104 Edward Phelps Allis jr., 



(1889) to have been first expressed by Allen Thompson, in 1831. 

 Ayees does not himself accept this proposition, at least in so far 

 as it relates to fishes; nor apparently does Eaepaele (1892), as will 

 be later explained. Dohrn, on the contrar}^, says (1890, p. 372) 

 that the development of these vessels, in fishes, definitely shows that 

 the primary dorsal longitudinal vessel, on either side, includes the 

 corresponding internal or "cerebral" carotid; and the use of the 

 word "cerebral", together with statements made on following pages^ 

 shows beyond question that Dohen considered that this primary 

 longitudinal vessel included some considerable portion of that part 

 of the internal carotid that lies anterior to the point where that 

 artery is joined by the mandibular aortic arch. And this conclusion 

 of Dohen's would seem to be confirmed by Kellicott's (1905) 

 figures of the anterior carotid arteries of Cemtodus, and his state- 

 ment (p. 140) that those arteries of that fish develop as anterior 

 prolongations of the lateral dorsal aortae; and by F. W. Müllee's 

 (1897) figures of the vessels in embryos of Lepidosteiis. Hochstetter 

 (1906, p. 85), in the most recent discussion of this subject, accepts 

 this interpretation of the vessels, apparently for all vertebrates. 



On this interpretation of the vessels, my diagrams, and the 

 conclusions deduced from them, are wholly based, and the diagrams 

 are designed to emphasize the deductions. In the diagrams, the 

 order in which the several arteries actually arise from the aorta 

 and the aortic arches has in every case been strictly adhered to^ 

 but the course and direction that these vessels take, and their 

 relative distances from each other, are all purely diagrammatic. The 

 relations of the arteries to each other, as to internal or external, 

 and anterior or posterior, are given as nearly as the descriptions 

 and figures permit, but there is no pretension that they are correct. 

 And I should here call attention to a statement made by Kellicott 

 (1905, p. 195) that "among the Ganoids (Müller, 1897, Allis, 1900) 

 the original single branchial vessel loses its connection with the 

 dorsal aorta and becomes the aiferent branchial artery, while a 

 single efferent branchial artery appears anterior to it, receiving 

 its blood through the gill-capillaries and opening into the dorsal 

 aorta". The word "anterior" is put in italics by Kellicott, and in 

 so far as Ainia is concerned, is based on the figures that I gave 

 of the aortic arches in that fish. But that particular feature of 

 my figures, while it may show the correct relations of the vessels, 

 was not based on observation. I could not readily show one artery 



