Sequel to a paper on the Reduction of the Thermometrical Observations made at the Apartments of the Royal Society .
By JAMES GLAISHER , Esq. , F.R.S. , of the Royal Observatory , Greenwich .
Received December 14 , 1849 , -Read February 28 , 1850 .
IN a paper which the Royal Society did me the honour to publish in the last volume of its Transactions , I gave the results found from all the thermometrical observations which have been taken at the Apartments of this Society ; and I stated that I had made some progress in the connection of this series of results with those deduced from the observations at the Royal Observatory , Greenwich .
Since that time I have reduced the two series of observations to one and the same series , and I have now the honour to lay the iesults from their combination before the Society .
In my former paper I stated that no observations had been taken between the years 1781 and 1786 .
Had the particulars of these years been about their average values , their omission would not have materially affected the final results , but on examination I found that those years were distinguished by very severe weather , and that their omission would have a sensible effect ; I have therefore supplied these particulars , as detailed below .
I also stated that it was doubtful whether the temperatures , as determined for Somerset House , were influenced by local causes .
I have endeavoured to collect information upon this subject , and of which I shall speak presently .
I shall adopt the same plan in the arrangement of the final results , which I pursued in my former paper ; and present them for monthly , quarterly and yearly periods .
The numbering of the Tables is continued from the former paper .
It may tend to clearness if I speak of each preliminary investigation separately .
Determination of the Miean Temperature of the Air at Lynidon in Rutlandshire , the longitude of which place is 0§ 3 ' East of Greenwich and the latitude is 52 ?
0 32 ' North , for every month in the years from 1771 to 1799 .
The results of meteorological obserlvations taken by THoMAS BARKER , Esq. , at Lyndon in Rutlandshire , and which seems to have been taken with care , wele published in the volumes of the Philosophical Transactions for the years 1772 to 1799 .
From these papers I have detetmined the rean monthly temperatures of the air , and which are shown in the following Table : By taking the mean of the numbers in each column of differences , we find that the temperature of the air at the Apartments of the Royal Society in 0 January was higher than at Epping reduced to the same level by. .
13 February was higher than at Epping reduced to the same level by 12 March was lower than at Epping reduced to the same level by less than 01 April was lower than at Epping reduced to the same level by ... 15 May was lower than at Epping reduced to the same level by ... 10 June was lower than at Epping reduced to the same level by ... 06 July was higher than at Epping reduced to the same level by. .
0'2 August was lower than at Epping reduced to the same level by ... 03 September was lower than at Epping reduced to the same level by .
10 October was higher than at Epping reduced to the same level by. .
0-2 November was higher than at Epping reduced to the same level by .
1-4 December was higher than at Epping reduced to the same level by .
15 And the mean temperature of the whole period was nearly of the same value at both places .
Having obtained these results , and finding that during the time the sun was situated north of the equator the temperature at the Apartments of the Royal Society was lower than at Epping , and that it was hligher during the time the sun had south declination , I requested Mr. SQUIRE to furnish me with full particulars with respect to the position of his instrument to the sun and to surrounding objects : the following is the information he gave me:"The thermormeter hangs near the north angle of a small projecting pier of a wall , nearly close to the brickwork , facing the N.E. , and an open country .
At the back of the wall is a grape-vine , and when in leaf , it so shades the wall that its temperature is not much affected by the sun 's rays ; but , before the vine is in leaf , it may raise the temperature a trifle ; yet from some casual observations scattered over my journal as tests , I do not find that the said thermometer is sensibly influenced by the heat of the sun at 8 A.M. , the time of reading the instrument .
On the 13th of May , 1847 , 1 moved the thermometer a few feet from its former position on the wall , with the same aspect , but , at the back of this part of the wall , there is a sort of grotto or summer-house , which is covered by thatch and completely interrupts the sun 's rays from the wall ; hence its present situation may perhaps be considered more eligible for a mean temperature than its former position , but I do not find any difference worth notice .
" The general fact , however , of a higher winter temperature , and of a lower summer temperature at the Apartments of the Royal Society , is satisfactorily proved by both sets of comparisons , and it is evident that the same cause has been in operation at both times , and to the same amount .
There can be no doubt that this cause is the vicinity of the river Thames to the locality of the observations .
Determination of the Mlonthly Mean Temperature of the water of the Thames by night and by day , from the year 1846 to the year 1849 .
The observations to determine the temperature of the Thames water are made by Lieut. SANDERS , R.N. The instruments consist of one maximum thermometer and of one minimum thermometer , suspended from the sides of the Dreadnought Hospital Ship , in a perforated trunk placed at about 2 feet below the surface of the water .
The range of temperature during the day is usually about 2 ?
, and the simple arithmetic mean of the readings of the maximum and minimum thermometers shows the mean temperature of the water .
TABLE XIII.-Mean monthly temperature of the water of the Thames .
1846 .
1847 .
; 1848 .
1849 .
-Mean of all the !
Mean a Mean of all the Mean Mean of all the Mean t. Mean of all the Mean Month , readings in each tempera- &gt; readings in each temperareadings in each temperareadings in each tempera mt month , u of th .
tur f month .
ture of the ... ... ... ... ... ... ... ... . .
the .
the Max. Min .
water .
Max. Mi .
water .
Max. water .
Max. Mil ater .
~~ Max. ' Min .
Max. Min .
Max. Min .
' Max. Mi:_ January ... .
44'3 42'0 432 3 371 35 35 33 35*7 3'51 35-4 41'6 39'6 406 February ... 45-3 42'5 43*9 38-9 37'38'1 41-8 40'3 41-1 44-2 42-6 43-4 March ... .
48-2 46'3 47'3 42'1 41-4 41-8 ... ... ... .
45'7 44*1 44*9 April ... ... 51-5 49'4 50*5 46.9 46-4 46-7 1 51-1 50 ; 2 50*6 ' 47-9 44-8 46-3 May ... . .
599 586 57.02 56 57 57-8 62.5 61-0 61*8 58-9 55-7 57-3 June ... ... 73'0 70'8 71*9 i 655 61.9 63'7 63.6 62'6 63-1,65'2 63.3 64'3 July ... ... 674 66-1 66'7 70'6 66-5 68-6 66-0 65-0 i 6s5 67'8 66-1 67-0 August ... 68-3 66-7 67'5 66-1 64-4 653 630 62-0 62 5 64-9 627 1 63-8 September .
64-7 63-5 64-1 57'0 56 56-5 59'5 58'8 i 59*1 61-4 58'7 60*0 October ... 542 52'8 53*5 53-3 530 53'2 53'7 508 52'2 52-8 49'5 51'2 Noveimber .
47'6 46-0 46-8 47*9 47-3 47-6 ... ... ... ... .
471 44-1 45'6 December .
37'6 34'9 36-3 4S 41 4250 419 5 4205 4392 40-3 36-8 38-6 By comparing the means of these numbers for the four years , with the means of the readings of the maximum and minimum thermometers in air at the Royal Observatory , for the same months , we find that the mean lowest readings of the water were higher in the twelve months respectively by 3 ?
09 ; 40 ?
6 ; 7 ?
7 ; 12 ' ?
3 ; 11 ?
*5 ; 160 ?
7 ; 12 ?
04 0'4 ; 10-4 107 ; 60 8 ; 60 ?
3 and 4 ?
09 , than the mean of the lowest readings of the air ; and it was lower than the mean maximum readings of the air by 30 ?
2 ; 40.6 ; 50.3 ; 9 ?
2 ; 6 ?
-0 ; 70 ?
5 ; 801 ; 6 ?
07 ; 70 ?
4 ; 6 ?
0 - ; 40 ?
7 and 30-0 , in the respective months from January .
These numbers are very large , and will fully account for the little higher temperature possessed by places in the vicinity of the river , and these differences of temperature are probably the fiuitful source of the London fogs .
Mr. SANDERS , at my request , has taken daily observations of the temnperature of the air at 32 feet above the water of the Thames at the hours of 6 A.M. and 6 P.M. during the years 1847 and 1848 , and at the hours of 9 A.M. and 9 P.M. in the year 1849 .
The result of these observations , compared with simultaneous observations taken at the Royal Observatory , is as follows:-that the temperature of the air 32 feet above the water , exceeds that at the observatory at 6 A.M. by 10 ?
6 ; 10i0 ; 00'8 ; 00 ?
3 ; 00 ?
6 ; 0 ?
7 ; 0 ?
'9 ; 0 ?
08 ; 002 ; 0 ?
0 and 0 ?
-8 , in the twelve months respectively ; and at 6 P.M. by 10 ?
2 ; 0 ?
08 ; 1 ?
0 ; 0 ?
-8 ; 0 ?
07 ; 0 ?
08 ; 0 ?
06 ; 0 ?
08 ; 1 ?
-3 ; 1 ?
00 ; 1 ?
07 and 0 ?
-9 , in the twelve months respectively ; that at 9 A.M. it was in excess in January by 1 ?
03 ; FebruaLy by 1 ?
'5 ; March by 00 ?
6 ; April by 0 ?
4 ; May by 20 ?
2 ; June by 0§ 4 ; and in October by 0 ?
5 ; that it was of a lower temperature in July by 007 ; August by 0 ?
5 ; and in September by 0 ?
'1 ; that at 9 P.M. it was always of a higher temperature : the excesses were 0 ?
'1 ; 0 ?
03 ; 0 ?
7 ; 0 ?
03 ; 1 ?
09 ; 2 ?
'9 ; 1 ?
05 ; 3 ?
'2 ; 1 ?
02 and 1 ?
03 respectively .
From these numbers , it seems that during the night hours , at all seasons of the year , the temperature of the air at the Dreadnought Hospital Ship is higher than at the Observatory , and that it is below only during the midday hours .
At times of extreme temperature the effect of the water upon the temperature of the air is very great .
On February 12 , 1847 , the temperature of the air at my house , which is situated about one mile and a half from the river , was 6 ` ?
0 ; the lowest reading , 32 feet above the water of the Thames , was 16 ?
0 ; the temperature of the water was 330 ; its heating effect upon the air in its imrnmediate vicinity amounted to 10 ?
; at the Observatory the reading was 100 ?
5 ; and the heat of the water of the Thames seems to have influenced the temperature of the air at the Observatory to the amount of 4 ? .
Some time since , on comparing the temtperatures of the air as recorded in the Philosophical Transactions in the year 1814 , with corresponding teniperatures as observed at Greenwich , I doubted the accuracy of the former in many instances , on account of their much higher values ; these investigations have now led me to believe that the temperatures , as lrecoirded in the Philosophical Trlansactions for that year , are correct .
TABLE XIV.-Comparison of the mean temperature of the Air at St. John 's Wood , and at the Royal Observatory , Greenwich .
1841 .
1842 .
1843 .
Mean temperature of the air .
Mean temperature of the air .
Mean temperature of the air .
Month .
At At In At At In At At In Royal ObSt .
John 's excess at Royal ObSt .
John 's excess at Royal ObSt .
John 's excess at servatory , Wood St. John 's servatory , Wood .
St. John 's servatory , Wood St. John 's Greenwich .
Wood .
Greenwich .
Wood .
Greenwich .
Wood .
January ... ... 336 34 3 +07 3 29 324 -5 399 3 -0 1 -0-8 February ... 35-3 36-5 + 1'2 408 40-3 --05 36-0 36-1 +0-1 March ... ... 46-2 46*6 + 0'4 44-9 44-3 -0-6 42'9 42'9 0.0 April ... ... . .
47-0 46.6 -0'4 45-2 46'3 + 11 47-1 47'6 +0-5 May ... ... ... 56-8 57-0 +0.2 53'2 53'2 0'0 52*2 51-2 -10 June ... ... ... 56-i 55-9 -0.5 62-9 62-5 -0-4 56-3 55'2 11 July ... ... ... 57*8 56*9 -0 9 60-2 59-5 -0-7 60-9 60-1 -0-8 August ... ... 605 59'5 -1'0 65-4 65-5 +0'1 62'1 61-9 -0*2 September ... 58-5 57-8 -0-7 56-4 56-2 -0'2 59-5 60-3 +0*8 October ... ... 48-8 49-1 +0-3 454 45'8 +0-4 48-0 47'6 -0-4 November ... 42-7 42-5 -0*2 428 43-3 +05 43-8 43 ?
9 +0-1 December ... 40-5 40-2 --03 45-0 44-6 -0-4 43-9 442 + 0-3 R~~~~~ 4 .
' IDCCCL .
4 E MtR.'GLAISI-IER ON THE THERMOMETRICAL OBSERVATIONS TA .
KEN AT 1TABLE XIV .
( Continued .
) 1844 .
1845 .
1846 .
Mean temperature of the air .
Mean temperature of the air .
Mean temperature of the air .
Month .
At At In At In At In Royal ObSt .
ohn 's excess at Royal ObSt .
John 's excess at Royal ObSt .
John 's excess at servatory Wood St. John 's servatory , Wood St John 's servatory , Wood St. John 's Greenwich .
Woodd .
Greenwich .
Wood .
Greeich .
ood .
- ... ._ ... ... . .
, January ... ... 39I 36-9 -22 38*3 38'7 +0'4 4317 431 -0*6 February ... 35-2 35'7 +0-5 32-7 32'9 +0'2 43-9 43-7 -0'2 1 March ... ... . .
41-5 41-6 +0-1 35-2 35-3 +0-1 43.3 44'0 +0 7I April ... ... ... 51-7 52-2 +05 46*3 47-5 +1*2 47-1 46-8 --03 ' May ... ... . .
52'9 52'5 --04 49-4 49 ?
2 -02 54-6 55.6 +1 0 June ... ... . .
60-7 59'9 --08 60'7 60 6 -0'1 65*3 65-2 -0'1 July ... ... . .
61-4 61-0 -0-4 59'8 59-4 --04 64-5 63-6 --09 August ... ... 57-7 57'4 -0-3 57-3 57-6 +0'3 63-2 62-5 -0-7 September ... 56-9 57-4 +0'5 53-6 53'8 +0-2 60-1 60-1 0'0 October ... . .
49 49492 -0-3 50-2 50'1 --01 50-5 49-9 -0'6 November ... 44-0 43-3 -0'7 45-8 45-3 -0-5 46-0 44-9 -1-1 December ... 33-0 33-8 +0-8 41-7 40-5 -1-2 32-9 32^7 -0'2 The observations at St. John 's Wood consisted of one observation of the maxitnum thermometer and of one observation of the minia umn thermormeter , daily , as well as three observations at different times every day of other therlamotneters .
They were made by GEORGE LEACH , Esq. , to whom I am indebted forthese results .
The mean .
values for the month were obtained by the application of corrections from my tables .
By examining the columrn of differences , no cei-tain difference exists between the temperatures of the air at Greenwich and at St. John 's Wood .
TABLE XV.-Meani monthly temperature of the Air at Fleet Street , London , as determined from obser-vations taken by Mr. W. SIMMS , optician .
Year .
January .
Feb. March .
April .
May .
June .
July .
August .
Sept. Oct. Nov. Dec. 1838 ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 526 42 3 39-9 1839 .
38S5 40-4 40-4 41]2 50O8 61i2 61-5 62-0 57-1 51-1 43-6 41-3 1840 .
41-4 38-9 39-8 52-3 55-5 58-9 59-4 65-1 54-8 48-9 45-8 36-6 1841 .
35-0 36-2 1842 ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... . .
47'6 1843 .
426 37-8 4366 46-2 494 53-4 58-4 621 ... . .
473 45-9 47'1 1844 .
43-6 35-8 43-5 49-3 50-0 58-2 60-8 56-8 57-5 50-4 44-8 36-7 1845 .
41-8 35-2 37-0 46-0 46-9 55-6 58-6 56-2 54-3 51-2 47-8 44-3 1846 .
46-1 45-4 47-8 45-9 53-9 62-3 1848 .
35-9 44-3 43-0 47-9 57-9 55-9 63-4 57-4 58-0 52-1 44-1 45-0 These observations were taken with good instruments , and believed to be taken with every care : Mr. SJVIMMS very kindly lent me the oiriginal observations , form which I have deduced the above values .
By comparison with the simultaneous observations taken at Greenwich , it seems that the temperature at Flee Set et upon the whole year is 00 ?
7 higher thain at Gireenwich .
By taking the means of the numbers in each column of differences , we find that the temperature of the air at the Apartments of the Royal Society was Higher than at the Royal Observatory , Greenwich , in January by. .
16 Higher than at the Royal Observatory , Greenwich , in February by .
1P4 Higher than at the Royal Observatory , Greenwich , in March by. .
1'1 Higher than at the Royal Observatory , Greenwich , in April by. .
11 Higher than at the Royal Observatory , Greenwich , in May by ... 11 Higher than at the Royal Observatory , Greenwich , in June by ... 0-9 Higher than at the Royal Observatory , Greenwich , in July by ... 0 Higher than at the Royal Observatory , Greenwich , in August by. .
13 Higher than at the Royal Observatory , Greenwich , in September by .
11 Higher than at the Royal Observatory , Greenwich , in October by. .
13 Higher than at the Royal Observatory , Greenwich , in November by .
13 Higher than at the Royal Observatory , Greenwich , in December by .
1'4 And upon the whole year the excess of temperature at Somnerset House was 1 ?
02 and to reduce readings taken at Somerset House to those at Greenwich , it is necessary to subtract the preceding numbers from them ; and it is necessary to increase the readings of the Royal Observatory by the above numbers to make them comparable with those taken at the Royal Society .
One part of these differences is owing to the difference of elevation , and which will amoiunt to about 00 ?
3 ; the greater part of the remaining difference is most probably owing to the vicinity of the water of the Thames , whose tenmperature during the night hours , at all seasons of the year , is several degrees higher than that of the air ( see remarks following Table XIII .
) .
The general result of the preceding investigations , with respect to the temperatures of London and the country is , that those parts of London situated near the river Thames , are somewhat warmer upon the whole year than the country , but that those parts of London which are situated at some distance from the river , do not enjoy higher temperatures than those due to their latitudes .
I proceed now to reduce the results at Somerset House to those of the Royal Observatory , Greenwich , by applying the numbers following the preceding Table to the numbers in Table I. of my former paper , and to those in Table X. till the year 1840 .
After this date the numbets are extracted from the several volumes of the Greenwich Meteorological Observations .
By taking the means of the numbers in each column without respect to sign , we find that the variability of temperature is greatest in the winter months ; its mean value in January is 3 ?
1 ; in both February and March is 2 ?
06 ; in April is 2 ?
'3 ; in May and June is 20.0 ; in the months of August , September and October , whose temperatures are the steadiest , it is 10 ?
9 ; in November it is 2 ?
0 , and in December it is 3 ?
'1 , as in January .
The numbers in the preceding Table very clearly show that causes exist at different times , which raise or depress the temperature , and which continue through long periods .
As in the distribution of the positive and negative signs in the space of seventynine years , we perceive a gradual increasing preponderance of the positive signs over the negative signs , it seems that the temperature of the climate during this period has increased .
As the mean results from so long a series of observations may be considered as true , having the advantage of being free from errors of observation and from those arising from imperfect instruments , we may really consider the numbers in the above Table as abnormal values ; yet as it seems most desirable to have those at the beginning of the series confirmed by the description of each year , made without instrumental means , for this purpose , as well as for the comparison of the character of the climate at the beginning and at the end of the series , I have collected the following brief particulars of every year till that of 1800 ; after this time the general characters of the years are well known .
1771.-There were frequent and very sharp frosts till April 20 .
On February 12 the reading of the thermometer was as low as 40 ; the month of May was warm ; the summer was cool and dry ; October was a wet and windy monthl , and the weather was mild to the end of the year .
The severe weather of the beginning of the year caused a bad seed time , and the harvest was very late .
1772.-The beginning of the year was mild ; from the middle of January frosts and great snows were frequent , and continued to the middle of March .
The summer was very fine ; the autumn was mild but wet , and there was no frost till Decemnber 22 .
1773.-With the exception of the latter part of February , which was stormy and wet , there was much fine weather till the beginning of May ; then many mornings were frosty , after which heavy rain fell frequently till June .
The summer was fine ; the autumn was wet .
Thiere were sharp frosts at the end of November and at the beginning of December .
1774.-'The year began with severe frost , and for nearly two months the ground was frost-bound ; occasionally there were great rains or snow ; the weather was more moderate in April ; the summer was cool with heavy rains .
The autumnal months were wet , particularly in September .
Some snow fell in November and beginning of December .
This year was remarkably wet .
1775.-The weather was mild at the beginning of the year .
The summer was dry and hot ; thunder-storms were frequent in autumn .
The year was very fine , and grain was cheaper than it had been for many years past .
1776 .
In January there fell a greater quantity of snow than had fallen for some years , and the frost was supposed to have been the most severe since 1740 .
The frost went away at the beginning of February , and the weather following was mild and wet ; it became hot about the middle of April .
May was cold and dry , with north winds ; after this the weather was mostly fine till the end of December , when there was a sharp frost .
1777.-The year began with a sharp frost , and heavy falls of snow continued till towards the end of February ; for a few days about the end of March the weather was unusually hot , the reading of the thermometer being nearly 70 ?
; after this the weather was windy and cold till June .
The latter part of the summer and autumn was fine .
The year ended with frost and snow .
1778.-There were frost and snow at the beginning of the year ; the beginning of April was fine .
The summer was fine and hot , supposed at the time to have been as fine a summer as that of 1762 , if not as fine as the summer of 1750 .
Frosty mornings began in September , but were less frequent afterwards .
On the last day of this year there was a violent storm , supposed by some to have been as violent as that of 1703 .
1779.-After the beginning of January there was no frost ; the spring months were remarkably warm .
In February wall fruit flowered ; the middle of April was quite hot , as was the summer and autumn ; about the middle of November there was a little frost , and again on December 22 : there was much sickness this year .
1780.-This year began with a frost almost as severe as that in 1772 ; there was not much snow , and the weather continued severe till near the end of February .
The month of March was warm ; it was hot from July to September , and mostly mild till Christmas , when a frost set in .
The year was sickly .
1781.-There was a little frost at the beginning of the year ; the spring was mild , the summer was hot , and the ground was much burned .
Autumn was fine and pleasant , and there were only a few frosty mornings during the remainder of the year .
1782.--The beginning of the year was mild , but in February it was frosty , and the remainder of the winter was severe ; the spring was cold ; nearly 12 inches of rain fell in April and May ; the weather was fine in June , but bad afterwards ; the autumn was cold ; it was severe in November , and during the first half of December .
1783.-The spring was pleasant , with frosty mornings very constant till near April .
A remarkable haze was prevalent all over Europe during the summer .
The autumn was fine , and the weather was mostly mild till the last week in December , when a great fall of snow took place .
1784.-There was steady frost with snow till February 21 , and till the end of March the mornings were frosty ; and at the end of March there were cold winds with snow .
This weather continued till the middle of April ; and till the first week in May frosty mornings were frequent , and the remainder of May was exceedingly hot .
There were a few hot days in July , but the weather was precarious throughout the autumn ; and in December the frost was as severe as it was in January .
1785.--The severe frost of the preceding month broke early in January , but on the last day of that month a second very severe frost set in and continued till the middle of March .
This winter was most severe .
The summer and part of autumn were showery ; a heavy fall of snow took place at Christmas , with severe frost .
1786.-The frosts at the beginning of the year were of short duration .
From the beginning of March there was a severe frost of a fortnight 's duration , and cold E. and N.E. winds were prevalent with fiosty mornings till the beginning of May .
June and July were moderately fine ; August was cold and showery ; and from this time to the end of the year there was a great deal of rain .
1787.-The year began with open weather .
April was cold with N. winds , and vegetation was stopped ; during April and May frosty mornings were frequent , and there was a sharp friost on the morning of the 7th of June ; it was a cold summer ; the autumn was mnild , and there was a heavy fall of snow and a week 's frost at the end of the year .
1788.-January and February were mild , the latter month being wet ; there was a fortnight 's frost in March ; there were several periods of hot weather in April , May and June .
The summer was in general dry ; autumn was fine ; there was a gentle frost at the beginning of December , then an exceedingly severe frost set in with heavy falls of snow , which continued to the end of the year .
This year was remarkable for abundance of fruits , &amp;c. 1789.-Very heavy storms of wind and snow took place till the middle of January ; and large rivers were frozen over ; there was a great loss of fish in ponds from the severity of the cold .
After the frost broke the weather was mild , but windy and wet .
During March there were nearly constant N. winds , and heavy falls of snow were frequent with sharp frost .
The summer was mostly wet ; August was fine , after which it was again wet , and continued so to the end of the year , with scarcely any frost .
1790.-The weather was mild and open till April , when the first snow fell in the year , and the weather , during the beginning of this month , was the most severe during the winter .
The summer was cool , cloudy and windy ; autumn was fine and pleasant ; December was stormy with very changeable weather .
1791.-Till January 6th there was frost ; after this the weathler was mild till towards the end of April ; there were many frosty mornings with cold N.E. winds in May .
The former part of the summer was cold ; frosty mornings were fiequent till the middle of June , the latter part of summer and autumn .
During November and December there were frequent storms and falls of snow and frost .
1792.-There were frequent sharp frosts till March , with stormy and wet weather ; the beginning of March was mild , after this there was a frost of a week 's durations The summer was wet and cold ; the autumn was wet , and December was cloudy , with very little frost .
This year was very wet .
1793.-January and February and beginning of March were mild ; a fiost set in at the end of March ; there was a great fall of snow in the first week in April .
The former part of the summer was cold , with frequent frosty mornings till June ; July was wet ; the autumn was fine , mild and calm , and there was no frost till the end of the year .
1794.-The year began with slight frost , which continued till the end of January ; February was very mild ; the spring was warm till May , which was cold ; July was hot ; the autumn was wet but mild , as was the first part of Decetnber , but the weather during the latter half of the month was severe with heavy snow .
1795.-The frost began about the middle of December 1794 , was excessively severe in January , and continued till the end of March .
There were very lairge falls of snow , and the consequent floods were so great that nearly all the bridges in England were injured .
Some snow fell in April .
The summer was cold , with frequent fiosty mornings till June ; there were some hot days in July , but it was generally cold ; after this the weather was fine till autumn .
In December much injury was done to shipping by the strong S. and S.W. winds ; there was no frost .
1796.-January was remarkably warm , with occasional thunder-storms ; there was no frost till March , and then of no long duration .
The summer was cool ; the autumn was fine with a few frosty mornings at the end of November ; in December a severe frost set in , and the reading of the thermometer in many places on the 24th was below zero of FAHRENHEIT'S scale .
1797.-During a few days in January the fiost continued ; after this , till the end of March , scarcely any rain fell , and the weather was fine with frequent frost .
From April to September there were frequent heavy rains .
The summer was cold ; there was some warm weather in July ; the autumn in general fine , and the weather continued open till the end of the year .
1798.-With the exception of a few slight frosts , which occasionally occurred till March , the weather was open and mild .
The summer was fine , as was autumn and the beginning of December after this a very severe frost set in , and the reading of the thermometer was as low as 5 ? .
1799.-The severe frost which set in about the middle of the preceding month continued to the middle of January , and again set in towards the end of the month with much snow , which continued during the first week in February ; some snow fell in March , and the mornings were frosty till the end of the month .
From April to the middle of November was wet ; December was foggy ; and after the 17th a severe frost set in with snow falling .
The whole year was remarkably cloudy .
If we compare the character of the preceding years with the abnormal differences shown for the same years in Table XX .
, the agreement is most satisfactory , and leaves no doubt upon the correctness of the numbers at the beginning of this series .
I do not think it necessary to lescribe the years from that of 180 , as mostof them ae well described by LUKE HOWARD in his ' Climate of London .
' ( Continued .
) Januar , April , y , Group of April , Group of of Group of For the Group of Y ear .
Februs ary , .
years .
May , year .
Auus years .
March .
JSne .
September .
December .
1820 .
--1.6 + 40-4 , -§ 19 ; -0-6 -0*9 1821 .
+ 0-5 O ; -08 J +0-3 +4-0 +1-0 1822 .
+5-3 , +2-9 + e0-5 r +2-1 +21823 .
-1-7 -1 2 -1-3 +01 .
--0 1824 .
-0-5 -2-7 0 , A +0+ va +^5 04 1825 .
0-0 &gt; + 0 +1-6 +27 +04 +-9 S+0 +0 28 -+13 +13 +05 1825.+1-+-+0 +18 3 +1-3 1826 .
+0-9 +1-9 +24 1 +1-3 +16 i 1827 .
23 + 0-3 ++0*4 ' +2-4 02 ' 2 1828.^ +3-0 X +1 5 +0-1 +2-8 +1-8 1829 .
-1-9 +0-3 -2-4 -28 -1-7 1830 .
-1*3 +0 -12 0-0 -0*5 1831 , +1-5 +1-3 +1*9 +317 +2-1 1832 .
0-0 +0-5 0 -02 +24 +0-8 1833 .
-0-1 +2-7 -2-0 +2*1 +7 1834 .
: +4-7\.HA +22 -1 8 +2740 1835 .
+18 + +10 +2-2 -+0 3 1836 .
+ -1 1 -0-4 -1-0 -0-6 ' -0-2 1837 .
-0-5 -3-8 -06 +0-9 -1-0 1838 .
38 --23 -1 2 -0-3 --19 1839 .
+0-1 -2-3 ' -1-2 +1-0 -0-6 1840 .
00 -+1-5 -1-3 + -2-2 ' -05 , 1841 .
+0-1 5 f+1-4 -0-6 +06 +04 ' 1842 .
+1-3 + 13 17 + 13 +1-0 +13 1843 .
+1-3 -02 +1-4 +18 +1-1 1844 .
+0-4 +3.0 -+1-5 -01-2 0+ 0-4 1845 .
-29 +* 00 1 -2-5 +0-3 +25 1*.7 &gt; +1*0 1846 .
+5-4 +3*6 +3-2 -0-3 +30 1847 .
-1-1 + 11 12 +4-1 1 +13 1848 .
+2-3 +32 ; -0-8 +31 ; +19 1849 .
+36 -0-4 9 +14 +1-6 The sign denotes that the temperature of that period was below the average , and the sign + denotes that it was above the average .
These numbers do not at all confirm the idea that a hot summer is either preceded or followed by a cold winter , or vice versed ; on the contrary , it would seen that anyhot or cold period has been mostly accompanied by weather of the same character .
The cold year of 1771 was followed by two cold years .
The hot year of 1779 was preceded by one warm year and followed by two others .
In 1780 the extreme cold of January was more than counterbalanced by the extreme heat of March .
The cold year of 1782 was followed by a long series of cold years .
The very cold year of 1799 was followed by a cold autumnn and winter .
The warm year of 1806 was preceded by a warm winter .
The very cold year of 1814 ( the last very cold year we have had ) was preceded by a cold sumrmer , autumn and winter .
The hot .
ear of 1818 was preceded by a moderate winter , and was followed by a warm one .
The hot year of 1822 was preceded by a warm winter and was followed by a amnoderately cold one .
The hot year of 1834 followed a very mild winter and was followed by another .
The hot year of 1846 was preceded by a warin winter and was followed by a moderate one .
The warm year 1848 was both preceded and followed by warm periods .
The mean telmperatures of the years 1771 , 1782 , 1784 , 1786 , 1799 and 1814 , were all below 46 ?
; the coldest was 1784 , and its value was 450'1 .
The mean temperatures of the years 1779 , 1818 , 1822 , 1834 and 1846 , were all above 50 ?
'5 ; the year of hlighest temperature was 1846 , and its value was 510 ?
3 .
Thus seventy-nine years , from 1771 to 1849 inclusive , gives a mean temperature of 480 ?
3 , with a variation , between one year and another , from 450 ?
1 in 1784 to 51 ?
0.3 in 1846 ; the difference is 6 ?
'2 .
TABLE XXIII.-Showing the mean temperature of the Air in Spring , Summer , Autumn , Winter , and for the year from March , and the same for successive groups of years .
Spring .
Summer .
Autumn .
Winter .
The year from March .
Year .
March , ; June , Group of September Gup f Deceber Group of ee ole Group of April , Group of : July , years October , vear January , , a May .
years .
August .
'November .
February .
1771 .
42 67 573 447 359 45 1772 .
43-1 59-7 i503 37*1 47-5 1773 .
' 44-2 58-3 ; 47'1 35'9 46'4 1774 .
46-9 1 607 : 47.6 i 39 8 48'7 1775.484 61 489 &gt; 49'0 : 35.4 5 36.9 48.8 &gt; 48 ' 1776 .
47-2 60.7 49-6 36'6 48-5 1777 .
47-0 59*7 51*0 35-4 48'3 1778 .
47-3 63-9 48*0 41*0 50.0 1 779 .
50*8 Jl 6 !
'-3 51'5 1 34-7 J 49-8 1780 .
49-6 : 6277 ^ 49-6 ' 37-7 49*9 ' 1781 .
47-3 63'6 49-0 4985 A 1782 .
' 425 57-9 45'1 37-4 45'7 1783 .
44-6 61-6 48*7 32.0 46-7 1784.45-0 57'3 473 47-5 325 36 455 47 1785 .
442 ' 59'8 u 48'0 !
35'8 49 ' 1786 .
43-7 59'2 44-'2 374 r 46'1 1787 .
46'2 60o1 &lt; 475 38-6 48-1 1788 .
48'-1 " 59-7 : - ; 485 -34'1 47'6 1789 .
43-9 J 57'9 46-7 40'5 473 1790 .
45-6 ' 58'6 48*5 392 480 1791 .
46-859'5S 48 ' 35-7 476 179-2 .
46-9 i 58-4 49-1 38-3 48'2 1793 .
44-1 5918 ' 49'2 39'7 48-2 1794 .
48-42 r 60-8 489-41 s 631'6 1 37'.0 47'2 47'7 1795 .
452 57-8 * 51 6 .
43-2 49-5 &gt; 1796 .
45-858- ' 48-6 33-'8 46-6 1797 .
44-6 59'0 47-7 39'0 476 1798 .
47*8 61-3 48-8 34-5 481 1799 .
427 57-6 482 34-6 45-8 1800 .
4 '6 64 60'7 1 and 1845 , and the mean of their temperatures was 430 ?
1 .
The coldest spring , during the whole period , was in the year 1837 , and its mean temperature was 400 ?
9 .
The years distinguished by hot springs were 1779 , 1780 , 1811 , 1822 , 1830 , 1841 and 1848 , and the mean of their temperatures was 50 ?
'0 .
The hottest spring , during the whole period , was 1779 , and its mean temperature was 500 ?
8 .
The mean temperature of summer , from all the observations , is 60 ?
00 .
The years distinguished by cold summers were 1771 , 1784 , 1799 , 1812 , 1813 , 1814 , 1816 , 1817 , and the mean of their temperatures was 560 ?
9 .
The years distinguished by hot summers were 1778 , 1779 , 1780 , 1781 , 1818 , 1826 , 1831 , 1834 , 1835 , 1842 and 1846 , and the mean of their temperatures was 63 ?
02 .
The coldest summer within the period was that in the year 1816 , and its mean temperature was 55 ?
02 .
The hottest summer within the period was that in the year 1846 , and its mean temperature was 64 ?
03 .
The mean temperature of autumn , from all the observations , is 490 ?
3 .
The years distinguished by cold autumns were 1771 , 1782 , 1786 , 1789 and 1829 , and the mean of their temperatures was 450 ?
'9 The years distinguished by hot autumns were 1779 , 1795 , 1804 , 1806 , 1811 , 1818 , 1821 , 1822 , 1831 and 1846 , and the mean of their temperatures was 520 ?
3 .
The coldest autumn within the period was that in the year 1786 , and its mean temperature was 440.2 .
The hottest autumn within the period was that in the year 1718 , and its mean temperature was 54 ?
*5 .
The mean temperature of winter , from all the observations , was 370 ?
6 .
The years distinguished by cold winters were 1783 , 1784 , 1794 , 1796 , 1813 and 1829 ; the mean of their temperatures was 32 ?
'6 .
The years distinguished by warm winters were 1778 , 1795 , 1803 , 1806 , 1821 , 1827 , 1833 , 1845 and 1848 , and the mean of their temperatures was 42 ?
01 .
The coldest winter within the period was that in the year 1794 , and its mean temperature was 31 c6 .
The warmest winter within the period was that in the year 1795 , and its mean value was 430 ?
2 .
The winters of the years 1833 and 1845 were remarkably warm , being both of the value of 43§ 1 .
By taking the difference between the mean temperature of each period from all the observations , and the mean temperature for the same period , in every year , the next Table is formed .
From the particulars in this Table , it seems that the highest reading of the thermometer within the year has occurred three times in May , seventeen times in June , twenty-four times in July , ten times in August , and three times in September .
The lowest reading of the thermometer in the year has occurred twenty-eight times in January , nine times in February , once in March , once in November , and fourteen times in December .
TABLE XXVIII.-Showinge the highest and lowest temperature during the period in each month .
At the Apartments of the Royal Society .
At the Royal Observatory , Greenwich .
Month .
The lowest The highest VMean of all the The lowest The highest Mean of all the reading of reading of Diffeenc renading of reading of ofDifference the thermothe thermoof readings .
Lowest Highest the thermothe thermoof readings .
Lowest Highest meter .
meter .
readings , readings .
meter .
meter .
readings .
readings ' _ 4 : ... Januaryr .
7.0 : 5707 2670 0 2i 590 February ... 17'7 61-0 43-3 26'1 153-7 7'7 623 54'6 19'9 54,3 March ... 23'0 66'0 43'0 29'2 585 s 13'1 71-5 58 4 24-6 62*7 April ... ... 270 77'0 50'0 32'8 66'5 27'0 76'5 49-5 -298 70'3 May ... ... 31-0 84*0 53'0 39-3 73*8 33-5 86-2 52'7 36-4 77-9 June ... ... 40-0 87-5 ' 47-5 45-5 79'5 38'6 87*6 49'0 42'5 83'0 July ... ... .
445 93'5 49'0 493 80'5 422 2 93'3 51*1 45-5 852 August ... 45'4 89'0 43-6 7 49'4 79'3 F 4-0 920 ' 500 0 44*5 i 82'6 September 37-0 83'0 46-0 429 73*6 32'0 86'4 544 363 78 October ... 30-0 75-0 45-0 35-5 65-2 28-3 74'0 45*7 31-4 68-4 November 23-8 63-0 392 30-4 57-4 22-2 66-3 44'1 26-0 59'6 December 4-0 58'0 54-0 25'9 53-6 164 62'8 46*4 23'5 55-5 The lowest reading as observed at the Apartments of the Royal Society , was 40 ; it occurred in the night of December 24 , 1796 ( at this time the reading in the environs of London was-6 ?
) .
The highest reading was 89 ?
; the difference of these readings is 85 ? .
The lowest reading , as observed at the Royal Observatory within the years 1841 to 1849 , was 4 ?
, and the highest was 93 ?
03 ; the difference of these readings is 89*'3 , XXXI.-Showing the mean monthly minimum temperature by night , the mean monthly maximum temperature by day , the mean daily and monthly range of temperature , at the Apartments of the Royal Society , and at the Royal Observatory , Greenwich , together with the increase of temperature month by month , by night and by day .
II Monthly increase in the meamn , Mean of all the maxiLower The mean monthly mum readings , or of mean mum readings , or of mean The mean daily range of emperatu thy f the lowest temperaOf the higest tempe.the lowest temperature minimum the highest temperature maximum of temperature at the range at at the mon ture by eight ratur byda Month .
by night at the reading at by die at the reading at the oyal em , ... ..t ... .- .
thh ; the Royae l. hRy Society .
t yl teof At the Atthe At the At the Royal Royal ObSoc y Royal OhS iety .
Royal Royal OhRoyal Royal Oyal Royal ObRoyal ; Royal Obseratory .
Society .
servatory .
Society .
servatory .
, Society .
servatory .
Society .
servtoey ato .
et .
ervaory .
December ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ..:. .
, ~ *--3'1 --3.1 --1 ' 15 ' -2-7 -1-8 January ' ... ... 34-4 33-1 13 'I 46-2 42-2 i -3-0 6-8 8-5 1-7 28-4 3-7 43 .
-2-5 +2-0 ' +0-5 +1 +18 February ... ... .
36-4 33-6 2-8 / 44-3 44-0 -03 7-9 99 2.(0 76 343 6-7 +31 -+2-7 +1-4 -+2'7 +3-6 +6-3 March ... . .
37 ' 8 36-3 1-5 48-2 50-3 2-1 10-2 13-4 3-2 29-3 38-2 8-9 *:ril ... ... ... : 425 i Bg ?
f 1 5 ?
3 ?
ii : -{ ' , ! .
+-4'8 +4-7 -.-2-9 -+7'0 --6 ' 5 April ... ... ... 42-5 39-2 33 54-7 56-8 21 12-3 16-7 4-4 3317 40-5 1 6'+8 +6-9 +5-8 +6-5 -+-8 ' 2 +8-4 May ... ... . .
48-3 ; 45-7 2-6 j 63-1 66-3 3-5 14-8 18'9 4-1 4 348 415 : 67\ , \:\ .
^:i~~~~~~~~ -{-5 .
; '- ?
^5.4 +4'8 +3'8 +55 +7 ' 4 June ... ... ... .
53-1 49'5 3-6 f8'6 72-6 4-0 15'5 19-4 3-9 34-0 40-5 65 : '~ ~ ~~~~~~~ !
-~~~~~~~~~~~~ ... &lt; ~-E +3'3 +3'*5 +3'7 +-2'8 +0-5 July ... ... ... : 56'6 53.2 3'4 71'4 73-1 1-7 14'9 18-4 353 31-2 397 .
85 -0-8 0'0:+02 -0 ' 6 --10 August ... . .
56-9 53-4 3-5 ' 70-8 72-1 1-3 14-1 17-6 3.5 29-9 38-1 8-2 -4-2 : : : -42 -40 --4-0 ' -5-7 -4-8 September ... 52-6 49-4 3-2 65-1 67-3 2-2 12-5 17-0 4.5 30-7 41-8 11-1 " ' : : ' '.-7'0 -5-8s -5-8 -8-4 -9-7 October ... ... 46-8 43-6 3-2 56-7 57-6 0-9 9'9 13-3 3'4 i295 36-9 74 --6-9 -6-4 -4-9 -8.5 79 November. .
40-4 38-6 1-8 48-2 ~ 49-7 1.5 12 8-0 105 27-0 34-0 70 3-6 -2-9 -4-1 --4-3 ~ -5-7 December ... .
37-5 34-6 2-9 43-9 44-0 0-1 6-4 9-0 2-6 27-1 323 52 The successive differences shown here between the results at the Apartments of the Royal Society and at the Royal Observatory , are harmonious with the differences exbhibited in Table XIII .
and following remarks , between the temperatures of the water of the Thames and those of the air at Greenwich , but of less amount , and they indicate the great influence the presence of a tidal river has upon the meteorological elements of the district through which it passes .
a0 E- ?
02 ss ?
r3 t'l O ttlf ?
0 S3 f0:S fP3 ?
0 ; o 133 0 .
tz OD Lt Ili GREENWiCH .
607 The subjects of this paper are the determination of mean numerical values , and the establishment of the laws of periodic variation from the long series of observations which were taken under the direction of this Society , combined with that still being made at Greenwich .
I have not attempted to deduce any rules for non-periodic variations .
It is most fortunate that through all reports this series of observations continued unbroken for so many years , and that it did not cease till that at Greenwich had been in operation for two or three years .
The number of observations treated of in this paper exceeds 200,000 , spread nearly equally over seventy-nine years , and the results generally are important additions to science .
I consider the determination of the mean temperature at Greenwich as a real addition ; it is probably the best determination of this element of any spot on the globe , and it will help to an accurate knowledge of the mean temperature of its surface .
I may however here remark , that none of the mean results in this paper could have been calculated if observations at equal intervals had not been taken throughout the twenty-four hours and continued for a few years .
The observations at Greenwich have supplied this want .
It was upon them , taken for five years , I based my determinations of curve of hourly mean temperature , and by this means have made all the observations available .
As it is difficult to have instruments read more frequently than twice or thrice in a day , yet , to make these available , it is necessary that the laws of diurnal changes of temperature be known ; and where such is not the case , an effort should be made in all countries to have a series of hourly or bi-hourly observations made for a few years , so as to be able to deduce usefil results .
Such series of frequent observations need not be made near each other , as it is found that the observations made over a considerable extent of country are subject to the same general laws .

