(104) In a word, I could easily multiply proofs of this kind if I were not afraid of abusing the patience of the reader. (105) Perhaps I shall be asked how I became acquainted with the fact that all these expressions are obsolete. (106) I reply that I have found them in the most ancient Hebrew writers in the Bible itself, and that they have not been imitated by subsequent authors, and thus they are recognized as antiquated, though the language in which they occur is dead. (107) But perhaps someone may press the question why, if it be true, as I say, that the marginal notes of the Bible generally mark various readings, there are never more than two readings of a passage, that in the text and that in the margin, instead of three or more; and further, how the scribes can have hesitated between two readings, one of which is evidently contrary to grammar, and the other a plain correction. (108) The answer to these questions also is easy: I will premise that it is almost certain that there once were more various readings than those now recorded. (119) For instance, one finds many in the Talmud which the Massoretes have neglected, and are so different one from the other that even the superstitious editor of the Bomberg Bible confesses that he cannot harmonize them. (110) "We cannot say anything," he writes, "except what we have said above, namely, that the Talmud is generally in contradiction to the Massorete." (111) So that we are nor bound to hold that there never were more than two readings of any passage, yet I am willing to admit, and indeed I believe that more than two readings are never found: and for the following reasons:-(112) (I.) The cause of the differences of reading only admits of two, being generally the similarity of certain letters, so that the question resolved itself into which should be written Beth, or Kaf, Jod or Vau, Daleth or Reth: cases which are constantly occurring, and frequently yielding a fairly good meaning whichever alternative be adopted. (113) Sometimes, too, it is a question whether a syllable be long or short, quantity being determined by the letters called mutes. (114) Moreover, we never asserted that all the marginal versions, without exception, marked various readings; on the contrary, we have stated that many were due to motives of decency or a desire to explain obsolete words. (115) (II.) I am inclined to attribute the fact that more than two readings are never found to the paucity of exemplars, perhaps not more than two or three, found by the scribes. (116) In the treatise of the scribes, chap. vi., mention is made of three only, pretended to have been found in the time of Ezra, in order that the marginal versions might be attributed to him. (117) However that may be, if the scribes only had three codices we may easily imagine that in a given passage two of them would be in accord, for it would be extraordinary if each one of the three gave a different reading of the same text.