Report on the Bacteriology of Water. 431 



possible to make the sowings approximately uniform say 12 25 

 spores in each drop. More can hardly be hoped for. 



7. The spores ought always to be germinated out at the same 

 constant temperature, though beyond the confirmation of the fact that 

 some temneratures favour germination better than others, my ex- 

 periments throw little light or further details on this matter. I select 

 22 C. as a useful temperature. 



8. It seems probable from the foregoing that the germinated 

 culture ought always to be transferred to the temperature, &c., 

 at which the growth is to be examined at the same age, so that the 

 filaments to be compared, as to their growth-curves under the new 

 conditions, shall have been exposed to uniform conditions for tho 

 same period of time previous to the study of the new reactions, and 

 so that we may always select filaments of equal length and age from 

 the spore, for comparison. This is undoubtedly a difficult condition 

 to fulfil, and I have by no means always succeeded in fulfilling it, 

 because it requires that the investigator shall be able to begin his 

 work at any stated times. 



9. Lastly comes the question of what we shall compare on the 

 growth-curves. The outcome of the experiments given suggests that 

 it may not be sufficient to compare any two doubling periods, as has 

 been done. Probably it would be better to decide to compare either 

 (1) the doubling of equal lengths e.g., the elongation from 100 to 

 200 ft. at the given temperature, &c., or (2) the doubling period 

 starting always at the same age e.g., whatever the length of the 

 filament after one hour's sojourn at the given temperature, see 

 how long it takes to double that length. Clearly we cannot do both 

 with the same series of cultures, though perhaps the ideal to be aimed 

 at is that both be done eventually. Looking at the importance of the 

 subject, it seems worth while to throw out the suggestion at any rate. 



The following set of cultures in normal gelatine show very clearly 

 that excellent and clear results are to be obtained by observing the 

 above precautions as closely as possible, and adopting the method of 

 recording now to be described. 



1. The culture, when put into the Sachs' box, is allowed half-an- 

 hour for the organism to accommodate itself to the new temperature. 



2. The first doubling period selected is that starting from the half- 

 hour accommodation just mentioned, and all previous growth (i.e., 

 growth during the accommodation period) is neglected. I term these 

 doubling periods " normal doubling periods," in contradistinction to 

 the foregoing doubling periods selected anywhere on the curves. 



3. If we plot out these normal doubling periods, as already described 

 (Curves AB, CD, and EF), we obtain a much more regular curve, 

 which, as will be seen, comes evidently into the category of respira- 

 tion curves. 



