THE PRESIDENTIAL ADDRESS 15 
reefs; that Central America and the Mediterranean are a difficult 
obstacle ; and that the known distribution of the Karroo fossil 
reptiles is not by any means what the hypothesis demands. 
If the idea of drift be accepted it cannot be regarded as a royal 
road out of all our difficulties, nor can it be the only form of 
earth-movement to be reckoned with. The late J. W. Gregory, 
whose life was sacrificed to geological discovery, has studied ex- 
haustively the geological history of the Atlantic and Pacific Oceans, 
both as revealed by the sedimentary rocks and fossils on their 
borders, and by the distribution of life to-day. He finds that, 
according to our present knowledge, in the two oceans, facilities 
for migration have fluctuated from time to time, periods of great 
community of organisms alternating with periods of diversity. 
Again, at some times connexion seems to have been established 
north of the equator, at others to the south ; and we cannot ignore 
the possibility of migration across polar lands or seas when terrestrial 
climates haye differed from the present. The facts of life distribu- 
tion are far too complex to be explained by any single period of 
connexion followed by a definite breaking apart, even if that took 
place by stages. Mrs. Reid, too, has pointed out that resemblances 
between the Tertiary floras of America and Europe actually increased 
at the time when the Atlantic should have been widening. Unless 
continental drift has been a more complicated process than anyone 
has yet conceived, it seems impossible to escape from some form of 
the ‘ land bridges ’ of the older naturalists : 
‘ Air-roads over islands lost— 
Ages since neath Ocean lost—’ 
We have no right to expect greater simplicity in the life of a planet 
than in that of an organism. 
As the question of drift must in the last appeal be one of fact, 
it is not unnaturally expected that the real answer will come from 
measurements of longitude and latitude with greater exactness and 
over periods longer than has yet been possible. None of the 
measurements hitherto made has indicated variations greater than 
the limits of errors of observation. ‘T'wo things, however, may 
militate against a definite answer from this source. Many parts of 
the crust, such as the shield-like masses of Archzean rock, may have 
completed their movement, or be now moving so slowly that the 
movement could not be measured. Careful selection of locality 
is essential, and at present we have little guidance. Also, as the 
displacement of crust must be dependent on the condition of its 
substratum, it will be a periodic phenomenon and the rate of move- 
ment may vary much in time. According to the theory of thermal 
cycles the sub-crust is at present solid, and may not permit of drift. 
