At the spur of the cliff immediately to the east the terrace is laid on to limestones of the Birket el Qurûn series at a height of 32 metres above the canal. Enormous blocks of Carolia-limestone, some exceeding 3 × 2 × 1·5 metres in measurement, are included in this deposit; such blocks cannot have been transported far, and were doubtless derived from formerly-existing higher beds in the immediate neighbourhood. The matrix of the deposit is usually crushed limestone with sand and gravel. Occasional fragments of silicified wood (and further west large logs), evidently derived from the Fluvio-marine beds, also occur.

In favoured localities the relation of the gravels to the Eocene is still better seen; the lower beds of the terrace here are sometimes formed of sands and sandy beds dipping 10° eastward.

At the little promontory 9 kilometres east of Sersena the same deposits reach an altitude of 157 metres above the canal below.

Further north another cake of gravel caps the summit, attaining here 60-70 metres above the canal-level. Probably these deposits were originally more or less connected and continuous, but since their deposition denudation must have removed the greater part, as they are now only found here and there capping the highest points of the escarpment along the east side of the Fayûm. Such isolated gravel-capped hills occur notably 12 kilometres east of Roda, 16½ kilometres east and 17½ kilometres north-east of Tamia.

Along the north side of the Fayûm the same deposits are found, in some cases covering large areas.

Twenty-five kilometres N.N.E. of Tamia the Eocene beds, here an alternating series of clays and limestones, are capped by a deposit consisting of coarse rolled gravel, with blocks of silicified wood enclosed in a sandy gypseous base, some 10 metres thick. A larger and similar deposit caps the next high ground four kilometres to the west, and about 9 kilometres N.N.E. of Garat el Faras; in this case it forms a round-topped gravelly hill-range, attaining a height of about 165 metres above the canal to the south-east. The loose gravel at many points passes into hard conglomerate, notably in the hills 9 kilometres east and 4 kilometres north-east of Garat el Gindi. At the former spot the conglomerate is composed of blocks of limestone, with round pebbles of flint and quartz, sandstone and quartzite, and fragments of silicified wood, cemented by sand and calcareous material. Blocks of silicified wood also occur strewn on the surface of these gravel deposits.

In the hills north-east of Garat el Gindi the gravel deposits do not occupy the summit of the escarpment but occur laid on to a platform of beds belonging to the Qasr el Sagha series. Behind, another escarpment, that of the Fluvio-marine series, rises to the plateau summit.

[Fig. 9] will show the general relation of the different formations in this part of the district.

Near Elwat Hialla the deposits contain numerous blocks of basalt in addition to the usual constituents. The basalt is derived from the sheets interbedded at the base of the Oligocene a little to the north. As these gravels are here close to that formation, blocks of sandstone, basalt, and silicified wood now form a large proportion of the constituents.

Along the north side of the Fayûm depression, to the west of Elwat Hialla, the gravel terraces are almost absent, having been removed nearly completely by denudation. That the terraces once existed throughout this region is however shown by the small patches met with to the north-east of Widan el Faras, the eastern extremity of Jebel el Qatrani, and at several points high up on the escarpments as far west as the western end of the lake. Beyond the latter point these terrace gravels have not been noticed; the slopes of the depression become more and more obscured by loose superficial flints washed down from the plateau, and the existence of underlying terrace gravels could only be shown by detailed mapping.