AXUMITE COINS 417 



country, and those minted afcer that event. The pagan pieces bear 

 on the margin the symbol of a crescent and globe : the earliest 

 known of these is attributed to King Aphilas, but his identity and date 

 cannot be traced. In the Christian coins the crescent and globe are 

 replaced by crosses, which are generally placed at the four cardinal 

 points, and interrupt the legend in an arbitrary way, adding to the 

 difficulty of interpreting it. The coins of King Okhsas are the first 

 inscribed with the cross. It was probably to his sons, Aeizanas and 

 Sazanas, that the Emperor Constantius II., 337-361, addressed a 

 letter in 356 which is still extant. There are coins of Aeizanas, both 

 when he reigned in conjunction with his father, and as sole monarch 

 after Okhsas's death. He was the author of the inscribed stone dis- 

 covered by Salt in 1805. 



Tliese conclusions, so far as the evidence supplied by 

 the coins alone is concerned, seem clear and satisfactory, 

 but when examined by the side of the bilingual inscription 

 found on the stone of King Aeizanas, they present 

 serious difficulties. Of the two inscriptions, the Semitic 

 or Sab^ean one is much the lonoer, but, as stated 

 above, the greater part is so much obliterated, that only 

 about one-third of the whole can now be deciphered. 

 Salt copied it to the best of his ability ; but, as he was 

 totally unacquainted with the language, his transcript 

 was of no practical value. The Greek inscription he 

 transcribed correctly, and the interpretation by Dr. 

 Vincent, which he gives in the third volume of Valentias 

 Travels (p. 184) leaves nothing to be desired ; nor have 

 the conclusions he draws from it been upset, but rather 

 confirmed, by subsequent research. It was not till 1893, 

 when Mr. Bent took some careful impressions of the 

 Sabsean inscription, that Professor Miiller of Vienna was 

 able to decipher and translate that part which still 

 presents coherent words and sentences, proving that the 



2 E 



