722 Reviews. [Oct., 



But there is abundant reason independently of this for hesitation before 

 admitting Mr. Proctor's premises. Thus micrometric measurements 

 are liable to so many sources of uncertainty, that scarcely any two 

 observers agree in giving the same result for the same object. Com- 

 paratively little weight can be attached to the breadth of the black 

 division, in consequence of the great difficulty and uncertainty of the 

 contacts of the points of reference with its borders. Other circum- 

 stances besides these show strongly how such values may be affected. 

 The value of the semi-diameter of the ball, for instance, made when the 

 rings are visible, is sensibly smaller than those made during its dis- 

 appearance, a circumstance arising probably from the glare of the rings 

 preventing the true border of the planet from being seen. Moreover, 

 the following curious observation was made by Mr. Main, who fully 

 investigated this subject some few years since, viz. that in two series of 

 measurements which he made of the globe and rings in two distinct 

 years, several of the first measures of each year made the breadth of 

 the rings conspicuously less than that of the dark interval, that the 

 difference between the two steadily decreased till they were equal, and 

 that then the breadth of the ring became greater and that the ratio 

 between them steadily increased. Now as it will not be denied that 

 observations made continuously by the same person, with the same 

 instrument, in the same way (supposing observer, instrument, and 

 method to be of the best kind, which will be readily granted in the 

 present case), are far more reliable than those which, being made by 

 different persons, instruments, and methods, are subsequently reduced 

 to a uniform standard by a presumed allowance for all variations, 

 the legitimate conclusion from the above would seem to be that 

 Saturn's ring system underwent, during each of the two years in 

 which these examinations were made (and, if these are to be taken as 

 particular instances of a general case, is continually undergoing), 

 periodical changes of very great magnitude. Yet there are very few 

 who, considering the uncertainties to which such measures are liable, 

 would not hesitate to draw an inference of this kind, and who would 

 not rather suspect the changes to be due to optical causes. But, as 

 we have said, these results greatly affect the importance of a theory 

 on the nature of the rings advanced by Mr. Proctor, and it is 

 therefore probable that he has allowed his mind to be unduly biassed 

 in the direction that would give it support. What are the rings ? 

 The chapter devoted to the discussion of this question is a highly 

 interesting one, and the conclusion arrived at, after a review of the 

 various conjectures that have been entertained on the subject, is 

 that " they are composed of flights of disconnected satellites so small 

 and so closely packed, that at the immense distance to which Saturn 

 is removed, they seem to form a continuous mass." This hypothesis, 

 which was first advanced by Cassini, may possibly be as true as it is 

 ingenious, and the most recently expressed views in cosmical philo- 

 sophy, if they do not support, certainly cannot be said to contradict 

 it. This, however, is not saying much. A hypothesis so strange 

 requires very strong evidence to render it acceptable, and we shall 

 here inquire how far Mr. Proctor has proved his case, and whether the 



