

174 Prof. F. 0. Bower. Studies in the [June 15, 



head ; (ii) by formation of entirely new archesporia, having no direct 

 connexion by descent from pre-existent ones ; (iii) by partitioning- 

 of a continuous archesporium ; this might readily result from partial 

 sterilisation and formation of septa. It has been one chief object of 

 this investigation to see what evidence may be gathered from Vascular 

 Cryptogams of one or of all of these modes of origin. The question 

 has been first approached by examination of the " strobiloid " forms. 



The frequent presence of synangia in eusporangiate Vascular 

 Cryptogams suggests either coalescence accompanying reduction in a 

 descending series, or partitioning by means of septa in an ascending 

 series ; the first question in connexion with such synangia will be 

 whether in any natural sequence of Vascular Cryptogams the pro- 

 gression from a non- septate to a septate condition can be traced ; or 

 the converse. Though the facts at hand do not amount to an actual 

 demonstration, the Lycopodinese and their allies are believed to be 

 an ascending series, and they are seen to supply important evidence. 

 The series PJiylloglossum , Jjycopodium, and Selaginella, Lepidodendron^ 

 and the Psilotaceee show natural affinities. In a paper shortly to 

 appear in the ' A_nnals of Botany ' I have drawn attention to the 

 remarkable anatomical similarity which links the Psilotaceae to 

 Lepidodendron, while no one would doubt the kinship of the latter to 

 other Lycopods. To this series Tsoetes may be added, for, though the 

 anatomical correspondence is not so close, I think its affinities with 

 the Lycopods are nearer than with any other family. 



As regards the sporangia, there can be no doubt of the homology 

 of the sporangium of Phylloglossum, Lycopodium, Selaginella, and 

 Lepidodendron ; similarity of position, structure, development (not 

 traced in Lepidodendron), and function all show this. Within the 

 genus Lycopodium differences of detail have been observed analogous 

 to such differences as would result in the production of more bulky 

 sporangia, such as those of Lepidodendron and Isoetes, though it is 

 true these differences are not so extensive. In these very large 

 sporangia trabeculee are found, as rods or plates of sterile tissue, 

 which may project far upwards into the sporangial cavity (Lepido- 

 dendrori), or may extend the whole way through it to the upper wall 

 (Isoetes). In the latter case it has been shown by Goebel that the 

 trabeculae are the result of differentiation of a potential arche- 

 sporium, part of which is sterilised and forms the trabeculse. But 

 these are at most only partial septa. 



The next step is to the Psilotaceae ; and the first question is that of 

 the real nature of the synangium in these plants. While Goebel and 

 Juranyi look upon the sporangiophore as an abbreviated axis bearing 

 two leaves, the synangium occupying its apex, Graf Solms, from 

 external observation of Psilotum, maintains the older view, that the 

 whole sporangiophore is of foliar nature with two lobes, while the 



