D.—ZOOLOGY. 79 
In every river and lake that it enters the trout forms freshwater 
colonies, and on the other side of the Atlantic the salmon does so fairly 
readily, although not nearly so generally as the trout does on this side. 
In Europe, trout being present, the salmon forms freshwater colonies 
only in exceptional circumstances. Thus Lake Wenern in Sweden, now 
cut off from the sea by inaccessible falls, has a stock of salmon ; there can 
be no doubt that in former times salmon entered the lake and bred in its 
tributaries, and that some of the smolts, when they reached the lake on 
their seaward migration, considered this very large lake a sufficiently good 
substitute for the sea to stay there, and so founded a lacustrine race. 
Freshwater colonies of trout are found in the Atlas Mountains and in 
the countries north of the Mediterranean eastwards to the Adriatic, 
proving that in glacial times the range of sea-trout extended southward 
to the Mediterranean. The rivers of Dalmatia and Albania are inhabited 
not only by trout but by fish of another species, known as Salmo obtusi- 
rostris. This little fish, which never grows larger than fifteen inches long, 
has all the structural characters that distinguish salmon from trout, and, 
_ indeed, looks very like an overgrown salmon parr; but when compared with 
salmon of the same size it is seen to differ in having a considerably smaller 
mouth, weaker teeth, and more numerous gill-rakers (15 to 18 instead of 
11 to 14 on the lower part of the first arch). In fishes generally the 
number and length of the gill-rakers—projections from the gill-arches 
that prevent food from entering the gill-chamber with the respiratory 
current—are related to the nature of the food ; thus, in exclusively pis- 
civorous fishes, such as the pike, they are represented by a few short 
knobs, and in feeders on minute plankton organisms they are very numer- 
ous, long, slender, and close-set. It has been recorded that Salmo obtusi- 
_rostris subsists mainly on the larvee of Ephemeride, which are very abun- 
dant in the rivers it inhabits, and there can be no doubt that the small 
‘size of the mouth, the feeble dentition, and the increased number of gill- 
takers are related to this diet. 
The presence of this fish in the rivers of the east side of the Adriatic 
seems to me to point to the probability that in glacial times salmon, as 
well as trout, occurred in the Mediterranean, and that in these rivers 
some of the salmon parr, tempted by the abundance of parr food, pre- 
ferred to continue the parr life instead of going to the sea as smolts, thus 
orming a freshwater colony in quite a different way from the salmon 
of Lake Wenern. The question may be asked—if these fishes are derived 
from salmon and live in the same way as salmon parr, how can their 
‘differences from salmon be adaptive ? The reply to this is that the size of 
the mouth in the salmon parr must have some relation to the fact that 
it is going to become a salmon, feeding on fishes in the sea, and that, as 
S. obtusirostris grows to twice the length of a salmon parr, we should expect 
‘the number of gill-rakers to be increased, for it is not number but the size 
of the interspaces that is important in relation to food. 
__ The work of Dr. Johannes Schmidt on the Viviparous Blenny (Zoarces 
wiparus) is of great interest. He had found that in the European eel 
he average number of vertebre was 115, and that from whatever part of 
‘its area samples were taken, whether from Iceland, Denmark, the Azores, 
or the Adriatic, the range of variation and the mean were exactly the same. 
_ This he considered as a confirmation of his view that all the eels from these 
