C. Barus — Effect of Magnetization, etc. 1T5 



dium arseniate has been mentioned recently as occurring in a 

 mineral water in Asche County, North Carolina. 



Hygeia Spring, which supplies the bath houses at the hotel 

 in the Lower Greyser Basin, is an alkaline siliceous water carry- 

 ing *3 of a grain of sodium arseniate to the gallon. In the 

 amount of sodium arseniate held in solution the Yellowstone 

 Park waters fall below the celebrated arsenical springs of La 

 Bourboule in the volcanic region of the Auvergne, which 

 within recent years have acquired a well deserved reputation 

 for their alterative properties in skin diseases. While the Yel- 

 lowstone Park waters are somewhat less rich in arsenic than 

 those of La Bourboule they must greatly surpass the latter in 

 their enormous overflow. The entire supply from the springs 

 of La Bourboule amounts to 1500 gallons per minute. Dur- 

 ing the past season the cauldron of Excelsior Greyser in Mid- 

 way Basin alone poured into the Firehole river, according to 

 the most accurate measurements which could be made, no less 

 than 4400 gallons of boiling water per minute. According to 

 the analysis of the sample collected August 25, 1884, the water 

 of this geyser contained .19 grains per gallon of sodium arsen- 

 iate. As yet we know very little about the remedial proper- 

 ties of the Yellowstone waters. At no distant day experience 

 may show that they are decidedly efficacious in external appli- 

 cations and under proper medical guidance may take high rank 

 as arsenical waters for the cure of certain forms of nervous 

 affections and cutaneous eruptions. 



Art. XXI. — The Effect of Magnetization on the Viscosity and the 

 Rigidity of Iron and of Steel ; by C. BARUS. 



Mr. Herbert Tomlinson* has recently communicated re- 

 sults on the changes of viscosity and of elasticity produced by 

 magnetizing iron. As both classes of data are obtained by the 

 vibration method, it seems not undesirable to attempt to verifv 

 them by some static method, and the one described in my last 

 paperf is so easily applicable that I have made use of it. My 

 original purpose was to confine the investigation to measure- 

 ments of viscosity ; but I found in the course of the work that 

 the incidental data on the rigidity of iron and steel could be 

 grouped together. Taken collectively in this way, they led to 

 inferences which the well-known and comprehensive researches 



* Tomlinson: Beiblatter No. 3, p. 176, 1887; original in Proc. Roy. Soc, xl, p. 

 447, 188G, is unfortunately not at my disposal. 



f This Journal, xxxiv, pp. 1 to 4, 1887. In equation (1), p. 1, of this article i/> 

 is obviously to be replaced by dip. 



