XXIV. The experiments to prove this, were to the admiration of all the by-standers, repeatedly made with the burning-glass of Mons. Villete, which we have spoken of before; and it was observed at the time of making them, that by so much the more cold the burning-glass was, by so much the quicker and stronger was the operation of the focus; and that by so much the warmer it was, by so much the more tardy and remiss were its effects. Among those who read these accounts, some looked upon them as wonderful, and others as incredible. XXV. But with all this, the reason of the phenomenon is not very hidden. It is certain, that cold condenses bodies, and that heat dilates them. It is also certain, that the more dense a body is, the more apt and fit it is to cause reflection, and that it is least so, the more lax it is. From these two premises it may be clearly inferred, that the effects we have mentioned were things of course; but for a further explanation of this matter, I will say they were so from two causes; for the burning-glass by being in a warm state, must consequently be less compact and hard, and the operations of its focus more remiss and languid. In the first place, because a great portion of the rays were absorbed in the pores of the metal, which had been dilated by the heat, and made no reflection at all. Secondly, by their being dilated, and as we may say, become more spungy, the concave superficies of the metal were less smooth and equal, from whence it would follow, that many rays by being obstructed in their descent by some insensible prominences, would not make their reflection in a right line on the point of the focus. This may be better understood, by attending to the reflections of a ball, that is struck against a wall with an unequal surface; and it is a position agreed to by all the mathematicians who have treated of catoptrics, that light and heat in their reflections, follow exactly the same rules, that heavy bodies do. XXVI. Nor is there are any difficulty in supposing, that a body as hard as metal, may undergo some insensible rarefaction when it grows warm; for in the first place, if a very intense heat can dilate metal, so as to make it break all its ligatures and become fluid, a less heat would have the same effect, in proportion to the degree of it, and might dilate and rarefy it a little. Secondly, experience teaches us, that every sort of metal is more sonorous when it is cold, than when it is warm; from whence it may be evidently inferred, that heat and cold do somewhat alter its texture; it being certain, that upon the texture, depends its being more or less sonorous. PARADOX VIII. _A flame extending itself upwards in a pyramidical or conic form, is occasioned by a violence done to the flame itself._ SECT. VIII.