34 REPORT OF DEPARTMENT OF BACTERIOLOGY OF THE 
In 1905 the selection of the land and the laying out of the plats were 
ieft to the farmers except that an uninoculated portion or check 
was required. In 1906 general directions for selecting and fitting 
the land designed for the experiment were furnished and it was sug- 
: gested that the experimental 

acre be laid out according to 
No lime, no Lime, no aes 
the adjoining diagram. In 
inoculation. inoculation. 
1907 a compliance with the 
é; latter suggestion was insisted 
No lime, Lime and con h fore 
: ; upon except where, for lo 
inoculation. inoculation. P pr ‘ ea 
reasons, a different arrange- 

ment was evidently prefer- 
able: 
The uninoculated area was intended to furnish evidence as to the 
amount and distribution of any inoculation originally present in 
the field as well as serving as a basis for judging the effect of the 
artificial inoculation or the lime applied to the adjoining areas. 
The experimenters were supplied with blanks for recording the 
nature and condition of the soil at seeding time as well as the con- 
dition of each plat at intervals during the first two seasons. In 
most cases the reports were supplemented by personal visits each 
season, when the records were verified and enlarged as much as 
possible. . 
In attempting to draw conclusions from experiments conducted in 
this manner we have been sensible of the difficulties connected with 
obtaining accurate data and we have excluded incomplete experi- 
Down-hill side of field. 
ments so rigidly that our actual calculations are based upon only 
67 of the 118 experiments which were actually begun. These 67 
experiments have, for the most part, been personally examined and 
are known to have been so conducted as to give dependable re- 
sults. Among them are a few which were not seen but which were 
so well arranged and the reports from which are so clear as to leave 
little room for doubt as to the actual conditions. 
It will be noted from the plan of the experiments begun in 1906 
that lime was to be applied to one-half of the experimental area. 
The results already obtained show that lime is often an important 
factor in the growing of alfalfa. A detailed discussion of our re- 
sults with lime will be given in connection with our report, next 
‘year, upon the experiments which were begun in 1907. 
