268 Rhodora [NOVEMBER 
appearance, while the earlier phases vanish wholly. Such a qualita- 
tive development of the individual has been shown to be in many 
cases essentially a recapitulation of the historical evolution of the 
species or family or larger group to which the individual belongs. 
And in a general way it may be said that the generations of animals 
and plants perpetually repeat the stories of their several races. 
Yet it is not safe to judge that whenever the infantile condition of 
a plant differs from the adult state, the former is due to reversion. 
In each instance regard must be had for inherent or adducible prob- 
ability. To illustrate, and at the same time to come directly to the 
case in hand: when we find that seedlings of Drosera intermedia 
begin with rotund leaves and bear only round-bladed leaves until 
they are considerably advanced in age, we may suspect that the 
youthful leaf reproduces an ancestral type. For evidence which 
may throw more light upon the problem we should, however, study 
ontogenesis in other species. As a matter of fact we do find that 
several species which in the adult state differ widely in the leaf agree 
at an earlier period and bear rotund leaves like the infantile leaves 
of D. intermedia. From all the cases of ontogenetic progress which 
I have been able to observe, with one merely negative exception, the 
indications are the same, and point to the existence of a fundamental 
type such as that which Drosera intermedia realizes in its earliest 
phases, from which the several species considered have probably 
arisen. 
Concordant ontogenetic evidence certainly has great value. Upon 
this principle and facts which cannot here be presented, the two- 
forked type embodied in Drosera binata and flagellifera, and the 
peltate type seen in Drosera lunata and fourteen related species, 
become derivatives from a rotund original. The African Drosera 
cistiflora, in its highest state characterized by long narrow lanceolate 
or linear-lanceolate leaves, and several closely allied species, may be 
traced back to a spatulate source; as may also D. filiformis. And 
the spatulate form in turn reverts to a rotund original. 
I have now outlined the reasoning by which I am persuaded that 
the small leaves with orbicular blades bearing marginal tentacles of 
a curious structure to be described below, which seedling plants and 
small adventitious plants of Drosera intermedia put forth, are rever- 
sionary. Being satisfied as to their nature, I have made a number 
of experiments with a view to determine some of the conditions of 
