


LOTSY, CURRENT THEORIES OF EVOLUTION. 415 
deal can be said; it assumes that elimination in nature always or 
a least in a majority of cases hits the less fit, while the great 
destruction going on in the egg-stage f.i. speaks very much against 
this view. 
We shall presently hear of Dr. Tower’s experiments on this 
momentous question, here I will only remark, that the great diversity 
arising atter a cross and the diminution of that diversity in suc- 
ceeding generations doubtless shows that elimination takes place, 
on a considerable scale in nature. 
On the principles enunciated, it seems to me, that the peculiar 
distribution of different closely related birds and plants over a 
limited area with many natural barriers, such as the Galapagos- 
islands, on which DARWIN laid so much stress, is much more easily 
explained on the assumption that the segregates of a cross have 
been distributed over these islands, and part of them isolated on 
each of them, than on the basis of anv theory of variability. 
I need not dwell any longer on the fact that the groups which 
taxonomists call species are no units, the experiments of JORDAN 
with Draba verna and, notably, the recent ones of MACALLUM with 
Parthenium argentatum show that these are composed, when there 
are no barriers between their members, of a large number of 
different forms frequently extending over a vast territory. This is 
exactly what we would expect to occur, if the group of forms 
which we are wont to consider as a species, had arisen from a 
cross, because we know, that if no isolation sets in, the diversity 
of the Fo of a cross is apt to be maintained to a considerable 
degree in succeeding generations. 
I do not of course, for a moment, mean to say that every species 
— arisen by hybridization, as all are, if my theory be correct — 
maintains the diversity of the Fa generation of the cross to which 
it owes its birth, as it would do, if the breeding had been promis- 
cuous and all descendants had survived. Every cross in nature has 
undergone the influence of elimination and therefore none has 
escaped the isolation of particular segregates, be it because some 
segregates flowered at different times or because others died, thus 
isolating the remaining ones, for instance because some were killed 
by frost or drought and others were not, but I do maintain that if, 
at the present moment, certain parts of Mexico in which Parthenium 
