[vanderleck] bacteria OF FROZEN SOILS 3 



February, and sampling had to be postponed until the second half 

 of April. 



In each experiment samples were taken on the five, ten and 

 fifteen inch levels and the depth of the frost line determined. The 

 levels were slightly changed from the year before, when all samples 

 were on four, eight, twelve and sixteen inch levels. In frozen soil 

 the differences are not sufficiently important to make a change of one 

 or two inches. Thus the sample of the five inch level practically 

 represents the surface conditions; the ten inch level the area of 

 greatest bacterial activity, and the fifteen inch level subsoil. 



All bacteria found in the samples were again classified in three 

 groups suggested by Conn, but this interesting phase of the investiga- 

 tions will be taken up in detail next winter. 



As was done in the previous winter, the plots were again selected 

 to afford a large range of conditions. In Experiemnt I a rich arable 

 soil, well manured, can be compared with the rich arable soil of Experi- 

 ment II which was not manured and not cropped, but fallowed for 

 four years. In Experiment III the arable soil of Experiment II is 

 cropped but not manured. Finally the plot on the permanent lawn 

 was subjected to the full severity of a Canadian winter from the very 

 first. In the 1916-17 experiment this exposure to the frost did not 

 commence until the middle of February. One sample was also taken 

 on the outdoor rink as a comparison with a similar sample taken 

 the previous year. 



Methods. 



The methods employed were those of the previous year, but the 

 soil was frozen to such hardness that sampling wich pick and crowbar 

 was no sinecure. With the exception of the first and last samples 

 the soil was frozen solidly throughout the winter, and the samples 

 were collected in large chips or flakes. These were broken up and 

 the stones picked out. In a few cases samples partly thawed in trans- 

 portation to the laboratory and were allowed to thaw completely. 

 For water determination three samples of 50 gr. each were taken; for 

 bacteriological determination 2 grams of the frozen powder were 

 put in 100 c.c. sterile water and shaken for at least five minutes. 

 Further dilutions were made so that in each case 1 c.c. of the proper 

 dilution were made so that in each case 1 c.c. of the proper dilution 

 was added to the plates. All samples were plated in duplicate on 

 beef peptone ager containing 1% glucose sugar and ^% blue litmus, 

 beef peptone gelatine and soil extract gelatine. Only one dilution 

 was used for each set of six plates. 



The beef peptone agar and beef peptone gelatine were prepared 

 according to the formulae given by the American Public Health 



