Chase.J -^i^ [April 21, 



and tetrads, and gives 40911 : 1 for tlie combined ratio of probabilities. 

 The otlier sub-group is liexad, giving the aggregate ratio 11611 : 1. The 

 total aggregate ratio of the artiad elements is more than 81585(10)'" : 1. 



is, therefore, comparatively slight, and far less satisfactory than in the 

 grouping according to valency. 



174. General Test of Atomic P/iyllotaxy. 



Computors who are accustomed to calculations of probable error, and 

 who have not given any special attention to the liarmonic influences of 

 setliereal vibrations, may, perhaps, question the propriety of making any 

 allowance for an a jyriori probability of division in extreme and mean ratio. 

 For the satisfoction of all doubts upon this point it may be well to apply 



