T. Russell — Prediction of Cold-waves. 475 



Tables and charts have been prepared showing the lowest 

 and highest 7 a. m. temperatures that have occurred in the 

 months of November, December, January, February and 

 March. These are of use in locating the areas of fall and 

 in estimating by differences the low temperatures that may 

 occur to the east, by comparison with what has already occurred 

 where a cold- wave is prevailing. 



There is very little time for extensive or elaborate computa- 

 tions in the work of predicting cold-waves. This fact has 

 been borne in mind in devising the method. It will not 

 require more than half an hour to apply it in any particular case. 

 Plan i meter measurements of the extent of high and low areas 

 of pressure were resorted to in the special cases used in deter- 

 mining the constants of the formula. But this is not 

 necessary in determining the extent of an area of low pressure 

 in the prediction of cold-waves. The area computed from the 

 measured lengths of the greatest and least axis of the outside 

 isobar of the area will be sufficient. Considering this area as 

 the base of a cone, and the altitude as the difference between 

 the outside isobar and the lowest barometer reading, its con- 

 tents can be computed with sufficient accuracy for the purpose 

 required. 



The predictions of cold-waves according to this method will 

 be better than those of the past in that part of the country 

 south of the States of Missouri and Kansas and south of the 

 Ohio River. Not much improvement can be expected in the 

 far northwest, where there is no opportunity to measure the 

 extent of the " high," where it is apt to be over a country not 

 covered by observation. Neither can the method give much 

 improvement in the New England States, where the " low " is 

 often out over the ocean and the "high" to the north of the 

 Dominion of Canada. The use of the rule for computing- 

 maximum fall will, however, make some improvement in the 

 predictions for these regions. Though only adapted for giving 

 the temperature at the place of greatest fall, it can nevertheless 

 be used for other places, and will give a value for the fall that 

 will certainly not be exceeded. 



Fuller details for the use of this method in prediction of 

 cold waves will be found in the annual report of the Chief 

 Signal Officer for the year 1890. 



