larity of the Zend word Athro is obvious, and we need therefore seek no refined subtlety in admitting it to worship as the etherial essence of the sun, since it can with so much more simplicity be understood as a common denomination of the solar orb itself. It should be remarked that the effigy of APAHOРo, like that of Aero, has flames on his shoulders.

ΑΘΡΟ,

I will not stop to inquire, whether the change from the Sanskrit OKPO (Arka), to the Pushtú or Zend Apo (Aita), has any possible connection with a parallel change in the family designation of the Saurashtra princes who were, in the first centuries of the Christian era, marked by the affix 'Bhatárka,' ('cherished by Arka,') but afterwards, for a long succession of reigns, were known by the surname of Aditya; but will proceed to describe the immediate contents of the plate now under review.

Figs. 1-3 are placed at the head of the series, because in them the head bears the nearest analogy to its prototype. In fig. 1, indeed, the letters behind the head may be almost conceived to belong to KWAOY. In the centre of the reverse is the so-called Chaitya symbol; which, had it only occurred on these descendants of a Mithraic coin, I should now be inclined to designate a symbol of the holy flame, trilingual and pyramidal, of the Sassanian fire-worship. The marginal writing may with certainty be pronounced to be an ancient form of Sanskrit; but I cannot attempt to read it. In figures a, b, c, I have copied the lines from three other coins, and have thus ascertained that a portion of the legend is the same in all, while the remainder varies. The former doubtless comprehends the regal titles; but in it there is no approach to the ordinary Indian terms of Kája, Rao, etc. The six parallel letters may be read परकृतपस.

Figs. 4 and 5 differ from the preceding in the central device, which now bears a rude resemblance to the human figure. The letters and general execution are very imperfect.

Figs. 6-9 are one step further removed from perfection. The legend, where best preserved, as in fig. 9, appears a mere repetition of the letter p, with the suffix r, ri, and y. There are three letters behind

the head in fig. 7, which may be taken either for corrupted Greek, or for the Pehlví of the Sassanian coins, ouo. The central symbol has the form of a trident. Lieut. Burnes informs me that several hundred of these three species of coins were found in Katch in 1830 in a copper vessel buried in the ruins of Puragarh, twenty miles west of Bhoj, a place of great antiquity, and yet marked by the ruins of a palace and a mint.

Figs. 10-12 are of a different type, though nearly allied to the former; they are not only found in Gujarát, but at Kanauj, Ujjain, and generally in Upper India. Lieut. Cunningham has just sent me impressions of five very well-preserved specimens procured at Benáres, on which, in front of the face, are seen some letters, very like the Pehlví character, S. The Sanskrit, too, is not of the elongated form of the upper group, but exactly like that of Wathen's Gujarát inscriptions. Not having yet succeeded in deciphering them, it is needless to copy out the mere letters at present. The symbol in the centre will be recognised as the peacock, sacred to Kumára, the Mars of the Rájputs, alluded to in the preceding observations.

[I extract from the Jour. As. Soc., Beng.', (vi. of 1855, N.S., No. lxxvi.), my decipherment of these legends, together with some further observations on the coins and their local associations::

CL. 1. KUMÁRa Gupta, xxvii., figs. 10-12.

LEGEND - देवं जयति विजितावनिरवनिपति कुमार गुप्तो

Dev(o) jayati vijitávanir avanipati(h) Kumára Gupto.

His Majesty, Kumára Gupta, who has subdued the earth, rules.'

CL. 2. SKANDA GUPTA, J.R.A.S.', xii., pl. ii., figs. 52, 53.
LEGEND —– विजितावनिरवनिपति जयति देव स्कन्द गुप्त य
Vijitávanir avanipati(r) jayati deva(h) Skanda Gupta-y.

CL. 3. BUDHA Gupta, ‘J.R.A.S.', xii., pl, ii., figs. 55. 57.

LEGEND — देव जयति तिजितावनिरवनिपति श्री बुध गुप्तो

Dev(o) jayati vijitávanir avanipati(h) Srí Budha Gupto.

When once fairly deciphered, these legends will be seen to present but few difficulties. The lapidary inscriptions have already proved that the Gupta artists indulged in faulty Sanskrit orthography as well as in grammatical errors, so I need not detain my readers by any comments upon minor imperfections, while the general sense of the legend is sufficiently clear. I must mention that, in my Devanagarí transcripts, I have adhered servilely to the original legends impressed upon the coins; the version in the Roman type is corrected up to Sanskrit requirements.

There is a superfluous (or possibly an ) after the on Skanda Gupta's coins,

the use of which is not apparent, but which clearly takes the place of the final o in Kumara's legends.

I may note that Kumára Gupta's coins display both the old form of and the more specially Gupta outline of that character. ('J.A.S.B.', iv., pl. xlix., figs. 10 and 12.) The is also seen in its transition state from the triple-lined letter of early days to the almost modern form; while, at times, it appears on Skanda's money as a character not easily distinguishable from the later Kumára Я just adverted to. This extensive modification of the , in the numismatic alphabet, is the more curious, as the corresponding lapidary character retains all the essentials of its ancient outline throughout the Gupta inscriptions, from the Allahabad pillar to Budha Gupta's record at Eran; and even on to Toramana's inscription at the same place.

The weights of these coins run as follows:

Nine fair specimens of Kumara's mintages average 30.1 grains: highest weight, 33 grains.

Eight fair specimens of Skanda's coinage average 31.7 grains: highest weight, 35 grains.

Dr. Swiney's coin of Budha (No. 55, pl. ii., 'J.R.A.S.', xii.) weighs 32 grains. Their relative rarity may be approximately inferred by the numbers of the coins of each prince in Col. Stacy's, my own, and Mr. Freeling's collections.

These coins of the 'peacock' type would seem to have formed the recognised silver currency of the central and eastern provinces of the Gupta dominions; for, though they are found in limited numbers amid the hoards of the local coinage of the western states, the relative proportions seem to be reversed in the 'finds' of the eastern districts of the Gupta empire. Those in my own collection have chiefly been obtained from Kanauj itself, while Mr. Freeling's centre of operations is Hamirpur on the Jamná; but all his nine pieces were obtained from the eastward of that river.

It may be useful to summarize the proved dates, discovered on the various specimens of the 'peacock' coins; the consideration of the grounds whereon these determinations are based is to be found at p. 551, 'J.A.S.B.', No. vii., 1856.

Of Kumara's coins, some bear ciphers for one hundred, twenty, and one = 121; others display figures for 124; one piece discloses a unit for 5, and another a sign which I doubtfully class as a 9, but in both these instances the decimal cipher is obliterated.

Skanda Gupta's money continues the series in the use of the same figures for one hundred, with the addition of two new symbols in the decimal place.

Budha Gupta, in like manner, dates in the first century of the given era, but the

value of his second cipher is undetermined.

His inscription at Sanchi, it will be remembered, bears date 165.

CL. 4.--TORAMÁNA.

Silver: wt. 32 grains : unique and unpublished.