on the lower and eastern part, nor by what means 

 the ancient bed of the Nile has been filled up, and also 

 the lake of Mareotis ; while those of Brulos, and Men- 

 zala are more slowly progressing to the same, or a si- 

 milar state. Neither need we wonder why the anci- 

 ent Canopic branch, from the ancient Milesian wall, 

 or the city of Deirut to the sea should be filled up, and 

 a new channel formed ; since, besides what I have 

 stated, Dr. Pocolce says, that " all the country here, 

 (near the Canopic branch,) is a sandy desert ; it might 

 be otherwise, when this branch of the Nile annually 

 overflowed, but there being a ridge of low sandy hills 

 running from north to south, near the Nile, it is possi- 

 ble that the fruitful soil may have been covered with 

 sand blown from these hills. The sand changes so 

 often, that it would be difficult to find the way, if they 

 had not built eleven pillars across the plain, which I 

 conjectured might be about half a mile apart, in order 

 to direct the way, which otherwise it would be difficult 

 to find at such times, as the wind raises great clouds of 

 sandy as it often does in Egypt."* 



It is, almost exclusively, to this cause, that we are 

 to look for the extension of the land between Alexan- 

 dria and the Canopic branch ; for Dr. Shaw observes, 

 that there are few or no tokens of the Nile's inunda- 

 tion, to be met with from Alexandria to Rosetta, the 

 whole tract appearing to have been, originally, either 



* Pocoke's Travels, vol. I, page i3. 



