LOUGH KT AL.: AGE AND GROWTH OF LARVAL ATLANTIC HERRING 



field-caught larvae, which describes the length 

 at age from an initial mean hatching size of 5.7 

 mm to an upper asymptotic mean length of 30.9 

 mm, agrees well with average growth rate esti- 

 mates from other studies. Our field data begin 

 with a 7-increment larva of 12.6 mm SL, which 

 also is nearly identical to the mean length at 

 increment age estimated by the growth curve. 

 From the growth model a 7-increment larva is 

 estimated to be on average 25 d from hatch (5.7 

 mm) having grown at an average rate of 0.28 

 mm/d. This implies that increment deposition 

 does not occur daily over these 25 d or that varia- 

 tion in the timing of first increment deposition is 

 high. If one assumes daily increment deposition 

 from yolk-sac resorption (4.5 d), a 7-increment 

 larva would be 11.5 d old, inferring the larva has 

 grown at an average rate of 0.60 mm/d, which is 

 rather high based on field and laboratory esti- 

 mates. Herring larvae <15 mm have estimated 

 growth rates typically in the range of 0.25-0.30 

 mm/d with an upper limit of about 0.35 mm/d. 

 The apparent delay in increment formation 

 observed in the laboratory-reared herring larvae 

 after the first increment at yolk-sac resorption 

 may be due to rearing conditions, although we 

 have no reason to suspect they were less than 

 optimal. Other studies have shown that the for- 

 mation of daily growth increments can be af- 

 fected by variations in food ration, temperature, 

 light-dark cycle, age of fish, and stressful condi- 

 tions in general (see references in first section of 

 paper). Increment formation appears to be spe- 

 cies-specific and, for clupeoid species like En- 

 graulis mordax (Brothers et al. 1976) and Clupea 

 harengus (this study) with relatively small eggs 

 and short incubation period, the initial incre- 

 ments begin at the time of yolk-sac resorption 

 (Radtke and Waiwood 1980). A dark band or 

 check observed around the nucleus of most of the 

 larval otoliths collected in the field, but not ap- 

 parent in the laboratory-reared larvae, may cor- 

 respond to the time of yolk-sac resorption as 

 Radtke and Waiwood (1980) found for larval cod 

 otoliths. The nuclear check may be the result of 



i 



several thin increments grouped together. Uchi- 

 yama and Struhsaker (1981), working with 

 Pacific tunas, found that countable growth in- 

 crements were formed only when the fishes were 

 fed to satiation throughout the day. The nuclear 

 check and the succeeding 10 or so thin incre- 

 ments observed for the field-caught herring lar- 

 vae may be related to the inability of a first- 

 feeding larva to meet its maximum daily ration 



during the transition from its yolk supply to 

 exogenous feeding. Initial feeding efficiency is 

 low for herring larvae, <5% success at yolk-sac 

 resorption, but increases to about 40% 2 wk after 

 hatching and 70% after 5 wk (Blaxter and Staines 

 1971). Karris (1959) observed a rapid leveling off 

 of growth after hatch in four species of fish and 

 Zweifel and Lasker (1976), after fitting a two- 

 stage Laird-Gompertz growth curve to a number 

 of larval fish species, one from hatching to yolk- 

 sac resorption and another to more rapid growth 

 at the onset of feeding, suggested that this phe- 

 nomenon was almost universal in larval growth. 

 It is conceivable that during this period of re- 

 duced growth, increment deposition also may be 

 delayed or diminished until the larva learns to 

 capture sufficient numbers of prey and begins 

 growing rapidly again. 



Although larval herring appear to be very re- 

 sistant to the range of temperatures normally 

 encountered (Blaxter 1960), the effect of tem- 

 perature on increment formation is not known. 

 Water temperatures observed in herring spawn- 

 ing areas in the Gulf of Maine-Georges Bank 

 region are typically as high as 12°-14°C in early 

 autumn and decline to near 0°C in winter (Table 

 1), approaching their lower lethal limit (Graham 

 and Davis 1971; Chenoweth 1970). Yolk-sac utili- 

 zation in herring larvae is directly related to 

 water temperature (Blaxter 1956; Blaxter and 

 Hempel 1963, 1966; Blaxter and Ehrlich 1974) 

 and variations in water temperature at hatch 

 can reduce or extend the time to first feedi ng and 

 consequently, otolith increment formation. Yolk- 

 sac resorption is completed at 4-5 d at 10°C and 6 

 d at 8°C. Feeding of larvae is believed to com- 

 mence at or prior to the end of yolk-sac resorption 

 when the maximum body weight (excluding yolk 

 sac) is reached after about 3 d at 8°C and 2 d at 

 12°C. Larvae reared at 10°C would initiate feed- 

 ing 2-3 d after hatch. There is some evidence to 

 indicate that early larval herring growth is 

 better at higher temperatures (Blaxter 1962), 

 although food availability is considered the more 

 important factor in controlling growth processes 

 and survival of larval fish in general (May 1974). 

 Increment formation of the green sunfish, Le- 

 pomis cyanellus, could be stopped when growth 

 was slowed sufficiently by simulated winter con- 

 ditions (Taubert and Coble 1977). The slowing of 

 growth during the winter period observed for 

 larval herring in the Gulf of Maine-Georges 

 Bank region also may affect their increment for- 

 mation but further research will be required to 



197 



