216 Br. Harold Jeffreys — Causes of Mountain-Building. 



allowed. Now it is obvious that such a contraction, if it occuried, 

 would shorten every diameter of the earth by this amount, and 

 could therefore raise a mountain chain as high, as broad, and as 

 complicated as the Alps, and extending most of the way round the 

 earth, whereas the estimate in question refers only to a chain some 

 1,000 kilometres in length. This simple method of evaluating 

 the contraction needed to produce a single range is evidently 

 unsatisfactory ; what we really need to determine is the reduction, 

 not in breadth, but in area, if the aggregate contraction needed to 

 raise all the mountains of the earth is to be found. Taking the 

 llockies and the Coast Range in California as standards, according to 

 the breadth of the particular range considered, and assuming the 

 lateral compression to be proportional to the mean height above the 

 surroundings (which would be exactly true if the contortion in all 

 ranges was geometrically similar and on a scale proportional to the 

 height), I have indicated elsewhere that all the known mountain 

 ranges could probably be produced by a contraction in circumference 

 of some 70 km., only about half of that shown to be available. 

 This will, of course, have to be revised continually as direct geological 

 information about the larger ranges, especially those in Asia, becomes 

 available ; it will further have to be increased to allow for suboceanic 

 mountains and old ranges now almost denuded away. At the same 

 time the estimate of the compression that could be produced by 

 cooling may need to be increased, as the coefficient of expansion 

 may increase more rapidly near the melting-point than I assumed ; 

 in any case the available compression and that needed to account 

 for mountain-building are of the same order of magnitude, and 

 a categorical statement that the theory is inadequate is clearly 

 unjustified. 



The theory of compression is complicated by the effects of 

 denudation. When a mass of radio-active matter is removed from the 

 surface of a continent, matter at a considerable depth is enabled to 

 cool more rapidly in comparison with the average over the whole 

 earth. In consequence it tends to contract more, and thereby 

 increases the curvature of the outside, just as a bow becomes more 

 curved when the string is tightened ; the effect of this would be to 

 raise the continents considerably. Similarly, the ocean beds would 

 be depressed on account of the radio-active sediments acting as 

 a blanket. A definite limit must of course be imposed on these 

 effects by the weakness of the crust ; thus the elevation and lowering 

 could never become much greater than was necessary to giveisostatic 

 compensation, and afterwards more mountains would be raised within 

 the continents and fewer on the ocean bed. 



The alternative hypothesis offered is that crust movements liave 

 occurred owing to heavy matter sinking into lighter matter below it 

 and causing it to spread out horizontally. This is physically quite 

 possible, but it is difficult to see how the heavy matter could have 

 got to the top in the first place except by compression. Isostatic 

 readjustment, subsequent to denudation, might easily produce gentle 

 anticlines and synclines, but the violent contortion implied by the 

 structure of the great mountain ranges seems difficult to account for 



