287 
the decrement of human life. 
earlier portions of the register would require to be augmented, 
not only on account of the smaller number of persons who 
have actually contributed to furnish it, but also on account of 
the greater proportion that these persons bear to the corres- 
ponding number at birth, when compared with the survivors 
at more advanced ages, who represent a population still more 
exceeding their own numbers. On the other hand, since 
the register in question relates only to a limited number of 
years, immediately following a very rapid increase of the 
Society, it is evident that the deaths must have occurred at 
earlier ages than if it had been continued until all the lives 
had dropped. 
Of these three modifications, it may be sufficiently accurate 
for the present purpose to omit the two latter as nearly 
counterbalancing each other, and to augment the earlier 
numbers in the proportion only of the members of the Society 
to whom they must necessarily have belonged, supposing 
that the admissions had taken place about the same ages at 
all periods ; assuming also the number of survivors at 45 to 
be in the same proportion to the births as in the Carlisle 
table. We may then proceed to take a mean between the 
mortality thus obtained, with proper interpolations, and the 
observations at Carlisle, as the second of the three principal 
bases to be afterwards incorporated with the mortality of 
Northampton and of London. Further than this, it is im- 
possible to place any great reliance on Mr. Morgan's docu- 
ment, which makes the annual deaths, in “a population 
exceeding 150000,” not quite 1 in 1500. 
Of the mortality of London, taken for the ten years from 
1811 to 1820, it may be observed, that its results bear the 
