TRANSACTIONS OF SECTION B. 499 



5. The Aerorthometer, an Instrumerit for correcting the Measure of a Gas.^ 

 B]i A. Veknon Harcoukt, M.A., F.B.S. 



In determining the mass of a gas, it is generally more convenient to measure than 

 to weigh. The volume multiplied by the density, which is known, gives the mass, 

 if the volume has been measured under standard conditions. Otherwise, the actual 



measurement must be corrected. The usual formula for correction, F= -^ — ~^ ■ 



760( l + a t) 



requires the observation of (1) v the actual volume, (2) P the height of the baro- 

 meter, (3) t the temperature, (4j p the tension of aqueous vapour at t given by a 

 table. 



A reading of the aerorthometer gives at once the quantity * — — ! ; and 



thus, to find V, the normal volume, it is only necessary to divide v by this reading. 



The instrument consists of two narrow vertical tubes, the one open above, the 

 other terminating in a a bulb whose capacity, including that of the stem down to 

 the first graduation, is 1,000 of the units into which the stem is divided ; both tubes 

 are connected below with a reservoir from which mercury can be driven up the 

 tubes by the pressure of a screw. When the mercury stands at the same level in 

 the open and closed tubes, the air in the closed tube, which at 0°0. and 7G0 mm. 

 occupies J ,000 volumes, is under the existing atmospheric pressure. It has also the 

 temperature of the surrounding air. It is thus under the same conditions as the 

 gas in any vessel near it. The volume read upon the aerorthometer is to l.COO as 

 the observed volume of the gas in the measuring vessel is to its normal or corrected 

 volume. 



For use in the ordinary case of measuring a gas over or in presence of water the 

 aerorthometer is charged with a minute drop of water. If filled with dry air its 



reading gives L_tiL_'. For technical purposes the graduation ' 1,000 ' marks 



the volume which the inclosed air occupies at 30 inches Bar. and 60° Fahr. 



6. A Revision of the Atomic Weight of Rubidiwm. 

 Bij Charles T. Heycock, B.A. 



The element rubidium has an atomic weight assigned to it usually of 85-4, and 

 consequently is removed from a whole number. Its revision therefore became of 

 interest, to see whether with improved methods it could be brought in accordance 

 with the hypothesis of Prout. 



The experiments here detailed are merely preliminary to a large number shortly 

 to be undertaken. The crude rubidium chloride was obtained from Dr. Theodore 

 Schuchardt of Goerlitz ; it was separated by Godeffroy's method with the alums. In 

 order to further purify it, the author converted the crude chloride into sulphate, 

 and decomposed this with an equivalent quantity of barium bitartrate. After 

 filtration, the mixed bitartrates of potassium, rubidium, and caesium were repeatedly 

 crystallised, the crops being separated. The bitartrate of rubidium separates from 

 its concentrated solution with ease, provided it is constantly stirred. 



The pure rubidium bitartrate was then decomposed by heating in a crucible of 

 platinum, and the resulting carbonate dissolved out and filtered from the carbon, 

 was divided into two portions, (a) and O). 



(a) was neutralised with excess of pure HCl. evaporated to dryness and fused. 

 The titration was made exactly as detailed by Stas, in his celebrated work, {Mem. 

 de VAcdd. de. Belgique, torn. xxxv. 1865), i.e. by adding to a known weight of silver 

 nitrate an equivalent quantity of rubidium chloride, calculated on the assumption 

 that Rb = 8iV0, in other words conforms with the hypothesis of Prout. 



The author then found in a beam of yellow light the amount of RbCl in excess 



' A fuller description of the aerorthometer, with a drawing, will be found Proc. 

 Roy. Soc. vol. xxxiv. p. 166. 



K K 2 



