280 PROCEEDINGS : BOSTON SOCIETY NATURAL HISTORY. 
embryonic epidermis are thus the primordia of the trichomes. For 
the sake of brevity I shall apply to them in this paper a distinc¬ 
tive term, and shall speak of them as trichoblasts. 
The distance from the apex at which trichoblasts begin to appear 
varies with the size of the root and the rapidity of growth. I have 
often found them forming within 10 or 12 cells of the growing point, 
in the case of rootlets line enough to be studied without sectioning 
(pi. 18, fig. 71). 
In the greater number of cases, the trichoblast is cut from the 
upper or proximal (basal) end of the mother cell, e. g., Aneimia 
adiantifolia (pi. 16, fig. 1), Isoetes, Selaginella, Lycopodium, Lim- 
nocharis, Commelina, Nymphaea; in some cases, however, from the 
lower, e. g., Azolla, Juncus, Luzula, Scirpus, Alopecurus (pi. 18, fig. 
65). 
The relative sizes of the trichoblast and the atrichomic sister cell 
vary in different groups of plants, and to some extent even in the 
same species. The disparity may be slight, as in many Gramineae 
(pi. 18, fig. 68). 
In Azolla it would appear from the ascertained position of the 
newly formed daughter nuclei and of the connecting fibers that the 
axis of the mitotic figure, in the mitosis giving rise to the tricho¬ 
blast, is inclined (pi. 16, fig. 5). In Azolla, Lycopodium, and prob¬ 
ably in other cases, the obliquity of the segmenting wall is fixed in 
the process of cell division. It is not due to subsequent changes in 
the forms of the cells. 
In most species of Lycopodium the division wall does not extend 
to the inner face of the mother cell, but abuts internally upon the 
proximal end wall of the cell (pi. 16, fig. 20). The trichoblast 
therefore has a sharp inner edge, and appears like a wedge inserted 
between two adjacent epidermal cells. It was mainly this peculiar¬ 
ity which drew the attention of the earlier anatomists to the special¬ 
ized character of the hair-cells in Lycopodium, while the no less 
specialized structures to be found in so many groups have escaped 
general notice. Nageli and Leitgeb explained the origin of the 
trichomes in Lycopodium (see note under Lycopodium below). 
They neglected the almost equally striking trichoblastic elements 
of the young root of Equisetum, the development of which root 
they described in detail. De Bary (Comparative anatomy, 1884, p. 
60) says: “Only in Lycopodium can special hair-cells be distin- 
