426 SPORANGIOPHORIC PTERIDOPHYTES 



surface of the sporophyll. This is illustrated by the living Psilotaceae, 

 and by some species of Sphenophylhtm, notably S. tnajus, which shows 

 other characters held to be primitive. But it is departed from in 

 S. Dawsotii and S. Romeri, where the number of the sporangiophores 

 is in excess of the sporophylls, while the leaf-whorls are deeply webbed 

 into a cup : Cheirostrobus is also an exception, but there the three sporangio- 

 phores correspond in position and number to the lobes of the tripartite 

 sporophyll : this condition, together with the vascular connections, suggests 

 a parallel amplification of the sporophyll and of the sporangiophore, to 

 which we shall see modern correlatives later among the Ophioglossaceae. 

 Thus, though liable to modifications, the characteristic position of the 

 sporangiophore in the Sphenophyllales is in a median position on the 

 upper surface of the subtending bract. 



Here I must enter my dissent from certain " interpretations " which 

 have been given of the leaf-borne sporangiophore. In cases where it 

 is inserted on the upper surface of the leaf, as in the Sphenophyllales, it 

 has been designated a "ventral lobe." If "ventral lobes" were of 

 common occurrence on the vegetative leaves of these or of other Pterido- 

 phytes, there might be some meaning in the term. It lies with those 

 who use this expression to show that such " ventral lobes " exist normally, 

 other than these spore-producing bodies which they so designate. If they 

 do not normally exist, then calling a leaf-borne sporangiophore a " ventral 

 lobe " merely leads to confusion, and provides no explanation of its real 

 nature. It introduces the idea that the sporangiophore is a result of 

 " metamorphosis " of some pre-existent vegetative structure, of the nature 

 of a "ventral lobe," an opinion untenable in the absence of proof that 

 such bodies existed in the vegetative state. 



But, on the other hand, it has been shown above that in the Equisetales, 

 a series undoubtedly related to the Sphenophyllales, parts similar to the 

 sporangiophores of the Sphenophyllales in structure and in function are 

 borne upon the axis and have no constant relation to the bracts : for 

 reasons assigned above (p. 382, etc.) these are not themselves to be held as 

 foliar. Study of such sporangiophoric types, not separately but collectively, 

 thus leads to a conception of the sporangiophore as a non-foliar structure, 

 which may be inserted either on the axis or on the leaf, though in certain 

 groups it shows a regular relation to the latter. It is, in fact, a part sui 

 generis as much as the sporangium is, and not the result of modification 

 of any other part. 



The history of individual development of the sporangiophore, as traced 

 in Tmesipteris and Psilotiim for leaf-borne types, and in Equisctiun where 

 they arise directly from the axis, gives a clue to their nature. The sporan- 

 giophore first appears as a broad cushion of tissue, in the peripheral parts 

 of which the sporangia are early initiated : these are from the first orientated 

 outwards from the centre of the outgrowth. In the Psilotaceae (as also 

 in S. majus) they maintain this, which may probably have been their 



