94 THE CYTOLOGY AND LIFE-HISTORY OF BACTERIA 



initially small, increase in size, while the central cells continue to multiply 

 rapidly, and the process is repeated. This corresponds to simple fission in 

 unicellular bacteria, and the two processes are not readily distinguishable by 

 routine bacteriological methods. 



In mycobacteria the difference in size, between the central and terminal 

 cells is not so great, but the mode of division is otherwise similar. 



The alternative method of vegetative reproduction may be found simul- 

 taneously, in the same culture, although different strains of bacteria may 

 show a preference for one method or the other. In this case the individual 

 cells grow until they arc all spherical or oval in shape. They then separate 

 completely from one another and develop into typical, multicellular bacilli. 

 This is superficially comparable with sexual, vegetative reproduction in 

 eubacteria, but the analogy is by no means complete. There is no apparent 

 sexual process involved, and the filament which fragments to produce the 

 new bacilli is not unicellular, but is a chain of cells, differing only in their 

 larger size from those composing the original bacillus from which the filament 

 arose. It appears, in fact, to be a vegetative process as simple as, or simpler 

 than the first. 



Wyckofi and Smithburn (1933) have claimed that the resting stage of 

 M. phlei arises from the fragmentation of the bacilli, and consists of a small, 

 coccal body. But it is not easy to determine from their illustrations whether 

 these workers were concerned with vegetative fragmentation or with the 

 sexual processes which Lindegren and Mellon (1933) showed to precede a 

 more complex mode of reproduction in M. tnhcrailosis (Section D above). 

 The fragmentation method of reproduction serves, in all probability, to 

 explain a considerable number of reports of complex reproductive cycles in 

 tubercle bacilli, including a filterable stage. Criteria of filterability arc purely 

 comparative, but this quality has so often been claimed for M. tuberculosis 

 that it is exceedingly probable that the bacilli may fragment into cells of which 

 a proportion are so tiny as to be capable of passing coarse bacterial filters. 



It must, at the same time, be recognised that reproduction by " gonidia,"^ 

 sometimes of filterable dimensions, has been reported in many other species 

 of bacteria (Chapter VII). 



