Feedback Case Number: 66974
IsNumeric() doesn’t spot Unicode full width variants of digits.
In unicode there are the “traditional” digits from ascii but there are also full width versions to match the width of Kanji characters. These are &uFF10 - &uFF19. There are also full width versions of - (&uFF0D) and a full width version of the full stop &uFF0E
They are documented here:
Halfwidth and Fullwidth Forms (Unicode block) - Wikipedia.
Val( ) .ToDouble .ToSingle etc don’t take account of these variants either.