i 3 2 SCIENCE PROGRESS 



age, on Prince William Sound, Alaska, and adduces evidence 

 which conclusively proves that these lavas were subaqueous 

 flows. Each flow has a flat base, with an abundance of mud- 

 filled cracks. 



In the Bulletin of the Geological Society of America for 

 December 1914, J. V. Lewis reviews all known occurrences of 

 pillow-lavas, and compares the structure with the lava structures 

 known as " pahoehoe " and " aa." He propounds a theory of 

 " bulbous budding " in explanation of the origin of the curious 

 pillow or sack-like forms of these lavas. In the declining stages 

 of a large flow a lava breaks up into a multitude of small flows, 

 each of which forms an elongated bulbous mass. On exposure 

 to the atmosphere a tough elastic skin of solidified lava forms 

 on each of these masses. The further expansion of the lava 

 within produces the concentric flow structures frequently seen 

 in the " pillows." Continued expansion may rupture the crust 

 and give rise to another little pillow-form extravasation. The 

 flow of a lava may thus continue through a series of pillows 

 connected by very short necks, which are indeed seen in many 

 occurrences. This theory seems to explain satisfactorily most 

 of the peculiar features of these flows. 



A most important piece of experimental work by N. L. Bowen 

 (" Crystallisation-Differentiation in Silicate Liquids," Amer. 

 Jour. Science, February 191 5) demonstrates the effectiveness of 

 gravity-separation of crystals in silicate melts, and furnishes 

 strong evidence in favour of the process of gravity-differentia- 

 tion as a factor in producing the observed diversity of igneous 

 rocks. Olivine crystals were produced in a molten glass at a 

 temperature of 1,460° C, contained in a crucible 15 mm. in depth. 

 After 80 minutes all the olivine crystals were collected in a 

 layer 1*5 mm. thick at the bottom of the crucible. Similar 

 results were obtained for pyroxene crystals, but tridymite, when 

 formed, tended to rise in the melts. Those results are applied 

 to the case of an olivine-rich layer in the quartz-diabase sill of 

 the Palisades of the Hudson River, and it is concluded that the 

 accumulation of the olivine crystals may have been accomplished 

 in 200 to 300 hours. This experimental demonstration of 

 sinking of crystals is welcome in view of the many recently 

 described cases of ultrabasic layers in igneous rocks which 

 have been ascribed to some form of gravity-differentiation. 



Dr. Leigh Fermor (Gcol. Mag., January and February 1915) 



