440 THE POPULAR SCIENCE MONTHLY, 



under the front edge of the dorsal fin, or even in front of it ; while in 

 the herring, though the position of the ventral fin varies a little, it lies 

 more or less behind the front margin of the dorsal fin. The anal fin 

 is of the same length as the dorsal, in the herring, longer than the 

 dorsal in the sprat. But the best marks of distinction are the absence 

 of vomerine teeth in the sprat, and the smaller number of pyloric 

 caeca, which do not exceed nine, their openings being disposed in a 

 single lonijitudinal series. 



Shads and pilchards have a common character by which they are 

 very easily distinguished from both sprat and herring. There is an 

 horizontal fold of scaly skin on each side of the tail above and below 

 the middle line. Moreover, they have no teeth in the inside of the 

 mouth, and their pyloric caeca are very numerous a hundred or more 

 their openings being disposed five or six in a row. 



The shads have a deep narrow notch in the middle line of the up- 

 per jaw, w^hich is absent in the pilchard. The intestine of the shad is 

 short and straight, like that of the herring ; while that of the pilchard 

 is long, and folded several times upon itself. 



Both of these fishes, again, possess a very curious structure, termed 

 an accessory branchial organ, which is found more highly developed 

 in other fishes of the herring family, and attains its greatest develop- 

 ment in a fresh-water fish, the Heterotis, which inhabits the Nile. This 

 organ is very rudimentary in the shad (in which it w^as discovered by 

 Gegenbaur *), but it is much larger in the pilchard, in which, so far as 

 I know, it has not heretofore been noticed. In Chanos and several 

 other Clupeoid fishes it becomes coiled upon itself, and in Heterotis the 

 coiled organ makes many turns. The organ is commonly supposed to 

 be respiratory in function ; but this is very doubtful. 



Herrings which have attained maturity, and are distended by the 

 greatly enlarged milt or roe, are ready to shed the contents of these 

 organs, or, as it is said, to spawn. In 1862 we found a great diversity 

 of opinion prevailed as to the time at which this operation takes place, 

 and we took a great deal of trouble to settle the question, with the 

 result w^hich is thus stated in our report : 



'^ We have obtained a very large body of valuable evidence on this 

 subject, derived partly from the examination of fishermen'and of others 

 conversant with the herring-fishery ; partly from the inspection of the 

 accurate records kept by the fishery officers at different stations, and 

 partly from other sources ; and our clear conclusion from all this evi- 

 dence is, that the herring spawns at two seasons of the year, in the 

 spring and in the autumn. We have hitherto met with no case of full 

 or spawning herrings being found, in any locality, during what may 

 be termed the solstitial months, namely, June and December ; and it 

 would appear that such herrings are never (or very rarely) taken in 



* " Ueber das Kopf skclet von Alcpocephalus restrains " (" Morphologisches Jahrbuch," 

 Bd. iv., Suppl., 1878). 



