Dr. Bird, relying on some Persian authorities, including the Táríkh-i Sind, tells us that the Súmras, who became first known in the Indian history in the reign of Mahmúd of Ghazní, were originally Muhammadans descended from Aboulahil, an uncle of the Prophet, and that one of the tribe who, in the beginning of the eleventh century of our era, obtained power in Sind, married into the family of Samma, and had a son named Bhaonagar. The chief who had been thus placed at the head of the tribe was named Hallah, the son of Chotah, a descendant of Omar Sumra, first of the family mentioned in their history. Contemporary with Chotah was Deva Ráí, sometimes called Dilu Ráí, the ruler of Alore. “The son born to Hallah had for his descendants Dodar, Singhar, Hanif, and others, who appear to have originally possessed the Dangah per­gunnah in the Registan, or sandy desert, from whence they extended themselves into the pergunnahs of Thurr, Sammawati, Rupah, and Nasirpur.” Dr. Bird adds, that nothing satisfactory regarding them is to be found in any Indian author, except the statement of their descent from the family of the Prophet, in which, therefore, he seems to concur. “They derive their name,” he continues, “from the city of Saumrah, on the Tigris; and appear to have sprung from the followers of Tamim Ansari, mixed with the Arab tribes of Tamim and Kureish.” * * * “In Masudi's time, many chiefs of the Arabs descended from Hamzah, the uncle of the prophet, and Ali, his cousin, were then subject (to the chief of Mansúra.). To these ancestors we may trace the Saiyids of Sinde, and the family of the Sumrahs.”*

The difficulty of solving this question is shown by so confused a statement written by a well-informed author.

Elphinstone observes that, “Kásim's conquests were made over to his successor Temím, in the hands of whose family they remained for thirty-six years, till the downfall of the Ummayides, when, by some insurrection, of which we do not know the particulars, they were expelled by the Súmras, and all their Indian conquests were restored to the Hindús; part of the expelled Arabs, according to Firishta, having found a settlement among the Afghans.” And, again, that “after the expulsion of the Arabs in 750 A.D., Sind, from Bhakkar to the sea, was ruled by the Súmra Rájpúts, until the end of the twelfth century; that it is uncertain when they first paid tribute to the Muhammadans, probably, the beginning of that century, under Shahábu-d dín, or his immediate successor.” Here, the whole period of the 'Abbáside governors, and of the independent rulers of Multán and Mansúra and the Karmatians, is entirely neglected. So important an omission by such a writer teaches us, as in the pre­ceding paragraph, how obscure are the annals with which we have to deal.*

In calling the Súmras Rájpúts, Elphinstone is without doubt correct, for notwithstanding the assertions of the local writers, the real fact must be admitted, that the Súmras are not of Arab descent at all, and that this fictitious genealogy was assumed by them, when the majority of the tribe were converted to Islám; and that, as the name of Sámarra offered a sufficiently specious resemblance, that town was adopted as the probable seat of their origin, though it was not built till after the supposed period of their emigration.*

That the Súmras were not Moslims during at least the early period of their sway, seems to be proved by their names, though this argu­ment is not quite decisive, for down to modern times in Sind, Mu­hammadan converts have been occasionally allowed to retain their Hindú names. Still, reasoning generally, the retention of Hindú names points, primâ facie, to the probability of the retention of the native religion. Now, when we come to examine the Bhúngars and Dúdás among the Súmras, we find that even to the latest period, with one, or at most two, doubtful exceptions, they are all of native Indian origin. The fact of their being called “Hamír,” in Sindian ballads (a probable corruption of “Amír”) scarcely militates against this, as it was, both in ancient and modern times, a distinctive appel­lation of the rulers of Sind, and was only superseded where, as in the case of the Jáms, there was a more familiar title of local origin. The ascription of so honourable an address and so high a lineage, is easily accounted for by the natural tendency to aggrandisement which has actuated all bards and minstrels, from Demodocus and Tyrtæus to the last prizeman of the Cambrian Eisteddfodd. That many of the tribe still continue Hindús, roaming as shepherds through the thals of Jesalmír and the Upper Dhat country to the east of Sind, we know from personal communication. Even if it might be admitted that, in the present day, they had forgotten their Arab origin, and lapsed into Hindúism from their former creed; still, that could not have occurred at the very earliest period of their history, within a century or two of their emigration, and before their high and holy origin could possible have been forgotten.

The Súmras of the desert are one of the subdivisions of the Pramára Rájpúts, and from frequently combining with their brethren the 'Umars, gave name to a large tract of country, which is even still recognized as 'Umra-Súmra, and within which Alor is situated. Renouard surmises that they may be “Som-Ráí,” that is, of the Lunar race, but, being without question of the Pramára stock, they are necessarily Agni-kulas. Their successors and opponents, the Sammas, were of the Lunar race.

It is not improbable that the Lúmrís, or Númarís, of Bulúchistán may be of the same stock, who, when they derive their lineage from Samar, the founder of Samarkand, may have been originally nothing but Súmras. This, however, would not be admissible, if they really have that consanguinity with the Bhátís which they profess, and which would throw them also into the Lunar family.*

It is not only from passages which professedly treat of the Súmras that we know them to be Hindús, but from an incidental notice in foreign historians, such as the authors of the Jahán-kushá and the Jámi'u-t Tawáríkh; where, in writing of the expedition of Jalálu-d dín to Sind, in 621 A.H. (1221 A.D.), they mention that, when he was approaching Debal, the ruler of that country, Hasrar, took to flight, and embarked on a boat, leaving the Sultán to enter the place with­out a contest, and erect mosques on the sites of the Hindú temples which he destroyed. This Hasrar is, in Firishta's account of the same expedition, named Jaisí, which, if it be correctly written, is more probably a titular than a personal designation; for we learn it was the name borne by the son of Dáhir, who ruled in the same province, and was so called from the Sindí word jai, “victory.” It seems, however, not improbable that the name is neither Hasrar, nor Jaisí, nor Jaisar, but Chanesar, the popular hero of some of the Sindian legends respecting the Súmra family. Neither of the three other names is to be found amongst those of the Súmra rulers, and written without the diacritical points, they all vary but little from one another. Admitting this to be the case, we obtain an useful synchronism in the Súmra dynasty, notwithstanding that the local ballad of Dodo and Chanesar makes them contemporaries of 'Aláu-d dín, a name more familiar to native ears than Shamsu-d dín, the actual ruler of Dehlí at that period, and his predecessor by nearly a whole century.