206 Mr. William Hiujgins [Feb. 20, 



cover at one side is a circular bole/, 3;^ iucbes diameter, for tbe ligbt 

 to enter ; below is a similar bole over wbieb is fitted a framework to 

 receive tbe " backs " containing tbe pbotograpbic plates, and also to 

 receive a frame witb fine-ground glass for putting tbe apparatus into 

 position. Immediately below, towards tbe speculum, is fixed a 

 sbutter witb an opening of adjustable widtb, wbicb can be made to 

 pass across more or less rapidly by tbe use of indiarubber bands of 

 different degrees of strengtb. In front of tbe opening / is fixed a 

 tube c, six feet long, fitted witb diapbragms, to restrict as far as 

 possible tbe ligbt wbicb enters tbe telescoj)e to tbat wbicb comes from 

 tbe sun and tbe sky immediately around it. Tbe telescope-tube a, a, 

 is also fitted witb diapbragms, wbicb are not sbown in tbe diagram, 

 to keep from tbe plate all ligbt, except that coming directly from tbe 

 speculum. It is obvious tbat, wben tbe sun's ligbt entering tbe tube 

 at / falls upon tbe central part of tbe speculum, tbe image of tbe sun 

 will be formed in tbe middle of tbe second opening at d, about two 

 incbes from tbe position it would take if tbe tube w^ere directed 

 axially to tbe sun. Tbe exquisite definition of tbe pbotograpbic 

 images of tbe sun sbows, as was to be expected, tbat tbis small devia- 

 tion from tbe axial direction, two incbes in seven feet, does not affect 

 sensibly tbe performance of tbe mirror. Tbe wbole apparatus is 

 firmly straj^ped on to tbe refractor of tbe equatorial in my observa- 

 tory, and carried wdtb it by tbe clock motion. 



Tbe performance of tbe aj^paratus is very satisfactory. Tbe 

 pbotograpbs sbow tbe sun's image sbar^^ly defined ; even small spots 

 are seen. Wben tbe sky is free from clouds, but presents a wbity 

 appearance from tbe large amount of scattered ligbt, tbe sun's image 

 is well defined upon a uniform background of illuminated sky, with- 

 out any sudden increase of illumination immediately about it. It is 

 only wben tbe sky becomes clear and blue in colour tbat coronal 

 appearances present tbemselves witb more or less distinctness. 

 [Several negatives taken during tbe summer of 1883 were sbown on 

 the screen.] In our climate tbe increased illumination of tbe sky 

 where there is a background of coronal ligbt is too small to permit 

 the pbotograpbs which show tbis difference to be otherwise than very 

 faint. A small increase of exposure, or of develoj^ment, causes it to 

 be lost in the strong pbotograpbic action of tbe air-glare. For this 

 reason, the negatives should be examined under carefully arranged 

 illumination. They are not, therefore, well adapted for projection on 

 a screen. [A negative taken with a wbity sky, showed a well-defined 

 image of the sun, with a sensibly uniform surrounding of air-glare, 

 but without any indication of the corona. In the case of the other 

 negatives exhibited, which were taken on clearer days, an ai^i^earance, 

 very coronal in character, was to be seen about the sun.J 



On May 6, the corona was photographed during a total eclipse at 

 Caroline Island by Messrs. Lawrence and Woods. Tbis circum- 

 stance furnished a good opportunity of subjecting the new method to 

 a crucial test, namely, by making it possible to compare tbe photo- 



