CRUSTACEA MALACOSTRACA. III. 2 2I 



pconiscus, shows some relationship to that genus, especially in the antennulse, the two anterior pairs 

 of thoracic legs and the ventral cleft. But having no subadult female I am unable to interpret the 

 anterior pairs of lamellae with certainty. The male differs much in several features from all males of 

 this family hitherto known; especially the extreme slenderuess and the antennulse are interesting. 

 The first larval stage, taken from the incubatory cavity, shows fine characters in outline and length 

 of some appendages and seta;, but no structural feature of peculiar interest. 



Occurrence. The animals are found in the marsupium of Cumacea belonging to the genera 

 D/as/ylis and Hemilamprops, and inhabiting depths of several hundred fathoms. Formerly (1897) I 

 published descriptions of several parasitic Copepoda of the family Choniostomatidse living either in 

 the marsupium or in the branchial cavity of Cumacea. Thus members of this order are infested with 

 parasites of the same two families as were previously known from the marsupium of Isopoda and 

 Amphipoda. — The specimens I have found in different species of hosts seem to belong to the same 

 species. 



157. Cumoechus insignis n. sp. 

 (PI. XVI, figs. 5 a- 5 k). 



Female. Scarcely anything of interest can be added to the generic description. The animal 

 figured (fig. 5 a) is 2 - 5 mm. long and 2 - 8 mm. broad. 



Male. Body (fig. 5 c) somewhat less than five times as long as broad. Head nearly longer 

 than broad, much produced, with the front end narrowly rounded. — First joint of the antennulse (fig. 

 5 e) produced extremely backwards, so that it is several times broader than long; its anterior angle is 

 somewhat broadly rounded, and the anterior margin is somewhat short and very oblique; the outer mar- 

 gin, counted from the front angle to the first comb-tooth, has its anterior half deeply concave; the 

 posterior margin, which is very oblique, has four very long and slender comb-teeth directed backwards 

 and a little outwards, and a fifth still longer and distinctly stronger process from the posterior end; 

 the intervals between the four first-named teeth are narrow incisions, but between the last tooth and 

 the terminal process is found an interval as if an intermediate tooth had been lost, but instead of a 

 tooth a seta is found; the inner margin of the joint is extremely long and very feebly convex, with 

 its short anterior part straight, and behind this part nine or ten processes are found; these processes 

 are extremely oblique and acute, and increase in length but not in breadth from the anterior first tooth 

 to the eighth or ninth, which is long. Second joint is directed somewhat outwards and much back- 

 wards; it is long, more than three times as long as broad; its terminal margin is divided into four 

 acute processes increasing much in length from the first to the fourth, which is long and somewhat 

 robust; the lower surface of the joint has, before the base of the posterior process another, moderately 

 long, acute process, and a tooth in front of the base of the last-named process. Third joint, excepting 

 the rami, not visible from below, but it has numerous very long sensory filaments (omitted on fig. 

 5 e, but shown on fig. 5 d), and the posterior ramus is very long. — The antennae (figs. 5 c and 5 d) 

 reach beyond the middle of fifth thoracic segment; the terminal margin of second, third, and fourth 

 peduncular joints is divided into teeth. -- As to the thoracic and abdominal segments and their appen- 

 dages scarcely anything needs to be added to the description in the diagnosis of the genus. 



