Mutinies caninus (Hud si), Fr. 355 
this rudiment run in very irregular and zigzag courses in all 
directions, and give out short lateral processes or branches, 
but these are probably in part anastomosing connexions 
which have been cut in sectioning. These hyphae stain 
deeply. Higher up the stipe the hyphae of the stipe and 
those of the column R are less interlocked than in this figure. 
Relationship of Phalleae and Clathreae. 
We may here consider the bearing of the facts of early 
development upon the views held as to the relationship 
between the Phalleae and Clathreae. As a stage of develop- 
ment has now been reached which is about the same as that 
illustrated by Ed. Fischer for /. impudicus 1 , in his Figs. 20 
and 31, it becomes possible to compare the actual course of 
development with the ideas that are prevalent. It should be 
understood, however, that Fischers statement of the course 
of development was advanced tentatively, until a fuller series 
of stages should be available for study, and that in his late 
paper on M. caninus 2 he outlines and favours in some im- 
portant features the course of development which I have 
presented here, but which I had already worked out indepen- 
dently in my inaugural thesis, before the publication of 
Fischer’s paper. While advancing such newer views, he also 
states his inability to .determine from his preparations whether 
his earlier scheme of development might not be the true one 
after all. 
In Fischer’s earlier account the central column of the inter- 
mediate stage was regarded as identical with the medullary 
bundle M of my Fig. 1, although Fischer was not sure that 
a portion of the upper end of the column might not be formed 
from a part of the sheaf-like head of M. The gelatinous layer 
G, the dense portion i, and the tissue of the gleba, were re- 
garded as arising from the sheaf-like head N. By a vigorous 
interstitial growth, and the production of radially arranged 
1 L. c., 1890, p. 22. 
2 L. c., 1895, p. 130. 
