30 Dew-Ponds. 



down. Many others are scattered about — they have 

 become more numerous of late years. Several are sit- 

 uate on the lower plateau, which is also dry enough. 

 Toiling over the endless hills in the summer heats, I 

 have often been driven by necessity of thirst to taste 

 a little of the water contained in them, though well 

 knowing the inevitable result. The water has a dead 

 flavor : it is not stagnant in the sense of impurit}', 

 but dead, even when quite clear. In a few moments 

 after tasting it, the mouth dries, with a harsh un- 

 pleasant feeling as if some impalpable dusty particles 

 had got into the substance of the tongue. This is 

 caused probably by suspended chalk, of which it 

 tastes ; for assuaging thirst, therefore, it is worse than 

 useless in summer : very different is the exquisitely 

 limpid cool liquid which bubbles out in the narrow 

 coombes far below. 



The indirect bearing of the phenomena of these 

 dew-ponds upon the water-suppl}^ of the ancient fort 

 is found in the evidence they supply that under differ- 

 ent conditions the deposit of moisture here might 

 have been very much larger. The ice formed upon 

 the branches of the beech trees in winter proves that 

 water is often present in the atmosphere in large 

 quantities : all it requires is something to precipitate 

 it. Therefore, if these hills were once clothed with 

 forest, as previously suggested, it appears possible 

 that the primitive inhabitants, after all, may have, 

 carried on their agriculture with less difScult}', and 

 have been able to store up water in their camps with 

 greater ease, than would be the ease at present. This 

 may explain the traces of primeval cultivation to be 

 seen here on the barest, bleakest, and most unprom- 



