CATARACTS, AND INUNDATIONS. 89 



counterbalance the quantity which is drawn off by the crooked 

 canal ; and the external well will assume the character of a common 

 fountain under these circumstances. 



I have now explained the principles, on which the common theory 

 of reciprocating springs is founded ; and the necessary consequences 

 of the theory are stated in the eight preceding propositions. This 

 has been done, to shew with what ease a natural apparatus on the 

 construction of Tantalus's cup elucidates the appearances, which 

 have been ascribed by writers to the fountains of Dodona, Coma, 

 and Paderborn. The operations of these springs are happily illus- 

 trated by the instrument in question ; on which account I do not 

 hesitate to pronounce the theory to be a good one, so far as it 

 relates to these fountains alone; provided they are faithfully 

 described. The simplicity of the preceding explanation, and its 

 coincidence with the narratives of the two Plinys, as well as the 

 history of the inconstant brook in Westphalia, disposed me to 

 admit the common theory, and to imagine it to be equally applicable 

 to reciprocating fountains in general; until an instance occurred to 

 my notice, which proved that, fluctuating fountains do not univer- 

 sally exhibit the periodical operations which are described by the 

 writers already quoted. I made a visit to Giggleswick Well, in the 

 autumn of 179^5 which taught me to value this once favourite 

 theory not so highly, and in particular to dispute the universality of 

 its application. The causes of these doubts will be easily perceived 

 from the following description of the well and its operations. 



This spring lies at the foot of Giggleswick Scar, which is a hill of 

 limestone in the West Riding of Yorkshire. The water discharged 

 by it, falls immediately into a stone trough ; in the front of which 

 are two holes near the bottom ; these are the outlets of two streams, 

 that flow constantly from the artificial cistern. An oblong notch is 

 also cut in the same side of the trough ; which extends from the 

 brim of it, nearly to the level of the two holes already mentioned. 

 This aperture is intended to shew the fluctuations of the well : for 

 the water subsides in it when the stream issuing from the rock be. 

 comes languid ; on the contrary the surface of the water rises again 

 in the notch, so soon as the influx into the trough begins to be more 

 copious. The reciprocations of the spring are easily observed by this 

 contrivance ; and they appear to be very irregular both in respect 

 of duration and magnitude. For the interval of time betwixt any 



