I want you, if you will, to go a step further The heat of the flame is due, as I explained in my last lecture, to the clashing of molecules. But what is the light of my candle and gas due to? The light is due to the solid matter in the flame, brought to a state of white heat or incandescence by the heat of the flame. The heat is due to the clashing of the particles, the light is due to the heated solid matter in the flame. Let me see if I can show you that. I am setting free in this bottle some hydrogen, which I am about to ignite at the end of this piece of glass tube (Fig. 38 A). I shall be a little cautious, because there is danger if my hydrogen gets mixed with air. There is my hydrogen burning; but see, it gives little or no light. But this candle flame gives light. Why? The light of the candle is due to the intensely heated solid matter in the flame; the absence of light in the hydrogen flame depends on the absence of solid matter. Let me hold clean white plates over both these flames. See the quantity of black solid matter that I am able to collect from this candle flame (Fig. 38 B). But my hydrogen yields me no soot or solid matter whatsoever (Fig. 38 A). The plate remains perfectly clean, and only a little moisture collects upon it. The light that candle gives depends upon the solid matter in the flame becoming intensely heated. If what I say be true, it follows that if I take a flame which gives no light, like this hydrogen flame (Fig. 39 A), and give it solid particles, I ought to change the non-luminous flame into a luminous one. Let us see whether this be so or not. I have here a glass tube containing a little cotton wadding (Fig. 39 B _a_), and I am about to pour on the wadding a little ether, and to make the hydrogen gas pass through the cotton wadding soaked with ether before I fire it. And now if what I have said is correct, the hydrogen flame to which I have imparted a large quantity of solid matter ought to produce a good light, and so it does! See, I have converted the flame which gave no light (Fig. 39 A) into a flame which gives an excellent light merely by incorporating solid matter with the flame (Fig. 39 B). What is more, the amount of light that a flame gives depends upon the amount or rather the number of solid particles that it contains. The more solid particles there are in the flame, the greater is the light. Let me give you an illustration of this. Here is an interesting little piece of apparatus given to my predecessor in the chair of chemistry at the London Hospital by the Augustus Harris of that day. It is one of the torches formerly used by the pantomime fairies as they descended from the realms of the carpenters. I have an alcohol flame at the top of the torch which gives me very little light. Here, you see, is an arrangement by which I can shake a quantity of solid matter (lycopodium) into the non-luminous alcohol flame. You will observe what a magnificently luminous flame I produce (Fig. 40).