416 Dr. T. Williams on the Mechanism of Aquatic 



of the distal margin, they become more and more separated 

 from one another (tig. 6 d) . The interposed substance becomes 

 more and more pellucid, until at length at the margin the vas- 

 cular loops {c, c) stand out in unmistakeable eminence. So mi- 

 crocosmic is the mechanism however, that it is impossible by 

 direct view to state whether these vessels are separated by a 

 water-fissure, such as that which exists in the gills of the Lamel- 

 libranchs, or whether they are joined by an intervascular mem- 

 brane. In the former case, water would penetrate into the space 

 (fig. 7 e) between the two constituent layers of the lamina ; 

 in the latter this space would be entirely closed from the 

 external element. On a superficial view this point may appear 

 very insignificant. On a deeper insight it becomes pregnant 

 with functional and homological meaning. If the water could 

 find a ready entrance into the interlaminar space (fig. 6 e, g)y 

 the column of blood flowing in each afierent and eff'erent blood- 

 vessel {a, b), that is, in the trunks bounding that space, would 

 be aerated on both sides, and the respiratory value of the organ 

 would at once be doubled. It is almost certain that it does not 

 and cannot. If the external water cannot and does not penetrate 

 into this interlaminar space, then it must be filled by the vital 

 fluid of the animal ; for the space exists beyond doubt. If this 

 latter supposition be true, it is quite certain that this fluid must 

 be distinct from that which circulates in the laminar or proper 

 branchial vessels. This doubt must for the present be left 

 undetermined. Whether it be soluble or not, enough of the 

 essential structure of the gasteropodan gill has been demon- 

 strated to establish between it and the branchise of the Lamelli- 

 branchs an extraordinary resemblance ; not a mere outline-simi- 

 larity, but a closeness, almost amounting to an identity of con- 

 structive plan. If everything else in the patelloid organism 

 exhibited the same degree of similitude to the system of the 

 Lamellibranch as that which obtains in the branchial apparatus, 

 the naturalist would not hesitate to define the Patella at once as 

 the highest Acephalan and the lowest Encephalan. It must 

 however be admitted, that the judgement of the classifier should 

 not be swayed exclusively by special affinities. The question 

 now arises, if these minute branchial leaflets of Patella really 

 consist each of two layers, how are these layers tied together ? 

 They are fixed to one another by an intermediate system of 

 threads {trabecules) crossing each other in such a manner as to 

 afi'ord the most effectual mutual support (fig. 6 A ; fig. 1 g). The 

 points (fig. 6/; fig. 7 d) to which these connecting fibres or bands 

 (fig. 7/; fig. 6 h) are attached to the vessels are swelled into 

 nodules (fig. 7 d)^ which present a singular resemblance to the 

 fleshy nodules on the branchial bars of the Mussel. Of course 



