CAUSES OF VOLCANIC ACTIVITY^DAY 261 



heavenlj^ body passing too near, condensed within a comparatively 

 short time to a liquid nucleus. This nucleus in its normal con- 

 vection cooling then acquired a crust Avhich broke up and sank as 

 soon as formed, its specific gravity being measurably greater in the 

 solid than in the liquid state. It is no more than a generation since 

 distinguished physicists were still in doubt about the relatiA^e den- 

 sity of solid and liquid lava, and some then believed that the crust 

 ought to float as readily upon the molten interior of the earth as 

 patches of crust float upon a liquid lava basin, like that at Kilauea 

 (pi, 2, fig. 1). Closer observation, hoAvever, has since established the 

 fact that the floating is but temporary and is due to nothing more 

 than the accident of gas accumulations beneath these patches of 

 crust, Avhich give sufficient buoyancy to support them on the sur- 

 face for a long time, but only for so long as they remain perfectly 

 poised. The rise of a bubble more violent than the rest, Avhich is 

 sufficient to tilt one of these blocks but a very little, or an over- 

 lapping Avave in the lava surface, Avill quickly cause them to lose 

 the supporting gases and to glide o'ut of sight beneath the surface 

 much as a piece of sheet iron sinks in Avater. 



The normal behavior of a crust forming upon a fluid mass highly 

 charged Avith gas, such as Ave must believe the nucleus of the earth 

 to haA^e been, is to sink, and so Jeffreys concludes that though the 

 first blocks to sink Avere undoubtedly remelted presently in the 

 course of their doAA^iward moA^ement and so merely helped to cool 

 the magma, eventually the sinking crust accumulated at the center 

 and the resulting earth became in consequence something of a honey- 

 comb structure, the rigidity of which may quite properly be equal 

 to that of steel, as has been so completely shown by the tidal studies 

 of Michelson and Sir George Darwin, and that any portions remain- 

 ing liquid can be nothing more tlian the product of local conditions 

 which presumably are slowly passing aAvay. 



This concept of the formation of the earth as a Avhole readily lends 

 itself to the interpretation of A'olcanoes as we find them. It is 

 extremely difficult to conceive of a volcano as the safety valve of a 

 molten interior, as the older geologists Avere accustomed to think of 

 it, or of a molten layer concentric Avith the crust as it was later con- 

 ceived to be, so long as tAvo such openings continue to exist side by 

 side in the Island of HaAvaii, one of AA-hich (Mauna Loa) is 10,000 

 feet higher than the other (Kilauoa) and yet liquid lava emerges 

 more frequently and in greater quantity from the higher opening, or 

 its subsidiaries, than from the loAver. It is also a matter of record 

 that greatly increased activity in the one basin rarely finds an echo 

 in the other. The liquid lava of HaAvaii is at least tAvo and one-half 

 76041—26 IS 



