There is, however, the element of uncertainty of which I have spoken above.
We cannot be quite clear what text the writer had before him.

This difficulty also exists, though to a less degree, when we come to consider quotations from the New Testament in writers of an early date whom we know to have used our present Gospels as canonical.
The text of these Gospels is so comparatively fixed, and we have such abundant materials for its reconstruction, that we can generally say at once whether the writer is quoting from it freely or not.

We have thus a certain gain, though at the cost of the drawback that we can no longer draw an inference as to the practice of individuals, but merely attain to a general conclusion as to the habits of mind current in the age.

This too will be subject to a deduction for the individual bent and peculiarities of the writer.

We must therefore, on the whole, attach less importance to the examples under this section than under that preceding.
I chose two writers to be the subject of this examination almost, I may say, at random, and chiefly because I had more convenient access to their works at the time.