THE ART OF SCIENTIFIC INVESTIGATION 



This is an important prerequisite to any experimental work, 

 and occasionally investigators who have neglected it undertake 

 laboratory work which has little relation to the real problem. 

 Appropriate laboratory examination of specimens is usually 

 carried out as an adjunct to this field work. 



Farmers, and probably lay people generally, not infrequently 

 colour their evidence to fit their notions. People whose minds are 

 not disciplined by training often tend to notice and remember 

 events that support their views and forget others. Tactful and 

 searching enquiry is necessary to ascertain exactly what they have 

 observed — to separate their observations from their interpreta- 

 tions. Such patient enquiry is often well repaid, for farmers have 

 great opportunities of gathering information. The important 

 discovery that ferrets are susceptible to canine distemper acose 

 from an assertion of a gamekeeper. His statement was at first 

 not taken seriously by the scientists, but fortunately they later 

 decided to see if there was anything in it. It is said that for 

 two thousand years the peasants of Italy have believed that 

 mosquitoes were concerned with the spread of malaria although 

 it was only about fifty years ago that this fact was established by 

 scientific investigation. 



It is helpful at this stage to marshal and correlate all the data, 

 and to try to define the problem. For example, in investigating 

 a disease one should try to define it by deciding what are its 

 manifestations and so distinguish it from other conditions with 

 which it may be confused. Hughlings Jackson is reported to 

 have said : " The study of the causes of things must be preceded 

 by the study of things caused." To show how necessary this is, 

 there is the classical example of Noguchi isolating a spirochaete 

 from cases of leptospiral jaundice and reporting it as the cause of 

 yellow fever. This understandable mistake delayed yellow fever 

 investigations (but the rumour that it led to Noguchi's suicide 

 has no basis in fact). Less serious instances are not infrequently 

 seen closer at hand. 



The investigator is now in a position to break the problem 

 down into several formulated questions and to start on the 

 experimental attack. EKiring the preparatory stage his mind will 

 not have been passively taking in data but looking for gaps 

 in the present knowledge, differences between the reports of 



10 



