712 CONCLUSION 
found at different heights on the same plant: or the plane of the whorls 
may be set obliquely to the axis. It would appear probable from such 
facts that the original type had whorled leaves, and that the spiral 
arrangement was acquired by secondary disturbance of it, a point of some 
considerable interest for comparison with the sporangiophoric Pteridophytes. 
The other character is seen in Jsoezes (p. 318), and in Lepidostrobus Browntt, 
both of which had very large sporangia (p. 322). In these a partial 
sterilisation of sporogenous tissue producing trabeculae meets a mechanical 
and nutritive requirement following on their large size, and the structure 
thus approaches a state of septation: such septation is indeed technically 
completed in the megasporangia of /sve¢es, but no Lycopod shows a septate 
state of the sporangium as a permanent character. The interest in this is 
in comparison of these sporangia with the similarly placed synangia of the 
Psilotaceae and Sphenophyllaceae. 
These two series, together with the Equisetales, have been included 
under the general designation of the Sporangiophoric Pteridophytes (p. 423). 
Though differing in detail, the main plan of their sporophyte is similar to 
that in the Lycopodiales, as regards axis and leaves, branching, and 
anatomical structure ; but the sporangia of the latter are replaced by 
sporangiophores, while the relations of these to the bracts is not 
uniformly so regular as that of the sporangia in the Lycopodiales. More- 
over, both bracts and sporangiophores show evidences of fission, sometimes 
independently, sometimes together. These relations have been considered 
above (p. 694-5), together with the similar variations of exact position of 
the sporangia and sporangiophores relatively to the axis: such facts, com- 
bined with the arguments already advanced in Chapter XXVIII., lead to 
the conclusion that che functionally identical parts designated sporangiophores 
and sporangia are cognate parts ; it appears probable that the sporangtophore 
zs ttself a consequence of elaboration of a simpler type of spore-producng 
member, of which the sporangium of Lycopodium is an example, while the 
trabeculae in Isoetes and Lepidostrobus Brownii suggest a mode of origin of 
the septate state. Lf this were so, then the sporangiophore would have been 
distinct in its phyletic origin from the bract-leaves, whith habitually subtend 
the spore-producing members, whether they be sporangia or sporangiophores. 
The Sporangiophoric Pteridophytes (which include the “ Articulatae” of 
Lignier together with the Psilotaceae) are primarily characterised by the 
presence of the sporangiophore. The fact that the leaf-arrangement is often 
whorled, which is a leading feature of the Articulatae, while that in the 
Psilotaceae is alternate, is here regarded as a point of secondary moment. © 
The reasons for this are, first, that the leaf-arrangement varies from the 
whorled to the alternate in the very natural phylum of the Lycopods, and 
secondly, that a similar change appears from the ancient Sphenophylleae 
to the modern Psilotaceae—groups clearly related to one another. /¢ 
seems probable that the whorled arrangement was initially general for the 
strobiloid types, but that the regularity kas been secondarily abandoned. The 
