F. W. Very— A Cosmic Cycle. 105 



comparatively early stage of condensation, before the loss of 

 peculiarly unstable substances. 



Out of a total of 111 Orion stars (Groups I to Y), Miss 

 Maury gives 36 as of Div. h, or having hazy lines implying 

 rapid axial rotation. The Sirian stars (Groups YII to XI) 

 show 46 stars in Diy. h, out of a total of 185 stars. * None of 

 this division occur among the solar stars. If the Orion stars 

 are much more expanded than the Sirian, an equal superficial 

 velocity will not be sufficient to give centrifugal separation at 

 the earlier stage. The Sirian stars are more highly condensed 

 — witness the dense hydrogen atmospheres — and hence are the 

 most likely to giye birth to planets by rotary velocity. 



That some of the planetary births from stars as late in stellar 

 development as the hydrogen-iron type, may occur by centri- 

 fugal expulsion, appears certain. Lockyer obtains for a 

 Aquilae a minimum equatorial surface velocity "of about 45 

 miles per second supposing that the axis is perpendicular to 

 the line of sight. "f Considering that this is 34 times as great 

 as the present solar surface rotation, it would not be surprising 

 if some of the hazy -lined stars should furnish us with a new 

 type of novae. That rapid rotation ceases in the solar stage 

 implies that rotational moment has then been destroyed by a 

 previous expulsion of planets. 



At present our knowledge of systems in rapid revolution, 

 and therefore presumably young, is confined to Algol variables 

 and spectroscopic binaries. Professor Pickering says: "It is 

 noteworthy that all known variable stars of the Algol class 

 have spectra of the first type, except perhaps P Canis Majoris, 

 whose spectrum is either of the first or second type.":); 

 U Ophiuchi, X Tduri, and ft Persei are helium stars, and the 

 low average density of the Algol variables, from J to ^ that of 

 the sun,§ points to an early stage of condensation as normal in 

 stars of the Orion type. It must also be presumed from the 

 short period of the Algol variables that the separation of the 

 components can be but little beyond Poche's limit, and conse- 

 quently that these bodies are at a very early stage of stellar 

 evolution. Since the eclipsing body is invisible, or at best a 

 red star (U Cephei), there must be either a great diversity of 

 composition or an entirely different life-history for the com- 

 ponents of an Algol variable. This diversity does indeed 

 suggest a capture theory, but the small eccentricity of the 

 Algol-type orbits is decisive against such a supposition. The 

 eclipsing body cannot be a meteor-swarm, or a companion dis- 

 rupted by tidal strain, for only a single body of regular figure 



*Miss Antonia C. Maury, Ann. Harvard Coll. Obs., vol. xxviii. pt. 1, p. 10. 

 f J. Xorman Lockyer, Proc. R. Soc. London, vol. lxvi, p. 237, 1900. 

 % E. C. Pickering, ith Annual Report of the Photographic Study of Stellar 

 Spectra. Hear y Draper Memorial, p. 5, 1890. 



S, H. X. Russell, Astrophysical Journal vol. x, p. 317. 1899. 



