The conservation of energy is a legitimate enough generalisation: we do not really doubt its conservation and constancy when we admit that we are not yet sure of having fully and finally exhausted the whole category of energy. What we do grant is, that it may hereafter be possible to discover new forms; and when new forms are discovered, then either the definition may have to be modified, or else the detailed statement at present found sufficient will have to be overhauled. But after all, this is not specially important: the _serious_ mistake which people are apt to make concerning this law of energy is to imagine that it denies the possibility of guidance, control, or directing agency, whereas really it has nothing to say on these topics; it relates to _amount_ alone. Philosophers have been far too apt to jump to the conclusion that because energy is constant, therefore no guidance is possible, so that all psychological or other interference is precluded. Physicists, however, know better; though unfortunately Tyndall, in some papers on Miracles and Prayer, thoughtlessly adduced the conservation of energy as decisive. This question of "guidance" is one of great interest, and I emphasise the subject further on, especially in Chapter IX. _Conservation of Matter._ Take next the "conservation of matter"—which means that in any operation, mechanical, physical, or chemical, to which matter can be subjected, its amount, as measured by weight, remains unchanged; so that the only way to increase or diminish the weight of substance inside a given enclosure, or geometrically closed boundary, is to pass matter in or out through the walls. This law has been called the sheet-anchor of chemistry, but it is very far from being self-evident; and its statement involves the finding of a property of matter which experimentally shall remain unchanged, although nearly every other property is modified. To superficial observation nothing is easier than to destroy matter. When liquid—when dew, for instance—evaporates, it seems to disappear, and when a manuscript is burnt it is certainly destroyed: but it turns out that there is something which may be called the vapour of water, or the "matter" of the letter, which still persists, though it has taken rarer form and become unrecognisable. Ultimately, in order to express the persistence of the permanent abstraction called "matter" clearly, it is necessary to speak of the "ultimate atoms" of which it is composed, and to say that though these may enter into various combinations, and thereby display many outward forms, yet that they themselves are immutable and indestructible, constant in number and quality and form, not subject to any law of evolution; in other words, totally unaffected by time.