The spelling of Arabic names varies greatly in all editions, and I have made no alterations. It should be explained that only three vowels are recognized in Arabic, and that some of the consonants have no equivalents in English. The general practice of orientalists in recent years has been to adopt one of the various sets of conventional signs for the letters and vowel marks of the Arabic alphabet, transliterating Mohamed as Muhammad, muezzin as mu'edhdhin, and Koran as Qur'an or Kur'an. This method is useful to those who know what it means but this book follows the old fashion of writing the best phonetic approximations according to ordinary English spelling.

He then prints a list of publisher's queries re spelling, and T. F. Lawrence's answers; for instance: .Query: "Slip [galley sheet] 20. Nuri, Emir of the Ruwalla, belongs to the 'chief family of the Rualla'. On Slip 23 'Rualla horse', and Slip 38, 'killed one Rueli'. In all later slips 'Rualla'." .Answer: "should have also used Ruwala and Ruala." .Query: "Slip 47. Jedha, the she-camel, was Jedhah on Slip 40." .Answer: "she was a splendid beast." .Query: "Slip 78. Sherif Abd el Mayin of Slip 68 becomes el Main, el Mayein, el Muein, el Mayin, and el Muyein." .Answer: "Good egg. I call this really ingenious." .If such are the difficulties of transcribing modern Arabic, confusion becomes worse confounded when orientalists turn to mediaeval texts, which pose additional problems owing to mutilations by careless copyists. The first English translation of "Ebn Haukal" (or ibn-Hawkal) was published AD 1800 by Sir William Ouseley, Knt. LL.D.*[Ibn Hawkal wrote his book in Arabic, but Ouseley translated it from a Persian translation.] In his preface, Sir William, an eminent orientalist, uttered this touching cri de cour:

Of the difficulties arising from an irregular combination of letters, the confusion of one word with another, and the total omission, in some lines, of the diacritical points, I should not complain, because habit and persevering attention have enabled me to surmount them in passages of general description, or sentences of common construction; but in the names of persons or of places never before seen or heard of, and which the context could not assist in deciphering, when the diacritical points were omitted, conjecture alone could supply them, or collation with a more perfect manuscript.... .Notwithstanding what I have just said, and although the most learned writers on Hebrew, Arabick, and Persian Literature, have made observations on the same subject, it may perhaps, be necessary to demonstrate, by a particular example, the extraordinary influence of those diacritical points [frequently omitted by copyists]. .One example will suffice - Let us suppose the three letters forming the name Tibbet to be divested of their diacritical points. The first character may be rendered, by the application of one point above, an N; of two points a T, of three points a TH or S; if one point is placed under, it becomes a B - if two points, a Y and if three points, a P. In like manner the second character may be affected, and the third character may be, according to the addition of points, rendered a B, P, T, and TH, or S.*[The original of this quote is enlivened by letters in Persian script, which I have omitted in kindness to the publishers.]