366 SCIENCE PROGRESS 



whirlpool in the atmosphere, could catch a balloon. 1 have 

 watched this subject carefully since Mr. C. F. Pollock and Mrs. 

 Assheton Harbord made a journey in a strong north-westerly 

 wind from London to the Continent in a free balloon. They 

 were brought to earth in a forest by a squall accompanied by 

 snow which tilted the balloon and loaded it with snow, so that 

 they had to descend in circumstances of great difficulty, fortu- 

 nately without mishap. It seemed difficult to account for a squall 

 catching a balloon and in consequence I have taken advantage 

 of opportunities to arrive at some reasonable explanation of 

 the unpleasant fact that a line-squall can catch a floater. 



We have examined the details of a number of line-squalls in 

 the Meteorological Office (8) and I am now prepared to offer 

 a suggestion as to the course of procedure by which the result 

 is arrived at (9). Let us go back to the maps of a line-squall. 

 We find that the line of the squall is athwart the " following" 

 wind, i.e. the wind behind the squall front ; if we take a 

 section across the line of the squall we have the following wind 

 with its velocity v and the wind in front which is inclined at a 

 finite angle to the following wind ; if we take only the component 

 of the wind at right angles to the squall front, that also appears to 

 be of the same magnitude, viz. v. 



But the line of the squall itself advances with a velocity 

 somewhat greater than the surface wind behind it ; it therefore 

 leaves the surface wind behind and sweeps away the wind in 

 front. The question is, How is this managed ? Let us take 

 into account the undeniable fact that in the upper air, over both 

 front and rear, the wind velocity is greater than that of the 

 surface wind and is, in fact, to be represented at a moderate 

 height by the gradient velocity, which we may take to be about 

 twice the surface velocity. Let us suppose that the velocity of 

 advance of the squall is half-way between those of the surface 

 wind and the gradient wind. Hence we get the state of things 

 representing the motion in the vertical cross-section across the 

 squall line. We have surface wind with velocity v, upper wind 

 with velocity 2V, and the line-squall, marching with a velocity 

 3/2 v, gaining | v on the front and leaving the rear wind behind 

 by the same amount. The air behind is colder than that in front. 



Now clearly the upper air is gaining on the surface air and 

 the crest of the advancing cold wave must topple over, driving 

 up the warm air and producing rain or snow and sweeping away 



