2 THE VOYAGE OF H.M.S. CHALLENGER. 



while in others they care absent from the pereion, or attached to the pleon also ; conse- 

 quently nearly every carcinologist who has attempted to construct a natural classification 

 has made use of characters founded upon the branchial apparatus. 



The broad division in the general structure of the branchial organs has long been recog- 

 nised, and its full value appreciated. Dana ' says, "The branchial system is one from 

 which we should particularly expect important distinctions and valuable characteristics of 

 the highest significance, and such distinctions exist. They are. at the basis of some of 

 the primary subdivisions, as exhibited in the systems of Milne-Edwards, and to a large 

 extent also in the system of De Haan." 



It is, however, very remarkable that, with this full conviction and desire, Dana has not 

 utilised his observations beyond those of previous writers, who divided the Decapod 

 Crustacea into two groups, — one having the branchiae protected by a carapace, the other 

 having them uncovered and pendent. Dana's terms of " Eubranchiata " and " Anomo- 

 branchiata " are synonymous with " branchies cachees " and " branchiogastres," the first 

 and second orders of the Malacostraca of Latreille's earlier classification, and the Decapoda 

 and Stomapoda of his later. 



The system of De Haan is based on the arrangement of the branchiae to such an extent 

 as to divide the Macrura into two portions, separating those in which the organs consist of 

 a series of long cylindrical filaments from others in which the structure is foliaceous, con- 

 sisting of a series of leaf-like plates. 



But De Haan appears to have appreciated the numerical value of the branchial 

 character rather than the position of the plumes in relation to the general structure of 

 the animal. 



The great object of a natural systematic arrangement is to determine the internal 

 structure by external evidence, without which it appears to me no classification can be 

 perfect, especially in the future, when extinct forms must be studied in their relation to 

 existing species, and this can only be done in the Crustacea through the preservation and 

 knowledge of the harder or external parts. 



The classification of Latreille separates the Macrurous Crustacea, in which the branchiae 

 are attached to the anterior limbs, and protected by the carapace, from those which have 

 the branchiae attached to the posterior limbs, or unprotected ; that is, those in which the 

 branchiae belong to the pereion from those in which they arc attached to the pleon, or 

 absent. 



This general arrangement has been adopted by Milne-Edwards and Dana with scarcely 

 a variation in the general outline, and the subdivisions of their classifications also closely 

 correspond. Thus the "Astacini" of Latreille agree closely with the " Astaciens " of 

 Milne-Edwards, and the two tribes, Thalassinidca and Astacidea of Dana, correspond 

 respectively with two divisions of the " Astacina " of De Haan. 



1 United States Explor. Expedition, vol. i. p. 61. 



