METHODS OF FORECASTING THE WEATHER. 165 



assertions should prove them. It was the professional meteorologists 

 themselves who undertook the accurate exanunation of all the various 

 cosmical hypotheses, and particularly that of the influence of the moon, 

 and it was they who found a slight influence of the moon on storms, 

 thunderstorms, the direction of the wind, atmospheric pressure, etc. 

 Now, do vou say, "1 told you soT' Well, tir.st of all, observe- — and 

 I can not in.sist upon it too strongly — that it is the professional 

 meteorologists, and they alone, who have made these investigations 

 which point to a slight influence of the moon. Next, I must direct 

 your attention to that little word "slight." The influence thus dis- 

 covered by them is indeed so small that we can not even state with 

 certaint}" whether it really does exist at all; or whether, perhaps, 

 it was only perceptible in these investigations because the period 

 of time included in them is still too short to furnish us with an 

 unexceptionable result. However, let us assume that this slight influ- 

 ence really does exist, and let us examine the amount of this influence 

 a little more closely. Its magnitude is expressed by the percentages 

 of the favorable cases. We will, however, for once greatly exaggerate 

 and assume that these favorable cases amount to a surplus of 6 per 

 cent. That is to sa}', that in 100 cases 55 succeed and 45 fail. Now, if 

 you use such lunar rules for weather predictions, what does it advan- 

 tage you in isolated single cases? For instance, 3^ou are in doubt as 

 to whether the rain is to be expected or not; the influence of the moon 

 indicates rain with a weight of 0.05. In spite of this small weight, if 

 now you forecast bad weather, 3"ou will, if 100 such cases occur, have 

 a failure in 45 cases. Had you paid no attention to the influence of 

 the moon you would possibl}^ have had 50 failures. Thus, in this case 

 of 5 per cent of surplus, that would be the whole efl'ect of your con- 

 sideration of the moon's influence. But we have in fact assumed an 

 exaggerated case, and the real influence of the moon is in every case 

 less than one-half of this, if indeed it reall}^ exists at all. 



You may rest assured that the professional meteorologists accept, 

 nay, even seek for, everything that can give them any assistance what- 

 ever in their weather predictions. By constant investigation and study 

 we ma}' hope to advance step by step and per cent 1)}' per cent. Every 

 single per cent of agreement that is gained is an important advance 

 and success. 



