540 Fraser and Brooks . — Further Studies on the 
even at a late stage of mitosis, by a more or less definite line on which the 
centrosomes lie. In Ascobolus furfur aceus, however, the nuclear area, 
especially in the third division, is ill-defined and shades off gradually into 
the cytoplasm, and the spindle often lies to one side of or partially 
outside it (Fig. 30). This arrangement brings the centrosome nearer 
the centre of the dense mass of cytoplasm which constitutes the aster (cf. 
Fig. 22), and it also affects the relative distribution of the vacuoles and 
nuclei. 
After the third division the cytoplasm is traversed by irregular series 
of vacuoles (Fig. 32), which may be termed lines of cleavage. These play 
an important part in the delimitation of the spore, or rather in the segrega- 
tion of the masses of spore-plasm. During spore-formation, as during 
karyokinesis, the cytoplasm near the centrosome takes at first a deeper 
stain than that which is more remote (Fig. 31). As development proceeds 
this dense area extends, and the impression is given of an outward flow of 
some substance which emanates from the centrosome and is capable of pro- 
ducing alterations in the cytoplasm. It has already been suggested ( 9 ) that 
this substance is not improbably an enzyme. The difference between 
Ascobolus and the forms hitherto investigated lies in the much more impor- 
tant part here played by the vacuolate areas which we have termed lines of 
cleavage. Such lines no doubt originate at the nuclear areas, but we think 
it not unlikely that their further development is the result of new tensions 
set up in the cytoplasm by the changes going on around the centrosome. 
In the forms we have studied the astral rays and the limiting layer of the 
spore in its early stages are alike less definite than those illustrated by 
Faull (6). 
As development proceeds the dense mass around the centrosome and 
nucleus becomes more regular, and is bounded by a definite line (Figs. 35, 
36), while a similar membrane limits the neighbouring epiplasm. Between 
the two is a clear space related to the old lines of cleavage, and its 
boundaries both towards the spore and towards the epiplasm may well be 
such ectoplasmic layers as delimit an ordinary vacuole. This interpretation 
is somewhat remote from that given by Harper, though we indorse his con- 
clusion that the aster plays an essential part in spore-formation. In Asco- 
bolus the wall of the mature spore is of great thickness, and this fact may be 
related to the wide vacuolate area which separates the two limiting mem- 
branes at an earlier stage. 
Divisions in the ascogenous hypha (Fig. 14) were several times observed 
in Ascobolus , but the cytoplasm is dense and contains various granules, and 
examples clear enough to allow a satisfactory determination of the chromo- 
some number were not obtained. Our observations point to the occurrence 
of four chromosomes at this stage. This would indicate that pairing 
of chromosomes in fertilization takes place here as in Humaria gramdata. 
