218 On the Arabian Frontier of Egypt. 



We cannot subject this hypothesis to a more decisive test, 

 than that of weighing its consistency with physical and with 

 historical fact, by considering, with the help of the diagrams 

 which embody the physical facts, how far the details of the 

 works, attributed by Pliny to Ptolemy, agree with the main 

 design Ptolemy would have in view in executing those works, 

 namely, that of keeping up as entire a line of water commu- 

 nication with the southern sea, as could be done with ease 

 and safety, under the altered state of things. 



The superficial communication of the main with the upper 

 gulf-basin being now, as we suppose, cut off (vide section 1), 

 the contents of the latter must have sunk a little by evapora- 

 tion. Were it not for the annual supply from the Nile, in- 

 troduced by Darius's canal, they would have dried up entirely; 

 and so long as this was taken from Bubastis, where the in- 

 undation, at its greatest height, was yet six feet lower than 

 the sea, that level is the utmost the water of the gulf-basin 

 could attain.* Under such circumstances, the greatest part 

 of the intervening shoal must have been laid bare (vide sec- 

 tion 1, Plate v., ante, p. 19), and to open a communication 

 by water as far as Arsinoe, it would be necessary to excavate 

 about xi. M.P. along it. 



Although it would be difficult to suggest how far below 

 their maximum height the waters of the gulf-basin would 

 sink annually by evaporation, we may be sure that the loss 

 of so large a siu-face, in such a climate, would be too great 

 to admit of a passage to Arsinoe navigable all the year, unless 

 a very deep channel were cut in the hard nucleus of the shoal. 

 It was easier to dig a trench of little depth in its softer sur- 

 face only ; and this done, to supply the whole line of naviga- 

 tion from a point in the Nile where the water is about 5 feet 

 higher than at Bubastis (vide section 2), by conducting it 

 from the Pelusiac branch near " Scenae,'' along the Etham 

 river's former course, where the half-obliterated bed of the 

 river was still sufficiently max'ked to render excavation in 

 that quarter unnecessary. t For high banks to lead the wa- 



* Vide the diagram B, plate VI., upper line A. PP. 

 t Vide ground plan, Plate VI., PP— PP. 



