relation of the characters in the particular (poikilodyna- 
mous) pairs varies greatly, whilst the number of interme¬ 
diate forms becomes reduced.* 
The formation of the gametes is described by Comma as 
schizogonous when it takes place according to Mendel’s 
Law. In the homoogonous method of development there is 
no separation of allelomorphs in this process. Unlike the 
different modes of vegetative development, the schizogon¬ 
ous and homoogonous methods of formation of the germ 
cells are regarded by Correns as being sharply distinct. 
Good examples of the “ mosaic ” development of the hete- 
rozygote are to be found in maize-crosses, especially in 
crosses between forms with a blue and with a white aleurone 
layer respectively. Webber (73) has suggested that such 
cases might be accounted for by parthenogenesis of both the 
nuclei which usually coalesce to give rise to the endosperm, 
the patches of colour being derived from the two. Correns 
pointed out that the appearance of a pale blue aleurone 
layer with darker patches as the result of the cross blue by 
white is inexplicable on this hypothesis. 
Bateson and Saunders (6) suggested, on the other hand, 
that the pied condition might be due, not to dominance 
failing sometimes and succeeding sometimes, but to the 
existence in the mosaic of islands of the recessive character 
U 6 or unresolved state (the phenomenon then 
not being described in terms of dominance at all). But 
this suggestion also fails to explain the same observation. 
he fact itself I have been able to confirm in the case of a 
cross between “Black Mexican” sugar corn and “Boone 
county White’’dent corn. 
“ ^ 0rrens Buds the power of development of a 11 
mage to vary greatly, depending upon the individu¬ 
ality of particular germ cells, plants or races, as well as in 
JZ£Si!**.*LW** ^ view is quite opposed to that which 
upon the sobiect. 7 
