The pseudobranchial ci'rculation in Âmia calva. 123 



there were several small eminences on it that looked exactly like the 

 points of origins of small branches. 



No indication of any nerve going definitely to the pseudol)rancli 

 could he found in this series of sections any more than in the many 

 other series of younger specimens examined. The nerve that most 

 closely approaches it is the ramus praetrematicus of the glossopharyn- 

 geus, which, in 50 mm larvae, is a large nerve that passes outward 

 behind the diverticulum of the pseudobranchial chamber, across the 

 hind end of the pseudobranch and quite close to its dorsal surface. 



The hyo-opercularis artery of my earlier descriptions arises from 

 the common carotid at the point where that artery has its origin 

 from the efferent artery of the first branchial arch. From there it 

 runs forward along the lateral surface of the skull until it reaches 

 the truncus hyoideo-mandibularis facialis. There it separates into 

 two principal parts one of which accompanies the truncus facialis, 

 while the other turns upward in front of that truncus and then back- 

 ward above it, and accompanies the ramus opercularis facialis in its 

 backward course along the dorsal surface of the adductor hyomandi- 

 bularis muscle. At the hind edge of the hyomandibular the artery 

 gives off a branch which first runs upward, perforating the large venous 

 vessel that lies in the angle between the hyomandibular and the 

 lateral wall of the skull, then turns laterally and downward along 

 the lateral surface of the vessel, passes external to the opercular 

 process of the hyomandibular and is lost on the inner surface of the 

 gill cover. The main artery continues backward, with the ramus oper- 

 cularis facialis, along the external surfaces of the adductor and levator 

 operculi, in the dorsal part of the gill cover. 



The only important difference in the course or disposition of the 

 hyoideo-mandibular arterial vessels of 12 mm and 50 mm larvae of 

 Amia thus is, that in the older larvae a secondary afferent pseudo- 

 branchial artery has been established, the primary afferent artery 

 diminishing, concomitantly, in relative importance, and perhaps, in 

 part of its course, entirely aborting. 



In tracing the changes that take place in embryos younger than 

 12 mm ones it will be best to begin with the youngest stage examined 

 and go upward to the 12 mm ones. 



In 6 mm embryos (Fig. 1), of which I have but a single and 

 quite imperfect series of transverse sections, there is a median ven- 

 tral vessel which connects autero-ventrally with, an organ that later 



