464 THIRD PART.—BACTERIA OR SCHIZOMYCETES. 
The rods are usually not quite straight but slightly curved. The curvature 
often appeared to me to change from one direction to another in the same individual, 
but it was impossible to feel sure of this point in presence of other movements to be 
described presently. When two sister-rods begin to separate transversely from one 
another, the curvature usually becomes somewhat more pronounced at the extremities 
where the division takes place, and the ends of the rods become slightly oblique to 
one another and overlap each other a little, or one thrusts itself laterally past the 
other, like the short commencement of a so-called false branch in Scytonema and 
similar genera of the Nostocaceae (Fig. 194 4, m). In these conformations the rods 
may either separate from one another in rapid succession, or remain united but 
always loosely united by their extremities into chains of varying numbers of members, 
seldom more than ten (Fig. 134 @). It follows from what has been already said, that 
a chain of this description is never quite straight; its general outline, when it is sup- 
posed to be at rest, is different in individual cases ; it is irregularly undulated, and kneed 
where the members join one another, often forming more or less acute angles, and 
never approaches the form of a regular spiral. But in the stage of development of 
which we are now speaking the chain of rods is never in a state of rest; the rods 
are ink. constant though comparatively slow movement. There is first a movement of 
rotation,.each.rod rotating about its longitudinal axis. Secondly there is a swinging: 
movement; one or both extremities of the rod describes a nearly circular path in 
suchia manner, that in the first case the outer surface of the rod moves in the path 
of the side of a cone the apex of which is formed by the other extremity of the rod; 
in the second case it moves in the path of the sides Of two cones the apices of which 
meet in the middle of the rod after the fashion of an hour-glass. Thirdly, there is a 
movement of progression, the rod moving through the fluid in varying and still undeter- 
mined directions. If the rods are united together into a chain, all the cells move 
together through the fluid in the same direction, but each separate rod shows the 
tendency to the rotation and swinging mentioned above. These movements are 
disturbed or prevented in proportion to the resistance encountered from the cohesion 
of the gelatinous substance which binds the cells together, and from the mass and 
movement of the cohering cells. Hence chains formed of ‚several members move 
forward in irregularly varying directions and with not less irregularly varying 
undulations, and with varying degrees of articulate movement at the points of 
junction. Vibratile cilia or flagella have never been observed in this species. 
If a pure culture is obtained on a microscopic slide from the smallest possible 
number of spores, the Bacillus will be found in 24-28 hours to have multiplied to such 
a degree, that the nutrient fluid, which at first was perfectly clear to the naked eye, 
has become turbid and milky; the organisms themselves are diffused with tolerable 
uniformity through the fluid, or some of them are settling to the bottom. Most but 
not all of the little rods now separate from one another, and each consists of about 
4-6 isodiametric cells. The transverse walls between the cells, though always 
delicate, become gradually more defined. The protoplasm fills the cell-cavity less 
perfectly, and appears to line the wall of the cell and to inclose a clearer central 
space; it is also dotted with a number of granules, some of which are highly refrin- 
gent (Fig. 194 ¢, d). In this stage the cells have ceased to increase in size and to 
divide, and are preparing to form spores. The commencement of the formation of a 
