117 
lar bundles, which proceed to the large cylindrical rootlets 
of the plant. The radiating series of vessels which are im- 
mediately vertical to each of the quincunctially disposed 
lenticular medullary rays are projected downwards for a short 
distance, like a tongue, into the lenticular spaces. Down to 
this point, the component vessels are disposed vertically, 
hut they became suddenly deflected outwards, at right 
angles to their previous course, to reach the rootlets for 
which they are severally destined. The deflected vessels 
are very numerous, but the greater part of them disappear 
in succession, only a limited number finally constituting the 
bundle occupying the centre of each rootlet. 
Professor Williamson pointed out the important bearing 
which these facts have upon the affinities of the Sigillaria of 
which Stigmaria is the root. He showed that not only the 
true Lepidodendra, but also the Lepidodendroid stems which 
Mr. Binney has described under the name of Sigillaria Vas- 
cularis, never could have belonged to the same plant as these 
Stigmarian roots. In the plants indicated the central or 
medullary axis is occupied by scalariform vessels intermingled 
with remarkable forms of scalariform cells, as already shown 
in the case of Lepidodendra by Mr. Carruthers, and which 
equally characterise the other plants referred to. It appears 
improbable, being contrary to all known facts, that the serial 
stem should have such a structure, whilst in the roots its 
vascular scalariform tissues were replaced by cellular paren- 
chyma of an altogether different type and character. The 
conclusion to be drawn from these observations is that we 
are yet as far as ever from all actual knowledge of the inter- 
nal organisation of the Sigillarne. For the two principal 
specimens from which the above conclusions were drawn, 
