502 Cresswell Shearer 



I bave examined this is not the case , as I have alreadv stated, for 

 in transjiarent larvae such as those represented in figures 17, 18 

 and 49, the dorsal wall of the coelom can be seen quite plainly in 

 the living state. In the development of Asterina gibbosa Mac 

 Bride (16 pag. 368) has drawn attention to the number of abnormal 

 ways in whlch the coelomic sacs frequently arise in this Echiuoderm. 

 One abnormalitv frequently found, consists in the coelomic epithe- 

 lium of the gut breaking up into a mass of cells having the appcar- 

 ance of mesenchyme , which choke up the lumen of the coelom 

 This abnormality may take place at any stage in the differentiatiou 

 of the coelom. I think this Observation is of great interest when 

 taken in connection with the frequency with which the epithelium 

 of the coelom in Phoronis hing against the gut is found wanting 

 (fig. 34). Mac Bride has also noted the irregulär manner in which 

 the various portions of the coelom often arise in different parts of 

 one and the same larva, and we might justly expect some similar 

 Variation in Phoronis which possesses so remarkable a life-history. 

 The embryology of animals that pass through a more or less 

 protracted larvai development has repeatedly shown us that every- 

 thing in early development is sacrificed to the immediate nceds 

 of the larva. For instauce in Unio the Organisation of the Glochi- 

 dium is distinctly foreshadowed in the segmentation long before the 

 actual appearance of the Glochidium stage. So in Piwronis the 

 development of the coelom may be somewhat modified. 



As already stated the nephridia are well developed while the 

 coelom is yet small. They are from the first outside this structure, 

 and ventral and lateral to it. As the coelom grows its lateral por- 

 tions soon force the nephridia against the inner wall of the ectodcrni, 

 and away from the gut. In section (pl. 32 fig. 47) a few cells of 

 the nephridial canal of one side are shown dose to this lateral 

 portion of the coelom. 



At this time the relative proportions of the nephridia and the 

 coelom are well shown in pl. 31 figs. 13, 18, pl. 32 figs. 44, 49, 53 and 54. 

 Fiually in pl. 33 figs. 56 and 58 these relationships are again shown 

 in a diagrammatic figure. From the inspection of these ligures it will 

 be seen that the nejdiridia with their solenocytes project some 

 distance into the blastocoelic space in front of the coelom, and that 

 it is only by the growth of this structure in the fully fornied Actino- 

 trocha that the anterior wall of the coelom comes close up to the 

 heads of these organs. By the time this stage is reached a con- 



