\section{From coordinates to distinct pairs}\label{sec:count} For each \(d\) fix \(\rho_d\in R\) of norm \(d\), obtained by multiplying one factor above each split prime of \(d\). Embed the residual shell at \(m=n/d\) in the original shell by \((z,y)\mapsto(\rho_d z,\rho_d y)\). This preserves the fractional norm coordinate, the prescribed orientation at \(3\), and inert divisibility above \(Z\). Each original value pair recovers its triple \(d,u,v\). At split primes up to \(Y\), the only possible positive valuation is one, occurring in both norms at the primes of \(d\), only in the first at \(u\), and only in the second at \(v\). The images for distinct triples are consequently disjoint. This makes it possible to bound the remaining losses on the original shell without summing an error over the triples. \subsection{Large inert primes} If an inert prime \(p>Z\) divides the first component norm, that component is \(p\beta\). The norm of \(\beta\) belongs to \([(x-H/2)/p^2,(x+H/2)/p^2]\). For the other component use \(n-x\) in place of \(x\). Each resulting annulus has squared-radius width \(H/p^2\) and outer radius at most \(\sqrt n/p\). Covering each lattice point by a translated fundamental cell, of fixed area and bounded radius, gives \(O(H/p^2+\sqrt n/p+1)\) possible \(\beta\). Indeed the cells lie in the annulus obtained by moving the two boundaries by a fixed distance, truncating the inner radius at zero. The complementary positive norm has at most \(6\max_{a\le n}\tau(a)=n^{o(1)}\) coordinate choices. Since all weights are at most one, summing on the original shell over the two components and \(Z0\). The block probability bounds hold as before. Choose \(c_1\) so that the corresponding divisor level is \(D'\le n^{1/4}\). The total error is \(O(D'\log(2D'))=o(H/\log n)\), since \(\nu(\ell)\le\tau(\ell)\) and \(\sum_{\ell\le D'}\tau(\ell)\ll D'\log(2D')\). The upper main term is at most a constant times \(H\) times \[ \prod_{\substack{5\le p\le Z'\\ p\ {\rm inert}}}(1-2/p) \prod_{\substack{5\le p\le Z',\ p\mid n\\ p\ {\rm inert}}} \frac{p-1}{p-2} \ll \frac{J(n)}{\log n}. \] This proves the upper bound and completes Theorem~\ref{thm:main}. The argument for this upper bound in fact works in any interval whose length is at least a fixed positive multiple of \(\sqrt n\). \begin{proof}[Proof of Corollary~\ref{cor:ranges}] In a fixed interval of shares, take a smaller fixed interval of length at most the distance of its center from the nearest endpoint of \([0,1]\). This satisfies the hypotheses with \(H\asymp n\) for large \(n\), giving the lower bound. Use the upper sieve on the full interval and on \([0,n]\) for the unrestricted count. At \(x=\lambda n\) with \(0<\lambda<1\), a width \(H=n^\theta\) satisfies the condition for any fixed \(0<\varepsilon<3\theta-2\). Finally, inside \([a n^\theta,b n^\theta]\) take a subinterval of width comparable to \(n^\theta\) and at most its midpoint. Its distance to the nearer endpoint of \([0,n]\) is comparable to \(n^\theta\), so one can use \(0<\varepsilon<2\theta-1\). The upper sieve on the full interval gives the matching bound, also for \(a=0\). \end{proof}