LIFE HISTORY STUDIES OF ANIMALS. 493 



the larval skeleton is replaced. Then the fish begins to feed again, 

 comes to the surface, enters the mouth of a river, and, if caught, is 

 immediately recognized as an elver or young eel. It is now a year old 

 and about 2 inches long. 



This history suggests a question. Are the dejjths of the sea free 

 from severe competition? The darkness, which must be nearly or alto- 

 gether complete, excludes more than the bare possibility of vegetation. 

 A scanty subsistence for animals is provided by the slowly decomposing 

 remains of surface life. When the dredge is sunk so low, which does 

 not often happen, it may bring up now and then a peculiar and specially 

 modified inhabitant of the dark and silent abyss. There can not, we 

 should think, be more than the feeblest competition where living things 

 are so few and the mode of life so restricted. Going a step further, we 

 might predict that deep-sea animals would lay few eggs at a time, and 

 that these would develop directly — i. e., without transformation. The 

 risk of general reasoning about the affairs of living things is so great 

 that we shall hold our conjectures cheax) unless they are confirmed by 

 positive evidence. Happily, this can be supplied. The voyage of the 

 Challenger has yielded proof that the number of species diminishes 

 with increasing depth, and that below 300 fathoms living things are few 

 indeed.^ Dr. John Murray gives us the result of careful elaboration of 

 all the facts now accessible, and tells us that the majority of the abyssal 

 species develop directly.^ 



We seem, therefore, to have some ground for believing that the depths 

 of the sea resemble the fresh waters in being comparatively free from 

 enemies dangerous to larvfe. The eel finds a safe nursery in the depths, 

 and visits them for the same reason that leads some other fishes to 

 enter rivers. It may be that the depths of the sea are safer than 

 rivers, in something like the same degree and for the same reasons tliat 

 rivers are safer than shallow seas. But we must be careful not to go 

 too fast. It may turn out that deep recesses in the shallower seas — 

 holes of limited extent in the sea bottom — enjoy an immunity from 

 dangerous enemies not shared by the great and continuous ocean floor.-^ 



After this short review of the facts I come to the conclusion that the 

 general rule which connects the presence or absence of transformation 

 with habitat is well founded, but that it is apt to be modified and even 

 reversed by highly special circumstances. The effect of habitat may, 

 for instance, be overruled by parasitism, parental care, a high degree 

 of organization, or even by a particular trick in &gg laying. The direct 

 action of the medium is probably of little consequence. Thus the 

 diflerence between fresh and salt water is chiefly important because it 

 prevents most species from passing suddenly from one to the other. 



' Cliallenger Reports. Summary of Scientific Results (1895), pages 1430-1436. 



2 Nature, March 25, 1897. 



^I am aware that other things aifect the interests of animals and indirectly deter- 

 mine their structure besides danger from living enemies. So complicated a subject 

 can only be discussed in a short space if large omissions are tolerated. 



