156 TEMPSKYA. 
cylinder. We must remember, however, that under natural con- 
ditions of growth the case may be otherwise. Be that as it may, 
my examination of the lower parts of cultivated Dicksonia antarctica 
stems does not afford any strong support to Velenovsky’s view. 
Could the thick enveloping mass of roots readily become separated 
from the central vascular axis of a tree-fern stem, and thus account 
for the occurrence in a fossil state of thick bundles of adventitious 
roots without the central stem axis? We know how it has fre- 
quently happened in the case of Lepidodendra stems from the 
Coal-Measures that the central vascular axis has been completely 
separated from the outer cortical tissues, and the latter have thus 
been compressed together, forming an apparently complete 
specimen. This separation in Lepidodendron is easily explained 
by the two concentric lines of weakness which exist in the meris- 
tematic layers, the cambium of the central cylinder, and the 
meristematic zone in the outer cortex. In Dicksonia and other 
tree-ferns we have no such zones of delicate cells, along which the 
tearing apart of tissues might readily take place; on the contrary, 
we have the central vascular tissue with numerous spirally placed 
petioles bound together by the plexus of roots, and it is not easy to 
understand how any separation could be effected during fossilization. 
In Museum specimens of Dicksonia antarctica stems, I am unable 
to detect any tendency to a clean separation of the surrounding 
roots from the central axis. 
Again, it does not seem probable that the central axis would be 
disorganized, and the roots remain as mineralized structures; the 
thick resistant bands of strengthening tissue which accompany the 
vascular plates would be far more likely to withstand weathering 
influences than the smaller root structures. 
Velenovsky suggests that the roots of Zempskya probably drew 
up from the soil the mineralizing solutions, which eventually 
replaced their organic cell-walls. It is perhaps conceivable that 
the roots on the base of the stem may have taken up calcareous 
or siliceous solutions, and that the central vascular axis did not 
offer any such convenient path for their ascent; if this were so 
the axis would gradually decay, and subsequent compression of the 
root envelope might close up the vacant space and leave no signs in 
the mineralized mass of any axial structure. This, however, is 
mere speculation, and probably of little or no value in the solution 
of this difficult question. 
