Page:Proceedings of the Royal Society of London Vol 69.djvu/246

230 Result. In the event I used A about 40 seconds after the beginning of totality, and found no trace of polarisation in any position of the instrument. To make sure that nothing was wrong I used also polari- scope B, and corroborated the first observation.

Towards the end of the third minute 2 m 50 s from the beginning of totality I again observed with B. At the end of the fifth minute I used C ; and in neither case could I detect a trace of atmospheric polarisation. About 3.50 P.M., after the eclipse was over, I used the polariscopes again, and found them all as adjusted in the morning before the eclipse ; each one showed showed strong bands on the patches of blue sky visible in the field of view.

It was intended to attempt to get

(i) Photographs of the ultra-violet spectrum of the flash at the

beginning and end of totality, (ii) A photograph of the ultra-violet spectrum of the corona.

The instrument was fixed on the top of the double tube, arid was supplied by light from the ccelostat. The arrangement was such that the slit could be set, without readjustment of the image, to catch the crescents both at beginning and end of totality, and the corona during totality. No other medium but quartz was used in the lenses and prisms, and the dimensions were such that a 2-inch beam was trans- mitted. The image lens had a focal length of 20 inches, and the collimator and camera lenses were each of focal length 18 inches; the prism was compounded of two 30 prisms, one of right-handed and the other of left-handed crystal ; all the optical parts were constructed by Hilger. Unfortunately it was found impossible to get the adjustments completed before the eclipse.

On the morning of the eclipse a final attempt was made, and it was thought well to make the exposures according to programme. These were all duly made, but the adjustment was so imperfect that the photographs are useless, which is all the more regretable as the plates show strong spectra.

It was intended to attempt to get

(i) Photographs of the " flash" spectrum with high dispersion in a limited range of spectrum, to assist in the discussion of the origin of lines, whose wave-lengths it was hoped would be determined accurately by means of the 4-prism spectroscope.

(ii) A photograph of the blue coronal ring near A 4231, for com- parison with the visual observations of the green coronal ring.