310 CARNEGIE INSTITUTION OF WASHINGTON. 



ever the agency — to give rise to these juvenile forms affecting so many 

 of the highest order of known organizations. The immediate problem 

 in hand, the rehabihtation of the Herschelhan hypothesis by such help 

 as dynamic encounter may offer in such deployment, raises the special 

 question whether appropriate encounters between galactic systems 

 can occur often enough to maintain, in proper proportion to the whole 

 number of aggregations susceptible of such interaction, the observed 

 number of spiral forms. 



Before emphasizing the unpromising aspect of the question put in 

 this form, it is worth while to consider the subconscious concepts likely 

 to lurk beneath such a question and to vitiate its answer. An analo- 

 gous question has been raised relative to the adequacy of close approach 

 of stars within our galactic system to give the observed supply of neb- 

 ulae assigned to close approach. 



(a) The immense extension of knowledge of celestial space, forced by 

 progressive observation, has also forced the conclusion that collisions 

 and close approaches are far apart. No equally imperative force has 

 deploj^ed our concepts of time in like degree. On the contrary, the old 

 speed-concepts of evolution have largely hngered and have introduced 

 incongiiiities into current concepts of time in its relations to concepts 

 based on space. As a matter of fact, consistent concepts of evolution — - 

 especially adequate concepts of the evolution of great star organizations, 

 "universes" in our inherited phraseology — clearly require enlargements 

 of our working concepts of the lifetime of clusters, of stars, and of 

 nebulae to a degree comparable to the enlarged concepts of space forced 

 by observation. Such a parity of concepts, when realized, removes 

 almost the whole cogency of the challenge of adequate frequency, so 

 far as planetary genesis is concerned. If there are 150,000 nebulae 

 in our stellar galaxy and 150,000,000 suns susceptible of becoming 

 producers of nebulae by close approach or collision, it is necessary only 

 that a nebula should live one-thousandth of the generation interval 

 to keep the supply of nebulae up to the present figure. The question 

 of adequacy hangs as much on the longevity of the nebulae as on the 

 rarity of close approach. An analogous revision of time concepts gives 

 some relief at least to the much more formidable challenge of the 

 adequacy of close approach or penetration to give a spiral state to the 

 hypothetical star systems supposed to masquerade as white nebulae. 

 The high velocities, the powerful attractions, and the vast expanses 

 assigned the star-clusters under this view, favor frequency of inter- 

 action, when frequency is measured in appropriate terms based on 

 the hfe-period of a disturbed galactic state. 



(6) If our galaxy has recently been disturbed, let us say by the 

 penetration of the Magellanic cluster, the assumption that the present 

 is a normal state in respect to the number of nebulae is scarcely war- 

 ranted. The galaxy is presumably passing from its disturbed state 



