Thence forth descending to that perilous porch Those dreadful flames she also found _delayed_. F. Q. iii. 12. 42. Here the poet probably wrote _allayed_, and the printer transferred to it the _d_ of _found_. See on Temp. iv. 1. While treating of elision, it may not be useless to remark that when a word beginning with _h_ is monosyllabic, or is not accented on its first syllable, the _h_ is not sounded. Any one who will observe will find that _his_, for example, is usually pronounced _is_; so that there is no occasion for the _'s_ of the dramatists. So we should write and pronounce _a history_, but write _an historian_ and pronounce _an 'istorian_; for _a historian_, as it is too often written and pronounced, makes a most unpleasant _hiatus_. We may observe how constantly the aspirate is suppressed in the poetry of Greece and Rome. The errors of transcribers and printers are _Omission_, _Addition_, _Transposition_, _Substitution_. Of these I will now give instances, chiefly from my own experience. I must, however, previously notice the rather curious fact, that these four sources of error, which I had traced out in our printed works, are all, and no other, acted upon by that, in my mind, most able of the German critics, J. Olshausen, in that _chef d'oeuvre_ of criticism, his Comment on the Book of Psalms, thus showing how universal they are. 7. _Omission._--If any one will examine a proof-sheet as it comes from the hands of a compositor, he will find abundant instances of this source of error. One would think that such could hardly escape the author's eye; and yet, in the edition of Ben Jonson's plays corrected by himself, almost the only errors detected by the editors have been those of omission; and a few more have, I think, escaped _them_. I may add that the words supplied by them seem to me to be almost invariably the very words which had been omitted. In Notes and Queries (3 S. vii.) there is a list of the principal _errata_ in the reprint of the First Folio, in 1808; and the far larger portion of them are omissions which can easily be supplied. From my own experience in the case of reprints, I can further assert that corrections of this kind are by no means a matter of hap-hazard; for where I have supplied the words I supposed lost or altered I have almost invariably found, on referring to the original edition, that I had hit on the exact word. As an instance, in a late reprint of Akenside's Poems, the following line occurs in the Hymn to the Naiads: Your _sultry_ springs, through every urn. Here I at once corrected _salutary_, and on looking at the original edition I found I was right. Yet, had this correction been made in Shakespeare, we know how it would have been disputed. As to omissions I can give the following instances. In a reprint (Lond. 1816) of Fletcher's Purple Island, we find (xii. 74, 85) the following lines: