1832.] to Civil Servants in Bengal. 281 



necessary to reject the year of appointment entirely from the calcula- 

 tions, and to assume the first year for the men appointed in 1790, as 

 commencing from the 1st January, 1791, and so with all following years. 

 The compiler of the tables, in the article of the Gleanings before 

 referred to, has included this year in the computation of 852 as the num- 

 ber of appointments up to 1828, for he would otherwise have rejected 

 the nominations of the last year of the series. His first year being, as 

 is evident from this circumstance, that of nomination, he sets down the 

 deaths upon the complete number of persons nominated as being, in the 

 year of their appointment, only 2. The above table, however, shows 

 six deaths in the year of appointment, or five, if the years subsequent to 

 1828 be thrown out; and this, without making any allowance for the 

 eleven cases of appointed servants who never came to India, several of 

 whom may have been prevented by death from availing themselves of 

 their appointments. Whence the author of that article took his numbers, 

 it is beyond me to guess; but it will stand to reason, that two deaths upon 

 850 appointments made in the course of 40 years, mostly a period of 

 war, can be no near approximation to the risks of the voyage out> 

 and the numberless accidents to which headstrong youth is liable. 

 Six deaths, in the year of actual appointment, have been traced ; 

 but, as there is reason to doubt, whether this even can be the 

 whole number, while, as above remarked, the period is never a complete 

 year, it has been deemed better to take, as the first year for calculation 

 of the value of life, that dating from the 1st January after appointment. 

 The deaths of our next year are 17, which I confess my inability to 

 reconcile in any way with the number of the Gleanings, which is 9 

 only for the two years' risk of life upon 809 appointments. It is not my 

 business, however, to point out these discrepancies. The essential 

 thing is to make out a statement that shall be free from error, or as 

 nearly so as the case will admit ; and that given above is offered as the 

 best which our present materials afford the means of framing. In the table 

 which follows, No. Ill, the ratio of deaths to survivors, in each year 

 of residence, is attempted to be exhibited, the deaths being taken from 

 the table above, and the number of persons whose life was at risk be- 

 ing thus computed : the total appointments between 1790 and 1831, 

 were 951, of whom 11 died in England, or changed their mind, or from 

 other cause never came to India. There remain 940, from which 

 number must be deducted the ascertained deaths (5)*, and retirements 



* The deaths upon 940 in the year of appointment were 6 by our table, but one 

 of these belonged to the 29 nominations of 1831, all which being 1 rejected from the 

 calculation for the first year after appointment, and being deducted separately, 

 this death must not be subtracted again. The deaths upon 940 — 29 or 911 

 were only 5. 



