98 JOURNAL OF SCIENCE. 
outset to get a clear idea of what selection can do, and what it 
cannot do; for this is a point very commonly misunderstood, 
and incorrect ideas on the subject have made some people reject 
the principle altogether. 
If there are a number of competing individuals, differing 
slightly from each other, and 
No compound of this earthly ball 
Is like another all in all,* 
then the principle of selection will determine which of those 
individuals shall live to propagate its kind, and which shall die 
out. But selection has no power if the individuals are not com- 
peting, and it is in no way concerned with the origin of the 
differences ; the differences must be there before selection can 
act. Those individuals which are best adapted to the circum- 
stances will survive, and, by the law of inheritance, the 
chances are in favour of the offspring inheriting the difference 
that caused their parents to survive. If we take a sufficiently 
large number of cases, the probability of inheriting this differ- 
ence becomes a certainty. If now these offspring again differ 
among themselves in the same way that their parents differed, 
selection will again pick out those in which the difference—or 
variation, as it is usually called—is most favourable ; and these 
selected individuals will, in their turn, hand the variation down 
to their offspring in a better form than they originally received 
it. Selection, therefore, in combination with inheritance and 
variation, is cumulative in its action; but it is important to 
notice that it is cumulative only by selecting the best varieties 
of each generation; it is inno way a cause of the variations 
themselves. 
An illustration will perhaps make my meaning clearer. 
Take the evolution of the eye. Suppose that in a number of 
eyeless animals the nerve of some portion of the skin of one 
individual was slightly sensitive to light, and thus, being able to 
distinguish day from night, it had an advantage over other indi- 
viduals in escaping its enemies. This individual will be one of 
those that survive on the principle of selection, and in the next 
generation there will be, by the law of inheritance, several indi- 
viduals endowed with the same power of distinguishing day frem 
night. These, in their turn, will survive, and at. last all tke 
individuals of the species will have the same faculty. This will 
be the result of selection, but selection did not make the nerve 
of the first individual sensitive to light, and it is powerless to 
improve the nerves of the offspring ; what it can do is to bring 
all the individuals up to the level of the best. If, however, an 
improved variety appears, the improvement is at once made 
permanent, and diffused through the species by the action of 
selection; but selection itself, I repeat, cannot originate nor 
improve. 
How variations arise we are profoundly ignorant. To sup- 
* Tennyson, 
