Mr. J. Ralfs on the Mode of Growth in Calothrix, ^c. 349 



In Scytonema, Calothrix, Arthroncma, &c. the sheath is some- 

 what cartilaginous and closely surrounds the contained filament. 

 As its texture is comparatively firm, it admits only a slight degree 

 of dilatation : it neither separates as in Oscillatoria, nor allows 

 the bundling of the filaments as in Microcoleus. 



In all these genera the structure of the filament, irrespective 

 of the sheath, is alike, and consists of a single, longitudinal series 

 of disciform cells which are often confluent or have indistinct 

 dissepiments. 



If a specimen of Calothrix or Canocoleus be examined we may 

 frequently observe, especially near the extremities of the branches, 

 short separated portions of filaments in every respect similar to 

 those which sometimes occur in Oscillatoria. At first such a 

 portion is separated from the original filament by a short in- 

 terval ; but as there is no division of the sheath and both portions 

 continue to elongate, they are soon in contact again. In the act 

 of passing each other the extremities sometimes become atte- 

 nuated. In this state the filament looks as if it had divided 

 obliquely, and the upper portion becoming impacted between 

 the filament and the sheath presents the appearance of a branch. 

 From this explanation it will be evident that the branches in 

 these genera are produced, not by an adhesion of other fila- 

 ments, but by a dislocation of the filament itself. 



Both portions continue to elongate upwards, and branches are 

 thus repeatedly formed by dislocation. The upper portions or 

 branches, however, always retain their original advantage and 

 extend beyond the trunk. This fact seems to me a strong proof 

 of the correctness of the view I have given, for it could scarcely 

 be constant if the branches originated in any other manner. 



The frond or sheath is itself truly branched or divided in the 

 ordinary way. Sometimes, as in Calothrix, it is forked as soon 

 as the upper portion becomes impacted, and the plant presents no 

 peculiarity to the eye in its mode of branching except that the 

 branches at the base are not united to the trunk. 



In Canocoleus the branching of the sheath does not occur at 

 the same spot as the dislocation of the filament. Upon this cir- 

 cumstance depends the peculiar character of the genus, for after 

 the dislocation the inferior portion as it elongates necessarily 

 pushes itself up by the side of the superior one. Sometimes the 

 filaments are again branched by dislocation before the sheath 

 divides, and thus from two to four (or even more) filaments pass 

 up side by side within a common sheath. Where the sheath 

 forks the filaments are in general equally distributed between its 

 branches. 



From what I have stated it will be seen that in Calothrix and 

 Ccenocoleus the dislocated ends pass each other without any 



