STUDY OF PELOMTXA PALTJSTEIS. 379 



were to be seen, but a very large Eotifer was discovered, wMcli 

 had escaped my notice previously, but now attracted it by an 

 unusually milky-white appearance. 



The suspicion that the Eotifer had devoured the missing 

 Felomyccce was confirmed by microscopic examination, and as no 

 more material could then be obtained, it was determined to tease 

 up the Eotifer in water, in the hope of finding the bacteria and 

 refringent bodies in useable condition. This was done, and the 

 contents of the Pelomyxce were found to be practically unchanged. 

 The refringent bodies separated out in the water, and the bacteria 

 were found to be quite alive, and were seen to be attached to 

 the walls of the refringent bodies in great numbers, and to be in 

 active " wobbling " movement on these. Seen " end-on," they 

 presented such a curious and interesting appearance as to induce 

 continued careful study. In the course of 1| hour's watching a 

 very curious thing happened ; the debris of the Hotifer contained, 

 among other things, a good many Eotifer ova, and the bacteria 

 gradually detached themselves from the walls of the refringent 

 bodies and swarmed upon the ova of the Eotifer instead, and this 

 in such numbers that the ova soon were quite covered and could 

 only be distinguished from the refringent bodies (in the former 

 condition of these) by focussing through to find the nuclei. The 

 refringent bodies were eventually quite deserted, and remained 

 clear and homogeneous. Clearly, then, the ova possessed an 

 attraction for the bacteria, which might be of a purely chemio- 

 tactic nature ; but on the other hand the bacteria might be 

 attracted to the ova either as a source of oxygen or as a food- 

 supply. It seemed unlikely that bacteria whose natural habitat was 

 in the interior of living protoplasm should be highly aerobic, and 

 therefore the hypothesis that they regarded the ova as a fresh 

 source of food seemed the most likely. At any rate this was taken 

 as a working hypothesis, and the first conclusion to which it led 

 was that the theory which applied to the ova might also apply 

 to the refringent bodies, and that the long-observed relation 

 between these and the bacteria might be that of food-supply. 

 A second inference drawn from the observation was that, if the 

 bacteria were attracted to feed upon albuminous bodies, such 

 as ova, some form of albuminous liquid would be the proper 

 medium in which to attempt their cultivation ; and further, that 

 some light was now thrown upon the probable composition of 

 the refringent bodies themselves. Subsequent investigation (in 



