March 24, 1916] 



SCIENCE 



413 



grees the electric cliarge of the metal changed 

 very little, but beyond 200 degrees the tube 

 became more electropositive with an increase 

 in temperature. It was impossible to measure 

 the induction of the tube much beyond 200 

 degrees, since at higher temperatures the hot 

 tube ionized the air and allowed the induced 

 charge of the cylinder to discharge to the tube. 



It is interesting in this connection to note 

 that the internal cohesion of iron and steel 

 seems to change with a change in the fixed 

 electric charge of the metal. In a paper on 

 " Contact Electromotive Force and Cohesion " 

 written several years ago it was shown that 

 when the metals are arranged in their proper 

 order in the contact electromotive series they 

 are arranged in the inverse order of their co- 

 hesion, so that the more electronegative a 

 metal is in the contact series the greater is its 

 cohesion. Since in the case of the steel tube 

 used in the experiment described above the 

 metal became more electronegative up to a 

 temperature of about 150 degrees, it would 

 seem to be a legitimate deduction that the ten- 

 sile strength of the tube should increase up to 

 this temperature and then begin to decrease 

 with a rise of temperature. 



In a series of experiments made by C. Bach 

 and described in Zeitsch. d. Deutsch. In- 

 genieure, 1904, p. 1300, the tensile strength of 

 iron was actually found to be much greater 

 at 200 degrees than at 20 degrees. From 200 

 to 300 degrees it decreases, but it is still 

 greater at 300 degrees than at 20 degrees. At 

 400 degrees it is only a little less than at 20 



In the Valve World of January, 1913, is an 

 article by I. M. Bregowsld and L. W. Spring 

 on " The Effect of High Temperatures on the 

 Physical Properties of Some Metals and 

 Alloys." In this article it is shown that 

 samples of cast iron, both soft and hard, have 

 a greater tensile strength at 300° F. than at 

 70° F., and that at 750° F. the tensile strength 

 is still within one per cent, of as great as it is 

 at 70° F. In the case of a sample of Crane 

 Ferrosteel the tensile strength is greater at 

 750° F. than at 70° F. 



In a dissertation by A. Lantz, entitled " Ein- 



wirkung der Temperatur auf die Biegfahigkeit 

 von Flusseisen und Kupferdraehten," Berlin, 

 1914, the author finds that what he calls the 

 BiegfdhigJceii of iron, i. e., its malleability or 

 toughness as measured by the number of times 

 it can be bent short forward and backward at 

 a given point before breaking, increases with 

 the temperature to about 220 degrees and then 

 decreases. In some cases the wire would stand 

 twice as many short bondings at 220 degrees 

 as at room temperature, while at 350 degrees 

 it would stand only one fifth as many as at 

 room temperature. The toughness of copper 

 measured in this way continued to increase to 

 320 degrees, which was the highest tempera- 

 ture of the experiment. 



The above mentioned experiments all seem 

 to indicate that the cohesion of iron increases 

 with its increase of temperature so long as the 

 iron continues to become more electronegative, 

 and that the cohesion begins to decrease at 

 about the temperature at which the iron begins 

 to lose its negative charge. 



This is what we should expect if cohesion 

 is an attraction between positive and negative 

 charges. An increase of cohesion would then 

 mean a greater attraction of the positive sub- 

 atoms of the metal for movable electrons and 

 a consequent increase of the negative charge of 

 the metal. So far as we know, such a change 

 in the attraction of the positive sub-atoms for 

 electrons can be brought about only by a 

 change in the specific inductive capacity of the 

 metal. 



It would seem that all known phenomena of 

 contact electrification may best be explained 

 on the hypothesis that different metals when 

 in electrical contact with the earth or with 

 the inside of a hollow conductor, although by 

 definition at zero potential, still actually re- 

 tain characteristic charges which are capable 

 of inducing a charge upon a different metal 

 when brought near it. It is these characteris- 

 tic charges which I have ventured to call the 

 natural charges of the metals. 



When two metals are brought near together 

 while in electrical contact with the earth, 

 their natural charges are increased or dimin- 

 ished by the bound charges due to their mu- 



