Thereupon, to find a way out of the difficulty, we resolved to modify completely the conditions of our experiments, and to apply to the research that we had in view a mode of cultivation that might completely, or nearly so, obviate the sole cause of error that we suspected, namely, the possible fall of cells or germs of yeast during the manipulations. We secured this by the use of flasks with two tubes, the right hand one of which was closed by means of a piece of india-rubber tubing with a glass stopper, the other one being drawn out in the shape of a swan’s neck. The use of these flasks, which was then new to us, permitted us to grow mycoderma and to study it under the microscope without fear of disturbance from exterior particles of dust. This time we obtained the results given in the first part of this paragraph. We no longer observed yeast or alcoholic fermentation following the submersion of the efflorescence, either in the flasks themselves, or in the test-flasks attached to them, as represented in Fig. 19. We observed, however, that kind of alcoholic fermentation of which we have already spoken and which is due to the mycoderma itself, a fermentative action that is still more instructive than the one which we thought we had determined, and certainly not less calculated to support the theory of fermentation which we have already briefly sketched. In an age when ideas involving transformation of species are so readily accepted, perhaps in consequence of their requiring no rigorous experimental work, it is not without interest to consider that, in the course of our researches upon the growths of microscopic plants in a state of purity, we once were inclined to believe in the transformation of one organism into another—the transformation of _mycoderma vini_ or _cerevisiæ_ into yeast, and that, on that occasion, we were altogether wrong, through having ourselves fallen a victim to the identical source of error which confidence in our theory of germs had led us so frequently to detect as affecting the observations of others. § III.—Growth of Mycoderma Aceti in a State of Purity. The study of _mycoderma aceti_ has not escaped the numerous causes of error which are apt to attend all observations made on microscopic organisms. This little fungus is still believed by many authors to be one of those polymorphous species capable of great modifications, according to the conditions of their cultivation—it could be, in turns, bacterium, vibrio, yeast, &c. Respecting it, we have seen resuscitated under a modern name, in the course of the last few years, the old hypothesis of Buffon concerning _organic molecules_, that of Turpin concerning the _punctiform globulines_ of barley, milk, and albumen, and the theory maintained by Dr. Pineau, of Nancy, and by Pouchet concerning _proliferous pellicles_.[71]