Forecasting the Weather and Storms 289 



this paper Professor Garriott divided 

 the tropical storms of September into 

 three classes, namely : First, those that 

 recurved east of the sixty-fifth meridian ; 

 second, those that recurved between 

 the sixty-fifth and ninetieth meridians ; 

 and, third, those that passed west of the 

 ninetieth meridian or reached the United 

 States without a recurve. Of the first 

 class of storms, all of which first ap- 

 peared east of the fiftieth meridian or 

 north of the twentieth parallel, Professor 

 Garriott observes that only two appeared 

 far enough to the south to render their 

 advance over or near the West Indies a 

 probability, and that in every instance 

 the westward movement of the cyclones 

 which recurved east of the sixty-fifth 

 meridian was apparently prevented by 

 anti- cyclonic areas which moved east- 

 ward over the Southern states and ob- 

 structed the westward advance and 

 forced a recurve to the northward. He 

 states that the recurve of storms of the 

 second class — i. c. , those that recurved 

 between the sixty-fifth and ninetieth 

 meridians — was apparently due to the 

 obstruction offered to a westward course 

 by anti- cyclonic areas which had ad- 

 vanced or had been drawn from the 

 continent over the west Gulf and South- 

 western states. A large proportion of 

 the third class of storms advanced 

 westward from the eastern West In- 

 dies. On their arrival in about longi- 

 tude west 80 degrees, the average lon- 

 gitude in which September tropical 

 storms recurve, the pressure over the 

 west Gulf began to decrease and rain 

 set in, while the interior eastern districts 

 of the United States were occupied by 

 an extensive area of high pressure. As 

 storms prefer to follow the path of least 

 resistance, the centers moved toward 

 the region of decreasing pressure and 

 avoided the high and increasing pressure 

 to the northward. When the pressure 

 continued high over the eastern districts 

 of the United States the storms were 

 unable to recurve, and were penned in 



over Mexico or the Southwestern states. 

 In such instances, Professor Garriott 

 states, the cyclones developed great vio- 

 lence before disappearing. Similarly 

 cyclones of this class that advanced 

 northwestwardly toward the Middle or 

 South Atlantic coast of the United 

 States were apparently prevented from 

 recurving by high pressure over the 

 ocean to the northward and northeast- 

 ward, and, being forced upon the coast, 

 developed destructive energy. 



From the foregoing it appears that 

 the effect of distribution of pressure in 

 determining a storm's path is recognized 

 in practical forecasting. 



NEW METHOD FOR DETERMINING 

 THE DIRECTION AND VELOCITY 

 OF STORM MOVEMENT 



Local Forecaster Edward H. Bowie, 

 in charge of the local office of the U. S. 

 Weather Bureau at St Louis, Mo., has 

 devised a new method of estimating the 

 future course and rate of translation of 

 storms, which, while not being absolute 

 in its determinations, is a marked ad- 

 vance over anything heretofore accom- 

 plished in this direction. The unusually 

 high degree of accuracy that has at- 

 tended Mr Bowie's forecasts for the past 

 several years attests the value of his sys- 

 tem. Storms follow the lines of least 

 resistance ; but the trouble is that with 

 the movement of vast systems of air, 

 due to the excessive heat of the equator, 

 combined with the rotation of the earth 

 and the continual breaking up of the 

 currents on the outer edges of these sys- 

 tems into cyclonic or anti-cyclonic vor- 

 tices, the lines of least resistance are 

 always changing, sometimes slowly and 

 again rapidly. The usefulness of Mr 

 Bowie's work lies in the fact that while 

 some of his values are but roughly as- 

 signed he has been able by a study of 

 the pressure gradients about the base of 

 the storm, in connection with the gen- 

 eral drift of the upper air, to obtain a 

 resultant that approaches with close pre- 



