Eddington has speculated in a most interesting way on this possible source of stellar heat in his recent presidential address before the British Association for the Advancement of Science (see _Nature_, September 2, 1920). He points out that the old contraction hypothesis, according to which the source of solar and stellar heat was supposed to reside in the slow condensation of a radiating mass of gas under the action of gravity, is wholly inadequate to explain the observed phenomena. If the old view were correct, the earlier history of a star, from the giant stage of a cool and diaphanous gas to the period of highest temperature, would be run through within eighty thousand years, whereas we have the best of evidence that many thousands of centuries would not suffice. Some other source of energy is imperatively needed. If 5 per cent of a star's mass consists originally of hydrogen atoms, which gradually combine in the slow process of time to form more complex elements, the total heat thus liberated would more than suffice to account for all demands, and it would be unnecessary to assume the existence of any other source of heat. [Illustration: Fig. 40. Spiral nebula in Ursa Major (Ritchey). Luminous matter, in every variety of physical and chemical state, is available for study in the most diverse celestial objects, from the spiral and irregular nebulae through all the types of stars. Doctor van Maanen's measures of the Mount Wilson photographs indicate outward motion along the arms of spiral nebulae, while the spectroscope shows them to be whirling at enormous velocities.] COSMIC PRESSURES