TRANSACTIONS OF THE SECTIONS. 



29 



ness of the bad conducting rock between the lowest of our thermometers and any 

 part of the earth where its substance can be fluid with heat, and be capable of 

 assuming more sudden changes of position or temperature than a solid can, we may 

 safely in a first examination consider the internal or terrestrial effect as constant at 

 each depth for the whole period from 1838 to 1854. The effect is small, but veiy 

 sensible, as thus : — 



Mean of each Thermometer for the whole period, from 1838 to 1854. 



ti 3-feet thermometer 46*27 



^3 6 „ 46-55 



ts 12 „ 46-94 



ti 24 „ 47-24 



where we find each thermometer to tell the same story of, and to point to, a heated 

 terrestrial centre, even by approaching so small a space as 3 feet ; and on the whole 

 they show an increase of 1" Fahr., with 21 feet of difference of depth for the in- 

 ternal influence, or the terrestrial source of surface temperature. Subtracting the 

 differences between (4 and the other thermometers from each in turn, we obtain the 

 following Table, wherein the terrestrial effect being eliminated, the variations from 

 cosmical influences become more apparent : — 



If these numbers be projected with the times, the curves they form are most in- 

 teresting, for they contain appearances of periodical waves distributed over a secular 

 swell, with so long a period, that only a small portion of it appears in the seventeen 

 years. 



If then we can depend on our observations being strictly cleared of every instru- 

 mental and terrestrial cause of disturbance which can logically affect their accuracy, 

 we have at once an indication of our sun being amongst the number of variable 

 stars. Can we then depend on them to this extent? The only possible room that 

 I can see for doubt, is the question of the constancy of the zero-points of the thermo- 

 meters ; and having no means of inquiring into this practically, I can only combine 

 with the general experience of the unalterability of spirit-thermometers, with the 

 very thick glass bulbs and tubes here employed, after a certain period, the particular 

 observations by Professor J. D. Forbes on a thermometer made at the same time as 

 our set, and in the same manner, and filled with the same spirit. The result of 

 examination was, that after nine years no appreciable change (certainly not -J^jth 

 of a degree) was found. 



This is very satisfactory ; and if further evidence be required that there is some 

 natural and cosmical cause acting on our thermometers, tending to produce an effect. 



