The set P of all probability measures s on the unit circle T splits into three disjoint subsets depending on properties of the derived set of {|j n | 2 ds} n \ 0 , denoted by Lim(s). Here {j n } n \ 0 are orthogonal polynomials in L 2 (ds). The first subset is the set of Rakhmanov measures, i.e., of s ¥ P with {m}=Lim(s), m being the normalized (m(T)=1) Lebesgue measure on T. The second subset Mar(T) consists of Markoff measures, i.e., of s ¥ P with m¨Lim(s), and is in fact the subject of study for the present paper. A measure s, belongs to Mar(T) iff there are e > 0 and l > 0 such that sup{|a n+j |: 0 [ j [ l} > e, n=0, 1, 2, ..., {a n } is the Geronimus parameters (=reflection coefficients) of s. We use this equivalence to describe the asymptotic behavior of the zeros of the corresponding orthogonal polynomials (see Theorem G). The third subset consists of s ¥ P with {m} v Lim(s). We show that s is ratio asymptotic iff either s is a Rakhmanov measure or s satisfies the Ló pez condition (which implies s ¥ Mar(T)). Measures s satisfying Lim(s)={n} (i.e., weakly asymptotic measures) are also classified. Either n is the sum of equal point masses placed at the roots of z n =l, l ¥ T, n=1, 2, ..., or n is the equilibrium measure (with respect to the logarithmic kernel) for the inverse image under an m-preserving endomorphism z Q z n , n=1, 2, ..., of a closed arc J (including J=T) with removed open concentric arc J 0 (including J 0 ="). Next, weakly asymptotic measures are completely described in terms of their Geronimus parameters. Finally, we obtain explicit formulae for the parameters of the equilibrium measures n and show that these measures satisfy {n}=Lim(n).