452 W. DE MORGAN AND THE LATE G. HAROLD DREW. 



is evidently similar to the perisarc of the normal animal. Once this 

 perisarc has been secreted, the restitution mass appears to be immune 

 to the attacks of flagellates or bacteria, which are unable to penetrate it. 

 From this stage onwards the external changes that occur take place 

 slowly. The turned-up edges become thicker and more nodulated at the 

 expense of the central part, and sometimes may project in the form of 

 spherical or ovoid nodules connected at the base with the main mass 

 by a comparatively small isthmus of tissue. (In Fig. 3 a mass, 48 hours 

 old, is shown, and in Fig. 5 one of 8 days.) There is later a slow 

 but continuous shrinkage of the restitution mass away from its perisarc, 

 leaving a clear space between the two (Fig. 6), after from three to four 

 weeks irregularly circular spaces begin to show in the cell mass, which 

 then presents a somewhat sponge-like fenestrated appearance (Fig. 6). 

 Later changes are extremely slow, and consist of a further slight shrink- 

 age, and increase in size of the spaces in the tissue. During all this time 

 the restitution mass retains its yellowish colour and definite outlines, and 

 sections show that the cells are healthy and undegenerated : in the case 

 of masses that die, the yellowish colour is rapidly lost, and they appear 

 as dirty white, soft, floccular bodies, which soon fall a prey to bacteria 

 and other parasites ; it is thus easy to distinguish the living from the dead 

 masses by the eye. 



At the time of writing some of these masses have been kept alive for 

 a period of 60 days, and during this time none of them have shown any 

 tendency to regenerate a stem or hydranth. 



In the case of the smaller nodular masses that had not been shaken 

 together so as to form larger aggregates, a perisarc was secreted in from 

 12 to 18 hours, and where several nodules were joined by their strands of 

 tissue a complete tube of perisarc was secreted around these connecting 

 strands. The appearance presented in such cases was often peculiar 

 and somewhat suggested an attempt at the formation of hydranths 

 which had aborted through not being able to burst the surrounding 

 perisarc ; but observations made from the earliest stages, when the 

 nodules arose from simple aggregations of the cells show that there is 

 no justification for such a view, and this was borne out by the internal 

 structure of these nodules as shown in sections of fixed preparations. 

 When the process of contraction of the restitution masses had proceeded 

 a little further, the connecting strands of tissue between individual 

 nodules were often completely retracted, thus leaving the nodules merely 

 connected by empty tubes of transparent perisarc. 



