86 REPORT—1848. 
cing the variations of the months in single years from the means of the same 
months drawn from many years. It thence appeared that important varia- 
tions are never merely local, but that the same character of weather prevails 
over large portions of the globe; that the anomaly reaches its maximum in 
one spot, in receding from which it lessens more and more, until, passing 
through places where the thermic conditions are in their normal state, an 
opposite extreme is reached, which so compensates the first, that the general 
sum of warmth distributed over the earth at any particular time of year is 
the same in different years, although the values which make up the sum may 
be very different. Knowing the prevailing character of the weather in par- 
ticular places in the different years, we are enabled to deduce from the devia~ 
tionsat a few normal stations, where the observations extend over a long series 
of years, the quantitative corrections to be applied to the results of observa- 
tions continued for only a few years. The fourth memoir contains the cor- 
rections calculated for nineteen such normal stations :—Madras, Palermo, 
Milan, Geneva, Vienna, Regensburg, Stuttgard, Carlsruhe, Berlin, Copen- 
hagen, Torneo, London, Kinfauns Castle, Zwanenburg, Paris, Salem, Al- 
bany, Gothaab and Reykiavig. These four memoirs also contain the com- 
plete data derived from observations at 700 stations ; or the monthly means 
during the respective years of observation. 
The second necessary correction is that required for eliminating the 
diurnal variation, and reducing the observations made at particular hours to 
the mean of the whole twenty-four hours, as it is only at a few stations that 
observations were made hourly. These latter stations, twenty-nine in number, 
supply the values required to reduce the observations at any particular hour 
to the mean of the twenty-four hours, and are given in the memoir entitled 
“ On the Diurnal Variations of the Temperature of the Atmosphere.” They 
are :—Rio Janeiro, Trevandrum, Madras, Bombay, Frankfort Arsenal, To- 
ronto, Rome, Padua, Kremsmiinster, Prague, Muhlhausen, Halle, Gottin- 
gen, Salzuflen, Brussels, Plymouth, Greenwich, Leith, Apenrade, Christiania, 
Drontheim, Helsingfors, Petersburgh, Catharinenburg, Barnaul, Nertschinsk, 
Matoschkin Schar, the Karian Gate, and Boothia Felix. 
It still remained to deduce from single years the monthly means for periods 
of many years. The temperature tables in the volume of the Transactions of 
the Berlin Academy for 1847, contain the means for the months, for the seasons, 
and for the year, as they follow directly from the observations without cor- 
rection for diurnal variation. These tables have also been calculated in Fahr- 
enheit’s scale, and are published in the Report of the Seventeenth Meeting 
of the British Association, held at Oxford, 1847. Since the publication of 
this work several stations have been added, and for other stations the means 
have been determined from longer series of observations. 
Lastly, it remained to fill up the wide intervals between the stations. by 
the help of points in the intervening seas. This last work consumed a great 
quantity of time, as generally speaking the single observations are not even — 
put together in daily means; and besides the mean place of the ship must 
be determined for each occasion from the continually varying latitude and 
longitudes. It is only in Beechey’s ‘ Narrative of a Voyage to the Pacific 
and Behring’s Straits,’ (which is a true model in point of redaction,) that this 
has been done. Besides the above work, I have made use of the following, 
viz. “The United States-Exploring Expedition” (in which however, as the 
distinct Meteorological Appendix has not yet been published, I could only 
employ the notices found in the text); Captain James Ross’s ‘ Voyage of 
Discovery and Research in the Southern and Antarctic regions ;’ and Du- 
mont D’Urville’s ‘ Voyage au Pole Sud et dans ’Oceanie sur ]’Astrolabe 
