14 DUNCAN, ON THE IMPREGNATION 
the pollen-tubes in their passage from the axis to the micro- 
pyle; neither the axis, the placentz, nor the papillary struc- 
ture, are hollow for the passage of the pollen-tubes; on the 
contrary, the tissues are remarkably cellular and well supplied 
with moisture. 
The ovules, when ready for impregnation, are large, very 
cellular and transparent; and a very simple manipulation 
splits off the external coat from the long and narrow nucleus. 
Each ovule is attached to the axis by its hilum, is a long 
oval in shape, and the orifice through which the pollen-tube 
has to pass is external to the hilum. Transverse sections show 
this orifice very well. The orifice of the micropyle is not at 
the extremity of the ovule, but is close to the hilum; the 
ovule is therefore “ anatrope” in appearance, but not so in 
reality, for there is no reflection of the ovule during its 
development, but one half of it is, from the first, devoted to the 
vascular system, and the other to the formation of the coats 
and embryo-sac. There is no space between the ovule and 
the walls of the ovary until long after impregnation. The 
micropyle is very distinct, being situated in the long mam- 
miilary end of the nucleus, which projects considerably below 
the external coat (figs. 1, b, e; fig. 4, c). 
For all the purposes of the study of its impregnation, the 
ovule may be first examined in the rudimentary flower, whose 
anthers are as yet uncovered by the perianth ; secondly, when 
the anthers are covered; and thirdly, immediately before 
impregnation, or when the flower is in bloom. 
1. A transverse section (through the axis) exhibits the 
ovules adherent by their cellular and vascular tissue to the 
axis (at the placenta), shows the projection of the nucleus 
cropping out of the external cellular coat, and presents to 
the eye the situation of the upper and globular part of the 
nucleus, covered now by the external coat, distinguishing it 
by a track of transparent tissue. By gently removing one of 
the ovules, and placing a piece of thin glass over it, and 
pressing the glass gradually with the handle of the knife, the 
nucleus may, in the majority of instances, be slipped out of 
its external cellular coat.* Then the nucleus is proved to 
be cylindrical, rounded at one end, and tubular, with the 
micropyle at the other (fig. 2). It is very tender, and 
consists of two parts—a body and a neck. ‘The neck is 
homologies of the sexual apparatus in animals. The anatomist may well be 
perplexed at a placenta inside an ovary, and part of it perforated by the 
male element. 
* This external coat is cellular, the cells being square in their outline ; it 
leaves more than two thirds of the neck of the nucleus uncovered, 
