PHYSICS. 467 



from a vessel if the vessel is lower than the machine; i. e., if the air is 

 pumped from above. The author, by the advice of Mendeleeft", is con- 

 tinuing his researches upon heavy gases. {N'aUire, December, xxvii, 

 p. 183.) 



Amagat also has experimented upon the elasticity of rarefied gases 

 and comes to a different conclusion. He says: "All that can be said is 

 that under the feeblest pressures that can be produced (one millimeter 

 or less, though I have obtained 0.2 millimeter) there does not appear 

 to be produced any sudden change in the law of the compressibility of 

 gases; they follow still the law of Mariotte, within the limits of ex- 

 perimental errof." {Comptes Eendus, August, xcv, p. 281.) 



Delacy h^s invented a barometer which records automatically the 

 variations on an enlarged scale. The tube, which has a capacious res- 

 ervoir at the top, is fixed, and opens into a cistern nearly as long, made 

 of a somewhat wider tube. This cistern carries on one side an index, 

 and on the other a pencil working on a moving cylindrical surface. It 

 forms the upper part of a kind of areometer, the lower part being 

 closed and floating in mercury in a still wider tube connected with a 

 reservoir kept at constant level. The variation of pi'essure is marked 

 by the variation of the height of the mercury in the reservoir, and this 

 latter is to that of the total height in the barometric cistern (or to the 

 path of the float or of the pencil) in the ratio of the section of the cis- 

 tern to that of the reservoir (about a sixth in the author's instrument), 

 thus realizing an amplification. {Nature, January, xxv, 1882, p. 290.) 



Dafour and Amstein have devised a registering barometer, now in 

 use in the Lausanne Meteorological Observatory, depending on the dis- 

 placement of the center of gravity of a glass tube containing mercury. 

 The form of the tube may be described as that of an L leading down 

 to a U by a vertical portion, the lower end being open. The tube 

 swings in the ]dane of its angles on a horizontal axis placed above the 

 center of gravity ; with increased barometric pressure it inclines to the 

 right, with decreased pressure to the left, and these movements are re- 

 corded by means of a style attached to the U part and applied to a 

 moving strip of paper. By a simple contrivance the jjendulum of a 

 clock is made to impart a slight shock every second to the tube so as 

 to destroy any adherence of the mercury. [Nature, February, xxv, p. 

 374 ) 



Joly has described a barometer, the readings of which may be made 

 electrically at a distance from the instrument. It requires but two 

 wires and reads to one-fiftieth of an inch. {Nature, Aj^ril, xxv, p. 559.) 



Brown has given a restime of the progress of the barometer and the 

 various forms it has undergone, in which, however, the excellent form of 

 instrument designed some years ago by Daniel Draper, and used in the 

 Central Park- Meteorological Observatory, is not mentioned. {Nature, 

 July, XXVI, p. 282.) 



Cailletet has suggested a new pump for condensing gases, in which 



