2IO DISCOVERY REPORTS 



limb and the limb behind. The whole of the filter apparatus is supported by the carapace 

 which completely covers the trunk limbs. 



In Nebaliopsis the trunk limbs project obliquely forwards, their endopodites being in 

 a line with the rest of the limb. The limbs are wide apart and the median space between 

 them is open ventrally. The eighth trunk limbs are without setae. The second to seventh 

 trunk limbs bear setae which form a complete wall to the median space but the setae 

 of one limb are not hooked on to those of adjacent limbs. Further, the exopodites are 

 minute and, with the exception of the first trunk limb, the epipodites do not cover the 

 spaces between successive limbs. The carapace, in an average specimen, extends back 

 only to about the fourth trunk limb. In all specimens the lateral margins of the carapace 

 are wide apart and not close together in the middle line as in Nebalia. 



There are thus very few points of resemblance between the two forms, but I believe 

 it is possible, with a fair degree of certainty, to suggest the mechanism by which 

 Nebaliopsis obtains its food. 



Nebaliopsis is, I believe, entirely a filter feeder. The molar process of the mandible 

 is represented by a small soft protuberance (Thiele, 1904, p. 21) which must serve as 

 a lateral lip rather than as a masticatory process. The whole mouth armature is ex- 

 tremely soft and quite unsuited for dealing with large food particles. In addition, there 

 is a complicated structure which, in my opinion, must be a filter. 



The consideration of the feeding mechanism can be divided into two parts. The action 

 of the second to seventh trunk limbs, and the action of the maxilla and first trunk limbs, 

 which I believe represents the actual filtratory mechanism. 



The eighth trunk limbs, from their separation from the more anterior limbs and 

 from the fact that they bear no setae, most probably take no part in the feeding process. 



In this analysis I make two assumptions. Firstly, I assume that the typical trunk 

 limbs exhibit a metachronial rhythm in their movement of the type that I have de- 

 scribed in detail for Nebalia (1927, p. 357) and Chirocephaliis (1928, p. 816). That is, 

 every limb commences its backstroke just before the limb immediately anterior to it. 

 I have previously suggested (1928, p. 816) that all Crustacea exhibit primitively this 

 type of movement, and I see no reason for assuming that Nebaliopsis is, in this respect, 

 different from Nebalia. Secondly, I assume that the typical trunk limbs depend to a 

 great extent on their turgidity, that is on their blood-pressure, for rigidity. I believe 

 that this assumption is justified from the fact that the distal two-thirds of each limb is 

 devoid of musculature. Also the cuticle is so thin that it is difficult to imagine any 

 movement being possible in so large a limb without some internal pressure to inflate it. 

 The trunk limb, in these respects, is markedly diff"erent from the powerful swimming 

 pleopods which are full of musculature and carry a comparatively thick cuticle. 



In Fig. 5 C I have drawn an outline sketch of the seventh pair of trunk limbs seen 

 ventrally and a median aspect of one of these limbs to illustrate the musculature. It will 

 be seen that each limb articulates on a rigid boss or prominence from the body wall (see 

 also Figs. I, 3). The limb contains two chief muscles. One comparatively powerful 

 muscle is attached in the limb close against the ventral body wall at the base of the first 



