492 MISS K. M. PARKER ON THE 



the heart primovdia lie at the medial mai'gin of the pleuro- 

 pericardial canals, we see that the inward pi-ogression of the edge 

 of the splauchnopleural fold in the direction indicate<l by an 

 arrow in text-fig. 24 A, will bring about the conditions shown in 

 text-fig. 24 B. (Oompai-e also text-fig. 8, where the relations are 

 essentially the same and probably approximate more closely to 

 those in the living embryo.) 



From this stage, it is evident that when the lateral limbs of 

 the pleuro-pericardial canals become incorporated in the median 

 pericardium by the backgrowth of the cardia.c fold, the heart 

 primordia will lie in the dorsal wall of the pericardium and will be 

 situated towards the middle line of the gut. (See text-fig. 8.) 

 Thus, as Rouviere (14) indicates in describing similar con- 

 ditions in the lubbit, the position of the heart primordia in 

 such a stage as is represented in text-fig. 24 B, is brought about 

 " by the inward extension of the lateral prolongations of the 

 cardiac fold." There is no evidence of reversal of the peri- 

 cardium, nor is there adequate ground for assuming that such 

 a process occurs. 



We can, therefore, gain a clear conception of the way in 

 which the lateral heart primordia attain the position they occupy 

 in Stage III. (PI. I. fig. 4 ; text-fig. 8) lying side by side below the 

 closed fore-gvit. 



In order to complete the history of the early development of 

 the heart, it is now necessary to consider the mechanism wliich 

 brings the heart-tubes into contact in the middle line. 



If we compare figs. 4 and 5 (PI. I.) we see a.t once that Avliile 

 the pericardium has increased i-apidly in the anteio-posterior 

 direction, it has not increased in transverse width and, in fact, at 

 the point of closest appi'oximation of the heart-tubes, an actual 

 decrease in width has occurred ; that is to say, the pericardium 

 at this stage is growing in the antero-posterior direction at the 

 expense of its transverse width. Tlds fact suggests a simple 

 explanation of the approximation of the heart-tubes after gut- 

 closure, for it may be supposed that if the total width of the 

 pericardium is reduced by this process of sti"etching, the distance 

 between the heart-tubes will decrease until they meet each 

 other in the middle line *. 



The heart-tubes, in the period following immediately on their 

 reaching the middle line, grow very rapidly, so that, in the next 

 stage (V.), we find various forms of curvature which serve to 

 accommodate the increased length of the heart. In the embryo 

 of Pe/rameles obesida described in this stage (PI. II. fig. 6), the 

 heart-tubes are in contact through a great portion of their length 

 and here follow a parallel curved course. The sejDarate heart-tubes 

 lying in the lips of the anterior intestinal portal, however, show 

 a marked difference from eaeli other both in their length and the 



* The suggestion that the approximation of the heart-tubes is due to such a 

 growth in longth without compensatory growtli in width was made to me by 

 Professor Hill. 



