FALLEN ON THE EARTH. 157 



pofed to a degree of heat fufficiently ftrong to oxidize, to a 

 confiderable depth, the iron crucible in which it was placed, not fufible in an 

 without its having undergone any alteration, except that of 

 having acquired a greater degree of intenfity in its colour. Its 

 tranfparency was not at all diminished. I think, therefore, 

 there is not the fmalleft reafon to allow any probability to the 

 opinion that it ought to be confidered as a kind of glafs. 



Of all fubftances hitherto known, that with which it feems it refembles thft 

 to have the greateft analogy, is the peridot, (the chryfolite ofP endot > 

 Werner,) to which fome mineralogifts have referred it. The 

 refult of Mr. Howard's analyfis of it, Is nearly the fame as that 

 of the analyfis of the peridot, made by Mr. Klaproth. 



The hardnefsand infufibility of this fubftance are nearly the in hardnefs an* 

 fame as thofe of the peridot ; but it feems to have a rather lefs infufibilit y- 

 degree of fpecific gravity : that of two very perfect cryftals of 

 peridot, I found to be from 3340 to 3375. The cryftaiiine 

 forms of the fubftance here defcribed, if ever we fhould be 

 able to determine them, would clear up our doubts refpecling 

 the analogy between the two fubftances. If we confider the 

 compact part of the fpecimen now treated of, particularly the 

 ftrong connection that appears to exift between the iron and It is curious that ■ 

 the tranfparent fubftance, and the great refiftance we ^pe-^^^^ 

 rience when we attempt to feparate them, we cannot help weather j 

 being furprifed, that almoft all the fpecimens of this mafs of 

 metallic iron that have been brought to Europe, are in the 

 cellular ftate already defcribed, owing apparently to the total, 

 or almoft total, deftruclion of the tranfparent fubftance. But, 

 betides the fragility of this fubftance, the fpecimen in queftion 

 helps very much to explain the above cirumftance, inafmuch 

 as many of the nodules of the tranfparent fubftance belonging 

 to it, are in a ftate of real decompofition. In that ftate, they 

 are changed into a white opaque fubftance, which, upon being 

 lightly prefled or fqueezed between the ringers, crumbles into 

 a gritty dry powder. This decompofition may be obferved to but it undoubt- 

 have taken place in various degrees : in many of the nodules, edl y 5s fo « 

 the fubftance is merely become friable, without being much 

 altered in its appearance ; whereas, fome of thofe which are 

 in a ftate of complete decompofition, are of an ochreous red- 

 difh yellow colour; it is, however, eafy to diftinguim that this 

 colour does not belong to them, but is owing only to the oxi- 

 idizement of the adjacent particles of iron. 



Froa| 



