io6 



the end curved inwards, but I have not seen any similar thing, as these setae show the same 

 shape in P . laiifroiis as in the species of Euphaitsia. 



Distribution. — Sars stated that his specimens were from the south-east coast of 

 Australia, the Arafura Sea and off Mindanao, Philippine-Islands. 



Nematoscelis G. O. Sars. 



This genus is difficult. Sars established 4 species: A^. megalops, N. microps, N. tenella 

 and A^. rostrata. N. megalops is easily distinguished by having the long spines on the elongated 

 pair of legs projecting both from the penultimate and from the last joint, while in all other 

 species hitherto found these spines project only from the last joint. .Sars founded his N. tenella 

 and A^. rostrata on half-grown or still younger specimens; in 1905 I cancelled A'^. rostrata as 

 the young of A^. microps and redescribed A^. tenella from adult specimens of both sexes. In 1S87 

 Chun established A^. Sarsii on a specimen from the Mediterranean, but it is quite impossible 

 from the characters given to guess anything on the relationship of the animal, and if the type 

 does not e.xist the species ought to be cancelled as unrecognisable. In i S96 Chun established — 

 described and figured — N. mantis on a male taken oft" Funchal (Madeira); in 1905 I cancelled 

 it as belonging to A^. microps, but in spite of the shape of the eye on Chun's fig. i (Taf XII) 

 this interpretation is wrong: the animal belongs certainly to A"^. tejiella G. O. Sars, as according 

 to Chun's fig. 8 the proximal process of the copulatory organs has several very conspicuous 

 saw-teeth on the distal part of the inner margin, and the lower antennular flagellum is straight 

 at the base (not curved as in N. gracilis n. sp. ; see below). In 1907 Illig attempted a revision 

 of the genus, re-establishing A', rostrata G. O. Sars as a valid species, but the characters used 

 by him for separating A'^. rostrata from A^. microps are (jf no value. 



Later I began to examine the copulatory organs, and then I arrived at the result, that 

 my vast material from the three Oceans comprises 5 species, viz. A^. megalops G. O. Sars, 

 A'', microps G. O. Sars, N. tenella G. O. .Sars and 2 undescribed species closely allied to N. 

 microps \ one of these new species, for which I propose the name A^ atlantica, I possess only 

 from the Atlantic, while I have specimens of the other. A", gracilis, in the "Siboga" collection 

 and from the Pacific. The adult males of all species are easily separated by examining the 

 copulatory organs, while the adult females and especially immature specimens of three of the 

 species (A^. microps, N. atlantica and iV. gracilis) are more difficult, and half-grown or still 

 smaller specimens are sometimes impossible to determine with real certainty. — A general view 

 of the male copulatory organs in this genus is given above on p. So. 



The "Siboga" collections contains 3 species, A^. microps G. O. S., N. gracilis n. sp. and 

 ^V. tenella G. O. S. But it may be useful to give an analytical key to the males of all species. 



A. Pirst elongated pair of legs with long spines both from the penul- 

 timate and the last joint. Inner lobe of the copulatory organs with 

 the proximal process much shorter and less thick than the terminal 

 process A^. megalops G. O. S. 



