XIX] CIRCULATORY SYSTEM 431 



the hinder part of the body. The forward extensions of the two 

 longitudinal vessels carry blood to the head and are known as 

 the carotid arteries. There can be little doubt that the impulse 

 leading to the evolution of the heart came from the necessity of 

 having a strong force to drive the blood through the capillary 

 channels on the walls of the gill-sacs. 



In the embryos of all Craniates the number of these paired 

 connections between the ventral aorta and the roots of the dorsal 

 aorta is equal to the number of visceral arches, but the two anterior 

 pairs, viz., those traversing the wall of the pharynx parallel with 

 those parts of its supporting skeleton known as the mandibular 

 and hyoidean visceral arches respectively (see p. 456), are found 

 in adult forms only as remnants in connection with the carotid 

 arteries. Whatever may have been the case in primitive forms, 

 these first two arterial arches have now no part in aerating the 

 blood, this function being performed by the succeeding pairs of 

 arches, along whose course only are gill-sacs developed. We shall 

 find that the arterial system near the heart is in all groups of 

 Craniata a modification of the arterial arches just described. 



The fore-limb is supplied by a vessel called the subclavian 

 artery, bat the origin of this differs in the several classes of 

 the phylum. In Amphibia, Lizards and Mammalia other than , 

 Cetacea, it arises from the epibranchial artery near or behind the 

 fourth visceral arch, while in Crocodiles, Turtles, Birds and 

 Cetaceans its origin is from the ventral end of the fourth visceral 

 arch. As in both Lizards and Cetaceans these two vessels exist 

 side by side, but only one of them supplies the fore-limb, it is 

 clear that the subclavian arteries are not homologous throughout 

 the group. 



Each chamber of the heart is separated from the one behind by 

 valves, which are flaps of membrane free to move in one direction so> 

 as to open and admit blood from behind, but restrained by tendinous 

 chords from being driven further back than so as just to meet when 

 the chamber contracts, and thus prevent any backward movement of 

 the blood. In the conus there may be several transverse rows of 

 "pocket valves. These valves as their name implies are loose 

 pockets of membrane which are pressed flat against the wall of the 

 conus during the forward movement of the blood, but which when 

 the conus contracts become filled with blood and swollen out so 

 as to meet one another and prevent the reflux of blood into the 

 ventricle. 



