lO Turner, Total Solai^ Eclipses. 



take proper photographs and study them at leisure than 

 to attempt visual observations in the limited time and 

 under the stress of excitement. 



I shall not attempt to give even a brief description 

 of all these observations ; but shall, with your permission, 

 select one special point for consideration from a line of 

 investigation to which my own work at recent eclipses 

 has been directed. My special object will be to shew 

 how a quantitative measure helps us in studying natural 

 phenomena. The point is not a new one — it is as old as 

 science itself; but every new illustration of it comes with 

 a certain freshness. A beautiful illustration was given 

 nearly 30 years ago by Professor Schuster in almost 

 exactly the same domain as that of which I ain about to 

 speak. He attacked * the problem of the distribution of 

 particles round the sun which might, either by scattering 

 the sun's light, or by themselves becoming incandescent, 

 give rise to an appearance such as the corona, and wrote 

 thus :— 



" Our problem is an inverse one, and seems at first 

 sight very hopeless. From the observed polarisation of 

 light we are to find out what part of it is due to scattering 

 particles, and, as will be seen, we cannot do this without 

 finding out at the same time in what way the scattering 

 particles are distributed round the Sun, and in what way 

 the light due to other causes varies with the distance 

 from the Sun. I began the calculation in the hope of 

 getting a rough idea only of the amount of polarisation 

 which we might expect. But it appeared that even such 

 observations as we can make during the short time avail- 

 able during a total solar eclipse may yield most important 

 inf(.M-mation as to the constitution of the solar corona. 

 I shall shew that combined measurements {a) of the 



* Moil. Not. Riy. Astroii. Soc, vol. 40, p. 35. 



