and the second (17 seconds) shows very little more than the fifth (4 seconds).
The form of the corona is clearly of the type associated with the minimum in the sunspot cycle.
The eastern equatorial region shows great disturbance, and there are many striking prominences on the eastern limb and several marked arches. In particular, reference may be made to (i) two dark rifts radiating from the neighbourhood of a prominence near the sun's equator, on the east side ; (ii) bright and dark arches in no less than three of the quadrants, extending about 4' from the limb ; (iii) a dark region surrounding a large prominence whose position-angle is approxi- mately 20 measured from the X. point of the sun ; and (iv) a remark- ably definite edge as of a chromospheric layer for a considerable arc of the sun's limb, in the N.W. quadrant, such as Sir Norman Lockyer described as seen in the photographs secured in Spain last year.
A summary is given below of the extension noted in the different photographs extension being measured beyond the moon's limb, and expressed in minutes of arc :
 | A table should appear at this position in the text. See Help:Table for formatting instructions. |
Plate No. 1. i 8 seconds.
No. 2.
17 seconds.
No. 5.
4 seconds.
No. 6.
2 seconds.
i
15'
21'
5'
East eq u atorial 8
23
19
g
West 5
22
12
9
South polar 3
10
8
^13. The Polariscopic Camera (Savart Plates and Nicol Prism}.
The Savart camera, which was constructed for use in Algiers, was to be used on the present occasion with a view to getting information as to the amount of polarisation in the outer regions of the corona, and, in particular, in the search for the position of maximum polarisation.
Professor J. J. Thomson very kindly put the large Nicol prisms from the Cavendish Laboratory collection at my disposal again, and the Savart plates were arranged in the same way as last year.*
The camera was put in the second compartment of the double tube, side by side with another camera which was provided with a similar lens, but had no polarising apparatus in front. In this way images of the same scale were obtained side by side on the one plate, one image being taken through the polariscopic camera, the other through the plain camera, each camera having an aperture approxi- mately F/27.
- ' Proc. Roy. Soc.,' vol. 67, p. 363 ; ' Monthly Notices, Roy. Astro. Soc.,' TO!. 60,
App, p. [27].
