O. N. Rood — Reflection of the Rontgen Rays. 179 



by diffused ordinary light, had a certain character of its own, 

 in the disposition for example of its horizontal bands, and a 

 study of the 90-hour negative convinced me that it also contained 

 decided evidence of the presence of similar diffused horizontal 

 bands, and that in point of fact upon the same plate were two 

 impressions superimposed, one due to regularly reflected 

 X-rays, the other to the same rays in a diffused condition. 



It now remained to attempt to make an estimate of the rela- 

 tive amounts of the regularly reflected and diffused X-rays in 

 the case of the 90-hour negative. This I undertook to accom- 

 plish by mixing together ordinary diffused with ordinary 

 specularly reflected light, till the mixture should furnish a 

 negative corresponding in appearance with that obtained with 

 the X-rays. It was first ascertained by an ordinary photo- 

 metric determination that the cylindrical mirror sent to the 

 plate-holder a certain amount of light, and that when this 

 mirror was covered up by unglazed white paper, the plate- 

 holder received only one-quarter of this amount, the original 

 source in both cases being the end of the Crookes tube. Con- 

 sequently, when the cylindrical mirror was covered up with a 

 sheet of this paper from which one-half of the surface had been 

 removed, chess-board fashion, the sensitive plate would be 

 acted on by four parts of regularly reflected ordinary light, 

 and by one part of light that was equally diffused in all direc- 

 tions, that is, it would receive 80 per cent, of regularly 

 reflected ordinary light, and 20 per cent, of diffused. A nega- 

 tive taken after this fashion, when compared with the 90-hour 

 X-ray negative, was found to have received too small an 

 amount of diffused light. The mirror was then covered by a 

 different sheet of the same paper, perforated after the same 

 geueral fashion, but furnishing to the sensitive plate a mixture 

 of 36 per cent, of regularly reflected and 64: per cent, of dif- 

 fused light. This second negative showed that too much 

 diffused light had been employed, and that the most appro- 

 priate mixture would not be far from equal parts of the two 

 kinds of light. Consequently it was concluded that, according 

 to this imitative experiment, the 90-hour negative had been 

 acted upon by approximately equal parts of regularly reflected 

 and diffused X-rays. The experiment was not pushed farther, 

 as a study of these negatives convinced me that 7io mixture of 

 regularly reflected and of equally diffused light would furnish 

 more than a coarse imitation of the X-ray negative, there being 

 details on the latter which no such mixture could reproduce. 

 The explanation of this lies, as I take it, in the fact that pol- 

 ished metallic surfaces do not furnish a mixture of regularly 

 reflected and of equally diffused X-rays ; they appear, accord- 

 ing to my observations, to furnish, at an angle of 45°, a certain 



