302 
of the organ was maintained at its highest level ; it was only the 
minus elements which were then eliminated. (2). As soon as selection 
is withdrawn and heredity is simply left to itself, any failures either 
in the force or in the precision of heredity will be allowed to survive 
and perpetuate themselves.”—(Contemporary Review. vol. Ixiv 
pp. 611-2) 
In considering the first proof, we must remember that 
the variations which are necessarily connected with repro 
duction are represented by different writers as taking place 
in three different ways; (1) as diverging in all directions 
from a common centre ; (2) as clustering round a central 
line, with a few variants on either side of a central column; 
(3) as constituting two columns on either side of a central 
line. The first view of the subject is the one generally 
advanced, and it is urged as the strongest of all possible 
arguments for the necessity of selection. The second view 
has been urged theoretically by Mr. Wallace; while the 
same writer has shown that in one instance which he inves 
tigated, the third view was the one which actually cor 
responded to the facts of the case. To make ourselves 
perfectly safe, we will assume the possibility of all these 
variations occurring in connection with a species whose 
stability has been long maintained. What part could 
Natural Selection play in these different circumstances? 
If we suppose that the variations are in all directions, 
then it seems probable that only a few variations will 
represent an ideal perfection ; and if Natural Selection is 
to maintain the highest efficiency of the species, all the 
rest must be destroyed. But Mr. Romanes does not take 
this view. He contends that there will be a large output 
of the unuseful. 
“According to the Darwinian theory, it is only those variations 
which happen to have been useful that have been preserved ; yet, 
even as thus limited, the principle of variability is held to have 
been sufficient to furnish material out of which to construct the