Fuly 9,.1885] 
Mammont; on February 15 a letter was read from 
Benzelius, stating that Kagg had received the drawing 
from a Capt. Tabbert, and that he could give no informa- 
tion as to its correctness. Again, on October 3, Benzelius 
exhibited a large bone, almost petrified, which was the 
jaw of a Mammont, or as it was called Behemoth, received 
from Tobolsk in Siberia, through Capt. Clodt von Jurgens- 
burg, and, on November 22, Benzelius exhibited “ part of 
the tusk of a Behemoth, which was exactly like ivory.” 
Finally, Benzelius communicated with the Russian Chief 
of Mines, Tatischew, who, in a letter dated May 12, 1725, 
had given long and important information of the history 
of the mammoth. This letter is printed in “ Acta Literaria 
Sueciz ” (vol. ii. p. 36, 1725). 
A. E. NORDENSKIOLD 
NIAGARA FAILS: THE RATE AT WHICH 
THEY RECEDE SOUTHWARDS 
| 52 diagrams are from the map issued by the New 
York Commission for the establishing a State 
reservation at the Falls, based on surveys made in 
August and September, 1883, by Thomas Evershed, 
under direction of Silas Seymour, State Engineer and 
Surveyor. The scale of the diagrams is one half that of 
TRUE NORTH 
rN 
the map, which is on a scale of four chains to the inch. 
To have given all on one diagram with the intervening 
Goat Island would take up nearly an entire page of 
NATURE, and if the scale were smaller it would fail to 
show clearly the distinctive features of the changes in 
progress. Fig. 1 shows the Canadian or Horse-Shoe 
Fall, Fig. 2 the Eastern or so-called “ American” Fall— 
a misnomer too deeply rooted in usage to be now sup- 
planted by some more fitting name. 
The rate at which the Falls are receding has been a 
matter of interest to geologists for over fifty years, but 
the results so far reached have been conflicting and in- 
conclusive. The manner in which the Falls work back- 
ward, undermining their brink, is so well known from 
Lyell’s clear description, that I shall not repeat it. 
In 1830, Bakewell, on the basis of such information as 
he could gather from old inhabitants and from his own 
observations, concluded that during the previous forty 
years the Falls had receded at the rate of three feet per 
annum. 
NATURE 
229 
Lyell, from such materials as he could obtain during 
his own visit in 1841 and 1842, estimated the annual 
retrograde motion at only a foot. It is sufficient to recite 
such discordant results arrived at by two. careful investig- 
ators to show how imperfect were the materials at their 
disposal, nor will any one who has been on the spot 
wonder at their differing so greatly. It would be possible 
to roughly compute the southward movement of the 
innermost recess of the Canadian Fall by referring its 
position from time to time to some fixed points on the 
adjoining shore, but any conclusive determination of the 
movement of the entire Fall could not be obtained in this 
way. The map referred to gives the outline of the Falls 
as determined by three surveys: the New York Geo- 
logical Survey of 1842, the U.S. Lake Survey of 1875, 
and Evershed’s Survey of 1883. The contours of the brink 
as established by these enable us to measure the total 
movement. 
I divide the contour from 8 to Goat Island into thirty- 
three sections, disregarding for obvious reasons the over- 
flow north of 8, on the Canadian shore. From § to « are 
eleven sections, from « to ¢ are twelve sections, from ¢ to 
Goat Island are ten sections. It is obvious that much 
the greater work has been done between @ and ¢,and that 
the innermost recess has kept in the same relative posi- 
tion. 
SCALE OF FEET 
° ._200 400 
528 FT. TO THE INCH 
Fig.2. 
Vooe Z 8 
mil ° © Auli 
fy ie LUNA ISLET 
Mf GoatY N 
AL ASLAND % 
y 
The means of the measurements on the sections, along 
perpendiculars from the contour at the date of each 
survey, measured on a tracing of the published map, give 
the following results for the Canadian Fall :— 
33 years 8 years 41 years 
endingin endingin ending in 
1875 1883 1883 
_ ft. ft. ft. 
Mean aggregate recession along con- 
tour of 2000 feet, from 8 to Goat 
island = 3000 sites. assy ie ccuee st CO mana 114 
Mean aggregate recession along con- 
tour of 1200 feet, Bto€=... ... — 60 — 
Mean annual rate of regression along 
the whole contour where a visible 
change was effected=... .. ... 2% ... eee 23 
Total maximum regression at the in- 
mermost recesS =...  ... «2. .5 
Annual rate of maximum regression = 
The “ American” Fall, measured in ten sections, gave 
a total mean recession of 374 feet in the 41 years ending 
in 1883, which is at the rate of about 10 inches per 
annum. 
I do not know that I have seen any estimate attempted 
of the relative volumes of water passing over the two 
falls. From such imperfect data as I have, referring to 
depth and swiftness, 1 should think that the rate of 
erosion for each fall gave some approximation to the 
