312 Mr Atherton's Description of Machinery for 



regret that satisfactory evidence cannot be obtained of the peculiar 

 combination and changes of circumstaHces which preceded and 

 gave rise to such remarkable events. 



It is peculiar to such events, that, not being subject to perio- 

 dical recurrence, they cannot possibly be foreseen ; and, similar 

 to the progress of a disease, its approach is only known sub- 

 sequent to the period at which the creating causes commenced 

 their operation, whereby the vigilance of practical observers is 

 rendered abortive, the result of their labour being merely a 

 minute historical record of facts, conveying such an insufficient 

 account of the circumstances from which those facts resulted, as 

 to give but little aid to the genuine advancement of science. 



The mode usually adopted of registering natural phenomena, 

 consists of noting down the state of things at particular periods 

 of the day, best suiting the general convenience of the observer. 

 Were the changes of natural phenomena subject to periodical 

 occurrence, this system, diligently attended to, may be suffi- 

 ciently satisfactory ; but, in the existing state of nature, nume- 

 rous circumstances, deeply involved in the production of subse- 

 quent remarkable events, pass unheeded. 



For instance ; the wind is frequently known to blow from all 

 points of the compass in the space of a ^^^ hours. This must ne- 

 cessarily create strange atmospherical disturbances in some shape 

 or other; and yet, should it happen during the night, or if the 

 attention of a recorder of atmospherical changes be otherwise 

 demanded for a few hours, the event would not appear upon his 

 record, which omission would probably lead to false conclusions 

 as to the origin of a subsequent matter of investigation. Thus 

 the laborious researches and happy conclusions of theoretical 

 speculators may frequently be condemned as useless, upon the 

 fallacious testimony of pretended practical observations. 



Natural phenomena are so constantly in action, and subject to 

 such incessant fluctuations, that the most persevering diligence 

 of man is scarcely competent to furnish data sufficiently com- 

 plete to form the basis for a standard rule, or test the truth or 

 fallacy of theoretical deductions. 



The numerous difficulties attending the present system of re- 

 cord, and the objections to which it is liable, are too evident to 

 require further remarks in order to shew the comparative advan- 



