a theory. Dr. Weismann asserts that we cannot imagine 
the process of Natural Selection :— 
“ I must say that, in respect of warrant to assume the process of 
Natural Selection, it does not seem to matter much whether we can 
easily, or with difficulty, or only with great difficulty, imagine it ; and 
for this reason, that I do not believe that we are in any case able to 
conceive in detail the actual morphological metamorphoses con 
cerned.”—(The Contemporary Review, vol. Ixiv., p. 320.) 
He asserts that we cannot demonstrate the truth of 
Natural Selection :— 
“Just as in this instance, so is it in every individual case of Natural 
Selection. We cannot demonstrate any of them, and there is no 
use attempting to make them seem unanswerable.”—(p. 323.) 
“It is true the results of artificial selection are in favour of the 
occurrence of Natural Selection, but, as Herbert Spencer justly 
observes, the two processes, though they may be analogous, are 
certainly not identical. The struggle for existence plays the part of 
breeder in the case of Natural Selection ; and how this factor works 
we are unable to determine in any single case. Who would say of 
any little variation in the form of any existing species that it is 
sufficient to give its possessor the victory in the struggle for existence, 
and so may become the starting-point of an advantageous metamor 
phosis of the part? Even in the simplest of cases that is impossible ; 
no one, for instance, could decide how much the colour of a green 
insect must vary so as to originate a process of selection dependent 
perhaps on adaptation to a new and somewhat differently coloured 
fodder-plant. We cannot estimate what Romanes has recently very 
well called the ‘selection value’ of variations, which Lloyd Morgan 
had previously spoken of as the ‘elimination value’; we can only 
say generally with Darwin that selection works by the accumulation 
of very slight variations, and conch/de from this that these ‘ slight 
variations' must possess selectio?i value. To determine accurately 
the degree of this selection value in individual cases is, however, as 
yet, impossible.”—(p. 324.) 
“As soon as an attempt is made to think out in detail the process 
of selection by which perhaps the little bristles or the small baskets 
of the worker-bees have arisen, it is seen that all and every one of 
the data are wanting. Moreover, in my opinion, we cannot hope 
that we shall ever possess them, either in these cases or in any 
yet simpler process of Natural Selection. Not only would it be 
necessary to form an estimate of the smallest variations so as to know