NEBALIOPSIS TYPICA 219 



and, more important, on to the boat, without being crushed, and further, if it is subsequently handled 

 and preserved by an expert, such specimens as my F 2 and the one I illustrate in this paper will be 

 obtained. 



THE FILTRATORY FEEDING MECHANISM 



In Nebaliopsis I described in my earlier report (1931, p. 210) a fihratory mechanism unique amongst 

 the Crustacea, for it was based on the joint action of the maxilla and the first trunk limb. I gave 

 evidence that this type of filter-feeding must have evolved from that of Nebalia, which I had previously 

 shown (1927) resulted from the combined action of all the trunk limbs, the maxilla being minute and 

 taking no part in the filtering process. I suggested (1931, p. 216) that along the lines leading to 

 Nebaliopsis the ancestral form ' developed the maxillary-first trunk limb filter mechanism, at first, to aid 

 its more posterior trunk limb filter. . . . Then, when this became sufficiently advanced, it opened up the 

 carapace — maybe to allow a greater inflow of water on to the maxillary region. . . . The maxillary filter 

 now became the chief feeding mechanism. . . . As it developed so the carapace widened out and water 

 came to be sucked in from all directions. The trunk limb filter was then almost abandoned, the trunk 

 limbs swinging forwards to act as a subsidiary mechanism supplying water to the . . . maxillary-first 

 trunk limb filter.' 



Linder (loc. cit. p. 30) accepts my description of the new filter apparatus, but considers that in 

 addition the trunk limbs together still form an efficient filtering mechanism. He even goes further 

 (loc. cit. p. 31) and suggests that the filtering power of Nebaliopsis is more effective than that of Nebalia, 

 a point which I most strenuously deny. The filter process of Nebalia is, as far as I can judge, the most 

 efiicient of all those crustacean feeding mechanisms that I have studied, for water carrying suspended 

 food particles is sucked into a filter chamber just as if it were being sucked into a cylinder by a movable 

 piston, and after being filtered is passed to the exterior through a valve as efficient as a rubber gas valve. 



The filter chamber of Nebalia into which the water is sucked is the median ventral space between the 

 trunk limbs. The entrance lies anteriorly in the mouth region. Posteriorly the chamber is closed by the 

 eighth trunk limbs uniting medially to form a wall. Laterally the trunk limbs form its walls, the spaces 

 between the limbs being spanned by continuous sheets of filter setae. Dorsally it is roofed by the mid- 

 ventral body wall, while ventrally there is a complete and thick floor formed by the endopodites of the 

 trunk limbs which recurve sharply backwards and slightly inwards so that their tops touch in the 

 middle line. Thus the filter chamber is a simple laterally compressed space with but one entrance, and 

 that is relatively small. It is a slit extending from the lower edge of the labrum to the 'elbow' of the 

 first trunk limb. Now the first trunk limb is itself small — it is only about two-thirds the length of the 

 middle trunk limbs which are the main limbs acting as pistons sucking water into the filter chamber. 

 Thus we have a relatively large filter chamber with a small opening into it. Clearly suction will be very 

 powerful, so that once water has been sucked in it will not escape out again through the same opening — 

 it must remain to be filtered. Obviously then any enlargement of this entrance will lead to a diminution 

 in the force with which the water is sucked in. In Nebaliopsis the whole floor of the filter chamber has 

 opened up by the disappearance of the long recurved posteriorly projecting endopodites of Nebalia. 

 Hence, even if the trunk limbs were acting as in the latter form, the suction into the filter chamber 

 would be relatively weak. However, it is wrong to call this median space between the trunk limbs of 

 Nebaliopsis the filter chamber. Apart from the fact that I showed that another filter chamber had 

 developed between the maxilla and the first trunk limb, this space, now that it is completely open 

 ventrally, is directly comparable with the mid-ventral space of a branchiopod. Thus, again supposing 

 that the trunk limbs were still filters, as they are in Nebalia, the efficiency of their combined efforts in 



