114. TRANSACTIONS OF THE ROYAL CANADIAN INSTITUTE. [VOL. XI 
they were plentiful around the station wharf and at other points in the 
neighborhood, like other young fish, with apparently ravenous appe- 
tites, feeding day and night on anything that could be swallowed. On 
the night of June 26th they were so numerous and were making so 
much disturbance in the water that I went out about 10 p.m. with a 
plankton net to find out what formed the food supply. There was 
scarcely anything but rotifers in the catch, but these were there in 
plenty. The stomachs of the fish were filled up with these. (Earlier 
in the evening the dog salmon fry were feeding almost entirely on 
Diptera.) 
These were undoubtedly Clupeids, but were so unlike the herring 
of 6 or 7 cm. in some respects that I could not feel sure as to their iden- 
tity. Anyone who has tried to diagnose very young fish from the descrip- 
tions given in the manuals, no matter how satisfactory they may be for 
mature fish, will know what a difficult job it is. The fry is often so 
unlike the adult in many particulars. One of the common differences 
occurs in a characteristic on which much stress is laid as a means of 
identification, 2.e., in the number of fin rays in the various fins. In 
comparing the fry of a number of species with the adults in this respect 
the fin rays have been found to be more numerous in the fry, the reason 
apparently being that from 2 to 4 of the anterior rays coalesce to form 
a stiffer support for the forward portion of the fin. The number of 
fin-rays, which one might suppose would provide one of the easiest means 
of identification in the fry, can in no sense be depended upon. This 
was one of the difficulties here. Other diagnostic features, such as the 
absence or presence of teeth on the vomer, can scarcely be made out 
with certainty, as such accessories, although present in the adult, may 
not yet be developed in the fry. The color cannot be depended on, for 
the greatest variation occurs. In this case the fish in question were 
much lighter in color than the older herring, and had not the metallic 
sheen. There was nothing in all this to prove conclusively that these 
fish were herring and not some other species of the Clupeid family, 
Clupanodon ceruleus, for instance, as that is found in the vicinity also. 
Young as the fry were, the scales were already formed and developed 
sufficiently to show something of the permanent structure. The scales 
of the adult herring and pilchard bear considerable resemblance, although 
they are markedly different. The pilchard scale is much larger and has 
characteristic scars passing in obliquely from the margin at the sides. 
Since the scales of the fry were very small and corresponded to only a 
small portion of the adult scale, the difference in size of the adult scale 
did not help in the diagnosis, and as the characteristic scars of the pilchard 
do not pass all the way to the nucleus, it was possible that although 
