The second purpose of the notes is to set forth the evidence of the manuscripts. I have not attempted to give anything like a full account of the variant readings of these, but have recorded so much as is sufficient for four different purposes. (1) To vindicate the text of 1633. I have not thought it necessary to detail the evidence in cases where no one has disputed the 1633 reading. If the note simply records the readings of the editions it may be assumed that the manuscript evidence, so far as it is explicit (the manuscripts frequently abound in absurd errors), is on the side of 1633. In other cases, when there is something to be said for the text of the later editions, and especially when modern editors have preferred the later reading (though I have not always called attention to this) I have set forth the evidence in some detail. At times I have mentioned each manuscript, at others simply _all the MSS._, occasionally just _MSS._ This last means generally that all the positive evidence before me was in favour of the reading, but that my collations were silent as to some of the manuscripts. My collators, whether myself or those who worked for me, used Mr. Chambers' edition because of its numbered lines. Now if Mr. Chambers had already adopted a 1635 or later reading the tendency of the collator--especially at first, before the importance of certain readings had become obvious--was to pass over the agreement of the manuscript with this later reading in silence. In all important cases I have verified the reading by repeated reference to the manuscripts, but in some of smaller importance I have been content to record the general trend of the evidence. I have tried to cite no manuscript unless I had positive evidence as to its reading.