MALACOSTRACAN FAUNA OF THE MEDITERRANEAN. 301 
carpal process reaches to the extremity of the posterior margin 
of the hand, as quoted by Norman, but that it goes beyond it 
(‘“raggiunge ed oltrepassa”’)*. This I have never seen in any 
British specimen, but it is the case with one of the Hyéres speci- 
mens, while in the other it just reaches the palmar angle, as is 
the case with many British examples. On the other hand, I 
have a specimen from Jersey in which the process only reaches 
to the middle of the posterior margin; while in another, taken at 
the same time and place, it covers about two-thirds of it. Other 
specimens (from N. Wales) have the process longer in various 
degrees; and it may therefore be taken to be an unreliable 
character for purposes of classification. 
Besides the difference in the carpal process, Della Valle says 
that the hand of the 2nd gnathopods in A. neapolitanus is much 
wider (“molto piu larga”) than that of the Ist, while in A. brunneus 
they are almost alike except that the 2nd is larger. The figures 
correspond with the description. I find, however, that this, as 
well as the extent to which the palmar angle is rounded off, is 
another variable feature. 
When A. melanops was published as a species, 7. e. before the 
discovery of the intermediate forms, the three species were 
easily definable as follows :— 
( (1) the middle of the post. | 
margin of the propodos. J 
(2) beyond the middle but 
not beyond itsextremity. 
| (3) beyond the extremity ; | 4. neapolitanus, D. V. 
~ hand very wide. 
A. brunneus, Della Valle. 
Carpal process of | 
2nd gnathopods < 
reaching to 
A, melanops, Walker. 
Finally, it is a question whether all three species should not 
be un‘ted to A. Marionis, Stebbing (‘ Challenger’ Amphipoda), 
notwithstanding the immense distance between their respective 
habitats. 
GITANA Sarsit, Boeck. 
St. 4a, 6. Five specimens. 
Fam. Cressipz, Stebbing. 
CRESSA DUBIA (Bate). 
St. 3. One specimen. Length 15 mm. 
* It is also more slender and pointed than in A. brunneus and A. melanops, 
as correctly figured by Della Valle. 
