456 DISCOVERY REPORTS 



plasm of a large muscle on the inner side of the protopodite. This muscle, which is 

 the largest in the limb, consists of a number of pillars of striated myofibrils arranged 

 in the form of a fan. Their cytoplasm, however, forms a continuous sheet on their inner 

 face, and it is into this sheet that the antennal artery can be traced as a thin slit which 

 gradually tapers and disappears. 



I have now described all the arteries which can be traced from the heart. In addition 

 to this system of vessels, blood leaves the heart by the hepatic valves and is pumped 

 directly into the gut parenchyma. This tissue is described later (p. 464). The actual 

 passage of blood through the parenchyma cannot be traced in Dolor ia, but in another 

 unnamed Cypridinid from South Georgia I traced definite sinuses extending from 

 the hepatic valves to a region under the gut. Presumably the blood forced into the 

 parenchyma oozes out through the crevices between the parenchymatous cells. 



Blood returns to the heart by the pericardium, and this has two entrances. The first 

 I have already described, namely, via the posterior pericardium. The second occurs 

 anteriorly, and leads from the cavities of the carapace valves. Blood enters the carapace 

 directly from the body cavity at the attachment of the adductor muscles. In addition, 

 a tongue of gut parenchyma passes into the valves alongside the adductor muscle, and 

 so blood may pass into the valves from this tissue. From this point there is a system 

 of radiating channels in the thickness of the carapace valves. These were mentioned by 

 Claus (1876, p. 93) in a footnote. They extend outwards to the edges of the valves, 

 where they join to a marginal vessel. This is most marked anteriorly and has been 

 figured by Miiller (1894, PI. 2, fig. 2) who, however, does not describe it. These marginal 

 sinuses extend up the anterior edges of the valves, past the antennal notches, and open 

 directly into the pericardium at the combined attachment of the aortic and nauplius 

 eye muscles (Fig. 4 A). The entrance of each sinus into the pericardium is guarded 

 by a minute muscle which runs from the floor of the sinus obliquely backwards and 

 outwards to the outer layer of the carapace, and attaches alongside the attachment of 

 the second dorsal longitudinal muscle. Contraction of this muscle must pull up the 

 floor of the sinuses, and so close the entrance to the pericardium. 



CIRCULATORY PROCESS 



From the arrangement of the muscles associated with the pericardium and arteries, 

 I believe it is possible to analyse the process by which the blood is pumped round the 

 body. 



The attachments of the pericardial dilator indicate that, on contraction, the peri- 

 cardial floor must be pulled down and the pericardial space thus enlarged. Similarly, 

 the attachments of the pericardial compressors show that they must function by pulling 

 in the walls of the pericardium, thus diminishing the pericardial space. A regular 

 alternate contraction of these two muscles would then produce a rhythmical diastole 

 and systole of the pericardium. At first sight this seems the obvious method of working, 

 but, as I shall attempt to show, such a method is not possible. 



I assume that these muscles contract rhythmically and synchronously with the heart. 



