cp's OEIS Frontend

This is a front-end for the Online Encyclopedia of Integer Sequences, made by Christian Perfect. The idea is to provide OEIS entries in non-ancient HTML, and then to think about how they're presented visually. The source code is on GitHub.

A387045 Positive numbers k with property that the largest circle on the xy-plane enclosing exactly k lattice points in its interior does not exist.

Original entry on oeis.org

5, 6, 17, 18, 33, 34, 35, 36, 38, 50, 53, 54, 63, 70, 71, 72, 73, 89, 90, 97, 98, 102, 109, 110, 125, 126, 127, 128, 129, 150, 151, 165, 167, 168, 178, 188, 198, 209, 210, 217, 218, 219, 220, 221, 222, 242, 243, 257, 258, 259, 260, 277, 278, 285, 286, 294
Offset: 1

Views

Author

Jianqiang Zhao, Aug 14 2025

Keywords

Comments

Conjecture: This sequence is infinite.

Examples

			It can be proved that the largest circle enclosing exactly 5 or 6 lattice points in the interior on the xy-plane does not exist. Number 5 is the smallest nonnegative integer having this property and 6 is the next. Therefore, a(1)=5 and a(2)=6.
Here is a brief argument. For details, please see my arxiv paper 2505.06234.
First, let C be the circle going through (-1,0) centered at (1/2,1/2). It passes exactly 8 lattice points and encloses exactly 4. Now with (-1,0) fixed on the circle we can shrink C by an infinitesimal amount to circle C' so that C' only goes through one lattice point (-1,0). Then another infinitesimal perturbation will move C' to include exactly 5 lattice points in its interior. Another infinitesimal perturbation will move C' to include exactly 6 lattice points in its interior. Therefore, if the largest circle enclosing exactly 5 or 6 interior lattice points exists, then its radius is at least sqrt(10)/2.
Second, a geometric argument shows that if the radius of a circle is at least sqrt(10)/2 then it encloses either exactly 4 interior lattice points or at least 7 interior lattice points.
		

Crossrefs

Cf. A387044, complement of A387045; A192493, A192494, A128006, A128007.