570 
MR, J. NORMAN LOCKYER ON THE 
of the mirror, and seeing that it did not leave the cross wires; then the spectroscope 
was placed on a large packing case, loaded with gravel for rigidity, with its collimator 
in a North and South line directed to the centre of the mirror, and the prisms set to 
minimum deviation for F (H/3). 
Conditions of Eclipse. 
At the Brazilian station the diameters of the sun and moon were 31' 55"'4 and 
36' 6' /- 6 respectively; this difference of diameters gave a duration of 4 minutes 
43 seconds, hence the relative movement was at the rate of *253" per second. 
The first contact took place at an angle of 136° W. of the North point, and the 
last contact at an angle of 45° E. of the North point; therefore the direction of the 
moon’s motion cut the North and South line of the sun very nearly at an angle 
of 45°. 
The local mean times of commencement and ending of totality were 11 hours 
40 minutes 5L seconds and 11 hours 45 minutes 34 seconds respectively, giving a 
duration of 4 minutes 43 seconds, and the altitude of the sun was 76°. 
The Greenwich mean time of totality was 2 hours 17 minutes, thus preceding the 
total phase at the African station by 1 hour 20 minutes, and following that of the 
Chilian station by 1 hour 10 minutes. 
Dark Slides Employed. 
Three dark slides were provided ; they were made so as to travel up and down by 
means of a rack and pinion, in a frame fitting into the back of the camera. Each 
slide had eight compartments for plates (4^ X If”), and the plates were successively 
brought into position by raising the slide (fig. 6). 
Corresponding to each compartment there was a notch at the side, which indicated 
the exact place for the plate when a spring detent in the frame fell into the notch, 
the detent being so arranged as to allow the slide to move only in one direction. 
The number of each slide was cut in deep Roman numerals, so that it could be 
known by touch when it was in the velvet bag, without bringing it into the light, for 
it was necessary to know which one was being dealt with, as they contained different 
kinds of plates. 
Method of Focussing. 
The collimation for parallel rays had been effected before leaving England, but this 
was again verified, and trial photographs were then taken till an excellent focussed 
spectrum was obtained from D to beyond K in the ultra-violet. To obtain the same 
conditions as would hold during the eclipse, the collimator was then removed and an 
attempt made to photograph the spectrum of a star (Arcturus); as perhaps the best 
