GENERAL RESULTS— SPHENOPSIDA 631 



From the great constancy of this character throughout 

 the Articulatae I am inclined to attach considerable 

 importance to it in this case, though in other groups 

 it may be variable. The mode of branching is also 

 very different, dichotomy occurring in the Psilotaceae, 

 while monopodial branching prevails in the Spheno- 

 phyllales. For these reasons, among others, it seems 

 best to regard the Psilotaceae as forming a class of 

 their own, the Psilotales, while including them under 

 the main division Sphenopsida. 



Miss Sykes, in an interesting memoir on the anatomy 

 and morphology of Tmesipteris} adopts the axial theory 

 of the fructification of Psilotales, and extends it to the 

 Sphenophyllales and the Equisetales, regarding the 

 sporangiophore as representing a branch throughout 

 all these groups. She thus accepts the general view 

 of the relationships maintained by Thomas, Bower, 

 and myself, and endeavours to reconcile it with the 

 axial theory of Bertrand, Goebel, and others. In the 

 special case of Tmesipteris, the latter theory is sup- 

 ported, among other arguments, by the observation that 

 the vascular strand entering the synangium -pedicel 

 branches into three, the central strand soon terminating, 

 while the two lateral branches run round the periphery 

 of the septum. The axial interpretation might be a 

 tenable one if the Psilotales stood alone, but through- 

 out the Articulatae there appears to be no satisfactory 

 evidence that the sporangiophore is a branch, while 

 there is much in favour of its foliar nature, as has been 

 shown above. The axial theory seems, on present 

 evidence, to rest on too slender a basis. 



1 Annals of Botany, vol. xxii. 1908, p. 63. 



