FISHES IN THE SOUTH-WESTERN DISTRICT. 47 



Larv. V. Fisch. d. deustch. Buchf, i., 1897, p. 317). The dimensions are 

 no obstacle to the identity of the two forms, since, if Species F is truly a 

 single species, its diameter ranges from '75 to 'OOOG mm. The extremes 

 are rather far apart, though I am not prepared to say that the variation 

 is too great, since one is constantly encountering fresh evidence of the 

 elasticity of the dimensions of Teleostean ova. In comparing the larval 

 P. tinionaculatus with my own notes and drawings of Species F, I find no 

 discrepancies of pigment and general conformation that are in themselves 

 of specific importance. 



In discussing the affinities of the egg and larva, Mcintosh* and 

 Ehrenbaum have laid stress on the absence of F. unimaculatus from 

 the fauna of their respective districts. With all respect, I would submit 

 that such absence may well be more apparent than real, recalling the 

 use which Cunningham has shown the Topknots to make of their 

 marginal fins. A fish which habitually clings to the vertical faces of 

 rocks is well fitted to elude the ordinary collecting apparatus, and is 

 taken, if at all, largely by chance. 



I do not say that P. unimacidatus is present in the neighbourhood 

 of Heligoland or St. Andrews, but I think that it might be, without 

 attracting human attention. 



The real objection to the identity of Species F with the Topknot now 

 under discussion seems to me to lie in the epidermal structure. My 

 solitary larva certainly did not exhibit the reticulo-papillate arrange- 

 ment of epidermal cells in a degree nearly as well marked as in the few 

 specimens of F which have come under my notice; and formerly 

 I should have considered this difference an absolute bar to the identity 

 of the species. Eecently, however, I have had the opportunity of 

 studying the development of two species of Arnoglossus, in the larval 

 condition of which the reticulo-papillate condition may be even more 

 strongly marked than in Species F, and I find that the condition is vari- 

 able, not only in individuals, but in the same individual. Certain circum- 

 stances, which I cannot clearly associate with the health of the 

 individual, operate in such a way as to render the degree in which 

 the epidermal peculiarities are apparent a matter of no value for 

 specific identification. Thus, while I do not feel in a position to 

 positively affirm the identity of all examples of Species F with P. 

 unimacidahis, I am strongly of opinion that some, at least, of them 

 may have belonged to that species. 



I have remarked that the ova of P. unimaculatus, taken from one 

 female, had yellow oil-globules. Probably all observers of Teleostean 

 embryology will agree that the presence or absence of colour in the 

 globules is of no specific moment. 



* McInto.sh and Masterman, British Marine Food Fishes, p. 348, 



