Fishes of the JVestern North Atlantic 581 



The larvae at hatching are 6— 7 mm long (Fig. 136 e),^" very slender, and wholly 

 transparent except for a double ventral row of black pigment spots between the heart 

 and the yolk sac, thence rearward a single row to the base of the future caudal fin, and 

 a larger spot above the anus; and soon an additional row develops a little farther up 

 along each side. The pectoral fins are already formed at hatching; the yolk sac has been 

 absorbed at a length of about 7 mm; at 9 mm (Fig. 136 f) the first indications of the 

 dorsal and adipose fins are apparent; at 20—30 mm all of the fins are clearly outlined; 

 and at 30—40 mm the young try begin to assume the outlines of the adult. It seems 

 that most of the larvae are 20—40 mm long by their first winter, 40—65 mm early in 

 the following spring, and 60— 80 mm 



bv midsummer, when they are about ^^'^ble V. Total Lengths (mm) of Newfoundland Capelin at 

 ^ r T^ 1 1 • 1 Two through Five Years of Age, Taken Inshore in Late 



one year of age. Table v gives the j^^^ ^^^ f^I^ . 



subsequent relation between age and ^ j^^j^^ ^ ^ Females ^ 



total length for Newfoundland fish 

 taken inshore in late June and July, 

 according to Templeman's detailed 

 studies of the scales and otoliths of 

 extensive samples (no). These values 

 correspond closely to lengths of 60— 

 80 mm at one year, 135— 150 mm for males and 105— 125 mm for females at two 

 years, and 155—205 mm for males and 130—195 mm for females at three years, as re- 

 ported in 1933 for Newfoundland Capelin by Sleggs from length-frequencies combined 

 with scale studies (lOO: 37). Pitt found much the same rate of growth for Grand 

 Bank Capelin {yg: 295-311). 



Comparison of Capelin Larvae With Other Larvae. Larvae of Clupea harengus (At- 

 lantic herring) might be mistaken for those of v. villosus if taken together. But the 

 position of the yolk sac, with its anterior end at least as far forward as the origin of 

 the pectorals in C. harengus and definitely posterior thereto in v. villosus, will serve to 

 distinguish them from each other until the yolk has been absorbed. During the earlier 

 stages also, the anus is about 83 "/o of the way back along the body in C. harengus but 

 only about 75 "/o of the way back in v. villosus. At lengths greater than 15-20 mm 

 the presence of the adipose fin marks v. villosus as such. 



Should larvae of Osmerus eperlanus (American smelt) be taken with those of v. vil- 

 losus, an event not reported as yet, the two can be separated up to a length of about 

 I 5 mm by the presence in O. eperlanus of a ventral pair of black spots anterior to the 

 pectoral fins. At larger sizes the very evident air bladder of O. eperlanus sets it apart 

 trom V. villosus. 



Other northern fish larvae that parallel the larvae of v. villosus in their slenderness 

 are those of Pholis gunnellus (rock gunnel) and Ammodytes (launce). But there should 

 be no danger of confusion here, for the anus in these is near the middle of the body, 

 as compared with about 75 "/o of the way back in v. villosus {no: 68, 69). 



30. For excellent illustrations of Newfoundland larvae, see Dannevig {20 : pi. 3, figs. 26, 27); for descriptions and illustra- 

 tions, see Templeman {110: 70-75, figs. 18-20); for Icelandic larvae, see Schmidt {()6: 16-18, pi. i, figs. 24-28). 



