XiPHOCEPHALUS ARMATUS. 121 



naturelle» in Paris were all Xiphocephalus ar m a t us, after my 

 diagnosis; one of these specimens, that marked aO.vycephalus armatus, 

 MiKNE Edwards, Quoij et Gaimanli)^ and therefore most likely his ori- 

 ginal specimen, has been the type for the diagnosis and description 

 I have given here, and from it the drawing has also been made 

 (PI. VI, fig. 1). 



Adams and White gave no specific description, bnt a good draw- 

 ing, the first published of this species. A copy of it on a reduced scale 

 is given below (PI. VI, fig. 14). 



Spence Bate in 1858 proposed the new generic and specific name 

 Macrocephalus longhvstris, taking for his type the very specimen that 

 was the original of the drawing given in 1848 by Adams and White. 

 This being the case there is not the slightest doubt as to the identity 

 of Xiphocephalus armatus with Macrocephalus longirostris. 



The description given by Spence Bate in 1862 is very good, and 

 points out some valuable characteristics, as for instance the carpal pro- 

 cess of the first pair perœopoda having a secondary tooth on the hind 

 margin, the sixth pair being shorter than the fifth, and the second pair 

 of uropoda not being longer than the last ural segment. But with respect 

 to the first pair of uropoda he is in error, as the original drawing of the 

 animal, which was made at the capture, clearly shows (PI. VI, fig. 14); 

 and as Adams and White as early as in 1848 stated that the specimen 

 then was in bad order there is no wonder that the later drawing and 

 description are somewhat incorrect. 



Th. Street's description of the species, published in 1878, is 

 valuable, as it records two good specific characteristics, not appreciated 

 before, namely the form of the under part of the seventh perseonal seg- 

 ment, and the dilation of the femur of the fifth and sixth pairs of 

 perseopoda; the accompanying drawing is not good. His contention 

 for maintaining a Rhabdosoma longirostre, which is based on the incorrect 

 figure in Spence Bate's Catalogue, pi. 54, fig. 6, has no great weight, 

 as I have said above. We must rely rather upon the original drawing 

 from the fresh specimen than upon the later figure made from the same 

 specimen in a bad state. 



Glaus in 1871 (S, p. 155), 1879, and 1887 argued that Bhab- 

 dosoma Whitei was the male form of Rh. armatum^ and thus his descrip- 



Nova Acta Reg. Hoc. Sc. Ups. Ser. III. 16 



