NEBALIA 221 



The antennae, during digging, will be drawn forwards as far as possible. This will 

 give the claws a bigger pull and so increase their efficiency for burrowing. 



Now the combined action of the eyes, rostrum and antennae will have a further result, 

 and that is, that the tip of the eye must catch on the peculiar hook on the hinder margin 

 of the second joint of the antenna. The hook is so placed that, as far as I can see, the 

 eye cannot rotate backwards without catching on it and, further, it is of just such a 

 shape that the tip of the eye fits into it perfectly. It has been described and figured by 

 Thiele (1904, p. 5) but he does not suggest any function for it. It occurs only in Nebaliella. 



The eyes are thus depressed by the rostrum and, at the same time they are gripped 

 from behind by the antennal hooks. As a result they are held firmly in position to form 

 with the rostrum a rigid shield covering the anterior opening of the carapace which will 

 guard this entrance against choking by mud during burrowing. 



The function of the pleopods is probably to cause a backward current in the burrow 

 which will carry away the mud kicked up by the antennal claws. 



The three main differences between the two species of Nebaliella are the form of the 

 rostrum, the number of claws on the third and fourth joints of the antenna and the 

 setation on the outer margins of the pleopods. In all these respects the Discovery 

 specimen agreed with the description of A'^. extrema. 



The antennal flagellum of A^. extrema has not been figured, but, from Thiele's 

 description (1905, p. 63) it is to be presumed that it agrees with that of N. antarctica 

 (Thiele, 1904, PI. i, fig. 8). The Discovery specimen, however, shows a marked differ- 

 ence from Thiele's figure. He describes the flagellum as of thirty-five joints, of which 

 the second to the fourteenth, the sixteenth, the twentieth and the twenty-sixth carry 

 bundles of sensory setae ("Riechschlauchen"). In the Discovery specimen there are 

 about twenty-three segments of which all except the terminal five carry bundles of 

 these setae. If the segments on Thiele's figure are counted as the gaps between suc- 

 cessive bundles and the terminal part counted as five segments then there are twenty- 

 two segments, which agrees fairly closely with the Discovery specimen. It will be seen 

 from Fig. 7 A that the proximal segmentation is obscure. The diff^erence is, I believe, 

 due to the fact that Thiele's specimen was probably a male while the Discovery speci- 

 men was a female, for Claus (1889, PI. 2) has shown that in the male oi Nebalia the true 

 segments of the antennal flagellum show a secondary segmentation which greatly 

 increases the apparent number of segments. 



Nebalia longicornis, G. M. Thomson 

 Specimens of Nebalia were collected from three widely separated localities : 



