NIMSAR, OR NIMKAR-BARIKHAR, OR VAIRATKHERA. 351

partly brick and partly kankar blocks, which betray their origin by their carvings and by the presence of the Swastika symbol, or mystic cross. The walls were originally of brick, but they have long ago disappeared, and the only parts of the old fort now standing are the gateway and the Shah Búrj. The foundation of the latter is, however, of Hindu construction, and as there are many carved bricks lying about, I presume that it was a temple. The fort is provided with a well 81 feet broad and 514 feet deep to the water level.

The tradition of the place is, that the building of the fort was finished on Friday, the 9th of the waxing moon of Chaitra, in the Samvat year 1362, or A. D. 1305, by Háhâjál, a renegade Hindu, who is said to have been the Vazir of Ala-ud-din Ghori. For Ghori we must read Khilji to bring the King's name into agreement with the date, and as the people are in the habit of styling all the Pathans as Ghoris, the alteration is perfectly allowable. But who was Háhájál? As a renegade Hindu and the Vazir of Ala-ud-din, he might perhaps be the same person as Kafur, who in A. D. 1305 was appointed as Malik Naib to the command of the army for the conquest of the Dakhan. I procured several of Alaud-din's coins at Nimsar, and in his reign I conclude that the fort passed from the hands of the Hindus into those of the Musulmans. The original fort is said to have been as old as the Pândus; and if the derivation of the name of the place has been truly handed down, it must have been occupied even earlier than the time of the Pândus.*

XXII. BARIKHAR, OR VAIRATKHERA.

Barikhar is the name of a village on the top of an extensive old mound called Vairátkhera, which is situated on the high road between Nimsar and Pilibhit, at 42 miles from the former, and 68 miles from the latter place. Barikhar is said to be a corruption of Bariyakhera, or Vairát-khera, and its foundation is attributed to Vairat Râja in the time of the Pândus. The ruined mound is 1,000 feet in length at top from east to west by 600 feet in breadth, and from 16 to 20 feet in height. But the dimensions at the base are much more, as the slope is very gentle, being 200 feet in length on

* On the opposite bank of the Gumti there is an old mound called Ora-jhar, and Oradih, as well as Benunagar, which is said to have been the residence of Benu Raja.

the north side, where I measured it. This would make the base of the mound about 1,400 feet by 1,000 feet, which agrees with the size of 50 bigahs, or 1,400,000 square feet, which is popularly attributed to it by the villagers themselves. But the fields are strewn with broken bricks for upwards of 1,000 feet to the northward, and for 500 or 600 feet to the eastward, where there are the remains of several temples. The area actually covered by ruins is not less than 2,000 feet square or upwards of 13 miles in circuit, which shows that Barikhar must once have been a good sized town, but I strongly doubt the story of the Brahmans which attributes its foundation to Vairât Raja. The name is written by the people themselves Badishar, although it is pronounced Barikhar, and I believe that similarity of sound alone has led to the identification of Barikhar with Bariyakhera and Vairât Raja.

XXIII. DEORYIA AND DEWAL.

I couple these two places together because they actually form parts of the old nameless capital of the Bachhal Rajas, who ruled over Eastern Rohilkhand and Western Oudh before the time of the Katehriyas. Dewal itself is a small village which has received its name from a temple in which is deposited a very perfect inscription dated in Samvat 1049, or A. D. 992. The opposite village is called Ilahábás by the Muhammadans, but this name is scarcely known to the people, who usually call it Garh-Gajana. The inscription is chiefly remarkable for the clean and beautiful manner in which the letters have been engraved; and its perfect state makes it the more valuable, as it furnishes us with a complete specimen of the alphabet of the Kutila character in which it is said to be engraved. James Prinsep gave a specimen of the characters along with a translation of the inscription in the Asiatic Society's Journal for 1837, page 777. But the copy from which he framed his alphabet was made by hand, and although it is wonderfully accurate as a mere transcript of the words, yet it is very faulty as a copy of the individual letters. This is the more to be regretted as the alphabet thus framed from an inaccurate copy has become the standard specimen of the Kutila characters. Now the term Kutila means "bent," and as all the letters of the inscription have a bottom stroke or tail, which is turned, or "bent," to

the right, I infer that the alphabet was named Kutila from this pecularity in the formation of its letters. But this peculiarity was unnoticed by the original transcriber, and conscquently the print types of the Kutila characters, which have been prepared both in Germany and in England, are entirely wanting in this special characteristic which gives its name to the alphabet. The letter 7 and the attached vowels are perhaps the most faulty.*

The village of Dewal is situated 16 miles to the south south-east of Pilibhit, on the west bank of the Kau, or Katni Nala. There are two or three plain brick rooms which are called temples, and in one of these the inscription is deposited; but it is said to have been found amongst the ruins of Garh-Gajana, or Ilâhâbâs, on the opposite bank of the stream. Garh-Gâjana is a large ruined mound, about 800 feet square, which includes two small tanks on the east side; but although it is called a Garh, or fort, it was most probably only the country residence of Raja Lalla, who founded it. The small modern village of Ilâhâbâs is situated close to the south-east corner or Garh-Gâjana, and near it on the the soutli side are the ruins of a very large temple, amongst which the inscription is said to have been discovered. The figure of the Varâha Avatar of Vishnu, which is now in the Dewal temple, was found in the same place. The mound of ruins is 200 feet square at base, but the walls of the temple are no longer traceable, as the bricks and kankar blocks have been carried away by the villagers. I traced the remains of at least six other temples around the principal mass of ruin, but there was nothing about them worth noting. To the south there are two larger mounds, which appear to be the remains of an old village.

The Kau or Katni Nala continues its course to the south for three miles, until opposite the large village of Deoriya, when it turns sharply to the east for two miles to the south end of a large ruined fort with is now called Garha-Khera, or the "fort mound." The Katni Nala here turns to the north, and after running round the three other sides of the ruined fort returns to within a few hundred yards of the point from whence it took its northerly course. It thus forms

* See Plate No. LI. for a photograph fac-simile of this inscription. The translation by Prinsep was published in the Bengal Asiatic Society's Journal, 1837, page 777.

a natural ditch to the old stronghold of the Bâchhal Rajas, which is only approachable on the southern side. The fort has been deserted for many centuries, and is covered with dense jungle, in which several tigers have been killed within the last few years. A single cart tract leads to the nearest portions of the ruins which have afforded materials for all the buildings in the large village of Deoriya. The exact extent of the fort is not known, but the position enclosed by the Katni Nala is about 6,000 feet in length from north to south and 4,000 feet in breadth, and the fort is said to be somewhat less than half a kos, or just about half a mile in length. The bricks are of large size, 13 by 9 by 2 inches, which shows considerable antiquity, but the statues of kankar are all Brahmanical, such as the goddess Devi, Siva and his wife, as Gauri-Sankar, and two arghas of lingams. These figures are said to be discovered only in the foundations of the buildings, which, if true, would seem to show that the existing remains are the ruins of Muhammadan works constructed of Hindu materials.

The Katni Nala is an artificial canal drawn from the Mala River near Sohás, 10 miles to the south-east of Pilibhit, and 6 miles to the north of Dewal. Its general course is from north to south, excepting where it winds round the old fort of Garha-Khera, after which it resumes its southerly course and falls into the Kanhaut Nala, about 3 miles to the south of the ruins. Its whole course is just 20 miles in length. All the maps are wrong in giving the name of Katni Nala to the Mala River, instead of to the artificial canal which joins the Mala and Kanhaut Rivers. The canal varies in width from 30 and 40 feet to 100 feet, and even more at the places where it is usually forded. Its very name of Katni Nala, or the "cut stream," is sufficient to prove that it is artificial. But this fact is distinctly stated in the inscription, which records that Raja Lalla "made a beautiful. and holy Katha-Nadi." That this was the Katni Nala, which is drawn from the Mâla River, is proved by the previous verse, which records that the Raja presented to the Brahmans certain villages "shaded by pleasant trees, and watered by the Nirmala Nadi." This name is correctly translated by James Prinsep as "pellucid stream," which, though perfectly applicable to the limpid waters of the Mála River, is evidently the name of the stream itself, and not a