9 c. Pale green, transparent; diameter 0·74 × 0·68 inch, weight 38.6 grains.

On one side a bearded face looking to the right.

On the other side half length figure with wings.

This is probably of Greek manufacture.

There are also two glass discs which at first sight might be mistaken for weights of the same class as No. 9 c.; but on further examination of them they are found to be broken on the top edge, as though they had been originally cast as pendants.

The larger of the two represents an animal suckling two small creatures, which I think represent Romulus and Remus. Above the she-wolf are a star

and crescent.

The other specimen represents a lion passant surmounted by a star and crescent.

These two discs are both of a brownish yellow colour, and transparent.

In the three collections here above described it will be seen that the weights of the majority of the discs correspond very closely with the recognised weights of dînârs and of dirhams, of their multiples and subdivisions. Those which do not so correspond are in some instances badly preserved specimens, having suffered abrasion from some cause; whilst others are probably the representatives of altered standard weights. The two specimens of Greek weights are well preserved, and represent so accurately the weights of the solidus and half solidus, that I think there can be no doubt as to their use; and if we find certain slight discrepancies in the Mohammedan weights, we must take into consideration the rather unsettled state of the Mohammedan empire, and must not expect in Arabic weights to find quite the same accuracy that we meet with amongst those of the more civilised Greek nation.

In the following table I have divided the weights in my collection, and placed their numbers under the subdivisions which, I believe, they respectively represent :

In reply to Mr. S. Poole's first argument, that the "glass discs are circular, thin, flat, and are therefore convenient for currency, whereas weights might be, and are, rings or blocks of metal of any shape," I would direct attention to the two glass discs with Greek monograms in my pos

session, which are represented in Pl. II., Fig. 10 and Fig. 11. They are respectively of the exact weight of a solidus and a half solidus. These discs are in a very perfect state of preservation, and I am inclined to believe that they were standard weights used by merchants and dealers, not for weighing their wares, but for weighing the coins which were received or paid by them.

I believe that weights of this description were in use amongst the Greeks, and even amongst the Mohammedans, so long as Greek coins were current; that when the Khalifah Abd ul Malik had made coins of a purely Mohammedan type, he, or perhaps some subsequent Khalifah, was induced to copy the Greek custom of making glass weights for the purpose of testing the weight of the current coins.

Metal weights, moreover, were not always rings or blocks. The Rev. Greville Chester has shown me a number of old Byzantine weights, which are circular discs of bronze, of different sizes.

For many years I have carefully examined numerous Arabic histories, in the hope of finding some allusion to the use of glass by the Mohammedans as a material of which coins or weights were made at some period of their history. It is only recently that I have found what I have been so long seeking.

When in A.H. 75 or 76 the Byzantine Emperor sent to the Khalifah Abd el Malik ibn Merwan, threatening that he would cause dînârs to be engraved with inscriptions insulting the name of the Prophet Mohammed, the Khalifah was greatly perplexed, and he summoned the chiefs and nobles of the people to advise him how to act. None gave him any satisfactory advice, until Mohammed, son of 'Ali, son of Hussein, spoke as follows: "You shall immediately call the workmen and order them to make dies for dînârs

and dirhams, putting on one side of them the Chapter of the Unity, and on the other side the Mission of the Prophet. Place on the margin of every dirham and dînâr the name of the town and year in which it is struck. Determine the weight of thirty dirhams of the different kinds, take ten whose weight shall be ten mithķâls, ten whose weight shall be six mithķâls, and ten whose weight shall be five mithkâls. The united weights of these thirty dirhams is twenty-one mithkâls. This you will divide by thirty, and the result will be that every ten dirhams should be equal to seven mithkâls. And you shall cast weights of glass, which cannot alter either by increase or by decrease, and you shall make the dînârs to the weight of ten mithkâls, and the dirhams to the weight of seven mithķâls. The dirhams, in those days, were the Kesrawîyeh, which are now called Baghalîyeh, because Ras el Baghl coined them for 'Omar, with the die of Chosroes, in the days of Islam. There is engraved on them the portrait of the

نوش خور king, below the throne is written in Persian

which means 'eat with health.' Their weight before the Mohammedan era was a mithkâl. The dirhams, of which ten were of the weight of six mithķâls, and those of which ten weighed five mithkâls, were called the Samariyeh. Of both light and heavy the inscription was Persian."6

Notwithstanding a passing doubt as to the entire accuracy of this narrative, we cannot help being struck by this allusion to glass weights. I have translated the entire passage, firstly, that it may be the better understood; and,

See Hayât ul Heiwân, vol. i. page 80; the original mention

وتصب صنجات من قوارير - of the glass weights is in these words لاتستميل الے زیاده ولا نقصان

secondly, because M. Queipo states that the dirhams, ten of which were equal to five mithķals, are not mentioned by any Arabic author; and he assumes that D'Herbelot made his calculation, and that he arrived at the result that such dirhams must have existed.7

It must be borne in mind that Demiri wrote this book, Hayât el Heiwân, in A.H. 773, more than seven hundred years after the time at which it is stated that Mohammad, son of Hussein, gave such important advice to Abd el Malik. The striking of purely Mohammedan coins only began in the year 76, and was gradually developed into its subsequent importance. We cannot therefore believe that all the advice attributed by Demîri to Mohammad ibn Hussein was really given by him at first, and before any coins were struck. We must consider that the place of the mintage being found, together with the date on dînârs and dirhams of a later period, and glass weights being at that subsequent date used for weighing them, the author of Hayât el Heiwân, or the author from whom he quotes, assumed that even these subsequent improvements and developments were originally suggested by the same Mohammad ibn Hussein, who probably merely advised the Khalifah to abolish the foreign coinage, and to strike dinârs and dirhams of purely Mohammedan type, in order to circumvent the Greek Emperor who had threatened to coin dînârs containing derisive epithets as applied to the Prophet Mohammed. But at any rate I look upon the allusion to glass coin weights as a confirmation of my theory that these discs were not intended for current coins. Moreover, Ibn ul Athîr mentions special weights for

See Essai sur les systèmes métriques et Monétaires, vol. ii. page 130.