362 



SCIENCE 



[N. S. Vol. LIV. No. 1398 



THE AMERICAN PHILOSOPHICAL 



SOCIETY 



II 



Propylene glycol dinitrate: Chakles E. Mdnroe. 

 During the Great War not only was the demand for 

 glycerine for use in making nitroglycerin greatly 

 increased, but, as glycerine is normally produced 

 from fats and oils which grew in demand for use 

 in food, there was a special shortage in the supply. 

 A promising substitute, though not a full equiva- 

 lent for this, was found in propylene glycol dini- 

 trate. Propylene glycol dinitrate is a nitric ester 

 produced by the nitration of one of the isomeric 

 forms of propylene glycol, which latter is the sec- 

 ond member of- the group of dihydroxy alcohols or 

 glycols. The nitration of the glycol to produce this 

 explosive is carried on in the same manner and with 

 the use of the same acids as that of glycerine to 

 produce nitroglycerine and the product has a simi- 

 lar appearance to the latter. The results of tests 

 reported show that this propylene glycol dinitrate 

 may be used as nitroglycerine is in the manufac- 

 ture of dynamites and blasting gelatins. It is 

 found to be less sensitive, to have a lower freezing 

 point, to be decidedly more volatile, and to develop 

 less strength than nitroglycerine, but in an emer- 

 gency it may be efficiently used as an explosive, 

 especially in mining and other industrial operations. 



Further investigations concerning the relations 

 hetween terrestrial magnetism, terrestrial elec- 

 tricity, and solar activity: Louis A. Bauer. The 

 following chief facts have resulted from the present 

 investigation: (1) The earth 's average intensity of 

 magnetism, as well as the strength of the electric 

 currents circulating in the earth's crust, decreases 

 with increased solar activity. The change between 

 minimum and maximum sunspot activity in the case 

 of the former may amount to six per cent, and 

 more and in the case of the latter one hundred per 

 cent, and more. (2) The atmospheric potential- 

 gradient, or the deduced negative charge on the 

 surface of the earth, increases with increased solar 

 activity, the range in the variation between mini- 

 mum and maximum sunspot activity being about 15 

 to 20 per cent. The electric conductivity of the 

 atmosphere, on the other hand, shows but little, if 

 any, systematic variations during the sunspot cycle. 

 Accordingly, since the vertical conduction-current 

 of atmospheric electricity is derived from the prod- 

 uct of the potential-gradient and the electric con- 

 ductivity, it is found that this vertical current 

 increases in strength with increased solar activity; 

 the range of the variation between the minimum 

 and maximum sunspot activity is about 20 to 25 



per cent. It would thus appear that atmospheric 

 electricity, like terrestrial magnetism, is controlled 

 by cosmic factors. These new results have an im- 

 portant bearing upon theories of atmospheric elec- 

 tricity. (3) Regarding the daily and monthly fluc- 

 tuations in terrestrial magnetism, earth currents, 

 and atmospheric electricity, as measured by the 

 quantity, H B, where H is the intensity of the field 

 and B the range in the element during the period 

 considered, it is found that while in general, the 

 magnetic and earth-current fluctuations increase 

 with increased solar activity, the electric fluctua- 

 tions, as shown by potential-gradient observations, 

 apparently decrease with increased solar activity. 

 (The latter result, however, should be regarded as 

 but a preliminary one and it is receiving further 

 investigation.) (4) Instead of using the sunspot 

 numbers direct for comparison with magnetic and 

 electric variations, it is found that a more satisfac- 

 tory measure of solar activity may be based upon 

 the monthly range of sunspot frequency, or upon 

 the average numerical departure of the daily sun- 

 spot numbers from the mean of the month. In 

 brief, there is indicated that a better measure of 

 the radiations and emanations affecting the earth 's 

 magnetic and electric conditions is some quantity 

 measuring the variability, or rate of change, in the 

 sunspot numbers, rather than the numbers them- 

 selves. By measuring in this manner the variations 

 in solar activity, and adopting a similar measure 

 with regard to the solar constant values obtained 

 by the Smithsonian Institution at Calama, Chili, for 

 the two years 1919 to 1920, a good agreement, on 

 the whole, is found between the two sets of meas- 

 ures of solar activity. 



On mean relative and absolute parallaxes : Keivin 

 Burns. This paper shows that the mean parallax 

 of a group of stars, distributed at random, is 3.56 

 times the mean total proper motion divided by the 

 mean total (uncorrected) radial velocity. By this 

 formula the mean parallax was computed for the 

 bright stars of each spectral class. The results are 

 m good agreement with those obtained by Camp- 

 bell, who used radial velocities freed from the 

 motion of the sun and the tau components of 

 proper motion. The newer method is much less 

 laborious. 



The mean parallax for those stars whose relative 

 parallax has been observed was computed and the 

 correction to reduce to absolute was derived. This 

 was found to be 0."010. This correction is the mean 

 parallax of the comparison stars, which is in fair 

 agreement with the value derived from the mean 

 proper motion of these stars. 



