[52] A short notice of them first appeared in the third number of the "Journal of Science and the Arts," edited at the Royal Institution. He detects, in the first instance, the general principle of inflammable gas, in a state of combustion, being arrested in its progress by capillary tubes; he next applies it to the construction of a Safety-lamp, and then, by observing the phenomena which this lamp exhibits, is led to novel views respecting the nature and properties of flame.--I shall endeavour to offer a popular view of the curious and interesting truths disclosed by this latter research. He had observed that, when the coal gas burnt in the iron cage, its colour was pale, and its light feeble; whereas the fact is rendered familiar to us all by the flame of the gas lights, that in the open air carburetted hydrogen burns with great brilliancy. Upon reflecting on the circumstances of the two species of combustion, he was led to believe that the cause of the superiority of the light in the latter case might be owing to a _decomposition_ of a part of the gas towards the interior of the flame, where the air was in the smallest quantity; and that the consequent deposition of charcoal might first by its _ignition_, and afterwards by its _combustion_, contribute to this increase of light. A conjecture which he immediately verified by experiment.[53] [53] This theory of Davy is well illustrated by the change produced in the flame of gas-light, when acted upon by the wind, as may be seen during an illumination. The loss of light under these circumstances evidently arises from the more rapid combustion of the gas, by its more complete admixture with air; in consequence of which the decomposition above described does not take place. The intensity therefore of the light of flames depends principally upon the production and ignition of _solid_ matter in combustion, so that heat and light are in this process independent phenomena. These facts, Davy observes, appear to admit of many applications; in explaining, for instance, the appearance of different flames--in suggesting the means of increasing or diminishing their light, and in deducing from their characters a knowledge of the composition of their constituent parts. The point of the inner blue flame of a candle or lamp urged by the blow-pipe, where the heat is the greatest and the light the least, is the point where the whole of the charcoal is burnt in its gaseous combinations, without previous ignition. The flames of phosphorus and of zinc in oxygen, and that of potassium in chlorine, afford examples of intensity of light depending upon the production of _fixed_ solid matter in combustion; while on the contrary, the feebleness of the light of those flames, in which gaseous and volatile matter is alone produced, is well illustrated by those of hydrogen and sulphur in oxygen, or by that of phosphorus in chlorine.