June 5, 1908] 



SCIENCE 



895 



and the sand used usually contains more . or 

 less iron. During tlie process of its manu- 

 facture the iron enters into the composition 

 of the glass, and if present as a ferrous com- 

 pound gives to the glass a green color. To 

 dissipate the green color manganese peroxide 

 is added to the melt for the purpose of oxidiz- 

 ing the ferrous iron to the ferric state. 

 Under these conditions if only a small 

 quantity of iron he present the pale yellow of 

 the ferric salt will not be observed; besides, 

 the yellow will be neiitralized by the violet of 

 the manganese salt, thereby producing a color- 

 less glass. Now it is well known that glass 

 decolorized with manganese slowly becomes 

 red-violet when long exposed to light, but re- 

 mains colorless when protected from the light. 

 The phenomenon is merely an instance of 

 chemical action in solid solutions; the effect 

 of heat shows it to be a reversible reaction. 

 The amethystine color is due to the presence 

 of a manganese salt. Light promotes the de- 

 velopment of the manganese salt; heat re- 

 verses the reaction. 



Henry Winston Harper 

 The Univeesitt of Texas, 

 Austin, Texas, 

 May 25, 1908 



To THE Editor of Science: Professor 

 Cockerell's note on the coloration of glass, 

 published in Science of May 22d, seems to 

 call for a word of discussion. It is not neces- 

 sary to go to arid regions to observe the phe- 

 nomenon. The globes of the street-lamps 

 used in the City of Philadelphia, colorless 

 when first put up, become in the course of two 

 or three months distinctly violet, and in a 

 year very strongly so. That the effect is the 

 direct result of exposure to sunlight is proved 

 by the fact that those surrounded by trees 

 require a much longer time for the appear- 

 ance of the color. 



As to recent literature on the subject, five 

 or six extensive papers, as well as several 

 brief notes, have appeared within the last 

 three years; it seems unnecessary to give a 

 list of these here, as they are fully recorded 



by Mr. Eoss A. Gortner.' The general con- 

 clusion from these various studies is that the 

 development of the color is due to the oxida- 

 tion of the manganese in the glass, although 

 the exact mechanism of the change is not 

 understood. Whether the tints produced in a 

 great variety of substances by exposure to 

 radium preparations are of the same character 

 as those brought about by sunlight and ultra- 

 violet light in glasses ' has never been defi- 

 nitely ascertained; but it appears more prob- 

 able that they belong in the class of colloid 

 colors, such as the red of glass containing 

 metallic gold, the blue of sodium chloride 

 heated in sodium vapor, and, possibly, the 

 violet of the amethyst-quartz found in nature. 



Edgar T. Wherry 

 Philadelphia, Pa. 



the ITALIAN archives OF BIOLOGY 



To THE Editor of Science: The publica- 

 tion of the Archives Italiennes de Biologic, 

 founded by Professor Mosso, after having 

 reached forty-eight volumes and its twenty- 

 fifth year met with a grave interruption in its 

 career, owing to the strike of the typographers 

 at Turin, and it has been found necessary to 

 make new arrangements for the continuation 

 of the journal. The next number is shortly 

 to be issued under the auspices of the new 

 administration. The publication remains, as 

 in the past, under the direction of Professor 

 Angelo Mosso, with Professor V. Aducco and 

 Professor U. Mosso as coeditors. It will still 

 have the cooperation of biologists in sundry 

 Italian universities. The original articles and 

 summaries published in the Archives repre- 

 sent faithfully the progress of biology each 

 year in Italy. The appearance of the journal 

 will be improved, and the editors make an 

 appeal for increased support from America. 

 The publication has acquired a high standing 

 and ought certainly to be among the journals 

 taken by every university in the country. The 

 subscription price is 40 frs. for the two annual 

 volumes. Subscriptions should be sent to the 



' " Some EfiFects of Smiliglit on Colorless Glass," 

 American Chemical Journal, Vol. 39, 1908, 157- 

 162. 



