86 On the 0Hgi7i of Meteoric Stones. 



motion. In the forming mass, all these movements combine to- 

 gether in a simple impulse, the direction of which may vary inde- 

 finitely, according to the position the vapours may have before and 

 during the reduction. For example, should they, before their 

 reduction, have arranged themselves in a vertical column, the 

 ignition may proceed either from above or below, and conse- 

 quently, the mass may move either perpendicularly upwards or 

 downwards. The same explanation, it is obvious, may apply to 

 all the cases. 



In the first number of Gilbert's Annals for 1817, p. 91, 

 there is a paper by Chladni in which this indefatigable inquirer 

 again mentions a number of observations, whence it indubitably 

 appears that fire-balls have often a bounding motion, similar to 

 rockets. This appearance becomes conceivable when we reflect 

 that, in the perfect state of ignition, affecting the whole mass, 

 new decompositions and internal changes may proceed without 

 interruption. If, during the first reduction, (which is not an 

 effect of heat but of electricity, so that the ignition of the meteor 

 is not the cause but the consequence of its formation), substances 

 have been found internally which are susceptible of being vola- 

 tilized by heat, explosions may take place, by which either the 

 whole, as is frequently the case, will be dispersed; or, as also 

 happens, the exploding matter issuing from one or more aper- 

 tures through which it forces itself, will assume the appear- 

 ance of streams of fire. But such explosions must, at the same 

 time, also change the direction of the motion, and when many 

 succeed each other in a short interval, such a bounding motion 

 may well be supposed to arise. In the April number of Gil- 

 bert's Annals, for 1818, page 299, Chladni mentions a fire-ball, 

 accurately observed and described, of 17th July 1771, which, 

 after descending, exploded, and then rose anew. 



But the most singular circumstance appears to me to be one 

 which Chladni has published in the Annals, for 1817, No. 1, 

 p. 96, which redounds very much to the honour of his unbiassed 

 love of truth, since it is not very favourable to his hypothesis of 

 the origin of meteoric stones. He remarks, whenever any one 

 has had an apportunity of observing such a phenomenon from its 

 first commencement, it has generally, at the place, '^een preced- 

 ed by a very extensive flash oi light, of which a striking exam- 



