The Rev. Edward Hincks on a Tablet in the British Museum. 41 



" One" is represented by ^yy ^ ish . din, both in the inscriptions at Perse- 

 polis and at Khorsabad. The form more commonly used, Y«-TT, is interchanged 

 with the foregoing word at both places in different copies of the same text. I 

 am, therefore, bound to suppose that this word signifies " one," but I can com- 

 pare it with nothing else either in Assyriac or in any other language. The second 

 part of the other word, which I have stated to be more common, is used for 

 yj ^y^ a . di " together with," which I believe to be the genitive of TinX. I 

 therefore, suppose that adu (for akhdu), or in construction ad (for akhd), may 

 represent " one" also, but only in the masculine ; the feminine is represented 

 at Behistun by ?E.<^y, that is, ill (for ikhd). The same change of the initial 

 vowel occurs in Arabic. 



I cannot offer any conjecture as to the number expressing " two." 



I have already given my reasons for thinking that ashlata = yyy J^yyy, that is 

 " three," in the feminine. The context in which it occiu's proves that it is a nume- 

 ral, and not a very large one. It has some resemblance to the Hebrew root for 

 " three." The masculine form would be ashla or asJial, 7ti'S, which contains 

 what Gesenius considered the essential or original part of the Hebrew root ; 

 that which connects it with the Irish Celtic ^r/, with the Polynesian talu, and 

 with hundreds of intermediate languages and dialects. The prefix i^ is the 

 same that we have in the following numeral. I feel, on the whole, great confi- 

 dence in this being the genuine Assyriac numeral for " three," and, at the same 

 time, an interesting relic of a language anterior to Assyriac. I have already 

 given arha' and arba'at for " four" — as to tliese forms I think there can be no 

 doubt. 



Colonel Rawlinson has given from an Assyrian Tablet |[|^ j^yy >-Ay-«^ with 

 the transcription khamisti ("Early History of Babylonia," pages 5, 6); but there 

 seems to be a mistake. The three characters given would be read kha . is . ti, 

 without an m. Possibly the printer has omitted Y- mi. If the word be printed as 

 it stands in the Tablet, it must be read klia . mil . ti ; the second character having 

 the value mil, as well as is. I incline to think that this is the correct reading; 

 for I is often used for s before a dental. Whichever be the true reading, the 

 form does not harmonize with 'isirat, " ten," which precedes it. From the loose 

 way in which Colonel Rawlinson speaks, it is impossible to tell whether the 

 numerals that he gives on page 5 are from the same, or from different Assyrian 



VOL. XXIII. !■■ 



