MR DALMAHOY ON A DIFFICULTY IN THE THEORY OF RAIN. 35 



lower gauge ; and one or two circumstances which might tend to produce such 

 an equality may be mentioned. If, for example, the upper gauge were placed 

 rather close to the top surface of a tower or building, and nearer to one side of it 

 than to the other, then, in accordance with Professor B ache's experiments, when 

 the wind blew from the direction of the more distant side, the indication of the 

 gauge might be expected to be greater than if the wind blew from the near side ; 

 and thus it would not be surprising if, in certain directions of the wind, the indi- 

 cation of the upper gauge became nearly or quite equal to that of the lower one. 

 Again, there are two other causes, opposite to each other in character, which, it 

 might be expected, would tend to equalise the indications of the upper and lower 

 gauges. The first of these is when the rain falls, more or less plentifully, but in 

 extremely small drops ; and the second is when the rain falls in large drops and 

 very copiously. In the former case, the very small terminal velocity of the drops 

 would give rise to a downward current of air so slow as scarcely at all to hinder 

 the entrance of the minute globules of water into the upper gauge, or accelerate 

 their coalescence into drops as they approached the ground ; and the ultimate 

 effect would be, a tendency to equalise the indications of the two gauges, by 

 allowing moi^e than the usual quantity of water to enter the u2Jj)er gauge. In 

 the latter case, again, the downward current of air might be so strong as to reach 

 the level of the ground, and this also would tend to equalise the indications of 

 the two gauges ; but, on this occasion, it would do so by allowing less than the 

 usual quantity of water to enter the lower gauge. 



Having explained what I conceive to be the twofold agency of the downward 

 current in producing the paradoxical results under consideration ; and having 

 also noticed some of the causes which may serve occasionally to modify these 

 results ; it would now have been desirable to test the hypothesis quantitatively ; 

 but unfortunately the want of data renders it impossible to do so in a satisfac- 

 tory manner. There is, however, one point on which a numerical estimate, even 

 if it were only a probable one, is essential. I mean respecting the quantity of 

 w^ater, in theshape of minute globules, which a given volume of the atmosphere 

 must be assumed to contain, in order to account for some of the more remarkable 

 results recorded at York or elsewhere. I shall, therefore, now endeavour to make 

 a rough approximation to this quantity, the chief object in view being to deter- 

 mine whether the quantity of moisture would be so great as to render necessary 

 the supposition of visible cloud throughout the space where rain is falling. 



In selecting a case for such a purpose, recourse cannot be had to those rare in- 

 stances in which 30 inches or more of rain fell in the course of twenty-four 

 hours ; for in none of these does it appear that the observations were made at 

 more than one level. 



Perhaps the winter observations at York will furnish as severe a test of the 

 hypothesis as can be found, at least as regards the ratios of the quantities of rain 



