The throne representations are of two kinds. On the jambs of the great doors leading out upon the porch, we see in the top compartment the monarch seated under the canopy, accompanied by five attendants, while below him are his guards, arranged in five rows of ten each, some armed with spears and shields, others with spears, short swords, bows and quivers. Thus the two portals together exhibit the figures of 200 Persian guardsmen in attendance on the person of the king. The doors at the back of the building present us with a still more curious sculpture. On these the throne appears elevated on a lofty platform, the stages of which, three in number, are upheld by figures in different costumes, representing apparently the natives of all the different provinces of the Empire. It is a reasonable conjecture that this great hall was intended especially for a throne-room, and that in the representations on these doorways we have figured a structure which actually existed under its roof (probably at t in the plan)--a platform reached by steps, whereon, in the great ceremonies of state, the royal throne was placed, in order that the monarch might be distinctly seen at one and the same time by the whole Court. [Illustration: PLATE XLVIII.] The question of the lighting of this huge apartment presents some difficulties. On three sides, as already observed, the hall had (so far as appears) no windows--the places where windows might have been expected to occur being occupied by niches. The apparent openings are consequently reduced to some fifteen, viz., the eight doorways, and seven windows, which looked out upon the portico, and were therefore overhung and had a north aspect. It is clear that sufficient light could not have entered the apartment from these--the only visible--apertures. We must therefore suppose either that the walls above the niches were pierced with windows, which is quite possible, or else that light was in some way or other admitted from the roof. The latter is the supposition of those most competent to decide. M. Flandin conjectures that the roof had four apertures, placed at the points where the lines drawn from the northern to the southern, and those drawn from the eastern to the western, doors would intersect one another. He seems to suppose that these openings were wholly unprotected, in which case they would have admitted, in a very inconvenient way, both the sun and the rain. May we not presume that, if such openings existed, they were guarded by louvres such as have been regarded as probably lighting the Assyrian halls, and of which a representation has already been given?