*>7 
make some such preliminary test as I have described, or, if the en- 
tire product of the plots were harvested, to keep the different parts 
of it distinct, rejecting those data which show an evident influence 
of one plot upon the other. 
In an ordinary plot experiment, if I infer from the outcome of 
the treatment of a lot of trees in the midst of an orchard what will 
happen if an entire isolated orchard were similarly treated, I 
tacitly assert that such a plot, so situated, is so like an isolated 
orchard in all that concerns my experiment that the one may be 
substituted for the other. As a matter of fact, however, while a 
small part of an orchard is analogous to the whole of it in many 
particulars, it may differ from the whole in just those particulars 
and conditions which most strongly affect my problem. 
Since reasoning by analogy is the only process possible in such 
a case, we must first make this analogy practically complete — must 
establish the essential similarities between our experimental plot and 
an entire orchard — before we can reach any safe conclusion as to 
the practical application of the experimental results. This I have 
done to the best of my ability in the experiment here referred to, 
and I venture to think that like results may be expected, under sim- 
ilar conditions, from a repetition upon entire orchards of the treat- 
ment here described. 
Orchard I. Sprayed Plot, Rows 1 to 11; and Cbeck, Rows 12 to 22 
Row 
Trees 
examined 
Per cent, 
injured 
Averag"e 
1 
2 
3 
4 
5 
6 
7 
8 
9 
10 
2 

1 

3 
4 
1 

1 

28.95 
29.89 
29.73 
28.59 
16. 
26. 
26.53 
11 
3 
43.33 
43.33 
12 
3 
81.5 
81.5 
13 
14 
15 
16 
17 
18 
19 
20 
21 
22 

2 

3 
1 
2 

1 

1 
92.23 
88.08 
87.7 
87.19 
93.79 
98.96 
90.33 
