530 CRUSTACEA. 



tively large in some species, and quite short and slender in others ; 

 and there is, also, a gradation from species with a three-jointed 

 palpus, to those in which this appendage is almost obsolete, consisting 

 of but two very small joints. Such changes occur among species that 

 would be called true Palsemons. And it is a fact worthy of note, 

 that the length of the palpus varies with the disjunction of the fla- 

 gella of the inner antennas. If the two flagella that are conjoined at 

 base are united only for a very short distance, the palpus is long ; if 

 nearly to the summits of these flagella, the palpus is reduced to two 

 joints. Again, if the union is almost or quite through the entire 

 length of the flagella, the palpus is altogether obsolete. Hence, the 

 existence or absence of a palpus is equivalent among the Palsemonidse 

 to the existence of three or two flagella to the inner antenna ; and no 

 other essential characters, higher than those of generic value, accom- 

 pany such variations. 



It is, therefore, true, both on general and specific considerations, 

 that the presence or absence of the palpus is a fact of far less taxono- 

 mic value than the differences in the form of the crown or triturating 

 portion of the mandible. It may guide in subdividing into genera, 

 but cannot be used for grouping the genera themselves. 



The differences in the mandibles lead to a subdivision of the groups 

 into families. The slender incurved form of the mandible of Crangon, 

 Nika, and Gnathophyllum, is of a wholly different character from the 

 stout straight form and broad oblique crown of Atya, or the two- 

 branched summit of the mandible in Palaemon, Hippolyte, and Al- 

 pheus : and the science is much indebted to De Haan for bringing for- 

 ward these characters in his classification. 



Besides the characters based on the mandibles, there are others 

 which bear on the arrangement of the Caridea. 



The development of the two outer pairs of maxillipeds into slender 

 legs, which is sometimes observed, is a character of great value. It 

 is commonly true that the outer pair is pediform, and thus the 

 Macroura show their low rank ; but when a second pair is also pedi- 

 form, it marks another step of degradation, evincing a further diffusion 

 of the forces posteriorly, along the ganglionic cord. This peculiarity 

 is a source, therefore, of important distinctions among the species. 



The presence of the palpus of the thoracic legs, and its enlarge- 

 ment to a natatory appendage in some species, a prominent charac- 

 teristic of the Schizopods, can hardly be employed in subdividing the 



