FISHERY BULLETIN: VOL. 81, NO. 4 



increments in the otolith core of French grunt, 

 Haemulon flauolineatum, and speculated that they 

 may have been deposited during the egg stage, and 

 Radtke and Dean (1982) found that laboratory- 

 reared mummichogs, Fundulus heteroclitus, deposited 

 2 or 3 increments prior to hatching. It therefore 

 seems reasonable to infer that the 2 or 3 irregular 

 increments found inside the nucleus of cod otoliths 

 by SEM analysis also were formed at some point dur- 

 ing the approximately 20-d incubation period 

 (Radtke and Waiwood 1980) prior to hatching. The 

 subsequent exclusion of these innermost increments 

 from the growth model, combined with the calculated 

 intercept values, supports the findings of Radtke and 

 Waiwood (1980) that cod larvae begin daily growth- 

 ring formation at hatch. The similarity of the otoliths 

 and growth patterns of cod and haddock warrants the 

 conclusion that daily increment formation for had- 

 dock larvae also begins at hatch 



Checks or check rings, exceptionally dark and thick 

 discontinuous zones, are the primary landmarks pre- 

 sent in larval otoliths and have been used extensively 

 to date growth disturbances, injury, poor feeding, 

 and environmental conditions (Taubert and Coble 

 1977; Radtke and Waiwood 1980; Lough et al. 1982; 

 Radtke and Dean 1982; Geffen 1982). If the discon- 

 tinuous zone represents a fixed period during the day 

 when calcium deposition into the organic matrix 

 slows or ceases as postulated by Watabe et al. (1982), 

 check rings would be a reflection of a disturbance in 

 the daily metabolism of the larval fish. Change from 

 one life history stage to another would represent one 

 such possible disturbance. The nuclear check 

 formed at hatching and the yolk-sac check were the 

 two major check rings found in the otoliths of cod and 

 haddock larvae. Radtke and Waiwood (1980) noted 

 the presence of a check deposited at the time of yolk- 

 sac absorption in laboratory-reared cod larvae, and 

 they remarked on its similarity to the metamorphic 

 checks found by Bailey et al. (1977) in capelin, 

 Mallotus uillosus. Although the increments deposited 

 between the nuclear and yolk-sac checks were faint 

 and difficult to discern with the light microscope, it 

 was possible to make counts on the clearer 

 specimens and by SEM techniques. The average 

 number of increments in this zone for both cod and 

 haddock larvae was 3 with a range of 2-8. Radtke and 

 Waiwood (1980) found that it took cod larvae 4 d at 

 4°C to reach yolk-sac absorption, and in Laurence's 

 (1978) rearing experiments cod and haddock had 

 completed yolk-absorption by 7 d at 7°C, Ellertsen 

 et aL (1980) found that yolk exhaustion for Arcto- 

 Norwegian cod occurred within 8 d of hatching at 5° C 

 and that feeding was initiated 2-4 d prior to this when 



the jaw apparatus became completely functional. 

 This close correspondence between rearing experi- 

 ments and the present field study supports the pro- 

 position that the observed check was deposited at, or 

 shortly before, the completion of yolk absorption and 

 that the increments between it and the nuclear check 

 approximate the number of days spent in the yolk- sac 

 phase. 



Because the early life history of the Atlantic cod and 

 haddock larvae analyzed took place over an extended 

 area and period of time, it is impossible to accurately 

 know the temperatures at which hatching and larval 

 growth occurred. However, based on the tem- 

 peratures recorded during the two collecting cruises, 

 it can be estimated that water temperature ranged 

 from 4° to 6° C for cod and 6° to 8° C for haddock. The 

 derived growth models estimate that cod grew in 

 length at an average of 2.6%/d and haddock at 3.4% d 

 Anderson (1982) recently found that a rise in the 

 growth rate of cod larvae collected on the Flemish 

 Cap "closely paralleled increasing surface water 

 temperatures." Laboratory rearing experiments 

 (Laurence 1978; Laurence et aL 1981) also found 

 that growth for cod and haddock was positively cor- 

 related with temperature and that cod larvae grew in 

 lengthabout0.62%/dfasterat7°Cthanat5°C.Ifthis 

 percentage increase is incorporated into the cod 

 growth model, the average rate of growth would be 

 3.2%/d, and most of the difference between the cod 

 and haddock growth curves could be accounted for 

 (Fig. 8). The similarity of these results agrees with 

 that of Laurence (1978) in which he found no major 

 differences between the growth rate of the two 

 species at similar food levels, although haddock 

 growth was slightly higher than cod at all 

 temperatures. 



Ring deposition appears to be endogenously con- 

 trolled with the daily light-dark cycle acting as the 

 primary triggering mechanism (Watabe et al. 1982). 

 Environmental variables, such as temperature and 

 day length, and feeding success are thought to be 

 reflected in the relative width of the individual 

 increments (Taubert and Coble 1977; Radtke 1980; 

 Methot 1981). Since the thin otolith increments 

 deposited immediately after the yolk-sac check ring 

 represent a short period of time and are a constant 

 feature throughout the extended hatching season, 

 these thin daily increments were probably a reflec- 

 tion of the adjustment to active feeding rather than to 

 an abrupt change in the environment. It is also known 

 that the length of the yolk-sac phase and the adjust- 

 ment to completely active feeding are somewhat 

 dependent on temperature (Laurence 1978; 

 Ellertsen et al. 1980); therefore, the greater number 



834 



