NEBULAE, CLUSTERS, ETC. 33 



Herschel, W. : Abstracts of Memoirs — Continued. 



A.D. Vol. P. 



particulars are readily explained by admitting them to 

 be more distant. This argument, however, does not 

 extend so far as to exclude a real difference which tbxro 

 may be in different clusters, not only in the size, but also 

 in the number and arrangement of the stars. 



XVIII. Of still more distant globular clusters of stars. 

 1814 104 280 It has frequently happened that I saw three objects in 



succession, the first of which was a brilliant globular 

 cluster of stars, the second a miniature of the former in 

 which the stars could just be perceived, and the third in 

 every respect a similar miniature of the second, as the 

 second was of the first, but in which the stars, though 

 suspected, were no longer to be distinguished. Five such 

 objects given, all in the milky way. 



XIX. Of a recurrence of the ambiguous limit of observa- 

 tion. 



2S1 It has already been shown [VIII] that in passing from 

 faint nebulosity to the suspected siderial condition wo 

 cannot avoid meeting with ambiguous objects, and the 

 same critical situation will again occur, when, from the 

 distinctly siderial appearance, we endeavor to penetrate 

 gradually further into space. 



The effects of clustering power have been gradually traced 

 from the first indication of clustering stars through ir- 

 regular as well as through more artificially arranged 

 clusters up to the beautiful globular form. 



The extended views I have taken in this and in my former 

 papers of the various parts that enter into the construc- 

 tion of the heavens have prepared the way for a final 

 investigation of the universal arrangement of all these 

 celestial bodies in space. Tne scale is still wanting by 

 which distances are to bo measured. 



282 XX. Of the breaking vp of the Milky Way. 



Its whitish tinge has been proved by star-gauges to arise 

 from accumulated stars. It does not now consist of 

 equally scattered stars. 



283 One hundred and fifty-seven instances have been given of 



clusters situated within the milky way. Sixty-eight 

 more are in the borders. Now, since the stars of the 

 milky way are permanently exposed to the action of a 

 power whereby they are irresistably drawn into groups, 

 we may be certain that from mere clustering stars Ihcy 

 will be gradually compressed through successive stages 

 of accumulation till they come up to what may be called 



