Smith—The Organization of the Colony . 1191 
form;, but the general axial relationship® cannot be changed by 
external conditions. 
In another connection I have described the manner of forma¬ 
tion of the young colonies (37), suggesting that in the variation 
of the cleavage planes forming the daughter cells lies the chief 
cause for the variations in the cellular arrangement of the coe¬ 
nobe. The four daughter cells are formed by a primary trans¬ 
verse cleavage of the mother cell, this division being followed 
by simultaneous cleavage of the two daughter cells, at right 
angles to the primary cleavage plane. The four cells thus 
formed then elongate and become arranged in two tiers within 
the old mother cell wall. During the elongation of the daughter 
cells the planes separating them commonly become parallel to 
the long axis of the mother cell. In this way the linear type of 
colony is formed. At other times the planes of separation do 
not become parallel to the long axis of the mother cell but re¬ 
main at an angle to it. Under these conditions the cells of the 
coenobe will not have the linear arrangement, when they unroll, 
but the alternate arrangement. It is difficult to say why these 
separation planes become parallel to the long axis of the mother 
cell in some cases and do not in others. These cleavage planes 
vary considerably in the angle they make with the long axis of 
the mother cell and so as a result there are certain colonies of the 
alternating type that are much more pronounced than others. 
That the type of coenobe which will be formed can be pre¬ 
dicted before the liberation of the daughter colony is seen in 
cells just ready to liberate young coenobia (Figs 73-75, Plate 
LXXXVIII). Figure 74 shows a cell thei contents of which will 
unroll into a coenobe whose cells are arranged alternately, while 
the cells of one of the young colonies in Figure 79 will form a 
linear series. All of the cells in a single colony do no produce 
the same type of coenobe but one cell may produce a coenobe in 
which the cells are arranged alternately, while the one next to 
it may produce the linear type of colony (Figs. 73 & 79). 
The colonies whose cells will be arranged in an irregular 
mass can also be determined at the time of the liberation of the 
young coenobe. Thus the young coenobe shown in Figure 69 
will form a colony of the irregular type shown in Figure 59. 
