Heredity and its Bearings on Atavism. 145 
the descendants seems to increase as we ascend in the 
scale of life, the modes of reproducing the individual 
becoming more and more complicated. Variability is quite 
conceivable in a budding Hydra, but it does not strike us as 
does the disparity in bodily features and mental traits so 
often shown by father and son. 
Above (pp. 186,137) I have shown that in each species the 
number of “ cell organs ” is a fixed one, and that in fecundation 
the nuclear chromatin segments,and probably most of the other 
organs, are always reduced to one-half their normal number. 
One readily sees that a zygote containing the normal 
number of “organs,” which are derived from two different 
sources, will be liable to greater variations than a cell derived 
asexually (a spore), for the following reasons :— 
(1) There will be a reaction upon one another of the 
various “cell organs” derived from the sperm and ovum; and 
(2) A reaction of the environment, which may be more fav- 
ourable either to the set of paternal or to the set of maternal 
organs, throughout ontogeny or during part of the same. 
(3) We have to take into account the relative vigour of 
individual cell organs handed down from the two parents, 
eg., all the organs of one parent may be relatively weaker 
than those of the other parent, or only some organs of one 
parent may be weak, while the others are specially strong. 
This variability must lead of necessity in extreme cases to 
great difficulty in elaborating food-materials to the mutual 
benefit of all the cell organs, and such a difficulty makes 
itself felt in the case of cross-fertilisation, the hybrid being 
' in many cases sterile, probably because the food elaborated is 
not of such a nature as to satisfy the requirements of any 
one cell, which therefore cannot remain non-functional and 
thus become a sex cell, or even if the sex cell be formed, 
the vigour of the individual cell organs will be so much 
weakened or disturbed as to make fertilisation impossible. 
ATAVISM. 
Atavism, or the recurrence of ancestral traits, may be 
explained, if the above conjectures be correct, thus:—We 
know that cell organs are transmitted during fecundation, 
VOL, XII. . K 
