This is the reason why the Talmudic commentaries,[94] which are, as it were, the summing-up of Rashi's teachings, of his own studies, and of the observations of his pupils, have a more mature, more thoughtful character than the Biblical commentaries. They undoubtedly represent a greater amount of labor. It seems that Rashi himself made two or three recensions of his commentary, at least for many of the Talmudic treatises. Testimony to this fact is given by the variations of certain passages in the extant text and that cited by the ancient authors, notably the Tossafists. Moreover, the Tossafists explicitly mention corrections made by Rashi in his own work. The query naturally arises whether the corrections indicate that Rashi worked the entire commentary over and over again. The answer is no; for certain treatises remained incomplete, and others seem never to have been begun. Presumably, then, Rashi revised a treatise according to the needs of the occasion, as, for instance, when it came under his eyes in the course of instruction. However that may be, the work that we now possess is a mixture of the first and the last recension, though we cannot always tell which is the later and which the earlier. Another fact explains the difference I have pointed out between the Biblical and the Talmudic commentaries. For the Biblical commentaries there had been no precedent, and if they possess the merit of originality, they also illustrate the errors of a man who tries his powers in a field of work devoid of all tradition. For the Talmudic commentaries, on the contrary, models were not lacking. The example of Gershom was sufficiently notable to evoke imitation, though his work was not so complete as to discourage it. We must not forget Rashi's predecessors because he eclipsed them. This would be contrary to his intentions, since he frequently cites them, rendering value in return for value received. In fact, he knew well how to use their works to advantage. He submitted them to a judicial and minute examination, collecting all the material he needed furnished by the Geonim as well as by his immediate masters. It would be as inexact to assert that he only made a <I>resume</I> of their works as to say that he worked along entirely original lines and relied solely upon his own resources. If we could compare his commentaries with previous commentaries (for some this comparison has been made), we should be forced into the admission that his part is smaller than one would suppose. The best proof of this fact is that the usual basis of his commentary for each treatise was the explanation of the master under whom he had studied it. He often cites the writings of his masters, to which he gives the title <I>Yesod,</I> "Foundation," probably either collections made by the teachers themselves or notebooks edited by their pupils. As a result of the love of brevity which is one of Rashi's marked characteristics, he does not quote in its entirety