They present no special features.  Butyl alcohol is, according to our observations, an ordinary product, although it varies and is by no means a necessary concomitant of these fermentations.  It might be supposed, since butylic alcohol may be produced and hydrogen be in deficit, that the proportion of the former of these products would attain its maximum when the latter assumed a minimum.  This, however, is by no means the case; even in those few fermentations that we have met with in which hydrogen was absent, there was no formation of butylic alcohol.

From a consideration of all the facts detailed in this section we can have no hesitation in concluding that, on the one hand, in cases of butyric fermentation, the vibrios which abound in them and constitute their ferment, live without air or free oxygen; and that, on the other hand, the presence of gaseous oxygen operates prejudicially against the movements and activity of those vibrios.  But how does it follow that the presence of minute quantities of air brought into contact with a liquid undergoing butyric fermention would prevent the continuance of that fermentation or even exercise any check upon it?  We have not made any direct experiments upon this subject; but we should not be surprised to find that, so far from hindering, air may, under such circumstances, facilitate the propagation of the vibrios and accelerate fermentation.  This is exactly what happens in the case of yeast.  But how could we reconcile this, supposing it were proved to be the case, with the fact just insisted on as to the danger of bringing the butyric vibrios into contact with air?  It may be possible that life without air results from habit, whilst death through air may be brought about by a sudden change in the conditions of the existence of the vibrios.  The following remarkable experiment is well-known:  A bird is placed in a glass jar of one or two litres (60 to 120 cubic inches) in capacity which is then closed.  After a time the creature shows every sign of intense uneasiness and asphyxia long before it dies; a similar bird of the same size is introduced into the jar; the death of the latter takes place instanteously, whilst the life of the former may still be prolonged under these conditions for a considerable time, and there is no, difficulty even in restoring the bird to perfect health by taking it out of the jar.  It seems impossible to deny that we have here a case of the adaptation of an organism to the gradual contamination of the medium; and so it may likewise happen that the anaerobian vibrios of a butyric fermentation, which develop and multiply absolutely without free oxygen, perish immediately when suddenly taken out of their airless medium, and that the result might be different if they had been gradually brought under the action of air in small quantities at a time.