Iieporl on tJte Bacteriology of Water. 371 



depends ; this causes the cell-chains to be so modified in length, 

 direction, and rapidity of development, that the colony formed from 

 the iusolated germ is weaker and more condensed, or compact, than 

 normally. Thus result the differences in the naked eye characters of 

 the colonies, which may go so far that the total aspect on the gelatine 

 plate is altered. 



Some clue to the action may perhaps be got eventually by following 

 up the fact that one consequence of the light action is to weaken the 

 enzyme action of the bacterium for the enfeebled liquefying power 

 is an expression of enfeebled enzyme power either by so altering the 

 protoplasmic machinery that less enzyme is secreted, or by so acting 

 on the enzyme that its power of converting the medium is altered. 

 The lessened enzyme power of course implies less power to obtain its 

 food from the medium, and so the progeny developed from the germ 

 started with are also feebler than the normal one. 



In the cases quoted, however, the colony gradually becomes more 

 normal as regards its enzyme power, especially in the dark, because 

 the successively developed new cells become stronger and stronger as 

 they are fed by the nutrient gelatine, and at last the differences are 

 equalised. 



The only source of error in the above conclusion that I could think 

 of, was the possibility that the rapid running of the colonies into 

 thin filaments in the first case is facilitated by the quicker liquefac- 

 tion of the gelatine, and that this liquefaction is, in turn, more rapid, 

 because there are so many more colonies per drop in the unexposed 

 flasks, because the majority of the bacilli in the exposed flasks are 

 killed. 



I accordingly made plates in which I diluted the samples from the 

 unexposed flasks five, ten, and even twenty times as much as the 

 samples from the exposed flasks, and so brought the numbers of 

 colonies on each plate approximately equal. Of course it may be 

 replied here that the differences of dilution may bring about diffe- 

 rences in development; but experience shows that increased dilution 

 tends to inhibition of colonies, and so I think the fact that I still get 

 the differences in the colonies already described, strengthens rather 

 than weakens the evidence that the alteration in the character of the 

 colonies from exposed flasks is really due to the action of the light. 



On November 7, which turned out a beautifully fine day, with 

 clear blue sky and bright sun, tubes of sterile distilled water were 

 infected with cultures of a yellow bacillus marked *, a large -white 

 one marked </, and a violet bacillus, common in the Thames. In each 

 case, two similar tubes were prepared, exactly alike, one of which was 

 exposed over a mirror, from 10.30 A.M. to 3.30 P.M. ; while the other 

 was wrapped in foil and black paper, and placed by the side of the ex- 

 posed one. The temperature recorded in control tubes was 10 12 C. 



