350 Mr. J. Ralfs on the Mode of Growth in Calothrix, &c. 
alteration of their direction. This is not the case in Seytonema 
myochrous, which acquires a very different habit owing to the vari- 
ation in the direction of the dislocated extremities. In that plant 
the new ends are curved towards the same side of the sheath ; 
they do not pass each other, but issue from the side together and 
at right angles to its axis. As both portions encounter equal 
resistance they elongate equally, and consequently the branches 
are said to be in pairs. Sometimes the dislocation does not take 
place until after a loop has been formed by a lateral protru- 
sion. Occasionally also the dislocation occurs without any cur- 
vature of the newly formed ends, which then pass each other as 
they do in Calothrix; but this rarely happens except im the case 
of lower or basal dislocations. The presence in the same speci- 
men of both modes of branching proves that they depend on 
modifications of the same law, notwithstanding their very dif- 
ferent appearance. 
Calothriz mirabilis presents another variation in the direction 
of the dislocated ends. At first sight the mode of branching 
appears similar to that of Scytonema myochrous, and different 
only in having more frequent divisions; but closer examination 
detects an essential difference. The filament indeed separates 
as in Calothrix and the ends pass each other; but instead of 
remaining within the same sheath, they immediately pass out 
obliquely in opposite directions ; consequently as both portions 
are free and continue to elongate, they seem merely to anasto- 
mose by cohesion at the convexities of their sheaths. As this 
plant divides at short intervals, it has the appearance of intricate 
network. 
In Rivularia also the branches are the result of dislocation, but 
in that genus a globule is formed at the base of the branch at 
the time of dislocation. 
A similar globule is present in the lower branches of some 
species of Calothriz and Cenocoleus; in these however it is 
usually developed only after the impaction of the branch, but 
sometimes during the division of the filament. If the lower 
portion of the filament elongates and passes the vesicle, its ap- 
pearance does not differ from one formed after dislocation. If 
the lower portion ceases to grow at the time of division, the plant 
is like a simple filament here and there interrupted by a vesicle 
or sporangium. 
In this group, however, the branching of the filament is not 
invariably accompanied by dislocation: in Stigonema I believe 
it never occurs, and even in Scytowema I have seen some species 
allied to myochrous in which the branches were apparently pro- 
duced in the usual manner by lateral protrusion without inter- 
ruption of continuity. 
