222 PBOF. HUXLEl ON THE 



figures give no idea of the regularity of the plaits, or of the manner 

 in which the cells of the epidermis line the sides of the folds, which 

 in transverse sections, have the appearance of glandular cseca. 

 It is this organ which I conceive to be the renal organ, function- 

 ally, and to represent the Wolffian ducts, morphologically. These 

 ducts are now known to be formed in the higher Vertebrates by 

 involutions of the lining of that part of the peritoneal cavity 

 which lies external to the generative area. Taking the raphe in 

 AmpMoxus to represent the line of union of the lateral laminae, the 

 development of which into the walls of the " perivisceral " cavity 

 has been observed by Kowalewsky, the space between each lateral 

 half of the plaited integument and the ventro -lateral fold of its 

 side, will answer to an involution of the epithelium of the soma- 

 topleure, such as that by which the Wolfiian duct of osseous 

 fishes * commences ; and the position of the reproductive gland 

 low down on the wall of the somatopleure is in accordance with 

 this interpretation. 



On this view, the wall of the respiratory chamber of Ain- 

 pJiioxus is strictly comparable to the somatopleure of a higher 

 Vertebrate embryo. On the other hand, the cells which line it 

 and represent the peritoneal epithelium must, from the mode of 

 formation of the cavity, occupy the place of the epiblast, and re- 

 present a continuation of the epidermis. Thus the respiratory 

 chamber of the Amphioxus is an epiccele, a cavity of the same 

 fundamental nature as the atrium of the Tunicata ; and this 

 circumstance constitutes another curious point of resemblance 

 .between the Tunicata and AmpMoxtis. 



On the other hand, it is such a cavity as would be formed by 

 the growth and extensive union in the middle line of the lateral 

 prolongations of the wall of the body in Balanoglossus. 



To what does the respiratory chamber of Amphioxus answer in 

 the higher Vertebrata ? In the manner of its formation it cor- 

 responds, as I have elsewhere f suggested, very closely with the 

 respiratory chamber into which the gill-clefts open in the Tad- 

 pole, and which, in most Anura, communicate with the exterior 

 by only a single external opening on the left side of the body, 

 though there are two symmetrical apertures in the Tadpole of 

 Dactylethra. But, in its relations to the alimentary canal, and to 



* Rosenberg, " TJntersuchungen iiber die Entwickelung der Teleostier-Niere,'' 

 1867. 



1" Manual of the Anatomy of Vertebrated Animals, p. 121. 



