So it appears to have been with Bagdad in the days of its glory. The walls of the palaces seem to have been of sun-dried bricks, and the whole constructive parts of wood, covered, it may have been, with gold and silver plates, and painted with all the hues of the rainbow, but nevertheless perishable, and certain to go to decay as soon as the maintaining hand of a resident sovereign was withdrawn.

From what we know from history of the age of Haroun el Rasheed, it is probable that no Moorish court ever reached a higher pitch of enlightenment and magnificence than that of Bagdad during his reign (A.D. 786-809). It was also so far removed from the direct influence of the Byzantine style, that it is probable we should find in his buildings the germ of much which now comes abruptly before us without our being able to trace it back to its origin.

In the whole architectural history of the world there is scarcely so complete a break as this, and scarcely one so much to be lamented, considering how great and how polished the people were whose art is thus lost to us. Let us hope, however, that it is not entirely lost; but that some fragments may yet be recovered by the first who earnestly searches for them. But in the mean while we must pass on to the age of Ghengis Khan, in the middle of the 13th century, before we can find one certain example of Mahometan architecture in this important district.

With such imperfect materials it will of course be impossible to attempt anything like a continuous history of Persian architecture, but we must be content with illustrating two or three of the principal buildings; by which means we shall be able to define the style with sufficient distinctness, and explain all its leading characteristics, though without being able either to trace the origin of its peculiarities, or to point out in how far they influenced the style of other countries.

One of the earliest buildings of which anything like correct illustrations have been published is the Imaret or Hospital of Oulou Jami, at Erzeroum—an arcade of two stories, surrounding on three sides a court-yard 90 ft. by 45. It is broken in the centre by what in a Christian church would be called a transept. The woodcut here given (No. 324) shows the general appearance of the arcade, and also the upper part of two minarets which flank the external porch. This porch is ornamented in the richest manner of the style. Opposite to the entrance a long gallery leads to the tomb of the founder, a circular building of very considerable elegance, the roof of which is a hemispherical vault internally, but a straight-sided Armenian conical roof on the outside. These dispositions make the plan of the building so similar to that of a Christian church, that most travellers have considered it as one-mistaken the court for the nave, and the tomb, with the gallery leading to it, for the apse and choir. There can, however, be no doubt but that it was originally built by a Mahometan, for the purpose of a hospital, or place of rest for pilgrims, during the sway of the Seljukian princes in the 12th and 13th centuries; and that its similarity to a Christian church in plan is accidental, though its details

very much resemble those of the churches of Ani and other places in Armenia. This, however, only shows that the inhabitants of the same country did not practise two styles, but only arranged the same forms in different manners to suit their various purposes.

324.

There is another mosque of about the same age as this at Ani, which would show even more clearly this close analogy; but it has never been drawn with sufficient correctness to admit of its being used for the purpose of demonstrating the fact which has been pointed out. But, indeed, throughout Armenia, mosques and Christian churches alternate with one another, borrowing details from one another, and making up one of the most curious mixed chapters in the history of the art; but one that remains yet to be written by some one who may visit the spot with sufficient knowledge and enthusiasm to accomplish it.

MOSQUE AT TABREEZ.

The next building that may be chosen for illustration is the ruined mosque at Tabrecz, which, when perfect, must have been one of the

most beautiful in the country. Its history is not exactly known; but it certainly belongs to the Mogul dynasty, which on the death of Maugu Khan, the son of Ghengis Khan, was founded in Persia by Hulaku, the brother of the first named. He and his sons generally retained the faith of their forefathers till Ghazan Khan, who succeeded in A.D. 1294. He zealously embraced the Mahometan faith, and it was apparently to signalise the conversion that he began the mosque; but whether it was finished by him or his

successors is not quite apparent. As will be seen by the plan, it is not large, being only about 150 ft. by 120, exclusive of the tomb in the rear, which as a Tartar it was impossible he should dispense with.

In plan it differs also very considerably from those previously illustrated, being in fact a copy of a Byzantine church carried out with the details of the 13th century. This fact confirms the belief that the Persians before this age were not a mosquebuilding people. In this mosque the mode of decoration is what principally deserves

attention, the whole building, both ex- 325. Mosque at Tabreez. Scale 100 ft. ternally and internally, being covered

to 1 in.

with a perfect mosaic of glazed bricks of the most brilliant colours, and wrought into the most intricate patterns, and with all the elegance for which the Persians were in all ages remarkable.

326.

Europe possesses no specimen of any style of ornamentation with which this can be compared. The painted plaster of the Alhambra is infinitely inferior, and even the mosaic painted-glass of our cathedrals is a very partial and incomplete ornament compared with the brilliancy of a design pervading the whole building, and entirely carried out in this style. From the time, however, of the oldest Assyrian palaces, to the present day, colour has been in this country a more essential element of architectural magnificence than form; and here at least we may judge of what the halls of Nineveh and Persepolis once were, when adorned with colours in the same manner as this now ruined mosque of the Tartars.

Though of course it is impossible to represent it properly in a woodcut, the view (woodcut No. 326) of its principal portal will give some idea of the form of the mosque, and introduce the reader to a new mode of giving expression to portals, which after the date of this building is nearly universal in the East. As will be seen, the entrance-door is small, but covered by a semi-dome of considerable magnitude, giving it all the grandeur of a portal as large as the main aisle of the building. The Gothic architects attempted something of this sort, by making the outer openings of their doors considerably larger than the inner; in other words, by splaying widely the jambs of their portals. By this means, in some of the French cathedrals, the appearance of a very large portal is obtained with only the requisite and convenient size of opening; but in this they were far surpassed by the architects of the East, whose lofty and deeply recessed portals, built on the same plan as the example here shown, are unrivalled for grandeur and appropriateness.

The mosque seems to have been deserted long before its destruction by an earthquake in the beginning of the present century, owing to its having belonged to the Turkish sect of the Sonni, while the Persians have during the last five centuries been devoted Shiites or followers of the sect of Ali and his martyred sons.

327. Tomb at Sultanieh. Scale 100 ft. to 1 in.

TOMB AT SULTANIEH. (A.D. 1303-1316.)

Mahomed Khodabendah, the successor of Ghazan Khan, the builder of the mosque at Tabreez last described, founded the city of Sultanieh, and, like a true Tartar, his first care was to build himself a tomb which should

Both the plan and view are taken from Baron Texier's Arménie et la Perse,' which gives also several coloured plates of the mosaic decorations, from which their beauty of detail may be judged, though not the effect of the whole.

Texier, from whose work the illustrations are taken, ascribes the building to

another Khodabendah of the Sufi dynasty, A.D. 1577-85. A little knowledge, however, of the style should have shown him that the monument was 200 or 300 years older than that king; and besides, the Sufis, not being Tartars, would not build tombs anywhere, much less in Sultanich, where they never resided.