FISHERY BULLETIN: VOL. 71, NO. 3 



as evidenced externally by the change in snout 

 shape, pectoral fin rays form, the gas bladder 

 disappears, pigment patterns change becoming 

 concentrated on the eyed side, and scales begin 

 to form. 



Fish of both species become sexually mature 

 by about 50 mm. 



Ratios of body measurements related to 

 standard length, frequently used as taxonomic 

 aids, were determined for larvae of both species. 

 Data for these ratios, which change with growth, 

 are listed in Appendix Tables 1 and 2. Consid- 

 erable overlap exists between the two species 

 although depth at anus/SL and depth behind 

 anus/SL are usually greater for E. microstomus 

 larvae than C. arctifrons larvae of equal size. 

 Depth behind anus/depth at anus related to 

 standard length (Leonard, 1971a, Figure 9) 

 provides a character useful for separation of 

 larvae over 8 mm of the two species. This ratio 

 approaches 1 in E. microstomus larvae of a 

 smaller size than C. arctifroiis corresponding 

 to the earlier metamorphosis of E. microstomus. 

 The ratio changes from < 1 to > 1. the condition 

 in the adults, during metamorphosis. Data for 

 the adults (Appendix Tables 3, 4) agree with 

 those given by Parr (1931). E. microstomus is 

 proportionately deeper bodied than C. arctifroiis 



but the latter has a proportionately longer 

 upper jaw. 



Changes in body form were further analyzed 

 for growth patterns. Body measurements (Ap- 

 pendix Tables 1-4) were used for this purpose 

 because methods using ratios "are inefficient 

 and may often lead to erroneous interpreta- 

 tion" as explained by Marr (1955) and illus- 

 trated by Ahlstrom and Counts (1955) and 

 Mansueti (1958). Statistics for the regressions 

 of head length, snout length, snout to anus 

 length, eye diameter, upper jaw length, depth 

 at anus, depth behind anus, and peduncle depth 

 on standard length are presented in Tables 4 

 and 5. The columns headed b list the rate of 

 increase in size of body part per 1 mm increase 

 in SL. Data on larvae and adults could not be 

 combined because of a gap in specimen size 

 between larvae undergoing metamorphosis and 

 juveniles of 30-40 mm in E. microstomus and 

 40-50 mm in C. arctifrons. The regressions all 

 appear to be linear (Leonard, 1971a, Figures 

 10-17). Correlation coefficients were >0.93 

 for all regressions except those of snout length, 

 eye diameter, and upper jaw length which had 

 correlation coefficients >0.87 in larvae and 

 >0.80 in adults. 



All body parts measured increase more in 



Table 4. — Statistics describing the regressions of body measurements on standard length for larval stages of Eiropiis 

 microstomus and Citharichthys arctifrons. The independent variable X is standard length in all cases. 



X. - Mean value of X 



Y - Mean value of Y 



N - Number of specimens examined 



b - Rate of increase of Y with respect to X 



a - K^intercept of regression line 



5,, - Standarcj deviation of the regression 



r- Correlation coefficient 



750 



