196 MESSES. E. W. L. HOLT AND L. W. BYENE— BISCATAN PLAXKTON : 



The most advanced specimens measure 15 mm. and have the form shown in fig. 2. 

 The head is a little longer than the abdomen, and head and abdomen together com- 

 prise a little less than half of the total length without the caudal fin. The eye is 

 relatively large, but not remarkably so for a larva. It is somewhat forvvardly directed, 

 as in many deej)-sea larvae and, to a less extent, in the adults of deep-sea fishes. The 

 snout is long and sharply pointed in lateral view. Its sides are subvertical and sharply 

 defined from the flattened upper surface. The angle of the jaAV readies or passes the 

 vertical from the hind edge of the eye. The teeth appear to be rather small and, as 

 usual in larva?, set at considerable intervals. The pectoral fins are large and set low- 

 down on the body; they would extend if perfect at least to the anus. The ventrals are 

 small and situated a little way in front of the anus. The dorsal shows from 10 to 12, 

 and the anal 13 to 17 rays, but when the larvse are examined from the ventral aspect it 

 appears probable that in the adult condition the anal is continued further back than 

 appears in the figure. The dorsal commences a little behind the origin of the anal. 

 There is a low fold of embryonic marginal tin in front of the dorsal, commencing at 

 some distance behind the head, and a similar ventral fold between the bases of the 

 pectorals and ventrals. The caudal is of moderate length and rather deeply forked. 

 An adipose fin, if present in the adult, would probably be conspicuously indicated by 

 embryonic rays at this stage of development, but cannot be detected in any of the 

 specimens. There are about 15 abdominal and about 21 caudal myomeres, but the exact 

 count is uncertain. 



On the lower edge of each gill-cover, or on the isthmus internal to it, is a large photo- 

 phore, consisting of a base of black matter and a rounded opaque white ajoex. ' ~Fonv 

 similar bodies, set in two closely apposed pairs, occur on the abdomen immediately in 

 front of the ventral fins. The larvse are devoid of dark pigment, other than that of the 

 eyes, except for a black spot in front of the caudal base. 



A specimen of 13 mm. shows 12 dorsal and 15 anal rays, and has the snout somewhat 

 arcuate in dorsal profile. This last feature appears to connect the stage of fig. 2 with 

 the larva depicted in fig. 3, which measures 9 mm. So far as we can count, the 

 myomeres are of the same number as in the larger specimens. The position of the 

 alimentary canal, which is slung obliquely from the trunk in the ventral marginal fin- 

 fold, may be due in some measure to post-mortem distortion. About midway in the 

 caudal region the trunk shows a ventral opaque excrescence, of which the nature is not 

 apparent. It does not look like any phase in the development of an anal fin and may 

 probably be due to injury. 



Though the collection comprises a number of small elongate larvcc of which some may 

 probably be specifically identical with " Prsescope," none have any characters of a nature 

 sufficiently definite to permit of their reasonably certain association with this form. 



A process of elimination causes us to regard this larva as that of a member of the 

 GonostomatincB. The suggestion of a probable parent is not easy : a comparison with the 

 adults of known genera and species points to Gonostotna itself ; and, although it would 

 prima facie appear improbable that a Gonostoma would attain its adult characters 

 through an early stage resembling the "Prsescope " larva, we cannot premise that fishes 



