[mleod a BARNES] ON THE DIFFERENCE IN TEMPERATURE jl45 



servatory. On February 21st, a cold wave was indicated 5 hours, 

 in advance as sho"\vn by an increasing negative difference. On February 

 v'ord it warm wave was observed to be approaching 12 hours before any 

 change occurred at the lower station. On February 28th, a rapidly ad- 

 vancing warm wave swept across the high level thermometer 5 hours 

 ahead of any change on the observatory thermometer. 



The extent and' duration of these ditierences would appear to 

 indicate that estimates can be made for much longer periods ahead. 



On two occasions when a heavy fall of snow occurred the precipi- 

 tation and consequent heat development in the upper atmosphere was 

 noticed on the upper thermometer causing a decided trace far over to 

 the positive side of the zero. 



It is evident that much further study must be made in which direc- 

 tion and velocity of the wind must be considered together with other 

 climatic conditions. 



Enough has been learned however, to make it of interest to carry on 

 the woric in the hope that useful material may be gathered which will 

 aid local meteorological temperature forecasting. 



Part II: — (presented May, 1906.) 



Since the preparation of the previous note which was too late for 

 publication in the Transactions of last year, we have obtained another 

 set of records during the winter just past (1905-06). 



We anticipated further trouble in our long distance cable with the 

 advent of the warm weather during the spring of 1905. ISTone 



developed, however, until the middle of June, when the insulation became 

 insufficient for maintaining accurate records. On account of the absence 

 during the summer of one of the authors, who has devoted special 

 attention to the cable testing, no attempt was made to rectify the fault 

 until the autumn, when the repairs having been completed, and tests 

 made, the insulation was again found almost perfect, and the apparatus 

 started. 



Zero traces were obtained with both thermometers in the circuit at 

 the observator}', in a similar manner to the tests described in Part I of 

 this paper. 



The records obtained show an average monthly difference of tem- 

 perature in degrees Fahrenheit at the two stations as follows: — 



December, 1905 —.3.0 (Lower Station warmer) 



January, 1906 —2.2 " ■-' 



February, " —1.6 



March " —2.5 



April ' —3.7 



Sec. III., 1906. 10 



