33^ Signs of the Eruption. [Book VI. 



that the wheels of a corn-mill, worked by that water, 

 moved very flowly ; it was neccfiary in all the other 

 wells of the town and its neighbourhood to lengthen 

 the ropes daily, in order to reach the water; and 

 ibine cf the wells became quite dry. Although moft 

 of the inhabitants were fenfible of this phenomenon, 

 not one of them feems to have fufpeclied the true 

 caufe of it. Eight days alfo before the eruption, a 

 man and two boys, being in a vineyard above Torre 

 del Greco (and precifely on the fpot where one of the 

 new .mouths opened, whence the principal current of 

 Java that deflroyed the town iffued) were much 

 alarmed by a fjdclen puff of fmoke which iffued from, 

 the earth clofe to them, and was attended with a flight 

 explofion. 



Had this circumftance, with that of the fubterraneous 

 ncifes heard at Rdina for two clays .before the eruption 

 (with the additional one of the decreafe of water 

 in the wells) been communicated at the time, it 

 would have required no great forefight to have been . 

 certain that an eruption of the volcano was near -at 

 hand, and that its force was directed particularly to- 

 wards that part of the mountain. 



On the 1 2th of June 1794, in the morning, there 

 was a violent fall of rain, and foon after the inhabitants 

 of Refma, fituated directly over the ancient town of 

 Herculaneum, were fenfible of a rumbling fubterra- 

 neous noife", which was not heard at Naples. 



From the month of January to the month of May, 

 the atmofphere had been generally calm, and there was 

 continued dry weather. In the month of May there 

 was a little rain, but the weather was unufually fultry. 

 For fome days preceding the eruption, the Duke della 

 Torre, a learned and ingenious nobleman, who pub- 

 liflied two letters upon the fubject of the eruption, 



obferved 



