1868.] 



Solar Eclipse 0/I868. 



113 



nebulae, made fresh exercise necessary. Among other things, I concluded 

 not to alter the pendulum, long ago adjusted for sidereal time. The difference 

 of rate being only 1 in 365 for mean time (and 1 in 388 for solar time at 

 that date), the telescope would only gain on the sun by less than one 

 second during the oh minutes of totality ; so that even supposing I 

 should wish to keep it directed on one and the same point the whole time, 

 the practical effect would only be that that point would move along the 

 slit by perhaps fa part of its visible length (estimating that length, or 

 the width of the field, at 5'). I mention this as the " Instructions " direct 

 the adjustment to apparent solar time. 



26. Disuse of the Barlow Lens accounted for. 

 In one other respect, too, I must plead guilty to a departure from 

 the letter of those instructions, which hardly perhaps needs justifica- 

 tion ; I allude to the disuse of the Barlow lens. My reason was prin- 

 cipally this, that its insertion keeps the observer some 6 inches further 

 from the body of the instrument, and, besides involving a complete dis- 

 turbance of equilibrium, puts him out of reach of the declination screw — 

 results which I could not but think had not been contemplated. I should 

 add that I was quite confident of the practicability of catching a promi- 

 nence, without having its image doubled in size, though I was by no 

 means so sure that I could spare any of the light, which would be reduced 

 one-fourth. 



27. Care in adjusting the Pointer during the approach of the Moon. 



During the advance of the moon, and up to the last available moment, I 

 paid particular attention to the collimation (I use the word in its true 

 sense of aim) of the needle-point, being perhaps unnecessarily anxious to 

 avoid my old difficulty of finding my object in the spectroscope. The 

 sharp cusps were well suited to this purpose, and the sun-spots were 

 good tests. I had been fortunate in getting the pointer very exact, and 

 was therefore not troubled with any collimation-error to allow for. 



28. Spectrum at the Moon's centre. 

 While thus employed I had occasion to remark that at the centre of the 

 moon, some nine or ten minutes before totality, the intensity of the solar 

 spectrum was much about the same as that of the full moon. 



29. Measurement of Solar Lines. 

 Lntensity of Spectrum of Limb at D. — The principal solar lines were 

 measured at intervals during the advancing eclipse. A few minutes before 

 totality, in going over these lines for the last time, the slit being as wide as 

 was allowable for full sunlight, i. e. very narrow, I recorded an increasing 

 brilliancy in the spectrum in the neighbourhood of D, so great in fact as to 

 prevent any measurement of that line till an opportune cloud moderated the 

 light. I am not prepared to offer any explanation of this. The clouds were 



