CRUSTACEA MALACOSTRACA. III. 



terminates at least a little, and frequently far, from its end, but the excavation containing the pleopods 

 is continued behind these to the end as a channel (PI. VI, fig. 6 a; PI. VII, figs. 2 g, ah, 6g), and on 

 its bottom the two doors closing the anal aperture are generally seen. The female operculum is ob- 

 long (though never so narrow as figured by Sars; its shape affords specific characters). - - Uropods 

 uniramous, long to extremely long, with two or many joints. 



Remarks. In the shape of the mouth-parts Macrostylis is allied to Nannoniscus G. O. S., 

 but in other features, especially the thoracic segments and their legs, it occupies an isolated position. 

 The above-mentioned organs in the abdomen are unknown in other Asellota; they are evidently a 

 kind of statocysts to a certain extent analogous with the organs, or unpaired organ (see later on), in 

 the telson of at least some forms of the family Anthuridse, and with the organs in the uropods of 

 the family Mysidae. 



46. Macrostylis spinifera G. O. Sars. 

 (PI. VI, figs. 6 a 6 c; PI. VII, figs, i a i c). 



1864. Macrostylis spinifera G. O. Sars, Forh. Vid. Selsk. Christiania f. 1863, p. 219. 

 ! 1897. G. O. Sars, Account, II, p. 121; PI. 51. 



The specimen from the "Ingolf" Stat. 32 is normal and agrees with a number of specimens from 

 Skager Rak, but the specimen from Stat. 80 differs in a couple of features. We may begin with re- 

 marks on the typical form, thus a kind of supplement to the description and figures of Sars. 



The antenmtlse are 5-jointed, in the female slender with the fifth joint minute (fig. i a). Sars 

 says that they are 4-jointed, which is incorrect; according to two of his figures the antennulae in the 

 male are similar to those in the female excepting that they have three sensory filaments, while the 

 female has two. But in a male from Skager Rak the antennulae are considerably longer and 

 thicker than in the female, and fourth joint has four sensory filaments, while on the fifth, which is 

 about as long as fourth and third joints combined, five such filaments were found; in another male 

 from the last-named locality the antennulae are intermediate between those in the male mentioned and 

 a normal female. The antennae vary somewhat in length, in two ovigerous females the antenna; do 

 not respectively reach to the posterior margin of third thoracic segment and to the middle of fourth 

 segment, in both cases the penultimate joint of the peduncle is only a little longer than the terminal; 

 the long antennae have 7, the others only 5, joints in the flagellum. 



Fourth thoracic segment has each postero-lateral angle a little produced as a foot for a some- 

 what small, robust spine; the three posterior segments have the postero-lateral parts somewhat pro- 

 duced backwards and a little outwards, each bearing the leg below at the end, and on the angle a 

 somewhat small spine; Sars' figures of male and female show the spines too thick and the majority 

 not marked off, so that they afford the erroneous impression -- not contradicted in his text -- of 

 being strong processes. -- In the full-grown female without marsupium each thoracic segment has a 

 ventral, spiniform process, which is long or very long on the two posterior segments, while the pro- 

 cess on first sternite is moderately large, originating from the middle of the sternite and directed 

 downwards and somewhat forwards; in females with marsupium the processes on the two posterior 



