S3 



priving air of part of its elasticity, and that for this reason it is more 

 compressed by the external air. In order to extricate myself from these 

 doubts I first supposed that such air ought to be specifically heavier 

 than common air, both on account of the i!)hlogiston it had gained, and its 

 greater density. But how great was my astonishment upon finding that 

 a very thin retort filled with this kind of air and Aveighed in the nicest 

 manner, was not only not heavier than an equal measure of common 

 air, but even somewhat lighter ! I then imagined that the last suppo- 

 sition might be applicable, and then it would follow that the lost air 

 might again be separated from the materials employed in the experi- 

 ment." With this view he tries whether " the lost air had not been 

 changed into fixed air," and failing in finding any, as the next resource, 

 alludes to the idea that " the phlogiston when united with this air 

 might make it less ponderous ;" " however, since phlogiston," he adds, 

 "is a substance (which always supposes some weight), I very much doubt 

 whether this hypothesis is founded on truth ;" and he concludes at last 

 by saying, " I will prove that by union of air [oxygen] with the inflam- 

 mable principle (phlogiston) a compound is formed so subtle as to pass 

 through the fine pores of the glass and disperse all over the air." This 

 subtle compound Avas the matter of heat and light : in the shape of these 

 incoercible and imponderable agents, oxygen, and the elements which 

 enjoyed the title of phlogiston, became free to pass to and fro 

 unquestioned ; and thus Scheele cut the knot which he was not able to 

 untie. He afterwards informs us that " this generation, or new com- 

 position of heat and light by an union of air of fire with more or less phlo- 

 giston, obtained" not much applause ; " but could I," he adds, "naturally 

 conclude otherwise, when I saw a mixture of air of fire and inflamma- 

 ble air after its explosion totally disappear, but that the fire air, when 

 united with the inflammable, must have penetrated through the glass ? 

 because nothing could be observed on the outside of the glass in which 

 the explosion was effected but heat and light ; also in the water over 

 which it was effected in close vessels, notwithstanding I varied the ex- 

 periment in different ways, I could perceive nothing uncommon." " I 

 had often burnt a mixture of dephlogisticated and inflammable air, and 

 have always observed a dew in the glass immediately after the explo- 

 sion ; but I believed that air always contains some moisture and that 

 the inflammable air might also contain some moisture from the vi- 

 triolic acid, nay, that even the flame of the candle may yield some 

 dampness*." 



* Crell's Chem. Annals, 1785, vol. i, part 3, p. 229. 



