GEOPHYSICAL LABORATORY. 157 



2:1. The ternary compounds are likemse made up of the simpler binary- 

 compounds in simple proportions. The impression given by a general review 

 of the silicates and other compounds of the refractory oxides is that these 

 compounds are of the so-called "molecular" class. Little is to be expected, 

 therefore, from structural formulas for the silicates until the nature of the 

 chemical union in "molecular" compounds is better understood. The bear- 

 ing of the X-ray analysis of crystal structure upon the problem is shown, and 

 it is suggested that certain of the more complex crystalline compounds may 

 be geometrical rather than chemical products. Various examples of the 

 practical aspects of the properties discussed are added. 



(31) A precision projection plot. Fred. E. Wright. .J. Wash. Acad. Sci., 6, .521-.524 (1916). 

 For the graphical solution of spherical triangles and of certain crystallo- 

 graphic-optical problems a projection of the sphere, such as the stereographic, 

 is commonly used. Chauvenet in 1854 published such a stereographic pro- 

 jection plot; in 1885 Sigsbee prepared an exceedingly exact stereographic 

 projection plot and recommended that the solution be made by means of 

 tracing-paper placed over the accurate base-plot. In 1902 Wulff proposed the 

 same procedure as a new method, which is now known imder his name. 

 The method should, however, be called the Chauvenet-Sigsbee method. The 

 precision plot described in this paper is simply a mechanical device for rotating 

 a piece of tracing-paper over an accurate projection-base printed on thick 

 celluloid. A metal stand supports an electric light which illuminates a disk 

 of frosted plate-glass; this plate in turn supports the equatorial projection-net 

 which can be accurately centered to the axis of rotation of an outer steel ring 

 which runs in an accurately turned bearing and carries the tracing-paper on 

 which the measurements are plotted in the positions indicated by the under- 

 lying projection-base. The tracing-paper is held in place b}' means of hinged 

 iron bars which pass over the outside square ends of the rotating ring and 

 clamp the paper securely. 



(32) The lava eruption of Stromboli, summer-autumn, 1915. F. A. Ferret. Am. J. Sci. 

 (4), 42, 443-463 (1916). 



This paper describes one of the most violent eruptions of Stromboli that has 

 been recorded, and one which is especially noteworthy because of the almost 

 continuous emission of lava down the Sciarra during a period of about 5 

 months — the emission of lava being a rare phenomenon at this volcano, where 

 the activity is normally of the explosive type. 



The record of observed phenomena is based upon personal observations 

 from November 9 to November 30, 1915, supplemented by notes made 

 during the preceding months by an Italian official on the island. Four 

 vents were active, three being on the upper edge of the Sciarra and the other 

 farther back on the terrace. The lava issued from a tunnel-like opening close 

 to one of the permanent explosive vents (A), near the center of the upper part 

 of the Sciarra, and at times it floM^ed in a continuous stream to the sea. A 

 remarkable feature of its entrance into the sea was the absence of commotion 

 whenever the lava flow was still incandescent on entering the water. This 

 phenomenon, also observed at Sakurashima, is attributed to the formation of a 

 protective, thin, poorly conducting, and quasi-flexible lava sheath through the 

 sudden chilling. Chemical analyses show that this lava is a basalt, and closely 

 resembles in composition those of previous eruptions. 



Observations were made of the changes in the volume of visible vapor (the 

 "smoke" cloud) above the vents concomitant with the hygroscopic state of 

 the air — a change from a dry to a damp atmosphere immediately resulting in 

 a great increase in the amount of the volcanic "cloud," without any cor- 



