ii8 The American Geologist. February, 1901. 
In the earth the major watersheds have exactly this arrange- 
ment. The great watershed of Eurasia dividing the northern 
and southern drainages, runs, not along the main mountain 
axis, but far to the north of it, between the parallels of 50° and 
60°. The northern and southern slopes of North America are 
separated by a divide along the same latitude. The southern 
watersheds, instead of following the lines of highest moun- 
tains, or the middle line of the continents, run close to the 
coast-lines ; the three watersheds mark the three vertical tetra- 
hedral edges, and they occur at almost the theoretical distances, 
120° apart. 
Similarly with the mountain chains. As Sir John Lubbock 
has pointed out, "in the northern hemisphere we have chains of 
mountains running east and west, the Pyrenees, Alps, Carpath- 
ians, Himalayas, etc. — while in the southern hemisphere the 
grea't chains run north and south — the Andes, the African 
ridge, and the grand boss which forms Australia and Tas- 
mania." That is to say, the northen mountains are parallel to 
the upper edges, and the southern mountains are parallel to 
the meridional edges of the tetrahedron. 
THE CAUSE OF THE TETRAHEDRAL PLAN. 
The statement that the elevations of the lithosphere are tet- 
rahedral in arrangement is not a hypothesis, but a simple state- 
ment of geographical fact. Is the fact a mere coincidence? 
On the contrary, there are good reasons why the earth should 
acquire such a tetrahedroid symmetry. When the earth solidi- 
fied, it would (neglecting the influence of rotation) have con- 
tracted into a spherical shape. It would have tended to acquire 
this form because the sphere is the body which encloses the 
greatest volume for a given surface. But as the earth contracts 
it tends to acquire a shape in which there is a greater surface 
in proportion to its bulk. Now, among the regular geometrical 
figures with approximately equal axes, the tetrahedron is that 
which contains the smallest volume for a given surface. Hence 
every hard-shelled body which is diminishing in size, owing to 
internal contraction, is constantly tending to become tetrahedral 
in form. Fairburn's experiments (quoted by Green) illustrate 
this tetrahedral collapse for short tubes ; and that it is consid- 
ered probable by some geodists is shown by the following quo- 
