IN THE SOUTII-WESTEKN DISTRICT. ~ 133 



is assigned witli some reserve to Bh. norvcgicus. No St. Andrews larva 

 is assigned to P. unimaculatus, because that species has never been 

 recorded in the district, but the capture of a single specimen of Bh. 

 norvcgicus seems to be considered to have atlbrded sufficient warrant for 

 changing the determination of the spined larva. 



The Irish larva without periotic spines, Sp. xiii., is, if one takes into 

 account the stages of the metamorphosis which the two forms exhibit, 

 much smaller than the spined Sp. xiv. Thus at a length of 1062 mm. 

 the latter is still nearly symmetrical, with a heterocercal tail, while at 

 9'37 mm. the former has the eye at the ridge, and the tail quite 

 homocercal. I should imagine that the larva which, at any given size, 

 has the metamorphosis most advanced, would be universally held to 

 belong to the smaller species. Yet IM'Intosh and jMasterman put 

 forward the same comparison as an argument in favour of an exactly 

 converse conclusion. 



Cunningham's observation of the later stages of the spined larva 

 seems to me too positive to permit of any doubt as to the spinigerous 

 nature of the larva of Bh. punctatus, unless, as is most unlikely, he was 

 dealing with a specimen with an exceptionally large number of fin-rays. 

 The St. Andrews authors, however, refer the larva without spines to that 

 species, with the simple remark that they are uuable to concur with 

 Cunningham's opinion. 



During the present season we have twice taken a larva with 

 periotic spines, corresponding exactly in this respect, and in the 

 disposition of the pigment, with my Irish series, but less advanced 

 in metamorphosis. They measure respectively 5'il and 4'5 ca. mm., 

 the latter specimen being bent and difficult to measure with accuracy. 

 The body is still elongated, and shows no signs of elevation. The 

 contour of the head is still rounded, although the jaws protrude 

 somewhat, and the general appearance is that of a larva not long after 

 the final absorption of the yolk. Yet the periotic spines are conspicu- 

 ously developed, the upper one being somewhat the larger, and rather 

 backwardly deflected. Tiie larvie, which will be figured in the Annales cle 

 Mnsde de Marseille, are certainly identical with the Irish forms, and I 

 refer them without hesitation to BJi. punctatus. 



They occurred on the 8th and 24th of April {vide record), that is to 

 say at the end of the period of occurrence of the Topknot ova, 

 while Cunningham's advanced larva3 were taken on the 4th May. 

 Taking into consideration dates and localities, one is led to suppose that 

 these larvie must be derived from ova similar to those which have been 

 referred to the Topknot generally. In other words the said eggs are in 

 part, at least, those of Bh. 2'>unctatus. 



From the date, locality, and dimensions uf the ova, it is obviously 



