Darbishire. — Observations on Mamillaria elongata. 389 
bouring set is 4 to 5-5 mm. The various sets of spines therefore overlap 
considerably (PI. XXV, Fig. 2). 
I will now describe the full-grown spines and the cushion of tissue they 
are inserted on. 
The cushion and its spines are quite definitely separated from the 
underlying tissue of the tubercle. The latter, although it includes 
the vascular tissue, may be considered the living and active part of the 
whole lateral organ ; the former may be called the dead or passive portion 
(PI. XXV, Fig. 15). These two parts are separated by a layer of cambium, 
which is continually adding to the cells of the outer dead tissues. It is not 
at all unlikely that a few cells of the living tissue of the tubercle are also 
derived from this cambial layer. The last few small cells of this tissue at 
least seem to run in regular rows, which end in one of the cambial cells 
(PI. XXV, Fig. 15). 
The cambial layer runs right across the whole tubercle, but it does not 
appear as a straight line in section. It is practically continuous with the 
epidermis of the tubercle, and it joins the latter just inside the small rim, 
with which the tissue of the tubercle surrounds the set of spines. 
The cushion in which the lower ends of the spines are inserted is 
entirely made up of cork-tissue (k). The separate cells are arranged in the 
typical way, namely regular rows, the walls being often much contorted 
owing to unequal pressure. The cork-cells of the cushion at first show 
a clear cell-cavity, but as they get further away from the phellogen they 
become so much twisted about that it is almost impossible to recognize 
any cavity or even the thickness of the cell-wall. The whole cushion may 
reach a thickness of about 600-700 ju, being at the most about 800 //, broad. 
It is entirely cut off from the living cells of the tubercle, even portions 
of the epidermis giving rise to cork-cells (PI. XXV, Fig. 15, g, k ). 
On the cushion are inserted the spines and a number of multicellular 
hairs. Inside the rim formed by the upper end of the tubercle, and growing 
from the corky tissue of the cushion, we meet with a ring of numerous dried 
hairs. These hairs may be no more than 4 \i thick, but about 20 to 40 ju, 
broad (i). They have collapsed almost entirely in one plane and the 
transverse walls project like prominent ridges (PI. XXV, Fig. 20, i). They 
are about 800-900 [jl long, but often are much crumpled and twisted. These 
hairs consist almost entirely of cellulose, except a very thin outer wall 
of cuticle. This circle of hairs is followed by a ring of spines, then follows 
another circle of hairs, and finally a solitary central spine. 
The spines are all of practically the same structure, but the solitary 
central one is larger than the others. It will, therefore, be sufficient to 
describe either one or the other (PI. XXV, Figs. 15, 16). 
In each spine three different tissues can be well distinguished, at least 
at the lower or basal end. At this point the core of the spine, whether it 
E e 2 
