494 BOTANICAL GAZETTE [DECEMBER 
definite relation between the length of the cortical cell and the 
epidermal cell which decides whether the epidermal cell shall 
become extended into an outgrowth or not. Fig. 3 shows a haired 
cell longer than the adjoining cortical cell, and fig. 4 shows two 
short epidermal cells adjoining one cortical cell. These measure- 
ments are corroborated by observations made 
throughout the work, many forms and all 
possible variation in length of cortical and 
' epidermal cells being found. 
Position of nucleus.—The position of the 
nucleus has been found to have no relation 
whatever to the hair or to its initial forma- 
tion. When the wall first swells, the entire 
wall bulges; later there is a greater swelling 
fe ae. in one portion of it, the nucleus may not be 
haired cell longer ies near that portion; if 
adjoining cortical cell; the nucleus is near 
fig. 4, two short epider- that portion, it may 
mal cells adjoining one 
cet al not follow the growth 
of the hair and be 
found near the tip of the hair, but remain 
in the base of the hair or in another part 
of the cell. There appears to be no rela- 
tion between the position of the nucleus 
and the formation of the hair. 
Fic, 3 Fic. 4 
I. IN AIR MEDIUM 
a) The physical character of walls 1, 2, Fic. 5 rn, 
4, 6.—The first evidence of the formation Fics. aan ener 
of a root hair is the bulging of the entire Sa ae on? ee 
wall z, as shown in fig. 5; then a portion — .yolten portion. 
of the wall stretches more than the rest 
and there is a narrowing of the swollen portion, as in fig. 6; but it 
may be noted that the base of a root hair is always greater 
in diameter than the rest of the hair, as in fig. 7. Seedlings of 
endive and lettuce demonstrate this fact. Branching of the hairs 
is common, indicating a difference in the physical character of the 
