26 Mr. William Sutherland on the 



3t* and 4", and tabulate the results as F r /2 + F„ F r /3 + F„ 

 and F r /4-f F s , according to the type of compound. 



Table XVI.— (F r /2 + F 5 ). 



Be. Mg. Ca. Sr. Ba. 



F 2 ... 3-9 3-9 4-6 51 



Cl 2 42 4-8 53 58 63 



Br 2 5-2 5-8 59 60 71 



I 2 57 6-4 67 64 74 



There are some irregularities among these numbers, the 

 most pronounced being where the value for Srl 2 falls below 

 that for Cal 2 ; but ignoring these as due to no fundamental 

 error, we get, taking the data for the chlorides as the best, 

 with 2'1 as the value for CI, the following values for F r /2 or 

 F/2, and approximate values of B derived from the best 

 values of M/p in Table XIV. 



Table XVII. 



Be. Mg. Ca. Sr. Ba. 



F/2 =2-1 27 3 2 37 4 2 



B =1-0 56 86 10 6 16-6 



Table XVI. {continued).— F r /3 + F.. 



B. Al. P. As. Sb. Bi. 



Cl 3 27 ... 2-9 30 3-2 37 



Br 3 31 4-0 34 34 3 7 4'1 



I 3 , 52 ... 4-5 4-5 



Table XVI. (continued).— F r /4 + F,. 



C. Si. Ti. Sn. 



Cl 4 2-6 27 30 30 



Br 4 31 3-2 38 35 



I 4 40 ... ... 41 



The values just given for the types RC1 3 and RCI 4 are the 

 means of those given by both the melting-point and boiling- 

 point methods, the latter being reduced first by dividing by 

 1*08 the mean ratio of results by the two methods, and then 

 converted to equivalent values by dividing by 3' and 4*. 



Again, amongst the types RC1 3 and RCI 4 , taking the values 

 of F r /3 + F, and F r /4 + F, for the chlorides as the most 

 reliable, but allowing for the others where necessary, and 



