178 



MR THOMAS HEATH ON 



The new thermometers were similarty treated for the twenty years' observation avail- 

 able at the end of 1899. It is a somewhat unfortunate circumstance, however, that 

 there exists no record of the relative capacity of the different parts of the tube of which 

 these new thermometers are made, so far as could be discovered after enquiry. This 

 appears to be due to the lamented deaths of Mr Richard Adie, the head of the firm 

 of Messrs Adie & Sons, and of Mr Thomas Wedderburn, who was their responsible 

 manager when they held the contract for the construction of the thermometers. It is 

 known, however, that the new instruments were made with, as nearly as possible, the 

 same size of tube as the old ones. We may therefore be considered justified in 

 supposing that the relative capacities of the capillary parts and the wide parts of 

 the tubes are, approximately, the same in the new set as they were in the old, 

 and that the only difference of importance is to be found in the different lengths 

 of the two sets. By far the greater part of the total corrections applied to the old 

 readings was that depending on the difference of temperature of the bulb and of the 

 wide part of the tube above ground, or the correction for the temperature of the air, 

 and only a small part, never more than 0*03 of a degree F. for the longest thermometer, 

 was due to the temperature of the stem. The corrections to be applied to the 

 new set may then without risk of important error be assumed the same as those of 

 the old. 



The following table gives the values of the monthly mean temperatures and the 

 equi-mensual dates for the new thermometers from the twenty years' readings, 1880-1899. 



NEW THERMOMETERS. 



Equi-meusual Dates. 



h 



h 



*3 



U 







°F. 



°F. 



°F. 



°F. 



January, 



3-95 



47-097 



45-788 



42-082 



40-128 



February, 



3-39 



46-767 



44-597 



40-949 



39-465 



March, 



5-57 



46-343 



43-780 



40-534 



39-421 



April, . 



5-01 



45-920 



43-314 



41-154 



40-748 



May, . 



5-45 



45-553 



43-470 



43-211 



43-853 



June, . 



4-89 



45-352 



44-395 



46-382 



48-045 



July, . 



5-32 



45-405 



45-873 



49-545 



51-638 



August, 



4-77 



45-716 



47-341 



51-170 



52-815 



September, . 



4-20 



46-198 



48-358 



51-158 



52-295 



October, 



4-64 



46-700 



48-681 



49-815 



49-387 



November, . 



. 4-08 



47-088 



48-149 



46-963 



45-541 



December, . 



4-51 



47-238 



47-273 



44-386 



42-477 



T = 



46-2814 



45-9182 



45-6124 



45-4844 



From the numbers in this table the equations to the curves for the new thermometers 

 have been formed. 



In the reduction of the readings of the new thermometers, the corrections have 

 not been applied to the single readings, nor to the monthly, quarterly, and annual means 



