mo Table of Mortality. [Sept. 



had entered on the 1 st December, there would have been the risk of 

 only one month in their case, and the number of casualties upon the 

 number admitted would have been one-twelfth only of the annual ratio. 

 To provide accurately for this I furnished the writer, employed in re- 

 casting the registers, with a table giving a decimal value for every day 

 of the year, and thence, according to the date of admission, I made 

 him enter the risk, as of the fraction for the period of the year remain- 

 ing to 31st December. Thus, in the re-cast of the registers, each 

 admission will be seen indicated by a fraction to three places of deci- 

 mals : and the number of risks is by addition of the whole brought to 

 the true annual sum for computation of the ratio of mortality from the 

 actual casualties. 



Thirdly. When a life lapsed, its risk was lost for the remainder of 

 the year. To provide for this, I made a reversed decimal table show- 

 ing the fraction of the year to the date of the casualty, and by entering 

 the lapsed life not as an entire year's risk, but according to the fraction 

 to the date of occurrence, effectually removed this source of error. 

 But those who follow this plan must be careful, when a life lapses in 

 the very year of admission, to take both fractions from the same table 

 for computation of the value of the risk : otherwise a child admitted on 

 the 2nd January and dying on the 30th December, would have the 

 same fraction to represent both dates, and would stand as 0, though 

 the risk of his life was an entire year, less only two days. The writer 

 employed in re-casting the Orphan School registers made this mistake 

 in the first instance, which is the reason of my noticing the point. 



Fourthly. Having thus settled the mode of entering admissions 

 and casualties, I caused books to be prepared for each year of life. 

 In that for age 0, I caused to be entered successively, all who were 

 admitted at an age less than one year, taking their names in succession 

 from the register of each year from 1798 to the present time. The 

 number of names thus entered in this book for age 0, is 5930, but 

 each being reduced to its fraction of the year of admission, and the 

 death cases being doubly reduced, the number of annual risks, for this 

 age is diminished to less than half, being 2646, which is what might 

 have been expected. The names of the whole being thus looked out 

 in the successive books, and entered in a fresh register for age 0, the 

 page was ruled for forty years of life from 0, and each name was 

 marked as a year of life in the columns following 0, as it was found in 

 the successive registers, until the date of decease, or of removal from 

 the institution. 



Fifthly. The book of those who entered at an age less than one 

 year being completed, and the individuals followed out, a similar book 



