1894.] Niagara Falls as a Chronometer of Geological Time. 145 



it constituted iu my own experiments the chief and most striking 

 phenomenon : and it was to the colours which appeared during the 

 bright phase that my attention was exclusively directed, the tints of 

 the relatively insignificant " luminous trails " being too faint to be 

 distinguish able . 



It is clear that the momentary excessive brightness of the positive 

 image is no less essential than the dark interval (or negative after- 

 image) for the generation of the phenomenon of recurrent vision 

 which forms the subject of the present paper. 



III. " Niagara Falls as a Chronometer of Geological Time." By 

 J. W. SPENCER, Ph.D. Communicated by Professor T. G. 

 BONNET, F.R.S. Received March 16, 1894. 



(Abstract.) 



1. Conjectures as to the Age of Niagara Falls. Prior to the writing 

 of the present paper, most of the conjectures as to the age of the 

 Falls have been based simply upon the supposed uniform rate of 

 recession. Thus, in 1790, Andrew Ellicott assigned 55,000 years as 

 the age of the Falls. In 1841, Sir Charles Lyell allowed 35,000 years ; 

 in 1886, Professor B. S. Woodward, after three surveys had been 

 made, calculated the age as 12,000 years ; and later, Mr. G. K. Gilbert, 

 supposing the recession to progress at' the maximum axial retreat 

 lone, reduced the age of the Falls 6,000 years. This latter was not 

 itended as an estimate, as he fully recognised that such a time must 

 lave been greatly lengthened by many changing conditions. The rate 

 iopted by the first two writers was only conjectural, as no surveys 

 id then been made. Three surveys had been completed before the 

 writings of the latter two writers, and I have had the benefit of a 

 fourth. Woodward's calculation was upon the mean mathematical 

 ilargement of the Horseshoe gulf at the end of the chasm, which 

 ite was less than the geological rate of retreat. The author's 

 method differs from the others in that it takes into consideration the 

 rate of recession throughout the changing episodes of the river, 

 which have been entirely discovered by Gilbert or himself. His com- 

 putations make the age surprisingly near to the conjecture of Lyell. 



2. Modern Topography. This section of the paper gives such 

 etails as bear upon the subject, some of which do not appear else- 

 where. 



3. Geology of the District. Besides what may be found in other 

 forks, there are several measured sections and descriptions showing 

 le amount of work the river had to do. Several figures illustrate 



the varying conditions. 

 VOL. LVI. L 



