INSTRUMENTS AND METHODS. 



17 



tho dry bulb thermometer as already described. These daily comparisons gave a good check 

 on the behaviour of the thermometers and the results are shown in the following table, 

 in which the mean values found by the daily comparisons from February, 1911, to February, 

 1912, are collected, the numbers in brackets giving the number of comparisons involved. 



Table 2. 



Thermometer Coinjxiritions. 



The mercury dry bulb thermometer No. 8474 was in use throughout the expedition 

 and was the standard for the reduction of the screen thermograph. The above comparison 

 shows that it and the ma.ximum tliermometer No. 3136 had retained their Kew corrections. 

 On the other hand the spirit minimum thermometer had changed its correction by -f- 03. 



In practice the method of correcting the maximum and minimum thermometers was 

 to consider the dry bull) thermometer No. 8174 to be correct, then at the end of each month 

 the mean corrections of the other two thermometers were determined from the daily com- 

 parisons during the month and applied to the individual readings. 



After February, 1911, the daily comparisons were not made and there were several changes 

 in the maximum and minimum thermometers used. The dates of the changes were not al- 

 ways recorded, and there are other small uncertainties. After a careful f-tudy of tho records 

 the appropriate corrections to apply throughout have been determined, I believe, with a fair 

 degree of certainty, but as some doubt still remains the maximum and mininuim temperatures 

 during the second year may be in error by ± -5° F. 



Method of Measuring Temperatures on Sledging Journeys. — Experience on previous expedi- 

 tions had taught that there are considerable practical difficulties when the air temperature 

 has to be measured on sledge journeys. As it is quite impossible to transport a suitable 

 screen some form of artificially ventilated thermometer becomes a necessity. An Assman'a 

 ventilated psychromoter would appear at first sight to meet the case, but these instruments 

 were tried on Captain Scott's first expedition and found to be quite unsatisfactory. The actual 

 difficulties will be found described in 'The Voyage of the Discovery,' Vol. I, page 276; it 

 is sufficient here to mention the impossibility of getting clockwork mechanism to work in low 

 temperatures, and the appreciable weight of any but the smallest pattern, which are too 

 small for accurate reading under the difficult conditions met with in the Antarctic. Some 

 form of the sling thermometer was the only alternative. The simple method of swinging 

 a thermometer on the end of a string becomes practically impo.ssible when one has to work 

 in thick fur mitts, and in bulky clothes nearly as stiff as thin sheet metal owing to the 

 frost. After considerable thought and experiment a form of sling thermometer based on a 

 very old principle was designed, which proved to be admirable in every way when put to 

 the severe test of continued use on sledging journeys. 



