OF THE ATOMIC WEIGHT OF ALUMINUM. 
1035 
five intervals of ‘1 each between any integer and the '5 which divides it from the next 
higher (or lower) integer, the result is given by the expression 
5-^1 + 
l g , 
17 X4 - 
X4 2 + 
JJL 
I IS | 3 
X4 3 + 
18 
I 10 I 8 
X4 8 
and the probability in question is found to be only equal to 1 : 109 7'8. 
Of course this result might be easily varied by assuming other limits of precision as 
marking accordance with the law within the range of possible constant error, but this 
example seems to be based upon not unfairly assumed ground in this respect, and 
seems sufficiently to illustrate the point that not only is Promt’s law not as yet 
absolutely overturned, but that a heavy and apparently increasing weight of pro¬ 
bability in its favour, or in favour of some modification of it, exists and demands 
consideration. 
