PHILOSOPHICAL SOCIETY OF WASHINGTON. 49 



"two ordinates, and the intercepted axis of abscissae, showing that 

 by the methods of Gauss, in which the ordinates are taiien at 

 unequal but selected distances, greater accuracy could be ob- 

 tained with fewer ordinates, than where the ordinates are taken 

 at equal distances. In the case of a semicircle the degree of 

 approximation was twice as great with three ordinates specially 

 selected, as with five equidistant ordinates. The barometric 

 curve could be better determined from three observations made 

 at ^, ^, and | of the interval from sunrise to sunset, than from 

 £ve observations at equal intervals during the day. 



Mr. B, Alvobd made remarks on 



THE mortality AMONG ARMY OFFICERS 



during fifty years, from 1824 to 1873, as derived from Army 



Registers. 



(^This paper is published in the Proceedings of the American Association 

 for the Advancement of Science, at its session in August, 1874 ) 



He also described a proposed plan for government providing 

 for the life insurance of persons connected with the army. 



Remarks were made by Mr. E. B. Elliott on the desirable- 

 ness of extending such a plan to the civil service ; and by Mr. 

 Taylor on the policy of government providing for all kinds of 

 insurance. 



91st Meeting. June 12, 1815. 



The President in the Chair. 

 Twenty-eight members and visitors present. ^ 



Mr. J. R. Eastman gave an abstract of a paper on 



THE comparison OP RAIN-GAUGES AT DIFFERENT ELEVATIONS ; 



describing observations made at the Naval Observatory in 

 Washington. Rain-gauges less than 2 inches in diameter gave 

 unsatisfactory results. During a year those at an elevation of 

 34 feet from the ground gave 88 per cent, of the amount of rain 



