Tyndall was an experimenter and laboratory apparatus builder, not an abstract model builder. But in his experiments on radiation and the heat-absorptive power of gases, he had an underlying agenda to understand the physics of molecules. Tyndall said in 1879: "During nine years of labour on the subject of radiation [in the 1860s], heat and light were handled throughout by me, not as ends, but as instruments by the aid of which the mind might perchance lay hold upon the ultimate particles of matter." This agenda is explicit in the title he picked for his 1872 book Contributions to Molecular Physics in the Domain of Radiant Heat. It is present less explicitly in the spirit of his widely read 1863 book Heat Considered as a Mode of Motion. Besides heat he also saw magnetism and sound propagation as reducible to molecular behaviours. Invisible molecular behaviours were the ultimate basis of all physical activity. With this mindset, and his experiments, he outlined an account whereby differing types of molecules have differing absorptions of infrared radiation because their molecular structures give them differing oscillating resonances. He'd gotten into the oscillating resonances idea because he'd seen that any one type of molecule has differing absorptions at differing radiant frequencies, and he was entirely persuaded that the only difference between one frequency and another is the frequency. He'd also seen that the absorption behaviour of molecules is quite different from that of the atoms composing the molecules. For Example, the gas nitric oxide (NO) absorbed more than a thousand times more infrared radiation than either nitrogen (N2) or oxygen (O2). He'd also seen in several kinds of experiments that – no matter whether a gas is a weak absorber of broad-spectrum radiant heat – any gas will strongly absorb the radiant heat coming from a separate body of the same type of gas. That demonstrated a kinship between the molecular mechanisms of absorption and emission. Such a kinship was also in evidence in experiments by Balfour Stewart and others, cited and extended by Tyndall, that showed with respect to broad-spectrum radiant heat that molecules that are weak absorbers are weak emitters and strong absorbers are strong emitters. (For Example, rock-salt is an exceptionally poor absorber of heat via radiation, and a good absorber of heat via conduction. When a plate of rock-salt is heated via conduction and let stand on an insulator, it takes an exceptionally long time to cool down; i.e., it's a poor emitter of infrared.) The kinship between absorption and emission was also consistent with some generic or abstract features of resonators. The chemical decomposition of molecules by lightwaves (photochemical effect) convinced Tyndall that the resonator could not be the molecule as a whole unit; it had to be some substructure, because otherwise the photochemical effect would be impossible. But he was without testable ideas as to the form of this substructure, and did not partake in speculation in print. His promotion of the molecular mindset, and his efforts to experimentally expose what molecules are, has been discussed by one Historian under the title "John Tyndall, The Rhetorician of Molecularity".