(304 TRANSACTIONS OF SECTION A; 



No slides are used, but the iustnunent is guided by paiailBl-u!otioii rudi?, 

 whicli rods can bo adjusted slightly out of parallelism, so that the iiistrumeut, 

 instead of sliding in a perfectly straight lino, describes an arc of a circle, of which 

 the object-glass used for forming the sun's image (22|-foot focus) is the centre, 

 the collimator always pointing to the centre of the object-glass. 



Almost the entire weight of the instrument is supported by four steel wires, 

 ■which hang from counterpoised levers pivoted on the roof of the building, so that 

 only a very small portion of the weight is allowed to bear on the planed surfaces. 



The motion is effected by a frictional clock acting on two screws, which drive 

 the instrument simultaneously at the forward and backward ends. 



5. On the Relation betioeeu Intensity of Light, Time of ExposurCi and 

 Photographic Action. By Professor H. H. Turner, D.Sc, F.R.S. 



1. If /be the intensity of a source of li'.'jt, and i the time of exposure, it was 

 not unnatural to assume that the photigraphic effect is proportional to the 

 product It. But experiments have shown that this is not the case. Sir William 

 Abney, who had previously expressed an opinion in favour of the law, announced 

 its 'complete breakdown' under certain conditions in 1893.' The conditions 

 specified by him were that a slow plate should be used, more particularly a 

 bromide ; and his Paper is written on tlie assumption that deviation from the law 

 will vary with the kind of plate used. 



2. The object of the present note is to suggest that the deviation is, at any 

 rate for stellar work, uniform ; that in fact a new law can be formulated, the 

 photographic effect being proportional to 



instead of the first power of t. This result was arrived at by the pfe.^ent writel" 

 from a discussion of numerous Greenwich plates in lOO.'i,- but as it was supposed 

 to apply merely to particular conditions, no great importance was attached to it. 

 On reviewing tlie work of the late Mr. E. L. J. Ellerj' it was noticed that he had 

 arrived at identical figures in 1891,^ under conditions which probably- differed in 

 several essentials from those at Greenwich {e.g., the manufiicture of plates Las 

 changed a good deal in the interval). It was then remembered that the work of 

 Dr. Schwarzschild, using a quite diii'erent method of measuring stellar magnitudes, 

 gave nearly identical results. Finally, on looking up Sir William Abney 's paper of 

 189.3 (loo. eit.), it was seen that the only numerical result given by him is closely 

 in accord with the suggested formula. 



3. As a purely empirical formula, therefore (for at present no physical basis 

 has been suggested, so far as I know), and chiefly with the view of eliciting 

 opinions and experiences which may throw further light on the matter, it is 

 suggested that 



CO 



photographic effect - intensity x (exposure)'; 

 or stellar magnitude = C — 2'5 log 1—2 -0 log t 



or, with an increase of exposure equivalent to five magnitudes, we only get four. 



G. Systematic Motion of the Starn. By Professor F. W. Dyson, F.R.S. 



Eighteen hundred stars from all parts of the sky whose proper motions were 

 greater than 20" a century were examined. It was found that 1,100 were moving 

 in directions within 60" of one apex, 600 within 60" of the second apex, and 100 

 Avere outside these limits. .The motions of the 100 stars were irregular and not 

 directed to one part of the sky. The nufnber of stars of large proper motion 



' Monthly Notices, liv. p. 65. - Ibid., Ixv, p. 7G4. 



^ Ibid.. Hi. p. 205. 



