388 Prof. Draper on the Ckemicol Action of Light. 



was no difficulty in obtaining an impression ; a proof not only 

 of the great sensitiveness of that surface, but also of the extreme 

 brilHancy of the light ; which lasting, it is said, scarcely more 

 than the millionth of a second, can impress so clear and perma- 

 nent a trace of its passage. I may obsei-ve, though it scarcely 

 belongs to the subject here under investigation, that the stain is 

 of unifomi whiteness, and the spark therefore throughout its 

 whole path of uniform brilliancy. 



But to describe the singular fact here alluded to. Let there 

 be placed over a sensitive Daguerreotype plate two metal balls 

 connected respectively with the inside and oiitside of a Leyden 

 phial, in such a way that the discharge may pass from one of 

 the balls at about half an inch distance to the sensitive plate, 

 and from the sensitive plate to the other ball at about the same 

 distance. It is to be understood that the balls overhang the 

 sensitive surface, and are about two inches apart. One spark is 

 quite sufficient to produce the cftect. The experiment should be 

 made in a dark room. 



The plate haviug been mercurialized in the usual way, then 

 presents the following appearances. On that part of it which 

 received the spark there is a solarized blue-white spot, about 

 7^^th of an inch in (hameter. Immediately round this is an 

 annular space about y^th of an inch in diameter, which is per- 

 fectly black, the rays from the spark liaA-ing there had no action ; 

 then follows a white ring about ith of an inch wide, and then 

 another black. Finally succeeds a whitish stain of an indistinct 

 circular form, which can be tracetl by inclining thepiate as ha\-ing 

 a diameter of about li inch. ' i Ii-ovk- 



On that part of the plate from which the spark escaped to the 

 other ball the same phfenomena are repeated ; a central bluish- 

 white dot, suiTOunded by alternate black and white rings ; the 

 last white ring degenerating into a hazy stain, which, when the 

 balls have not been more than two inches apart, encounters the 

 other stain, and becoming confounded with it, forms a mark 

 somewhat resembling the figure 8. 



That the Leyden spark is of sufficient brilliancy to decompose 

 the sensitive surface has already been proved ; the mere staining 

 the plate in the instance before us is therefore nothing extraor- 

 dinar)'. But how shall we account for the production of these 

 alternate white and black spaces, rings of action and of inaction ? 

 Some may at first be led to suppose that this is only an interest- 

 ing form of Priestley's experiment of " the fair}' rings," formed 

 by receiring the shock of a batteiy on a polished steel surface, 

 when, by the oxidation that ensues, a film is formed of variable 

 thickness, and giving therefore the coloiu's of thin plates. But 

 a little consideration mil show that it is impossible that this 



