FISHERY BULLETIN: VOL. 74, NO. 4 



presumed breeding period (Ponomareva 1963), 

 though eggs were abundant in August nearby in 

 the Sea of Okhotsk. To the north and south of the 

 eastern Aleutian Islands, Nemoto (1957) found 

 females of E. pacifica with attached spermato- 

 phores during July. 



Dominance of the southern California popula- 

 tion by the particular cohorts followed in the 

 analyses of growth tends to obscure the regular 

 contribution of small classes, including those of 

 fall-winter in which densities of larvae are usually 

 1,000-2,000/1,000 m'. 



Such continuous recruitment of variable inten- 

 sity is seen as an adaptation to midlatitude ir- 

 regularity in oceanographic conditions, both sea- 

 sonal and year-to-year, as compared with cycles at 

 high latitudes. Continuous recruitment permits 

 the stock to always include a wide spectrum of 

 sizes and maturity stages, providing a potential 

 for one or another to adapt to periods of poor 

 climate or food availability, of differing duration 

 or amplitude. For example, in 1954, a year of weak 

 upwelling, recruitment was all but limited to 

 June-July; nevertheless, spawning resumed at 

 high intensity during four different periods in 

 1955. 



Periodicity was observed in maxima of spawn- 

 ing and recruitment, and recruitment is appro- 

 priately out of phase with the inferred spawning 

 (Figure 4c), implying substantial synchrony 

 among breeders. Spawning apparently pulses at a 

 2-mo frequency during the period of maximum 

 gamete generation, which also must be the period 

 of maximum food use by breeders and larvae. This 

 is to be compared with the annual (or at most, 

 semiannual) frequency of breeding noted in the 

 subarctic North Pacific. Thus it appears possible 

 that, under optimal feeding conditions off south- 

 ern California, a female might spawn every 2 mo: 

 first at about 11.5 mm length (20-50 eggs), second 

 at 16 mm (50-200 eggs), and third at 20 mm 

 (100-400 eggs), during which time an individual 

 might be expected to produce a maximum of 650 

 eggs. This is compatible with an observation of 

 1,400 oocytes (all stages of development) in ovaries 

 of an E. pacifica in the springtime in the north- 

 eastern Pacific (Ponomareva 1963) where spawn- 

 ing is concentrated into one season, and with 

 Lasker's observation, reported in Mauchline and 

 Fisher (1969), that an E. pacifica from southern 

 California shed 230 eggs after capture. 



The long duration of maturity— probably half of 

 this species' life expectancy-further contributes 



to population stability and continuity. In conjunc- 

 tion with substantial horizontal transport, the 

 capacity to breed several times enhances genetic 

 integration across the distributional range. 



The first observations on growth in E. pacifica 

 were from specimens maintained in the laboratory 

 by Lasker (1966) at 10°C with excess food. In small 

 juveniles, growth was steady at 2.5 or 2.9 mm 

 during 2 mo, from about 5 to 8 mm length. In the 

 southern California field populations, growth of 

 juveniles of this size was consistently in the range 

 of 3-3.5 mm /mo. However, the 5309 cohort, having 

 reached 5 mm by the start of the fall period of 

 reduced growth, then grew only 3 mm in 1.5 mo. 



Larger E. pacifica were observed by Lasker to 

 grow somewhat slower. A 6.5-mm specimen grew 

 1.5 mm in 70 days, but added only 1.5 mm in 230 

 more days before dying, not having reached fully 

 adult size. A 7.9-mm specimen grew 1.5 mm in 75 

 days, an 8.0-mm specimen grew 1.5 mm in 130 

 days, and an 8.4-mm specimen grew 1.0 mm in 160 

 days. These rates are smaller than those for the 

 local field populations. They are closer to those 

 supposed for E. pacifica in the northeastern Pacific 

 where environmental enrichment is not by inter- 

 mittent upwelling but by winter mixing followed 

 by spring stability in the water column, hence not 

 a continuing process. 



In the analysis of growth, cohorts are considered 

 as normal L-F distributions representing broods 

 continuously hatched during a few days to a month 

 or more. Observation on duration of reproduction 

 is limited by the character of the sampling, here in 

 approximately 1-wk period with a 2-3 wk interval 

 between surveys. Only in a few of the months can 

 a pulse in recruitment by recognized as distinct to 

 that month. In most months, the larvae derive 

 from the beginning, continuation, or end of a 

 period of cohort formation which extends beyond 

 one survey period and into another. Recruitment 

 found less than in past or succeeding months is 

 neither recognizable initially as a cohort nor 

 traceable thereafter. 



The area's population, therefore, is constantly 

 polymodal in character, being compounded of 

 individuals belonging to different age-groups and 

 sexes. The possible difference in size between the 

 sexes after about 15 mm length was not taken into 

 consideration in the growth study. 



The simplest method of analyzing growth and 

 survival is that of following obvious modes, survey 

 to survey. This is probably the most significant 

 means biologically. Nevertheless, certain im- 



758 



