536 



Notices respecting New Books. 



persons familiar with this subject ; availing himself of certain data 

 collected by competent persons, with great care and precision, from 

 the records of several of the principal assurance companies, he has 

 endeavoured to show the effect of selection amongst assured lives, 

 that is, to what extent the rate of mortality is diminished during a 

 given year amongst persons of all ages pronounced to be eligible for 

 assurance at the commencement of that year. A table formed from 

 such observations might, as Mr. Farren says, " be expected to display 

 the extreme value of selection, and thus afford a minimum rate of 

 mortality;" for it n.ust be remembered, that, in the ordinarj' tables, 

 the numbers exposed to the chance of death at every age are made 

 up of persons selected at that and almost all previous ages. Com- 

 pared with such tables, and with one exhibiting the rate of mortality 

 prevailing amongst the male population of England and Wales, the 

 results obtained by Mr. Farren are very curious. We give the fol- 

 lowing brief abstract to enable the reader to judge of them : — 



It will here be seen, that selection operates at all ages with con- 

 siderable force in diminution of the ordinary mortality, and that, as 

 might be expected, the force has a tendencj' to increase as the age 

 increases. It is probable that the general effect would be still greater, 

 but that it is counteracted b^^ the efforts continually made to effect 

 assurances on not very good lives, " the more precarious the life the 

 greater the inducement, as Mr. Milne observes, for persons interested 

 in its continuance to get it insured." One remarkable feature in 

 the tables is the comparatively low rate of mortality amongst the 

 general population in the last decade but one, that obtaining amongst 

 assured lives being actually higher. Mr. Farren accounts for this, 

 partly by a peculiarity in the construction of Mr. Farr's table (the 

 mortality amongst the male population), and partly by the supposi- 

 tion, " that persons seeking life assurance at the ages in question 

 form a better criterion of their own health than at other periods of 

 life, and thus render the task of selection by parties having an oppo- 

 site interest more than usually difhcult." It will probably be con- 

 ceded, that about the period of life referred to, intimations of latent 

 diseases are of more general occurrence than at any earlier one, and 

 that in later life their development makes them apparent ; but these 

 considerations scarcely seem to suffice for explanation of the anomaly. 



We have not space to follow Mr. Farren through many other in- 

 genious deductions, hut we must not pass by without comment the 



