FRUCTIFICATION OF CHOANEPHORA. 411 
leaving the former structures full of watery fluid. When the ripe conidia become 
detached, the thin membrane of the upper portion of the capitellum usually breaks up, 
leaving only delicate threads attached to the rim of the lower portion, or collapses and 
disappears entirely. Occasionally, however, it persists in the form of a filmy sac pro- 
jecting from the cup of the lower portion (fig. 4). The persistent portions of the capitella 
now come to appear as a series of pedicillate funnels; and as conidia not unfrequently 
fall into them, or are carried down attached to the collapsing upper portion, it might 
readily be supposed that they had been developed within the funnels. Taking the deve- 
lopment and structure of this form of fructification alone there would seem to be very 
conclusive grounds for regarding the plant as of a mucedinous nature; but the facts in 
regard to its anatomy and life-history, still to be described, show what false conclusions 
may be arrived at where one form alone of asexual fructification is employed as a basis 
for the classification of fungal organisms. 
On proceeding to examine those portions of the plant which remain concealed within 
the tissues of its host, the following results are arrived at. The mycelial system is com- 
posed of a series of main tubes, which are branched, devoid of septa, and full of a granular 
yellowish protoplasm, in which a continuous streaming motion may be detected, and of 
a set of branched radicles or haustoria, taking origin from the branches of the main 
tubes. These radicles are at first full of contents similar to those of the main tubes; but 
they soon become emptied of these, and are, when mature, full of watery fluid and pro- 
vided with a basal septum (fig. 20). The mycelium is therefore in every respect charac- 
teristic, not of the Mucedines, but of the Mucorini. The cells of the corollar tissue of 
the host do not appear to be penetrated either by the main tubes or by the radicles; but 
the latter are closely applied to them, and in some cases appear to adhere to them by 
special dilatations. 
At certain points, where they approach the surface, the mycelial filaments give origin 
to slender twigs, which generally follow a course more or less parallel to the parent 
tubes. Such twigs are destined to produce the aerial conidiiferous filaments. After 
growing for some distance, they either terminate in a distinct dilatation or, more rarely, 
pass on directly into the acrial filaments. Sometimes the tips of the twigs force their 
way to the surface previous to forming their terminal dilatations; but the latter are 
generally produced immediately beneath the epidermis. In such cases the resistance 
which they encounter causes the filaments in many instances to follow a more or less 
contorted course. The dilatations are frequently directly continuous with their parent 
tubes; sometimes, however, they are eventually separated from them by transverse 
Sepia. 
Such are the features commonly presented by the plant; but in certain cases the 
. Mycelium is capable of producing the apparatus of sexual production. This happens 
comparatively rarely, and I have as yet been unable to determine what the precise con- 
ditions are under which it occurs. In these cases the mycelial tubes, after having given 
origin to the common conidial fructification, produce a number of short branches, which, 
in place of growing in length, swell out into thick club-shaped processes (fig. 11). In 
most cases these arise directly from the mycelial tubes, and remain continuous with: — S 
