134 
R. A. HARPER 
the previous type, although the nucleus itself may have a diameter 
•| to I greater. The chromatin shows very little affinity for ordinary 
chromatin stains and appears almost homogeneous throughout the 
nuclear cavity. There is no sharp distinction of chromatic strands 
and nuclear sap to be made out. The chromatin is apparently distri- 
buted in the form of fine granules throughout the nuclear cavity. The 
condition of these nuclei resembles somewhat that found in the nuclei 
of the onion root tip at some distance back from the apex where they 
are not dividing rapidly and pass over into a resting condition. 
The homogeneous dense stage in the development of the sporange 
is followed immediately by a condition which apparently represents a 
further contraction and loss of cell sap on the part of the protoplasm. 
At this stage we find abundant vacuoles, or what appear at first 
glance as vacuoles, arising in the peripheral region of the spore sack 
(figs. 3, 4, 5). These figures, as is also the case with figures 2, 6-io, 
15 and 17, represent sections cut more or less horizontally through the 
developing spore sack, the columella cavity appearing as a more or 
less circular opening in the center. A little closer examination shows 
that each of these apparent vacuoles is pierced by a capillitial thread 
which runs straight through its central axis or may be slightly curved 
(figs. 9, 15, 17). These apparent vacuoles are in reality cavities formed 
by the withdrawal of the protoplasm from the surface of the capillitial 
threads at particular points. It is plain that the protoplasm is 
bounded off against the surface of the capillitial thread by an osmotic 
membrane, and that it has exuded liquid through this membrane into a 
space formed about the surface of the thread. The increase in amount 
of this exuded liquid leads to the formation of an almost globular 
droplet, which on superficial observation has all the appearance of an 
intra-cytoplasmic vacuole. Several of these droplets may be formed 
in a series not very far apart on the same capillitial thread. The row 
of watery droplets is in a sense suspended on the capillitial thread. 
It is to be seen at once that these droplets are in the strictest sense 
exterior to the protoplasmic mass, since the membrane of the apparent 
vacuole which encloses them is in reality continuous along the capil- 
litial thread with the external plasma membrane of the whole sporange. 
The droplets are not of course perfectly spherical, but form more or 
less extensive catenoid-like series on the capillitial threads. Their 
formation seems to consist, as noted, in a further extrusion of what 
we may call cell sap from the protoplasm, the exudation now taking 
