[M'LEOD & HARNEs] ON THE DIFFERENCE IN TEMPERATURE 143 



sloAv travelling of the pen to the iright from the warming of the upper 

 surface of the cable by the sun. 



It is impossible to determine which of the wires was located nearer 

 to the top of the cable, but it is possible that either the main leads or tha 

 compensating leads run for a greater length near the upper part of the 

 cable, which certainly detracts from the full value of the apparatus 

 for temperature work. 



The present cable has 11 wires, 6 of which are used for the anemo- 

 graph connection and four for the temperature records. It is very 

 much to be desired that a separate lead covered cable having only four 

 strands be procured for the temperature work. The insulation should be 

 rubber and braid and the wires should be twisted m such a way as to 

 bring all equally to the upper and lower portions of the cable. Since 

 owing to lack of funds such a cable is quite out of the question at the 

 present time we decided to go on with the work, utilizing the existing 

 cable. Moreover the ditîerences which we had previously measured were 

 so large that an uncertainty in the leads of even a degree Fah. was of 

 secondary importance. 



We were specially interested in the daily variations so we decided to 

 pay less attention to the absolute readings for the present. The mag- 

 nitude of the difference between the temperature on the moimtain and 

 observatory was of less importance than the relative differences from 

 hour to hour. At the same time special watch had to be kept on a pos- 

 sible shift of zero. 



After devoting about two months to obtaining a series of zero traces, 

 the thermometers were put in place and on Jan. 9, 1-905, the record of 

 differences was resumed. 



Since the resistance of the leads is about 40 ohms and that of the 

 thermometers is only 10, special care had to be taken to watch the per- 

 mancy of the former. 



The Absolute Value of the Temperature Differences. 



We found that the differences were not so large as we had recorded 

 in our previous winter's work. JSTo difference was observed at any time 

 during the winter greater than 7 degrees Fah., while in our previous 

 winter's work we had obtained differences amounting to 20 or 25 degrees 

 as will be seen by referring to the charts published in our previous paper. 



This can hardly be explained at the present time without a much 

 greater study of the influence of climatic conditions, which it is hoped 

 that furtlier records may throw light upon. In general it may be said that 

 the past winter was not so severe in the North, just above Montreal, to 



