360 Mr. F. Y. Edgeworth on the 
The opinion that the great mass of the ice on the Antarctic 
continent and also on Greenland lies near to the outer edge, 
and that it gradually diminishes inwards till at last it disap- 
pears, is evidently one based on a misapprehension as to the 
physical conditions of continental ice. I cannot help believing 
that had Professor Nordenskjold duly reflected on the neces- 
sary physical and mechanical conditions of the problem which 
he is endeavouring to solve, he would not have undertaken 
the journey across the Greenland ice. 
Note, September 22nd. — Baron Nordenskjold has just re- 
turned, and, as might have been expected, he found the 
interior of Greenland a desert of ice. with the icy plain gra- 
dually- sloping upwards to the interior. The ice rose at the 
furthest spot reached (280 miles from the coast) to 7000 feet 
above the sea, and was still seen to rise to the east. The 
results of the expedition are, however, of the most important 
character, confirming, as they do, the true theory of conti- 
nental ice. 
L. The Method of Least Squares. By F. Y. Edgeworth, 
31. A., Lecturer on Logic at King's College. London*. 
I. 
ri^HE Law of Error and the Method of Least Squares do 
J- not traverse the same ground ; and the direction of the 
one course of reasoning is inverse to that of the other. In the 
former we derive the formation of a sourcef of error from the 
confluence of an indefinite number of small tributaries. In 
the latter we start from a position lower down on the stream 
of causation, from observations resulting from a source of 
error, and reason up from given observations (accompanied 
with some knowledge of the source of error from which they 
result) to (a more complete knowledge of) the source of error, 
the facility-curve under which the observations range them- 
selves. For example: — Given .%\, ,r 2 , &c. observations diver- 
ging according to a probability-curve of known modulus, but 
unknown centre, to determine the centre. These distinctions 
are likely to recommend themselves to those who have studied 
Mr. Glaisher's article and other first-rate authorities. But 
there is another distinction, more interesting to the philo- 
* Communicated by the Author. 
t According to the received view, one particular source — the probability- 
curve ; according to our views, a great variety of facility-curves (see pre- 
vious paper). 
