318 CRUSTACEA. 



us that there is only a distant relationship between this group and the Cope- 

 poda, and, taking into account the Cypris larva, we are justified in assuming 

 a derivation from a primitive Phyllopod with a bivalve shell. In these points 

 we agree with the view of Balfour {Text-book of Comparative Embryology, 

 Vol. I., p. 509) and Fowler (No. 43). 



Among the Malacostraca, the Leptostraca which, besides the now living genus 

 Nebalia and its relations, probably comprises a series of fossil forms such as 

 Ceratocaris, Dictyocaris, and Hymenocaris, occupy the most primitive position 

 and show morphological characters which connect this group directly with the 

 Phyllopoda. Nebalia, for the accurate knowledge of whose structure and 

 the definition of whose systematic position we are indebted to Claus, shows 

 decidedly the Malacostracan type in the division of the body into regions, 

 in the structure and number of the limbs, and in many points of internal 

 anatomy (presence of a masticatory stomach as part of the stomodaeum, number 

 and distribution of the hepatic tubes), so that there can be no doubt about 

 its connection with the Malacostraca. Even the eight similarly shaped pairs of 

 thoracic appendages, resembling the Phyllopodan limbs, are related to those of 

 the Malacostraca by the number of joints in the endopodites. Contrasted with 

 this, the presence at the posterior end of the abdomen of a supernumerary limb- 

 less segment is of smaller importance. Among the features which connect the 

 Ncbalidae with the Phyllopoda and thus constitute them direct descendants 

 of the hypothetical group of primitive Phyllopoda must be enumerated : the 

 presence of a long heart with a large number of venous ostia (four large and 

 three small pairs), the very primitive structure of the ventral chain of ganglia, 

 the maxillary ganglion having, as in Branchipus, remained separate, the flat, 

 lamellate form of the eight pairs of thoracic limbs which recalls the Phyllopoda, 

 and in which the separation of maxillipedes from the ambulatory or from the 

 swimming limbs has not yet taken place, the presence of a large bivalve shell 

 closing by means of a special adductor muscle, and finally the possession of 

 two long, independently movable furcal processes strongly resembling those 

 of Branchipus. 



Among the other Malacostraca, the Stomatopoda occupy an unusually isolated 

 position. We evidently have here to do with a stock which separated very 

 early from the primitive Malacostraca. While the form of the heart, and 

 perhaps also the arrangement of the hepatic tubes, indicate a primitive 

 condition, there is a development in many other directions of new and 

 evidently independently acquired characters. 



The principal stock of the Malacostraca, on the other hand, is to be derived 

 from the Leptostraca through the Schizopoda, among which, again, the 

 Buphmuiidae must be regarded as the most primitive group. Their primitive 

 character can chiefly be recognised in the shape of the thoracic limbs, which 

 are all developed as biramose swimming limbs, and more or less resemble one 

 another. The view that the Decapoda are descended from Schizopoda is 

 supported eh idly by the presence of a Schizopoda-like stage in the metamor- 

 phosis of many Decapods. Among these the Penaeidea, to which also the 

 cidea are nearly related, approach the Schizopoda most closely. The other 

 Decapoda seem to lie derived forms of the ontogenetic series proceeding from the 

 Bchixopoda. The Brachyura, which are connected with the Macrura through 

 many tranaitionary forms known as Anomura, must be regarded as the most 

 highly developed but most specialised group of this series of forms. 



A second scries of forms proceeding from the Schizopoda leads through the 



