MYSIDEA. 635 



central point near the front, as in the Caligi; the mandibles, maxillae, 

 maxillipeds, and two pairs of bifid legs were distinct, but no rudiments 

 of those posterior, neither of any abdominal appendages. 



The four antennae were nearly as in the preceding, except that the 

 first pair were three-jointed, the last not being subdivided, and the 

 multiarticulate branch of the second pair consisted of but four joints. 



TRIBE II. MYSIDEA. 



A SHRIMP-LIKE form and habit is strongly retained through all the 

 genera of Mysidea. There is, however, a slight prolongation of the 

 head in the Sceletininae, and a constriction of the carapax, across just 

 anterior to the mouth ; and this aberrant feature is excessively deve- 

 loped in Lucifer, where a long, slender segment precedes the mouth, 

 reminding us of the same in Alima. The species of Lucifer still 

 retain much of the Caridoid habit in the thorax and legs, and are 

 Mysidean; while those of Alima, true Squilloids, besides having a 

 carapax over the whole cephalothorax, are characterized by the pre- 

 hensile legs of the Squillse and Erichthi. 



There is a wide range among the species as to the legs, although 

 they are never prehensile, and also as to the branchiae. The thoracic 

 legs are either bifid, with a long ciliated outer branch, or they are 

 simple or undivided ; they are either complete in the normal number, 

 or one or two posterior pairs are obsolete. The first, second, and 

 third pairs of maxillipeds are more or less pediform, and they may be 

 either like the following legs, or the third alone may be thus similar 

 to the legs in form and direction. 



The differences in the branchial function are of a striking character. 

 In the higher species, approaching most nearly to the Macroura, there 

 are thoracic branchial appendages, uncovered by the sides of the 

 carapax, hanging in branching tufts or clusters of closed vessels from 

 the outer side of the base of several of the legs. In the greater part 

 of the species, such branchial appendages fail entirely, and this func- 



