GIBSON — CEPHALOCHOEDA : " AMPHIOXIDES." 227 



endostyle and club-shaped gland in A. valdivice is entirely unrepi'esented in A.pelagicus ! 

 But for all the " Zurtickschiebung " of the myotomes, greater extension of the mouth, 

 and shifting of anterior gill-slits onto the right wall of the phai-ynx in the former, the 

 morphological relations of organs in the mouth-region are essentially the same as in 

 A. pelaffictis, and it is hard to believe that segments attain a partial development in one 

 which are altogether suppressed in the other. Still more improbable is it that a variable 

 number of such segments should be developed. Their development is rendered possible, 

 according to Goldschmidt, by the elongation of the movith, causing the ventral ends of 

 the anterior myotomes to be carried ])ack; the ventral mesoderm sharing tliis transference 

 leaves room for the development of that belonging to the suppressed anterior segments. 

 Now the actual number of gill-slits found in relation to the mouth is constantly 9, and 

 probably the total number of 34 is equally constant in animals which have developed 

 their full number : it is therefore well-nigh imj)ossible to believe that a variable number 

 of these slits are pre-somital. Everything points rather to the conclusion that the gill- 

 slits of A. valdivice are not eumetameric. 



I may mention that I did not, except in a few cases, see the "halbmondformige 

 Falten" — segmental extensions of the fusion between the body- wall and "pars nutritoria" 

 of the pharynx — upon which Goldschmidt relies as indications of the strict intersegmental 

 arrangement of the gill-slits. 



The lateral folds separating dorsal and ventral portions of the pharynx are seen 

 to be practically obliterated in sections of an expanded specimen. In the fully 

 expanded condition they probably disappear altogether. Since, in life, the expanded 

 condition is, in all probability, the normal one, the function assigned to them by 

 Goldschmidt, that of preventing the fall of food-particles into the " pars respiratoria," 

 becomes somewhat doubtful. Willey has suggested that they may represent the peri- 

 pharyngeal bands of Amphioxiis. There can be little doulit that this is the true homology 

 of the bands of differentiated cells which occur on the folds ; these, however, are not 

 absolutely co-extensive with the folds themselves, and the latter have probably a purely 

 mechanical significance. 



Goldschmidt describes in A. pelagicus two bands of thickened epithelium, seen in the 

 wall of the "pars nutritoria" in the region of the most posterior gill-slits, passing back into 

 the similar epithelium of post-pharyngeal gut-wall. Very similar thickenings occur in 

 A. valdivice, but are here seen to form direct continuations of the tracts of cells which 

 we have regarded as peripharyngeal bands ; their position is more ventral than in 

 A. pelagicus. 



The blood-vessels in relation to the pharyngeal wall may be alluded to here. 

 Goldschmidt calls attention to the fenestrated character of the vena subintestinalis and 

 branchial artery. This is especially conspicuous in the anterior pharyngeal region of 

 A. pelagicus, where the latter is seen in sections as a number of separate vessels lying 

 side by side in the right wall of the " pars respiratoria." Anteriorly, in both species, 

 the branchial artery is continued into a plexus of vessels supplying the endostyle and 

 club-shaped gland ; in front of these organs they appear to be collected into a single 

 large trunk which passes upwards ventral to the dorsal portion of the pre-oral gut (vide 



