xvi LIFE-HISTORIES 327 



blade-like larvae (Leptocephali) are found which remain for 

 a year or more pelagic. They change into a cylindrical 

 shape and migrate shorewards, often for great distances, 

 it may be several thousand miles, e.g. to the Baltic. 

 They pass up the rivers in legions in a marvellous pro- 

 cession or eel-fare, and reach ponds and quiet-flowing 

 streams. There the eels feed and grow, and after some 

 years attain sexual maturity. But there is never any 

 breeding in fresh water, and as the constitution changes 

 the creatures become restless. They change in colour 

 and become large of eye, and there is an excited return 

 journey to the sea. So far as is known they go out into 

 deep water to spawn, and the individual life probably 

 ends in giving rise to new lives. In no case is there any 

 return. 



3. Prolonged Embryonic Life. Another type of life- 

 history implies a suppression of the larval period and a 

 lengthening out of the embryonic development within 

 the egg-shell or egg-envelope. A good instance is to be 

 found in a class of unfamiliar but widespread primitive 

 animals, the Onychophora, of which Peripatus (see p. 252) 

 is a representative genus. They are interesting ' syn- 

 thetic types," with certain features of insects (e.g. air- 

 tubes or tracheae) and others of segmented worms (e.g. 

 kidney-tubes or nephridia) ; they are without doubt 

 survivors from very ancient days, keeping a precarious 

 foothold in many parts of the world in virtue of their 

 quiet w r ays, their nocturnal habits, and their quaint device 

 of squirting on to small insects a jet of slime from two 

 papillae in the mouth. These characteristically "crypto- 

 zoic " creatures are viviparous, and the young they bring 

 forth are from birth miniatures of the adults. Moreover, 

 the period of embryonic development is remarkably pro- 

 longed, sometimes lasting a whole year. Contrasting this 

 with the life-history of one of the higher insects, like a 

 butterfly, we see that in Peripatus it is the embryonic 

 phase that is lengthened out, while in the butterfly it is 

 the larval phase. 



The time that birds can spend in brooding is limited by 

 the conditions of food-supply, temperature, and the like ; 



