GIGANTOCYPRIS MULLERI 217 



THE HEART VALVES 



The pair of postero-dorsal ostia call for no special comment. 



The paired hepatic valves lead directly into the gut parenchyma. I pointed out 

 (Cannon, 193 1, p. 449) that Liiders was undoubtedly incorrect in stating that blood 

 passed into the heart through these gaps from the gut region. My sections of Giganto- 

 cypris have supported this view, for in fixation the blood cells become grouped on the 

 heart side of the valve (Plate XLI, fig. 8) showing that they are efferent valves. The tubes 

 into which the valves open are very definite vessels whose walls are formed by the 

 parenchyma cells. They can be traced backwards where, at the sides of the stomach, 

 they bifurcate, the branches becoming smaller and finally disappearing in the interstices 

 of the parenchymatous tissue. 



Both the hepatic valves and the aortic valve are associated with very distinct gaps in 

 the heart musculature. The oval, bean-shaped opening leading to the right hepatic 

 valve is clearly shown in side view in Fig. 10. The right half of the narrow oval opening 

 leading to the aortic valve is seen in the same figure. 



In Doloria I stated that the edges of the gaps are fused with the pericardial floor. 

 In Gigantocypris, however, this is not so. The valves are actually at some distance from 

 the muscular wall of the heart, the intervening gap being joined by a squat tube (Fig. 15). 

 The walls of these efferent tubes appear to consist of a very thin membrane in which 

 there are a few striated muscles running approximately in the same direction as the 

 nearby heart musculature. They are not formed from accumulations of the parenchy- 

 matous cells which in other places extend over the heart surface, such as I described 

 in Doloria. The latter form a branching network, while the walls of these efferent tubes 

 are definite membranes. The tubes are closed at the outer ends by the pericardial floor, 

 and the valves consist simply of splits in the latter. The edges of the splits show a distinct 

 fibrillation (Plate XLI, fig. 8) but without any cross striation. 



The fibrils of the hepatic valve extend posteriorly into the most lateral branch of the 

 lateral subpericardial muscle. Anteriorly they do not appear to extend beyond the limit 

 of the valve itself, but from this point there is a radiation of fibrils in the pericardial 

 floor which undoubtedly affords the anterior support for this muscle. 



The fibrils of the aortic valve, which are much more distinct and extensive than those 

 of the hepatic valves, narrow down anteriorly to attach to the front of the upper end 

 of the aorta on a level with the top of the nauplius eye. This is the same level of attach- 

 ment as in Doloria, but in the latter the attachment went right through to the ectoderm. 

 Here the attachment is to the aortic wall and the aorta is free from the ectoderm. 

 Posteriorly the valve narrows down to a tendinous connexion to the pericardial dilator. 

 In Doloria I have described this muscle as being continuous dorsally with a median 

 muscular strip of the pericardial floor which I term the median subpericardial muscle. 

 This distinction should probably not have been made, as in Gigantocypris the pericardial 

 dilator only extends a short distance above the attachment of the aortic valve, where it 

 disappears into the pericardial floor (Fig. 10). 



