ee rr we 
1906] ' OLIVE—DEVELOPMENT OF EMPUSA 203 
conidiophore, flow out into the single, large, bell-shaped conidium 
(fg. 39). 
In E. culicis, which has binucleate conidia, septation of the conidi- 
ophore occurs, by which the protoplasm cuts itself off from behind; 
but, unlike the case of E. sciarae, I have found no branching with 
it, so that as a consequence, all of the binucleate protoplasm of the 
conidiophoric hypha flows into the one terminal conidium (figs. 30-35). 
The simple conidiophore of E. culicis resembles, therefore, that of EF. 
muscae in being unbranched, but the origin of the conidiophores may 
differ in the two cases. One may find, in fact, in E. culicis, two sorts 
of “hyphal bodies;”’ either small cells which, like the corresponding 
ones of E. muscae, give rise to but one conidiophore; or, on the other 
hand, larger cells, which may give rise to several conidiophores by 
budding from several points simultaneously, in a similar manner to 
Conidiobolus. In both instances, each conidiophore remains simple 
and ultimately bears, terminally, the binucleate conidiospore. 
In the species with uninucleate conidia, E. sciarae, E. americana, 
E. aphidis, and Empusa sp., the coenocytic conidiophore, as was 
indicated above, is cut up by septa, in a manner to be described later, 
into uninucleate segments (figs. 16-21). In all these cases, this results 
in a branching growth and successive abjunction of the acrogenous 
spores. Below the terminal cell the penultimate cell pushes out to 
one side, and thence bores its way to the surface of the host, where 
it abjoints a single uninucleate conidium (figs. 16, 23). From this 
habit in certain forms, arises a profuse system of branching, frequently 
of a digitate type (e. g., Empusa sp.) or corymbose (E. sciarae and 
others), to enable the uninucleate segments to reach the surface and 
to discharge their protoplasm by means of the abjointed conidia. 
But the last-formed vegetative cells of E. sciarae contain only 2-4 
nuclei (figs. 1, 12, 15), so that, in this instance, only a correspondingly 
small number of branches are formed. 
_ The process of abjection of the conidia of Empusa is apparently 
similar in a general way to that described for the sporangia of Pilob- 
olus, except that in these conidia there is no gelatinous collar visible. 
The formation of the partition at the base of the conidium in Empusa 
also is quite a different process from the formation of the columella 
in the case of Pilobolus. 
In Empusa, the vacuole which appears in the basal portion of the 
