what they could to level and obliterate the work, the lines can be clearly truced, and it has been shown that the town could be entered by eleven gates. Near the centre of the enclosure was an inner stronghold, and within this again was the palace of the Khans. The stronghold, shuped like a trupezium, was surrounded by thick walls, which were demolished at all ancient date, and now present the appearance of a rampull about ten feet high. Your circular bastions protected the four angles, and two double rectangular bastions guarded each of the four yates, one of which pierced each of the four walls. The walls were further strengthened by eight other pentagoml bastions. The main entrance was on the eastern side.

Within this fortress stood a group of buildings, which is undoubtedly to be identified as the palatial residence of the Khans. The principal editice, which may be distinguished is the Throne-palace, was curiously constructed. A large room in the basement, to which there seems to have been no entrance from without, except perhaps a narrow isste undernonth a staircase, points to the fact that the ground floor was only in substructure for an upp?r storey. This storey conto sisteil of it proclomos or entrance-hull on the south sidle, to which the chief staircase ascended, and a hall of audience. The hall Wirs nearly square, and was divided by rows of columns into three parts, resembling the nave and aisles of il church. The throne stood in a round anos, in the centre of the northern wall. Not far from this building stood at rectangular temple, which in the days of Krum and Omurtay was devoted to the heathen cult of the Bulgarians, but wils converted, after the adoption of Christianity, into a church.

The fortress and the palace, which seem to have been built much about the same time, certainly belong to no later period than the first half of the ninth century. The architecture of the Throne-palace bears the impress of Byzantine influence, and has a certain resemblance to the Trikonchos of Theophilus, as well as to the Maynaura.' It was doubtless constructed by Greek masons. The columns may have been imported from Constantinople; it is recorded that Krim,

I It resembled the Triklinos of the an upper storey and in being entered Maynaura by its throne-apise and the through the proclomos, it's the Tiikon row's of columns in the " navo"; it chos was ontered through the Sigmat, resembled the Trikonchos in being to which external stairs ascended.

when he attacked that city, carried off works of art from the suburlan buildings.

The title of the rulers of Bulgaria was kunas uvege, “sublime khan,” but even while they were still heathen, they did not scruple to have themselves described sometimes in their oflicial monuments as rulers by the will of God.” ? Of the political constitution of the kingdom little can be ascertained. The social fabric of the ruling race was based on the clan system, and the head of each clan was perhaps known as a župan. From early ages the monarchy had been hereditary in the clin of Dulu, but in the middle of the eighth century, Kormisos, who belonged to another family, ascended the throne, and after his death Bulgaria was distracted for some yeill's by struggles for the royal power. We may probably see in these events it revolt of the clans against the hereditary principle and an attempt to make the monarchy elective.

There were two ranks of nobility, the boilads and the bayains,' and amony the boilads there were six or perhaps twelve who had it conspicuonis position at the court. When a Bulgarian ambassador arrived at Constantinople, etiquette required that the foreign minister should make particular inquiry first for “the six

I kávas prvý, priedling the name (truyurt in the inscripcions). isnya hins ben siti tutorilyquated...this Tomaschek) with the Comman Turk viegloc"high, Glorious"; op. Mare quari, Shrid: 119", 1990; Chron. 10.

2. Omulting in the Chatiliar inscrips tion(s. 11.821.822), in Deoü äpxwv, .lbolu, 645 ; and Homil, o da 0, d., ib. 230

1.7.1i, 8091). The lines of the title luy Omurtalinporoves Uspenski's (un. jecolure (ib. 197198) that the Roman government contrred it on Malanmir buccaline Christianity luul spread in Bulgaria in his reign. Marquart's view in (Chron. 11.12) that the title was lucant as a translation of the Turkisha Tingriti bolmy's got,

192). Oklsun, of the family of kurigir, is described as o govtár (190); Okørses ils ó kutavós (whord K scolas to be all ditor for $, ib.); and in alle other inscription (No. 7, p. 19:2) in honour of suime one geveâs 'Ep... ápms, I would supply at the beginning Gourav Jos. As the title Zhupan Wils sol liy South Slavonic preoples for thou head of in tribe, it is in prisonbile conjecture that it designated a tribal

Among the Bulgarians. Sve posed 10 occur in the form courav in the marly inseription of Miroslı in Hungary, which is believed to relato to the Ciepiels lib.).

"]

great boiiads," and then for the other boilads, “ the inner and the outer.' There were thus three grades in this order. We do not know whether the high military offices of tarkan and kaulshan were contined to the boilads. The khan himself had a following or retinue of his own men, which seems to have reseinbled the German comitatus. The kingdom was divided into teu administrative divisions, governed by oflicers whose title we know only under the equivalent of count.

The Bulgarians used the Greek language for their official documents, and like the ancient Greeks recorded their public acts by inscriptions on stones. Mutilated texts of trenties and records of important events have been discovered. They are composed in colloquial and halting Greek, not in the diplomatic style of the chancery of Byzantium, and we may guess that they were written by Bulgarians or Slavs who had acquired a smattering of the Greek tongue. Among these monuments are several stones inscribed by the khans in memory of valued oflicers who died in their service. One of them, for instance, inet his death in the waters of the Dnieper, another was drowned in the Theiss. This use of the Greek language for

i In Constantine, Cer. 681, we find Const. Porph. De min. imp. 169,-, ther six great boilies (tomth cont.), alo-poyoroup, as Marquart correcis but in De culm. imp. 151, Wi learn of for á No90,30701p), the Turkish madur, the cipoluite of "the twelve great from which the Russian buyutyr boiluds" by the Servians (ninth cent.). ( = hiero) is derived; and for prou (1110), It seems plain the inner and outer in Mansi, xvi. 158 ; see Uspaski, ib. simply mean it higher and lower grade. 204). κολοβρος (κοιλοι βρος) hells to For we find exactly the same termis, have been it title of ritlik, 110 it post great, innrr, und unter applied to the or ollico ; Tomaschek quatus it with ilireu Bulgarias. There were the Turkislı qulunghw, it muille, au Mare Great Bulgarians on the Danube, the guarl (Chron.nl) compiles poi au la,spas Imer Bullgarians on the Sea of Azov, in Theoph. Simoratiit, i. 8. 2, who and the Duter Bulgarians on the explains it as máqos or de peers. Volga Ser blow, p. 110.847.

θρεπτοί άνθρωποι, fruit in the ? Therapkávos (inscriptions) was in. inscriptions. See Usikuski's long dis. doubseilly i military commander. We

cussioni, ib. 201 sqq. meet this Turkislı title in Menander's • dun. Birl, subu. 800 (p. 87), “ intrat account of oni ombassy of the Turkish drcom comitatus." Silistria was the Khan Dizabul to Justin II. (fr. 20). chief place of one of the counties: The ambassador's name wils Tarma, inserijilioni

, Simeon, 1:1. kpl. iii. 150, αξίωμα δε αύτω Τ'αρχάν. So also Coil. κόμης Δρίστρου. Cp, also Theoploy'. Τh. 113, καλουντερκάνος (eg. Καλού luctus, llist. mart., 1.li., 120, 201, 213. Tepkavos), in Const. Cor. 681, ó See alboba, 212. Bou Aíus tapauvos. See Uspenski, op. • Some mysterious opigraphic frag cil. 199-200 ; Marylart, (hron. 43-44. ments have also been discovered, For the kalxavos sue inscriptions, written, partly ani least, in Greek leuters, bulu, 220, 23:1, 1110 Simon (Cont. but not in the Greek tongue. They licory. ed. Muralt, -19, ed. Bom 893), aro very slight and little can be made άμα Λακάνω. Othern lignitics were of them. See allubu, c. viii. Bayarovp or 309 otop (inscriptionly; also "Ibuba, 190.1944.

their records in the most #triking wign of the influence which was exercised on the Bulgarians by tho civilizntion of Cone stiintinople. We can truc thin influence also in their buildings, and we know that they enlisted in their service Greek engineers, and learned the use of those military engines which the Greeks and Romans haud invented for besieging towns. Notwith. standing the constant warfare in which they were engaged against the Empire, they looked to Constantinople much :19 the ancient Germans looked to Rome. Tervel had been created a Caesar by the gratitude of Justinian II., and two of his successors found an honourable refuge in the Imperial city when they were driven by rivals from their own kingdom. Tseriy tled to the cont of Leo IV. (A.1). 777), accepted baptism and the title of latriciam, and was honoured by the hand of an Imperial princess.' It might be expuected that the Bulgarians would have found it convenient to a lopt the Romiin system of marking chronology by indictions or even to use the Roman cra of the Creation of the worlıl, and we actually find them employing both these methods of indicating time in their ollicial records. But they had also a chronological system of their own. They reckoned time by cycles of sixty lunar years, starting from the yeur A.D. 659, memorable in their history us that in which they had crossed the Danube and made their first permanent settlement in Moesia. For historical purposes, this system involved the same disavvantage as that of Indictions, thongh to a much sinaller degree; for instance, when an event Was vated by the year shegor ulem or 48, it was necessary also to know to what cycle the year referred. But for practical urposes there was no inconvenience, and even in historical records little ambiguity would have been caused until the Bulgarian anuals had been extended by the passage of time into a larger series. It is possible that the Bulgarian lunar years corresponded to the years of the Hijra, and if so, this would be it remarkable indication of Mohammadan influence, which there are other reasons for suspecting. We know that in the ninth century there must have been some Bulgarians who were acquainted with Arabic literature..

3

But the Bulgarians had other neighbours and foes busidlen the Romans, and political interests in other directious thun in that of Constantinoplo. It is recorded thut the anino prince who crossed the Danube und inaugurated a new period in Bulgarian history, also drove the Avar's westward,' and the record expresses the important fact that in the seventh century the Bulgarians succeeded to the overlordship which the Avar khans had exercised over Dacia in the reigns of Maurice and Heraclius. This influence extended to the Theiss or beyond. Eastward, their lordship was bounded by the Empire of the Khazars, but it is impossible to define the precise limit of its c'xtent. There can be no doubt that in the seventh and eighth centuries Bulgaria included the countries known in later times as Walachia and Bessarabia, and the authority of the khuns may have been recognised even beyond the Dniester. At all events it appears to be certain that in this puriod Bulgarian tribes were in occupation of the coastlands from that rivor wellnigh to the Don, and this Bulgarian continuity was not cleft in twain till the ninth century. The more easterly portion of the people were known as the Inner Bulgarians, and they were probably considered to belong to the Empire of the Kluzar's. But we cannot decide whether it Was it the Dniester or rather at the Dnieper that the authority of the Khazars ended and the claims of the Circat Bulgarians of Moesia began.

South of the Danube, the kingdom extended to the Timok, which marked the Servian frontier. The Bulgarians lived on

3 terms of unbroken friendship with the Servians, and this may perhaps be explained by the fact that between their territories the Empire still possessed an important stronghold in the city of Sarilica.

For the greater security of their country the Bulgarians reinforced and supplemented the natural defences of mountain

(Mosis of Chorene), Geography to limit the Bulgarians on their rastorn (seventh cent.), cited in Westberg, Brie frontier, and there is no probability triige, ii. 312; Marquart, Chron. 88. that the Khazars over exerted utlıor.

? Sor. Incertus, 3.15. Βουλγαρίαν ity further than the Dniestor, il as εκείθεν του "Ιστρου ποταμού (= Pseudo- far. Simcon, 615). There is no reason to : one point on the frontier (Com. suppose that when Isperikh settled stantine, T)e alm. imp. 155) seems to in the Dobrudzhin, he abandoned Bess. have been Rasa (Novi Bazar, Jirerek, arubja. Till the ninth century there Geschichte, 150). was no power but that of the Khazars

ጊ