NOTES ON MAJOR COATES' DISCOVERY. 87 



little leakage through it. There would be fissures in it, no 

 doubt, but the ashes or the Oxford Clay from Compton, mingled 

 with the first water sent down, and the chalk disintegrated by the 

 frost, might close them sufficiently.* Where the soil was more 

 open or, indeed, throughout the whole course earthenware 

 pipes might have been used. When water was conveyed through 

 pipes, it was customary to make reservoirs along the aqueduct 

 at distances of three or four miles, so that a section of the 

 aqueduct could be repaired without taking up the whole. The 

 reservoirs would also, by breaking up the course into lengths, 

 prevent any undue pressure being put upon the pipes. If pipes 

 were used, a much smaller reservoir at the spring head would 

 have been required. 



Major Coates, however, believes that the conveyance of the 

 water was by means of an open water course and not by pipes, 

 and that the aqueduct w?s used only when the water was in 

 flood. As to the low gradient, he informs me that, according to 

 Molesworth's tables, if the maximum mean depth of the channel 

 was two feet and the cross section twelve square feet, the 

 velocity of water along it would be two and a quarter feet a 

 second with a fall of four feet in a mile, and six feet a second with 

 a fall of one two-hundredths, as recommended by Vitruvius ; 



* Whether an aqueduct cut in the Chalk along the side of a hill would or 

 would not convey water need be no matter for speculation, because we have such 

 an aqueduct close to Dorchester which has been running for centuries, and we 

 can observe it in operation. It may be seen near the foot of the hill upon which 

 the gaol and Poundbury Camp stand ; along the foot of this range of hill it is 

 cut for some distance in the Chalk. Upon the bank of the channel below the 

 gaol is a footpath, and on the other side of the footpath a ditch, six or eight feet 

 from the channel, and below the level of the water which runs in it ; the leakage 

 both here and a little above, where the meadow is below the level of the water, 

 can be observed that it is not excessive. What has made this channel water- 

 tight V It may be answered : the alluvial soil brought down by the stream has 

 been carried into the fissures between the lumps of chalk and there compressed 

 by the weight of the superincumbent water. If this has rendered the channel 

 watertight, may not the Komans, as their practice was, have sent down the 

 aqueduct clay mingled with the water, and so rendered the chalk channel 

 sufficiently watertight to convey the water without excessive waste to Dor- 

 chester ? There is abundance of Oxford Clay at Comptou ; indeed, the village is 

 built upon it. 



