Azotic gas, the other constituent of atmosphe rical air, is chiefly recognised by its negative qualities. It possesses the mechanical properties of air ; it does not support combustion ; and no animal can breath it without death. It constitutes the basd of nitric acid, and is one of the constituents of ammonia. There is reason to consider it as a compound body, but hitherto chemists have not been able to ascertain its constituents ; though several extraordinary phe nomena, observed during the decomposition of am monia by Davy and Berzelius, cannot well be ac counted for, without supposing hydrogen to be a con stituent of it. It has been supposed a compound of hydrogen and oxygen ; but several circumstances militate against this opinion. Mr Davy has been for some time occupied incessantly in attempts to ascertain its composition, but hitherto without success. Till the discovery be made, some of the most interesting parts of chemistry remain involved in impenetrable obscurity. The specific gravity of azotic gas, according to Kirwan, is 0.985, that of air being 1.000 ; while, according to Lavoisier and Davy, it is 0.978. This last esti mate we are disposed to consider as most correct. If so, 100 cubic inches of it, at the temperature of 60° when the barometer stands at 30 inches, weigh 29.83 grains troy.
Reckoning the specific gravity of oxygen gas 1.093, and that of azotic gas 0.978, and supposing atmospherical air to be composed of 79 parts of azote ' and 21 oxygen by bulk, it follows, that 100 parts of ' it in weight are composed of 77.43 azote 22.57 oxygen 100.00 Though it has been ascertained, that these two constituents of air n..ver vary in their proportions, yet as the methods of analysing air are very useful in all chemical investigations of gaseous bodies, and have led to many discoveries of importance, it will be pro per to give an account of them lucre. They consist in the application of various substances to a given bulk of air, which have the property of absorbing •) and removing the oxygen, but which do not act upon the azote. The diminution of bulk gives the quan tity of oxygen ; the residue that of azote. The ap paratus contrived for these experiments, received the name of eudiometers, because they were considered at first as measurers of the goodness of air. For it was supposed that the proportion of oxygen was va-* riable, and that the salubrious or noxious qualities of the air depended upon that proportion. Ingenhousz thought he disco ered, that the atmosphere above the sea contained more oxygen than that above the land; hence he accounted for the supposed salubrity of the sea air, which has been highly extolled from the remotest times.
sThe first. eudiometer was applied, in consequence of Dr Priestley's discovery, that nitrous gas absorbs the oxygen from common air. When nitrous gas comes in contact with oxygen gas, they immediately combine and form nitric acid; and if the mixture be standing over water, the acid is immediately absorbed by the liquid. Hence the bulk of a mixture of ni trous gas and common air immediately diminishes, and the diminution is proportional to the quantity of oxygen in the air, supposing all other circumstances the same, and of course measures that quantity. Dr Priestley's method was, to let up into a gradua ted tube 100 measures of air, and then to add 100 measures of nitrous gas. The mixture became yel low, and its bulk diminished. He denoted the good ness of the air by the residual gas. Thus if 114
parts remained, he said that the goodness of the air, by the test of nitrous gas, was 114.; of course the smaller the residue, the greater was the goodness Of the air. This method did not ascertain the absolute quantity of oxygen. It was soon observed, that even when the air operated upon was absolutely the same, the residue was liable to considerable variation from apparently trifling circumstances. Thus, for example, if the tube was agitated during the mixture, it was observed that the residue was always much less than if no agitation was applied. If the tube was narrow, the residue was always more considerable than if the tube was wide. The purity of the water, too, over which the experiment was made, had considerable influence. Mr Cavendish observed, that if the water. was in such a state, that it frothed when agitated as if it had contained soap, then the residue was always less than it otherwise would he.
The apparatus was much improved by Fontana, who regulated the size of the tube and the manner • of mixing the gases ; hence the instrument is usually known by the name of Fontana's cudiometer. This eudiometer was employed by Ingenhousz, and the variations which he found in the compositions of the atmosphere, were obviously owing to the errors to which it was liable. Mr Cavendish first pointed out the precautions necessary, in order to ensure accuracy when this etn:;ometer was employed. But before nitrous gas could be used with advantage in the ana lysis of air, it was necessary to ascertain the propor tion of it which combined with a given bulk of oxy-• gen gas. This was first undertaken by Mr Dal ton, (Phil. Meg. xxiii. :351.) According to him, 21 parts of oxygen gas are capable of uniting either with 36 measures of nitrous gas, or with 2 x 36=72 measures. Both of these compounds are soluble in water. If the tube in which the mixture is made be wide, and if agitation be employed, the two gases come at once in contact, so that the oxygen com bines with a maximum of nitrous gas. If the tube be narrow, and if no agitation be employed, the oxy gen gas combines with a minimum of nitrous gas. In tubes of intermediate bore, the proportion of ni- • trous gas which combines with the oxygen, is inter mediate between 36 and 72. Hence his rule is to employ a tube of so small a 'bore, that water can just be poured easily out of it ; to put up into this tube the quantity of air to be examined, and then to let. tip a quantity of nitrous gas equal to about half the bulk of the air ; to allow this mixture to remain two or three minutes without any and then to observe the diminution of bulk. This diminution is to be multiplied by 44-, or 0.368. The product is the bulk of the oxygen gas contained in the air examin ed. Suppose we employ 100 measures of air, and let up 50 measures of nitrous gas, and that the dimi nution of bulk amounts to 57 ; then 57 x 0.368= 20.976, or very nearly 21. This indicates, that the. 100 measures of air contain 21 measures of.oxygen. gas. 'We have tried the method of Dalton very fre quently, and have. found that when the tube is suf ficiently narrow, the experiment carefully made, the mean error cannot be rated higher than 1 per cent. When the gas examined contains much more oxygen than common air, and above all, when it is almost pure oxygen, the error is greater ; so great, indeed, that the method cannot be depended on.