صور الصفحة
النشر الإلكتروني

hand from the entrance ; the interior measurements of the temple are about 50 ft. by 20 ft, eight feet high at the altar, and from 12 to 14 ft, high under the principal entrance. The doorways seem to have had doors or curtains and bars, once fixed to them. The workmanship is rude : from the position and general design of the temple its lateral chapels and central nave, it appears to have been once a Christian church.

About 400 yards lower down the same side of the Wadi, is another rock cut temple, much smaller in size, which contains the fragment of a Greek inscription cut into the stone along the cornice in its front.*

[ocr errors]




The sign of a cross and a piece of a shell, such as are used in modern Greek churches, were found in the temple.

The ruins of houses are scattered around, constructed of fragments of mica schist, steatite, quartz in part coloured green by the colouring matter of the emerald, and containing crystals of emerald.

The mines appear to have been sold out to companies of adventurers, who built magazines and houses around the entrances of the excavations, of which several hundred still remain. In many instances, wherever the position of the quartz veins, and schists permitted, the whole face of a mountain has been quarried down and exposed to research.

The Zubára mines are galleries run into the mica schist; layers, containing nodules of emeraldic quartz, are often disposed so as to have a layer of quartz for their roof. At Sakeyt, on the contrary, the emeralds had been searched for in veins and layers of quartz, which is tinged purple, yellow, and red. The more transparent and white sorts, found in the debris thrown out by the ancient miners, exhibit the light bluish green of the Egyptian emerald. The schists and steatites are variously coloured, and mica occurs in golden scales. White, greenish,

* The inscription contains apparently a dedication to the Egyptian goddess Isis and to the Greek god Apollo. The cross and shell indicate it having been subsequently used as a Christian chapel.-T. J. N.

black, light pink, and dark red coloured crystals occur as well as combinations of mica with quartz.

The schistose formations rest on the granitic: which last rocks have been made use of in the cynocephali, seen in the vicinity and lower down in the valley. These granitic rocks rise in elevations from 600 to 700 ft. above the surface of the valleys. They are overlaid toward the summit, by a layer of micaceous quartz containing a few emerald crystals : immediately over which is a kind of (argillaceous ?) schist, whose super-crust is strongly ferruginous. A large mass of calcined iron stone indicates the ancient working of this bed for iron ore; which, probably, furnished the material for the tools of the workmen.*

The mining district is confined to the hills which are enclosed in the basin of Wadi Sakeyt; and which are much lower than the high mural ridges forming the sides of the basin, and are of an irregular conical shape. They present parallel bands and waving lines of reddish quartzose matter and dark brown schists. The section of the hill over the temple presents,-in its lower portions—friable micaceous schists coloured with partial scales of iron ore; and covered with a tabular and highly sonorous bed of whitish brown quartz in parallel waved bands indicating volcanic or plutonic disturbance. There are no fissures in the curves.

Farther up the Wadi, rise hills of laminar talc with brown argillaceous layers : then mica schists of various colours, with crystallized garnet and quartz in needles. The highest summits appear to be of granite. The emerald is met with in the quartz debris of former mines, and occurs in crystals of a light bluish green attached to the quartz in mica schist.

Observations on the Language of the Goands, and the identity of many

of its terms with words now in use in the Telugu, Tamil and Canarese. By WALTER ELLIOTT, Esq., C. S., Madras.

A paper by Dr. Manger, on the language of the Goands, in the March No. of the Journal, offers some interesting grounds for ethnolo

* From the Cynocephali the Greek inscription and the Christian emblems found here it would appear that the mines have been worked from the era of the Pharoahs down to early Christian times.- T.J. N.

gical speculation. So long ago as 1812, a notice in the Oriental Christian Spectator, * alluded to the discovery of Tamil and Canarese words in the Goand language, by Mr. Loesch, a German Missionary, who soon after fell a victim to the climate. The same subject was alluded to by Mr. D. F. McLoed, in a letter to the Secretary of the Asiatic Society, in 1814 ;+ but until the present instance no vocabulary, it is believed, (with the exception of a short list in No. CXLV. of the Journal,) has been published, from which an opinion could be formed of the extent to which the admixture of the dialects of southern India, prevails in the present speech of these wild tribes.

A very superficial examination of Dr. Manger's list is sufficient to show that more than one half of the terms set down by him are identical with, or approximate very closely, to words now in use in the Telugu, Tamil and Canarese tongues. In a corresponding list which accompanies this paper I have noted such words as occurred to mind, and a more careful examination would doubtless elicit more.

The investigation of the different races that constitute the Hindu population of India has hitherto received less attention than the subject deserves. Beyond the fact that all the spoken dialects of India proper, are referrible to two great divisions, which the natives themselves recognize under the titles of Pancha-Goura, and Pancha-Dravira, -but little is known of the general relations and affinities of the people using them. According to Colebrooke, the central seat of the former was Canouj, I the capital of the Canya cubjas, from which point its cognate dialects spread both east and west, and then stretch far to the south. and southwest, over Maharashtra, extending down the Malabar coast even to the vicinity of Mangalore. $ The southern dialects have generally been considered to commence from the neighbourhood of Beder, near which the limits of the Mahratta, Canarese and Telugu, meet. Dr. Manger's Vocabulary at once carries us to the Nurbudda, and it is not improbable that similar dialects may be discovered in the mountainous region on its northern band, and even in Bundelcund.

The first question that arises is, whether these two classes of languages indicate the contemporaneous existence of two great aboriginal * Vol. iji. p. 240.

+ Friend of India, 1844, p. 203. # Asiatic Researches, vol. VII. p. 220. Goand.

♡ A dialect of Concani is spoken in all the tract north of Sadasheoghur, and Mahratta is the language of the mountains immediately above it.

races inhabiting the northern and southern portions of India, or whether the people using the southern language at one time occupied the whole extent and were gradually driven southwards by the pressure of a new race of invaders from the north. The isolated existence of a cognate dialect of the south, among a wild tribe in habiting unapproachable forests and fastnesses considerably to the north of the present range of these languages, is in favor of the latter supposition. But a single fact affords too narrow a basis on which to build so important a hypothesis.

The opinion of Mr. Colebrooke regarding the derivation of the Hindi or northern dialects from Sanscrit, has not found favor with Oriental philologists, and seems no longer tenable. But its influence on all the languages now in use, whether in the north or in the south bears incontestible evidence of the sway of a people vastly superior in power and civilization to the aboriginal races. All the written characters now in use, as has been proved by James Prinsep, have been derived from that source, and the very number of the letters, their classification and arrangement, are the same in all the languages of the north and of the south, except the Tamil, the most remote of the southern dialects. It is not only singular in wanting the regular series of aspirated consonants, but the number of simple consonants and vowels likewise falls short of those of all the others. It has besides, letters to express sounds peculiar to itself, and others which receive new powers by reduplication.* This fact would seem to indicate the gradual retrogression of the great southern race to the extreme verge of the peninsula, where it preserves the most distinctive marks of its original character. But whether this was owing to the growth of the power and the extended conquests of the Hindí tribes or to the silent progress of the Bráhmanical faith and literature, or, as is most likely, to both of these combined, -remains to be ascertained.

In looking at the comparative list of words it will be observed that a considerable number of Goand wordsť are derived from the northern stock, as was likely to happen from the influence of the surrounding dialects. But the same effect is observable even in Telugu and Cana

* Thus double rr and double dd become ti, double bb becomes pp, and double ss serves for ch.

+ As Boy, Girl, Ilorse, Ass, Goat, Twenty, Fifty, &c.

rese, where Sanscrit terms, * have in general instances superseded the original words. The influence of the Urdú on Goand is also perceptible in the ten Commandments, and Sand Sumjee's song, where we find such terms as admee man, hazar 1000, kúm business, labari-gohai false witness, khabar news, kuan well, tisra din third day, hath market, guttri bundle, ghossa laga become angry, pucha question, thera kya established, and the particles jub when, and keh that.

The similarity of grammatical construction between the Goand and southern languages is apparent in many respects, but in this part the Vocabulary seems rather defective.

The plural is formed by the addition of k in Goand, and by kal and gal in Tamil and Canarese. The objective cases which terminate in na and un in the former, are formed by in the genitive, and nu the accusative, in Tamil and in Canarese. The structure of the verb, as far as it can be learned from the examples given, also presents many analogies. The present and some of the other tenses in Goand are derived from the present and conjunctive participles without personal terminations, in this respect corresponding with the formation of the Malayalem verb, (Peet's Malayalem Grammar, p. 60.) which is also without inflections. The past and future tenses show some traces of resemblance to the Canarese and Tamil, both in their formation and in their personal terminations. The large employment of auxiliary verbs in the southern dialects is not perceptible in the Goand specimens of construction, unless it be in the formation of the passive by the use of howe.

These hasty remarks have been thrown together in the hope of showing how wide a field is open for further investigation, not with the idea of communicating valuable results. The Vocabulary and specimens are too scanty and imperfect to make the institution of more careful comparisons worth while at present. The Goand words too seem to have suffered considerable mutilations and changes at the hands of the printer, which renders it hazardous to venture on conclusions drawn from less obvious resemblances. It is to be hoped that more attention will be given to so interesting a subject, and the publication of a well selected Vocabularyt of terms for general adoption as promised by the Editors, will greatly facilitate the labours of future inquirers in this field. * As in day, man, twenty, &c.

+ See foot note, p. 286.

« السابقةمتابعة »