It will perhaps seem strange to class oxygen as a crystalloid. But inasmuch as the crystalloids are distinguished from the colloids by their atomic simplicity, and inasmuch as sundry gases are reducible to a crystalline state, we are justified in so classing it.
The remark made by a critic to the effect that in a mammal higher temperature diminishes the rate of molecular change in the tissues, leads me to add that the exhalation I have alleged is prevented if the heat rises above the range of variation normal to the organism; since, then, unusually rapid pulsations with consequent inefficient propulsion of the blood, cause a diminished rate of circulation. To produce the effect referred to in the text, heat must be associated with dryness; for otherwise evaporation is not aided. General evidence supporting the statement I have made is furnished by the fact that the hot and dry air of the eastern deserts is extremely invigorating; by the fact that all the energetic and conquering races of men have come from the hot and dry regions marked on the maps as rainless; and by the fact that travellers in Africa comment on the contrast between the inhabitants of the hot and dry regions (relatively elevated) and those of the hot and moist regions: active and inert respectively.
The increase of respiration found to result from the presence of light, is probably an indirect effect. It is most likely due to the reception of more vivid impressions through the eyes, and to the consequent nervous stimulation. Bright light is associated in our experience with many of our greatest outdoor pleasures, and its presence partially arouses the consciousness of them, with the concomitant raised vital functions.
To exclude confusion it may be well here to say that the word "atom" is, as before explained, used as the name for a unit of a substance at present undecomposed; while the word "molecule" is used as the name for a unit of a substance known to be compound.
On now returning to the subject after many years, I meet with some evidence recently assigned, in a paper read before the Royal Society by Mr. J. W. Pickering, D.Sc. (detailing results harmonizing with those obtained by Prof. Grimaux), showing clearly how important an agent in vital actions is this production of isomeric changes by slight changes of conditions. Certain artificially produced substances, simulating proteids in other of their characters and reactions, were found to simulate them in coagulability by trifling disturbances. "In the presence of a trace of neutral salt they coagulate on heating at temperatures very similar to proteid solutions." And it is shown that by one of these factitious organic colloids a like effect is produced in coagulating the blood, to that "produced by the intravenous injection of a nucleoproteid."