Causes of the Aurora Borealis. 235 



It is therefore evident that the load was not applied precisely at 

 the centre point, but upon a surface six inches long, the centre of 

 which corresponded with the centre of the bearing length. 



In every instance when the beam was broken, the fracture took 

 place at one end of the bar (b), and therefore at the distance of three 

 inches from the true centre. 



In the trials relating to stiffness, tables (A) and (B), it is not sup- 

 posed that this bar modified the results in any appreciable degree, 

 that is to say, the results given in these tables may, without any sen- 

 sible error, be considered as having been obtained by applying the 

 weight precisely at the centre. The case is different, however, with 

 the results given in table (C) ; but since the error, whatever it 

 may be, (and that it must be small is evident,) does not affect the 

 comparison of the woods, it is of no consequence as to the partic- 

 ular object had in view. 



Some of the general inferences to be drawn from the foregoing 

 experiments are : — 



1st. That within certain limits which extend much further than it 

 is proper in practice to suffer wood to bend, the deflexions are very 

 nearly as the weights producing them ; see table (B), column (g). 



2nd. When we have a given weight to sustain, so that the deflex- 

 ions shall not exceed ^^V of an inch to one foot, other circumstances 

 being equal, ten beams of southern pine are equivalent, nearly, to 

 thirteen of spruce or to fifteen of white pine. When the deflexions 

 to which the timber is exposed are very much less than ^'^-th of an 

 inch to one foot, or when we merely wish to guard against fracture, 

 the disproportion between southern pine and the other woods, is much 

 greater ; see 2nd and 1st lines, table (D) and column (A), table (C). 

 It was to obtain the results given in this paragraph, that these experi- 

 ments were instituted. 



Art. V. — On the Causes of the Aurora Borealis. 



The sublime and beautiful spectacles presented by the aurora bo- 

 realis, have been in all ages a source of admiration and wonder alike 

 to the savage and philosopher. The ignorant and superstitious view 

 the still mild northern light as a precursor of famine, and the brilliant 

 crimson and other colored flashes, as ominous of pestilence and war ; 



