Stigmarian Rootlets . 569 
top of the cotyledon, which forms the absorbing organ of the 
seedling, of a considerable development of fibrous cells. These 
spirally-marked cells are ‘ short and parenchymatous/ and re- 
mind him of the tracheids found under the epidermis of the 
glands of Drosera. In the latter the tracheids are no doubt 
to a great extent excretory as well as absorptive, but in this 
‘ Saugorgan ’ of Agave the function of these tracheids must 
be exclusively absorptive, and they would therefore readily 
compare with the cortical tracheids in the Stigmarian rootlet. 
The only roots which offer any counterpart to the vascular 
strands with tracheidal terminations in Stigmarian rootlets are 
the highly modified roots of certain parasitic Phanerogams. 
Here, as Solms (’68) has shown, the haustoria are often tra- 
versed by vascular strands quite similar in structure to the 
ends of bundles in the foliar expansions. We have, therefore, 
here an instance of roots becoming modified for special reasons 
very much in the same way as the Stigmarian rootlets were 
adapted to their peculiar conditions. 
That the roots of Stigmaria should require some special 
provision for facilitating the absorption of water might be 
expected from a consideration of their general structure. In 
some cases, as in the rootlet reproduced in Fig. 3, a thin band 
of parenchyma was the only connexion between the stele and 
the outer cortex, and through it alone could a passage of water 
take place from the peripheral to the central portion of the 
rootlet. In other cases there appears to have been only 
a delicate trabecular tissue constituting the middle cortex, 
and in some there was possibly no definite connexion between 
the stele and the outer cortex, except by such vascular strands 
as are described above. In all cases there seems sufficient 
reason for the existence of special vascular communication 
between the central cylinder and the peripheral tissues to 
ensure an adequate supply of water. 
It should be remembered that in the roots of Calamites, 
which probably existed in similar conditions of climate and 
habitat, the roots were not monarch, and had numerous stout 
rays of parenchymatous cells running from the epidermal 
