OF THE FOSSIL PLANTS OF THE COAL-MEASUIIES. 
315 
shown that, as each branch grew, a rapid and very large increase took place in the 
number of the vessels constituting the medullary cylinder of the Lepidodendroid plants ; 
but I have been unable to satisfy myself whether these new additions were developed 
centrifugally or centripetally as in living Lycopods. The difficulty of determining this 
point arises from a circumstance which marks an additional distinction between the living 
and the extinct forms. In the former, the vessels first produced in the procambial axis 
retain their relative positions permanently: the new vessels added to them merely 
occupy, in succession, the spaces intervening between older ones, without either disturb- 
ing the latter or enlarging the area which the entire bundle occupies. The case is 
wholly different with the Lepidodendroid plants : in them we have no evidence that 
the vascular elements developed in this manner. However we regard my medullary 
cylinder, whether as exclusively composed of foliar bundles or of a combination of 
foliar and stem-bundles as in living Lycopods, there is no perfect parallelism between its 
arrangements and those of the recent plants. Beginning in the fossil form at the tip of 
a twig as a small vascular bundle, we have seen that it gradually enlarged its area 
until it became a cylinder of considerable dimensions. Some of the primitive cells out 
of which the vessels were developed, and which I presume have been procambial, 
obviously increased by a prolonged meristem action until they produced a central axis 
of permanent parenchymatous tissue representing a medulla, the pressure occasioned by 
the growth of which was probably the cause of the centrifugal movements of the vessels 
composing the rudimentary vascular medullary axis. But whilst we may thus account 
for the displacement of the vessels, it is difficult to explain the origin of the numerous 
additions to their number in the vascular ring taking place coincidently with the growth 
of the stem. We find no traces of reproductive procambial cells interspersed between 
the vessels of this vascular zone. Were these additional vessels produced through the 
cells of the pith-meristem within the cylinder, or through those of the cortical one 
external to it X We must look to one of these sources for their origin; and my own im- 
pression is that their true source was the innermost layer of the cortical cells. If so, 
their development was centrifugal, or in the opposite direction to that of their living 
allies. But there is yet a further distinction to be recorded. We have seen that in the 
vascular bundles of the recent Lycopodiacem each central vascular bundle is flanked on 
either side by a layer of prosenchymatous fibre. I find no trace of any such tissue 
occupying a similar relation to the vessels of the fossil plants. In the latter these vessels 
are either unmixed with any cellular tissue whatever, as in the Diploxyloid forms of the 
cylinder, or they are distributed through a mass of mere parenchyma, which appears to be 
a permanent tissue, as in the medullary axis of Lepidodendron selaginoides*. Thus we 
find that even in those structures which are generally accepted as the representatives of 
the vascular bundles of the living Lycopods, the fossil forms differ very widely from the 
recent types, both genetically and in their composition. 
Such differences occurring in the central vascular axis (so universally accepted as 
* Philosophical Transactions, 1872, Part 1, Plate sxiv. fig. 1. 
MDCCCLXXII. 2 U 
