FISHERY BULLETIN: VOL. 71. NO. 2 



Table 1. 



-Summary of all morphometric comparisons 

 of Urophycis tenuis and U. chuss. 



V . ifiiuis ' 



N.S. 



U. chuss 



P.E.I 



NS 



* 



NS 

 NS 



NS 

 NS 



* * 



NS 



' NS. = eastern coast of Novo Scotia from vicinity of Halifax 



to north and east. 

 P.E.I. = Northumberland Straits in Gulf of St. Lawrence 

 adjacent to Prince Edward Islond. 

 NS = not significant. 



* = significant at 5°o level. 

 ** = significant at I°o level. 

 - N.E. = Gulf of Maine and southern New England waters. 

 Regression of: 



a. distance from snout to posterior margin of orbit on upper 

 jaw length. 



b. length of pelvic fin on distance from origin of pelvic 

 fin to anus. 



c. head length on standard length. 



d. preanal length on standard length. 



was not significantly different from U. teinds 

 N.E. or P.E.I. Regression coefficients for the 

 latter two samples were significantly different 

 at the 5% level. The adjusted mean for U. chuss 

 was significantly different from that of U. tenuis 

 N.E. at the 1% level and from that of U. tiniuis 

 N.S. at the 5% level. No significant differences 

 were found between U. cJiuss and U. toiuis 

 P.E.I. 



The contention of Bigelow and Schroeder 

 (1953:223), that this character "can be relied 

 on, even for very small fish," is not borne out. 

 If true, all values of U. tenuis would be on or 

 to the right of the line y = x, and those for 

 U. chnss would be to the left of the line 

 (Figure 9). 



Regression of Length of Pelvic Fin on Distance 

 from Origin of the Pelvic Fin to Anus (Figure 

 10, Table 1) 



This relationship was linear for U. chuss, 

 but the values for U. tenuis seemed to be 

 curvilinear. The slope of the regression changed 

 when the independent variable reached 150 to 

 160 mm. Most values for U. tenuis P.E.I, were 



! U tenuis NE O O 



2 u itnttis H% •--• 



3 U tenuis PE I O- -O 



4 U chuss A A 



DISTANCE FROM THE ORIGIN OF THE PELVIC FIN TO THE ANUS 



Figure 10. — Regression of pelvic fin length on the 

 distance from the origin of the pelvic fin to the anus 

 from three samples of Urophycis tenuis and one of 

 (7. chuss. 



distributed above the change in slope and could 

 not be compared to the other samples. An 

 analysis of covariance was performed among 

 the linear portions of the three remaining 

 sample regressions. U. tenuis N.E. and U. 

 toiuis N.S. were significantly different from 

 U. chuss at the 1% level and from one another 

 at the 5% level. The doubts of earlier workers 

 concerning the validity of this character are 

 borne out. If valid, all values for U. chuss 

 would be on or to the left of the line ij = .r, and 

 those for U. tenuis would be to the right of the 

 line (Figure 10). Although most values for 

 U. chuss do show this relationship, a few are 

 to the right of the line, and many values for 

 smaller specimens of U. tenuis are to the left. 



Regression of Head Length on Standard Length 

 (Figure 11, Table 1) 



Regressions for all samples were linear. 

 U. tenuis N.S. was not significantly different 

 from U. tenuis N.E. or U. tenuis P.E.I. The 

 adjusted means of the latter two were different 

 at the 1% level. V. chuss was significantly 

 different at the 1% level from all samples of 

 U. tenuis. This character may be useful in 

 distinguishing between U. chuss and U. toiuis 

 longer than about 150 mm, particularly when 

 used in addition to the gill raker and scale 

 characters discussed above. 



1 



484 



