186 NATURAL SCIENCE [September 
We cannot find that the rough tubercles of the flounder are useful 
to it because it lives in rivers or for any other reason, we know of 
no advantage which the plaice derives from its red spots, its smooth 
scales, or the bony tubercles on its head. Nor can we find any 
indication that these peculiarities are correlated with adaptive differ- 
ences. The only adaptive difference at present clear is that the 
plaice, has blunter teeth in its throat than the other species, and that 
these are suitable for crushing the shells of the bivalves on which 
the plaice feeds. But we know of no connection between this and 
the other characters. The theory then that these specific peculi- 
arities are due to the natural. selection of indefinite variations is 
unsupported by any evidence. 
How then are we to explain such specific characters? It seems 
to me we are forced to regard them as the necessary consequences of 
growth and of the conditions of life. It is evident enough that differ- 
ences of habit and the extension of a species into different regions 
will necessarily lead to its subdivision into groups between which 
little or no interbreeding takes place. The individuals which 
lived in estuaries would breed together and not with those that 
lived always in the sea. Thus any modification produced in the one 
group would remain confined to that group, and by interbreeding 
within that group would be kept approximately uniform. It is 
difficult in the present state of our knowledge to say how modifica- 
tions of the kinds here in question are determined. There are 
indications that continuous modification takes place in successive 
generations without any direct stimulus that can be detected. We 
must remember that the development of the individual is the growth 
and multiplication of groups of differentiated cells. This develop- 
ment is controlled partly by heredity, partly by surrounding condi- 
tions, but the development of every part and every organ is to some 
extent independent. The facts indicate that a particular part or 
organ may for some unknown reason obtain increased nourishment 
and develop with increased vigour, or on the other hand may show 
diminished vigour, and that the change may be progressive in suc- 
cessive generations. It would seem that ultimately the effect must 
be due to external conditions acting upon the properties of living 
matter. But the action is evidently very indirect, and the processes 
involved are so complicated and recondite that at present we know 
nothing about them. We have indications of the influence of ex- 
ternal conditions in the differences in the same species in different 
geographical areas. Thus the flounder in the Mediterranean has 
scarcely any tubercles, while in the Baltic and Arctic regions those 
structures are excessively developed, so that nearly the whole skin is 
covered with them. Whether this is due to the cold or not we do not 
know, but it is a fact that the plaice also shows greater roughness of 
