180 MR. T. H. CORRY ON ASCLEPIAS CORNUTI. 
excretion now takes place, and the amount of gummy matter exuded is only just enough 
to give rise to a sheet of gum occupying the intervals between the rounded ends of the 
papilliform cells, which still project through, it being covered only very slightly if at all 
with gum (figs. 16, 17, 18). When therefore the partially hardened, somewhat elastic, 
and still very viscid excretion of one of these diverging grooves is forcibly removed, its 
broad edges or margins are seen, when examined under the microscope, to have the 
appearance of a flat perforated membrane, which is exceedingly viscid, the perforations 
in which correspond to the position of the excreting cells (fig. 16). The two diverging 
furrows, each with its excretion of partially hardened gum, arise perfectly independently 
of the main furrow, and their excretions are for a very short time unconnected with the 
corpusculum, but afterwards they become attached to this body. These excretions are 
then known as the “ processes, arms, caudicles, or appendages ” of the so-called “ gland," 
which was formerly regarded, e.g. by Payer, as giving rise to them in the form of a 
secretion, which travelled downwards in the diverging furrows, but which, as has just been 
shown, does not do во. The term “ corpuscular processes,” or more preferably “appendages,” 
may be conveniently applied to them. Robert Brown recognized the independence of 
their origin from that of the corpusculum, and Hermann Schacht the fact that they are 
produced by a method generally comparable to that in which the corpusculum itself is 
formed ; it will be evident, however, from the foregoing account, that the actual details 
are somewhat different. If at this period the corpusculum is separated from its furrow, 
these appendages will remain firmly attached to it, and come away with it in the form 
of flat compressed membranous sheets, each with a central darker, more thickened, solid . 
portion (vide figs. 16, 19). The upper extremity of each of these slender elongated 
corpuscular appendages impinges or abuts laterally (fig. 16), as has been already stated, 
on the corpusculum seated in the corpuscular furrow, while the lower diverging extremity 
lies directly over the apex of one of the anther-cells. Since the furrow in which the 
corpusculum is formed is always placed in the interval between two stamens, it follows 
that the two anther-cells, on the apices of which the appendages of the same corpusculum 
abut, do not belong to one and the same stamen, but to two different though neighbouring 
stamens. | | 
At this period of the developmental history, however, the contents of each ovoid anther- 
cell are quite free and distinct from the corpuscular appendage; for the anther-cells arc 
completely closed, and it is quite possible to remove the corpusculum and its appendages 
without displacing anything else. The free unconnected extremities of the appendages, 
though still very viscid, are not thickened to any extent; while the contents of each 
anther-cell have at this period acquired the colour, form, and degree of solidity character- 
istic of them in the mature state; and it is possible easily to separate them from the 
anther-cell in a single mass, forming a pollinium. | 
The triangular anther-alee stand in such а manner as only to leave а very narrow 
longitudinal linear slit or chink between them, which may be termed tke alar fissure 
(a. f., figs. 5, 6). This fissure lies directly over one of the furrows which occur at each 
angle of the style-table, and in the upper part of which the corpuscula are formed. The 
result is that, between the outer margins of the ale on the one hand, and the surface of 
