METEOROLOGY. 



159 



Mean tem- 

 perature of 

 th,: interior 

 Jet 

 with tht 

 mean or the 

 duly ex. 



kssiisst 



F.ut the mot important fact established by the above 

 the remarkable Coincidence between the 

 average temperature of the earth at the depth of -'5 feet, 

 and the mean of the cl.ily extreme.-. \\.thout resting on 

 the n-Milt of IS 14, when the obscrvatio: .maxi- 



mum and minimum were for a time interrupted, that 

 of 1815 ii most sati Our observations on the 



pump water, did not iv. u-l to the conclusion of the 

 year 18 It!, but taking the averages of the \"2 month?, 

 from Julv 1815 to June 18lfi, during which time our 

 register was carried on with the most s,crupulous aUeii- 

 to accuracy, the coincidence is equally striking. 

 The results are as follows : 



In the one case, the average *f tbe wa- tenths 



of a degree lower, and in the other it is five- tenths of a 

 degree higher, than that of the atmosphere, which 

 gives a mean difference of only kulj'a letdk of a degree. 

 Coincident* But it appears that the temperature of t 

 bsUsBur- at any depth, if taken regularly t equal interval*, will 

 ijr at any give a reult, differing very litUe from the mean of the 

 daily extremes in the open air. iii.ice the beginning 

 of 18 18, we have observed th temperature of water 

 isaainff from a pipe, after passing through a distance of 

 about 300 yards, at tbe average depth of 3 feet below 

 the surface, aad fii d that its mean annual temperature 

 ceincidcs very nearly with that of the daily maximum 

 and minimum. The water, before being collected in 

 tbe first or highest cUlero, is brought in different di- 

 rections, and 1'r.nn a considerable distance, in covered 

 ditches, cut for the purpose of draining a Urge field. 

 After issuing from the cistern, it is conveyed by an 

 earthen pipe upwards of 200 yards before it How - 

 the stop cock, where the temperature is taken, and nt 

 every observation, which is made about the 5th, 15th, 

 mmA '/5U of each month, it is allowed to run five mi- 

 nute*, during which time tbe thermometer is found to 

 become stationary. This mode of estimating the ave- 

 rage heat of the intseisr, appears to us to be more ac- 

 curate, than sinking thermometer to the same depth 

 in any given spot, where the fluctuations) of temperature 

 muit be more or less rapid and irregular, according to 

 the nature of the soil. Tbe following are the 

 of our observaiiona. 



establishing this fact we are doing essential service toMeteorolo- 

 the science of meteorology. The <;reat desideratum in _^' _ J 

 thennometrical observations, is uniformity in the me- 

 t' determining the mean temperature, and until 

 such uniformity be introduced, this department of the 

 science will still continue, what it has hitherto been, 

 in a state of infancy. We admit, however, at the same 

 time, that it is a ve'ry interesting inquiry, at what hour 

 of the day the temperature approaches nearest to the 

 true annual mean, and we shall now state the results 

 of certain observations on this point, assuming the ave- 

 rage ef the daily extremes as the standard. 



If the register kept by the Royal Society of London 

 can be depended on, it would appear that the mean of 

 8 in the morning, and 2 in the afternoon, does not dif- 

 fer much from the mean of the maximum and mini- 

 mum. Unfortunately, however, the observations are 

 not always made at the same hours, nor has the self. 

 registering thermometer, which was accidently de- 

 stroyed inVovember 1810, been since replaced. Dur- 

 ie 16 years preceding that, viz. from 179* to 

 1809 inclusive, the annual mean temperature at 8 in 

 the morning, and 2 in the afternoon, was uniformly a 

 little higher than the mean of the daily extremes, but 

 the average difference for the whole period, amounted 

 onlj to half a degree. Could confidence be placed in 

 the accuracy of the observations, the result is a very 

 important one ; but strangely at variance with the opi- 

 nion of a celebrated philosopher, stated in the Philoso- 

 ial Transactions of Kdinburgh, that the mean tem- 

 perature at 8 in the morning, may be supposed to be 

 nearly the true annual mean. 



From our own observations we have found, that the Man tern- 

 average of 10 o'clock, morning and evening, coincides perature at 



th that of the daily extremes. Owing ? j n " 

 to accidental circumstances, we cannot give a compa- ,"**" 

 rative statement of the two for the wnole of every coinci a 

 year, since the commencement of our meteorological nearly 

 register, but the following appear to be sufficient fur with the 

 ig the fact. Taking the means as far as ">en of 

 .is, or one decimal place, the difference between the the c *' 

 average of 10 morning and evening, and that of the tr 

 daily maximum and minimum, is 



0.0 

 O.S 

 0.7 

 0.0 

 0.3 

 0.6 

 0.3 

 0.5 



0.3 



In both cases, the mean temperature of the 

 higher than tl> .iir, but the average 



ree. 



!mg facts, we thii." 



selves ju ' aintainiug. that the avenge 



facts. fa] y extreme 



U-r, may be wore certainly depended on, as giving a 

 near approxin- e true mean temperature, than 



observation* made at any particular Luur.i, and that in 



For 2 months of 1 - 



11 IT ISIS, 



. 



12 months of 1 . 

 8 m.. :slb', 



.nth* of 1817, 

 12 months of 1 . 



7 months of 1 819, (to this date,) 



~n 



being little more than ihrcc-tcntlit of a degree, on an 

 average of 71 months. In the general averages, the 

 of the extremes when there is any difference, is 

 found to l)e always above the mean of 10 and 10 ; but 

 it is not uniformly so, in the monthly average, except- 



i excess was always on the 

 'ii*, that where 

 > are wanting, the mean 

 : i racy, by 

 at 10 



vations me. ' :un's 



11 average 



.in the mean of S in the 

 her than the mean of 8 in 



ing 1M7 .711(1 IM'l. wh< 

 same side. It might 1 

 heif- registering: thermo 

 temperature may be fou 



of 6 y< .--, it 

 morning eingi] 

 the morning ai 



