184 H. S. H. WARDLAW. 



between the values of the quantities associated together. 

 Karl Pearson (1903) has regarded all phenomena as being, 

 to a certain degree, contingent upon one another. In 

 phenomena or quantities of the first type the degree of 

 contingency is high, and may be expressed numerically as 

 approaching unity, where a contingency of 1 represents 

 absolute dependence. In phenomena or quantities of the 

 second type the degree of contingency is low, and may be 

 represented as approaching zero, where a contingency of 

 represents complete independence. 



The values of the various quantities which may be 

 measured in milk, the percentages of the various substances 

 which occur in it, although they must a priori bear an 

 intimate relation to one another, being produced together 

 by living organism which we know by observations made 

 in other directions to have a wonderful power of co-ordin- 

 ating its activities, yet do not bear to one another a rela- 

 tion which we have so far been able to accurately and 

 briefly define. The amounts of the different substances 

 occurring in milk must therefore be regarded at present as 

 bearing only a low degree of contingency to one another, 

 and not very much information may be gained by the study 

 of the relation between series of observations of such 

 quantities, as will be seen later. 



If, however, each of these series be examined separately, 

 and the results be arranged according to their magnitude, 

 the series may be found to be capable of being divided into 

 two main classes: (1) those in which the numbers of results 

 are fairly evenly distributed over the whole range of the 

 values observed, or (2) those in which there is a certain value 

 in whose vicinity the results are much more numerous than 

 elsewhere, the numbers of results becoming smaller and 

 smaller as this value is receded from in either direction. 

 In a series of results of the first kind, any particular result 



