192 H. JOEL CONN 



at first glance is the regularity of the numbers of these organ- 

 isms in the unheated infusion. The highest count is 1,500,000 

 and the lowest 400,000. Compared with bacterial counts in 

 general, these show remarkable regularity, especially when it 

 is considered that the soils varied from poor sand to richly man- 

 ured loam and that the counts were made on plates of such 

 high dilution that comparatively few colonies were obtained on 

 each plate. The counts obtained from the heated infusion 

 are not quite as regular; but if the first eight tests are excluded — 

 in which the use of 85° may have killed a few spores — there is 

 scarcely any more variation than in the case of the unheated 

 infusion. 



Because of this regularity in the counts it is possible to ob- 

 tain general averages that can be fairly compared with each 

 other. The average count from the unheated infusion is 

 788,000, from the heated infusion 712,000. This slight differ- 

 ence indicates that there are very few if any of the organisms 

 present in soil in a form that can be killed by the temperatures 

 used. Studying the figures more closely it will be noticed that 

 the greatest differences between the two counts occurred in 

 the first eight tests, in which 85° was used. The average count 

 in these first eight tests, unheated, was 670,000, while the average 

 count, heated, was 445,000. In the last eighteen tests, how- 

 ever, both counts averaged nearly the name, 844,000 and 833, 

 000, respectively. 



A more careful analysis of the data yields similar results. 

 The last column of the table shows the difference between the 

 two counts with a plus sign before it if the count obtained from 

 the unheated infusion was the higher, with a minus sign if that 

 from the heated infusion was the higher. It will be seen 

 that there are eighteen cases in which the plus sign is used, 

 and in these cases the greatest difference was 530,000, or if the 

 tests are excluded in which 85° was used, it is 400,000. On 

 the other hand in the eight tests in which a greater count 

 was obtained from the heated infusion there is one difference as 

 large as 400,000. The average difference between the two 

 counts is 76,000 in favor of the unheated infusion; while if the 



