14 THE ROYAL SOCIETY OF CANADA 
livestock and enables us to propound improved methods for the 
conduct of the art or practice of farming. It seems clear to me that it 
is to agricultural science, and to this source alone, that we must look 
for that knowledge which can permanently advance our agricultural 
practice. 
The problems of agriculture naturally cover a very wide field, 
and as a result are exceedingly varied; further, they are frequently 
very complex in their character. They call for profound uninterrupted 
and continuous study. As a rule they are difficult to solve, chiefly 
from the large number of factors and limiting conditions usually 
involved and the necessity of noting and accurately gauging their 
significance, before any interpretation of results can be attempted. 
As in other departments, research in agriculture to be successful 
necessitates careful, skilful planning; there must be nothing of the 
haphazard in its conduct. It may proceed by experimental methods, 
but certainly not by a series of disconnected, unrelated experiments. 
It calls for a close, scrutinizing observation, clear thinking and a sound 
judgment, more especially in its field work, throughout its various 
stages and-an ordered marshalling of the facts before their final inter- 
pretation. Throughout the whole work, I repeat, there must be sys- 
tem and scientific thoroughness, if the results are to be of per- 
manent value and generally applicable. 
1 have spoken of experimental work as forming an integral part 
of investigation and research, for it is by experiment that we arrive 
at the facts, the facts necessary for the establishment of the prin- 
ciples. But the experimental work, taken by itself, is not necessarily 
of the nature of research. Isolated facts as the result, say, of field 
experiments, even if well proven by repeated and careful trials, seldom 
have any value for extended application. Much of the experimental 
work in agriculture to-day is of this nature, lacking scientific method, 
plan and completeness. The mere trying out of this and that in an 
indiscriminate matter, without any due regard to the laws of chemistry 
and biology that may be involved and without taking into account the 
numerous modifying factors and influencing conditions, without a 
study of the causes that may affect the results, while it may yield 
information of local importance, cannot add to the store of per- 
manent knowledge of wide application. It is of little value for the 
enunciation of principles; in a word, it cannot advance agricultural 
science. 
In saying this I do not wish to be understood as speaking deroga- 
torarily of the agricultural experimental work of which there is so 
much to-day, though certainly a good deal of it might be of greater 
value if conducted with more care and thoroughness. It has given 
