tion to estimate hours to midpoint of the hatching 

 interval. 



Methods 



On 4 April 1981, adult walleye pollock were trawled 

 in Stephens Passage, southeastern Alaska (lat. 

 58°17'N, long. 134°42'W). One sexually mature 

 female and one adult male were kept alive until 

 spawned artifically in the laboratory, about 3 h later. 

 The eggs were removed and fertilized according to 

 the "dry" method (Kinne 1977). 



Embryos were incubated about 4 h at 6°C to ensure 

 that only viable eggs were used in the experiment. At 

 the end of the 4-h period, each egg was examined 

 visually before being transferred to an incubator. 

 The incubators were then placed in water baths of the 

 various experimental temperatures. By the time of 

 the first observation, about 4 h later, water tempera- 

 ture in the incubators had reached experimental tem- 

 peratures. Six groups of about 200 embryos each 

 were incubated separately in identical incubators. 

 Two of the groups were incubated at 6°C to provide 

 an estimate of residual error. The other groups were 

 incubated at 2°, 5°, 8°, or 11°C. These incubation 

 temperatures fall within the range of incubation tem- 

 peratures walleye pollock embryos usually encounter 

 at sea. 



The incubators were 3.5 1 cylindrical containers 

 made of black ABS (acrylonitrile-butadiene-styrene) 

 plastic. Each incubator was filled with 2,500 ml of 

 seawater (salinity 32.5%o) and covered with a clear 

 Plexiglas 1 cover 3.2 mm thick. Seawater was not 

 changed during the experiment. The incubators were 

 kept in thermostatically controlled water baths at 

 temperatures within ±0.2°C of the treatment tem- 

 perature. Photoperiodwas 12-h illumination and 12- 

 h darkness. Illumination at the cover of each 

 container was 170 lux (15.8 fc) from 60- W Soft White 

 incandescent bulbs. 



'Reference to trade name does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



Embryos were removed from the containers at 

 about 4-h intervals for the first day, then at least daily 

 thereafter. The embryos were preserved in Gilson's 

 fixative and later classified according to a seven- 

 stage classification system, based on easily 

 recognized developmental features (Table 1). For 

 quantitative purposes, we used whole integers for 

 midpoint of stages (Ferraro 1980). The start and end 

 of each stage is quantified by adding or subtracting 

 0.5 to the stage number. We also recorded the time 

 when larvae were first observed in culture vessels 

 and the time when all embryos had hatched. 



TABLE 1. — Stages of embryo development used for walleye 



pollock. 



Stage I: Fertilized egg without germ disc. 



Stage II: From germ disc through 32-cell stage. 



Stage III: From 64-cell stage to completion of blastoderm. 



Stage IV: From germ ring to germ ring enveloping egg, but before completion 



of epiboly. 

 Stage V: From completion of epiboly to embryo extending at least three- fourths 



way around yolk and caudal region not directed off axis of embryo. 



Stage VI: Fromcaudalregionslightlyoffaxisofembryotomarkedlyoffaxts, and 



tip of tail just reaching embryo head. 



Stage VII: Tip of tail extending beyond embryo head to hatching. 



Statistical Analyses 



Because the embryos were incubated for 4 h at 6°C 

 before they were placed in the incubation containers, 

 the data were corrected (Table 2) for the delay in at- 

 taining experimental temperatures using Ferraro's 

 (1980) method. In our experiment, the correction fac- 

 tor was the ratio of development time to Stage VII for 

 embryos incubated at 2°, 5°, 8°, and 1 1°C relative to 

 development time to Stage VII at 6°C (see Table 2 for 

 derivations and Table 3 for corrected midpoint and 

 duration of each stage). 



We developed a general predictive equation with 

 temperature- dependent coefficients to estimate the 

 age of a walleye pollock embryo, given a stage of 

 development and incubation temperature over the 

 range of 2°-ll°C. For each experimental tempera- 

 ture, the midpoint age (in hours) for each develop- 

 mental stage was plotted against the stage (Fig. 1), 



Table 2. — Derivation of correction factors to adjust development data for differences between 

 experimental and pre-experimental temperatures (°C) in walleye pollock. See Table 1 for descrip- 

 tion of stages. 



891 



