room to spread themselves over the wide expanse of the lake before they can raise the latter to a dangerous level. The change which Suyya effected in the place of junction of the Vitasta and Sindhu, was thus closely connected with a scheme of regulating the course of the united rivers from their confluence to the Volur. It is fully in keeping herewith that Kalhana immediately after mentioning the above change, refers to the stone embankments constructed by Suyya along the Vitasta, and to the regulation of the Volur. In the lines which follow next, the reclamation of the land for new villages is directly mentioned as a result of these operations. Kalhana names particularly (v. 106) as such villages localities known by the name of Kundala. Villages thus designated (Marꞌkundal, Utskundel) are actually situated along the course, which the Vitastå at present follows from Shad'pur to the Volur. 10. Kalhana was evidently anxious to mark with all possible clearness the change which had taken place in the site of the river-junction. He has taken care to let us know in detail of the sacred buildings which stood by the side of the confluence before and after Suyya's regulation. The translation of the couplet, v. 99-100, has been given above as follows: : "On the two banks of the original (?) confluence there stood the [temples of Vispu] Visnusvāmin and Vainyasvumin situated in Phalapura and Parihāsapura [respectively]; whereas on the bank of the present [confluence] which has got to the vicinity of Sundaribhavana, [there stands the temple of] Vişņu Yogaśāyin, the object of Suyya's worship." The temples here referred to are not otherwise known, and we have to rely for their identification on this very passage. The interpretation of the latter is unfortunately complicated by the fact that the first words of verse 99 as found in our two MSS. contain un undoubted corruption. The adhyapyāstāṁ of A, gives no sense, as little as the correction of a later hand in that codex, adyapy. In the absence of something better I had adopted in the Ed. the conjectural reading of the Calcutta and Paris Editors adyapy, but a closer examination of the text shows that this conjecture is untenable. From the use of tu with adyatanasya in the next verse, it follows that Kalhana wants to refer there to the modern confluence in contradistinction from the earlier one. Hence the word adya, ' now,' in v. 99, would be meaningless; nor could it be brought into agreement with the use of the imperfect verb āstāṁ. The proper emendation seems to be indicated by the reading of L, abhyasyāstāṁ. The Aksaras and I, and being distinguished in S'aradă writing only by a slight difference, the restoration of the correct reading ādyasyāstāṁ is rendered palæographically very easy. We thus get the proper contrast between the 'original' (ādya) confluence described in the first verse and the 'present' (adyatana) one in the next, which the context imperatively requires. 11. In our discussion above, we have already shown that the temple of Vainyasvāmin must be identified with the extant temple ruin at Malikpur. The position of the latter corresponds exactly to that indicated in v. 97 for the Vainyasvamin shrine, with reference to the former river-junction. Kalhana, v. 99, speaks of this temple as situated in Parihāsapura,' and in order to understand this description we must refer once more to the topography of this neighbourhood. The Malikpur ruin, as already explained, lies on the narrow neck of raised ground which separates the Trigam swamp from the east end of the Badrihēl Nala. On the opposite or southern side of the latter rises with steep banks the plateau of Par'spor, the ancient Parihasapura, with the great temple ruins described in Note F, iv. 194-204.* See v. 103 sq. and note. The northern group of these temples (A, B, C on map) lies at a distance of less than one mile to the S. of Malikpür. As their ruins are far more conspicuous than the remains at Malikpur, the question might arise whether we ought not to look rather among them for the temple of Vainyasvamin. My reasons against the latter view are briefly the following: In the first place, I see at present no ground for doubting the correctness of the information The great town which King Lalitäditya had founded there, must have given its name at an early date to the surrounding district. We find the term Parihasapura used in this extended sense in Sriv. iv. 352; Fourth Chronicle, 554, and subsequently the Pargana of Paraspür' in Abu-l-Fazl's list of territorial divisions of Kasmir (Ain-i Akb., ii. p. 370). The villages around the Par spōr Udar continued, in fact, until some sixteen years ago to form a separate Pargana, which was officially known by the naine of Partspor, and is also correctly shown on the Survey map. It is now interesting to note that Malikpur, according to the uniform evidence of the villagers and small local officials examined by me, was included in the Par spōr Pargana, whereas already Trigam along with Shad'pur and other neighbouring villages belonged to the Pargana known as Sairu-l-mawazi' Pain. In view of the tenacity with which the old territorial limits have maintained themselves throughout in Kasmir until the recent changes in the administration, we have every reason to assume that the location of Malikpur in the Pargana of Partspōr is of old date. If, then, the Malikpur ruins mark the site of the Vainyasvamin temple, it is clear that Kalhana was justified in speaking of the latter as 'situated in Parihasapura.' 12. In order to ascertain the probable position of the second shrine which Kalhaņa mentions by the side of the former confluence, that of the temple of VISNUSVAMIN, we must return once more to the neighbourhood of Trigăm. As the Visņusvamin temple, is distinctly said to be situated in Phalapura, while the Vainyasvamin shrine opposite to it lay in Parihasapura, it is quite certain that we cannot look for the former on the Parꞌspōr plateau. Accepting Malikpur as the site of the temple of Vainyasvamin, the above indication must necessarily lead us to search for the position of the Vişņusvamin temple on that strip of raised ground which stretches to the S.W. of Trigam along the western bank of the Trigam swamp. Proceeding from Gund-i Khalil, the central hamlet of Trigam, in the above-marked direction by the path which leads towards Divar, we pass at short distances two old sites, known as Kanetsuth Masjid and Utar Khāv. There the foundations of small temples can still be traced. Going further in the same direction we reach a narrow neck of raised ground which projects between the Trigam swamp and the little lake known as Udan Sar. Immediately below it to the south is the embankment of the Kanyesuth stretching across the Badrihēl Nala, as described above. given to me as regards the traditional name of the Malikpur ruin (see above, para. 5). Secondly, it must be remembered that, as shown in Note F, all the chief ruins on the Par spor plateau must belong to the great sacred buildings which Lalitaditya constructed thore. Now it is worth noting that the ruined mound A, which falls nearest to the old place of junction, and accordingly would have to be thought of in the first place under that assumption, is by far the most extensive of all the ruins of Parihāsapura. If this had been the Vainyasvamin temple, we could scarcely explain why the latter name does not occur in the lengthy account Kalhana gives in iv. 194-214 of the numerous structures erected at Parihāsapura by Lalitāditya and his court. Thirdly, attention must be paid to the fact that the position of the Parihasapura ruins relative to the old river-junction is not the one which we should expect to be taken up for a shrine erected in honour of a sacred Saingama, This And that the Vainyasvamin temple had this The same position is assigned to the most sacred of Samgama Tirthas, that of Prayaga at Allahabad; see CONSTABLE'S Hand Atlas of India, pl. 44. On the top of this projecting neck I found scattered a quantity of large carved slabs and architectural fragments, which belonged undoubtedly to some ancient building. They are found chiefly near a spot where the outlines of a square enclosure or building can still be partly traced in the form of foundation walls. This spot, which seems at one time to have been used as a Ziarat and burying-ground, is now known to the villagers as Timbar Shāhun Marguzar ('the cemetery of Timur Shah'). It appears to me very probable that the temple of Viṣṇusvamin stood once on or close to this spot. The position corresponds exactly to the indications furnished by Kalhaņa's words. On the opposite or eastern side of the Trigăm swamp, which represents the former bed of the Sindhu (see above, para. 6), we have Malikpur with the remains of the Vainyasvamin temple, the direct distance between the two sites being scarcely more than half a mile. The scattered condition of the remains of Timbar Shahun Marguzar and their comparative insignificance is easily accounted for. The Kanyesuth which, as already stated, is constructed entirely of ancient stone materials, lies immediately below this site. It is evident that the latter, which of all the ruined sites lies nearest to the Kanyesuth, has supplied most, if not all, the sculptured slabs, etc., of which this embankment is composed. 18. The temple of Viņņusvamin was according to Kalhana's words situated in PHALAPURA. It follows from what has been said above as to the similar location of the Vainyasvamin shrine in Parihasapura, that Phalapura is probably used here as the designation of a cmall territorial division., Trigam has, as already explained, always been counted with the riveraine Pargana of Sairu-l-mawazi' Pain (map Salimozapaieen'), and Phalapura must hence be assumed to correspond to the latter tract or a part of it. Phalapura is referred to in iv. 184, 673, apparently as a town or village founded by Lalitaditya, and is placed by the gloss on the second passage, 'in the neighbourhood of Parihasapura.' This is very likely the original application of the name. Just as the town of Parihasapura gave its name to the Pargana of Partspor, so also the name of Phalapura evidently came to be used as the designation for a small district. The subsequent creation of a larger territorial division known by the Persian torm Sairu-l-mawāzi ' would account for the disappearance of the name Phalapura.7 6 14. It remains now for us only to examine the reference which Kalhana makes in v. 100, to the confluence of the two rivors such as it existed in his own time. We have already in the introductory remarks of this note indicated the evidence which proves that the river-junction which Kalhana know, has remained unchanged to the present day. It is, therefore, but a matter of minor importance that the locality of Sundaribhavana which Kalhana names as in the vicinity of the modern junction, can no longer be identified. The place is nowhere else mentioned, and I have not been able to trace in that neighbourhood a local name which might be derived from it. The temple of Viṣņu YOGASAYIN, which our verse mentions as situated by the side of the new junction and evidently as a foundation of Suyya, is also otherwise unknown. But its site is, perhaps, indicated by the remains of an ancient temple found on the right bank of the Vitasta, a little below its actual junction with the Sind and close to the park known as Naran Bagh. These remains have, during the late Mahārāja's reign, been built up into a Compare regarding the origin and significance of this term lin-i Akb., ii. p. 367. 7 TROYER and LASSEN (iii. p. 1002) evidently following WILSON, History, p. 63 note, identify Phalapura with Shahabuddinpur or Shād'pur. But the passage of the Ain-i Akbari to which they refer (ii. p. 364), merely mentions Shahābuddinpur as situated at the junction of the Bihat (Vitasta) and Sind Rivers. The modern village Pālapūr, about three miles below S'rinagar on the right bank of the Vitastā, which Wilson in his note on Moorcroft's Travels, ii. p. 219, took for Phalapura, cannot be thought of in this connection. It lies too high up the river, and its name is clearly derived from Pula, a Kram name widely spread among Kasmiri Muhammadans; see LAWRENCE, Valley, p. 307. solid platform to form the base of a new temple which Pandit Rajakāk Dar, of S'rinagar, intended to erect here. Judging from the character of the ornaments displayed on many of the slabs, these materials could well have belonged to a temple of Avantivarman's time. The Purohitas of the neighbouring Gayatirtha knew nothing of the origin or name of this temple, but had heard by tradition that it was dedicated to Viṣņu. 15. The spot which is actually held to mark the Tirtha, and where accordingly the ablutions and sacrifices take place, lies about 200 yards from the remains described, at the very angle formed by the uniting rivers. Opposite to this point there rises in the bed of the Vitastă a small island built of solid masonry. On it stands an old Cinar tree, and under the branches of the latter are placed a large Linga and a few old sculptures. The Tirtha to which this little island belongs is visited by numerous pilgrims at certain Parvans throughout the year, and is now known through the whole of Kasmir by the name of Prayāga. It is easy to explain how this name came to be attached to this particular Tirtha. The confluence of the Vitasta and Sindhu Rivers has, as the Samgama of the two chief rivers of Kasmir, enjoyed great sanctity evidently since early days. It is mentioned in the character of an important Tirtha by the Nilamata, vv. 297 and 1076. The first passage (Gangā Sindhus tu vijñeyā Vitastā Yamunā tathā | sa Prayāgasamo deśas tayor yatra tu saṁgamaḥ) identifies the Sindhu River with the Gangå,' and the Vitasta with the Yamuna. In accordance with this identification the term Prayāga, which applies to the confluence of the great rivers of the Indian plains, near Allahabad, has been transferred to the junction at Shad'pur. The old Cinar on the above described little island has taken the place of the famous Ficus Indica tree of the real Prayaga (see iii. 430). The only text in which the name Prayaga is actually given to the Shad'pur junction, is the Vitastāmāhātmya, xxi. 74. The recent date of this composition, or at least of its extant redaction, is sufficiently proved by the modern form of many of the local names found in it, e.g. the name Sāradāpura (!) given to Shad'pur itself. The passages, vi. 305 and vii. 214, in which Kalhana mentions the erection of shrines, Mathas, and other sacred buildings at the junction of the Vitasta and Sindhu by Queen Diddă and King Ananta, must, of course, refer already to the new junction created by Suyya. A pilgrimage to it is alluded to in viii. 3149. Also Mankha in his description of Kasmir, Srikanthacarita, iii. 20, does not fail to advert to this Samgama as a spot dear to S'iva. General references to the locality are found, Rājat. vii. 909, 1595; viii. 506; Sriv. i. 441. [It is curious to note that also tho Tirtha of the real Prayaga near Allahabad has quite recently been shifted by an artificial change of the confluence of the Ganga and Yamunā. I take the following from the Pioneer of January 27th, 1898 : "Any apprehensions which may have been entertained as to the acceptableness to the Magh Mela pilgrims of the artificial confluence of the Ganges and the Jumna recently constructed under the orders of the Government, have been entirely dispelled by the event.. It will be remembered that the purpose in cutting a canal from the Ganges to the Jumna was, by establishing a temporary confluence four miles nearer to Allahabad than the natural confluence, to facilitate the otherwise impracticable task of supplying filtered water from the municipal reservoirs to the huge assembly which gathers every year on the apex of shifting sand formed by the two rivers. So far from rejecting this artificial confluence, the pilgrims have utilized it to the almost entire exclusion of the natural confluence; and although on the great day of the Solar Eclipse the immense throng of bathers proved too heavy a tax on the limited capacity of Compare for a description of the Prayaga island at Shad'pur, HÜGEL, Kaschmir, i. p. 331; VIGNE, Travels, ii. pp. 96, 146; INCE, Handbook, p. 207. The Cinar tree of the island, according to a popular legend, does not grow in size or decay; comp. Tirthas., also the Rev. HINTONKNOWLES' Dictionary of Kashmiri Proverbs, p. 173. 9 Compare regarding this identification, which is also alluded to by Kalhaņa, note i. 57. the work, the surplus thousands overflowed peaceably along the northern bank of the Jumna, east and west of the canal, content to perform their ceremonial ablutions in sight of, if not in contact with, the actual mingling of the two waters. Thus a difficult measure has been successfully carried through; and the action of the Government, which might ignorantly or mischievously have been misinterpreted, has been recognized as a benefaction." It is evident that the pious visitors of the Kasmir 'Prayaga' must have acquiesced with equal ease in the transfer of their Tirtha consequent on Suyya's regulation.] NOTE J.-V. 152-155. THE S'AHI OF UDABHĀŅĻA. 1. The historical data furnished by this interesting passage together with other available information regarding the S'ahi dynasty have been fully discussed by me in my paper Zur Geschichte der Çāhis von Käbul, contributed to the 'Festgruss an Rudolf von Roth. Zum Doctor-Jubilaeum 24. August 1893, von seinen Freunden und Schülern,' Stuttgart (Kohlhammer), pp. 198-206. I have shown there that the rulers, which in Kalhaņa's narrative figure as the S'AHIS of UDABHAṇḍAPURA, belong to the Hindu Shāhiya dynasty' which we know from a famous passage in Albérūnî's India (ii. p. 13) to have ruled the Kabul Valley and Gandhara previous to the conquest of Maḥmūd of Ghazna. According to Alberuni's account this dynasty succeeded to the Turkish Shāhiyas,' who were traditionally believed to have ruled over the Hindu kingdom of Kabul for sixty generations. The last king of this race, Lagatūrman, was deposed by his Brahman Wazir who, having risen to great power and influence, seized the royal throne and became the founder of the Hindu Shahiya dynasty. 2. In this usurper who, in the extant text of Albērūni is called Kallar, we have in all probability to recognize the LALLIYA S'AHI of the Rajatarangiņi. Kalhaņa subsequently, v. 233, mentions Lalliya as the father of Kamaluka. The latter prince is certainly identical with the Kamali who, in Albērůni's genealogy of the Hindu Shahiyas, appears as the ruler next but one after 'Kallar.' The 'S'ahi,' whom Kalhana mentions in the last-named passage, without giving his name, as Kamaluka's predecessor, is probably the Sămand (Samanta) who in Alberūni's list follows immediately after Kallar. The description which the Chronicle gives of Lalliya S'ahi's great power and repute, agrees singularly with what Albērūni has to tell us of the energetic founder of the Hindu Shahiya dynasty. The proposed identification of 'Kallar' with Lalliya finds further support in the convincing conjecture by which Prof. Ch. Seybold, in his remarks on the above quoted paper (Zu Birūni's Indica, Z. D.M.G., xlviii. p. 700), has explained the apparent difference of the names. Prof. Seybold sees in the form of the single MS. which has preserved for us the text of the Indica, a misreador (i.e. Lalliya), and accounts for such a corruption by well-known paleographic peculiarities of Arabic manuscripts. [At the time of sending this note to the press, I notice that the above identification of 'Kallar' and Lalliya had suggested itself already to General Cunningham; see Archeol. Survey Reports, v. p. 83.] 3. Besides Kamaluka and the nameless S'ahi, the Chronicle mentions of Lalliya's successors BHIMA S'AHI, of Udabhāṇḍapura, vi. 178, vii. 1081, the grandfather of Queen Diddă, |