114 ME. F. DAT ON SOME TKISH GASTEKOSTEI. 



were deficient of any portion of the pubic arch ; but the tri- 

 angular pubic plate differed both in its form and size. The 

 height of the dorsal spines was subject to great variation, and 

 the comparative length of the ventral spine to the pubic plate 

 was inconstant, these variations not being due to the size of 

 the specimens (see figs. B and C). 



The lateral scale-like plates were either three long ones articu- 

 lating with the ventral plate, or else a fourth and even a fifth 

 were also present — ^this being an advance towards the variety 

 traclmrus, wherein these plates are found along the whole length 

 of the body of the fish. 



Those with three dorsal spines had the first two from 1| to 2, 

 25, and 3 times in the height of the body ; those with four spines 

 had the second 2 to 2| and 2^ in the height of the body. This 

 species appears to be subject to various deviations from the type, 

 seen in the greater or less amount of scaly plates along the sides, 

 and in the difference observed as to the comparative length of the 

 venti-al spine to that of the pubic plate. Desirous, however, of 

 testing these questions, I examined numerous specimens of this 

 fish captured at the same place ; and they give the following 

 results : — 



As an average, specimens 



2 "4 inches long, ventral spine from | to f length of pubic plate, 

 9.9 5 



^ ^ J) >) J) J) 6 " V 



9-1 S +n -1 



•^ -L )» » jj 3J 3 ''" 5 » " 



2-n 5 



1-5 inch „ „ „ I „ „ 



If we turn to the definition of this species, we find the ven- 

 tral spine given as nearly or quite as long as the pubic plate. 



There were other variations likewise perceptible in these Irish 

 specimens : in some the ventral spine was seri'ated on both sides, in 

 others only externally ; the dorsal spines were likewise much less 

 serrated in some than in others. But to me all appear to be va- 

 rieties of one species ; and if such can be discovered in so limited 

 an extent of country, one cannot but surmise that a search through 

 an extended area might be productive of greater results, and occa- 

 sion the suppression of several, at present, recognized species. 



