> 










will conform with the normal law of errors. The crust 
of the earth is not so constituted that each point can 
move independently of the rest, and the movements 
therefore are not analogous to the errors in a series of 
ingpendent observations. According to the geological 
gWidence the greater movements, which have most 
ence on the frequencies, are of a widespread char- 
acter, and their general effect is to throw the surface 
‘into broad undulations. Upon these broader move- 
_ ments are superimposed the more intense but more local 
untain-building movements. 
Mr. G. V. Douglas points out in a paper to appear 
shortly in the Geological Magazine that if we start with a 
level, or equipotential, surface, and suppose it affected 
by movements of the types referred to, the resulting 
altitudes will necessarily give a frequency curye showing 
two maxima. The actual frequency curve 1s, in fact, 
perfectly consistent with ordinary geological concep- 
tions and does not require the original existence of the 
two distinct surfaces postulated by Wegener. 
Wegener imagines that at the close of the Carbonifer- 
ous period the Sal formed one continuous patch covering 
about half the globe, and the Sima covered the rest. 
He professes that he has taken the forms of the existing 
land-masses, including their continental shelves : he has 
modified the present forms by unfolding the mountain 
ranges which have been raised since the Carboniferous 
period ; and he finds that the different patches can 
then be fitted together into one continuous whole, like 
the pieces of a puzzle. It is evident, however, that 
Wegener has given free play to his imagination. In 
following the edge of the continental shelf he has 
allowed himself a very considerable amount of latitude, 
and he has not hesitated to distort the shapes of the 
masses. Few geologists who are familiar with moun- 
tain structures will attach much value to Wegener’s 
estimates of the effect of Post-Carboniferous folding. 
It is easy to fit the pieces of a puzzle together if you 
alter their shapes, but your success is no proof that you 
have placed them in their original positions. It is not 
_ even a proof that the pieces belong to the same puzzle. 
If Wegener’s hypothesis rested solely on the evidence 
of fitting that he brings forward it might well be 
ignored. But there is more to be said for it than this. 
’ In the Indian Peninsula the oldest fossiliferous 
_ deposits are of terrestrial origin and contain remains of 
plants and of reptiles. The flora is commonly called 
the Glossopteris flora and is very distinct from the con- 
temporaneous flora of north-western Europe. There 
_ isa similar series of terrestrial deposits in South Africa 
and another in Brazil, both of which contain the Glosso- 
_ pteris flora and remains of reptiles. The Glossopteris 
_ flora occurs, moreover, in Australia, the Falkland 
Islands, the Antarctic continent, and in other parts of 
South America besides Brazil. In Wegener’s recon- 
struction all these areas are brought together, and it is 
easy to understand why they should have a common 
flora and why that flora should be different from the 
flora of the distant Europe. 
But the Glossopteris flora is found also in Kashmir, 
north-western Afghanistan and north-eastern Persia, 
Tonquin, northern Russia and Siberia. In Wegener’s 
reconstruction all these areas lie far from the masses 
that he has grouped together in the south. 
The Russian deposits are especially interesting. Not 
NO. 2781, VOL. 111] 
- 
NATURE 
| 

only do they contain representatives of the Glossopteris 
flora, but they also include reptiles of the same type as 
those which are found in South Africa, and several species 
of freshwater shells which are identical with those in the 
South African beds. Wegener’s explanation has not by 
any means simplified the problem of the distribution of 
the Glossopteris flora and of the fauna associated with it. 
In India, South Africa, South America, and Australia 
the system containing the Glossopteris flora begins with 
a boulder bed, which is universally admitted to be 
of glacial origin. These glacial deposits are now 
scattered over a wide extent of the earth’s surface. 
Even if we admit movement of the pole, on the most 
favourable supposition the ice must have spread much 
farther towards the equator than the ice-sheets of the 
Pleistocene Glacial period ever did. Nor is it possible 
to invoke the aid of icebergs, for the associated deposits, 
except in the case of Australia, are all of terrestrial 
origin. With Wegener’s reconstruction these diffi- 
culties disappear. The areas are grouped together and 
the pole may be placed conveniently in the middle of 
the mass. 
But the boulder beds of this period are not limited to 
these areas. There is a boulder bed in the Salt Range 
which appears to be of the same age as the Talchir 
boulder bed of the Indian Peninsula. In north-western 
Afghanistan Griesbach found a boulder bed similar to 
the Talchir boulder bed, and in the beds overlying it he 
found several of the characteristic plants of the Glosso- 
pteris flora. According to Wegener’s maps this boulder 
bed must have been deposited within 30 degrees of the 
equator of the period ; and it cannot have been laid 
down at a great elevation, for the beds that conform- 
ably follow it include both marine and terrestrial 
deposits. Wegener’s ideas have not very greatly 
reduced the area that must have been affected by the 
ice of the Permo-Carboniferous Glacial period. 
There is another line of evidence that Wegener puts 
forward. There are five geological features, according 
to him, which occur on the two sides of the Atlantic 
and are re-united when the patches of Sal are fitted 
together. : 
The strike of the ancient gneiss of the Hebrides and 
northern Scotland becomes, he says, continuous with 
that of the gneiss of Labrador. The former, according 
to him, now runs from north-east to south-west, the 
latter from east to west. But according to the Geologi- 
cal Survey of Scotland the prevalent direction in Scot- 
land is W.N.W.-E.S.E. or east to west. If Wegener’s 
direction fits the other side the real direction does not. 
The Caledonian folds of Scotland and Ireland, he says, 
become continuous with those of Newfoundland. But 
the Newfoundland folds are of considerably later date. 
If there was actual contact the earlier Scottish folding, 
in spite of its great intensity, must have ended abruptly 
at the line where separation was to take place ages after- 
wards, and on the other side of the line the commence- 
ment of the later Newfoundland folds must have been 
equally abrupt. 
Farther south the Armorican folds of Europe, in 
Wegener’s reconstruction, are continued by the 
Appalachian folds of North America, and no objection 
can be raised on the score of age. But a single coincid- 
ence of this sort has no value, for Wegener has adopted 
the simple plan of bending North America so that the 
