390 
PKOFESSOR W. C. WILLIAMSON ON THE OKGANIZATION 
panying the above structure in its outward course in tile shape of an investing 
cylinder. 
It thus becomes clear, not only from the specimens described, but from other similar 
ones in my cabinet, that though we discover no true dicliotomization or externally con- 
spicuous branching of the Dictyoxylon- stems, they gave off distinct lateral appendages 
almost at right angles to the axis of the parent stem. What were these diverticula'? 
whither did they go ? and why do they terminate so invariably at the circumference 
of the parent bark ? 
I think we must separate Plate XXIII. fig. 7 , n, from the other examples ; it differs 
alike in its structure and origin. My impression is that it may have supplied an adven- 
titious or aerial root, which impression is confirmed by the history of a similar appendage 
which I find in Dictyoxylon Grievii. The other examples in which we have a distinct 
central axis surrounded by a regularly constructed exogenous zone are manifestly young 
branches ; but whether they were prolonged into ordinary leaf-bearing branches, or 
whether they were half-abortive fruit-bearing ones, I am unable to determine. The 
invariable absence of any outward prolongation of them beyond the periphery of the 
parent stem, certainly seems to indicate a connexion with some deciduous appendage. 
Had they been prolonged into such persistent organs as true branches, it is difficult to 
believe that clear proof of the existence of such organs would not have been met with 
in some example amongst the hundreds of these stems which I have examined. 
That they were not the vascular axes of leaf-petioles seems clear from their exogenous 
organization. I know of no leaf, however large, in which such an organization exists. 
Hence, as already suggested, I am inclined to believe that they may have nourished 
spikes of fructification, which, being modified branches, might be expected to retain a 
branch-structure in their central axes. I have shown in my last memoir the existence 
of exogenous layers in the axis of a Leyyidostrobus , and I have in my cabinet specimens of 
Volhnannia Binneyi which have a similar radiating external woody cylinder. Hence 
there is no impossibility involved in the supposition that similar conditions may have 
prevailed in Dictyoxylon. These, however, are purely hypothetical suggestions ; and the 
further fact remains, that we have hitherto discovered no fruits that could, with reason- 
able probability, be correlated with the stems in question. 
Whatever may have been their functions, these branches throw a clear light upon the 
origin of the irregular vascular medullary cylinder seen even in very young Dictyoxylons, 
and which only remain, in a fragmentary state, as detached buttresses supporting the 
inner surfaces of the exogenous woody zones. But we must further study this subject in 
the light of the facts described in my last memoir on Lepidodendroid plants. I there 
showed that the compact central vascular axis of each young Lepidodendroid twig first 
expanded into a vascular cylinder, through the rapid fusion and multiplication of some 
of the more central cells located amongst the vessels, and which thus developed into a true 
cellular medulla, the growth of which was synchronous with that of the vascular cylinder. 
I further demonstrated that this compound axis became invested, at a still later period, by 
