74 
Mr. Aikin on the V allies and Watercourses 
sea. Data for determining this point with considerable accuracy are 
afforded by the levels of the Ellesmere canal, the main line of w T hich 
extends between Shrewsbury and Chester, and, though not entirely 
executed, has been completely surveyed. 
The summit line which separates the running waters lies nearly 
east and west. It approaches in one or two instances within a 
quarter of a mile of the valley of the Dee, and extends with a con- 
siderably indented outline from the village of Selatyn a Httle north- 
west of Oswestry, to the town of Whitchurch. The high level of 
the canal passes through this latter town, and after a descent of 128 
feet joins the Chester canal, which in its progress to tideway in the^ 
Dee at the city of Chester, descends 170 feet 10 inches; making 
the elevation of the summit line or water-shed 298 feet 10 inches 
above the Dee at Chester. By another line between the two extreme 
points, surveyed but not executed, the difference of level appears to 
be only 292 feet ; the average of the two gives 295 \ feet, which I 
think may safely be assumed as nearly approaching to the truth. 
If from the same summit level we compute the lockage of the 
canal to its intended junction with the Severn at Bagley brook, near 
Shrewsbury, we shall find it to amount to a fall of 140 feet. Hence 
the height of the Severn at Bagley brook is 1 55 1 feet above the 
tideway in the Dee at Chester. 
In order to compare this result with that furnished by levels taken 
between Shrewsbury and the Bristol channel, I must first state that 
the height of the basin of the Shrewsbury canal at this latter town 
above the Severn at Bagley brook, amounts (according to information 
furnished to me by Mr. Telford) to 24 feet 4 inches. From this 
basin to the termination of the canal at Coalport on the Severn, the 
descent on the whole is 59 feet ; the canal at Coalport is about 16 
feet above the Severn ; and the fall of the Severn itself from Coal- 
