10 



AUGUST BRINKMANN. 



[REP. OF THE "MICHAEL SARS" NORTH 



thus appears to be of very considerable horizontal distri- 

 bution. The "Michael Sars" investigations show it as 

 extending throughout great areas of the North Atlantic, 

 being, however, of most common occurrence in the western 

 half of that sea. 



As to the depth at which A/, mirabilis lives, this has 

 up to now been an entirely open question. Verrill (20) 

 writes "whether they occurred at the surface or near the 

 bottom I am unable to say;" he considers the form as 

 pelagic, but bases his opinion merely upon its shape and 

 structure, which he very correctly notes as highly adapted 

 to pelagic life. Jo u bin (12) succeeded in showing that 

 the animal really was pelagic, but neither he nor subse- 

 quent investigators have been able to say anything as to 

 the depths at which it lived. 



The "Michael Sars" expedition has now, thanks 

 to the thoroughness of the methods there introduced 

 by Dr. Hjort, furnished all requisite information for 

 elucidation of this question. The plan adopted was, it 

 will be remembered, to take a large number of horizontal 

 hauls simultaneously at the same station but from dif- 

 ferent depths, these hauls being of so long duration 

 — several hours — as to assure an overwhelming quanti- 

 tative superiority of the material actually collected at the 

 depth desired, when compared with the amount which 

 might accidentally find its way into the net while being 

 hauled in. It is thus possible, by comparing the results 

 of the different hauls, to determine the depth at which 

 an organism lives. This can of course, only be ascer- 

 tained with absolute certainty in the case of species which 

 are numerously represented : fortunately, however' N. mira- 

 bilis was taken in such quantities at certain stations as 

 to furnish thoroughly adequate material for the purpose. 



The accompanying table (p. 11) shows the distribution 

 of the material among the different stations and at the 

 various depths; all hauls made at these stations are here 

 noted, including also those with negative result as regards 

 N. mirabilis. 



The depth is reckoned as equivalent to 2 /a of the 

 length of wire out. This calculation is unfortunately 

 not based upon direct measurement by self-registrering 

 apparatus attached to the nets; no such attempt was made 

 during the expedition, and the experiment would also 

 undoubtedly involve considerable technical difficulty. In 

 thus disregarding Dr. Hjorts estimates of depth, according 

 to which the depth of the catch should be set at about 

 half the length of wire out, I am acting upon my own 

 experience from this and other expeditions, which leads me 

 to believe that the true position may be most nearly 

 gauged by taking -h or possibly even a little more. To 

 reckon with 1 . 2, or even less — down to Va - - as sug- 



gested by Lea (15) would, from the observations noted 

 below, be entirely misleading. 



At Station 102 of the present expedition (depth by 

 sounding 1098 metres) a series of pelagic hauls was made 

 with nine implements, the lowest working on 1500 metres 

 of wire. According to the estimate of Hjort and Lea, 

 this net would have fished some 300 metres above 

 the bottom; the catch, however included a nudibranch 

 (Cuthonella abyssicola) l ) which is a decided bottom or- 

 ganism, and (according to 0rjan Olsen 19 p. 6) three 

 specimens of a pycnogonide, Nymphon grossipes, which 

 is beyond all reasonable doubt a bottom form. This net 

 must thus, with 1500 metres of wire out, have reached a 

 depth of 1098 metres, i. e. a little over 2 /3 of the length 

 of wire. 



In the course of an expedition made in 1914 on 

 board the research vessel "Armauer Hansen" belonging 

 to the Bergen Museum, similar observations were made. 

 At Station 3 the soundings showed a depth of 1400 

 metres, and the lowest haul in a series corresponding to 

 that above mentioned brought up bottom organism in 

 large numbers, the length of wire out being 2000 metres. 

 At Station 4, the depth was taken immediately before and 

 after the haul, showing 830 and 770 metres respectively; 

 here also the lowest net in a series, working with 1200 

 metres of wire, brought up quantities of bottom material. 

 In both cases therefore, the nets must have fished at a 

 depth equivalent to about -l-i the length of wire. 



And finally, direct measurements (Jespersen 11) tend 

 in the same direction. 



I have been at some pains to justify my estimate as 

 to depth in proportion to length of wire, this factor being, 

 as we shall subsequently see, of the utmost importance 

 when considering the distribution of the species in connec- 

 tion with the hydrographical conditions. 



We may now commence with the hauls from Stations 

 80 and 81, as shown in the table here given. The nets 

 fishing here at depths of 0, 66, 133, 200, 400 and 666 

 metres brought up not a single specimen of N. mirabilis; 

 the hauls from deeper levels however, produced 29 

 and 22 specimens respectively. This in itself warrants 

 the conclusion that the upper limit of distribution at these 

 two stations must lie at about 1000 metres and that the 

 specimens were actually caught at this or greater depths; 

 not taken accidentally while hauling in. This is rendered 

 the more probable by the size of the implements used: 

 the young-fish trawl - - a very large net — worked at 

 666 metres with negative result for both stations whereas 

 the small ' A U metre nets fishing at 1000 metres brought 

 up 7 and 2 specimens respectively. 



') According to verbal statement by Konservator Grieg. 



