1870.] 



Atmospheric Lines of the Solar Spectrum. 



7 



class, In the absence of a micrometer for measuring small spaces, I have 

 assumed that in widening out at sunset A does so equally on both sides ; 

 on the contrary, B, I believe, maintains its left (or violet) edge constant, 

 and changes towards its right (or red) edge. The line 812 is the most 

 notably variable I have observed. Unlike other lines, it commences to 

 change quite two to three hours before sunset. The darkening is gradual 

 until near sunset, when it suddenly becomes black, and the band to the 

 right as suddenly appears. The variable lines are, however, so numerous, 

 that further notice of them might prove tedious reading. I therefore con- 

 tent myself by calling attention to the fact that, so far as I have the means 

 of judging, the variable lines at sunset and sun high superpose one another, 

 — more accurately speaking, on the former occasion, when they are gene- 

 rally wider, they include the position occupied sun high. I need hardly 

 dwell on the circumstance that the loss of light which no doubt occurs 

 near sunset cannot be the cause of the widening out in these lines ; for 

 constant lines, identified with KirchhofF's, stand mapped in the vicinity of 

 variable lines of greater and lesser refrangibility. 



22. Concluding remarks. — At the time of writing the air is so laden 

 with dust that you may gaze on the setting sun with impunity ; and spec- 

 troscopic observation, even at several degrees of altitude, is impossible. 

 But when the ensuing rainy season is approaching its termination, and the 

 exquisitely clear state of atmosphere once more returns, I hope to make 

 good progress with the map ; and also to try and ascertain*, by comparison 

 of the map already made with the actual objects, if the variable and atmo- 

 spheric lines and bands change their appearance from time to time. A 

 variation of the latter kind might open out a vast field of inquiry. 



23. Proposed alteration of spectroscope. — Though quite willing mean- 

 while to continue the work with the spectroscope as it now stands, I should 

 consider it a great boon if you would allow me to send it to England for 

 certain improvements. These I briefly indicate as follows: — 1. One or 

 even two additional prisms might be introduced into the drum, so as to 

 increase the dispersion. 2. The collimator should be made up into one 

 piece with the drum, and this may be done simply by uniting the present 

 three pieces firmly together. 3. A micrometer introduced at the eye-end 

 of the telescope would be invaluable. The stand answers every purpose, 

 so that it might be useless sending it home, unless any delay is likely to 

 occur in improvising a stand to adjust and test the instrument. Should I 

 receive a favourable reply by the return mail, the spectroscope ought to 

 reach Messrs. Troughton and Simms by the end of July ; and if they could 

 undertake to do the needful alterations by the end of August, it would, if 

 despatched immediately overland, reach me about the first week in October, 

 just in time for the fine weather. Unless, however, these dates are not ex- 

 ceeded, I should lose the observing season, and this would be lamentable. 



24. The same. — Even should the foregoing meet with approval, I am 



* Keport of the Committee, last paragraph of § 4. 



