282 SUMMARY OF MEASUREMENTS. Chai\ VII. 



plants. This having been done in each case in which 

 more than one generation was raised, it is easy to 

 calculate the average of the average heights of the 

 crossed and self-fertilised plants of all the species 

 included in Table A. It should however be observed 

 that as only a few plants of some sjDecies, whilst a 

 considerable number of others, were measured, the 

 value of the mean or average heights of the several 

 species is very different. Subject to this source of 

 error, it may be worth while to give the mean of the 

 mean heights of the fifty-four species in Table A ; and 

 the result is, calling the mean of the mean heights 

 of the crossed plants 100, that of the self-fertilised 

 plants is 87. But it is a better plan to divide the 

 fifty-four species into three groups, as was done with 

 the previously given eighty-three cases. The first 

 group consists of species of which the mean heights 

 of the self-fertilised plants are within five per cent, 

 of 100 ; so that the crossed and self-fertilised plants 

 are approximately equal ; and of .such species there 

 are twelve about which nothing need be said, the 

 mean of the mean heights of the self-fertilised being 

 of course very nearly 100, or exactly 99*58. The 

 second group consists of the species, thirty-seven in 

 number, of which the mean heights of the crossed 

 plants exceed that of the self-fertilised plants by 

 more than five per cent. ; and the mean of their 

 mean heights is to that of the self-fertilised plants 

 as 100 to 78. The third group consists of the species, 

 only five in number, of which the mean heights of 

 the self-fertilised plants exceed that of the crossed by 

 more than five per cent. ; and here the mean of the 

 mean heights of the crossed plants is to that of the 

 self-fertilised as 100 to 109. Therefore if we exclude 

 the species which are approximately equal, there are 



