MICROSCOPIC ORGANISMS IN INTESTINAL CANAL. 243 
conidia seem, as a rule, to become broader, and their extremities, 
which are at first in many cases more or less truncate, assume a 
rounded convex outline. In some cases the filaments appear to 
divide dichotomously at the extremity. The superficial layers 
at the basis were full of long, horizontal, sparsely-branched 
filaments, from which short vertical twigs arose, became aerial, 
and ultimately split up into conidial segments. After another 
interval of twenty-four hours the reaction of the basis was 
transiently but distinctly alkaline, and the surface was clothed 
with a thick, shaggy, greyish-yellow coating of curious gela- 
tinous, acutely conical tufts, composed of dense masses of oidial 
conidia and filaments. Most of the conidia were very short 
and broad, many being nearly spherical. They were full of 
dense shining protoplasm, only showing at utmost one or two 
minute vacuolar spaces. On being sown on a suitable medium, 
they rapidly germinated, undergoing a great increase in size, 
accompanied by extensive vacuolation ere doing so. The short 
rounded conidia measured from 9°4 x 6*ly to 6°5 x 6:0 or 
55 uw. The longer joints measured 23 x 5°5 yu, and all inter- 
mediate forms connected the two series with one another. On 
the following day the reaction of the basis was distinctly aud 
permanently alkaline, and no farther development of fungi 
occurred in it. 
That the Oidium in this and other cases owed its origin to 
fungal elements intrinsic to the basis, and not to extrinsic ones 
accidentally introduced from without, was proved by the follow- 
ing facts :—1st. Ocdiwm lactis is not a form which tended other- 
wise to occur spontaneously in any of the localities in which the 
experiments were conducted. 2ud. Boiling the excreta previous 
to isolation was invariably followed by a failure in the appear- 
ance of the fungus, and this not as the result of any change 
causing them to become an unsuitable medium for it, as an 
abundant crop appeared as usual on introducing oidial elements 
artificially. It cannot, moreover, be assumed that in the experi- 
ments on the effect of boiling, the excreta were originally fortuit- 
ously an unsuitable medium, as check experiments were tried 
with unboiled portions of the same material, which constantly 
resulted in the occurrence of the normal development. That the 
phenomenon is not one dependent on casual peculiarities of a 
particular season was shown by its uniformity at intervals of 
several years’ duration. Besides the experiments on a large 
scale in ordinary moist chambers, others were tried in which 
minute fragments of the material were hermetically sealed in 
wax cells, and the sequence of events in these cases was pre- 
cisely of a similar nature. There can, I think, be little doubt 
that the digestive canal in man in this country normally contains 
