240 BOTANICAL GAZETTE [September 



like process, straight or curved, sometimes stouter and distally 

 perforate, sometimes obsolete: the cells of the basal tiers similarly 

 modified, less prominent, flattened; the spinous process minute or 

 obsolete. The whole sporiferous pustule, in the type, 330 /x in 

 diameter by 210 /x deep. Spores 62-77X32 /i, the stalk (broken) 

 about 25X7 ju- 



On the inferior surface of the abdomen of Chromopterus delicatidum Beck., 

 Kamerun, West Africa. 



A single specimen of the apparently rare host, bearing this 

 very peculiar fungus on its abdomen, was found in a collection of 

 flies sent me by the Rev. George Schwab, to whom I am very 

 greatly indebted for this as well as for numerous similar favors, 

 and to whom I owe the remaining forms from Kamerun described 

 below. Since there is but a single specimen, I have been unwilling 

 to destroy it in order to determine the exact relation of the fungus 

 to its host, and have merely removed a certain number of spores 

 with a needle point from the general mass, which is firmly adherent 

 to the soft integument of the inferior surface of the abdomen. This 

 mass is somewhat diagrammatically illustrated in fig. 1, where it 

 is shown in situ on the insect's body; but whether the vegetative 

 hyphae penetrate the integument, or merely adhere firmly to 

 its surface, cannot be determined in its present condition. The 

 spores (fig. 2), which are in different stages of development in 

 different parts of the mass, the younger ones mostly near the edges 

 close to the substratum, are easily detached, and carry with them 

 a portion of the slender stalk, which is probably somewhat longer 

 than is represented in the figures. The cap-cells of the spore are 

 distinctly different from those of the paler mid-region, and their 

 terminal processes vary greatly in development. In some instances 

 (fig- 3) they are much stouter, and seem certainly to be distally 

 perforate; although in others this is quite evidently not the case, 

 and it is barely possible that they may be associated with some 

 viscous secretion which might aid the spore in attaching itself to 

 a new host. No such secretion, however, can be detected in the 

 present condition of the material; although, as in the case of the 

 Hormiscium just described, as well as in the species of the following 

 genus, it is difficult to imagine how the parasitism of such forms 



