A M M

A M M

( 50 )

surface of the vessebwhen exposed to a coldof—26°. But ammonia requires exposure du-ring some hours to potash to stand the testeven of the temperature of 0. and a singletransfer of the dried gas through mercury inordinary use, again communicates moisture toit. The gas employed in these experimentswas dried by sticks of pure potash fastened tosteel wire, so that they could be withdrawnthrough the confining mercury when theirdrying action had ceased, and this was ascer-tained by applying ether or salt and snow tothe outside of the glass. In some of thefirst experiments Dr. H. found that though thegas thus prepared deposited no moisture bycold before electrization, yet it gave a sensiblequantity afterwards ; but on applying heat justbefore the experiment both to the mercury andthe iron cistern containing it, this condensation ofmoisture after electrization no longer appeared;at least only in the slightest possible degree; sothat it seems necessary for perfect accuracy toheat the mercury and vessels as weil as to drythe gas, and even to boil the mercury in thetube before admitting the ammonia.

With these precautions given quantities ofammonia were subjected to long continuedelectrization, by which (as already mentionedin our former article) the bulk of the gas isnearly doubled, and when no further increasetakes place, the residue has now lost its alkalineproperties and is entirely resolvable into azoteand hydrogen. Dr. Henry found from themean of six experiments that 100 measures ofpure ammoniacal gas were resolved by electri-city into 198.78 measures of the mixed gasses,but the results of the experiment on which heplaced the most confidence, on account of thescrupulous precautions observed against the in-trusion of moisture, gave an increase of only180.6 from 109 of ammonia.

The proportion of the hydrogen and azoteto each other was ascertained by passing intothe mixture a known quantity of oxygen gas,firing it with the electric spark, and estimatingthe hydrogen thus consumed, the azote beingunaltered and constituting the chief part of theresidue. In making this estimation, two mea-sures of hydrogen are reckoned as saturatingone measure of oxygen for the composition ofwater. Even this part of the experiment isliable to some inaccuracy as Dr. H. observes,for it is necessary to use a redundance of oxygengas to explode the whole of the hydrogen, butit is by no means certain that a small portionof the azote will not also undergo combustion

into nitric acid which will be absorbed by thewater generated. On the whole, Dr. Henryestimates the entire mixed gasses produced bythe electrization of die ammonia, to be com-posed of about 74 measures of hydrogen to 23of azote.

The extreme labour of decomposing ammo-nia by mere electricity, * led the author to at-tempt a shorter mode of analysis. Ammoniaand atmospherical air will not explode by theelectric spark in any proportions, but Dr. H.found that ammonia and oxygen gas when min-gled in proper proportions, may be detonatedover mercury by the electric spark with as muchease as a mixture of hydrogen and oxygen, andthis experiment affords a very ready way of ana-lyzing ammonia with infinitely less labour thanby simple electrization, and with as much accu-racy, provided proper precautions are observed.

With a greater proportion of pure oxygen gasto ammonia than that of three to one, or ot am-monia to oxygen than that of three to 1.4 the mix-ture ceases to be combustible, bur when the pro-portions best adapted to inflammation are used,the oxygen.may be diluted with six times its bulkof common air, without losing its property ofburning ammonia.

The products of the combus ion of ammoniawith oxygen vary essentially, according to theproportion of the gasses used. If the oxygengas be as much as double the ammonia, the am-monia entirely disappears, aswellasthe hydrogen,and the only residue is azote with the excess ofoxygen. The moment after the detonation, awhite cloud settles on the inner surface of the tube,which is too small for analysis but is probablynitrat of ammonia, the nitric acid being formedby the oxygenation of a portion of the azote.Hence when the oxygen of the residue is removedby sulphuret of lime or any other eudiometricalprocess, the quantity of azote remaining is lessthan ought to be produced from the ammoniadecomposed.

On the other hand when less oxygen is addedat first to the ammonia than is- sufficient to satu-rate all the hydrogen, that is to say, when theammonia much exceeds the oxygen, no nitrat ofammonia appears, the mixture being free fromcloudiness, and the mixed gas remaining afterthe first detonation will completely lose its hy-drogen by a second admixture with oxygen andfiring, and the azote will remain untouched.

Dr. Henry gives the results of comparativeexperiments in the two* cases, that is, where anexcess of oxygen was first employed;, and where adeficiency was used. In the first case the oxy-