Jan. 1, 1886.] 



♦ KNOWLEDGE ♦ 



81 



But, 5. All the classes of nebulse are brought together 

 again by the discovery th?,t theiv gaseity is simply a 

 question of proportionate decree. 



6. All the classes (includintj; the irreguUr class) are 

 bruught together in the Magellanic Clouds. 



The main inference from this little set of facts, is that 

 tlie gaseous matter so common in the nebula; exists in all 

 part.s of the galactic system, and serves as a sort of index 

 of the oneness of the siderei'J and nebular systems. 



The second omitted fact is the discovery made in 

 recent times of the singular complexity of the solar 

 system. Forced to take the solar system with which we 

 suppose ourselves familiar as presenting in a general way 

 ii tj-peof the wider system of which it is'a part, we 

 clearly are led to form different views now than analogy 

 suggested when as yet astronomers knew so little of our 

 system. The di.-covery of the asteroids lias done much 

 towards this change of view ; but I think the knowlcdsre 

 recently acquired respecting meteor-systems tends much 

 more importantly to give us just conceptions of the rich- 

 ness and complexity of our solar scheme. I do not so 

 much refer to the strange discoveries recently made 

 respecting the orbits of the meteors, the association of 

 these bodies with comets, and so on ; though tlfe evidence 

 on all these points seems too definite and complete to 

 leave any room for (juestion respecting them. The point 

 to which Prof. Herschel has called so much attention— I 

 mean the large number of meteoric sj'stems which our 

 earth traverses— and the consequent argument from 

 probability that there exist millions of such svstems 

 within the solar .scheme— seems to me far more important. 

 It teaches us to look for an enormous wealth of relatively 

 minute bodies in other systems, and therefore prepares 

 us to look on the "suns" in the sidereal scheme as 

 relatively few, the minute orbs and the groujis cl' minuter 

 crbs as relatively numerous. 



extinction of light. It seems to me to have escaped 

 notice that the arguments in favour of the extinction of 

 light and those against it are equally irresistible on the 

 acce])ted theories of stellar distribution.* The argument 

 on which Struve founded his formula — brightness of a 



star= / (O-990G.51) ^'^'^*^'' ^)— has never, I believe, 

 (dist )' 



been disposed of ; though arguments of equal force have 



disposed effectually of liis extinction theory. Now I 



think that when irresistible arguments can be urged, on 



a given hypothesis, both for and against a certain theory, 



we may reasonably assume that there is something wrong 



about the given hypothesis. 



Ton disposed of Struve so completely, for instance, 

 that I believe every one has since looked on the theory of 

 extinction r.s exploded : yet it was rather by arraying 

 stronger arguments than his than by destroying' the 

 force of whi-.t he had put forwr.rd that this was^done. 

 Ton put the matter directly on this footing ; but others 

 who ha-\-e quoted the result have quite forgotten that a 

 difficulty T\.",s admittedly left unaccounted for. (I am 

 not referring to Struve's misinterpretation of some words 

 of your father's, but to his argument dr.-.wu from the 

 insufficient number of faint stars.) 



According to the views I have been led to form the 

 (luestion of the extinction of light seems an open one — 

 but several facts seem to suggest that there is appreciable 

 extinction even within such a distance r.s sep.ir.-^tes a 

 Centauri from us. 



I think on a reconsideration of my views you will see 

 that (according to them) increase of distance would not 

 necessarily lead to nebulous light. It seems to me that 

 neither does tlie presence of irresolvable nebulous light 

 necessarily indicate extreme distance, nor the converse. 

 If we liave a certain group of stars, and that group be 

 supposed to move continually away, the question whether 

 it will ever become nebulous (with ang power if there be 

 no extinction, or with a given power if there be extinc- 

 tion) depends wliolly on the relation between the size of 



Fig. 1. — A star-group. 



Fig, 2. -Imagined-biu i.npossible shapo of a star duster like that in Fi- 



Your father had no such analogy to guide him ; but 

 he was, I believe, so steadily progressing towards a 

 change of view respecting stellar distribution (I judge 

 from some of his later papers) that I think he mio-ht soon 

 have been able to reverse the analogy, and -iii/vr the 

 existence of multitudes of minute bodies within tlie solar 

 system from the analogy which the sidereal scheme 

 presents. 



Lastly 1 should have referred to the question of the 



association with the Milky Wav, we get the following three 

 columns : — 



1- H. III. 



Clusters. Planetary and Frrc- Clusters. 



gular Xebuhe. 

 Irresolvable Nebuhe. Irresolvable Nebuhu. I'lanetary and Irre- 

 gular Xebuls. 

 rianetary and Irre- Clusters. Irresolvable Xebuls. 



gular Xebulje. 

 Mr. Muggins's discovery (confirmed by Lieut. Herschels observa- 

 tions) that the gaseous nebuh-e show a faint continuous spectrum 

 as well as the bright-line one ; and Lieut. Herschel's discovery of the 

 converse fact that stellar nebula; show a bright-line spectrum as well 

 as the continuous one. make tables I. and U. highly signiticant. 



the componen^stars and the distanct-s which sejiaivte 

 them. If the two stars A and B, in the cluster shown in 

 Fig. 1, i-.re of such size and so distant that when just dis- 

 appearmg to the naked eye they aiv clear of each other 

 in ajipearane:', they will be equally .so when just dis- 

 appearing for any power whatever, unless there is extinc- 

 tion. 



We know there are ji-rts of the iMilky AVj'.y where 

 there is irresolvable nebulosity : and the question v.t once 

 suggests itself. Have we any evidence whether this is to 

 be looked upon es a proof of indefinitely vast extension 

 of the galaxy in direction of the nebulosity 1 It seems to 

 me that we have positive proof that this is not the case. 

 I will take the c;-,-e of the clustering aguregation in 

 Perseus. Your father's recount of this spot shows th<;t 



When I speak of accepted views I refer to those which are con- 

 tinually described m treatises on astronomy as the direct fruits of 

 your father's researches and your own : I know well that there is 

 not a single hne either of your own or your father's writing in which 

 the question of stellar distribution is spoken of as one on which 

 wc a'c ill a pusiliju lu forni a dctinitc tliu-jry. 



11 



