264 
2. On the Fallacy of the Present Mode of Estimating the 
Mean Temperature in England. By James Stark, M.Eh 
In the Meteorological Tables for England, published at the 
public expense, viz., those of Greenwich Observatory, and those of 
the fifty- five meteorological stations in England, appended to the 
Registrar- General’s Quarterly Reports, the mean temperature as 
given is not the true mean, but is an estimated mean, deduced 
from the united observations made with the self-registering and 
dry-bulb thermometers, the exact means of these instruments 
being first altered by certain tables, with the view of correcting 
them for diurnal and monthly range. The author objected to this 
mode of estimating the mean temperature, because the tables used 
for the purpose of correction were drawn up from insufficient data, 
and because the principle was bad of having observations made by 
one instrument, whose readings were liable to constant errors, made 
the basis for the correction of observations made by a more trust- 
worthy instrument. 
The tables for the correction of diurnal range used in England 
were drawn up from a five years’ series of observations ; but by a 
reference to the Scottish observations, it was attempted to be shown 
that a five years’ period was far too short to elicit even a true mean 
annual temperature ; while, by a reference to the mean temperature 
of corresponding months during five consecutive years, it was found 
that such a period gave still more incorrect results for the months. 
It was then demonstrated, by a reference to the Greenwich 
monthly and daily observations, that the readings of the dry-bulb 
thermometer, at every two-hourly period, failed to give the true 
mean temperature either of the day or of the month, as they 
always missed the period of the day when the maximum tem- 
perature was attained, which, as indicated by the self-registering 
thermometer, was from 1° to 5° higher than any of the readings 
taken at the two-hourly periods. The consequence was, that the 
mean temperature deduced from the dry-bulb two-hourly readings 
was always below the truth, and it was impossible from them to 
ascertain at what period of the day the highest temperature was 
attained, so that the chief elements for the calculation of the amount 
of diurnal range were wanting. 
The data, however, which were used for the formation of tables 
