“…In fact, we fix numbers 𝑆 0 , 𝑅 0 ∈ (0, ∞) with Ω ⊂ 𝐵 𝑆 0 and 𝑆 0 < 𝑅 0 , abbreviate 𝐴 𝑅 0 ,𝑆 0 ∶= 𝐵 𝑅 0 ∖𝐵 𝑆 0 , and require there are parameters 𝛾 1 , 𝛾 2 , 𝛾 3 ∈ [0, ∞], 𝑞 ∈ (1, ∞) such that the restriction 𝑢|𝐴 𝑅 0 ,𝑆 0 × (0, 𝑇 0 ) belongs to 𝐿 𝛾 1 ( 0, 𝑇 0 , 𝐿 𝑞 (𝐴 𝑅 0 ,𝑆 0 ) 3 ) , ∇ 𝑥 𝑢|𝐴 𝑅 0 ,𝑆 0 × (0, 𝑇 0 ) to 𝐿 𝛾 2 ( 0, 𝑇 0 , 𝐿 𝑞 (𝐴 𝑅 0 ,𝑆 0 ) 3 ) , and 𝑓|𝐴 𝑅 0 ,𝑆 0 × (0, 𝑇 0 ) to 𝐿 𝛾 3 ( 0, 𝑇 0 , 𝐿 𝑞 (𝐴 𝑅 0 ,𝑆 0 ) 3 ) . Actually our assumptions are somewhat more general (see at the beginning of Section 5 and Theorem 5.11), in view of applications in [8] to a nonlinear problem with (1.2) as special case. But with the preceding conditions, all the main difficulties of our proofs would already be present.…”