186 



THE AMERICAN MONTHLY 



[October, 



known that the color of a transparent 

 object is due to the absorption of cer- 

 tain colors from the light — thus, a red 

 glass permits the red rays of the 

 spectrum to pass through it while it 

 is opaque to all the others. Hence, 

 if we examine with the spectroscope 

 the light from such a glass we find 

 that the spectrum consists only of a 

 band of red, all the other colors being 

 absent. Such a spectrum might be 

 given by a large number of totally 

 different substances of a red color, 

 since there is no distinguishing char- 

 acteristic of a spectrum formed by a 

 general absorption of this kind. But 

 there are many colored solutions and 

 compounds that exercise a selective 

 absorption upon the light that enters 

 them, that is to say, when the trans- 

 mitted light is analyzed by the prism 

 it shows, not a single band ot color as 

 in the former case, but a continuous 

 spectrum with one or more dark 

 bands crossing it, indicating what col- 

 ors have been absorbed. The posi- 

 tion of these bands is constant for the 

 same compound ; hence, we have a 

 means of detecting certain com- 

 pounds, and analyzing mixtures of 

 them, by the character of the light 

 they transmit. Take, as an example, 

 the coloring matter of blood, the 

 spectrum of which will be shown this 

 evening. There are two distinct, dark 

 bands in this spectrum which always 

 occupy the same position. As an 

 illustration of the ease with which 

 two coloring matters may be distin- 

 guished in a solution, we may allude 

 to the experiments of Mr. P. Petit in 

 studying the coloring matter of dia- 

 toms. It has long been known that 

 when certain species of diatoms are 

 dried, the brownish, or yellowish, en- 

 dochrome changes to a bright-green 

 — such a change is particularly no- 

 ticeable in Melosira, and in the beau- 

 tiful Atdacodiscus Kittonii, of which I 

 have an excellent collection illustrat- 

 ing this fact, and doubtless it is true of 

 many other species. An examination 

 of the spectrum of the coloring mat- 

 terof diatoms known as "diatomine," 



has afforded an explanation of the 

 change of color, and has shown that 

 diatomine is a mixture of the two 

 coloring matters which are common 

 throughout the vegetable kingdom, 

 phycoxanthine and chlorophyll — the 

 former is yellow, the latter green. 

 It is clearly shown by Mr. Sorby's 

 investigations on the coloring matters 

 of plants, described in the Proceedings 

 of the Royal Society, 1873, p. 442, that 

 each of the above-named coloring 

 matters, as distinguished in diatoms 

 by Mr. Petit, is composed of sev- 

 eral distinct coloring matters, but to 

 illustrate the application of the micro- 

 spectroscope to analysis of this kind, 

 we may assume the correctness of the 

 distinction recognized by Mr. Petit. 



According to his experiments, the 

 different colors of diatoms are due to 

 the different relative proportions of 

 these two constituents. For example, 

 Melosira and Navicula contain a 

 larger proportion of the green than 

 Nitzschia or Diatoma. This is clearly 

 shown by the spectra, which are re- 

 presented on the board. 



The first spectrum represents the 

 absorption due to chlorophyll, the 

 second is that of phycoxanthine. 

 Observe the difference in the position 

 of the dark bands in the two spectra ; 

 in the latter the bands is further to 

 the left. 



The third spectrum represents the 

 coloring matter of Melosira Tiiiminu- 

 loides. The three fainter bands of 

 chlorophyll are present, and the 

 broad, black band indicates the pre- 

 sence of both phycoxanthine and 

 chlorophyll. In the fourth figure 

 there is not a sufficient quantity of 

 the chlorophyll present to develop all 

 of the faint bands characteristic of 

 that substance, but that some is pre- 

 sent is shown by the broad band on 

 the left, which also shows the presence 

 of phycoxanthine. 



By the aid of the spectroscope and 

 chemical analysis, Mr. H. C. Sorby, 

 in the article already referred to, has 

 shown the presence of twelve distinct 

 colors in the red, olive, and green 



