108 Mr. J. N. Lockyer. Use of Reflection Grating. [Feb. 14, 



the telescope that the image of the ring reflected by the silver surface 

 adjacent to the grating was thrown on to the screen. This done, the 

 grating was placed at right angles to the axis, and the spectrum of the 

 circular slit, illuminated by sodium vapour and carbon vapour, photo- 

 graphed for the first, second, and third orders on one side. The third 

 order spectrum, submitted to the Society with this communication, 

 showing the exquisite rings due to the carbon vapour flutings, 

 was produced in 42 seconds. The first order spectrum, also sub- 

 mitted to the Society, was produced in the same period of time, and 

 was very much over-exposed. Tt is, therefore, I think not expect- 

 ing too much that we should be able to take a photograph of the 

 eclipse, in the third order, in two minutes ; but let us make it four. 

 Similarly, we may hope for a photograph of the second order in two 

 minutes, and it is, I think, highly probable also that a photograph of 

 the first order may be obtained in one minute. 



Tt is clear then that, by mounting photograhic plates on both sides 

 of the axis, one solidly mounted equatorial of short focal length may 

 enable us to obtain a large number of photographs, with varying 

 lengths of exposure, of the next eclipse. I have insisted upon the 

 solidity of the mounting because, if any one plate is to be exposed 

 during the whole of totality, the instrument must not be violently dis- 

 turbed or shaken while the eclipse is going on. I think, however, it is 

 quite possible to obtain several photographs, of the lower order spectra, 

 without any such disturbance. The same plate may be made to record 

 three, or even four, exposures in the case of the first order in an 

 eclipse of four minutes' duration, by merely raising or lowering it 

 after a given time, by means of a rapid screw or other equivalent 

 contrivance, so that a fresh portion of the same plate may be exposed. 

 Similarly, the plates on which the spectra of the second order are to 

 be recorded may be made to perform double duty. 



If one equatorial thus mounted were to be devoted to each quadrant 

 of the coronal atmosphere, it is certain, I think, that most important 

 results would be obtained. 



