GIBSON— CEPHALOCHORDxV : " AMPHIOXIDES." 253 



Probably the backward extensions are a purely larval character, and become obliterated 

 at an early stage. 



The true mode of origin of the pterygocoeles still remains to be observed. Macbride's 

 later view of their " pseudocoelic " origin we have already seen (p. 228) to be untenable, 

 and there can be little doubt that they are true coelonQic cavities. Their relations in 

 Aniphioxides i^elacjlcns, where they apparently do not exist in the anterior portions of 

 tlie metapleures, and are much largest posteriorly, strongly suggest that they develop, 

 like the metapleures themselves, from behind forwards. Van Wijhe's suggestion (1906) 

 that they may arise as segmental diverticula from the splanchnocoele, which become cut 

 off from it and fuse with one another, seems much the most probable that has yet been 

 made. 



Since the metapleures and their cavities extend back much farther in the adult than 

 in the larva, their posterior portions must be formed separately from their anterioi', 

 during or after the metamorphosis. In Asymmetron and Meteropleuron it is not 

 improbable that the ventral fin grows into continuity with the right metapleure before 

 the metamorphosis (in some of the older Amphioocides pelagicus it was seen to extend 

 forward very nearly as far as the branchial region), and subsequently moves up onto the 

 right side and acquires a cavity. The occurrence of apparent connections between 

 metapleural cavities and splanchnocoele at the hinder ends of the former, in the young 

 metamorphosed animal described above, suggests that these cavities are derived directly 

 from the splanchnocoele rather than from a backward growth, of the cavities already 

 existing anteriorly. The posterior part of the left metapleure, in Asymmetron and 

 Heteropleuron, and of both in BrancMostoma, must be formed in a backward direction 

 from their original posterior termination. 



It follows from the view which I have taken of the origin of the pterygocoeles and of 

 the nature of the stomoccele, that I regard the adult connection between these cavities, 

 and with it the apparent continuity of left pterygial muscle and outer lip-muscle, as 

 entirely secondary developments. Von Wijhe admits the two muscles named to be 

 independent of one another ontogenetically. I believe that they are equally independent 

 morphologically, and that their connection is of purely physiological significance. 



References. 



[Papers dealing with " Amphioxides " are indicated by *.] 



1893. Andrews, E. A. — An Undesciibed Acraniate, Asymmetron lucayanum. Stud. Eiol. Lab. Jolins 



Hopkins Univ. vol. v. 

 1892. Boveri, Th. — (i.) Die Nierenkanalchen des Amphioxus. Zool. Jahrb. (Anat.) Bd. v. 



(ii.) Ueber die Bildungstiitte der Geschlcchtsdriisen und die Eiitstehung der 

 Genitalkammei'D. beim Amphiuxus. Anat. Auz. Bd. vii. 

 *'i903. Cooper, C. P. — Cephalochorda (Systematic). J. Stanley Giirdiner : Fauna and Geography of 

 the Maldive and Laccadive Archipelagoes, vol. i. part 4. (Caiiibridge.) 

 SECOND SERIES. — ZOOLOGY, YOL. XIII. 34 



