Susceptibility was found to diminish greatly with rise of temperature. According to G. Jäger and S. Meyer (Wien. Akad. Sitz., 1897, 106, II. a, p. 623, and 1898, 107, II. a, p. 5) the atomic susceptibilities k of the metals nickel, chromium, iron, cobalt and manganese in solutions of their salts are as follows:—

Metal.k × 106Metal.k × 106.
Ni4.95 = 2.5 × 2Co10.0 = 2.5 × 4
Cr6.25 = 2.5 × 2.5Fe(2)12.5 = 2.5 × 5
Fe(1)7.5 = 2.5 × 3Mn15.0 = 2.5 × 6

Fe(1) is iron contained in FeCl2 and Fe(2) iron contained in Fe2(NO3)6.

Curie has shown, for many paramagnetic bodies, that the specific susceptibility K is inversely proportional to the absolute temperature θ. Du Bois believes this to be an important general law, applicable to the case of every paramagnetic substance, and suggests that the product Kθ should be known as “Curie’s constant” for the substance.

Elementary Bodies and Atomic Susceptibility.—Among a large number of substances the susceptibilities of which have been determined by J. Koenigsberger (Wied. Ann., 1898, 66, 698) are the following elements:—

Element.κ × 106.Element.κ × 106.
Copper−0.82Tellurium− 2.10
Silver−1.51Graphite+ 2
Gold−3.07Aluminium+ 1.80
Zinc−0.96Platinum+22
Tin+0.46Palladium+50 to 60
Lead−1.10Tungsten+14
Thallium−4.61Magnesium+ 4
Sulphur−0.86Sodium+ 2.2
Selenium (red)−0.50Potassium+ 3.6

In a table accompanying Koenigsberger’s paper the elements are arranged upon the periodic system and the atomic susceptibility (product of specific susceptibility into atomic weight) is given for each. It appears that the elements at about the middle of each row are the most strongly paramagnetic; towards the ends of a row the susceptibility decreases, and ultimately becomes negative. Thus a relation between susceptibility and atomic weight is clearly indicated. Tables similarly arranged, but much more complete, have been published by S. Meyer (Wied. Ann., 1899, 68, 325 and 1899, 69, 236), whose researches have filled up many previously existing gaps. The values assigned to the atomic susceptibilities of most of the known elements are appended. According to the notation adopted by Meyer the atomic susceptibility k = κ × atomic-weight / (density × 1000).

Meyer thinks that the susceptibilities of the metals praseodymium, neodymium, ytterbium, samarium, gadolinium, and erbium, when obtained in a pure form, will be found to equal or even exceed those of the well-known ferromagnetic metals. Many of their compounds are very strongly magnetic; erbium, for example, in Er2O3 being four times as strong as iron in the familiar magnetite or lodestone, Fe2O3. The susceptibilities of some hundreds of inorganic compounds have also been determined by the same investigator (loc. cit.). Among other researches relating to atomic and molecular magnetism are those of O. Liebknecht and A. P. Wills (Ann. d. Phys., 1900, 1, 178), H. du Bois and O. Liebknecht (ibid. p. 189), and Meyer (ibid. p. 668). An excellent summary regarding the magnetic properties of matter, with many tables and references, has been compiled by du Bois (Report to the Congrès Int. de Phys., Paris, 1900, ii. 460).

Element 106k Element 106k Element 106k
Be +0.72 Cu −0.006 Cs − 0.03*
B +0.05 Zn −0.010 Ba − 0.02*
C −0.05 Ga − La +13.0
N  ? Ge − Ce +34.0
O + As ? Pr + Strong
F −0.01* Se −0.025 Nd +
· · · · · · · · · · Br −0.033 Sa +
Na −0.005* · · · · · · · · · · Gd +
Mg +0.014 Rb −0.02* · · · · · · · · · ·
Al + Sr −0.02* Er +41.8(?)
Si +0.002 Y +3.2(?) · · · · · · · · · ·
P −0.007 Zr −0.014 Yb +   (?)
S −0.011 Nb +0.49(?) Ta + 1.02(?)
Cl −0.02* Mo +0.024 W + 0.1
· · · · · · · · · · Ru + Os + 0.074
K −0.001* Rh + Ir +
Ca −0.003* Pd +0.55 Pt + 0.227
Sc  ? Ag −0.016 Au − 0.031
Ti +0.09 Cd −0.015 Hg − 0.030
V +0.17 In +0.01* Tl − 0.93
Cr + Strong Sn +0.004* Pb − 0.025
Mn + Sb −0.069 Bi − 0.023
Fe + Te −0.039 · · · · · · · · · ·
Co + I −0.040 Th +16.0(?)
Ni + · · · · · · · · · · U + 0.21
* Calculated.