NATURAL HISTORY AND PHYSIOORAPHY OF NEW RRUNSWICK. 453 
deviation from coincidence will afford a measure of the variation in 
weather conditions between base-station and place of observation. 
Local weather-changes are thus brought out with great clearness. 
One can therefore eliminate all readings showing a marked deviation 
from this coincidence and retain only those in which the base-station 
reading gives a correct index of the weather changes. This method 
is, of course, only applicable where several measurements are made of 
the same locality, and in such cases I have applied it in the calcula- 
tions yielding the results below, selecting from my total number of 
readings only those which are thus shown to be the best. 
Method. — The readings were in nearly every case made at the 
exact minute when the barometer was being read at the Fredericton 
station, and the temperature was recorded at the same time. The 
results have been worked out by Airy’s tables and with his formula 
for allowance for temperature. In the case of the readings compared 
with Chatham, however, owing to a misunderstanding of the time at 
which they were to be taken, our readings are some minutes earlier 
(Chatham taken at 7.50, 2.50 and 7.50, ours at 7.24, 2.24. and 7.24, 
standard), and hence they are, theoretically, less accurate than the 
series compared with Fredericton, though the error would be very 
slight, especially where several readings are averaged. 
Probable Accuracy. — The instruments and methods used, and 
the care exercised in all observations and calculations are such as to 
make me feel confident that the following measurements are as accurate 
as aneroid measurements can be made with Fredericton and Chatham 
as base-stations. 
Results. — The following figures express heights above mean-tide 
level at St. John. Unless otherwise specified, the places have not 
been measured before. 
Trotvsers Lake. Mean of two measurements, 1,286 feet. Chalmers 
gives (Summary Report for 1900) 1,350, and Mclnnes (Geological 
Map) 1,360 feet, both, in my opinion, impossibly high. Concerning 
my own lower result of last year (viz.: 1,243) some observations will 
be given below. 
Long Lake . — A single measurement, 1,302 feet, probably too high. 
Mr. Chalmers makes Long 30 feet lower than Trowsers, which seems 
to me incorrect. Mclnnes made it 10 feet higher, and I made it 13- 
feet higher last year, and 16 feet higher this year. 
