56 Lady Isabel Brownd. 
Lepidostrobus Mazocarpon and Mazocarpon might equally well, in 
accordance with his view, be regarded as stages in a process of the 
septation of a unilocular sporangium. As evidence in favour of his 
view Professor Bower claims that in the development of the 
Psilotaceous synangium, tissue that normally develops as tapetum 
may become sporogenous and vice versa. He also appeals to the 
incomplete septa found in Lepidostrobus Brownii and to the 
trabeculae of Isoetes as examples of sterilization leading to partial 
septation; on Miss Benson’s view these structures might be 
regarded as the reduced remains of former septa. Some support 
is afforded to the views of both these botanists by the recent 
description, by Nathorst, of a Rhsetic fossil, Lycostrobus Scotti, 
thought to be rather closely related to Isoetes, in which the micro¬ 
sporangium was apparently divided by narrow septa (compared by 
Nathorst to trabeculae) into roundish loculi (30). The attribution 
of this fossil, known in a single impression only, to the Isoetaceous 
cycle of affinity remains doubtful as no ligule or vegetative organs 
are known. The weakest part of the theory that the Lycopod 
sporangium is the result of coalescence and fusion of free sporangia 
lies in the fact that it is among the heterosporous forms ( Lepido¬ 
strobus Mazocarpon, Mazocarpon, Isoetes ), presumably less primitive 
than the homosporous types, that what are regarded as the remains 
of a septum are most strongly developed. On the whole the sterile 
tissue present in the above mentioned forms is much in excess of 
that found in Spencerites or in most species of Lycopodium. Similarly 
on Professor Bower’s hypothesis that the sporangia of the synangium 
of the homosporous Psilotaceae represent the loculi of a septate 
sporangium, it is curious that indications of intermediate stages in 
the process of septation should be more marked in several hetero¬ 
sporous, than in any homosporous members of the Lycopodiales. 
The analogy with the Sphenophyllales and Equisetales, in 
which the ventral lobes of the sporophyll seem (as explained in the 
first and second articles of this series) to have become free from 
the dorsal lobes and to have been displaced on to the internode 
above, makes it possible to regard the sporophylls of Lycopods as 
lobes of a dorsiventrally lobed sporophyll in which, as in Spheno- 
phyllum fertile, both dorsal and ventral lobes are fertile. But as 
the oldest known Lycopod fructifications show the same typical 
simplicity of the sporophyll as the recent cones, and as this Lycopod 
type is as old as, if not older, than that of any Sphenophyllaceous 
fructification known to us we should not be justified, in the present 
