Section III., 1909. [227] Trans. R. S. C. 



XIV. — Convection and Stellar Variation. 



By Ralph E. DeLt ry, M.A., Ph.D. 



Presented by Dr. W. F. King. 



§1. The object of this paper ^ is to outline an h3'pothesis to explain 

 certain general conclusions drawn from the photometric and spectro- 

 metric data of variable stars. 



§2. It is supposed that the star is a body condensing under the 

 action of gravity and developing great quantities of heat which give rise 

 to rapid radial convection currents bearing masses of hot gases from 

 within and cooler and condensed materials back to the interior, and 

 that, in the absence of disturbing agents, a " kinetic equilibrium " is 

 established resulting in a steady and practically constant total emission 

 of light b}' the star. Since tliie radiation from the star depends on the 

 velocity and character of its convection currents, any change in these 

 convections causes a change in the heat and light emitted. Consequently, 

 to account for the variations in the light of some stars, it is assumed 

 that there are changes in the convection currents of the stars caused by 

 the changing action of disturbing agents. 



§3. The nature of these convection currents is revealed to some 

 extent by the study of the sun's atmosphere. Short-exposure photo- 

 graphs of parts of the sun's surface taken on a large scale at intervals 

 of less than a minute by S. Chevalier - show that the granulation of the 

 photosphere is undergoing very rapid change, and we may attribute this 

 to the rapid radial currents which exist throughout the entire atmos- 

 phere of the sun. The spots, faculse and prominences, which may, be 

 Tegarded as accentuated developments of the general currents, change 

 continuously and frequently exhibit great velocities in their radial and 

 transversal movements. The number and areas of these disturbed 

 regions vary in a period of average length about 11. "3 years, and in about 

 the same period the regions in which the spots are most abundant change 

 in latitude in the north and south hemispheres. It is not yet known 

 whether the convection currents over the entire surface of the sun vary 

 periodically, yet at least the enlarged convections or their results — if we 

 may so term the spots, facuLT and prominences — undergo periodic varia- 



1 Abstract from a paper given at a meeting of the Royal Astronomical Society 

 of Canada in Ottawa, March 2.5, 1909. 



2 Astrophysical Journal. 27. 12-24. 1908. 



