August 2, 1883] 



XATURE 



3 2 5 



so well ascertained a fact. I have, however, observed 

 in Iceland that lava-streams very frequently terminate in 

 or flow under lakes or gulfs of the sea, though water pre- 

 sents no obstacle to their continued progress. Lakes 

 have been filled in solid by outpourings of lava, and had 

 those I observed existed previous to the flows the lava 

 must have entered them in a more abrupt manner, and it 

 seems therefore likely that they are depressions caused 

 by the weight of the lava. But there are also instances in 

 which the actual depression produced by the weight of 

 lava-streams can be seen. A great lava flow has at some 

 period debouched from Skjaldbreith, and from two other 

 nameless craters to the south-east, on to the historic plain 

 of Thingvdllir, forty square miles of which is water. At 

 its northern end the lava is still in its original position 

 upon the slopes, but the whole central mass in the plain 

 has torn itself away from the sides and sunk a hundred 

 feet, leaving vertical cliffs of solid lava of that height on 

 its flanks. 



Again, near Myvatn there is an immense tract of lava, 

 the latest contribution to which, estimated by Mr. Lock 

 at 3t,ooo,ooo,ooo cubic feet, welled out in 1875. This 

 tract, known as the Orcefi, presents a somewhat analogous 

 instance, for the centre of the flow has also broken away 

 from its flanks and sunk. Myvatn, a lake of some thirty 

 miles in extent appears to have been formed by the weight 

 of lavas which have poured on to the plain from nearly every 

 side. Another recent stream in the same neighbourhood, 

 whose source and age are unknown, follows the course of 

 the Skjdlfandafljdt to the sea and terminates in a deep 

 gulf. It appears that, as long as lava-flows occur in 

 mountainous regions or in narrow valleys, any subsidence 

 occasioned by the additional weight is difficult to detect, but 

 as soon as it enters on to plains the subsidence is marked. 

 It may also be that valleys in undulated or folded strata, 

 being inverted arches, would resist pressure, while even 

 in valleys of erosion much of the pressure must be ex- 

 erted obliquely against the mountain masses instead of 

 wholly vertically as on a flat surface. It has been sug- 

 gested that the discharge of masses of lava at the surface 

 would leave cavities in the interior and thus occasion 

 subsidence, and it has even been anticipated that Iceland 

 would bodily disappear from this cause, like the island of 

 Friesland, from the maps ; but there is no evidence of 

 any such cavities having existed in old basaltic formations 

 or in volcanic districts, and it is far more probable that 

 the escaped matter is pressed out by other lava which 

 ii'imediately replaces it. 



Dr. Fisher believes that all plains in proximity to 

 mountain chains, upon which the material provided by 

 their denudation is spread out, sink under the weight of 

 the material and cause a compensating elevation of the 

 neighbouring mountains. The sub- Himalayan range con- 

 sists of subaerial deposits from 12,000 to 15,000 feet thick 

 brought down by torrents, and which must have been 

 deposited on a level and continuously sinking plain. 

 " The conclusion seems irresistible that corresponding to 

 the long, though occasionally interrupted, depression of 

 these plains, a correlative elevation of the great range 

 which has supplied the deposits has been going on." i If, 

 as in the Himalayas, the region be one of approximate 

 equilibrium, and much sediment is brought off the moun- 

 tains and spread over the plains, the mountains become 

 after a while too light and the plains too heavy,- and ac- 

 cordingly the mountains rise and the plains sink to restore 

 the contour. This appears to be what has happened. 



These comprise nearly all observable instances in which 

 weight has been transported from elsewhere to areas 

 where it did not previously exist, and are sufficient to 

 prove that in such cases a subsidence more or less 

 equalling in amount the vertical thickness of such added 

 matter — except in the case of ice, which is of a much 

 lower specific gravity — nearly invariably follows. Can it 



1 "Physics of the Earth's Crust/' p. 81. 



! lb., p. 83. 



be reasonably maintained that these subsidences and the 

 reelevations, which seem invariably to accompany the 

 removal of weight, whether by melting of ice, as in the 

 glacial period, or by denudation, are not the result of the 

 increase or diminution in pressure? If the accumulation 

 of sediment were due to the subsidence, instead of the 

 subsidence to the accumulated sediment, as recently 

 suggested by Dr. Geikie, it would be most improbable 

 that they would so frequently bear such near proportion 

 the one to the other. In none of these instances has the 

 subsidence exceeded the accumulation, as must some- 

 times have been the case if the sediment merely accumu- 

 lated because a subsidence quite independent of it hap- 

 pened to be in progress. 



Such subsidences would only be possible with a sub- 

 stratum somewhere of viscous matter. Professors Shaler 

 and Le Conte and Mr. Fisher, and many other very able 

 geologists, have advocated the existence of a fluid or vis- 

 cous layer between a solid interior of great density and a 

 consolidated crust. If, Mr. Fisher maintains, it requires 

 great pressure to solidify the materials at the temperature 

 of the solid interior, a melting temperature may exist at 

 some depth before the pressure is sufficient to solidify. 

 Although Prof. Geikie and many other geologists do not 

 admit the continuous existence of such a layer, it is diffi- 

 cult to see how they escape the conclusion. In his text- 

 book ' Prof. Geikie states that " from the rate of increment 

 of temperature downwards it is obvious that at no great 

 depth the rocks must be at the temperature of boiling 

 water, and that further down, but still at a distance which 

 relatively to the earth's radius is small, they must reach 

 and exceed the temperatures at which they would fuse at 

 the surface." Further on he explains that the crystalline 

 rocks of the Highlands of Scotland and of the Green 

 Mountains of New England are mechanical sediments 

 metamorphosed chiefly where they are most highly con- 

 torted, or have been subjected to the greatest pressure. 

 Strata of sedimentary origin which have accumulated to 

 thousands of feet in thickness may be depressed deep 

 beneath the surface and brought within the influence 

 of metamorphosis, 2 and be eventually reduced to a 

 soft and pasty condition, and protruded into some >>f 

 the overlying less metamorphosed masses in the form 

 of granite veins, or be erupted to the surface in the 

 form of lava. This is an absolute admission that at 

 some depth, relatively not great, pressure converts 

 solid into viscous or fluid strata. He further states that 

 " There can be no doubt that the lines of equal internal 

 temperature ijsogeotbermal lines) for a considerable 

 deptli downward, follow approximately the contours of 

 the surface, curving up and down as the surface rises 

 into mountains, or sinks into plains ; " ' so that it seems 

 difficult to understand why the particular line of tem- 

 perature or of pressure at which most rocks melt, should 

 not be continuous.- 4 Like conditions must produce like 

 results, and if the mere pressure of overlying strata can 

 anywhere or at any depth render rocks molten or fluid, 

 they will become molten or fluid wherever the required 

 pressure occurs. A nucleus kept solid at a temperature 

 higher than its melting-point, through excess of pressure, 

 cannot pass into a crust whose solidity is due to lowness 

 of temperature, through absence of pressure, without the 

 existence of that interme-iiate stage of pressure or tempe- 

 rature requisite to produce a melted zone or layer. Prof. 

 Geikie in fact himself admits ''that the nucleus though 

 practically solid, is at such a temperature and pressure 

 that any diminution of the pressure by corrugation of the 

 crust or otherwise, will cause the subjacent portion of the 

 nucleus to melt." 5 But as the pressure diminishes gra- 

 dually throughout the crust from the enormous amount 

 on the solid nucleus to the merely atmospheric pressure 



* L.c, p. 289. 



2 Geikie, " Text-Book of Geology," p. 587. 3 L.c, p. 287. 



* Fisher maintains that mountain chains have solid roots, far exceeding 

 their bulk above ground, projecting into the liquid layer. 5 L c, p. 265. 



