IN THE TROPICS : III 



41 



If we arrange the figures obtained in Table 33 in groups 

 under the various percentages we get the following series 

 of numbers : — 



PerCent. 43 44 45 46 47 48 49 50 

 No. 113 3 4 5 11-5 14 175 



51 52 53 54 55 56 



12-5 8-5 10 6-5 — 1 



These numbers are plotted in the accompanying diagram, 

 in which the different percentages are represented by equal 

 distances along the base line, whilst the vertical distances 

 represent the number of individuals which showed each per- 

 centage of yellow grains. : — 



Fig. 1. 



The above curve (or rather polygon) is somewhat irregular, 

 for the number of " variants " (individual samples) included 

 is small. Nevertheless it bears some resemblance to a normal 

 curve of prob ability. 



By Galton's method the probable error or " mid departure " 

 of this polygon (the average distance from the mean of the 



