1878.] 
On Residual Phenomena. 
33 
instance of residual phenomena. The laws of “ natural 
selection ” and “ sexual selection,” collectively known as 
Darwinism, do, as far as we can judge, account satisfactorily 
for a large portion of the phenomena concerned ; but as, 
undoubtedly, there is another portion which they cannot by 
any amount of ingenuity be made to explain. For instance, 
it has been well said that before natural selection — or indeed 
selection of any kind — can be brought into play, variation 
must have already set in. This will be at once apparent on 
the following consideration : — Suppose a pair of animals, 
or, still further to simplify the matter, a single hermaphro- 
dite, being of low type, existing in the primaeval world, had 
produced a hundred fertile ova. Two cases then are only 
possible : the young animals springing from these ova must 
be either all absolutely alike, or they must exhibit certain 
variations, however slight. In the former alternative there 
is no basis for natural selection to work upon, the very idea 
of selection implying differences among the objects among 
which a choice is to be made. In the latter case, the 
varieties, being ex hypothesi antecedent to the aCtion of 
natural selection, cannot be its effects. Here, then, we have 
a residual phenomenon, a marginal faCt for which the 
Darwinian hypothesis is unable to account, and which will 
yet have to be explained before we can understand the 
origin of species. Nor in the instance we have supposed 
can the explanation be furnished by Lamarck’s hypothesis. 
The influence of all external circumstances aCting upon the 
parent animal must affeCt the ova, if at all, equally ; but, as 
regards the first beginning of variation among a number of 
creatures absolutely alike at birth, the Lamarckian view 
has the advantage over the doCtrine of natural selection. 
If the young animals wander away from the place of their 
birth they may become exposed to different influences, from 
climate or from the quantity and quality of their food, and 
among their progeny slight variations may thus possibly 
appear, and thus enable the principle of natural selection to 
come into play. Successive broods produced by one and 
the same parent may also be supposed to differ from each 
other to a small extent if any change has occurred in the 
circumstances to which the parent is exposed. But it is 
difficult to avoid the conclusion that the “ residual ” here is 
an internal tendency to vary, faint instances of whose aCtion 
we see in the faCts that no two children of the same parent, 
no two symmetrical organs of the same animal, no two 
leaves even of the same tree, are absolutely alike. 
VCI* VIII. (N.s.) 
D 
