402 PROFESSOR C. MICHIE SMITH ON OBSERVATIONS ON 



examined under the microscope by Dr T. K. Rogers, a skilled chemist, but no 

 trace of volcanic matter could be detected in it. I do not suppose that the 

 asserted discovery of volcanic ejecta in rain and snow in Europe will be allowed 

 to have much weight in the decision of the question, since, even if the volcanic 

 nature of the materials found is proved, it still remains an open question where 

 they came from. 



With regard to the water vapour theory, there is at least some definite 

 evidence upon which to found an argument. The spectroscopic evidence 

 detailed above shows that all the prominent lines in the spectrum which are 

 due to the presence of aqueous vapour were stronger than usual, and that there 

 was, in addition, a strong general absorption in the red. This general absorp- 

 tion might doubtless be produced by the presence of dust of a suitable fineness 

 suspended in the air, but that it may also be produced by water vapour, or at 

 least by clouds, is amply proved by observations which I have since made on 

 several occasions. In a recent thunderstorm, to which I have referred above, 

 and' which was predicted spectroscopically forty hours before it developed, I 

 found almost exactly the same spectrum as in the case of the green sun, but 

 the absorption in this case was produced by the passage of the sunlight through 

 a comparatively thin stratum of dense cloud, instead of through a fairly trans- 

 parent atmosphere. We know, too, that the absorption produced by aqueous 

 vapour in certain conditions is capable of producing the observed appearance. 

 On this point we have the observations made by Mr Lockyer, confirmed by 

 other observers, of the sun being seen green through the steam escaping from 

 the funnel of a steamboat, and his further observation of a green sun seen 

 through a mist on the Simplon. We have also an observation made by 

 Professor Piazzi Smyth (Edinburgh Astronomical Observations, vol. xiv. 

 Appendix, p. 29), represented in a plate of which the following description is 

 given : — " Two eye views of varieties of daylight, whereof the first represents a 

 remarkable case of the sun looking blue rather than yellow, red, or the colours 

 usually given to it by extra thickness of atmosphere ; and the second shows 

 a peculiarly green sunset sky— both of them proving to be forerunners of the 

 sirocco's stifling wind and its concluding warm rain." The spectrum of the 

 cloud light on this occasion shows a strong similarity to the spectrum observed 

 here, only the absorption in the red end is less, and in the blue end rather 

 more strongly marked. For my own part, I do not believe that a more or less 

 green sunset is so rare an occurrence as has been supposed, and since my 

 attention has been called to it, I have seen it several times and very markedly 

 on the 13th and 14th of May, when it was perhaps the earliest symptom of the 

 coming monsoon. 



The abundance of water vapour present is also shown by the very heavy 

 monsoon which followed the appearance under discussion, when in Madras 



