322 BOTANICAL GAZETTE [ OCTOBER 
opment. De Bary(1) says of Xylaria that the hyphal coil, or 
‘‘Woronin’s hypha” as it has been termed, seems to disappear 
before the formation of the asci begins. Hartig (10), in his study 
of Rosellinia, states that the asci seem to spring from a certain 
gelatinized zone, but whether they originate in the sexual appa- 
ratus or in the Wandparenchyma he ‘is unable to determine. Wor- 
onin(Q), too, fails to trace the exact origin of the ascus in Sor- 
daria or to find the connection between it and the ‘hyphal coil.” 
That a structure so prominent as this sexual apparatus, if it per- 
sists throughout the later development, should elude the notice of 
so many observers is quite incredible. It seems quite as improb- 
able that this fertilized archicarp should fail to play any part in the 
production of the ascospores Harper (5) finds in Spherotheca 
that the oosphere grows out into a branch of five to seven cells, 
from a certain one of which the ascus arises. The existence of 
the sterile cells of this branch suggests a solution of the problem 
in the true Pyrenomycetes, which has already been partially 
verified in the case of Teichospora, and is further confirmed by 
what is found here. Before the enveloping weft of threads, 
which are to constitute the wall, becomes too dense it may be 
observed that the coiled archicarp is becoming septate. Even 
in very young stages this septation has gone so far that in se 
tion the interior of the perithecium appears as a homogeneous 
mass of cells in which the outline of the coil cannot be dist 
guished. It now seems logical to conclude that this entire ene 
of parenchyma which forms the interior of the immature per 
thecium is formed by successive cell divisions in the archicarp. 
Instead of the five to seven-celled branch, giving rise 
ascus in Spherotheca, we have here an indefinite number of cells 
giving rise to a variable number of asci. This is strictly ae 
with the process in Teichospora, which was more easily tae 
we a 
because of the absence there of enveloping threads. It 11m ee | 
i 
contradictory to the observations of Woronin, De Bary, and H that 
though it fills in the gap which they have left open. It is ee ‘s 
¢ DUES 
only a comparatively very small number of these cells giv 
; el 
asci and the question as to what determines the fertile cells 1s : 
