Experiementally, suceptibility can be dervied from thermal evolution of the mgnetiztion under a weak field H (linear regim). In these condition X=M/H (with H=B/m0).

1.For ferro compounds :

X(T) diverges or exhibits a narrow peak at Tc (Curie-Weiss law). However, one must taking account on the demagnetization field correction. If we put, 1/X vs T, far away from TC, that be nearly linear curve which shuts down (vanishes) at TC: 1/X(T)~ T-TC

2. For antiferro (AF):2 cases 

2.1 when H is parallel ot the AF axis : X(T) grows up from 0 to a maximum value at TN (Néel Température) and deacrease when T>TN as a paramgnetic case.

2.2. when  H is perpendicular to c-axis, X(T) is nearly constant below TN, and it rejoign X(T) for H//C above TN. Avobe TN, 1/X(T) correponds to a linear curve 1/X(T)~ T+TN. 

TN is visible when X(T) // and X(T) perp accross each the other.

3. Ferri (FI)

is between Ferro and AF behavior : because the system contains tow non equivalent lattices A and B with opposite magnetizations MA and MB. Thus, the systme exhibits at compenstion Temperature  TComp when |MA| =|MB| belew  (TComp