J. Henderson Smith 35 



If we write the last table in a different form, taking as 1 the time 

 required to kill 75 per cent, of the spores, we get Table VI, 



The corresponding time-ratios for a unimolecular curve are added to 

 this table, and it will be seen that the observed figures approximate 

 more and more closely to the logarithmic type, the greater the strength 

 of phenol used. 



It is unfortunately not practicable to use higher concentrations of 

 phenol at this temperature, 25° C. With 0-8 and 1-0 per cent, the 

 reaction runs so fast in the early stages that it becomes impossible 

 accurately to sample quickly enough. In 0-8 per cent, phenol, for 

 example, 43-27 per cent, survive for 30 seconds, 5-7 per cent, for 1 minute, 

 3-6 per cent, for 1*5 minutes, and 0-39 per cent, for 2-5 minutes, Abeady 

 in half a minute the number of survivors falls from 100 to 43, and it is 

 impossible to observe the important early part of the curve. 



Effect of Low Teniferatures. An attempt was therefore made to work 

 at much lower temperatures, in the hope of so slowing the reaction as 

 to make observation practicable. All apparatus, the phenol solution, 

 the suspension of spores, etc., were previously cooled to the temperature 

 of melting ice, from 1-5° to 2-0° C, and the samples dropped into ice- 

 cold w^ater and centrifuged as quickly as possible. Control experiments 

 showed that germination w^as not altered in numbers by the exposure 

 to cold, but was rendered a Uttle slower during the subsequent incu- 

 bation. With 2-0 per cent, phenol at 2-0° all spores were killed in 20 

 seconds. With 1-5 per cent, phenol at the same temperature 1-32 per 

 cent, survived 21 seconds, and 0-52 per cent, survived 41 seconds. 



At the low temperatures it was possible to observe fairly satisfactorily 

 the course of the reaction ^\'ith 1 per cent, phenol. With higher concen- 

 trations the process was too rapid for detailed examination (see Tables 

 VII-XI). It is apparent, however, that with 1*25 per cent, the sigmoid 

 character is still present: and with 1 per cent., whether at 1-5° or 7-8°, 

 the curve is frankly sigmoid (v. Fig. 3), as markedly as with 0-7 per cent, 

 at 25°, even although the reaction proceeds more quickly. These results 

 will be referred to later (r, p. 49). 



3—2 



