46 1 Miscellaneous. 



22. Phoca minor, Cauis luarinus. Phoca major, Leo mariuus. Phoca iirsus, 



ursus marinus. 



23. Eosmarua. Bos mariuus, Vacca marina. 



E-EPTILIA. 



24. Testudo digitata. Testudo alata digitis alatis. Orocodilus. 



25. Caimanus. 



2(>. Lacerta, Lacertus. 



27. Guana, Iguan. Salamandrina. SalMmandra. Stellio, Jekko. 



28. Cordylus, Uromastix. Scincus. Chaleis, Seps, auoh Putria. Ciiamaileo. 



29. Rana. 

 m. Bulb. 



The Secon-Jary Appendage of the Upper Antemue as a Character 

 in the Amphqwda. By Alfred 0. Walker. 



From the time of Leach (1815) the presence or absence of an 

 appendage to the upper antennse in the Amphipoda has been used 

 as a character to separate not only species but genera and families. 

 As long as the antenna? were not examined too closely, and no 

 notice was taken of the appendage in a rudimentary condition, the 

 practice worked fairly well, though open to the objection that the 

 antennse are more often missing in an Amphipod in spirit than any 

 other part. But when, in conseqiience of the discovery by Stebbing 

 of the a[)pendage in certain species of Metopa in the ' Challenger ' 

 Collection (all from one locality), Delia Valle, regardless of the fact 

 tliat in most species of this genus no appendage has been found, 

 places two genera so nearly allied as Stenothoe and Metopa in 

 different families, it is time to enter a protest. Even as a specific 

 character the appendage is unsatisfactory. Young specimens often 

 have fewer joints in it than adults, and in Elasmopus soJcotrre, 

 Walker, all the males examined had the appendage, while the 

 females were without it. 



The question has been forced on my notice by the examination 

 of a species of Iphimedia in the ' Discovery ' Collection. A young 

 specimen, 15 mm. long, has a secondary appendage which is rather 

 conspicuous, being half as long as the second joint of the flagellum ; 

 while in another specimen from the same tube, but double the size, 

 the appendage is reduced to the most rudimentary condition. The 

 species is in every other respect a typical Iphimedia ; yet, according 

 to G. 0. Sars's definition of the family Iphiraedidoe, the immature 

 form does not belong to it! 



In conclusion, while we can hardly afford to dispense altogether 

 with the appendage in our diagnoses of species, yet I submit it 

 should be used with caution and only when it is conspicuous ; a 

 dift'erence in the number of joints should not be considered sufficient 

 to constitute a distinct species. 



Ulcombe Place, Maidstone. 

 July 16, 1905. 



