249 
Ellis . — The Life-history of Bacillus hirtus . 
solution of Fuchsin or Methylene-blue. The exact amount to be added can 
easily be ascertained after a little experience. The division-wall is seen 
as a very narrow band stretching across the cell, perpendicular to the 
long axis of the cell (Fig. 24). It appears to stain less intensely than 
the mature wall, when it first makes its appearance, but this may be due 
to an optical effect, the result of its narrowness. The next stage in 
the process is the formation of the constriction (Fig. 25). Two processes 
now seem to go on concurrently, the extension of the constriction and 
the partition of the division-wall. In some cases the constriction 
proceeds so quickly that the cell is already half cut off before the 
division of the wall has taken place (Fig. 26). In other cases division 
of the wall has taken place before the constriction has advanced a third of 
the way across (Fig. 27). The new transverse wall is divided by a clear 
space running longitudinally across the division-wall (Figs. 26, 27). The 
latter at this stage is seen as two very thin lines, one on each side of 
the clear line, and by their narrowness easily distinguishable from the 
wider, more deeply staining walls of the rest of the cell. This clear space 
is due to a mucilaginous change in the composition of the division-wall, at 
the place where a clear space is seen. There is no doubt that for some 
time there is at this stage protoplasmic continuity between the separating 
cells, though it cannot be demonstrated at this stage of the division. The 
next stage in development after the appearance of the thin clear line con- 
sists in the gradual strengthening of the new wall, with a concurrent 
rounding off of the end, until ultimately the constitution of the division- 
wall is not distinguishable either in density or in breadth from the remainder 
of the wall (Fig. 28). When two dividing cells have reached this stage 
they gradually draw away from each other, though they are often still con- 
nected. Sometimes two cells which have separated and are half a length 
apart move and wriggle about as if a rope connected the two. And such 
in fact is the case. In Fig. 29, stained by Grain’s method, the union 
between the cells by means of a protoplasmic cord is evident. In Figs. 30-34 
are shown on a larger scale the stages in the division of the cell, showing the 
two different ways in which the final stages in the division are completed. 
With regard to the time required for the whole process of division 
to take place, Meyer ( 9 ) asserts that in one of his observations an hour 
elapsed between the appearance of the division-wall and the formation 
of a definite constriction. It is natural to expect that there will be great 
variation in the rate, but in nature the process must, under favourable 
conditions, take place at a much faster rate. In one experiment which 
I made in connexion with this point a one-cell individual divided up until, 
roughly speaking, about 128-140 cells had been formed. This represents 
the seventh order of division. To form these cells about three hours were 
required in this particular instance. I could not follow it further because 
