In the second chapter of the book by R.P.Burn, titled: Groups: A path to geometry, it is given in q. 16 ( scanned image of question: https://i.stack.imgur.com/vtg24.jpg , scanned image of answer: https://i.stack.imgur.com/7iIxj.jpg ) to show that given a group $S_n, n \in \mathbb{N}$, and an arbitrary permutation $\alpha \in S_n$; the cycles formed by the sequence $1, \alpha, \alpha^2, \alpha^3, \alpha^4, \cdots$, are either identical or disjoint.
The approach taken by the question is to assume two non-disjoint sequences:
(i) $A : \ a, a\alpha, a\alpha^2, a\alpha^3, a\alpha^4, \cdots$,
(ii) $B : \ b, b\alpha, b\alpha^2, b\alpha^3, b\alpha^4, \cdots$,
such that $a\alpha^i = b\alpha^j$.
The book states to show:
(i) If $i\ge j$, identify second sequence with in the first.
(ii) Also, as the two sequences are non-disjoint, so identify first sequence with in the second.
Issue:
The assumption of two non-disjoint sequences $A, B$ with different indexes $i, \ j$ is confusing. I feel at most a contradiction can be used to show that $i=j$.
Unable to grasp the two liner answer stated:
If $a\alpha^i = b\alpha^j$, then $a\alpha^{i -j} = b$ and $a\alpha^{i -j+k} = b\alpha^k$. If $a\alpha^m = a$, then $b\alpha^{j-i+m} = a$.