FISHERY BULLETIN: VOL, 77, NO. 4 



which maintained temperatures within ±0.5° C of 

 10°, 15°, 20°, and 25° C. Before the experiments 

 began the embryos had been reared for 5 h at 20.5° 

 C and 23.5%o, and had reached the early blastodisc 

 stage of development. By the time of the next ob- 

 servations, about 5 h later, water temperature in 

 the culture dishes had reached the desired incuba- 

 tion temperatures. Development data were 

 mathematically adjusted for the delay in attain- 

 ing experimental temperatures — see Results. 



Developing embryos were observed with a 

 stereomicroscope at intervals of about 4-6 h. Dead 

 embryos were counted and removed, and stage of 

 development of most live embryos was recorded. 

 The basic nine-stage staging classification (Table 

 1 ) used in this research was similar to that used by 

 Farris (1958, 1961). I refined this staging scheme 

 by distinguishing among an early, middle, and 

 two late periods within the nine stages. Early and 

 late periods within a stage of development were 

 quantified by subtracting 0.2 from, and adding 0.3, 

 and 0.5, respectively, to the stage number. 



Table l.— Fish embryo stages of development used (orBrevoor- 

 tia tymnnus. 



Stage 



Description 



1 Fertilized eggs prior to cell division to 8-cell stage 



2 Eight-cell stage to completion of blastodisc formaton 



3 Blastodisc formation to germ ring halfway around egg 



4 Germ ring halfway around egg to |usl prior to blastopore closure 



5 Blastopore closure to tall bud beginning to separate from the yolk 



6 Tail bud free of yolk to caudal one-eighth of body free of the yolk 



7 Caudal one-eighth of body free of yolk to caudal one-fourth of body free of 



yolk 



8 Caudal one-fourth of body free of yolk to fin fold moderately wide and tail 



portion of embryo rotated out of embryonic axis and tail approaching head 



9 Tip of tail approaching head to hatching 



RESULTS 



Survival to hatching in theB. tyrannus rearing 

 experiments was low, particularly at the 10° C 

 incubation temperature. Temperature but not sa- 

 linity had a significant effect on embryonic mor- 

 tality in these experiments (Table 2). Testing by a 

 posteriori sum of squares simultaneous test proce- 

 dure (SS-STP) (Sokal and Rohlf 1969) revealed 

 that embryonic mortality was significantly great- 

 er {P< 0.05) at the 10° C incubation temperature, 

 and not significantly different tP 0.05) at 15°, 20°, 

 and 25° C. 



During the experiments it became clear that 

 most embryo deaths occurred during the first half 

 of embryogenesis, and in particular, just prior to 

 blastopore closure (prior to stage 5 in the staging 

 classification used in this research) (Table 3). 



Table 2. — Data on mortality (upper) and two-way ANOVA 

 (lower) with replication to determine the effect of temperature 

 and salinity onBrevoortia tyrannus embryos. Mortality data are 

 the proportion of dead embryos (p). ANOVA performed on angu- 

 lar transformed (arc sine p"^) data. 



•• =iP 0.01. ns = P 0.05. 



Table 3.— Percent (cumulative) mortality o( Brevoortia lyran- 

 nus embryos prior to and after blastopore closure (stage 5 1 reared 

 at four temperatures in the laboratory. 



Mortality to stage 5 



Mortality to stage 9 

 (%J 



Temperature 

 ( C) 



10 

 15 

 20 

 25 



When the difference in frequency of embryo 

 deaths prior to stage 5 and from stage 5 to stage 9 

 was tested at each of the four incubation tempera- 

 tures, mortality was not significantly different 

 throughout development at 10° C, but was sig- 

 nificantly greater prior to stage 5 at the 15°, 20°, 

 and 25° C incubation temperatures (Table 4). 



Data on embryonic age and stage of develop- 

 ment were virtually identical in replicate culture 

 dishes and for embryos reared at the same tem- 

 perature but different salinity. Therefore, 

 analysis was restricted to temperature effects on 

 development rate. 



Table 4. — Significance of difference in deaths of Breevortia 

 tyrannus embryos prior to and after blastopore closure (stage 5) 

 at four temperatures. Test is 2  2 test of independence using the 

 G-statistic with Yates' correction. 



ns = P -0 05: " = P<0 01. 



944 



