ABOUT THE 20tH OF DECEMBER, 1836. 



143 



racter is wholly changed. It would seem, then, that even if the south-east 

 wind could continue its course over this mountain barrier, it could carry no 

 rain with it. So far, then, as rain, at least, is concerned, the Rocky Mountains 

 must be supposed the western boundary of the storm. 



I am unable even to conjecture a probable limit of the storm on the north. 

 By inspecting the table on page 130 it will be seen that, with certain excep- 

 tions to be hereafter considered, the oscillation of the barometer increased pretty 

 uniformly from the most southern to the most northern station. The range 

 was greatest of all at Quebec. It would seem natural to infer that Quebec 

 could not have been south of the centre of the storm. Why should not the 

 storm have extended as far to the north as it did to the south of this station? 

 On this supposition the northern limit would be found near the arctic circle. I 

 have been unable to obtain observations of this date from any higher latitude 

 in North America. As some guide, however, to my conjectures, I have insti- 

 tuted a comparison between the barometric observations made by Capt. Parry, 

 at Melville Island in 1819-20, and similar observations made at New Haven. 

 The following table exhibits the times and amount of all the barometric max- 

 ima observed at Melville Island which reached the height of 30.2 inches. Op- 

 posite these times I have given all the instances in which the barometer at 

 New Haven rose to 30.1 inches within the ten days subsequent. I have also 

 added another column, giving all the instances in which the barometer at Lon- 

 don, by Professor Daniell's register, rose above 30 inches within the fifteen 

 days subsequent. 



Melville Island. 



New Haven. 



London. 



April 



27, 



30.86 



April 



29, 



30.27. 



[nterval 



2 clays. 



May 



1, 



30.33. 



Interval 



4 days. 









*May 



5, 



.35. 





8 





May 



12, 



.05. 





15 





Jan. 



16, 



.77 



Jan. 



21, 



.22. 





5 





Jan. 



22, 



.07. 





6 











Jan. 



23. 



.13. 





7 





Jan. 



29, 



.09. 





13 





Dec. 



30, 



.75 



* Jan. 



7, 



.32. 





8 





** Jan. 

 Jan. 



9, 

 14, 



.59. 

 .12. 





10 

 15 





Sept. 



3, 



.42 



Sept. 



6, 



.12. 





3 





Sept. 



9, 



.05. 





6 











* Sept. 



12, 



.43. 





9 





Sept. 



13, 



.23. 





10 





May- 



14, 



.41 



May 



16, 



.10. 





2 





May 



21, 



.32. 





7 











May 



19, 



.10. 





5 





















May 



21, 



.10. 





7 















Dec, 



14, 



.40 



Dec. 

 Dec. 



18, 

 22, 



.25. 

 .20. 





4 



8 















Jan. 



10, 



.35 



Jan. 



16, 



.30. 





6 





Jan 



14, 



.12. 



a 



4 " 





















Jan. 



22, 



.07. 



a 



12 ' 



i 



