January 19, 191 1] 



NATURE 



391 



'so that, just as from a morphological point of view the 

 .es may be characterised by a definite form or struc- 

 , so, too, it is possible to characterise it by certain 

 -defined conditions of existence. These conditions 

 acterise a given species quite as much as any morpho- 

 al description, and, in fact, for a proper conception 

 lie species both methods of investigation are supple- 

 :ary. 



jw, if we wish to investigate the conditions of exist- 

 under which animals live, we must naturally first of 

 Ascertain where they live, which, in the sea, will be 

 imount to discovering the depth where they reside. 



- alone can enlighten us regarding the conditions of 

 , the temperatures, and the salinities which are 

 isite for their existence. 



ur knowledge regarding the haunts of the smaller 



nisms has advanced greatly in recent years owing to 



fact that we have made use of small closing nets. 



with regard to the larger animals, especially fishes, 



-till possess little knowledge, despite the great exer- 



- made in this direction, above all others, by the 

 :e of Monaco and the \ald'\v\a expedition. 



hen fitting out the Michael Sars expedition, I kept 

 in mind that one of the most important of all our 

 ~ should be to try and develop a method which would 

 : more information regarding the vertical distribution 

 lep-sea fishes. ' 



e ideal instrument for capturing the larger pelagic 



nisms would be a big tow-net or pelagic trawl which 



1 be sunk closed to the requisite depth, then opened 



towed at a carefully ascertained depth, and finally 



i again and hauled in. Such an instrument would 



ipabie of capturing many of the larger animals, and 



ould secure them, too, at known depths. However, 



an instrument of this kind would naturally be extremely 



'complicated, too much so, in fact, to prove trustworthy 



r our present system of working, and it would further 



1 a great expenditure of time. It would not be 



ble to operate many of these tow-nets simultaneously 



■ngst other reasons, because of the slip-leads), and 



-rquently it would be necessary to undertake a series 



jof hauls at the different depths. 



This being the case, I gave up the idea of trying to 



-rruct any such instrument. A more practical plan 



^d to be to try and tow a number of instruments 



- uultaneously at different depths, and to compare the 



; catches thus made with each other. 



I Thanks to a practical arrangement, we succeeded in 

 I towing ten different instruments from two wire lines. 

 !The arrangement generally adopted was as follows : — 



I silk net of i metre diameter 



I young- fish trawl 

 I silk net of J metres diameter 

 I j'oung-fish trawl 

 I silk net of f metres diameter 

 I young- fish trawl 

 1 silk net of J metres diameter 

 I large tow-net of 3 metres diameter 

 (made of shrimp-net) or a young- 

 fish trawl 



U ith this, or a corresponding arrangement, we carried 

 ■out some long hauls at about thirty stations, as well as 

 ifrom the Canary Islands to the Sargasso Sea, and from 

 Newfoundland to Ireland. Some hauls were made in the 

 daytime and others at night. 



J In this manner we collected a very large material, con- 

 sisting of many kinds of pelagic organisms — fishes, 

 cephalopods, crustaceans, medusae, &c. I will here merelv 

 mention a few instances of the evidences our material 

 affords as to the occurrence of these animals at different 

 depths. To illustrate the method employed, I will begin 

 lyith the remarkable and well-known Sternoptvchid 

 ■■i^gyropelecus hemigymnus. Of this species we caught 

 ;286 mdividuals at the different stations. The bulk 

 I occurred at depths between 150 and 500 metres; no in- 

 ,aividuals were caught above 150 metres, and onlv about 

 j7 per cent, were taken at depths lower than 500 metres, 

 we assume, then, that these 7 per cent, were captured 

 1 NO. 2 15 I, VOL. 85] 



during the process of hauling in the appliances, and that 

 none of them live at depths below 500 metres, we will 

 have an idea of the accuracy of our method. By far the 

 greater part were caught at a depth of 300 metres, where 

 we generally had out a |-metre silk net, whereas at 150 

 metres and at 500 metres the appliance used was, as a 

 rule, a young-fish trawl, that would have had a far 

 greater capacity for catching these fish. It seems, accord- 

 ingly, that the preponderating majority of the individuals 

 of this species is very strictly limited to an "' intermediary " 

 layer, situated at a depth of about 300 metres. A closer 

 investigation of the individuals captured at a depth of 

 150 metres shows that they were all caught at night. 

 This may be due either to an upward nocturnal wandering 

 or to chance, though on this question the smallness of 

 our material makes it unsafe to hazard an opinion ; in 

 subsequent investigations, however, it will be worth while 

 taking this fact into consideration. Among the individuals 

 captured in 500-metres water there must, at any rate, be 

 a few that were taken in the process of hauling in the 

 young-fish trawl through the intermediary layer above ; 

 still, there were far fewer found in the young-fish trawl, 

 which was towed in 1000-metres water — it seems evident 

 that there must also have been some individuals swimming 

 at the 500-metres depth. 



This instance gives us a good illustration of our method, 

 with its deficiencies and advantages. It is obvious that 

 the greater the number of individuals we have to deal 

 with, the greater is the probability of obtaining trust- 

 worthy information, and the safer are the conclusions we 

 can deduce from our results. When, therefore, in what 

 follows, I proceed to give some instances of the distribu- 

 tion in depth of different kinds of fish, I will begin by 

 mentioning the commonest, or, at any rate, the most 

 numerously captured forms belonging to the species 

 Cyclothone microdon and C. signata. 



Of these two species we caught altogether more than 

 7500 individuals, which were all measured and arranged 

 according to their length and the instrument in which they 

 were captured, so as also to obtain information regarding 

 the occurrence of the different sizes at different depths. 



C. microdon was found during the cruise of the Michael 

 Sars in the northern Atlantic at every station where an 

 appliance was towed in depth below 500 metres. Above 

 500 metres it was only met with occasionally. A table 

 shows how, at a depth of 300 metres, we only came across 

 one individual (in the southern section). In depths from 

 500 metres down to 1500 metres its quantitative occurrence 

 appears to be fairly uniform. 



In our northern as well as in our southern section we 

 found approximately the same number of individuals in 

 each of the three young-fish trawls, which we towed 

 simultaneously, viz. at depths of 500 metres, 1000 metres, 

 and 1500 metres. 



When we next examine the size-distribution at the 

 different depths, we see that it is perfectly clear that the 

 smaller sizes are met with much higher up than the kirger 

 ones, which latter are mainly to be found at a depth of 

 1500 metres. In the northern section we find that at a 

 depth of 500 metres the greatest number of individuals 

 were 30 mm. in length, whereas at 1500 metres the 

 majority were 60 mm. .\t a depth of 500 metres we only 

 came across two that were more than 50 mm. in length. 



The smaller and younger individuals, of a length of 

 20-30 mm., live, accordingly, to a preponderating extent, 

 1000 metres higher up in the water-layers than the 

 majority of the largest and oldest individuals. 



Another remarkable fact which strikes us when we study 

 the table is that the average size of individuals is much 

 smaller at the same depth in the southern than in the 

 northern section. 



C. signata resides in an intermediarj- layer, with maxi- 

 mum in the number of individuals at about 500 metres. 

 In the case of this species, too, we note that the younger 

 individuals are mainly to be found high up in the water 

 (notice particularly the southern stations), and that the 

 same size is to be found deeper in the southern section 

 than in the northern. 



We have a remarkable parallel to the areas of vertical 

 distribution of these two fish species in the case of the 

 red-prawn species. These latter unite with the black 

 fishes in forming a populous and characteristic " com- 



