19 
ner, as are equal points of barometrie pressure, and the configuration 
of the figures so obtained indicate the different areas of high and low 
pressure. In practice Mr. Nowack uses blue for all points of higher, 
ic; it s 
As a matter of fact it often occurs that points of high or low pressure 
fall on the chart near together. Suppose, for instance, points of low 
charts are c rised by t igure 
different areas, and by the relatively large number of his primary and 
subsidiary centres. As I have said above, reference must be made to 
Mr, Scott’s report as to how far they represent the real state of affairs. 
I said that on the average the charts were drawn three days before the 
event, 7.e., a chart drawn from observations at noon on Tuesday should 
ee the actual distribution of pressure at noon on the following 
Friday. 
As a matter of fact, though Mr. Nowack professed himself often 
satisfied as to the agreement of his charts with the actual barometric 
the real chart of the day for which they were drawn up. When. this 
happened it was necessary to change the order of the charts, and » 
assume that for the day in question the plants had not indicated three 
days ahead but two, four, or five days ahead, and the charts would be 
PE so as to fit in most accordantly with the barometric | 
charts. pe. 
In speaking of the weather, I explained that local weather is not 
foretold with regularity, as described in his pamphlet, two days > 
but at intervals of one, or three or four days ahead, as dict west 
previous experience. Thus, usually on Monday the plant in Lo 
weather for Wednesday and barometric pressure for Thursday; <4 A. 
ever, the chart made from the plant for Thursday does -—— ‘da " 
actual Thursday's chart, that chart is regarded as being that tor y y» 
