PROF. K. PEARSON ON THE MATHEMATICAL THEORY OF EVOLUTION. 107 
Table II.—Ordinates of Normal Curve. 
D = Deviation. S = Standard Deviation. 
F = Frequency. 
P = Maximum Frequency 
D/S. 
F/P. 
D/S. 
r/p. 
I 
0 
I 
1-6 
•2780 
o-i 
•9950 
1-7 
•2357 i 
0-2 
•9802 
1-8 
•1979 1 
0-3 
•9560 
1-9 
•1645 
0-4 
•9231 
2 
•1353 
0-5 
•8825 
2-2 
•0889 j 
0-6 
•8353 
2-4 
•0561 1 
0-7 
•7827 
2-6 
•0340 
0-8 
•7262 
2-8 
•0198 1 
0-9 
•6670 
3 
•oiii ! 
I 
•6065 
3'2 
•0060 ; 
I-I 
•5467 
3'4 
■0031 
1-2 
•4868 
3-6 
•0015 
1-3 
•4286 
3-8 
■0007 
1-4 
•3753 
4 
•0003 
lo 
•3246 
5 
•000,004 
[Note, added February 10, 1894.—(1.) The importance of breaking up asymmet¬ 
rical frequency-curves into normal components has been recognized for a long time 
by anthropologists and biologists. Attempts at a solution have been made by 
P. Livi, ‘Sulla statura degli Italiani,’ Firenze, 1883 (see also ‘ Archivio 
VAntrojpologia e VEtnologia,’ vol. 13, Firenze, 1883, and ‘ Annali di Statistica,’ 
vol. 8, 1883, pp. 119-56). Also by 0. Ammon in his recent work ‘Die natUrliche 
Auslese heim Menschen^ Jena, 1893. These attempts can hardly be looked upon 
as serious. Professor Lexis and Dr. Venn have pointed out that the curve of 
deaths for each year for 1000 persons born in the same year—the true mortality- 
curve—is also in all probability a compound curve. 
Since writing the above memoir I have succeeded in resolving this mortality-curve 
into components which are not, however, all of the normal type, but become, as we 
approach infantile mortality, of the skew form (see p. 74 above). 
0. Ammon, in the volume cited above, endeavours to demonstrate an evolution in 
the length-breadth index of the skull of South-Germans since primitive times. He 
does this by comparison of the index as obtained from measurements on skulls from 
the Bow-Graves and on modern skulls. He has not, however, noticed that the 
frequency-curve for Bow-Grave skulls is asymmetrical. I have succeeded in 
breaking it up into two components, one of which practically coincides in mean 
and standard-deviation with the frequency-curve for the skulls of modern South- 
p 2 
