EXPERIMENTATION 



The careful recording of all details in experimental work is 

 an elementary but important rule. It happens surprisingly often 

 that one needs to refer back to some detail whose significance 

 one did not realise when the experiment was carried out. The 

 notes kept by Louis Pasteur afford a beautiful example of the 

 careful recording of every detail. Apart from providing an 

 invaluable record of what is done and what observed, note- 

 taking is a useful technique for prompting careful observation. 



The experimenter needs to have a proper understanding of 

 the technical methods he uses and to realise their limitations and 

 the degree of accuracy attainable by each. It is essential to be 

 thoroughly famihar with laboratory methods before using them 

 in research and to be able to obtain consistent and reUable results. 

 There are few methods that cannot at times go wrong and 

 give misleading results and the experimenter should be able 

 to detect trouble of this nature quickly. Where practicable, 

 estimations and titrations of crucial importance should be checked 

 by a second method. The scientist must also understand his 

 apparatus. Modem complicated apparatus is often convenient 

 but it is not always foolproof, and experienced scientists often 

 tend to avoid it because they fear it may give misleading results. 



Difficulties often arise in organising experiments with subjects 

 over which there is only limited control — human beings or 

 valuable farm animals. Unless the basic needs of the controlled 

 experiment can be satisfied it is better to abandon the attempt. 

 Such a statement may appear self-evident, but not infrequently 

 investigators find the difficulties too great and compromise on 

 some arrangement that is useless. Large numbers in no way offset 

 the necessity of a satisfactory control group. The outstanding 

 illustration is supplied by the story of B.C.G. vaccination in 

 children. This procedure was introduced twenty-five years ago 

 and was then claimed to protect people against tuberculosis; but 

 although a large number of experiments have since been carried 

 out, there is still to-day controversy as to its value in preventing 

 the disease in people of European stock. Most of the experiments 

 have proved nothing because the controls were not strictly 

 comparable. The review on B.C.G. vaccination by Professor 

 G. S. Wilson provides a good lesson in the difficulties and pitfalls 

 of experimentation. He concludes : 



17 



