OF THE FOSSIL PLANTS OF THE COAL-MEASURES. 
9 
spond very closely with mine from the Oldham Upper foot-coal and Ganister-bed. The 
principal difference between the French and English examples that I have seen is in 
their size, some of the former being nearly four times the diameter of the largest of 
those which I have obtained from our older coal-fields. I am indebted to M. Renault 
for a beautiful specimen from Autun which is fully two inches in diameter. I have, as 
yet, met with nothing like this in our Lancashire beds ; but it corresponds closely 
enough with the dimensions of the swollen bases of the petioles of Angiopteris erecta 
and others of the larger Marattiacese. M. Renault finds the gum-canals in his speci- 
mens more numerous in the centre of the petiole than at its circumference. In mine 
these canals are pretty equally distributed over the entire transverse section. Such 
canals are found abundantly in most of, if not all, the Marattiacese, though they are not 
confined to this aberrant type of the Ferns. I find them very well represented in the 
petioles of Cibotum princeps. 
In the petioles of Marattia fraxinea and M. laxa these canals are of large size, but 
chiefly found in the medullary parenchyma, especially in its more central portions, being 
wholly absent from the hypodermal prosenchyma. 
On the other hand, in Angiopteris Teismaniana , A. erecta , and Marattia ascensionis , 
they not only exist in the parenchymatous portions, but they are equally abundant, 
though of smaller size, in the dense layer of sclerenchyma * which encases the inner 
parenchyma. In several of my figures of the fossil forms I have represented these canals 
as being more or less filled with a cylindrical rod of black carbon, as in the example of 
fig. 17, c 1 . I have seen nothing like this in any of the vascular tubes, whilst in the 
canals it is the common condition. It appears to me that this difference is due to 
something more than mere infiltration of carbonaceous matter in the case of the inter- 
cellular canals ; in all probability the black substance is the carbonized residuum of the 
gum with which these canals were once filled. 
M. Renault has already called attention ( loc . cit.) to the fact that whilst in our fossil 
examples the hypodermal sclerenchyma forms an interrupted layer, in the recent 
Marattiacese it forms a continuous one. The inner margin of my sections of Angio- 
pteris erecta display an irregular dentate outline, which approaches in some slight degree 
to that of fig. 1. M. Renault has further pointed out that ordinarily in Angiopteris 
there are isolated bundles of sclerenchyma within and detached from the continuous 
hypodermic layer. As already shown, I find these islets to be wanting in Marattia laxa 
and M. fraxinea (that is, in the two species which exhibit no gum-canals in the sclero- 
dermic layer), whilst they are present in Angiopteris Teismaniana , A. erecta , and Marattia 
ascensionis ; they are also more conspicuous in the thick bases of the petioles than in 
their slender upper extremities. 
The differences in the thickness of the sclerenchymatous layer in my specimens 
* I have used this word, not in the limited sense in which it is employed by Mettenitts, but in the broader 
one suggested by Sachs in his ‘Lehrbuchder Botanik,’ 2nd edition, p. 76, where he proposes the application of 
the term to all hardened prosenchymatous as well as parenchymatous cells. 
MDCCCLXXVI. C 
