July, 1922] 
CLELAND OENOTHERA FRANCISCANA 
397 
parallelism may strike one at first, but it is quickly evident that this is 
merely the chance lying together of two whole threads which, when traced 
a little distance, are seen to have no immediate connection with one an- 
other. During this stage, also, the nucleolus begins to round off and 
leaves the nuclear membrane, where it has been lying through synizesis. 
It usually takes the stain very hghtly so that the endonucleus shows clearly 
within. It is an interesting fact that the latter body seems quite often 
to be clearly connected with a part of the spireme (fig. 16). 
During synizesis and the open spireme stage, the pollen mother cells, 
which during the resting period were packed closely together, begin to 
round off and to separate somewhat from each other, though they do not 
become entirely detached. The anther and the loculus in which they are 
placed become considerably enlarged during this time, but the cells them- 
selves grow only slightly. The nuclei also do not grow appreciably, meas- 
urements during the open spireme and second contraction stages showing 
an average diameter of 12-13 fx, as compared with lo-ii jj, in the previous 
resting phase. 
Second Contraction. After a time the nucleus begins to pass out of 
the open spireme stage and to enter upon the stage known as "second 
contraction." The spireme, which has hitherto been rather evenly dis- 
tributed throughout the nucleus, begins to thicken and shorten, the central 
portion condensing very rapidly, while the peripheral part becomes thrown 
into loops which radiate more or less clearly from the central region (fig. 
15). The threads in the center of the nucleus become thicker and thicker, 
until they gradually lose their thread-like form and become great, irregular, 
swollen masses which, as they increase in size, come to lie against one 
another so closely that the individual parts become lost and it is usually 
impossible to tell , whether they have amalgamated or not (figs. 16-18). 
This process results from the passage of material into the region from 
the rest of the spireme, a process which results also in a shortening of 
all the threads of the reticulum. 
As condensation goes on and this central mass becomes larger and 
larger, all the threads of the reticulum shorten, and it becomes more clearly 
evident that the threads which pass out toward the periphery are arranged 
in loops (figs. 15-18). Even in early stages of the second contraction 
it can be seen that the threads are forming loops which run out to the 
periphery and which are attached to each other in places (fig. 15). At 
first they are so long that they must bend in various ways, but as the 
threads shorten, the loops become smaller and smaller, so that at last 
they can stretch their full length without touching the periphery. The 
nucleus then begins to present a very striking figure, the chromatin mass 
resembling a large, irregular central hub from which the loops radiate 
out like spokes in a wheel (fig. 19). As the loops become shorter, their 
two sides may begin to approximate until in some cases they actually 
