RESEARCH IN THE EXACT SCIENCES. 1 85 



mass of data will only provide probabilities and suggestions. Any 

 man with ideas on the subject of meteorology would after a little 

 experience discard existing material and start afresh, or else waste 

 his best 3-ears in trying to reduce material to a common measure, 

 which is really a hopeless task. 



(i>) Medical Statistics. — These are made by each medical man and 

 each hospital on a separate plan, and without any idea, as a rule, of 

 the points which it is needful to observe in order that logical con- 

 clusions may be drawn. This is especially the case in inheritance of 

 disease tendencies. Further, immense masses of material are wasted 

 because one or other essential factor has escaped record in one or 

 other series. 



We have had to report recently on cancer .statistics, lunacy statis- 

 tics, and inoculation for enteric fever statistics. Only moderately- 

 definite conclusions can be drawn, because the material has usually 

 been collected without insight into the conditions requisite for draw- 

 ing definite statistical results. 



(c) Physical Measurements. — The same applies here, in perhaps a 

 less degree, but still quite definitely. Ob.servations on the strength 

 of materials exist in immense quantities. These are largely of no 

 value because the experimenters have had no clear idea a priori of 

 the points they wanted to elucidate. Further, this applies to a 

 whole mass of physical observations which have been made without 

 sufficient mathematical knowledge to realize the difficulties of the 

 problem. The failure on this account of physicists like Wertheim, 

 Savant, and Kupffer in the first half of the nineteenth century is 

 quite paralleled in recent work by men whom for obvious reasons it 

 is better to leave unnamed. 



(yd) Biological and Sociological Observations. — These are of the 

 lowest grade of value in too many cases. Even where the observers 

 have begun to realize that exact science is creeping into the biolog- 

 ical and sociological fields they have not understood that a thorough 

 training in the new methods was an essential preliminary for effective 

 work, even for the collection of material. They have rushed to 

 measure or count any living form they could hit on without having 

 planned ab initio the conceptions and ideas that their observations 

 were intended to illu.strate. I doubt whether even a small propor- 

 tion of the biometric data being accumulated in Europe and America 

 could b}' any amount of ingenuity be made to provide valuable re- 

 sults, and the man capable of making it yield them would be better 

 employed in collecting and reducing his own material. 



