1: \begin{definition}[{Minimal scale~$\Y_s$}]

2: \label{d.Ys}

3: For the rest of this chapter, we let $\Y_s$ denote, for each $s\in (0,d)$, the maximum of the random variables in Theorem~\ref{t.regularity} and Propositions~\ref{p.correctorsbasecase} and~\ref{p.basecase}. In particular, we have that 

4: \begin{equation*} \label{}

5: \Y_s = \O_s\left(C(s,d,\Lambda)\right)

6: \end{equation*}

7: and that $r\geq \Y_s$ implies each of the conclusions in these results. We also let $\Y_s(x)$ denote, for each $x\in\Rd$, the random variable $\a\mapsto \Y_s(T_x\a)$ where $T_x$ is the translation operator on $\Omega$ defined in~\eqref{e.Ty}. It follows that $\Y_s(x)$ satisfies the same estimate as~$\Y_s$ and that similar statements hold for $r\geq \Y_s(x)$, appropriately translated by~$x$. 

8: \end{definition}

9: