380 PROCEEDINGS OF THE AMERICAN ACADEMY 



" To render probable tbe supposition of olefiant gas being the 

 source of the gas-carbon, it has been generally stated that this bi- 

 carburet loses two of its four proportions of carbon by heat, and be- 

 comes converted into marsh gas, or light carburet of hydrogen, the 

 formula of which is C0H4 ; and thus the definiteness of an exact re- 

 sult is presented. 



" In the manufacture of gas for lighting, an increased temperature 

 in the retort diminishes the illuminating power of the gas, and hence 

 it has been assumed that the illuminating effect of the gas is dimin- 

 ished by a loss of the carbon contained in the olefiant gas, to which 

 a large part of the light-giving quality has been attributed. 



" It becomes an interesting point in general chemical science, to 

 learn how far the facts gained by observation and experiment will sus- 

 tain these assumptions which have been held in relation to the source 

 of gas-carbon as above alluded to, and to inquire into its connections 

 more particularly. 



" Gas-carbon, in its difficult combustibility under a current of heated 

 air, its relation to nitrates of the alkalies and sulphuric acid, must be 

 classed with the carbon found in crude iron, and called graphitic car- 

 bon, or carbon in an allotropic state. It differs as much from lamp- 

 black and charcoal as these do from diamond, and in the artificial 

 production of it, in all the cases hitherto observed, it has a certain rela- 

 tion to vapors. The fine specimens obtained when molten iron passes 

 over moist earth, the metallic-like glazing of coke, and the lustrous 

 residues of animal decomposition by heat, in presence of vapors, are 

 all instances of the existing connection between vapors and this allo- 

 tropic carbon. 



" Taking the suite of specimens before us, the microscope enables 

 us to see, in the early stages of deposition, that every part is vesicu- 

 lar ; that mammillary forms result from the aggregation of the vesicles ; 

 and, pursuing these observations, we often find the broken vesicles fill- 

 ing vacant spaces between those more perfect, and a consolidation re- 

 sulting from this arrangement. Where pendent parts exist, their sec- 

 tions show a perfectly regular building up from layers of sublimate, 

 each layer being composed of vesicles, more or less broken ; the thin 

 shell of each exhibiting the superposition of layers which belongs to 

 bubbles. The examination of hundreds of specimens will not show 

 any departure from this character of a sublimate, produced either 

 from its own vapor, or when transported by another kind of vapor. 



