5'36 PHILOSOPHICAL TRANSACTIONS. [aNNO 1779. 



birds not only will die considerably sooner tban tlie otber, but will show signs 

 of uneasiness as soon as it is come to that place. The inflammable air there- 

 fore, when breathed together with a considerable cjuantity of common air, must 

 always swim at the top of it, filling the cavity of the windpipe, &c. while the 

 common air occupies the lower place, and filling the smallest pulmonary vesicles 

 is subservient to the ordinary functions of the luniks. 



Here I put an end to my observations on inflammable air considered with 

 respect to respiration; but I beg leave to add a few words respecting a property 

 of the inflammable air, which, as far as I know, has been overlooked by the 

 most diligent observers. I mean here to speak of such inflammable air as is 

 extracted from metals, by means of oil of vitriol, especially that extracted from 

 iron and zinc. The air of these metals, when presented to the flame of a 

 candle, not only burns with a whitish flame inclining to green, as is well known, 

 but exhibits a kind of sparks or explosions, which may be easily distinguished 

 between the body of the flame by their vivid light. These sparks, which are of 

 a vivid colour, dart in every direction. They might be easily taken for those 

 sparks that are emitted from red-hot iron; or they miglit be compared to very 

 small grains of gunpowder, if these were inflamed successively, and without 

 smoke; or they might even be compared to charcoal that sparkles, but without 

 any noise. This phenomenon seems very interesting, as it respects the nature 

 of the inflammable air itself. What seems most singular is, that this appearance 

 forms a distinctive character between the inflammable air of metals, and that 

 extracted from animal or vegetable substances; at least I may safely say, 

 that I never found the inflammable air of animal or vegetable substances 

 sparkle like that extracted from metals. In several of the former kinds of 

 air I could observe no sparkling at all; in others the sparks were so few that they 

 might be considered as nothing in comparison to the sparkling of the inflammable 

 air from metals. 



The inflammable air of metals itself, if left in contact with water for a long 

 time, or shaken in it till it becomes less inflammable, will in great measure lose 

 its sparkling property, and at last loses it entirely, when it is become in a state 

 of being hardly inflammable. I have observed, that the inflammable air is the 

 more difficult to be decomposed, by being shaken in water, as the number of the 

 sparks it shows when burning is greater; and according to this number of 

 sparks, the inflammable air makes greater or weaker explosions when mixed 

 with the dephlogisticated air: so that it proves by experiments, that the phlo- 

 gistic principle is more fixed, and in greater quantity combined, with the in- 

 flammable air of metals, than with that of vegetable or animal substances. I 

 do not mean to deny the possibility of finding other species of inflaininable air. 



