Cysiidia of Coprinus atramentarius . 621 
so large a proportion of the interlamellar spaces that a great many spores 
would settle upon them, and in that way be prevented from escaping into 
the outer air. A beautiful arrangement is provided to make this impossible. 
The cystidia do not undergo auto-digestion at the same time as the basidia and 
paraphyses in their immediate vicinity , but a short time previously thereto. 
They destroy themselves in succession from below upwards on each gill, 
and each one disappears a few minutes before the basidia in its neighbour¬ 
hood come to be involved within the upwardly progressing zone of spore- 
discharge. The semi-diagrammatic drawing given in PI. LI, Fig. 19, shows 
what was made out by means of sections taken transversely to the gills 
or parallel to their edges. The section through three gills is supposed to be 
quite vertical. From above downwards one can distinguish the following 
seven zones: (a) a zone of less ripe spores in which the cystidia are fully 
turgid, and are acting as props to keep the gills apart; (b) a zone of riper 
spores in which the cystidia are disappearing ; (c) a zone of ripe spores from 
which the cystidia have already disappeared ; (d) a zone of spore-discharge 
where cystidia are absent; (e) a spore-free zone ; (/) a zone of auto-digestion ; 
and finally, (g) a liquid film, the products of auto-digestion, at the gill-edge. 
It is clear that, owing to their early auto-digestion, the cystidia cannot 
possibly hinder the fall of the spores and their escape from between the gills. 
Since the cystidia of Coprinus atramentarius are essential constituents 
of the fruit-body mechanism in that they alone prevent the very thin, very 
broad, and very flexible gills from coming into contact with one another, it 
is not surprising that we should find that they are retained between the 
gills as long as practicable without their becoming hindrances to the fall 
of the spores. They are removed only just in time to prevent them ever 
extending between those parts of the gills where spore-discharge is taking 
place. The cystidia-freed portions of the gills which hang downwards 
(PI. LI, Fig. 19, zones c , d , ef> and g) are only about 0-25 mm. in depth. They 
are kept apart by means of the cystidia above them, so that there is no 
danger of the interlamellar spaces between them becoming reduced in width. 
As a matter of fact the interlamellar spaces in the regions of spore-dis¬ 
charge appeared to be distinctly broadened out owing to a slight con¬ 
traction of those parts of the gills undergoing auto-digestion (cf. Fig. 19). 
The fate of the cystidia was found out by studying sections, about 1 mm. 
thick, which were cut transversely through the gills. Some of the sections 
were cut perpendicularly to the gill-edges and therefore resembled 
that shown in PI. L, Fig. 5, whilst others were cut in such a way that one 
side of each section was made up of the free auto-digesting edges of several 
gills. In all cases the sections were placed in a compressor cell in order to 
prevent undue transpiration. Where a section included the free gill-edges, 
it was turned upside down so that the edges looked upwards. On looking 
at the sections with the low power of the microscope, it was possible not 
