34 THE METHOD OF SCIENCE 



patent. If we are possessed of normal faculties, we have only to 

 observe in the right places and at the right times and we are able 

 at once to note them. Experiment would be useless and absurd, 

 indeed impossible, here. Thus we could not by experiment ascer- 

 tain that a species of butterfly was winged and contained several 

 varieties. There could have been no such sciences as zoology and 

 botany had we depended on experiment. Therefore their students 

 have relied almost exclusively on simple observation. The right 

 times and places in which to observe may be hard to find, but, 

 once found, the phenomena can be noted. 



59. In other sciences, for example, physics and chemistry, many, 

 indeed most of the facts are obscured. Thus we cannot by our 

 unaided senses discover the composition of water or alcohol, or the 

 way in which gases behave under pressure. Therefore physicists 

 and chemists have been obliged to rely very largely on experiment. 



60. The study of heredity, which draws its materials from every 

 science that deals with life, stands midway. Here the laboratory 

 cannot create a science ; it can only help to create one. Most of 

 what we know about life is patent to our senses and can be simply 

 observed. Especially is this the case with regard to facts ot 

 structure, for example, the shape, size, and anatomical relations of 

 the parts of an animal. But some facts, especially those relating 

 to function (e.g. the function of the thyroid gland), are obscured, and 

 to discover them, if it be possible, we must resort to a laboratory 

 method. Therefore, if we say, or imply that facts supplied by 

 simple observation are useless to the students of heredity, we 

 declare, in effect, that most facts relating to structure and many 

 relating to function are useless we declare, in effect, that when 

 studying heredity we must not draw our facts from systematic 

 zoology, botany, comparative anatomy, embryology, palaeontology, 

 indeed most sciences that deal with life. 



61. In contradistinction to simple observation, the laboratory 

 methods of inquiry have sometimes been termed ' exact ' ; but, 

 very obviously, the term, implying as it does that simple observation 

 is necessarily less exact, is erroneous. They are especially exact 

 on particular occasions only. For example, we do not need them 

 to make us entirely sure that men resemble apes physically more 

 than they do other animals, that the offspring of human beings 

 are human, that every generation follows closely but not exactly 

 in the developmental footsteps of the one that precedes it, that the 

 forms of life which have inhabited the earth have changed with the 

 geological epochs, that stags have antlers and elephants have tusks, 



