ole, i. A. MINCHIN. 
with only one cell on the spicule, this cell being situated on 
the lance-head in the position of the thickener. At first I 
regarded this state of things as normal, but I am now con- 
vinced that it arose simply by the brush having removed the 
founder, which is often situated very close to, or even among, 
the collar-cells. (Compare Figs. 48 and 71 6.) 
(2) Leucosolenia variabilis. In this sponge the 
ordinary monaxons vary greatly in size but agree in their 
characters, and form one class in which every gradation can 
be found between the two extremes of size. They have dis- 
tinct lance-heads and long, curved shafts. Two other types 
of monaxons are also found, but they are very scarce, and I 
have not found any stages of their development. 
The ordinary monaxons may conveniently be grouped into 
small (fig. 83), medium (fig. 86), and large (fig. 89), but these 
three sizes are connected by every possible gradation, and 
must not be regarded as indicating distinct subdivisions. 
There is nothing to note about their mode of development 
which has not been said already, but some conclusions of 
importance from a systematic point of view may be drawn 
from the stages of growth. It is sufficiently obvious from 
the figures that small monaxons like fig. 83 are by no means 
to be regarded as young stages of large monaxons like 
figs. 88, 89. On the contrary the future size of the spicule is 
determined at a very early stage—that is to say, as soon as 
the lance-head is formed (compare figs. 78 and 84). Further, 
the study of the development affords a means of distinguishing 
the spicules that have or have not attained their full growth, 
as already pointed out. In describing the spiculation of a 
sponge for purposes of specific identification it is most impor- 
tant that the description should apply only to fully-formed 
spicules, and not to those in which the growth is not 
complete. 
The mode of development of the monaxon spicules that I 
have described in the foregoing is one that I have found in 
all the monaxons of Leucosolenia that I have studied, and 
Woodland, whose preparations I have looked through, has 
