40 
Keport of the Director of the 
The series, as given, are not comparable for the purpose of obtaining 
information as to which seedsmen supply the best seed, as the number of 
trials differ. To make this evident, selecting the first six of our names, 
and arranging according to the magnitude of the germination per 
cents in this seed, we have the following .order: 
Per cent. 
Tillinghast 94.7 
Thorburn 81.8 
Sibley 78.8 
Landreth 78.1 
Henderson 76 . 9 
Gregory 71.8 
As we had but eight trials with Tillinghast, let us take the germin- 
ation results of the first eight trials of the seed from the other 
seedsmen; our order of arrangement then stands: 
Per cent. 
Tillinghast , . . . 94.7 
Landreth 86.2 
Gregory 80 . 
Thorburn 77^8 
Henderson 77.4 
Sibley 73.6 
This order would in turn be changed by any different grouping we 
could make from our. figures, and if we desire to obtain information of 
this kind, we must use the same number of packets of the same varie- 
ties from each seedsman, and germinate under like conditions. From 
the trials here presented, we can legitimately refer to the seed from 
Tillinghast being of superior quality, but we cannot justly discriminate 
between the others named. We should mention, however, that Tilling- 
hast's seeds were apparently in some cases selected for the purposes 
of trial, while the other tests were from strictly commercial sanrples. 
