THE CRUSTACEA 25 



possessing rudiments of some of the post-naiipliar somites and 

 appendages. 



!Most of the larval forms arc pelagic in habit, and many sho^v 

 special adaptations to this mode of life, especially in the develop- 

 ment of spines and processes which are probably less important as 

 defensive organs than as aids to flotation. 



Phylogexy. 



The study of fossil Crustacea has hitherto contributed com- 

 paratively little towards a precise knowledge of the phylogenetic 

 history of the class. Although their remains are abundant in 

 nearly all formations, from the most ancient up to the most 

 recent, in very many cases only the carapace or shell is preserved, 

 the limbs being lost or represented only by indecipherable frag- 

 ments. Many important forms must have escaped fossilisation 

 altogether owing to their small size and delicate structure. Further, 

 many important groups were already differentiated when the geo- 

 logical record l^egan. The Branchiopoda, Ostracoda, and Cirripedia 

 are represented in Cambrian or Silurian rocks l)y forms which seem to 

 resemble those now existing, so that palaeontology can have little 

 light to throw on the mode of origin of these groups. In the case 

 of the Malacostraca the material is a little more promising. It is 

 not improbable that the Phyllocarida, which are found from the 

 Cambrian onwards, may include the forerunners of the true Mala- 

 costraca, liut nothing is definitel}^ known of their appendages. The 

 recent discovery, in the Tasmanian Anaqmles and the Australian 

 Koonunga, of what are believed to be representatives of the 

 Carboniferous and Permian Syncarida, has given a clue to the 

 affinities of forms hitherto problematical, and may throw light on 

 the relations of other Palaeozoic fossils hitherto vaguely referred 

 to "Schizopoda" or Decapoda. Remains of undoubted Decapods 

 are first met with in Mesozoic rocks. They are abundant in many 

 deposits, and are sometimes sufficiently well preserved to render 

 possible fairly accurate determination of their affinities. The 

 Isopoda and Stomatopoda are known from Mesozoic deposits, but 

 have hitherto yielded no results of phylogenetic importance. 



In view of the scarcity of trustworthy evidence as to the actual 

 forerunners of existing Crustacea, ]ihyl()genetic conclusions based 

 on the data of comparative anatomy and embryology remain largely 

 speculative. They are none the less a necessary jireliminary to the 

 attempt to construct a natural system of classification. 



The earlier speculations on this subject started from the assump- 

 tion that the " theory of recapitulation " could be applied to the 

 larval history of the Crustacea. The various forms of larvae, moi'e 

 especially the nauplius and the zoea, were supposed to reproduce, 



