542 



Fishery Bulletin 105(4) 



Similar analyses were conducted when a significant 

 difference in %A^, was identified between two groups. 

 For these comparisons, however, days preserved was 

 used as the metric of time preserved, since fresh weights 

 were taken for all these samples. 



Results 



Effect of preservation on weight of ovarian material 



All four experimental treatments had significant effects 

 on subsample weight for cod and plaice (Fig. 1), except 

 for plaice samples in the freezing treatment (P=0.12), 

 which also had the highest standard error (SE) ofVcA^^ 

 (SE = 4.93). For cod, standard error of %A^, was also 

 highest in the freezing treatment (SE = 3.37) along with 

 mean %4iy, (mean = 51.32). Mean %A^, was lowest for 

 cod in the ethanol treatment (mean=-4.21) but high- 

 est for plaice (mean=25.3). In the lobe-formalin treat- 

 ment there was a significant difference in %A^t ^^^ ^^ 

 cod (P=0.036), GOM cod (P<0.001), all cod combined 

 (P<0.001), and haddock (P=0.002). These data were not 

 available for plaice. 



Mean %4iy/ ^^^ ^°^ significantly different between 

 cod and haddock in the lobe-formalin treatment re- 

 gardless of whether haddock was compared with cod 

 samples restricted to GB (P=0.24, o = 0.03) or when 

 all cod were combined (P=0.28, a=0.05). Mean VcA^, of 

 subsamples was significantly different between cod and 

 plaice for the ethanol (P<0.001), freezing (P<0.001), and 

 Gilson's (P=0.0042) treatments (Table 2). There was 



no difference in % 4,y, between cod and plaice in the 

 subformalin treatment (P=0.81). This test marks the 

 only comparison between cod and plaice where no dif- 

 ference was found. There was a significant difference in 

 mean %A^f (P<0.001) for cod between GB (mean=1.41) 

 and the GOM (mean = 6.79) in the lobe-formalin treat- 

 ment. For cod from the GOM, there was a significant 

 (P<0.001) and nearly threefold difference in mean %\fi 

 between lobe-formalin (mean=6.79) and sub-formalin 

 (mean=19.32) treatments. 



For the 17 ovaries collected from cod caught in the 

 GOM, all of which were weighed after 24 and then 48 

 hours on ice (fresh weights were not available), mean 

 percent change in weight during this interval was sig- 

 nificantly different from zero (P=0.02). However this 

 change was very slight (mean=-0.32, SE = 0.11). 



Effect of preservation on oocyte diameter 



Oocyte diameter in the experimental treatment was 

 significantly different from that of the control treat- 

 ment (Fig. 2), in all but three comparisons. Significant 

 differences were not detected for plaice (P=0.22) in the 

 ethanol treatment or for GB cod (P=0.59) or haddock 

 (P=0.37) in the split-formalin treatment. Of the four 

 experimental treatments, the freezing treatment pro- 

 duced the highest mean ^^^^for GB cod (mean=18.81), 

 GOM cod (mean = 26.99), GB and GOM cod combined 

 (mean=23.29), and haddock (mean=28.88). Plaice sam- 

 ples in the freezing treatment also exhibited a sig- 

 nificant positive change (P<0.001, mean=9.90), but a 

 larger change was exhibited in the Gilson's treatment 



