FISHERY BULLETIN: VOL. 85, NO. 4 



formis. Correlation values for these two curves 

 were r = -0.79 and -0.80, respectively (Table 2). 

 Pseudocyst counts of white K . thyrsitis were <10 

 and did not appear to adversely affect the texture 

 of the Pacific whiting (Table 3A). The low count is 

 thought by Kabata and Whitaker (1981) to be the 

 result of an evolutionary development of defense 

 mechanisms in the host fish. The correlation coef- 

 ficient representing sensory textures and white 

 K. thyrsitis infection was 0.12. This suggests that 

 K. thyrsitis plays an insignificant role in the tex- 

 ture quality of Pacific whiting, a view also held by 

 Kabata and Whitaker (1981). 



Fitted regression curves between sensory tex- 

 tures and black pseudocyst counts are shown in 

 Figure 5B. Mixed black pseudocyst counts did not 



Table 2. — Spearman correlation coefficient between sensory tex- 

 ture rating and Kudoa pseudocysts in Pacific wtiiting caughit during 

 the summer of 1983. 



Table 3. — Percentage of sensory texture ratings for (A) each 

 Kudoa species, (B) fork length, and (C) sex composition for Pacific 

 whiting caught dunng the summer of 1983. 



follow the pattern for black K . paniformis counts, 

 as observed with the white mixed and K . pani- 

 formis infections. This suggests that the black 

 pseudocyst counts cannot be used as a reliable 

 predictor of sensory texture. Magnitudes of black 

 K. paniformis counts and black mixed counts 

 were <320, while counts were <16 for black K. 

 thyrsitis. Like those infected with white K. thyr- 

 sitis, fish infected with black K. thyrsitis para- 

 sites had normal, firm cooked textures (Table 

 3A). The absolute value of the Spearman correla- 

 tion coefficient (Table 2) between texture quality 

 and black pseudocyst intensity was lower (-0.38) 

 than that of the coefficient for texture quality and 

 white pseudocyst intensity ( -0.74), both at signif- 

 icance level of 0.001. 



From the correlation coefficients for the white 

 and black pseudocyst counts of both species, the 

 square of the coefficients (coefficient of determi- 

 nation) was calculated to compare their relative 

 importance in terms of differences in their magni- 

 tude with respect to one another. Fifty-five per- 

 cent of the observed variability in all sensory tex- 

 ture ratings can be accounted for (predicted) by 

 the observed variability in the white pseudocyst 

 counts, while only 14% can be accounted for by 

 the observed variability in the black pseudocysts 

 counts. However, infections do not occur in Pacific 

 whiting as only white or black pseudocysts; they 

 also occur as a mixture of the two. Thus, when the 

 quantitative effects of the white or black pseudo- 

 cyst counts on sensory texture were evaluated by 

 multiple regression analysis, we found that only 

 1.5% of the variability in sensory texture rating 

 was accounted for by black pseudocyst counts, 

 and 45% accounted for by white pseudocyst 

 counts. These figures represent partial coeffi- 

 cients of determination which indicate the rela- 

 tionship between two variables while controlling 

 the effects of one or more other variables and are 

 consistent with the findings of Patashnik et al. 

 (1982) and Tsuyuki et al. (1982). 



Intensity of White and Black 



Pseudocyst Infection in Relation 



to Sensory Texture Ratings 



The magnitude of white and black pseudocyst 

 counts in relation to their corresponding sensory 

 texture scores are shown in Tables 4 and 5. 



A total of 214 fish, 38% of the fish examined, 

 did not have white pseudocysts. Table 4 shows 

 that texture scores for this group of fish range 

 from 4 to 9 with 97% of the scores in the range of 



750 



