14 Sigurd Johnsen. 
II. General Remarks. 
The normal habitat of M. glaciale in the open sea of the 
Atlantic is above great depths from the surface and downwards 
to about 500 metres depth (or perhaps more), the occurrence 
near the surface gene:ally supposed to take place at night. In 
the main this holds good also for the northern waters. 
Most of the records here are seaward of the 400 metre-line 
(see chart of distribution) and from the fjords of Western Norway 
where the depths are 300—500 metres or more. There are few 
records above the more shallow coastal plateau; we have here 
the capture of abt. 50 young-fish over the banks off Nordland 
where the depth was 200—300 metres; probably also the speci- 
mens from the ventricle of codfish (Coast of Finmark) are shore- 
ward of the 200-metre line. This is in accordance with the 
distribution west of Ireland where according to Holt & Byrne 
(1911 p. 17) most of the records are from water of more than 
400 fathoms while there is no evidence of its occurrence shore- 
ward of the 200-fathom line. 
The data from northern waters do not bear aie evidence 
of the vertical range of the species. Collections by means of trawl 
are of no use where the range of pelagic animals is considered. 
Even the method of horizontal hauls with a series of appliances has — 
its drawbacks. The actual depth of each gear is not known, only the 
number of metres of wire out. For the horizontal hauls of the “Arm. 
Hansen” I have, for reasons elsewhere stated (1921 p. 69), reck- 
oned with two-thirds of the length of wire to the gear. The 
depth of the gear gives us the lower limit of the depth of the 
capture. It must, however, always be borne in mind that the 
gear is fishing also during setting out and hauling in; by towing 
horizontally for a considerable time the possibility of the catches 
