96 BOTANICAL GAZETTE [august 



are related among themselves more remotely even than was long 

 supposed, for the evidence now in hand appears to indicate that 

 upon the whole the first of these is primitive, the second reduced 

 from possible pro-Calamariah^ ancestry, and the third, in the 

 descent of which both amplification and reduction probably played 

 minor successivie roles, specialized from pro-Sphenophyllaceous 

 stock. The belief once prevalent that the Lycopods are related 

 through the Psilotales to the Sphenophyllales and only remotelv 

 through the latter to the Calamariales will accordingly have to be 

 abandoned. 



Anatomical considerations 



With the discovery of structural material of Cantheliophorus we 

 may expect facts of great interest. If our interpretation of this 

 form is correct, we may expect to find within the brace the fibro- 

 vascular bundle which supplies the sporangia. In the large sporan- 

 giophores of C. linearifolius, at the insertion of the brace upon the 

 toe, there are minute striae which in passing from the proximal 

 extremity of the brace into the toe, instead of bending downward 

 as one might expect, toward the insertion of the pedicel, bend 

 upward, and this may perhaps be interpreted to signify that here 

 the brace at its insertion may already have been dragged slightly 

 downward during the evolution of the race from Calamarian or 

 pro-Calamarian stock, somewhat after the manner of the sporan- 

 giophore of Paleostachya vera Seward (12), and if such is the case 

 then we should not be surprised to find that the vascular strand 

 makes a slight loop forward in the toe above the insertion of the 

 brace, comparable to that in the species of Paleostachya just cited. 

 It would seem further to be quite in keeping with our theory of 

 reduction for Cantheliophorus to find 2 parallel vascular strands 

 in the midrib corresponding in a way with the 2 close parallel lines 

 sometimes seen over the median nerve upon the adaxial surface 

 of the blade, but such suggestions in the present state of our knowl- 

 edge are perhaps unduly speculative. 



3 This is a hj'pothetical Calamarian ancestor, not necessarily proto-Calamarian. 



