The omission of the letter led Lassen into one of the greatest blunders in his revised translation. His ‘progenies Arçis’ commits him to a definite historical error, while the rendering of the following words ‘ârija ârija’ is merely an instance of aberration to which the greatest scholars are occasionally subject. Rawlinson, who greatly excelled him in ‘intuition,’ had no difficulty in supplying the missing letter as _p_, and he translated the passage correctly: ‘son of a Persian, an Arian, of Arian descent,’ in the place of ‘Progenies Arçis, a venerabilibus stirpis auctoribus oriundi’! Lassen’s knowledge now enabled him to point out several instances in the Inscription of Artaxerxes Ochus that served to illustrate the decay of the language, though the interval from the classical age of Darius was not more than a hundred and ninety years. It is here that the two new signs—or rather contractions—for the syllables ‘dah’ and ‘bumi’ first occur. Of more interest is the evidence this inscription affords of the degeneration of the Persian religion by the admittance of Mithra into its worship. Artaxerxes the Third traces his genealogy through Artaxerxes the Second (Mnemon), Darius the Second (Nothus), Artaxerxes the First (Longimanus), Xerxes, Darius the First, and Hystaspes, to Arsames the Achaemenian; and neither of the two last are distinguished by the royal title. The most important publication after Lassen’s essay in 1844 was a criticism that appeared upon it by Adolf Holtzmann in the following year.[531] It was written with much personal animosity to Lassen, and this enlivens in an amusing fashion the extreme aridity of the subject-matter. Only two letters now remained to be correctly determined: 19 (𐎮), the _k’h_ of Lassen, and this Holtzmann successfully accomplished. The letter occurs in the words Lassen transliterated ‘jak’hija’ and ‘hak’hi(s).’[532] Holtzmann substituted _d_ and read the first word ‘jadij,’ which he compared with the Sanscrit ‘jadi,’ Zend ‘jedhi’—‘when’—instead of Lassen’s ‘venerandus,’ a meaning that turned out to be correct.[533] Finally, he reviewed all the words in which the letter occurs, and he found that the substitution of _d_ for _k_ enabled him to assign satisfactory meanings to the whole of them.[534] Holtzmann is also credited[535] with having slightly improved the value of 28 (𐎩), the _z_ of Jacquet, by giving it the sound of _g_—presumably _g_ soft, but as it always precedes _a_ the reader would naturally assume it to be hard, as in ‘gadija,’ ‘aga’mija,’ etc. It is in fact _j_ before _a_.