Ch. VIII. SOUTH AMERICA. 169 



from whence it was taken. Some time after, it un- 

 dergoes a fecond operation ; but without yielding fo 

 much juice as at firft and at the third or fourth very 

 little, by which means the fiili is exhaufted beyond re- 

 covery. In 1744, being in the lieutenancy of Santa 

 Elena, I had the fatisfadion to fee this liquor ex- 

 tradled according to the firft procels, and Ibme threads 

 died with it. This purple is far from being fo com- 

 mon as f®me authors have imagined for, though the 

 fiili increafes, yet fo large a quantity is neceflary to 

 dye a few ounces of thread, that little of it is feen ; 

 and indeed its great price is partly owing to its fcarcity. 

 Another circumftance worthy obfervation, and which 

 increafes or diminifhes the value, is the difference of 

 weight and colour of the cotton dyed with it, accord- 

 ing to the different hours of the day. I could not find 

 any fatisfadtory account of this property at Punta de 

 Santa Elena, where the inhabitants, being lefs curi- 

 ous, have not carried their fpeculations fo far as to 

 be acquainted v^ith this remarkable fingularity ; 

 whereas at Nicoya it is fo well known, that the dealers 

 in it, both buyers and fellers, are exadrly acquainted 

 with the times of its increafe or decreafe, fo that one 

 of the firfl: preliminaries to a contrad; is, to fettle the 

 time when it fhall be v/eighed. From this alteration 

 of the weight of the purple thread at Nicoya it may 

 be inferred, that the fame happens at Punta de Santa 

 Elena ; the turbines at both places being exadly of 

 the fame fpecies, and without the leaft vifible dif- 

 ference in colour. Another very remarkable par- 

 ticular relating to its tind, and which I have heard 

 from perfons of undoubted veracity, is, that the co- 

 lour of a thread of flax is very different from that of 

 a thread of cotton. It would therefore be proper to 

 make repeated experiments, on threads of filk, flax, 

 and vv^ool. 



Some, by faying that the fiOi, from whence this 

 dye is extracted, breeds in a ihell, by which either the 



fiat 



1 



