THE ART OF SCIENTIFIC INVESTIGATION 



old notes, one's memory of the experiments has become dim so 

 that the task is more difficult and cannot be done so well. Also, for 

 reasons discussed elsewhere, it is desirable to review the problem 

 periodically. However, work that has not produced significant 

 results is better not published. It cluttei^s up the journals and 

 does more harm than good to the author's reputation in the minds 

 of the discerning. 



When the work has been completed, it is wise to submit the 

 article to an experienced colleague for criticism — not only because 

 the colleague may be more experienced than the author, but also 

 because it is easier to see flaws in another's work or language than 

 in one's own. 



A word of caution might be given against publishing work that 

 is not conclusive and especially about making interpretations that 

 are not fully justified by the experimental results or observations. 

 Whatever is written will remain permanently in the literature and 

 one's scientific reputation can be damaged by publishing some- 

 thing that is later proved incorrect. Generally speaking, it is a 

 safe policy to give a faithful record of the results obtained and 

 to suggest only cautiously the interpretation, distinguishing 

 clearly between facts and interpretation. Premature publication 

 of work that could not be substantiated has at times spoilt the 

 reputation of promising scientists. Superlatives and exaggeration 

 are anathema to most scientists, the greatest of whom have 

 usually been modest and cautious. Faraday wrote to a friend in 

 1831 : 



" I am busy just now again on electro-magnetism, and think I 



have got hold of a good thing, but can't say. It may be a weed 



instead of a fish that, after all my labour, I may at last pull up." 



What he pulled up was the electric dynamo. In 1940 Sir Howard 

 Florey wrote to the Rockefeller Foundation for financial sup- 

 port for his work on penicillin, which he then had good reason 

 for believing could be developed into a therapeutic agent even 

 more effective than the sulphonamides. In such a letter one might 

 be expected to present the work in the most favourable light, but 

 this is all that Florey allowed himself to say : 



" I don't think I am too optimistic in thinking that this is a 

 very promising line."'^ 



What a classic piece of understatement that has proved to be ! 



136 



