140 NOTES ON THE IJEI'IIODUCTION OF TELEOSTEAN FISHES 



anil all have dendritic yellow pigment, so that it is probable that the 

 chromatophores always assume this form before hatching takes place. 

 The pigment has much the distribution shown in the figure of the Irish 

 larva, but in the most recently hatched specimens there is none on the 

 dorsal fin, which has no elevation anteriorly. Black chromatophores are 

 present, but are almost entirely masked by the yellow. I suspect that 

 they occurred also in the Irish specimen, but escaped my observation 

 owing to the cabin of the S. s. Fhujal being very badly lighted. In one 

 Plymouth specimen there is no post-anal pigment except a single patch 

 near the middle of the tail. The larvtc were exceedingly delicate, and 

 only one survived even to the early stage of the Irish figure. It had 

 acquired the same elevation of the anterior part of the dorsal marginal 

 fin, accompanied, as in the Irish larva, by pigment. A Plymouth larva, 

 apparently newly hatched, measures 2*02 mm. The Irish specimen, 

 about twelve hours old, measured 2 "6 8 mm. 



I imagine that the normal larva exhibits no epidermal peculiarity. 

 None was present in the Irish specimen, nor in one of the Plymouth 

 examples. In others the skin was tuberculated, while in one I observed 

 a reticulo-papillate condition exactly similar to that met with in Arno- 

 glossus and in Sp. F of M'Intosh and Prince. The absence of sub- 

 marginal pigment patches from the dorsal fin, coupled with the anterior 

 elevation of that fin, sufficiently distinguishes the form before us from 

 Sp. F, but otherwise a papillate specimen in which the connecting 

 ridges are also developed comes very near to that supposed species ; 

 a fact which illustrates the danger of relying on the reticulo-papillate 

 epidermal character for purposes of specific determination. The 

 Laboratory was often very hot during the months in which these 

 larvte were obtained. My specimens, necessarily confined in small 

 vessels for periodic observation, suffered great mortality, and I have no 

 doubt that the tuberculation of the skin was simply pathological. 



The question of the parent species must remain for the present quite 

 uncertain. I am not at all satisfied that we are dealing with a Gadoid, 

 but as the characters appear to connect the larva with that group rather 

 than any other, it may be as well to consider whether any local Gadoid 

 species can be reasonably regarded as the parent. It is unnecessary to 

 recapitulate the forms of which the young stages are known, since their 

 larvic cannot possibly be confused with the one before us. There 

 remain but a few species worthy of serious consideration. These 

 include some of the rarer rocklings, Motella. M. mustela and M. 

 tricirrata need not be considered. Their ova and larva?, however 

 difficult to distinguish from each other, are well known, M. cimlria 

 exists, and may be common in the district, though rarely observed. 

 M. maculata is known to me from a single specimen taken in Start 



