478 PROCEEDINGS OF THE AMERICAN ACADEMY 



did not crystallize, its opacity, as seen when magnified two hundred 

 diameters, being due to a dense aggregation of exceedingly minute 

 bubbles. Upon treatment with microcosmic salt, lime and magnesia 

 gave, with very slight flaming, nebulous films ; that of the former 

 being resolvable into a multitude of spicular crystals, when magnified 

 four hundred times ; and the latter into rectangular crystalline forms, 

 in most cases, very nearly square. With the same flux, baryta gave 

 hexagonal crystals, discernible by the naked eye ; and which, when 

 not rendered too opaque by prolonged flaming, viewed under the micro- 

 scope, resembled the common stellate form of snow crystals. Strontia 

 produced equally conspicuous crystals, but of a rectangular form, the 

 length being twice or three times the breadth. 



Several experiments with each of the bases just named, in both 

 fluxes, showing no variation in crystalline form from those I had 

 already observed, although in every instance I submitted the result of 

 my trials to microscopic examination, rarely using a less magnifying 

 power than eighty diameters, I passed on to such of the other earths 

 and metallic oxides as were within my reach, of which a list will be 

 found upon page 481. 



Before describing in detail the phenomena produced by flaming with 

 these several substances, a few words in reference to the methods pur- 

 sued may net be unacceptable ; premising, however, that these state- 

 ments, together with all included in these pages, are not predicated upon 

 sufficiently numerous experiments to be regarded as authoritative ; and 

 that they are offered rather as suggestions to persons better fitted, and 

 with more ample leisure and opportunity than the author, to follow 

 out the subject exhaustively. 



The loop of platinum wire should be about one eighth of an inch in 

 diameter, and quite circular, in order to give as uniform convexity to 

 the surface of the bead as possible. Two turns of the wire are better 

 than one where microcosmic salt is used, on account of the greater 

 fluidity of this flux than borax. The bead should not be much over 

 one twentieth of an inch thick at least, where examination by a high 

 magnifying power is desirable. Crystallization is best produced by 

 thrusting the tip of the blowpipe well into the flame of the lamp, 

 until a fine-pointed jet of flame, less than an inch in length, is ob- 

 tained ; then expose a small portion of the bead, close to the edge, (see 

 diagram,) for an instant to its action, until the part so exposed. 



