144 



REPOET — 1886. 



tube, and a vertical image of tlie scale is thrown by a gilded surface so 

 ■ as to coincide with the axis of the tube, and the level of the top of the 

 mercury can thus be read off by a microscope without parallax. The 

 height can be easily estimated with suitable illumination to '01 millimetre. 

 The barometer is enclosed in a metal vessel filled with water and con- 

 taining a thermometer reading to Jq° C. A variation of '01 millimetre 

 would correspond to a change of less than half a second per day in a 

 pendulum. 



The empirical relation between temperature and level of the mercury 

 was determined by preliminary experiments at the College de France. 



In the same volume of ' Comptes Rendus,' p. 631, M. Mascart gives an 

 ■account of observations which he made with this instrument at Paris, 

 Hamburg, Copenhagen, Stockholm, Drontheim, and Tromso, on the 

 occasion of a journey to the north. The Copenhagen results could not 

 be utilised through an accident. 



The four northern stations gave results which compared with Paris 

 differed from the theoretical values by the following amounts : — 



-where '-^ expresses the error in g. 

 9 



dl expresses the error in the length of the seconds pendulum. 

 dn expresses the error in the number of seconds per day. 



He suggests that there is a local -variation at Drontheim, as the pen- 

 dulum gives a variation from theory in the same direction, though only 

 of half the amount. 



M. Mascart remarks that the only conclusion which he draws is that 

 the gravity barometer is easily transportable, and that its precision is 

 not inferior to that of the pendulum. It only requires observations of the 

 level of the mercury and of temperature, and the installation can easily 

 be made in the room of an hotel in less than an hour. 



M. Mascart has been kind enough to inform the Committee that, 

 though lie has published no farther account of his investigations, he is 

 still pursuing them. He has found a difficulty in transporting the in- 

 struments since constructed owing to the breaking of the tubes by the 

 impact of the mercury. He thinks, however, that this difficulty may be 

 overcome. 



He is now observing with a somewhat similar instrument whether any 

 changes of gravity in the same place can be detected, using a column of 

 mercury about 4 metres long, balanced by the pressure of nitrogen in the 

 cistern, the nitrogen being at about 5 atmospheres. The cistern is about 

 3 metres below the ground. So far he has only detected an annual 

 variation correlative with the continuous variation of temperature. He 

 intends to register the height of the mercury continuously by photo- 

 graphy. 



