308 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1909. 
coherent, then no upward-pointed fissure could be formed. In prac- 
tice neither of these conditions is fulfilled. It is obvious that the 
crowding and crushing will be most complete in the shells of com- 
pression (pl. 1, C) where these are carried on a continuous block 
of contraction (pl. 1, A). I mean by a block a portion bounded by 
fissures formed in the contracting part of the earth crust. Where the 
shells of contraction are fissured (E, F) there the crowding of the 
superincumbent masses of cooled rock will not take place. As a 
result, each block or island of contractile crust, with its compressed 
burden, will tend to tear away from the adjoining blocks or islands, 
so that the limiting fissures in the contractile joints will extend up 
into the compressional shells (G, H). 
This exactly fits in with what is frequently found in the distribu- 
tion of volcanoes along the edges of areas of marked compression or 
mountain regions. It will explain also the presence of volcanoes 
having a linear arrangement between closely situated mountain 
‘chains or areas, as in South America. Great rifts, such as those of 
Central Africa and some canyon districts, are probably of such 
origin. In earthquakes of tectonic origin it has been pointed out ¢ 
that the piers of damaged bridges have usually been found to have 
approached each other. This would evidently take place in the 
areas of positive compression (pl. 1, C, C). On the other hand, in 
exceptional cases the piers have been found to have been separated. 
This might well occur in the area of negative compression (R), or 
what might well be termed the areas of retraction. The much larger 
proportion of the former effect on the bridge piers would no doubt be 
in the much greater ratio of compressional areas to retractional areas 
on the earth’s surface. 
May not ocean basins be in part due to blocks or islands of the 
contracting zones exerting that diminution of volume in a vertical 
more than in a horizontal direction, as we have so far been con- 
sidering it to be? The peculiar abysmal ocean troughs, often at the 
edge of ocean basins and parallel to chains of volcanoes or inter-, 
rupted by them, could well be explained by the same circumstances. 
I do not claim that ocean basins are alone due to this cause, but to a 
combination of these conditions, with perhaps the slipping, shearing, 
and corrugating of the primitive crust over a fluid envelope, and 
even the tetrahedral collapse of a cooling globe. I lay down here but 
a general principle to which there may be many exceptions, due to the 
vicissitudes of cooling and the variation in the materials concerned 
in any particular region, not to speak of the changing position of the 
earth’s axis, the crustal inertia of Prof. G. H. Darwin, etc. 
«Professor Hobbs, Ninth International Congress of Geography, 1908. 
