Gadomski and Caddell: Effect of temperature on egg development and survival 



43 



ture and egg stage (I-VIII). Stage IX (hatched) was 

 not included in these predictive equations because 

 this would not be a recognizable stage in field-col- 

 lected eggs. 



Chorion, yolk mass, and oil globule diameters were 

 measured for 30 eggs of each species over a range of 

 developmental stages. To allow for shrinkage, mea- 

 surements were made after eggs had remained in 

 4% formalin for at least one year. 



Results 



Eggs of all four species were buoyant and spherical 

 and had a smooth chorion, unsegmented yolk, and a 

 single oil globule. Mean chorion diameter was sig- 

 nificantly different between all species ranging from 

 0.75 mm (California halibut) to 0.85 mm (barred sand 

 bass) (Table 1; one-way analysis of variance 

 ( ANOVA), F 3 U6 =184.6; P<6.001 ). Only the mean yolk 

 diameter of California halibut differed significantly 

 from mean yolk diameters of the other species (Table 

 1; one-way ANOVA, F 3 108 =29.3; P<0.001). Mean oil 

 globule diameter varied from 0.12 mm (California 

 halibut) to 0.22 mm (white croaker) and was signifi- 

 cantly different between all species (Table 1; one-way 



ANOVA, F 3 108 =320.2; P<0.001 ). 



Embryonic development in relation to germ-ring 

 migration and blastopore closure occurred sooner for 



riod was obtained by combining eggs sampled 

 from the two jars held at each temperature. A 

 sample was determined to be at an egg stage 

 when >50% of the eggs in that sample met the 

 stage criteria. Because the duration of each 

 stage differed, we standardized our presenta- 

 tion of egg development in accordance with 

 Ketchen (1956), by spacing egg stages on the 

 vertical axis proportional to the rate of devel- 

 opment at a median survival temperature: 20"C 

 for barred sand bass and fantail sole (Figs. 1 

 and 2), and 16°C for white croaker and Califor- 

 nia halibut (Figs. 3 and 4). 



For each species at each temperature, the rate 

 of egg development was quantified as the slope 

 (p) of age regressed on stage, where age is time 

 (in hours) from fertilization to the nine devel- 

 opmental stages (I-IX). Embryonic developmen- 

 tal stages E, O, and S were not considered in 

 this analysis. Lines were fitted by using least- 

 squares regression forced through the origin. 

 Multiple regression equations were also fitted 

 for each of the four species to estimate age (in 

 hours) of an egg, given developmental tempera- 



2 6 10 14 18 22 26 30 34 38 42 46 50 54 58 62 66 70 74 

 Hours after fertilization 



Figure 1 



Development of eggs of barred sand bass, Paralabrax nebulifer, 

 at five temperatures. Eggs died at 8°C with no cell division evi- 

 dent. At 12°C, many eggs sampled were dead, resulting in a lack 

 of egg-stage observations for some time periods (unconnected 

 points). Hatched larvae from eggs at 12"C were abnormal and 

 soon died. Descriptions of egg stages are presented in Table 2. 



