6o 
W. C. Worsdell. 
there are numerous instances recorded in flowers, of which I need 
cite but one. Velenovsky describes a flower of ForsytJiia viridissima 
Lindl. in which the four diagonally placed petals of the normal 
flower are represented by 2 large lateral petals which resulted from 
congenital fusion of each pair of normal petals, producing a 
character which, in the case of Fraxinus dipetala, belonging to the 
same order, is the normal one. 
Very interesting are the instances ot “double apples” recorded 
from time to time ; all these cases 1 am rather inclined myself 
to regard as instances of partial reversion to the corymbose 
character of the fructification of the wild crab and other genera of 
Rosaceae, to which in fact a garden apple-tree sometimes actually 
reverts at its second flowering, where, instead of the terminal 
flower only, both this and the other flowers also of the corymb, 
produce fruit. In the cases figured by me there are obviously two 
flowers which have set fruit; in the first case (Fig. 17) the two 
Fig. 17. Longitudinal sections of “ double apples,” shewing three stages in 
the degree of “ coalescence.” A Apples quite separate, and of 
unequal size, borne at the apex of a common pedicel; B Apples of 
equal development and partially united ; C Apples of unequal 
development and much more intimately “ coalescent,” * 
