12 Mr. W. Galbraith on the Experiments 



because heat escapes from glass and many other substances, 

 when smooth or polished, chiefly in straight lines perpendi- 

 cular to the surface. Now if we attend to the position of 

 Sir William Herschel's prism and thermometer *, this will 

 help to explain why the thermometer indicated heat, even 

 when none of the illuminating rays reached it at all ; as also 

 why the heating power of the red rays seemed so much to 

 surpass that of the other colours, &c.: because the more di- 

 rect!}' opposite the thermometer was to the flat side of the 

 prism, the more of its heat would it receive ; and in the course 

 of such an experiment, there can be little doubt that the prism 

 became considerably heated by absorbing a portion of the 

 solar rays. 



The circumstance of heat emerging from an even or po- 

 lished surface chiefly in perpendicular lines will help to ex- 

 plain various phaenomena. From it we learn that Mr. Leslie's 

 lens, however hot, could have little or no tendency to melt 

 the wax; because the heat would emanate from its convex 

 surface chiefly in a set of diverging straight lines, such as 

 might be drawn from the centre of the sphere of which the 

 lens formed a part. 



For aught that is known to the contrary, the heating powers 

 of all the rays in the spectrum may be equal ; because the 

 seeming difference may be owing to a difference of density, 

 some of them being probably much dilated or attenuated by 

 refraction. I am, gentlemen, yours &c. 



Dec. 21, 1824. Henuy Meikle. 



in. On the Experiments of the Pendulum made hy Captain 

 Kater, M. Biot, 8fc. By Wm. Galbraith, Esq. A.M. 



To the Editors of the Philosophical Magazine and Journal. 

 Gentlemen, 



IN my paper published in your Magazine for September 

 1824', I gave, page 163, a general formula (7) to determine 

 the ellipticity of the meridian from observations on the length 

 of the pendulum made in different latitudes. In this formula, 

 however, it is necessary to substitute quantities which, with 

 the exception perhaps of the time of rotation, are not yet 

 sufficiently well determined. The accuracy of these is con- 

 stantly improving, and in a short time we expect to have the 

 length of the pendulum oscillating seconds on the equator as- 

 certained by different observers in several points of that circle 



* Diagrams or plates illustrative of this may be seen in so man}' books, 

 that it wouM be superfluous to infoducc them here. 



with 



