110 A. Hl. Graves, 
is a layer of small, thick-walled cells, lacking intercellular spaces 
—evidently a strengthening layer (Text-fig. 26, ex). This may be 
designated the epidermoidal layer, as described by Juel (1884), or 
exodermis, as defined by Vines (1898, p. 111) and Strasburger 
(1908, p. 102). 
(1) The Root Hairs. In a recent paper, Leavitt (1904) has pub- 
lished the results of his investigations on the root hairs of the higher 
plants, a subject which has hitherto been much neglected and con- 
cerning which only scattering references can be found here and 
there in the literature. 
Leavitt has ascertained that the root hairs are of two types, 
depending on their manner of development. The first type may arise 
from any ordinary epidermal cell. The second type of trichome; 
common to most monocotyledons, arises from a _ specialized cell, 
which, in its embryonic state, Leavitt has termed a trichoblast. 
It is to this second type that, as briefly noted by Leavitt, (1904, 
p- 292) Ruppia maritima belongs. According to my investigations, 
the root hairs arise here in the following manner. At a_ short 
distance back of the growing point, certain of the epidermal cells 
begin to differentiate. They are easily distinguished by their deeply 
staining, granular contents, and their larger nuclei, (Pl. VI, figs. 29, 
32, 33). As to their size, there is some variation, but on the whole 
they are of about the same size as the ordinary epidermal cells. 
These peculiar cells are the primordia of the trichomes, called the 
trichoblasts, which by the subsequent development of tube-like pro- 
jections from their free walls will produce the root hairs (Pl. IX, fig. 48). 
As to their mode of origin, there seems to be no regularity. 
Often two or three non-piliferous cells alternate with the tricho- 
blasts, but again, as many as a dozen or more ordinary cells may 
occur between two trichoblasts (Pl. VI, figs. 29, 32, 33). 
Not all of the trichoblasts develop root hairs. They may remain 
simply as large cells with granular contents and large nuclei. They 
are evidently, then, as Leavitt has also noted, potential trichomes, 
the development of root hairs depending possibly on some stimulus. 
This explains why we find zones, often at considerable distances 
apart, where the hairs are developed in abundance. 
A remarkable characteristic of the root hairs is their persistency. 
In roots which measured over 20 cm. in length, quite a number 
of hairs were noted still intact and apparently functioning even 
at the base of the root near the coleorrhiza. This condition is 
very common and may be due to the fact that the hairs are not 
worn off easily in the soft mud. Often, however, hairs do not 
