$ ix.] Kefir. Bacilhis Amylobacter. 99 



It appears, as has been already said, in the form of small slender 

 rods united into filaments, which are closely interwoven and held 

 together in Zoogloeae by means of a jelly. 



The source of the grains has not been traced further back 

 than the leather milk-bottles of the mountaineers ; their place of 

 birth is still unknown. They come to us in the dry state, and 

 are kept in this manner in the Caucasus also. They must be 

 dried quickly, and the best plan is to dry them in the sun. 

 Much of the dry imported material is dead when it comes, as 

 far as my experience goes. The softened living grain grows 

 slowly in the milk, as we have already seen (page 59), with 

 uniform increase in size and multiplication of all its parts. 

 This growth is accompanied by the separation from time to time 

 of single lobes of different sizes from the whole, and thus the 

 number of the grains increases. From isolated observations I 

 regard it as possible that Dispora-cells sometimes issue from a 

 grain, and may then develope into kefir-grains, but this is not 

 certain. Distinct formation of spores has not yet been observed. 

 Kern it is true has not only described such a formation, but 

 named the Bacterium of kefir Dispora, because two spores are 

 formed each time in a rod, one at each end. After repeated 

 observation I have never seen anything of the kind, though I 

 have very often seen figures which answer to Kern's represent- 

 ations, and which are due to the circumstance that a rod or 

 portion of a filament is curved and its middle portion lying 

 horizontally is seen in its length, but one or both of its 

 extremities which bend away from the horizontal plane are 

 viewed in cross-profile. It is by such appearances that Kern 

 has allowed himself to be misled. If we allow the name 

 Dispora to be used provisionally, it must not be forgotten that 

 the character which it is intended to express does not really 

 exist. 



7. We will close the series of examples of the Bacteria which 

 excite characteristic fermentations in non-nitrogenous com- 

 pounds by the consideration of a species of Bacterium which is 



H 2 



