104 AGANTHOPTERYGII. 



half the depth, and in one of 10 inches is as 1 to 2|. Owing to 'the species 

 varying very considerably in depth, as elsewhere shown, this must necessarily 

 be a very uncertain character." 



Although the period stated (from the middle of March to the begin- 

 ning of October) is the usual time of capture in Belfast Bay, I have occa- 

 sionally seen this fish in the market here in every month of the year. 

 During the Avinter season, they have been brought in fine condition from 

 Cushendall and Glenarm with salmon, of which a very few are there taken 

 in the sea. 



Food of the Mullet. — The stomachs of a few mullets from Portaferry, 

 opened by Dr. Drummond, 4th August, 1838, Avere filled with minute 

 larvae of which he informs me some were alive. 



23rd October, 1838. — The stomachs and intestines of six mullets from 

 Donaghadee, were filled Avith a minute Asperococctis, like A.jmsillis. Some 

 small pieces of other Algae occurred, as one specimen of Enteromorpha 

 compressa Avill shoAv. That Avith the Asperococcits I liaA'e preserved on glass. 



Larvae from stomach of mullet taken at Portaferry, Avere 2 and 2\ lines 

 long; 12 joints in body; head broAvn ; body colourless, except centre, 

 Avhich is dark. 



Of four specimens obtained at Donaghadee on 11th Oct., 1838, one 

 contained ova, Avhich, though minute, Avere apparent to the naked eye ; 

 another showed them hardly developed, and milt apparent to the unas- 

 sisted eye Avas in one of the males. The stomachs of tAVO Avere empty ; 

 those of the other tAvo filled entirely Avith a minute Asperococcus apparent 

 to the naked eye. I got Dr. Drummond to put some of it tinder his micro- 

 scope, Avhich proved the plant to be of this genus ; the size of specimens 

 ■was generally about half an inch in length. The lineated appearance ex- 

 tended in these fish to the ventral ])rofile, as Donovan represents. 



A. H. Haliday, Esq., after examining, at my request, a number of larvaB 

 taken from the stomach of a mullet in the month of November, favoiu'ed 

 me with the folloAving remarks : — 



" I have examined the larvae found in the mullet's stomach. They 

 seem to be all of one sort, btit from the diff"erence of size are evidently in 

 variotxs stages of groAvth, and perhaps none of them full groAvn. The 

 multitudes of them found favour the conclusion, of Avhich I have scarcely 

 a doubt from their form, that they belong to some species of the genus 

 Chironomns, several of Avhich occur in the greatest profusion on our sea- 

 coasts. On comparison Avith the larvae of Ch. jjhimosus, L. (the common 

 red worm of ditches) the chief difi'erence is in the form of the posterior 

 extremity as folloAvs : in place of tAvo long and divaricated branches of the 

 last segment Avith four shorter-pointed processes betAveen them in pairs 

 and the spiracles above their origin, prolonged into tAVo slender tapering 

 tubes, croAA'ned Avith a AA'horl of fine hairs (in Ch. jjlumosvs), I observe 

 in these only a single cylindric false leg inclined doAvnAvards Avith a Avart 

 at its base on each side, including the spiracle : but this difference is no 

 more than Ave may admit as specific, since Ave knoAV the terrestrial larvae 

 of other species of this genus to be totally deprived of these appendages 

 of the posterior extremity. The larvae of the other genera in this family 

 {Cliironomidce), even those of Ceratopogon, Avhich are least remote, are 

 more complex in their external structure : and among those of the re- 

 maining Nemocera, destitute of lateral spiracles, I knoAV none w'hich have 

 much resemblance to these. 



" The larvae in the families Bihionidce, Scatopsidcc, ^Ii/retophilida;, and 

 Cecidomidce having the lateral spiracles, are out of the question." 



