VI 



PREFACE 



calculation whatsoever. Each constant is a direct expression 

 of an observed fact. 



I have been asked several times in what way the described 

 method was to be applied to the investigation of certain given 

 properties of certain animals or plants. My answer is that in 

 each peculiar case a preliminary investigation is required in 

 order to find the primordia which are to be measured and the 

 way in which the measurements are to be carried out. This 

 preliminary work may be a long one. 



For example, I began in December 19 14 the study of the 

 British species of the genus Mnium (Acrocarpic Mosses). I 

 spent two months on the preparatory work before I had found 

 a dozen of the primordia and established their exact definition 

 and the method of measuring them. In a similar way it took 

 me in 1916 about three months to determine exactly twenty-five 

 primordia of the fertile stem of the Grasses. In 1907 I began 

 the measurement of thirty-eight primordia of the species of the 

 genus Carabus after a preparatory investigation of the simple 

 properties of the Coleoptera and other insects which lasted 

 several years. Such work is, I think, an excellent school for 

 the biologist, because he becomes aware of the vagueness of 

 many of our notions. As soon as we require to make measure- 

 ments, we feel again and again compelled to replace approxi- 

 mations by exact definitions. It may be hoped that the 

 preparatory work under consideration will become easier in 

 proportion as the number of examples investigated becomes 

 greater. 



The work needed for the measurement of the primordia of 

 animals and plants is rather lengthy, but biologists ought not 

 to shrink from the idea of undertaking such a laborious task. 

 It may be recalled that the physicists, the chemists and the 

 mineralogists have already collected hundreds of thousands 

 of constants at the cost of tremendous toil. They have per- 

 severed in spite of numerous difficulties. For example, in 

 order to establish exactly the constants (atomic weight, specific 

 gravity, melting-point, etc.) of a given metal, a pure specimen 



