480 PROCEEDINGS: BOSTON SOCIETY NATURAL HISTORY. 
branches. Certain species of Dictyostelium also often show, in 
luxuriant growths, branched fructifications, in which the axis bears 
at irregular intervals one or more secondary fructifications, usually 
of unequal length (pi. 8, fig. 109). In both instances, the branched 
portion is similar in all essential respects to the main fructification 
and is attached externally to it. In the case of Dictyostelium, the 
branching is sometimes evidently accidental; as, for example, when 
the stalk of the fructification has fallen, there may arise from late 
ascending colonies erect branches, which are fixed by means of mucus 
to the inclining main axis in a manner similar to the fixing of the 
principal stalk to the substratum. In other instances, however, when 
the growth is vigorous, small colonies, ascending below the larger 
pseudoplasmodium at the summit, become diverted from the axis 
and form a stalk placed at right angles or obliquely to the main 
stalk. The formation of such branches might be explained as the 
result of loss of water, so that further upward movement of the 
colony is impossible. Such a conclusion is made the more credible 
by the fact that tardy colonies, or even individuals having sufficient 
vigor, may sometimes be seen to reach the main aggregation at the 
summit. When food is especially abundant, a fructification of 
Dictyostelium sphaerocephalum^ for example, has been seen to bear 
as many as four or five branches of unequal length (pi. 8, fig. 109) ; 
and in this form, as well as in others, under similar conditions, it is 
not unusual for several diverging fructifications to spring from a 
common base. 
As mentioned above, Polysphondylium is the only member of the 
group in which ordinarily the fructifications are regularly branched. 
The fructifications of this organism, however, present many varia¬ 
tions as to the number of the branches, since, while some may even be 
simple and unbranched, others may bear but one, or, more often, a 
considerable number of secondary axes, usually arranged in whorls 
along the main axis. A mature fructification may sometimes have 
as many as ten whorls of branches, all of about equal length and dis¬ 
posed at approximately equal distances along the axis, the undermost 
whorl made up of five or six rows, the uppermost of from one to 
three (pi. 8, fig. 118). 
The phenomena connected with the formation of such branched 
fructifications in Polysphondylium may be described as follows. 
The erect pseudoplasmodium in its earlier stages cannot be dis- 
