Nevertheless, Liebig's desire to penetrate more deeply into the nature of the process of fermentation remained in many minds, and numerous endeavours were made to obtain further insight into the problem. In spite of an entire lack of direct experimental proof, the conception that alcoholic fermentation was due to the chemical action of some substance elaborated by the cell and not directly to the vital processes of the cell as a whole found strenuous supporters even among those who were convinced of the vegetable character of yeast. As early as 1833 diastase, discovered still earlier by Kirchhoff and Dubrunfaut, had been extracted by means of water from germinating barley and precipitated by alcohol as a white powder, the solution of which was capable of converting starch into sugar, but lost this power when heated [Payen and Persoz, 1833]. Basing his ideas in part upon the behaviour of this substance, Moritz Traube [1858] enunciated in the clearest possible manner the theory that all fermentations produced by living organisms are caused by ferments, which are definite chemical substances produced in the cells of the organism. He regarded these substances as being closely related to the proteins and considered that their function was to transfer the oxygen and hydrogen of water to different parts of the molecule of the fermentable substance and thus bring about that apparent intramolecular oxidation and reduction which is so characteristic of fermentative change and had arrested the attention of Lavoisier and, long after him, of Liebig. Traube's main thesis, that fermentation is caused by definite ferments or enzymes, attracted much attention, and received fresh support from the separation of invertase in 1860 from an extract of yeast by Berthelot, and from the advocacy and authority of this great countryman of Pasteur, who definitely expressed his opinion that insoluble ferments existed which could not be separated from the tissues of the organism, and further, that the organism could not itself be regarded as the ferment, but only as the producer of the ferment [1857, 1860]. Hoppe-Seyler [1876] also supported the enzyme theory of fermentation, but differed in some respects from Traube as to the exact function of the ferment [see Traube, 1877; Hoppe-Seyler, 1877]. Direct experimental evidence was, however, still wanting, and Pasteur's reiterated assertion [1875] that all fermentation phenomena were manifestations of the life of the organism remained uncontroverted by experience. Numerous and repeated direct experimental attacks had been made [p015] from time to time upon the problem of the existence of a fermentation enzyme, but all had yielded negative or unreliable results.