ORGANIZATION OF INDUSTRIAL RESEARCH. 
making readily accessible those vast stores of specialized knowledge which 
research has already accumulated, but which still require to be brought into that 
systematized and orderly arrangement which characterizes science. The research 
laboratory should be built around a library. These special libraries should be 
linked together, and closely affiliated with the great libraries of the wrold. Such 
admirable journals as Chemical Abstracts should be substantially endowed, the 
publication of monographs encouraged and assisted, while systematic reviews 
and reports of progress covering limited special fields in science and technology 
should appear far more frequently. The intensive collection of scientific and 
technical information throughout the world, its codification and its distribution, 
might well be made a governmental function to an extent not now approached. 
Experience has shown that, in the organization and conduct of industrial 
research laboratories, certain well-established principles are of general applica- 
tion. 
It is desirable that the laboratory be separately housed, and that an institu- 
tional atmosphere be developed and maintained. It should be recognised that 
research is necessarily expensive, and that results are not forthcoming over night. 
The ovulation of ideas takes time, and their material embodiment proceeds slowly, 
and seldom can be hastened without danger. Adequate equipment, which is often 
costly, is essential for good work, and a liberal policy must be followed where 
requisitions are concerned. It is well to have the laboratory expenditure 
arranged on the appropriation system, and to allow the director a free hand 
within the appropriation. If the laboratory is a part of an industrial organiza- 
tion, it should maintain close contact through conferences and reports with all 
departments of the organization; but no department should be allowed to dominate 
the laboratory, and its director should be responsible only to the executive of the 
corporation. Since no organization can include every department of human 
knowledge, the director should be encouraged to make frequent use of the advice 
and services of outside specialists. 
“The gods send threads to webs begun,” which is the classical way of saying 
that we have to get into the water to learn to swim. It is, therefore, well, in case 
of major problems, to start somewhere as well as may be, even though the starting 
ground be not wholly to one’s liking. It is easily possible to spend too much 
time in preliminary searches of the literature, and better to avail of early 
enthusiasms in some initial work at least, while carrying on concurrently the 
tedious task of reviewing and abstracting literature. 
Nothing is more expensive or demoralizing than experimentation in the plant. 
An industrial research laboratory should, therefore, be adequately provided with 
equipment of semi-commercial size. Infant mortality among processes is high in 
any case, and the most critical period in their young lives is that covering the 
transition from the laboratory to the plant. They require, and the research 
laboratory should provide, a nursery to protect and foster them during this period 
of their development. Some large manufacturers have even found it desirable 
to operate in connexion with and under the sole direction of their research 
laboratory a small plant in which actual commercial manufacture is regularly 
conducted, Such extension of the laboratory’s function permits the complete 
reduction to practice of new methods, and the commercial demonstration of the 
sufficiency of the product before the innovations are introduced into the main 
plant. 
Even when no such provision appears feasible, it is, nevertheless, ‘highly 
desirable to have the industrial research labo ratory actually engaged in some 
C.8706.—5 177 
