GEOPHYSICAL LABORATORY. 173 



light of a definite wave-length and constitute a transition phase between posi- 

 tive and negative crystals. 



Vogt has discussed the relations of akermanite and gehlenite as interpreted 

 from a study of slags. He concluded that they mix to form an isomorphous 

 series and found that the birefringence of intermediate mixtures decreases to 

 zero. However, the formulas assumed by Vogt for akermanite and gehlenite 

 differ from those used by us, and most of his materials contained iron, manga- 

 nese, soda, or other impurities, so that the data he obtained are the result of 

 more complex mixtures than the system discussed by us. 



(22) A note on the annealing of optical glass. L. H. Adams and E. D. Williamson. .J. 



Opt. Soc. Amer., 4, 213-223 (1920). (Papers on Optical Glass, No. 30.) 



This paper discusses the origin of internal strain in glass and its removal, 

 and specific directions for the practical anneahng of optical glass are given. 

 The subject of anneahng is treated more fully in another paper ("The An- 

 neahng of Glass"). 



(23) Optical properties of anthophyllite. N. L. Bowen. J. Wash. Acad. Sci., 10, 411-414 



(1920). 



The refractive indices of anthophylhte from Frankhn, Macon County, 

 North Carolina, are redetermined and found to be as follows: 



7 = 1.6404 ^ = 1.6301 a = 1.6195. 



These values are found to be in complete accord with those for anthophyllite 

 from Kongsberg and the pure artificial amphibole MgSiOs when the FeSiOa 

 content is plotted against refractive indices. 



(24) The rhyoUtes of Lipari. Henry S. Washington. Am. J. Sci., 50, 446-462 (1920). 

 Typical obsidians, pumice, hthoiclal rhyolite, and a hyalo-dacite of Lipari 



are described, with, six new analyses. As these rhyoUtes are regarded as typi- 

 cal and have never before been properly analyzed, this work is of use in char- 

 acterizing the type. The refractive indices of the obsidians and pumice, 

 determined by Dr. ]\Ierwin, are discussed. There is also given an analysis of 

 an obsidian of the island of Milos, of which the refractive index was also de- 

 termined. The data correspond well. This is compared with analyses of 

 obsidian of Nisyros by Martelh. 



The observation is made that ferrous oxide dominates ferric oxide in the 

 glassy forms of the same magna, while the converse is true of the crystalline 

 forms in the Lipari rhyoUtes. The same holds good for the rhyoUtes of Sar- 

 dinia and the pantellerites and basalts of Pantelleria, described some years 

 ago by the writer. The possible relation of this to the magmatic gases in 

 lavas is briefly discussed. 



(25) Certain relations between chemical composition and refractivity in optical glasses. 



Fred. E. Wright. J. Am. Ceramic Soc, 3, 783-832 (1920). (Papers on Optical 

 Glass, No. 31.) 



The manufacture of optical glass as a war-time problem required the solu- 

 tion of many different phases in order to establish it on a definite basis of rou- 

 tine quantity production. One of these problems was the development of 

 suitable batches for the several types of optical glasses needed in military op- 

 tical instruments. This necessitated a detailed study of the relations between 

 chemical compositions and optical constants of glasses and of the factors, such 

 as volatilization, pot solution, and state of annealing, which tend to change 

 the relations during the manufacture of the glass. The amount of accurate 

 information available at the time, especially of a chemical nature, was meager 

 but exceedingly valuable. The problem was accordingly attacked by graphi- 



