NATURAL SCIENCES OP PHILADELPHIA. 163 



any other similar change by any change in the atmosphere. If such an occur- 

 rence were possible, I believe I would have observed it during the past two 

 or three years. The effect of moonlight in obscuring the colors of the stars, 

 and giving them a yellowish shade, can hardly be called an atmospheric ac- 

 tion. Neither can the effect of the rays of the sun in the earliest daybreak of 

 the morning or in the latest twilight of the evening, be called an atmospheric 

 operation. Such an effect tends to impart a general whiteness to the stars, 

 obliterating their colors in part or in whole, the same as in the end it oblite- 

 rates all their light. 



The question now arises, How is it that the atmosphere, when hazy and 

 imperfectly transparent, has the power of depriving the stars of their colors, 

 whatever their colors may be, and reducing them all alike to a dull whiteness ? 

 The reason may be seen in the simple fact of the obstruction of their light. 

 Their light becomes diminished in amount to such a degree that it no longer 

 has the power to produce the sensation of color on the retina. Nearly all the 

 stars, when viewed through a telescope, are colored ; they are of some hues 

 other than white. Of this I adduced evidences in my communication for 

 these Proceedings in June, 1863. They appear colored through the telescope 

 because their light is collected by the instrument in a comparatively large 

 mass ; so large that it can make their colors readily perceived. Take away 

 the instrument from all except the larger stars, and the pencil of light becomes 

 so small as to be without the power of imparting the sensation of color. In 

 the same manner the pencil of light from the larger stars may be reduced by 

 haziness in the atmosphere to so small an amount as to be incapable of im- 

 parting the sensation of color, except a dull whiteness, whatever their real 

 colors may be. 



But how does it happen that a green star is changed by haziness to blue ? 

 I once thought that possibly this effect might be due to the same cause which 

 makes the deep ocean, the distant mountainns, and even the atmosphere, ap- 

 pear blue. After further observation and reflection I cannot adopt that ex- 

 planation ; for then all the stars, like the distant mountains, would be colored 

 blue. Then there would be no such contrasts of all colors among the stars 

 as we now behold. The true explanation seems to be that the mists of the 

 atmosphere, in acting on the light of a green star, first obstructs the yellow 

 rays, and after these are all absorbed then the blue rays alone will be visible, 

 and the star must appear blue. Ultimately the mist may become so impervi- 

 ous that the attenuated ray of light can no more excite the sensation of color 

 and the star must appear dimly white. 



Before it can be admitted as a scientific truth that the atmosphere of our 

 earth has the power of changing the color of a fixed star from one hue of the 

 rainbow to another totally different, there must be brought forward a number 

 of well authenticated facts as grounds for such a belief. We must have the 

 specifications of certain stars which have been seen to change, and the dates 

 of such changes, and the conditions of the atmosphere by which such changes 

 have been produced, and also the numbers and the names of the persons by 

 whom such phenomena have been witnessed. Such evidences of the changes 

 of the colors of the fixed stars by our atmosphere have never been seen nor 

 heard, and for my part, judging by my own observations, I never expect to 

 see them, nor to hear of them. An exception to this remark may be the case of 

 a green star turning to blue, as already explained. Perhaps another excep- 

 tion may yet be found, as indicated in the following passage from Humboldt. 

 The italics are not in the original : " We do not here allude to the change 

 of color which accompanies scintillation, even in the whitest stars, and still 

 less to the transient and generally red color exhibited by stellar light near the horizon, 

 a phenomenon owing to the character of the atmospheric medium through 

 which we see it, but to the white or colored stellar light radiated by each 

 cosmical body, in consequence of its peculiar luminous process, and the dif- 



1864.] 



