NEBALIOPSIS 215 



distance the filter apparatus extends behind the mouth. It is of no use a Hmb filtering 

 ofl^ food particles if there is no means of transferring these to the mouth and hence the 

 posterior part, at least, of the filter would be useless unless a suitable mechanism for 

 the purpose existed. The process of transference is, I believe, carried out entirely 

 by the mandibular palps. The length of these structures is just sufficient to allow them 

 to bend back as far as the hinder limit of the filter chamber. Also, they are jointed in 

 such a way that the tips can be folded back into the mouth entrance and the tips are 

 armed with comparatively stout claw-like setae armed with a brush of setules. I believe 

 they stroke upwards so that their distal setae project through the filter setae into 

 the chamber and then inwards and forwards to the mouth, thus combing the filtered 

 residue to the mouth entrance. Here it is sucked into the mouth by the powerful dilator 

 musculature of the oesophagus. 



I have said nothing as to the actual phase diffi'erences between the oscillations of the 

 mouth parts. I assume that the trunk limbs behind the first exhibit the typical meta- 

 chronial rhythm, but exactly how the mouth parts oscillate I think it impossible to say. 



To summarize the feeding mechanism as far as I have analysed it — it is suggested 

 that the second to seventh trunk limbs, by their oscillation, produce a drift of water 

 forwards over the mouth parts. This will cause a subsidiary current along the mid- 

 ventral line towards the mouth, and particles in this current will be retained there by 

 the filter setae arming the inner margins of these trunk limbs. Water arriving at the 

 mouth region will be sucked into a filter chamber between the maxilla and first trunk 

 limb when these limbs move downwards from the body wall. It will be sucked out by 

 the backward movement of the second trunk limb, the epipodite of the first trunk limb 

 acting as a valve closing laterally the inter-limb space between the first two trunk limbs. 

 Particles filtered oft' in this way on to the filter setae of the maxilla and first trunk limb 

 are scraped off and passed forwards by the mandibular palps into the mouth. 



Nebaliopsis is without doubt the most specialized of the Nebaliacea (Thiele, 1904, 

 p. 25). Of the other members of the group I have described the feeding mechanism of 

 Nebalia from actual observation (1927). In the next part of this report I give evidence 

 that the principle of the feeding process of Nebaliella is the same as that of Nebalia. 

 Although I have not been able to examine a complete Paranebalia I believe, from the 

 published accounts of this genus, that it also feeds in a similar manner. Thus the 

 Nebalia mechanism can be taken, in general principle, as representing the primitive 

 feeding mechanism of the Nebaliacea, and it must be assumed that the method of feeding 

 of Nebaliopsis has evolved from it. 



How this has come about is very difficult to say and will remain so until new forms 

 are found less divergent from other Nebaliacea than Nebaliopsis. But, that the change 

 took place in correlation with the planktonic habits of Nebaliopsis, is, I think, certain. 

 It is possible to imagine a Nebalia adopting a pelagic habit. In fact, certain of the 

 specimens recorded in this report were collected at the surface (p. 221), but I cannot 

 conceive a Nebaliopsis living in mud. For this reason I take it that Nebaliopsis evolved 

 from a planktonic Nebalia. 



