772 ^ CENTURY OF PROGRESS /N THE NATURAL SCIENCES 



the limits of the continental platforms, whieh drop off quite abruptly to the 

 abyssal ocean floor at about the 200-meter depth line. Wallace accordingly 

 framed the classification of islands into "oceanic" and "continental," the oceanic 

 islands being thought to have received all of their plant and animal life overseas, 

 whereas the continental islands were populated overland, by direct invasion. 



The directly demonstrable land connections between continents, and between 

 islands and continents, that have been available for the dispersal of land animals 

 are few. There is the existing Central American isthmus connecting North and 

 South America; the shallowness of Bering Sea and narrowness of Bering Strait 

 indicate that this region afforded a broad connection of North America with Asia; 

 and except for the man-made Suez Canal, Africa is connected with western Asia, 

 and was much more broadly connected in the past before the block-faulting that 

 produced the Red Sea. The past connection of the British Isles with the Euro- 

 pean mainland and of the Greater Sunda Islands (Borneo, Sumatra, and Java) 

 with southeastern Asia are undoubted, documented physiographically by the 

 drowned river valleys in the neighboring shallow seas. 



Land Bridge Speculations 



With these known connections to explain, the existence of animal life in 

 certain islands and to explain the resemblances and relations between animals 

 of one continent and another, it was perhaps natural to turn to hypotheses of 

 other transoceanic land connections to explain such facts as the predominance of 

 marsupials in Australia and South America, or of certain types of freshwater 

 fishes in Africa and South America. The trend to bold hypotheses of this kind 

 begins with the English naturalist Edward Forbes, as early as 1846. Forbes 

 analyzed the fauna of the British Isles as to its various components, and found 

 an element in the south of England and Ireland related principally to the life 

 of Spain and Portugal. To explain this relation, he supposed the former (but 

 quite recent) existence of a continental land mass projecting far out into the 

 Atlantic. At first thought, since it was known that the sea had widely trans- 

 gressed most continental areas for long periods in the past, it seemed logical 

 enough that land areas might equally as well have been present where the oceans 

 now lie. Hypothetical continents or isthmuses — "land bridges" — between conti- 

 nents were proposed and drawn in upon maps, and presently received names. An 

 "Atlantis" was thought to have occupied most of the North Atlantic, and an 

 "Archhelenis" the South Atlantic. Antarctica was renamed as "Archinotis." 

 "Lemuria," constructed to account for the distribution of lemurs, spread across 

 the Indian Ocean from India and Ceylon to Madagascar. "Pacelia" was a lobe 

 of hypothetical land connecting the Hawaiian Islands with North America. 



Such speculations were strongly re-enforced by the geological hypothesis of 

 a "Gondwana Land" uniting all of the southern continents in Paleozoic and 

 much of the Mesozoic time introduced in 1860 by the French- American geologist 

 Marcou for the Jurassic, modified by Neumayr (1883 and 1887), and by Suess 

 (1885), who placed its origin in the Paleozoic. The Gondwana Land hypothesis 

 was based on the distribution of the remarkable fossil fern Glossopteris of 

 Permian age, and on attempts to delimit continental borders in earlier geologic 

 ages in the light of evidence from marine fossils. The great work of Edward 



