74 ROYAL SOCIETY OF CANADA 
Much work has already been accomplished in this direction as the result 
of researches conducted elsewhere by Prof. Penhallow, who is now making 
a series of observations of a similar nature here. In this connection the 
importance of a continuous temperature record throughout the year, and 
particularly at the beginning and close of the vegetation period, should 
not be overlooked. 
CONCLUSION. 
We may conclude from the foregoing preliminary reduction of the 
observations that results of considerable meteorological and botanical 
interest might be readily obtained by the adoption of a similar method 
at many different localities and for longer periods of time. It would 
probably not be necessary to make the apparatus so elaborate and accu- 
rate as that in use at the McDonald Physics Building. It is possible to 
make a form of indicator reading to a tenth of a degree F., with which 
observations may be taken with greater ease and quickness than with 
mercury thermometers, and with at least equal accuracy. The same 
thermometers, in a slightly more sensitive form, reading with the same 
indicator, might also be readily applied to ordinary meteorological pur- 
poses. They are, for instance, particularly suitable for ohserving the 
temperature of the air at considerable heights above the ground, from 
which the initial rate of change of temperature with height, a factor of 
considerable importance, might be obtained. 
The form of apparatus which it is proposed to adopt for these and simi- 
lar purposes is here shown in action to illustrate the facilities which the 
method affords for the observation of temperatures with thermometers 
located in otherwise inaccessible situations. The cost of the thermometers 
is not greater than that of equally accurate mercurial instruments. The 
same indicator can be used with any number of thermometers, being 
placed in any convenient central situation in the observatory. The dis- 
tance at which the thermometers are placed is a matter of no consequence, 
except in respect of the cost of the leading wires to the indicator. 
