Causes of Water Spouts. 65 



brings with it all the vapors that are above toward the centre of the 

 primitive whirlwind. 



Thus, in the fourth experiment, (Fig. 3.) the descending spout of 

 oil, and the ascending one of blue dust, mingle in the zone which 

 passes through the centre of the mill. In asc'ending spouts, below 

 the clouds, the spectator can see only half the phenomenon, as the 

 other half passes above the clouds; by this mechanism, the lowest parts 

 of the atmosphere and those which are elevated above the clouds, 

 are drawn to the same point by the two opposite interior currents of 

 the spouts ; these strata of air, charged with vapour and often with a 

 different kind of electricity, unite, and produce the thunder and other 

 effects which sometimes accompany water spouts. 



In descending spouts, the spectator sees the entire phenomenon, 

 because the whirl which produces it, is formed between the clouds 

 and the sea. 



.The water, under descending spouts, is generally seen to rise and 

 boil, even, while the aerial current is still far from the sea ; sometimes 

 the two spouts join and form but one ; the primitive whirl is then at 

 the point where they join ; but if they remain separate, it is because 

 neither the one nor the other reaches the primitive and invisible whirl- 

 wind which produced them. 



One circumstance in the 2d experiment presents another remark- 

 able analogy ; it has been seen that sand produces an empty spout, 

 into which clouds of fine sand are gradually forced ; similar observa- 

 tions of empty spouts have been several times made by travellers, 

 who always represent them as opaque on the borders, and transpa- 

 rent in the middle.* In a description cited by Mr. BerthoUon, se- 



* What is the force which retains the particles of sand and water in the regular 

 "rcle in which they revolve as rapidly as if they had a centre of attraction ? The fol- 

 lowing explanation may, I think, be given, both as to the imitative experiment, or 

 the natural phenomenon. 



No portion of the surface can issue from the circle which it traverses without be- 

 ing replaced ; now the surface of the spout having throughout its height, the same 

 circular movement, cannot remove itself from its axis, without occasioning a vacuum ; 

 this surface is then retained by the pressure of the lateral columns. 



The replacement cannot take place in the interior of the spout except at the infe- 

 rior end of the axis, which is not submitted to the centrifugal force, and an equili- 

 brium is there established, between the energies of the centrifugal force, and the 

 weight of water withdrawn, and which determines the diameter of the spout. 



This explanation applies also to the descending spouts ; it is known that they never 

 have the regular and cylindrical form of the first, but rather that of a spindle or of a 

 cone widemng towards the clouds, and terminating in a point near the bottom. Here 



