The Plan of the Earth and its Causes. — Gregory. 139 
the bubble of a spirit-level marks its higher end ; and they give, 
therefore, evidence of the triangular lateral flattening of the 
southern half of the globe. 
But as, on the mathematical figure of the earth, such lateral 
flattening is more improbable than variations along the axis 
of rotation, let us consider whether there is any geodetic proof 
of these flattened faces and projecting edges. 
There has been a long controversy as to whether Bessel's or 
Clark's ellipsoid better represents the figure of the earth. 
Clark's figure was the later in date, and is generally considered 
as the more exact. Helmert therefore expresses some surprise 
that the gravitational observations in central Europe along the 
52d parallel of north latitude agree with Bessel's curve better 
than they do with Clark's ; this is the case all across the area 
on which Bessel's work was done. But as soon as we get into 
the Volga basin, the gravity line diverges from Bessel's curve 
and approaches that of Clark. The change comes due north 
of the Eurafrican meridional edge. The anomalies are at 
once removed if we assume that both ellipsoids are locally cor- 
rect; that Bessel's curve is true for Europe, and Clark's cor- 
rect for Asia; and that the two merge into one another north 
of the line of the Eurafrican tetrahedral edge. 
On the tetrahedral theory, there ought to be a projection 
north of this tetrahedral edge. And gravity determinations 
show a great deficiency in gravity in western Russia in an ap- 
propriate area along the Volga basin. It is true that the fig- 
ures have been queried. There is a natural tendency to query 
all facts that do not agree with theory, and the notes of interro- 
gation in this case may illustrate that tendency. But on the 
view that there is an upward deformation of the earth in this 
area, the anomalous deficiency in gravity observations is at 
once exjplained. 
It mav be replied that the existence of a normal gravity at- 
traction at Moscow negatives the assumption of a superficial 
deformation ; but the relative excess of attraction there is pos- 
sibly due to the outcrop of Palaeozoic rocks, of greater density 
than the loose sediments of the Russian lowlands. 
Passing from Russia to the area in North America, where 
the next tetrahedral corner 'should occur, there is another area 
of deficient gravity, which may also be due to that area being a 
