YOUNG JACK CREVALLES 



509 



dorsal and anal soft-ray relationships given by 

 Nichols and Roemhild sustains the same modal 

 value that my specimens exhibited (table 22). 

 Branching and lobation occur as in the second 

 dorsal fin. The first and second rays are longest 

 and nearly equal to about 32 mm. standard 

 length; above 32 mm., the first ray is longest. 

 The longest dorsal ray averages longer than the 

 first anal ray. The second anal spine is two-thirds 

 as long as the longest anal soft-ray at 15.3 mm., 

 one-half as long at 60 mm., about one-fourth as 

 long at 164 mm., less than one-fifth as long at 

 432 mm., and about one-ninth as long at 830 mm. 



Interneural and interhemal spines. — The poste- 

 rior lateral projections of these spines are well 

 developed and extend above the body surface 

 along the bases of the dorsal and anal soft-rays 

 at 15.3 mm. (fig. 81). They are also developed 

 on the damaged 12.7-mm. specimen. At 164 mm. 

 they are still externally visible, principally along 

 the posterior bases of the fins, but are not project- 

 ing in a 248-mm. specimen. 



Caudal. — 9 + 8 principal rays; about 9 or 8+8 

 secondary rays. Branching of the principal rays 

 has occurred by 15.3 mm. (fig. 81). 



Pectoral. — 1-19 or 20. The full complement of 

 rays has formed by 20.4 mm. standard length. 

 The distal end of the fin is rounded from 15.3 mm. 

 to about 27 mm., after which it becomes pointed 

 and falcation begins (figs. 81 to 83). Falcation is 

 pronounced by 80.5 mm. (fig. 84). 



The regression of pectoral length on standard 

 length is shown in figure 87 and table 20. Two 

 lines were fitted to this regression: for specimens 

 from 15.3 to 43.7 mm. and from 43.7 to 283 mm. 

 standard length. An extension of the lower line 

 intersects the upper line at approximately 54 mm. 

 indicating an inflection occurs at about 54 mm. 

 and that a faster pectoral growth rate prevails 

 above that size. The proportional rates of increase 

 for the two variates are 0.27-mm. (below 54 mm.) 

 and 0.39-mm. (above 54 mm.) increase in pectoral 

 length per 1.0-mm. increase in standard length. 

 Comparison of the upper regression line with 

 specimens of hippos up to 830 mm. (fig. 91) 

 suggests that a decrease in the pectoral growth 

 rate occurs at some size around 300 mm. 



Pelvic. — 1-5. The soft-rays are branched and 

 segmented at 15.3 mm. standard length. 



Body depth. — Depth at first anal spine averages 

 less than depth at pelvic from 15.3 mm. to about 



489286 O— 59 7 



30 mm. standard length, is about equal from 30 

 to 90 mm., and is greater above 90 mm. The 

 coordinates for body depth at pelvic on standard 

 length for specimens smaller than 30 mm. are 

 plotted in figure 8. 



The regression of body depth at pelvic on stand- 

 ard length is shown in figure 88 and table 20. 

 A line fitted to this regression for specimens from 

 15.3 to 164 mm. standard length shows a pro- 

 portional rate of increase for the two variates 

 within this size range (0.37-mm. increase in body 

 depth per 1.0-mm. increase in standard length). 

 The alinement of the coordinates of the specimens 

 in the "latus and/or hippos" series suggests 

 that a faster body-depth growth rate occurs 

 below about 10 or 12 mm. The position of the 

 coordinates of the six largest specimens in figure 

 88 below the extension of the calculated regression 

 line and of the coordinates of specimens up to 830 

 mm. (fig. 90) suggests that a decrease in body- 

 depth growth rate occurs at some size around 164 

 mm. 



Head. — The regression of head length on stand- 

 ard length is shown in figure 88 and table 20. 

 A line fitted to this regression for specimens from 

 15.3 to 164 mm. standard length shows a pro- 

 portional rate of increase for the two variates 

 within this size range (0.30-mm. increase in head 

 length per 1.0-mm. increase in standard length). 

 Comparison of the extension of the calculated 

 regression line with coordinates of larger speci- 

 mens (fig. 91) suggests that head growth may 

 continue at the same proportional rate up to 830 

 mm. 



Eye. — The regression of eye diameter on stand- 

 ard length is shown in figure 89 and table 20. 

 Two lines were fitted to this regression: for speci- 

 mens from 15.3 to 51.0 mm. standard length 

 and from 51.0 to 162 mm. The lines intersect 

 at approximately 55 mm., indicating that an 

 inflection occurs at about that size and that a 

 slower eye growth rate prevails above that size. 

 The proportional rates of increase for the two 

 variates are 0.09-mm. (below 55 mm.) and 0.06- 

 mm. (above 55 mm.) increase in eye diameter 

 per 1.0-mm. increase in standard length. The 

 position of the coordinates of the "latus and/or 

 hippos" specimens suggests that a faster eye 

 growth rate may prevail below approximately 15 

 mm. The position of the coordinates of the four 

 largest specimens in figure 89 below the extension 



