682 CONCLUSION 
in its simplest forms): it is habitual, though with occasional exceptions, 
in Jsoetes and the Ophioglossaceae, and it is seen in many Ferns, and 
especially in such early types as the Marattiaceae, and Osmundaceae. 
The whole plant in these cases consists of a single upright radial shoot, 
and there is reason to believe that this is itself a primitive condition ; 
such a view accords with the generally primitive character of the plants 
in which it is seen. It may, however, result also from reduction, as may 
often be seen in starved seedlings of annual flowering plants. 
But in all the more advanced types, branching of the shoot occurs, 
resulting in multiplication of shoots, and ramification often of a high 
order. It is necessary to put this in relation to the simple unbranched 
state. The terminal dichotomy of the shoot was probably a primitive 
mode of branching. It is characteristic of those species of Lycopodium 
and Selaginella, which are held as primitive; (viz. the Se/ago section of 
Lycopodium, and in Selaginella spinulosa): it is seen occasionally in Ssoetes, 
in the Psilotaceae, and in the Ophioglossaceae as a rare occurrence, also 
in the Osmundaceae, and in some other Ferns. Gradual transition from 
the dichotomous to the monopodial branching may be traced by comparison 
of the more primitive with the more specialised species of Lycopodium 
and Selagine/la, while in some cases the change may be traced through 
unequal development of the branches of the dichotomy in passing from 
the earlier to the later branchings of the individual life.) This makes it 
appear probable that the monopodial is a later and derivative mode of 
branching. 
It is a question what the relation of these terminal branchings of the 
shoot may be to such lateral branchings as are seen in Lguisetum, and 
Sphenophyllum, or in those Ferns where axillary branching occurs. It seems 
not improbable that these are in origin quite distinct modes of amplification 
of the vegetative system ¢rom those brought about by terminal branching, 
and that they are to be regarded rather as regularly recurring and early, but 
nevertheless accessory developments. A reason for this distinction is to be 
found in what is seen in Eguzsetum, for here terminal fissions of the strobilus 
are occasionally to be found, and are quite different in nature and origin from 
the formation of branches normal for the genus. Again, in the Ophioglos- 
saceae, in which family dichotomous branching has been seen as a rare 
occurrence, and in many Ferns such as /P%erzs, which show occasional 
dichotomy, buds arise at points remote from the apex of the shoot, in the 
former case upon the roots, in the latter commonly near to the bases of the 
leaves: these are clearly adventitious. All of these are probably of distinct origin 
and nature from the terminal branching which is fundamentally dichotomous. 
Moreover there is a structural difference between terminal ramifications and 
branchings which are accessory: the former carry on the vegetative con- 
struction with amplified stele, and fully formed leaves arranged as in the 
1This is believed by Bruchmann to be the rule throughout the genus Se/aginedla, 
their first branching being regularly dichotomous. Z.c., p. 18. 
