10 H. MARSHALL WARD. 



then by septa in other planes at right angles — and the sub-globular 

 body thus produced lies with the open part towards the epidermis. 



After this period, two events occur: 1st, the cells of the " core," 

 possessing very thin walls, acquire a different aspect from those of the 

 outer shell ; their finely granular protoplasm makes them appear 

 denser and more opaque, shining through the latter until this becomes 

 too thick to be transparent ; 2nd, the open part of the growing peri- 

 thecium becomes closed over, and the internal structures can no longer 

 be made out without the aid of actual sections. At this point my 

 observations have failed to decide which of two possible modes of 

 growth take place : Is the covering in of the " core " completed simply 

 by the extension and closing in of the edges of the outer layer ; or are 

 cells, cut off from the " core " below, intercalated, so to speak, into the 

 open gap ? One is led to expect by analogy that the former process 

 takes place, but some events lead to the suspicion that such may not 

 be the case. 



At the stage corresponding to fig. 19 the young perithecium 

 appears almost opaque, very little light passing through the dark- 

 coloured and thick outer walls ; from below, however, the larger cells 

 composing the " core " can be readily seen in the optical section, 

 shining by means of their dense, fine-grained contents through the 

 shell. In the next stages, the " core " can only be seen dimly through 

 the outer envelope (fig. 20), even after treatment with reagents, or, as 

 in figs. 21 and 22, after cutting or tearing off some of the outer cells. 



Nothing but a fortunate vertical section through the young fruit 

 at or near this stage will decide finally whether the lower side is 

 covered in by the meeting of the outer shell edges, or by partial 

 "delamination " from the lower side of the "core," and this I have 

 not succeeded in obtaining. The thick, dark outer walls have now 

 become so opaque, that optical sections fail to determine the course of 

 events ; and treatment w T ith reagents does not afford evidence suffi- 

 ciently satisfactory to decide the questions, since it seems impossible 

 to remove the colouring matter, Potassic hydrate or weak acids do, 

 it is true, render the structures a little more translucent after some 

 time ; but even the extreme resort of heating in weak chromic acid 

 has only yielded partial results, and w T ith this slight information on 

 the point I have reluctantly been compelled to content myself for the 

 present. A comparison of figs. 17 to 21 certainly suggests that the 

 process of envelopment is completed by the outer layer of cells derived 



