METALS TN THE ATMoSPHEKK. 245 



1)()1('. pyroxeno. varioiislv associalc^l. All of the itu'tcoric iniiuM-als 

 can be foiiiul in theuliists of the atiiiosphciv. 



Meteors iDay fall in a direction oi)posite to that of the eai'th's nio\-e- 

 ment, in which case their relative speed, heinii' the snni of the two 

 movements, will he \erv y-reat, perhaps TO km. a second. The resist-. 

 ance of the air to a fliifht of snch s])eed produces enough heat to com- 

 pletely burn and \-()latilize the matter. If. on the contrary, the move- 

 ment of the falling- mass is in the same direction as that of the earth, 

 its relative velocity — the diffei'ence between the two al)solute speeds — 

 is scarcely K) km. a second. In this case the heat developed is suffi- 

 cient only to fuse the mass and vitrify its surface, then perhaps to 

 l)reak it in such a way as to form a meteorite or aerolite. It is not 

 sufficient to make a shooting star with its great train of fiery particles. 

 Consequently, in coining into the earth's atm()s})here, aerolites, whether 

 big or little, encounter a friction that generates heat and incandes- 

 cence, c()nse(}uently combustion, fusion, volatilization: condensation 

 of the volatilized [)articles follows, and the dissemination of these con- 

 densed i^articles. Thus it is easy to understand how meteorites bi'ihg 

 into the atmosphere xarious metals, free or in combination, and why 

 metallifei'ous minerals, in coi"])Uscles so minute that it is impossible 

 to separate and identify them, may he found in the air entirely inde- 

 pendent of the j)ul\'erized minerals raised from the earth's surface. 



Mattel' ,sohihh' in inttci-. — As already mentioned, the "cinders" of 

 atmospheric sediments, when tivated w ith water, always ^deld a cer- 

 tain amount of S()lul)le salts. 'Miese are chlorates, alkaline sulphates, 

 or calcium sul])hate. and nitrates, particularly '.hat of annnonium. 



A III niDii'/K III lutratc. — A drop of rain allowed to e\aporat(' spon- 

 taneously on a bit of glass leaves crystals of various shapes on its 

 borders as the corpuscles are ilrawn toward the center. A star 

 with six points is well marked when the ci'vstallization has taken 

 ])lace slow!)-; more rarel_y the crystals assume the j)lunnde form. 

 Annnonium nitrate freciuently forms remai'kable grouj^s of ciwstals 

 in the sha[)e of crosses and swoi-ds, like those obtained by e\aporating 

 a drop of snow water. In no other nunnu'r can similar cr\stals be 

 obtained, neither by \arving the solution of the salts noi- the method 

 of ev^aporation. It will form only in regular ci'ystals ramifying 

 uniforjidy from a common stem oi- else in isolated prisms. Tissan- 

 dier attributes this pec-uliai- crystallization in meteoric watei- to some 

 organic nuitter dissolved in the rain oi- snow. The e\aj)ornt ion of 

 this water lea\es in the bottom of the N'essel a hai'd. fragile residue 

 somewhat similar in appearance to coagulated albumen. C'rvstals 

 of annnonium niti'ate are easily recognized bv their solubility in 

 alcohol and by the fact that heat decomposes them without residue. 

 Their presence in the air can not b(» verified, since, as everyone know's, 

 nitric acid and annnonia unite readily to form annnonium nitrate. 



