1906] CHAMBERLAIN—OVULE OF DIOON 353 
while the ventral canal nucleus remains at the apex, are questions 
still unanswered. A moment’s reflection will convince any one that 
neither temperature, light, nor gravity has any appreciable influence. 
The path by which nutrition reaches the egg may be an important 
factor, for the extensive vascular system of the ovule terminates near 
the top of the endosperm. On the other hand, the same position of the 
nucleus is found in gymnosperms whose ovules have scarcely any 
vascular tissue at all. Later, during the formation of the proembryo, 
the polarity is reversed, the nuclear activity being most vigorous at 
the base of the egg. At this time it is evident that nearly all the 
nutrition is coming from the endosperm at the base of the egg. Asa 
working hypothesis it may be suggested that the early position of 
the nucleus at the apex, and also the reversal of the polarity, are due 
to chemotaxis, the source of nutrition being the controling factor. 
The egg nucleus.—The egg nucleus of Dioon is the largest which 
has yet been found in plants. Its usual diameter is about 500 p, 
but nuclei sometimes reach a diameter of 600 #. Not infrequently the 
nucleus is elongated, and in such cases its bulk is likely to be greater 
than that of the usual spherical nucleus. One nucleus measured 1475 # 
by 3804. In spite of the immense size of this nucleus, its structures 
could not be interpreted. Its chromatin content is shown at the lower 
end of the spindle in fig. 32. Stages immediately following this were 
not secured, and in stages like jigs. 34 and 35 the chromatin can no 
longer be identified. From this time to the entrance of the sperm 
into the egg, the internal structure of the nucleus is shown in figs. 
42-44, which represent details as they appear under a magnification 
of 1300 diameters. The entire nucleus, drawn to this scale, would 
be more than 6™ in diameter! In all parts of the nucleus is found a 
delicate network of varying thickness, and upon it or imbedded in it 
are granules and globules of various sizes. The smallest granules, 
which are almost always associated with the network or clinging to the 
Surface of larger globules, stain with gentian-violet. Most of the 
larger globules stain with safranin, so that the general tone of the 
network is red; but some stain with the gentian-violet and in others 
the two stains blend. There are always a few which do not stain at all. 
In most preparations the network stains very faintly or not at all. 
The nucleoli, which are large and vacuolated, can usually be distin- 
