THE ART OF SCIENTIFIC INVESTIGATION 



giving significant results. Sketchy experiments are only justifiable 

 when preliminary to more elaborate experiments planned to 

 give a reUable result. Each stage of the investigation must be 

 established beyond reasonable doubt before passing on to the 

 next, or else the work may be condemned, quite properly, as 

 being "sloppy". 



The essence of any satisfactory experiment is that it should 

 be reproducible. In biological experiments it not infrequently 

 happens that this criterion is difficult to satisfy. If the results of 

 the experiment vary even though the known factors have not 

 been altered, it often means that some unrecognised factor or 

 factors is affecting the results. Such occurrences should be 

 welcomed, because a search for the unknown factor may lead 

 to an interesting discovery. As a colleague remarked to me 

 recently : " It is when experiments go wrong that we find things 

 out." However, first one should see if a mistake has been made, 

 as a technical error is the most common explanation. 



In the execution of the experiment it is well worth while 

 taking the greatest care with the essential points of technique. 

 By taking great pains and paying careful attention to the im- 

 portant details the originator of a new technical method some- 

 times is able to obtain results which other workers, who are less 

 familiar with the subject or less painstaking, have difficulty in 

 repeating. It is in this connection that Carlyle's remark that 

 genius is an infinite capacity for taking pains is true. A good 

 example is provided by Sir Almroth Wright's selection of the 

 Rawlings strain of typhoid bacillus when he introduced vaccina- 

 tion against that disease. Only quite recently, since certain 

 techniques have become available, has it been found that the 

 Rawlings strain was an exceptionally good strain for use in making 

 vaccine. Wright had carefully chosen the strain for reasons which 

 most people would have considered of no consequence. Theobald 

 Smith, one of the few really great bacteriologists, said of 

 research : 



" It is the care we bestow on apparently trifling, unattractive 



and very troublesome minutiae which determines the result." ^^ 



Some discrimination, however, should be used, for it is possible 

 to waste time in elaborating unnecessary detail on unimportant 

 aspects of the work. 



i6 



