332 DISCOVERY REPORTS 



Upogebia D.I. (Fig. 33) 



Discovery St. 91. False Bay, South Africa. 



Stage I. Length 4-15 mm. 



General form as in U. deltaura. Abdomen without spines. Telson of triangular 

 shape, with 7 + 7 spines and a small spine ventrally near anus. 



Antennule unsegmented. Antenna, exopod with small spine and ten setae ; endopod 

 stout, with three apical setae. 



Maxillule, palp of three segments. Maxilla, palp small, unsegmented, with five 

 setae ; four large inner lobes ; exopod with narrow proximal part bearing small setae. 

 Maxillipede i endopod of five segments, segment 2 with outer seta. Maxillipede 2, 

 endopod of four segments, without outer setae. Maxillipede 3, endopod rudimentary; 

 exopod without setae. 



Legs 1-3 with exopods without setae, endopods of all legs large. Pleopods large, 



on somites 2-5. 



In European Upogebia there may be four stages, but there are commonly only three, 

 so that development is slightly abbreviated. In U. savignyi an inert larval form hatches 

 from the egg (Gurney, 1937^) and moults very soon into the adult form. In the species 

 described here abbreviation of development has not gone quite so far. The antennules, 

 antennae and telson are those of a normal stage I larva, but the legs and pleopods are 

 as advanced as in the last stage of U. deltaura for instance. It is probable that the first 

 moult would, as it does in U. savignyi, give rise to the post-larva. The difference between 

 the two consists simply in the larva retaining its activity. 



Upogebia B.R. II (Fig. 34) 



Barrier Reef St. 49. 



This larva is a normal Upogebiid, except in one respect, namely that it has dorsal 

 spines on all the abdominal somites except the first. The complete absence of spines 

 from the abdomen is a characteristic feature of all Upogebiid larvae hitherto known ; 

 but it can no longer be taken as a character holding good for the group as a whole. 



LAOMEDIIDAE 



The family Laomediidae is one of rather special interest in view of the extreme rarity 

 of the species. Each of the three genera Laomedia, Jaxea and Naushonia is represented 

 by a single species, and of each very few specimens are known. The genus Axianassa, 

 Schmitt, likewise with one species only, is included in the family by De Man. 



The development oi Jaxea nocturna is well known (Cano, 1891 ; Caroli, 1924, etc.), 

 and that of Naushotiia crangonoides has been described by Thompson (1903); but 

 nothing is known of the development of Laomedia or Axianassa. I have described 

 (1924, p. 156) a larva which was assumed to belong to this family; but I have now 

 seen other specimens of the same form from the Discovery material and am convinced 

 that this identification was wrong. It appears to be a Pagurid. 



