IN III THE PHOTOSYNTHESIS OF NATURALLY my second lecture I laid some stress on the fact that every chemical reaction must consist of three separate and distinct phases. Since molecules normally exist in nonreactive phases, it is necessary before any reaction can take place to convert these phases into the reactive phases by the supply of energy. The first stage of every reaction consists in the supply of a definite increment of energy to each molecule, this increment being determined by the difference in energy content of the reactive and non-reactive phases. This first stage is followed by the second stage or reaction proper, in which the reactive phases of the resultants are produced. The third stage consists of the loss of energy by the reactive phases of the resultants, whereby their normal or non-reactive phases are formed. It follows from this development of our knowledge of chemical change that reactions may be divided into two classes, namely, those in which the increment of energy necessary for the activation of the reactant molecules is small and those in which this increment is large. This subdivision is very important, because the ordinary methods used in the laboratory for the activation of molecules, namely, the action of heat, solvent, or material catalyst, are only capable of supplying small quantities of energy and hence of stimulating reactions the necessary energy increment of which is small. The experience of chemists who use these methods is, therefore, entirely restricted to this class of reaction, and as long as we continue only to use these methods we cannot gain knowledge of those reactions the necessary increment of which is large. This latter class of reaction is, to a great extent, a new field, a study of which promises well, since results may be expected which are entirely outside the experience gained by a study of the former. In my two previous lectures I have brought forward evidence in favour of a specific reac tivity peculiar to each phase of a given molecule. The reactivities of the phases of a molecule with large energy content will be very different from those of the phases of the same molecule with small energy content; and I feel, therefore, that I am fully justified in saying that, if it be found possible to carry out reactions of the second class, the results obtained will be of more than ordinary interest. A moment's consideration will show that any highly endothermic reaction must belong to the second class, since it is obvious that the larger is the excess of the activating increment of energy over and above the energy radiated in the third stage, the greater must be that activating increment. In all probability, therefore, a reaction of the second class must be sought amongst the highly endothermic reactions. The question arises as to the possibility of supplying in such a reaction the very large energy increment necessary to activate the molecules, since, as has already been pointed out, the ordinary methods of laboratory practice are unable to supply to any molecule the large number of molecular quanta which together make this critical increment. It might be considered possible to supply the large critical increment by working at a very high temperature such as that of the electric furnace, but this is impossible owing to the fact that the products of the reaction will be unstable at those high temperatures. On the other hand the molecular phase hypothesis at once suggests a method of supplying the necessary energy increment which is free from the difficulties caused by intensely high temperatures. If the energy is supplied to the reactant molecules at the frequency characteristic of their non-reactive phase, the energy so gained will be sufficient to activate them for any reaction, however endothermic this may be, since it may readily be proved that the energy quantum characteristic of a phase is sufficient just to resolve it into its component atoms. From the theoretical point of view, therefore, it should be possible to bring about any reaction, even a highly endothermic one, by activating the molecules by light of the same frequency 'as that characteristic of the phases of the reactants which are present. A typical endothermic reaction is the conversion of carbonic acid into formaldehyde and oxygen, since the reverse reaction, the combination of oxygen with formaldehyde to form carbon dioxide and water, evolves about 127,000 calories per gram molecule. It is known that carbonic acid exhibits an absorption band with a central wave-length of about 2000 Angstroms and the energy quantum characteristic of that phase is 9.84 X 10-12 erg which corresponds to 144,000 calories per gram molecule. This is a measure of the energy absorbed by carbonic acid when exposed to light of the wave-length 2000 Angstroms and is much larger than the observed heat of the reaction. It follows that we have at hand in the absorption of light at the phase frequency a method of supplying the energy necessary to activate the molecules in a reaction of the second class, a method which is theoretically ideal in the sense that the activated molecules of the product are able at once to lose use these methods is, therefore, entirely restricted to this class of reaction, and as long as we continue only to use these methods we cannot gain knowledge of those reactions the necessary increment of which is large. This latter class of reaction is, to a great extent, a new field, a study of which promises well, since results may be expected which are entirely outside the experience gained by a study of the former. In my two previous lectures I have brought forward evidence in favour of a specific reactivity peculiar to each phase of a given molecule. The reactivities of the phases of a molecule with large energy content will be very different from those of the phases of the same molecule with small energy content; and I feel, therefore, that I am fully justified in saying that, if it be found possible to carry out reactions of the second class, the results obtained will be of more than ordinary interest. A moment's consideration will show that any highly endothermic reaction must belong to the second class, since it is obvious that the larger is the excess of the activating increment of energy over and above the energy radiated in the third stage, the greater must be that activating increment. In all probability, therefore, a reaction of the second class must be sought amongst the highly endothermic reactions. The question arises as to the possibility of supplying in such a reaction the very large energy increment necessary to activate the molecules, since, as has already been pointed out, the ordinary methods of laboratory practice are unable to supply to any molecule the large number of molecular quanta which together make this critical increment. It might be considered possible to supply the large critical increment by working at a very high temperature such as that of the electric furnace, but this is impossible owing to the fact that the products of the reaction will be unstable at those high temperatures. On the other hand the molecular phase hypothesis at once suggests a method of supplying the necessary energy increment which is free from the difficulties caused by intensely high temperatures. If the energy is supplied to the reactant molecules at the frequency characteristic of their non-reactive phase, the energy so gained will be sufficient to activate them for any reaction, however endothermic this may be, since it may readily be proved that the energy quantum characteristic of a phase is sufficient just to resolve it into its component atoms. From the theoretical point of view, therefore, it should be possible to bring about any reaction, even a highly endothermic one, by activating the molecules by light of the same frequency 'as that characteristic of the phases of the reactants which are present. A typical endothermic reaction is the conversion of carbonic acid into formaldehyde and oxygen, since the reverse reaction, the combination of oxygen with formaldehyde to form carbon dioxide and water, evolves about 127,000 calories per gram molecule. It is known that carbonic acid exhibits an absorption band with a central wave-length of about 2000 Angstroms and the energy quantum characteristic of that phase is 9.84 X 10-12 erg which corresponds to 144,000 calories per gram molecule. This is a measure of the energy absorbed by carbonic acid when exposed to light of the wave-length 2000 Angstroms and is much larger than the observed heat of the reaction. It follows that we have at hand in the absorption of light at the phase frequency a method of supplying the energy necessary to activate the molecules in a reaction of the second class, a method which is theoretically ideal in the sense that the activated molecules of the product are able at once to lose |