ON THE MEASUREMENT OF THE LUNAR DISTURBANCE OF GRAVITY. 103 



out the mercury, and it was left burning whilst the observation was being 

 taken ; but now that this operation is done entirely from above, the 

 phenomenon has disappeared. 



In a paper entitled ' Recherches sur la Verticale ' (Ann. de la Soc. 

 Scient. de Bruxelles, 1881), M. d'Abbadie continues the account of his 

 observations with his instrument, called by him a Nadirane. It was 

 described in the last year's Report, and some further details have been given 

 above. A portion of this paper refers to his old observations, and gives 

 further important details as to the exact method of making observations, 

 and of various modifications which have been introduced. 



Each complete observation consists of the following processes : — > 

 measurement of the distance between the cross-wires and their image, 

 (1) in the meridian, (2) in the prime vertical, (3) in the N.W. azimuth, 

 (4) observation of barometer, (5) of thermometer, (6) of direction and 

 force of the wind, (7) condition and movement of the image estimated 

 with the micrometer, (8) condition of the heavens, (9) of the breakers 

 called 'les Criquets,' which can be observed from the neighbouring room. 



This last is to determine whether it is possible to have a rough sea 

 with a calm image ; a condition which has not hitherto been observed. 

 This statement seems somewhat contradictory of the following : — 



' Aucunes des variations dans les circonstances coucomitantes n'a para 

 se rattacher a I'etat de I'image qui, pendant des journees entieres, parait 

 tantot belle, tantot faible, et parfois meme disparait entierement, bien que 

 ce dernier inconvenient ait ete evite en grande partie par I'usage d'un 

 recipient en bois a fond raine pour contenir le mercure.' I presume we 

 are to understand that the roughness of the sea and the badness of the 

 image is the only congruence hitherto observed. 



M. d'Abbadie's observations on the efiFect of the tides will be referred 

 to in the Appendix to this Report. He then discusses the various causes 

 which may perhaps influence the vertical. 



The variations of air-temperature are insufficient, because the vertical 

 has been seen to vary 2"*4 in six hours. If the effects are to be attri- 

 buted to variations in the temperature of the rock, it would be necessary 

 to suppose that that temperature varies discontinuously, which it is 

 difficult to admit. 



If it be supposed that the changes take place in the instrument itself, 

 the like must be true of astronomical instruments. And there is no reason 

 to admit the reality of such strange variations. 



Another cause, more convenient because more vague, is variation of a 

 chemical or mechanical nature in the crust of the earth. But if this be 

 so, why does the vertical ever return to its primitive position ? Another 

 cause may be variation in the position of the earth's axis of rotation. 



The azimuthal variations in astronomical instruments, referred to 

 by M. d'Abbadie (see a paper by Mr. Henry, vol. 8, p. 134, ' Month. Not. 

 R.A.S. '), are difficult to explain without having recourse to such variation 

 in the axis of rotation. 



He also tells us that Ellis (vol. 29, 1861, p. 45, ' Mem. R.A.S. ') has dis- 

 cussed the Greenwich observations from 1851 to 1858. A comparison of 

 the results obtained from two neighbouring meridian instruments seemed 

 to show that the azimuthal variations are partly purely instrumental. 



M. d'Abbadie's paper contains diagrams illustrating the variations of 

 the vertical observed with the Nadirane during nearly two years. He 

 sums up the results as follows : — 



