Vol. 67.] EASTERN DE8ERT OF EGYPT. 253 



The foregoing analyses show that only about a third of the rock 

 consists of sand, the remaining two-thirds consisting of soluble salts. 

 A fair amount of gypsum and magnesium salts is indicated along 

 with the rock-salt in the deposits, as would have been expected in 

 the circumstances. 



Both the Gulf of Jemsa and Zeit Bay are continuations of low 

 tracts of sandy and gravelly desert which lie to the north of them. 

 Where these tracts meet the sea there salt marshes exist. In 

 examining the country at the head of the Gulf of Jemsa, I found 

 that the marsh gradually gave place to firm ground as one pro- 

 ceeded landwards. But another change took place: the cementing- 

 material was now gypsum and not salt. The desert was curiously 

 marked by low concentric ridges only an inch or two high, which 

 ran somewhat parallel to the shore-line, as well as to the borders 

 of higher tracts of ground ; they suggested lines left by the 

 gradual desiccation of a shallow pool of water. The ridges were 

 covered with aggregates of gypsum crystals, some of which 

 reached 5 or G centimetres in length. When the ground was dug 

 into, it was found to be very moist underneath. A little farther 

 north the sandy desert presented a curious appearance, being 

 covered with a network of projecting veins of gypsum. These stood 

 out sometimes 6 inches or more from the surface, owing to the fact 

 that they seem to resist the wind — the sand in between the veins 

 having been removed by that agency. Similar veins are sometimes 

 very abundant in the recent limestone of the Gaysum Islands. 



Again, on South Gaysum Island, the surface in the centre of 

 the southern part of the island was found to consist entirely of 

 gypsum crystals. When the deposit was dug into, the gypsum was 

 found to be present in stout fibrous crystals closely packed together, 

 which penetrated from 6 to 12 inches or more below the surface. 

 When traced in any direction, the gypsum deposit was seen to 

 merge insensibly into raised coral-beach or shelly and foramini- 

 feral sands. The deposit itself contained practically no organic 

 remains ; it is evidently in process of formation, at the expense of 

 the more normal deposits. It appears, then, that both here and at 

 the head of the Gulf of Jemsa and of Zeit Bay the sea-water is 

 being drawn up by capillarity into the huge deposits of fine sand 

 which there exist. The gypsum seems to be deposited there by 

 this action, the salt probably being retained in solution. Where 

 the sands are siliceous, the result is the formation of gypsum crystals 

 among the sand at the surface. Where the sands are calcareous, 

 the change seems to be accompanied by a metasomatic replacement 

 in the calcareous matter, resulting in the destruction of organic 

 remains. That the shells of the raised coral-reefs do tend to become 

 converted into gypsum was proved by an analysis of a shell taken 

 from the raised reef of Jubal Seria. This yielded the following- 

 result : — .,-, , 



Per cent. 



CaCOo about GO 



CaS0 4 ' 20 



Other salts, chiefly sodium chloride 1 Qf . 



and salts of magnesium J'" 



