2 CARL ROVALLIUS, AMPHIPODA HYPERIIDEA. 



During some years of voyages in the European seas, and in the tropical parts of the At- 

 lantic and of the east Pacific, I had myself opportunities of studying and collecting 

 many interesting forms. Thus the materials I have had at my disposal have been tolerably 

 rich, and only five or six of the known generic forms have been unrepresented. 



In the "Historical account)) (part III of this treatise) I shall give a detailed exposition 

 of the previous systemization of the group; here I intend to begin directly with the system 

 such as, according to my opinion, it ought to be drawn up. In the "Morphological remarks*) 

 (part II) I shall deal with the most interesting morphological and anatomical features, 

 and the transformation of some of the organs in the different families. 



I have distributed the Hyperids into 16 families. The third family is synonymous 

 to the »Hyperines gammaroides» of Milne Edwards, the seven succeeding and the first 

 two families nearly correspond with his »Hyperines normales», and the last six with his 

 »Hyperines anormales», but as there are many transitions and gradual alterations between 

 the different families I do not think it convenient to divide the tribe into so sharply 

 distinguished sub-tribes as did Milne Edwards 1 ). If the increasing number of species 

 should make it desirable to establish any sub-tribes, they certainly ought to be three, 

 distinguished however by the different form of the first pair of antennas in the males; 

 and corresponding with the divisions A, B, and C in the diagram of the next page. 



As for the terminology, I follow that adopted in my previous carcinological papers, 

 only remarking here that I interpret the fourth joint of the first pair of antennae, com- 

 monly very largely developed, as the first joint of the flagellum. A sufficient corroboration 

 of this view is afforded by a comparison with the same organ in the genus Synopia, 

 Dana, where the appendicular flagellurn originates from the anterior margin of the third 

 joint of the antenna> 2 ). 



') For further information about this matter see the "Historical account", part III. 



2 ) See Amphipoda Synopiclea by Carl Bovallius. (Acta Societatis Scientiarum Upsaliensis. Seriei III, 

 Vol. XIII. 1886. 



