134 NOTES ON THE REPRODUCTION OF TELEOSTEAN FISHES 



probable enough that some of the latter may belong to Rh. norvegicus, but 

 I do not think that there is any satisfactory evidence of the nature of 

 the metamorphosing larva of that species. It may, as M'lutosh and 

 Masterman suppose, be characterised by the possession of periotic 

 spines, but I should say that the conformation of the St. Andrews 

 specimen of 11 mm. (if correctly represented by M'Intosh and Master- 

 man), referred to Eh. norvegicus, bears certainly a greater resemblance to 

 that of Bh. piinctati'.s. 



Apart from the spinigerous forms there is another larva or group of 

 larvre which can be definitely associated with the Topknots. 



Under Sp. xiii. a typical series of this form, from the symmetrical 

 condition to an advanced stage of metamorphosis, has been figured and 

 described by myself. It has no spines at all, and at parallel stages of 

 the metamorphosis is very much smaller than the spined form, and shows 

 moreover no trace of the bold pigmentation of the latter. From the 

 conformation at the most advanced stages, and from the fin-ray formula, 

 I considered that this form belonged to P. unimaculatus, and so far as is 

 possible, the appearance of the early vitelligerous larva of that species 

 confirms my opinion. 



Metamorphosing larvre, either identical with or at least very similar 

 to the Irish specimens, have been met witli at St. Andrews, and are 

 referred, as we have seen, by M'Intosh and Masterman to Rh. pundaius. 

 As that species has certainly a spined larva, it appears to me that the 

 spineless forms from St. Andrews must belong either to P. unimaciUatus 

 or to Rh. norvegicus. It is simply a question of whether the larva of the 

 last-named has periotic spines or has none. 



The few spineless sinistral larvto which have been taken at Plymouth 

 in 1897 leave the matter in doubt. One, 8 mm. in length, presents the 

 stage of metamorphosis of an Irish specimen of 8'87 mm. (op. cif., 

 PI. XL, Fig. 92). The two are very much alike, but the Plymouth 

 example is somewhat more profusely and generally pigmented. Does 

 the difference in size justify us in supposing that the Plymouth larva 

 belongs to a smaller species than P. unimaculahis ? I should say that 

 it is possible, but not certain, since individual larvae vary in the size at 

 which they assume the different phases of metamorphosis. jVnother 

 larva, about 3 mm. long after preservation, connects itself more readily 

 with the younger stages of the St. Andrews spineless larva3 than with 

 any of the Irish series. The head is large, but the trunk is narrow and 

 elongate, without any trace of Pleuronectid metamorphosis, but the 

 abdomen is relatively enormous, a condition apparently due to the 

 viscera being distended with food. Whether naturally or by accident, 

 the abdomen is laterallly compressed. Pigment is present in the form 

 of minute black chromatophores scattered over the general surface, but 



