[Footnote 33: Elmacin (p. 343, 344) assigns this probable
number, which is reduced by Abulpharagius to 15,000, (p.
227,) and by D'Herbelot (p. 102) to 12,000 horse. But the
same Elmacin gives 300,000 met to the emperor, of whom
Abulpharagius says, Cum centum hominum millibus, multisque
equis et magna pompa instructus. The Greeks abstain from
any definition of numbers.]