74 The Botanical Gazette. [March, 
separated, almost all entomophilous ones have stamens and 
pistils in the same flowers. Miiller therefore considers di- 
clinism to have been the original condition of phanerogams, 
from which hermaphroditism has been developed by natural 
selection. Darwin takes exactly the opposite view. Butit 
hardly seems necessary to adopt either exclusively, for 
both hermaphrodite and unisexual forms are common among 
the lower plants. Why cannot both forms have been tra 
mitted from the pteridophyte-like ancestors of phanerogams: 
Diclinism and bisexualism may be collateral branches, one 
not necessarily older than the other, though in many individual 
cases it is evident that one has been, or is being derived from 
the other. 
_wings of butterflies, the wings, tails and beaks of birds, etc- 
f low organization, seems to hol 
good here. Contrast the 
and“the variable number of petals or sepals of water 
Violacez, Composite, Labiate and Scrophulariacee. ss 
decrease in the number of the floral organs is often ac 
panied by further specialization by the union of their individu 
members to secure still better protection of the pollen, honey — 
and ovules and better adaptation to the agents of pollinatio™ 
senieesl 
