THE DATA OF THE SCIENCES 211 



must be systematized, or they are of no greater scientific value 

 than the amorphous mass of data with which casual experience 

 stores our minds. The methods by which facts are collected and 

 arranged differ according to circumstances. 



348. In some sciences, for example systematic botany, zoology, 

 and anatomy, the facts are nearly all patent and, once ascertained, 

 fall so naturally and easily into the right categories that the 

 difficulty of systematization, of discovering the relationships 

 between the facts, is reduced to a minimum. In others, the ex- 

 perimental sciences, for example physics and chemistry, very few 

 facts are patent to our senses : the main difficulty lies in ascertain- 

 ing them. Therefore special methods of investigation have been 

 devised. In yet other sciences, for example heredity, though 

 many facts are patent, others are obscured, and the relationships of 

 nearly all are hard to ascertain. Consequently we have difficulties 

 both in discovering information and in arranging it. The patent 

 facts can be simply observed. The obscured facts must be rendered 

 patent by means that clear away the obscuring conditions. The 

 difficulties in arranging the facts must be met by testing our think- 

 ing with especial care. In this way only is it possible to achieve 

 an approach to exactness. 



349. Very remarkable notions as to what constitutes science 

 are prevalent amongst sections of biologists. Thus medical men, 

 who are biologists in the sense that they are students of a living 

 being, frequently denounce deduction. Apparently they are un- 

 aware that the correctness of inductive thinking can be tested 

 efficiently only by means of it, and that mathematics, the purest 

 and most accurate of sciences, the model for the rest, is almost 

 purely deductive. Again, systematic zoologists and botanists 

 sometimes denounce 'speculation.' That is they denounce all 

 thinking which is not very easy, all attempts to ascertain relations 

 that are not very obvious. Yet, again, some experimental observers, 

 appealing to physics and chemistry, sciences that are both accurate 

 and experimental, often insist that biology must be made accurate 

 by being made experimental. But physics and chemistry are not 

 accurate because they are experimental, but because the data 

 on which they are founded are capable of being as precisely 

 measured as it is possible to measure anything, and because 

 the students of them test their thinking with great care. In 

 biology, we must, of course, when necessary, use experiment as 

 well as every other means of ascertaining truth. But that is one 

 thing. Quite a different thing is what has actually happened in 



