§2 
in cacao pods by cutting and raising pieces of tissue. After 
inoculation, the raised tissues were replaced, and the parts were 
kept moist with sterile cotton-wool soaked in sterile distilled 
water. After four days, mycelium appeared aerially around 
the edges of all the inoculation cuts (sixteen in number), and, 
after ten days, conidia of Colletotrichum were present in numbers 
on both sides of the cuts. After a further ten days, caespitose 
perithecia of Glomerella cingulata, S. and v. S., developed on the 
site of the Colletotrichum acervuli. Control pods, wounded as 
above, did not show any fungus growth. Similarly, cacao twigs 
were inoculated with Colletotrichum conidia in wounds. In their 
case, growth of the inoculum was slower than ia the case of the 
pods, but the results were the same. Conidia were also placed 
on unwounded surfaces of pods and twigs, but, although they 
germinated freely and produced numerous appressoria, no pene- 
tration of the unbroken surfaces resulted. Colletotrichum coffea- 
num was thus proved to be capable of vigorous, growth on cacao 
material. Its effects were indistinguishable from the usual pod- 
rot of Uganda, and the Colletotrichum and Glomerella on the pods 
and twigs were morphologically indistinguishable, the former 
from coffeanum or incarnatum, the latter from Glomerella on 
coffee twigs. Conversely, Glomerella ascospores from a cacao- 
pod when sown in prune-agar plates, gave an aerial mycelial 
growth in forty-eight hours, which was succeeded in five days 
by conidial formation, the conidia being identical with those of 
Colletotrichum coffeanum, and the same ascospores when inoculated 
direct into wounds in cacao-pods gave fifty per cent. positive 
results with the production of Colletotrichum conidia in abundance. 
Ascospores placed on pods germinated but did not penetrate the 
unbroken skin. 
Further work consisted of the transfer of Colletotrichum 
conidia direct from rotted cacao-pods found on trees to agar 
plates and tubes, and the employment of the resulting cultures 
to inoculate coffee leaves. The behaviour of the fungus in these 
_ cultures was exactly similar to that of the coffee Colletotrichum. 
Sixteen inoculations were performed with Colletotrichum conidia, 
and, in every case, positive results followed. An asetose Colleto- 
trichum, indistinguishable in other respects from Colletotrichum 
coffeaniim, was recovered from all. Controls remained healthy. 
Thus the cacad-pod Colletotrichum, whether incarnatum or 
theobromicolum, was shown capable of infecting coffee. The 
conclusion to be drawn from the cross-inoculation results and 
cultural evidences is that C. coffeanum, and either C. incarnatum 
or C’. theobromicolum, or both, are, if not the same species, closely ~ 
related forms which vary under natural conditions. If dis- 
tinctions are to be preserved between them, they would seem to 
be of a physiological rather than a morphological nature, and to 
be based on the existence of physiological varieties within the 
morphological species. The conclusion from the succession of 
forms mentioned above, viz., Colletotrichum conidia and Glome- 
