus. W. 

 J. Baldwin of the Hawaii Institute of Marine Biology has, however, 

 examined numerous specimens collected from the Marshalls during 

 Hida and Uchiyama's study and found them all to be H. 

 quadrimaculatus. 



ever, is not only larger than H. quadrimaculatus, but 

 has considerably higher relative fecundity (940- 

 2,090 ova/g, calculated from Ritterbush 1974). 

 These differences again indicate that reproductive 

 success of the large species is lower or less predict- 

 able than in the small, less fecund species. 



Neither our results nor those of Leary et al. (1975) 

 on the Hawaiian anchovy offer any unequivocal 

 evidence that these tropical clupeoids ever ripen and 

 spawn more than one batch of eggs. The age-length 

 data (Fig. 6) indicate that 90 mm SL gold spot her- 

 ring, some of which appeared ready to spawn, are on- 

 ly about 1-mo older than clearly immature fish 75-80 

 mm SL and that females live 3-4 mo longer after 

 reaching 90 mm SL. Thus the herring appear capable 

 of spawning several times during their reproductive 

 life span, but we have no evidence that they actually 

 do so. The wide range in maximum size of ova found 

 in herring > 90 mm SL could result either from multi- 

 ple spawning or variation in the size and age at which 

 females begin to ripen ova for a single spawning. The 

 absence of spent females indicates a single spawning 

 and high mortality of spawners; however, no females 

 with hydrated ova were observed either. Since 

 spawning appears to occur away from the areas 

 where we made our collections, it is possible that 

 both hydration and recovery from spawning are so 

 rapid that there is no trace of either in fishes about to 

 leave for or just returned from spawning. 



Our results, as well as those of the great majority of 

 studies on clupeoid fishes in which appropriate data 

 have been examined and presented, show that sex 

 ratio of fish caught by a given gear and at a given time 

 frequently deviates from 1:1. Assuming the actual 

 sex ratio of adult gold spot herring is 1:1, the day- 

 night differences in sex ratio indicate that males are 

 more likely to move offshore at night and perhaps to 

 remain there during the day. Such behavior is most 

 likely related to spawning. The gold spot herring, like 

 most other clupeoids, probably spawns at night, and 

 Hunter and Goldberg (1980) have presented evidence 

 that spawning schools of pelagic clupeoids are 

 dominated by males. 



Like many other clupeids, female gold spot herring 

 appear to reach sizes larger than the largest male. 

 Although there may be sexual differences in growth, 

 our data indicate that female herring reach larger 

 sizes because they live longer. If spawning incurs in- 

 creased mortality due to either energy requirements 

 or exposure to predation, the males may not live as 

 long because they spawn more frequently than do 

 females. 



Discussion of the introduction and spread of H. 

 quadrimaculatus in Hawaii is limited because we do 



595 



