180 PROCEEDINGS OF THE AMERICAN ACADEMY 



indentation of the shadow at x, in that part where it is projected on the 

 outer border of the ring C, is perfectly explained on the supposition 

 that this part of the ring is on a higher level. The same shadow, 

 as it appeared projected on the rings B and A, also clearly indicates that 

 the plane of these zones is on a lower level. 



In order to find the shape of the surface of the rings from the obser- 

 vation of the form of the shadow thrown by the planet, I have experi- 

 mented on a miniature representation of Saturn, illuminated by a lamp 

 occupying the position of the sun, while my eye occupied the position 

 of the earth. By successive trials in altering the shape of the minia- 

 ture rings, I have soon found what must be the form of the rings in 

 order to give to the shadow the same appearance which had been 

 observed on the planet ; and the result agrees with the explanation 

 already given. 



From the form of the shadow as it has appeared at different times 

 during the last four years, and from the experiment just mentioned, it 

 seems pretty clear to me, that, from the inner margin of the dusky ring 

 jP, the thickness gradually increases until it reaches the extreme border 

 of the ring C, where it gently decreases, as indicated by the rounding 

 of the shadow at this point ; after which it sinks perpendicularly down, 

 until it comes even with the general level of the rings B and A. The 

 slightly curved appearance of the shadow of the planet during the 

 present year, with its concavity turned towards its globe, also supports 

 this hypothesis. 



Though, in general, the level of the ring C is always higher than 

 that of the rest of the system, it does not seem, however, to be uni- 

 form and permanent, but varies, either by the rotation of the rings 

 upon an axis, or by some local changes in the cloud-forms themselves ; 

 as in several instances I have observed quite rapid and striking changes 

 taking place during the course of one evening in the indentation of the 

 shadow shown at x. Fig. 1. Sometimes the indentation appeared to 

 increase, indicating a higher level ; and sometimes to decrease, indicat- 

 ing a lower level. 



That the thickness of the rings is increasing from the interior margin 

 of the dusky ring to the outer border of tlie bright ring C, seems to be 

 corroborated by the phenomena which I have observed on the dusky 

 ring, and of which I shall speak presently. 



On all favorable occasions, I have made careful searches on the dusky 

 ring for the divisions suspected by Bond ; but I never had the faintest 

 glimpses of them. The dnsky ring appears to me to be continuous, 

 though it is certainly not of the same thickness throughout. Whatever 



