1886.] on Recent Researches on Meteorites. 543 



the same in every case, so that a short description will suffice for all. 

 The temperature was kept as nearly as possible the same in every 

 experiment, but no doubt differences of many degrees occurred in 

 some of the experiments, which was unavoidable in using an ordinary 

 combustion furnace. 



The meteorite or graphite, as the case might be, was broken up 

 into a coarse powder, introduced into a convenient length of 

 combustion tubing, and connected up with a Sprengel pump, a small 

 bulb-tube immersed in a freezing mixture intervening, so as to retain 

 any moisture or condensable volatile products that might come off. 

 The tube was first thoroughly exhausted and then heated in an 

 ordinary gas combustion furnace to a low red heat. The gases, during 

 the heating, were gradually drawn off by the Sprengel pump, and 

 when the tube had remained for several minutes at a low red heat it 

 was completely exhausted. The total quantity of gas collected was 

 in every case used for the analysis. 



The " Dhurmsala" specimen was an ordinary fragment of a much 

 larger original mass, but in the case of the Pultusk and Mocs 

 meteorites, comj)lete stones were fortunately obtained, weighing 

 respectively 57 and 103 grams, having the characteristic black glaze 

 on their surfaces. 



Such a large quantity of water was condensed in the bulb tube in 

 heating the Dhurmsala meteorite, it being the first one examined, 

 that it was thought it might be principally due to the great absorptive 

 power of these porous bodies, and that therefore the moisture might 

 have been condensed in the pores of the meteorite from the surround- 

 ing air. The Pultusk and Mocs specimens appeared to be especially 

 adapted for ascertaining whether this was the case, as the complete 

 covering of black glaze would probably prevent the moisture from 

 penetrating to the interior of the stones. The fragments of these 

 stones were therefore transferred as quickly as possible to the 

 combustion tube after they had been broken up. Notwithstanding 

 these precautions, fully as much water was condensed from them as 

 from the Dhurmsala specimen, which seems to suggest that the w^ater 

 is really combined in some form in the stone and not obtained directly 

 from the surrounding atmosphere, although it must be admitted that 

 the glaze on both the stones was not of a very glossy character, and 

 did not have the appearance of being absolutely impervious to 

 moisture. 



The pumice-stone was examined merely with a view to comparing 

 the gases occluded by a porous body of volcanic origin with those 

 contained in meteorites. The sample taken was a fresh piece of 

 stone, which had not been dried or purified in any way. 



It is evident that it differs considerably from the meteoric stones, 

 the total occluded gas being very small, only about half its volume, 

 the carbonic acid at the same time being much less, with a propor- 

 tionate increase in the carbonic oxide. 



The general method of analysis was as follows, and the accuracy 



(Vol. XI. No. 80.) 2 n 



