286 MAJOR A. E. OXLEY ON THE INFLUENCE OF MOLECULAR 



* 



experimentally, and point to the importance of the local molecular forcive in magneto- 

 optic phenomena. It should be noted that the value of the reverse field we have 

 taken is that of the intermolecular field. Within the atom the reverse field will 

 probably be greater than 6'5x 10 6 gauss. If it is of the order 5x 10 7 gauss, the 

 above ratios become 320x10* for iron, 200 xlO 6 for nickel and 700 x 10" for cobalt. 

 Intra-atomic fields of the order 1 8 gauss are required by HUMPHRIES* to explain the 

 pressure shift of spectral lines and by RlTzf in his theory of spectral series (see also 

 Part III., p. 100, and supra, p. 273, footnote). 



(8) SUMMARY OF CONCLUSIONS. 



(I.) The applications of the local molecular forcive, in diamagnetic, paramagnetic and 

 ferro -magnetic media, have in the present research been extended to interpret the 

 ultimate tensile strength of crystalline and vitreous media. It has been shown by 

 EWINU and UOSENHAIN that the permanent set which occurs prior to breaking is due 

 to slipping along the cleavage planes within the individual crystalline grains. We should 

 therefore expect that the material would be fractured when the applied mechanical 

 stress is equal to that produced internally by the local molecular forcive. The internal 

 stress within the material is shown to be of the order 2 x 10 9 dynes per square centimetre 

 which is approximately the mean value of the ultimate tensile strengths of crystalline 

 and vitreous media (pp. 250-259). 



(II.) As a consequence of this internal stress, the energy per unit volume will be 

 2x 10 9 ergs, and this energy, which is over and above that which exists in the fluid 

 state, should be a measure of the latent heat of fusion per cubic centimetre. This test 

 which was applied in Part III. to test the order of the local forcive, has been extended 

 to a variety of organic and inorganic media, including the metals and is found to 

 accord with the experimental values to the right order (pp. 253-4). 



(III.) Since the forces under which the molecules vibrate are those to which we 

 ascribe the elastic properties of crystalline media, the results obtained are consistent 

 with the theory of specific heats developed by DEBYE, in which the specific heat is 

 attributed to purely translational vibration, and it has been shown (Part III.) that, 

 near the fusion point, the rotational energy acquired by the molecules will give a 

 measurable departure from this theory which is actually observed experimentally. 

 As we should expect, it is found that the elastic constants of a variety of ferro- 

 magnetic, diamagnetic and paramagnetic media are of the same order, several 

 diamagnetic and paramagnetic media even surpassing steel in their power to resist 

 distortion (pp. 257-259). 



(IV.) Any change of internal pressure will be accompanied by a change of volume 

 defined by the compressibility of the medium and dependent as to sign upon 



* ' Astrophysical Journal,' vol. 23, p. 232, 1906; vol. 35, p. 268, 1912. 

 t ' Ann. der Phys.,' vol. 25, p. 660, 1908. 



