2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Jl 



day. A test of the clock was made on the day preceding the eclipse,* 

 but owing to the cloudiness it was not possible to follow quite up to the 

 time of the eclipse. Such observations as were made on the pre- 

 ceding day, however, indicated that the clockwork moved a little too 

 slowly and so the rate of the clock was increased about 3 per cent with 

 the expectation of more exactly following the sun at the time of the 

 eclipse. Unfortunately this proved to be too much of a correction, 

 so that the clock moved a little too fast during the eclipse, and the 

 images of the moon are not as truly round as they should be. Prob- 

 ably appreciable injury of definition has resulted from the slight 

 drifting of the corona during the taking of the photograph. This 

 might have been corrected, as the result proved, by mounting a 

 following telescope in connection with the apparatus so that the 

 observer might be in a position to guide the cameras during the 

 eclipse. However, the moon's motion relative to the sun makes the 

 moon unfit to follow upon, and it was supposed that the rays of 

 the corona would be too indefinite so this was not thought worth while 

 to arrange for. As it proved, there was a splendid prominence 

 visible during the eclipse which would have been satisfactory to 

 follow upon and it is perhaps to be regretted that some arrangement 

 for following was not provided. 



The two camera telescopes were rigidly fastened together. 

 Exposures were made by the removal of two pasteboard boxes which 

 covered the ends of the tubes but were separately mounted on hinged 

 supports independent of the cameras. As the requisite time of 

 exposure was not accurately known, it was arranged to expose one 

 of the telescopes for i minute 30 seconds, the other for 2 minutes 

 45 seconds. 



The program was carried through without any accident, and upon 

 developing the twb negatives both were found to be very good, but 

 the exposure of i minute 30 seconds seemed to show quite as much 

 extension of the corona as that of 2 minutes 45 seconds. As less 

 drift of the clock occurred during the shorter interval than during 

 the longer we give in the accompanying illustration only the result 

 of the shorter exposure. 



It is much to be regretted that the full excellence of the photograph 

 cannot be produced in the illustration. There were a great number of 

 sharp relatively narrow coronal streamers extending nearly two 

 diameters in almost every direction from the sun. Decided evidences 

 occur of coronal streamers at the north and south poles similar to 

 those which are found at times of sun-spot minimum. The corona 

 on this occasion was an intermediate type between a sun-spot maxi- 



