200 On the Mechanism of Aquatic Respiration. 



The ultimate laminae (fig.4y,^) are arranged in a dense parallel 

 series on either side of the secondary processes. They present 

 the same disposition as those ot* the gill of the Crab. Those of 

 the latter, however, are comparatively stiff leaves ; those of the 

 former are contractile and flexible in the highest degree. The 

 tissue of the Cephalopod gill is extremely extensile, that of the 

 Crab is fixed and passive. 



It is very difficult to obtain a satisfactory view of the fully- 

 unfolded fiat surface of one of these ultimate laminse ; but with 

 care, it may be obtained. Each lamina is constructed very much 

 on the plan of that of a Pectinibranchiate gill. It is a leaflike 

 structure, bearing two strata of vessels (fig. 6 a, b), an afferent 

 and an efferent, which double the one into the other at the free 

 margin {c, c). The aerating current laves thus the advancing and 

 the returning capillary streams of blood. The ultimate blood- 

 channels (figs. 5, 6) bear an exact resemblance to those already 

 described in the gills of the Pectinibranchiata. They are parallel, 

 non- dividing channels, of unvarying diameter. At the point of 

 curving (c,c), or the free margin, they do not dilate, as in the case of 

 the Lamellibranchiata. From all others they are distinguished 

 by the circumstance of their extreme distensile and contractile 

 capabilities. If these ultimate vessels united and divided in 

 a retiform manner, they would approach to the plan observed 

 in the branchiae of the Fish. They form, however, a straight 

 parallel series, laminarly blended together by delicate cellular 

 tissue (fig. 6). In this character they conform to the Molluscan 

 distinctive type of structure. No instance of departure from this 

 type is known, from the Tunicater to the Cephalopod. From 

 the Pectinibranchiate standard they differ in the absence of 

 cilia. The absence of cilia is compensated in the Cephalopod 

 by the extreme muscularity of the organ. 



It is scarcely possible in the branchial organs of this class to 

 arrive at a certain knowledge as to the point whether the ultimate 

 laminae are covered or not by an epithelium. The parts are 

 so irritable and transparent, that the question cannot with con- 

 fidence be determined. That they are not lined by a ciliated 

 epithelium is quite certain ; that they are not invested by any 

 epithelium at all, is improbable ; but the existence of this cover- 

 ing cannot be demonstrated convincingly. The conclusion must 

 therefore rest on general analogy : all analogy is unquestionably 

 in favour of its presence. 



The branchiae of Loligo (fig. 3, 3^) differ from those of Octo- 

 pus (fig. 1) in a remarkable manner. In the latter genus the 

 secondary divisions [c, c) are attached to the ventral aspect of 

 the organ (fig. 2 i, i). In the former they are loose, and float 

 freely in the branchial cavity (fig. 3 6, b, & 3^c, c). They thus form 



