- 193 — 



these also the alimentary canal was always empty and this holds further for several species 

 from tropical seas, which I have been able to examine in the collections of the Zoological 

 Museum at Copenhagen, and which were taken in surface hauls by sailing vessels. 



The, condition seems to be therefore, that the Leptocephali one usually obtains (that 

 is from surface layers in tropical seas)' are no longer taking food; they have already 

 passed through the stage of growing and feeding and like our specimens of L. breviro- 

 stris and of Synaphobrancfuis pinnahis are already at the end of the larval develop- 

 ment or have even perhaps begun their metamorphosis. 



If this explanation is correct, as there is reason to believe, since I have found both 

 for some North Atlantic and tropical, grown up, Leptocephali that they do not feed, it 

 seems to me that much that was previously mysterious in the biology of the Leptocephali 

 becomes comprehensible. We have thus learnt the following facts: 



(1) We know the Leptocephali from essentially the upper layers in the oceans, and 

 the Leptocephali usually occurring there are such as have already passed through the 

 stage of feeding and are prepared to begin (or have already begun) their metamorphosis. 



(2) The eggs of (all?) the eels (Muraenidae) are pelagic. 



(3) Pelagic fish eggs, which in size and structure greatly resemble muraenoid eggs, 

 occur in great quantities floating in deep water in the Atlantic and hatch out there to larvae. 



(4) The youngest (preleptocephalic) larval stages of the eels, i. e. the growing and 

 feeding stages, have hitherto not been found in the sea. That the earliest preleptoce- 

 phalic stages of the eels have not hitherto been found at the places from which the Lep- 

 tocephalus stages were obtained in quantities must be ascribed to the probability, that 

 these earlier stages like the larval stages of so many other deepwater 

 species. are bathypelagic, so that they could not be taken with the apparatus and 

 methods by means of which the Leptocephali have been caught in quantities in the 

 various tropical oceans. 



If this explanation is correct, as will be determined in the near future it is hoped, 

 the conditions are as follows: 



(1) The eggs of the eels, as they have almost the same specific gravity as the 

 water in the deeper layers, remain normally deep down and hatch out to larvae which 

 likewise live pelagically. 



(2) These early preleptocephalic larvae remain normally in deeper water during 

 the period of growth and feeding. 



(3) Only after the stage of growing and feeding has been passed through and the 

 metamorphosis, often of long duration, which brings the larvae to the form and appearance 

 of their parents, is about to begin, do they rise up, to the middle or upper water layers 

 where the Leptocephalus stage is passed through. This is the reason why the prelep- 

 tocephalic stages in contrast to the Leptocephalus have hitherto usually escaped the 

 fishing apparatus and why the developmental series is so abruptly broken off below. If 

 we seek at the right places with suitable apparatus there can naturally be no doubt that 

 the preleptocephalic stages of the eels will be found in just as great quantities as the 

 Leptocephalus. 



