382 JPROFESSOR C. PIAZZI SMYTH ON THE 



our exact knowledge of sidereal phenomena. It may therefore be of some 

 preliminary assistance, if I call attention to a few of the more noticeable featm-es 

 which have struck me, after having thus brought together the several authorities. 

 Beginning then with the columns on the left-hand side of the page, it may be 

 remarked, that — 



In Magnitude^ these stars show no very sensible alterations within the period 

 over which all the observations extend. 



In Colour^ a slight gradual change in both components may be suspected. 

 In Position and Distance, there is something to discuss. Looking up and 

 down these columns, the reader may first be struck by the greater differences, — 

 and, evidently by their + and — character, greater errors of observation, — among 

 the earlier than the later observers. This is much as might be expected in the 

 nature of things, and is not mentioned invidiously, but only as a point upon which 

 a practical judgment must be formed, and a case where w^e should try to avoid being 

 led away impulsively by the splendour attached to great names. Had the two 

 earliest sets of observations, by " H'" and " o-,"" been the only ones in existence, a 

 rapid change of the stars in both position and distance would have been implied, 

 and believed in to this day. But subsequent observations soon impressed both 

 "2" (the elder Struve at Dorpat), and my father in his " Cycle," that so far as 

 Position was concerned, no sensible change was going on ; and this view is 

 remarkably borne out by the Elchies observation, at an interval of twenty-eight 

 years after a mean of four of the best sets of preceding observers. Indeed, it 

 differs therefrom by only 1' ; a pure accident in an instance where the third part 

 of a degree would have been excusable ; and a coincidence certainly, wdiich I had 

 no idea of, until after the authorities had been collected, and that was very long 

 after the observations were made. 



In Distance^ my father thought the stars constant, but Struve (pp. 137 and 281 

 of his great Catalogue), after having rejected the H^ observation as erroneous, 

 thinks that a slow increase is going on, especially w^hen taking account of con- 

 stant errors ; which, being corrected for in earlier observations, w^ould make his 

 measure of 182I-93 = ir-21, and that of H' and S in 1821-91 = 10"-58, leaving a 

 large increase for his subsequent observations of 1832 and 1836. The Elchies 

 observation however again shows, and by an almost marvellous agreement, no 

 increase to have taken place in the last twenty-eight years. 



May we trust this result? I think w^e may; but am not so much guided 

 thereto by the Elchies result, to which the attributed "weight" is exceedingly 

 small, as by some considerations on the older measures. All the world knows 

 the elder Struve as a first-rate observer, and when he observes with the large 

 Dorpat telescope of 9 inches aperture, no one can come near him. This class of 

 his observations is marked with a 2, and the record of 1832 is the mean of no less 

 than seven sets, in five different years ; something of this sort too being gene- 



