49° BOTANICAL GAZETTE [DECEMBER 
specialized elements. These cells differ in the formation of their 
cell plate, the wall lies somewhat diagonally, and the cell differs 
from the other epidermal cells in shape, size, and content. This 
type he finds in the Schizaeaceae, Equisetum, Azolla, Lycopodium, 
Phylloglossum, Isoetes, Selaginella, and in Nymphaeaceae. SNOW 
(21) finds that no definite length can be given as the limit for the 
formation for hair development; that in some roots the average 
length of piliferous cells is less than that of the smooth cells, but 
that the differential elongation of the epidermal and cortical cells 
is important, and that hair formation depends upon their ratio, 
that is, between the capacity of the epidermal cells to elongate 
and their ability to do so. 
BARDELL (1) comes to the same conclusion as Snow. A few 
of the measurements given by BARDELL are shown in table I. 
TABLE I 
Plant a. —— ei gay ag a — 
Pisum sativum...... 231 161 (—71) | 191 (—40) 
4 Fr ae 179 139 (—40) | 172(— 7) 
. Tea ie. cS 69 66 (— 3) 77 (+ 8) 
Oi ten: 69 = 77 (+ 8) 
a re 40 55 (+15) 55 (+15) 
SS as 36 33 (3) 
eer ee ne 124 58 2 
It is difficult to draw the conclusion which is given, for in the 
one case a difference of 40 mm. in length between epidermal and 
cortical cells calls forth a root hair; while in another the same differ- 
ence accounts for the absence of a root hair, and measurements 
equal to, or greater than, or less than, seem to account for haired 
or unhaired epidermal cells. 
SUCCESSION.—SCHWARz (20) found that hairs were always 
produced in acropetal succession, while Devaux (2) asserted that 
new hairs could appear among the old ones; whereas HABERLANDT 
(8) agrees with Scuwarz (20), emphasizing the fact that no new 
hairs ever arise among existing ones. 
PosITION OF THE NUCLEUS.—HABERLANDT (8) noted the posi- 
tion of the nucleus in Brassica alba to be at the tip. In Pisum 
