ORIGIN OF THE FREE-LIVING STATE 253 
Selaginella, parts which are neither true roots nor axes, but serve as bases 
of attachment for roots. Though these bodies do not as yet greatly help 
to solve the question of origin of the root, they draw attention to the 
Lycopodiales in connection with any discussion how a_ subterranean 
absorptive system originated. In the vast majority of Vascular Plants the 
root is formed early, and is ready as soon as the embryo projects from 
the prothallus, to take up its physiological duty. There seems in them 
no need to assume that they achieved their independence through an 
intermediate “ protocorm” stage. It is quite as probable that the transition 
was in descent, as it is to-day, directly to the rooted state. 
The general conception of the rise of the sporophyte embodied in the 
preceding chapters differs in its tone and tendency from some of the writings 
which have preceded it. The attempt has here been made to treat the 
sporophyte consistently throughout: to apply the same point of view to 
the Vascular Plants as to the Bryophytes. One important difference 
between the morphological method adopted here and that of some other 
writers is that it gives a less prominent place to hypotheses involving 
reduction from a more complex ancestry. The general principle here 
has been to. assume that morphological characters are in the up-grade of 
development, unless there is good reason for holding ‘a contrary opinion ; 
and only to admit that an organ has been reduced from some more 
elaborate body when there is some assignable reason (comparative or 
physiological, but preferably both) for that conclusion (Chapter XIX.). 
A theory of reduction has never been resorted to merely as a means 
of resolving a difficulty of comparison. The position adopted has accord- 
ingly been to regard it as probable that the smaller-leaved types were 
themselves primitive as a rule, though in some there is evidence of 
probable reduction; and to contemplate it as probable that microphyllous 
have given rise to megaphyllous types. It is highly probable, on the 
other hand, that reduction of size and complexity has been highly effective 
in certain phyletic lines: for instance, thé recognised evolutionary story 
of the Cycads involves extensive reduction of the sporophylls. But for 
the primitive strobiloid forms, it would seem to harmonise better with 
their early occurrence, and their morphological characters, to hold that 
they represent a relatively primitive condition, rather than a down-grade 
of morphological complexity. 
The recent changes of view have been in great measure due to the 
extension of the knowledge of the fossils, both stratigraphically and 
morphologically. It is true that such data are seldom decisive on ques- 
tions of comparison (Chapter XVIII.), but the case here is a strong one. 
It is found that strobiloid forms are at least as early in occurrence as 
large-leaved Fern-like types. Further, they appear not only to have been 
present in the earliest fossil-bearing strata, but also well represented in 
numbers and in variety of type. This has brought with it the conviction 
that the strobiloid type has quite an equal right with any other to be 
