s6 DISCOVERY REPORTS 



prominent epistome, two setae on the inner lobe of maxilla i instead of several, shallower 

 side-plates, especially side-plate 2, the hind lobe of side-plate 5 deeper than the anterior 

 lobe instead of vice versa, and elongate dactyli on peraeopods 1-5. The first two of these 

 characters can be excluded from the discussion in view of the remarks made below on 

 the Discovery specimens. 



On the other hand, there is such an obvious agreement in the mouth-parts and other 

 appendages (v. infra) that the differences just mentioned would seem to be specific 

 rather than generic, with the exception perhaps of the dactyli. The enlargement of the 

 dactyli probably indicates a mode of life different from that of Eurythenes gryllus (cf . 

 Stephensen, 1915, pp. 37 and 43, and Schellenberg, 1926 b, p. 241). I would suggest 

 that this mode of life (in Katias) is predaceous on quickly moving animals, rather than 

 semi-parasitic, because the mouth-parts show no trace of degeneration as they do in 

 e.g. Chevreuxiella, Steph. If enlarged dactyli be regarded as a generic feature, are 

 Liljeborgia macronyx, Epimeriella macronyx, Hyperia macronyx also to have new generic 

 names created for them? The difference in length of the dactyli in the allied species of 

 these genera may not be as great as that between Katius and Eurythenes ; but that very 

 fact weakens the value of the character for generic purposes. 



The physiological argument was used in the Terra Nova Report (1930, p. 346) for 

 splitting up the Iphimediids, but in that case the separation of the genera was made on 

 the mouth-parts, which are not only more important physiologically, but showed far 

 greater morphological differences than the mere difference in the length of the dactyli 

 in Katius and Eurythenes. 



In the present family a great importance— possibly too great an importance — has been 

 attached to the characters of the mouth-parts. Where the characters of the mouth-parts 

 are identical, as in Eurythenes and Katius (even as regards the character of the spines on 

 the outer lobe of maxilla 1 , and the warts along the inner margin of the outer plate of the 

 maxilliped) there would seem to be no question of generic separation on morphological 

 grounds. The branchiae are said to be simple in Eurythenes, in Katius they are slightly 

 pleated. 



Yet it must be admitted that Katius has the head curved downwards (cf. Stephensen, 

 1925, p. 126) and a somewhat inflated and arched peraeon which gives it an appearance 

 not unlike that of a Lanceolid. 



Katius obesus, Chevr. (Fig. 21 and PI. I, fig. 1). 



Chevreux, 1905 {he. cit.), p. 1, figs. 1-3. 

 Stephensen, 1915, p. 37; 1925, p. 126. 

 Shoemaker, 1920 (Canad. Arct. Exp.), p. 8. 

 Schellenberg, 1927, p. 681, fig. 72; 1931, p. 16. 



Occurrence: 1. St. 78. South Atlantic. 1 juv. 16 mm. 



2. St. 101. South Africa. 1 ovig. 942 mm. 



3. St. 107. South-east Atlantic. 1 juv. 16 mm., 1 ^ 25 mm. 



4. St. 253. South-east Atlantic. 1 juv. 23 mm. (no penes, no brood lamellae). 



5. St. 288. Mid-Atlantic. 1 juv. 23 mm. (no penes, no brood lamellae). 



