of Kentucky and Afrka. 14t 



tive account of the nitre caverns, &c. of Kentucky, published 

 in the Transactions of the Philosophical Society, in Philadel- 

 phia, Vol. VI., and in Bruce's Journal, Vol. I. p. 100. The 

 following communication arose from a conversation on that 

 subject between Dr. Brown and the Editor. 



New-Haven, July 27, 1818. 

 Dear Sir, 



I have just found the passage I referred to the other day, 

 relative to the existence of native or sandrock nitre in the 

 interior of Southern Africa. It is in Barrow, and not in Vail- 

 lant, as I thought when I had the pleasure of conversing with 

 you concerning it. I am much obliged to you for recalling 

 my attention to that curious subject, as it has brought to my 

 recollection a fact, which I believe I omitted to mention in 

 my memoir, (viz.) the existence of a black substance in the 

 clay under the rocks, of a bituminous appearance and smell. 

 This I remember to have seen in a rock-house, near the Ken- 

 tucky river, where very considerable quantitites of sandrock 

 nitre Had been obtained. This substance was found in masses 

 of a few ounces weight, and in the crevices of the rocks near 

 the basis of the side walls. The smell was not wholly bitu- 

 minous, but resembled that of bitumen combined with musk. 

 I am quite unable to account for the formation of the nitre, or 

 the production of this black substance which sometimes accom- 

 panies it, both in Africa and America. Had I seen Mr. Bar- 

 row's travels, when I noticed the bitumen, I should certainly 

 have paid more attention to it. But perceiving no relation 

 between the rock nitre and the masses of this substance, my 

 examination of it was much too superficial. I do not very 

 well understand what Mr. Barrow means by saying, that many 

 wagon loads of animal matter lay on the roof of the caverns in 

 Africa. I saw no such matter on the roof of the rock-houses 

 in Kentucky. Certainly the caverns have been the habita- 

 tions of wild beasts, and great quantities of leaves, &c. have 

 been mixed with the debris of the superincumbent rocks, 

 but it does not seem probable, that much animal matter could 

 be filtrated through a roof of rock, perhaps forty or fifty fpct 



