SYSTEMATIC ACCOUNT 173 



fig. 13) shows that the length of the two somites together is undoubtedly the greater, and in this it 

 agrees with the specimens in the Discovery collections. 



The form of the coxal joint has already been dealt with under the characteristics of members of the 

 subfamily; the abdomen is illustrated in Text-fig. 130, the only point in which it differs from the 

 description of the type specimen is in having the middle part of the fourth segment fused with that of 

 the telson. 



The antenna (Text-fig. 13^) is considerably longer than the antennule (Text-fig. 13/), and its 

 flagellum consists of fifteen joints, the first of which is the longest. In Nierstrasz's specimen the 

 flagellum, which he says was probably broken, exhibited eight joints. The form of the cutting edges 

 of the mandibles (Text-fig. 13/;, /) does not agree with Nierstrasz's description. The left mandible 

 has, in addition to the five teeth on the lateral edge and four on the median, nine curved spines. 

 The right mandible is very similar to that of M. stebbingi, the median lobe, which is said to be straight 

 in the type specimen, is finely crenulated and eight curved spines are present on its inner side. 



The inner lamella of the maxillula (Text-fig. 13 c) which is missing in the type specimen, is slightly 

 expanded distally and bears two plumose setae; the maxilla (Text-fig. 13^/) differs from the type 

 specimen in having three instead of five spines at the distal end of the outer lamella. 



The maxilliped (Text-fig. 13^) is very similar to that of M. stebbingi, and has, like that species, four 

 joints to the palp. The modification of the second pleopod in the male is shown in Text-fig. 1 3 b, 

 the form of the uropods in Text-fig. 13 a. 



Distribution. The type specimen was collected by Dr J. H. Kruimel from waters off Punta 

 Arenas on the east coast of the southern tip of Chile ; the specimens in the Discovery collections come 

 from an area between the Falkland Islands and the mainland of South America, but considerably 

 nearer to the latter. 



Family Pseudidotheidae, Ohlin 



The members of this family, consisting of three genera Arcturides Studer, Pseudidothea Ohlin and 

 Holidotea Barnard, are intermediate in position between those of the Idoteidae on the one hand and 

 the Astacillidae on the other, that is to say, they possess the idoteid shape and the form of the first 

 four pairs of pereiopods with the astacillid uropod and modified first male pleopod. 



The diagnostic characters of the family are given by OhUn (1901, p. 274), and reference is also made 

 to them by Barnard (1920, p. 381), Nordenstam (1933, p. 113) and Hale (1946, p. 168). 



Ohlin (1901), in his diagnosis of the family, stated that the peduncle of the antenna consisted of four 

 joints. Studer (1884), in his description of Arcturides cornutus, also gave this number; in Holidotea, 

 however, there are five (Barnard, 1920). Nordenstam (1933, p. 113) considered that in Pseudidothea 

 the peduncle is better described as consisting of five joints, but that the first joint ' is indistinctly 

 marked off on the ventral side from the second ' (cf. Ohlin, 1901). I have examined specimens of both 

 P. bonnieri Ohlin and P. scutatus (Stephensen) and find that five joints are present. Hale (1946) 

 described a new species, Arcturides tribulis, which is very similar to A. cornutus Studer, and he gave 

 the number of joints of the peduncle as five. In the Discovery collections in another species, A. acu- 

 minatus, there is a five-jointed peduncle. It would appear that Studer's observation on this point is 

 incorrect, and that a five-jointed peduncle to the antenna is common to all members of the family so 

 far described. According to Hale (1946, p. 168), the only characters separating the genera Pseudidothea 

 from Arcturides are (i) the number of joints to the flagellum of the antenna which are two in the 

 former and three in the latter genus ; (2) the ' secondary ramus ' of the uropod is as long as the ' lateral 

 ramus ' in Arcturides, whereas in Pseudidothea the ' secondary ramus ' is only about three-quarters the 

 length of the ' lateral ramus '; and (3) the coxae are distinctly marked off on the second to the seventh 



