ROBERT E. WILSON 



generalize among industries, or among different companies in 

 an industry, as to what they can and should do to support basic 

 research. 



As to the first point, that of making profits, a free enter- 

 priser cannot quarrel with this objective, but he can point out 

 that much of industry tends to be too shortsighted in applying 

 this criterion to basic research. Companies which are ac- 

 customed to demand assurance of a payoff of any investment in 

 three to five years have to adjust their thinking before they are 

 willing to embark on a substantial basic research program. As 

 they become educated, however, they realize that a proper pro- 

 portion of basic research vitalizes their whole research program, 

 and also that the stakes, while involving a somewhat longer 

 time scale, are often of a much greater order of magnitude. 



Consider, however, the problem of the typical small com- 

 pany in an industry, say one which does about i per cent of the 

 total volume of the industry. That may seem unrealistically 

 small to a person in the automobile or steel industry, but it is 

 really typical, for many industries, of the company that has 

 started a real research program only since World War II. Even 

 if such a company budgeted as large a proportion of its sales 

 dollar for research as its larger competitors, which it seldom 

 does, its laboratory and staff are necessarily small, but its press- 

 ing problems may be almost as great as those of its larger com- 

 petitors; in fact, they are probably more immediate because the 

 company lacks the background of years of research and the 

 backlog of information built up thereby. In addition, it is some- 

 what handicapped in recruiting good men in competition with 

 larger and better equipped laboratories. Furthermore, it is ques- 

 tionable whether in many fields a laboratory of thirty or forty 

 scientists could profitably devote 8 or i o per cent of its research 

 effort to basic research — in nuclear language, such a group 

 would probably be "subcritical." 



216 



