68 THE QUANTITATIVE METHOD IN BIOLOGY 



each species. This work has not been a waste of time. 

 Although the figures of one species, taken separately, are of 

 little importance, from the whole set of specific figures interest- 

 ing conclusions may be drawn. At first is appears that the 

 number of hairs is not very variable in certain species, whereas 

 it is very variable in the majority of them. Secondly, all the 

 species (ninety in number) being arranged in a series according 

 to the value of the property under consideration, a gradual 

 transition is observed from the lowest figure to the highest. In 

 this series certain species which are (with regard to other prop- 

 erties) very different from each other are placed side by side, 

 and certain species which are looked upon as being allied with 

 one another are separated by a considerable distance. Thirdly, 

 the species which are at both extremities of the series are dis- 

 tinctly different by their figures, in spite of variation. There- 

 fore the extremes are easily distinguished by the primordium 

 number of hairs of the lahrum, without regard to any other 

 property.^ 



FOURTH EXAMPLE (INTER-SPECIFIC CORRELA- 

 TION) : We find in a classic book, which contains the descrip- 

 tion of about thirty-five species of the genus Carabus, that in 

 one of these species the ultimate joint of the maxillary palp is 

 shorter than the penultimate. For all the other species no 

 information is given about the relative length of the joints 

 mentioned. We are also told that the ultimate joint of the 

 maxillary and labial palps is broad in certain species and 

 narrow in others,^ but for the majority of the species this char- 

 acteristic is passed over in silence. In order to replace this 

 fragmentary information, which is rather irritating by its 

 vagueness and incompleteness, by more useful notions, I have 

 measured in ninety species five properties — viz. (i) length of 

 the ultimate joint of the maxillary palp ; (2) breadth id. ; 

 (3) length of the penultimate joint id. ; (4) length of the ultimate 

 joint of the labial palp ; (5) breadth id. All the figures being 

 arranged in order in the same way as in the third example, the 

 conclusions are similar with regard to each of the five properties. 

 Such material is exceedingly interesting with reference to the 

 correlation between several properties in a series of species. 

 This form of correlation might be called inter- specific. It is 

 something quite different from the correlation between proper- 

 ties of one species. 



FIFTH EXAMPLE: In the classical descriptions of the 

 leaves of the mosses, the length and the breadth of the leaves 



1 1 intend to give a complete account of these and many other similar facts 

 in a work on Carabus and Calosoma which I hope to publish later on. 



* The significance of the terms broad and narrow is, of course, an enigma 

 which ought to be guessed by the unfortunate reader. 



