270 Professor Ward [April 27, 



The red rays are made to fall across, say the left end of the slot, 

 starting about half an inch from its extremity, then come the orange- 

 yellow and green, then the blue and violet, the end of the visible 

 violet being about an inch from the right end of the slot. Beyond 

 the visible red, to the left, the invisible infra-red rays are falling 

 on the plate, and beyond the visible violet, to the right, fall the 

 invisible ultra-violet rays. 



Owing to the obstruction of the glass, however, very little of what 

 ultra-violet escapes absorption in the atmosphere reaches the agar- 

 plate itself; though, as we shall see, a good deal of the large propor- 

 tion of these rays which emanate from the electric arc are incident on 

 the plate provided no glass is employed at all. 



The plate which I now have thrown on the screen [Photographs of 

 solar spectra in bacteria] was thus exposed for five hours in August to 

 the spectrum obtained by decomposing a beam of solar light reflected 

 from the mirror of a heliostat, and you see that the result is a most 

 satisfactory proof (1) that the infra-red, red, orange and yellow rays 

 are totally without effect — as shown by the bacteria exposed to these 

 rays having germinated and developed as rapidly and as strongly as 

 those in the dark ; (2) that the bactericidal effect occurs in the blue- 

 violet region, diminishing in intensity at both ends as we pass into 

 the green (to the left), or into the ultra-violet (to the right) ; (3) that 

 the most destructive rays are the blue rays to the right and the violet 

 rays. 



These results are — if we allow for differences of absorption in the 

 screens, &c. — entirely in accordance with my screen-experiments, and 

 prove most conclusively that the rays which kill the bacteria are the 

 blue and violet ones. This explains why I find these organisms 

 destroyed so much more rapidly by the light of the summer sun than 

 in winter ; why a clear blue sky is so much more effective than a hazy 

 one ; why direct sunlight acts so much more quickly than reflected or 

 diffuse daylight; and many other experimental facts. It will be 

 noticed that no question of temperature comes in; the plates are 

 exposed on ice, and the hotter regions of the spectrum are totally 

 without effect — the bactericidal action is here emphatically due to 

 rays which affect our eyes as blue and violet. Now let us examine 

 the results obtained with the electric spectrum. 



The methods are exactly as before, excepting that the light, con- 

 centrated by quartz lenses, is passed through quartz prisms, (fee, and 

 is not allowed to pass through glass at all, unless, as in the case 1 will 

 select, a piece of thin glass is interposed for experimental purposes, 

 to determine the difference in effect between it and quartz. In this 

 case two slots were cut in the glass lid, one of which was covered 

 with thin glass, the other with quartz. Everything else was as 

 before. The plate was exposed for twelve hours to a very pure 

 spectrum, and it brings out very clearly the following facts. [Photo- 

 graphs cf electric spectra shown.] 



(1) That when there are plenty of ultra-violet rays, and no 



