THE LEPADOGASTER FAMILY. 197 



in August and September, and range from 5 to 5'5 mm. 

 They probably remain, on emergence from the capsule, more or 

 less in the shelter of their original home, until the yolk is 

 absorbed, and sufficient strength is gained for a free existence 

 and the capture of prey. At this stage the larval tail is still 

 present superiorly, with its embryonic rays, while nine true 

 rays occur beneath. Minute black specks appear on the top 

 of the head and under the jaw, with a few on the abdomen, 

 besides the rows along the sides. The prominence of the 

 alimentary canal inferiorly, with the posterior vent, is a 

 feature at this stage. A disc-like thickening occurs on the site 

 of the sucker, but the apparatus is little developed ; yet, even 

 this indication of it is comparatively early, since the pelvic fins 

 are generally late in making their appearance. 



At 60 mm. the larval tail remains as before. The rows of 

 closely arranged black specks go down to the ventral edge, but 

 do not pass to the median ventral region. Two marked pigment- 

 lines occur ventrally on the lower jaw — beneath the eye, and at 

 the opercular region. It is interesting that the young flying- 

 fish (Exoccetus) of the Mediterranean is similarly speckled all 

 over, though with less regularity, at the length of 7 mm. At 

 10 mm., however, the condition of the latter is wholly altered, 

 for it becomes somewhat piebald with rich madder-brown. 



At 10 mm. the lateral rows of chromatophores are still more 

 definitely arranged, forming about four complete longitudinal 

 rows on each side, besides minor specks, and the angle of the 

 mandible is more or less horizontal instead of being upturned. 

 The larval tail has almost disappeared, the only trace being the 

 slender tip of the notochord which slants upward to the dorsal 

 edge of the fin, a true fin-ray occurring immediately beneath ; 

 but the marginal fin with embryonic rays connects it with 

 the dorsal — in which true rays are developing. Rudimentary 

 true rays also mark the first part of the anal fin. The sucker 

 appears to be nearly complete, but is proportionally smaller than 

 in the adult, probably because the habits are less sedentary. 

 These characters remain at 10'5 mm., except that there are 

 now five short true rays on the proximal side of the slender 

 tip of the notochord in the tail, and the latter is evidently 



