672 



SCIENCE. 



[N. S. Vol. XXIII. No. 591. 



temperatures, and has misled several investi- 

 gators int« interpreting the volatilization of 

 the oxide to be due to its vapor pressure, when 

 in fact the oxide, heated in the absence of car- 

 bon or reducing papers, shows little or no 

 volatility. The volatilization of magnesium 

 oxide is of this class. 



Whether the pressure which is essential to 

 the preparation of good quartz glass in quantity- 

 acts upon the vapor pressure of the silica, or 

 whether it affects the reduction of the silica to 

 the metal, has not yet been determined. It is 

 not unlikely that both reactions occur, and 

 that the narrow temperature limits within 

 ■which we found it practicable to work, lie be- 

 tween the temperature of volatilization of the 

 silica and that of its reduction by carbon. 

 This question is not material to the successful 

 production of quartz glass, and will be consid- 

 ered in a later paper. 



One other conclusion appears to be reason- 

 ably certain from our work, namely, that air 

 once enclosed within the body of a charge of 

 quartz glass can not be displaced, either by 

 long-continued heating or by extremely high 

 temperatures. 



Our experience did not suggest that we were 

 approaching any necessary limit in the size of 

 the charge which could be handled. A furnace 

 of suitable size, provided with somewhat more 

 power, would undoubtedly produce clear quartz 

 glass in much larger units than we were able 

 to do in our small furnace. 



Summing up the conditions for preparation 

 of good quartz glass, we find them to be: An 

 initial temperature of 2,000° or more, without 

 pressure, to produce sufiicient quartz vapor to 

 drive out the air from between the grains, 

 followed by pressure (at least 500 pounds), and 

 a reduced temperature (perhaps 1,800°), with 

 time for the quartz to flow compactly together 

 without being attacked by the graphite. 



Arthur L. Day, 

 E. S. Shepherd. 



Geophysical Labobatoet, 

 Carnegie Institution, 

 Washington, D. C, 

 April 18, 1906. 



METEOROLOGICAL PHENOMENA ON MOUNTAIN 

 SUMMITS. 



Much of our knowledge of the upper air has 

 been obtained from observations made on the 

 summits of mountains. With the single 

 notable exception of the Prussian Aeronautical 

 Observatory near Berlin, where, for several 

 years, daily observations at great heights have 

 been obtained with the aid of kites and bal- 

 loons, we are still dependent upon the moun- 

 tain observatories for information concerning 

 annual and seasonal changes in the upper air 

 at different heights, and for other data not 

 easily secured except by means of continuous 

 observations made at the same place. The 

 chief error arising from any general applica- 

 tion of such observations is caused by the un- 

 known influence of the mountain itself upon 

 the meteorological conditions in its vicinity. 

 The results from observations in the free air 

 do not show the same vertical changes that are 

 observed on mountains, the diurnal periodic 

 change of temperature noticeable on all moun- 

 tains disappearing at a height of 1,000 meters 

 in the free air. 



A few approximate comparisons have al- 

 ready been made, of Ben Nevis (1,343 meters 

 high) in Scotland by Mr. Dines, and of the 

 Brocken (1,100 meters high) in Germany by 

 Dr. Assmann, but in both instances the kite 

 or balloon observations apparently were made 

 at a distance exceeding 90 kilometers from the 

 mountain observatory. Also, Mr. Clayton has 

 compared the temperature on Blue Hill with 

 that of the free air. 



The data obtained indicated that the tem- 

 perature on mountain summits is lower than 

 that of the free air at the same height. No 

 information as to differences of humidity or 

 wind velocity is available, although it appears 

 quite probable that the wind velocity is higher 

 on mountains than in the free air at the same 

 height. 



During the last week in August, 1905, I was 

 able to make a comparison of the weather 

 conditions on Mount Washington, N. H., with 

 those of the free air, by means of kites flown 

 in the Ammonoosuc Valley 16 kilometers west 

 of and 1,500 meters lower than the summit of 



