45G 
PROFESSOE K. PEAESOX OX .MATHEMATICAL 
Fills curve was drawn on a large scale and its areas read off with an integrator. 
The following theoretical frequencies were obtained : 
No. of lips. 
Observation. 
Calculation. 
6 and over 
50 
47 
5 
798 
762 
4 
123 
160-5 
3 
18 
20 
2 
5 
5 
1 
•> 
1-5 
Tliere would not lie any seiious divergence liere, were it not for the group with 
four li])s, which observation shows to he inucli under-represented. But it must he 
remenihei-ed that we have only seven groiqis, and that such a number is very 
insufficient for a good determination of the moments of a curve. Further, the 
variation is not really continuous, as indicated by the curve, but discrete. We have 
at jiresent no clear statement as to how tlie moments of a discrete system of variation 
should he modified or corrected so as to give the liest results for the moments of the 
continuous curve which is to theoretically represent the series. I am doubtful 
wliether SHErrAKTi’s corrections—tlie best for continnons variation—are equally 
aiiproprlate in this case. Above I have used merely the rougli moments, bnt I 
