8 



A Simple Model of 

 Technological Research 



In the previous chapters research was related to productivity. This 

 chapter presents an attempt to formulate a mathematical model of 

 research work. The basic concept of the model is the concept of 

 the experiment. The scientific experiment is the means for 

 verification or rejection of hypotheses; the field trials of the 

 agronomist or the plant breeder are experiments, so also farmers 

 gain knowledge from experiments— mostly of the "trial and er- 

 ror" type. 



Our approach was first suggested by the study of the develop- 

 ment of sugarcane varieties reported in chapter 3. It owes much to 

 George Stigler's work on the economics of information (Stigler 

 1961). Nelson's treatment of research and development (Nelson 

 1961) has also provided insight. In chapter 3, the history of sugar- 

 cane varietal development was demarcated into four stages. A 

 brief recounting will be useful. 



Stage 1 described the long period of history during which 

 genetic improements were based on rare cases of "natural'' sexual 

 reproduction. Only a small number of improved commercial 

 varieties emerged from this process over centuries. 



Stage 2 marked the beginning of organized technological dis- 

 covery effort in sugarcane varietal improvement. Researchers in 

 the experiment stations in Barbados, British West Indies, and in 

 Java discovered in 1887 that the cane plant could be induced to 

 flower and reproduce sexually. Crosses between parent plants 

 produced thousands of seedlings, each of which could be 

 vegetatively reproduced as a new variety. The analogy between 

 this discovery activity and search is relatively easy to draw. 



140 



