THE i'lSlIKS. 11)1 



The 'Research' made in all 128 tow-net hauls, of which 90 were either fished at, or 

 were opeu up to, the surface, while 37 were stopped by the closing of the net at depths 

 varyin!^ from 50 to 1500 fath. The tow-net hauls which closed below the surface include 

 18 -wiiich came within 300 fath. of the surface, viz. 3 closed at 300, 3 at 200, 3 at 150, 



8 at 100. and 1 at 50 fath. ; but none of these, nor any of the 19 hauls closed at srreater 

 depths, yielded evidence of the presence of larvae. 



Even ova taken in deep water often suffer from capture, as it is quite common to find 

 in the 'Helga' collections tough spinous shells of the character associated by Raffaelc 

 with J/rtc?/'»r«s, quite devoid of contents; audit is reasonable to suppose that broken 

 shells of a more delicate character might easily escape observation. Very young larvae 

 are much more delicate than ova, and even those of littoral species are usually mangled 

 beyond recognition in tow-net gatherings from shallow water. It may therefore be 

 said, without the least disrespect to the skill and care employed in the sorting of 

 'the 'liesearch' material, that the apparent absence of larvae from the sortings of the 

 deeper gatherings does not absolutely prove tiiat no young larvae came in the way of 

 the nets. 



The table given above (so far as any conclusion can be drawn from the few taken) 

 points to ova being fairly evenly distributed right up to the surface. In the case of 

 larvae it would seem that they were generally distributed between 100 and 25 fath., 

 somewhat scarcer in the upper 25 fath., and very scarce at the actual surface. The 

 paucity of hauls commencing at more than 100 fath. renders it impossible to say at what 

 depth below^ the surface the greatest comparative density of larvic commenced. The 

 genus Scopehis we suppose to be represented in 17 gatherings, and in 16 we have noted 

 the occurrence of Ilaurolinus or a very similar form (probably in all cases M. boreaUs) ; 

 but we think that many of the young or damaged larv£e to which we have not ventured 

 to give even a queried name may also be 3IanroUcus. 



Turning to the question of diurnal oscillation of larvae, the results seem to be negative. 

 It is true that the only captures made by purely surface hauls took place at night, but 



9 out of 11 night hauls were blank. In the hauls made from 25 to fath. larvge 

 were taken except at midnight, 2 .\.M., G a.m., and 8 A M. At 50 to fath. every 

 haul, except one of two made at !■ p.ii., produced larvie, and larvte were taken at 

 75 to fath. throughout tlie 21 hours. At 100 to fath. blank hauls were made 

 at 2 A.M., 4 A.M., 7 A.M., 11 A.M , 12 uoon, and 6 p.m. ; but the success and failure of 

 hauls in the hours of daylight and darkness is not in such contrast as to lead to any 

 certain conclusion. 



Taking all the superficial hauls (100-0 to 0-0 fath.) together, it is apparent that 

 relatively to the total numbers under each condition blank hauls were more frequent by 

 day than by night ; and if we interpret the somewhat sporadic catches in the hauls of 

 100-0 fath. to mean that the larvae were mostly in the upper 75 fath., there is some 

 approach to evidence of a descent during the hours of daylight. 



The fact is that the grouping of the larvae as a whole, due to our failure to recog- 

 nize the parentage of most of them, obscures the result; and we cannot treat them 



33* 



