F. W. Very — A Cosmic Cycle. 113 



In favor of this view, I note that there has never been a 

 recurrence of a nova. The blaze is final. These stars all 

 appear in the Milky Way, the locus of the youngest or explo- 

 sive stars. 



The phenomenon in its successive phases has somewhat the 

 same sequence as a volcanic eruption. There are preliminary 

 symptoms leading up to a climax. The violent commotion at 

 the center of the star at first produces nothing but increased 

 superficial circulation, exposing fresh heated surfaces of great 

 brilliancy, but as yet with little accompanying outburst of gas, 

 and with slight velocity in the line of sight. By the time the 

 eruption has become fairly started, the inertia of the outer 

 layers has been so far overcome that rapidly moving streams of 

 gas break through to the surface, and carrying everything 

 before them, launch out into space. For a short time the tur- 

 moil is tremendous. Yelocities of every degree from excessive 

 positive to extreme negative are met, but tending to particular 

 values along certain streams where the chief part of the motion 

 resides. Some portions of matter, having lost their velocity 

 through friction and interference, are left by the way and » 

 expand to comparative coolness, giving fine absorption lines. 



The broad paired dark and bright bands, in the order of 

 intensity, belong to hydrogen, calcium, helium, the substance 

 which gives the bright blue band in the spectrum of the Wolf- 

 Rayet .stars, magnesium, sodium, nebulum, and others of less 

 intensity. It is essentially a chromospheric spectrum, passing by 

 successive repressions into that of a nebula. The original 

 motion dies out by being frittered away through friction. 



Antecedent Stages of Stellar Growth. 

 The invisible preparation for a new star is probably of exces- 

 sively long duration, involving the gradual growth of meteorites 

 from innumerable centers of crystalline aggregation, the slow 

 beginnings of movement under gravitation, and at last the 

 clustering of meteor-swarms ; but the bursting forth into full 

 stellar life is comparatively sudden, although possibly not 

 attained without more than one fluctuation. The link between 

 nebulae and novse is demonstrated. As just stated, we have no 

 example of the recurrence of a nova. It is hardly possible 

 that vast interpenetrating meteor-swarms should pass through 

 each other, attaining stellar temperatures by collisions, without 

 experiencing a change of motion, perhaps becoming associated 

 in orbital connection, in which case the collisions might be 

 expected to recur with increasing frequency until the nova 

 became a permanent star ; but nothing of the sort takes place. 

 The speed of the meteors being very great, the temperatures 

 Am. Jour. Sci. — Fourth Series, Yol. XIII, No. 74. — February, 1902. 



