TWO-PHASE CONVECTION IN IGNEOUS MAGMAS 4Q1 
Gas-phase convection—This case is covered by Daly.t He 
assumes from observations on vesicular lava at craters 200 “‘stand- 
ard” (t mm. at surface pressure) bubbles per cubic centimeter. 
At a depth of 3,000 feet a magma is under a pressure of 200 pounds 
per square inch. This is a greater pressure than some magmas are 
subjected to, but it is to be noted that the gas is not only com- 
pressed, but more soluble under pressure—a fact which Daly does 
not seem to consider. There should also be mentioned some 
thermal effects connected with the separation, reaction, and expan- 
sion of the gas bubbles; but too little is known of the effect of these 
factors on the density to include them in the calculation. 
To obtain data comparable with those of other calculations 
in this paper the following case was selected: 
Assumed pressure, 200 pounds per square inch 
Assumed vesiculation, 200 ‘‘standard” bubbles per cubic centimeter 
Magma specific gravity, 2.70 
Vesiculated magma specific gravity, 2.638 
Density difference, .062 
Final rate of motion of a sphere of 1o meters radius, over 2,200 meters 
per hour. } 
Double liquid-phase convection.—If an intermediate magma splits 
into two immiscible liquids, consisting of granite and gabbro 
phases, the difference in specific gravity might be so great that a 
rapid separation of the two would occur; but it is not certain that 
the separation of immiscible globules is accompanied by any 
pronounced change in the aggregate specific gravity. 
Crystal-phase convection.—The effects of the development of 
crystals should be emphasized, because of the certainty of the 
when the viscosity is 5. From‘this, R=1.6 cm. When R is 1.6 cm. and V=s5, the 
Stokes formula becomes 
a 2(980)(r .6)2(.0r) 
Fale oat Ciao, 
45 : 
per second. For a sphere of 10 meters radius, the last formula 
D1 VaR b Tye V1.6 
Fl VR” ecomes INT ner 5 
From this «’’=27 cm. per second. This is 972 meters per hour. 
For the greater viscosity, 300, the radius R, of the sphere that will obey Stokes 
law, is greater, but the final rate of motion of a sphere of to meters radius is nearly 
the same as in the case of the lower viscosity. 
tR. A. Daly, Igneous Rocks and Their Origin, pp. 261-64. 
