SUMMARY 425 



evolutionary priority in face of strong comparative evidence to the contrary. 

 Accordingly the existence of the complex and definite cone of Cheirostrohus 

 as the earliest of the Sphenophyllales cannot be held as invalidating the 

 conclusion above stated. 



In all these sporangiophoric Pteridophytes the axis is the dominant 

 part of the shoot, and takes the lead in its development. This is shown 

 anatomically by t the typically protostelic structure of Sphenophyllum, while 

 the foliar strands insert themselves marginally upon it. In the Psilotaceae 

 the type of stem-structure is the same, but the xylem becomes hollow 

 and medullated in the upper region, and in Tmesipteris it may break 

 up into separate strands. Even in the apparently divergent case of the 

 Equisetales it has been shown that the vascular system is referable to 

 a primitive type of monostele, though greatly reduced in accordance with 

 a semi-aquatic habit. Notwithstanding such changes the structure of the 

 axis in all these forms indicates origin from a type in which the axis is 

 predominant over the appendages, the vascular supply of these being 

 inserted with the minimum of disturbance upon the cauline stele (Clado- 

 siphonic type of Jeffrey). This confirms the theory of a strobiloid origin, 

 with pre-existent axis and subsidiary appendages. 



The differences in number of the sporangia on the individual 

 sporangiophore call for remark. In the Sphenophylleae they have been 

 observed to vary from six in Sphenophyllum majus to one only in 

 S. Dawsoni, while Psiiotum and Tmesipteris take a middle position with 

 three and two respectively. In the Equisetales the number in the 

 fossil form appears to be commonly four, but in recent species of Equisetum 

 the number may be much larger and variable. 1 In the Sphenophyllales 

 the variations present some points of interest : the number six occurs 

 only occasionally in the one species named, while four is the usual 

 number in that species, which it shares with Cheirostrobus. But in the 

 species of Sphenophyllum with compact strobili the number may be two, 

 or only one. These low numbers go along with a larger number of the 

 sporangiophores, which may be twice (S. Dawsoni}, or three times 

 (S. Romeri] the number of the subtending bracts. In the former species 

 the frequent juxtaposition of the stalks, and the insertion of the vascular 

 supply of the stalks upon ' the strands supplying the bracts, suggests that 

 fission has been operative, as in the chorisis of stamens : and it seems 

 probable from the facts that with an increase of number of the sporangio- 

 phores, however brought about, there has gone a decrease in the number 

 of the sporangia which each bears. Accordingly S. Dawsoni and S. Romeri 

 may be held to bear sporangiophores of a type reduced from the original : 

 and a central type of sporangiophore would appear to be one with about 

 four sporangia. 



The typical position which such a sporangiophore holds in the 

 Sphenophyllales is one of attachment in a median position to the upper 



] In Calamites paleaceus the pendulous sporangiophore bears a solitary sporangium. 



