56 CRYSTALS AND LIVING UNITS 



attributes of living matter, may well be marked in the products 

 which separate from the most fermentable solutions ; and it is 

 precisely this attribute which is the principal factor in bringing 

 about the self-multiplication, or ' discontinuous growth,' of living 

 units. 



Thus it is that rapid growth and rapid fission frequently go on 

 simultaneously ; so that although the total amount of living matter 

 which separates from a solution may be large, the individual living 

 units may be very small. Discontinuous growth is in excess, and 

 therefore the fact of the growth being really rapid is apt to be 

 overlooked. 



All the differences in size and form recognisable between 

 Micrococci and small Bacteria, between Bacteria and Bacilli or 

 Spirilla, between the latter and Vibriones, whether jointed or un- 

 jointed ; between Vibriones and Leptothrix filaments, plain or 

 segmented in various ways ; and between Leptothrix and the very 

 delicate mycelial filaments of many Moulds, are easily explicable 

 in accordance with these considerations. The several forms, many 

 of which frequently occur together in the same solution, depend, in 

 the main, upon the frequency with which segmentation tends to 

 occur, and upon the degree of completeness of the process. 



The forms of Torulas, again, are almost infinite in variety as met 

 with in different situations, and they are often notably different 

 even in the same solution. They may vary in size from the 

 minutest visible speck to a vesicle ao^io " or more in diameter. 

 They may after a time become spherical, ovoidal, ellipsoidal, or 

 cylindrical in form, and produce buds in different directions. 



It is impossible for us to assign any ultimate reason why one 

 rather than the other of these forms should manifest itself. We 

 can only observe that in some solutions different forms of Bacteria 

 most frequently present themselves, while in others Torulae are 

 most prone to occur. It has been known, for instance, since 

 the time of Dutrochet that the organic forms met with in acid and 

 alkaline or neutral solutions vary ; and it has been frequently 

 observed by others, that Torulae are most apt to present themselves 

 in slightly acid solutions. Again, while the most putrescible 

 solutions almost invariably yield Bacteria, the same fluids, after 

 their fermentability has been greatly impaired by the influence of 

 heat, may produce nothing but Torulae. Torulas are generally more 

 frequent in saline solutions than Bacteria, and in some of these, 

 after they have been boiled, no Bacteria ever present themselves. 



