424 SPORANGIOPHORIC PTERIDOPHYTES 
Psilotaceae, and apparently axillary in the Sphenophylleae. But the analogy 
of Eguisefum is interesting in this relation, for there, though the normal 
branching is monopodial, a terminal branching of the strobilus sometimes 
occurs. Thus the points of resemblance appear greatly to outweigh the 
differences, and the Psilotaceae and Sphenophylleae may well be grouped 
together as representing one phylum—the Sphenophyllales. 
Examining the plants thus designated from the point of view of a 
strobiloid theory, the Psilotaceae show in the most pronounced way the 
“ Selago” condition, while about the limits of the fertile zone in Zmesipéeris 
imperfectly formed sporangiophores are often seen, which bear a similar 
interpretation to the abortive sporangia in Lycopodium. The matter is 
complicated here, it is true, by the marked difference between the simple 
sterile leaf and the forked sporophyll: it has been shown, however, that 
in the first stages of the individual development these parts are indis- 
tinguishable. Though in Sphenophyllum the strobilus is definitely marked 
as a tule from the vegetative region, it is important to note that 
Sphenophyllum majus, which most nearly resembles the Psilotaceae in the 
form of its appendages, has also an indefinite strobilus, with continuation 
again upwards into a vegetative state. Thus in both families the shoot 
shows examples of imperfect differentiation. This goes along with a 
development of the sporophyll, both in Zmesipterzs and in Sph. majus, 
as an effective organ of assimilation, which is a further mark of a low 
differentiation. These facts may be held as justifying for the Sphenophyllales 
a line of argument similar to that for the Lycopodiales: that a definite 
strobilus has been the result of differentiation in a shoot in which the 
vegetative and reproductive functions were not originally separate. It is 
true that the case is not so clear for the Sphenophyllales as for the 
Lycopodiales: this is consequent on the number of the known species 
and of individuals avaflable for comparison being much less, and the 
knowledge of the fossils more limited. It will perhaps be objected that 
the earliest known cone of this series, Cherostrobus, was perfectly definite 
and highly specialised, while the earliest indication of a less specialised 
type is in Sph. majus. But the fact that so often the earliest known 
fossils of any phylum are very complex does not prove that such types 
were earliest in evolution. Thus Chetrostrobus among the Sphenophyllales 
and Pseudobornia among the Equisetales, though respectively the oldest 
representatives known, are both extreme forms, as compared with the other 
members of the phyla to which they respectively belong. In estimating such 
facts we should reflect that at the present day primitive and recent forms 
grow side by side, and both or either might be preserved as fossils; also 
that the chances of this happening depend upon many varied factors, of 
opportunity, texture, habitat, etc. The chances of discovery at the present 
day are equally varied. When these points are fully considered it will 
be clear that stratigraphical position of those isolated fossils which happen 
to have been discovered and described should not suffice to prove an 
