57 
which is the reason assigned by the theory for its transmission, 
but could not possibly be of any advantage at all. 
If, again, the theory be sound, we have a right to anticipate 
tnat where an experiment has extended over at least 6 000 
years— some would say 60,000— where the struggle for exist- 
ence has been severe, and favourable variations have often 
occurred, some definite advance would have been produced, 
feuch a case is that of man; no one can say he has had no 
^ ru Sgl e f° r existence. Take the case of the labourer, where 
development of muscle is so advantageous, and where use does 
develop certain muscles in a high degree. Now here is a dis- 
tinctly useful modification ; but are his children born with a 
more fully developed muscle than their father? Is the race 
ot such men steadily growing more muscnlar? The reverse 
seems nearer the truth. Once more, therefore, the theory lacks 
the evidence needed for verification. But Tyndall says, and 
rightly, that “ the function of the experimental philosopher is 
to combine the conditions of nature and produce her results '' ; 
but, he adds, “ this was the method of Darwin.” Here I differ 
from him, because I consider Darwin's experiments on pigeons, 
to which Tyndall refers, as being quite distinct from the 
methods of nature. He selected a variety that struck his fancy, 
and with his eye directed to the particular appearance which 
e wished to exaggerate, he selected it as it reappeared in suc- 
cessive broods, and thus added increment to increment, until, as 
he says, an astonishing amount of divergence from the parent 
type was effected. Here, then, we have wish, observation, in- 
telligence, and voluntary selection, every one of which is a 
conscious state, and every one of which is wanting in nature. 
Ain I justified from the evidence, that a conscious intelligence, 
having an end in view, can produce some slight useless varia- 
tions, foi such are those of pigeons, in inferring that nature 
without consciousness, without intelligence, and without a pur- 
pose, can produce endless beneficial variations ? Am I warranted 
in inferring that, because a compositor can, by selecting the 
particular type he requires, arrange them into a connected 
statement ; therefore, if you fling them on the floor, they will 
^^ emse ^ ves a more difficult and longer statement? 
If I be, then I strangely misapprehend the nature of evidence; 
but if I am not, Darwin's experiments are of no evidential 
value whatever as to nature's method ; and his hypothesis is 
not a good one, because in this case at least it is not in agree- 
ment with fact, does not allow of deductive iuference, and 
conflicts with known laws of nature. 
He also instances Darwin's investigations into the cell-making 
