OF THE EDDYSTONE LIGHTHOUSE. 45 



a complete model of the rock and building, which was 

 entirely approved. 



Then before setting to work, he had to examine 

 what stone and cement would answer best for his 

 purpose. Portland stone was by far the best for 

 working, and easy to obtain; but was subject to a 

 very important objection that, at the Dock- Yard, it 

 had been penetrated by sea worms, and rendered quite 

 porous. Granite, which best, and in fact completely 

 resisted the action of worms, was difficult and expensive 

 to work. He finally resolved to employ both these 

 stones ; using granite for the exterior, and building 

 the inside with Portland. 



He then entered into an extensive investigation, 

 accompanied with chemical analysis, of the best mate- 

 rials for making his cement ; which was required to 

 set quickly ; to resist the action of salt water ; and to 

 become harder and firmer with age. The result of 

 these experiments was, that equal measures of slaked 

 lime, from the blue lyas stone, and of Puzzolana an- 

 swered best, and that they did just as well, when mixed 

 up with salt water, as with fresh. He found too, that 

 the same ingredients, mixed with water into a liquid, 

 and poured into crevices, would fill them up just as 

 hard, as if put in stiff but it was necessary to add it 

 repeatedly, as the excess of water leaked off, leaving 

 the crevices only part full. 



These preliminaries being settled and a work yard 

 prepared, on the West side of Mill Bay; contracts 

 entered into for granite and Portland stone, to be sup- 

 plied in blocks of a ton each, cut according to moulds 

 sent to the quarries ; and suitable vessels hired for the 

 conveyance of the stones, apparatus and workmen to 

 the rock ; they began to prepare the foundation on the 

 6th of August, 1756. Subject to continual interrupt- 

 ions from the sea ; whereby they were often prevented 

 from working more than a few hours in a week, they 

 did not complete the foundation till late in November, 

 after which no more work could be done on the rock, 

 until June 3rd, 1757. At this time they fixed the 



