495 



somewhat shorter than the penultimate one, flagelhini about the length of 

 that joint, and 7-artieulate. Anterior gnathopoda fully as large as the 

 posterior, and, like the latler, having the carpus comparatively small, and 

 produced below to a narrow setiferous lobe, propodos very large and 

 considerably expanded distally, palm scarcely longer than the hind margin 

 and slightly curved, being defined below by an obtuse angle armed with 

 several spines, one of which is rather elongated, dactylus long and curved. 

 Posterior gnathopoda with the propodos scarcely larger than that of the 

 anterior ones, and exhibiting a rather different form, being oblong oval, 

 broadest at the base, and gradually tapering distally, palm very oblique and 

 much longer than the boldly-curved hind margin. The 2 anterior pairs of 

 pereiopoda but little longer than the gnathopoda, and having the 2 outer 

 joints about equal-sized; the 2 succeeding pairs extremely slender and elongated, 

 those of penultimate pair attaining nearly ^h of the length of the body, both 

 having the basal joint compai'atively small and narrow, and the carpal joint 

 nearly twice as long as the meral one; last pair scarcely reaching to the 

 end of the carpal joint of the preceding pair, and having the basal joint 

 somewhat expanded in its proximal part. Last pair of uropoda rather large, 

 rami much longer than the basal part, and oblong oval in form, being 

 edged round by strong, ciliated setae. Telson very small, nearly twice as 

 broad as it is long, cleft rather wide and angular at the bottom. Colour 

 in the living state of the animal not yet stated. Length of adult 

 female 35 mm. 



BcmaiJiS. — As above stated, this form, first detected by Prof Loven 

 in the lake Wetter in Sweden, has hitherto been recorded as only a variety 

 of G. loricatus Sabine, and was described and figured as such by the autlior 

 in his work: «Histoire naturelle des crustaces d'eau douce de Norwege.» 

 Though it has, most probably, originally developed itself from that 

 arctic form, it exhibits at present, on a closer examination, several very 

 striking differences, and may therefore, with equal reason, be regarded as a 

 distinct species, as was done with the sjiecies of Mysis occurring in the 

 same lake, and now generally termed M. rclicta, although it is quite certainly 

 a descendant of the arctic form, M. ocidala of Fabricius. Li order to 

 indicate the paralellism between the 2 above fresh-water crustaceans, as to 

 their genealogical relation to arctic marine forms, I have felt justified in 

 choosing the very same specific name for the Amphipod here treated of, as 

 that proposed by Loven for the fresh-water Mysis. I have had an opportunity 

 of more closely comparing this form with 2 fine specimens of G. loricatus in our 

 Museum, collected off Novaja Semlja, and have foTind the following differences : 

 The arctic form attains a much larger size, being in one of the 2 specimens 



