78 REPORT — 1876. 



In the Anomura, of which TAtliodes may form the best example, the coxaj 

 may best be dissected out ; and it does not require any very extreme care to 

 separate the frame of one appendage from those by which it is compressed 

 both anteriorly and posteriorly, bj' which compression the joint partakes of 

 a quadrilateral form. The plates are in many places reduced to an extreme 

 tenuity, and practically fulfil the office of a ^^single wall, although in reality 

 they are produced by two lamellae closely compressed but not united. The 

 i7iferior or ventral wall, that forms the sternum, is very much more strong, 

 and extends until it meets the corresponding plate upon the opposite side. 

 In Lithodes this simple condition extends from the anterior to the posterior 

 extremity of the percion. 



In the Brachyura, of which we may take Cancer as the type, the walls of 

 the coxal joint form the floor of the percion from the anterior extremity to 

 the fourth or tetartosomite, from which posteriorly an upright wall in the 

 median line separates the right side from the left, and encloses the muscles 

 of the four posterior pereiopoda within as many corresponding chambers, 

 forming a strong arch that supports the internal viscera and precludes their 

 sinking into the ventral cavity. 



If, as I contend, this condition of the structure may be demonstrated 

 beyond doubt, it follows that the episternal pieces lose their homological 

 signification, as defined by Prof. Milne-Edwards, in the same way as the 

 epimera of the dorsal arc. 



The episternal plates are parts of the first or coxal joint of the legs pro- 

 duced as plates, valuable as supporting tlie articulations of the next succeed- 

 ing joint with the first. It is interesting to observe that these so-called 

 episternal plates can be traced back to large spinal processes in the young 

 animal, and to less important processes in the pupal or thii'd stage in the 

 process of the development, where they can be distinctly seen as parts of the 

 coxee of the appendages attached to the poreion (fig. 7). 



This appears to be the anatomical condition in the Brachyura, and also 

 in some of the Anomural groups. 



But in the Macrural tjpe the ventral surface of the pereion is formed of 

 the lower arc of the several somites which belong to this division of the ani- 

 mal. Some slight variations of form and apj^earance exist in separate genera. 

 In Palinurus the anterior part of the sternvim is narrow and longitudinally 

 longer than broad, while the posterior part gradually increases in width 

 from the anterior to the posterior extremity. Each somite is completely 

 fused with those with which it is in contact at the centre, while deep lines 

 of fissure define their separation on each side, the posterior process of which 

 somites corresponds analogically with the so-called episternal plates in the 

 Brachyura, but homologically they are distinct, being, in this form, parts of 

 the true somite, and not a portion of the coxa of the leg incorporated with it 

 (PL II. fig. 1). 



In the genus Astacits the sternal plates are all narrow, being scarcely 

 broader posteriorly than they are anteriorly, while in the genus Ilomanis the 

 sternal plates are still more narrow and less important. This appears to be the 

 general characteristic of the ventral plates in Nephrops, Palamon, Crangon, 

 &c., but more delicately and feebly constructed, so far as the external condi- 

 tions ; but in the lower forms of Crustacea, such as the Amphipoda and the 

 I^opoda, the sternal plates are broader than they are long, and consequently 

 the several pairs of appendages are widely separated from each other, cor- 

 respondingly so throughout the entire length of the pereion. 



The internal structure in the Podophthalmous types is more complex than 

 the same parts in the lower or sessile-eyed forms. 



