Main Page.

Unicode allows adding diacritical markings to ordinary characters, primarily to support the texts of many languages. In most cases these indicate details of pronunciation, but mathematicians sometimes find it convenient to employ the markings for entirely different purposes. Unicode has a code range (0300-036F) of various diacritical marks that combine with other characters. Many of them are shown here, with a few markings from other code ranges. They are not intended to be an orthogonal set.

Unfortunately, many implementations do not yet support these well. In most cases, a marking should be ordinarily be centered over or under the primary character's glyph, but sometimes appears elsewhere, or is missing entirely. Sometimes the extra marking overlaps the main character when it ought to be completely separate.

In the examples below, the "Y" (&#x59;), "y" (&#x79;), and "o" (&#x6F;) characters are the ordinary letters from page zero of Unicode, as inherited from long-established ASCII. The "o" is included merely to to give the diacriticized "Y" and "y" a typographical context for better understanding. Surrounding all that in the source code are the <tt> and </tt> markup symbols of HTML, in order to effect a monospace font. On the author's computer, this gave better results than a variable-pitch font.

Authors should remember that some Unicode characters have diacritical marks "built in"; The choice between combinings and built-ins requires consideration of the context. Example:

All the official Unicode names begin with the word "combining", which is here abbreviated as "c.".

In table D-1, each row has markings in above-below pairs:

Table D-2 has some above markings:

Table D-3 has some below markings, with related markings in the same row:

Table D-4 has some overlay markings:

If the markings of table D-5 are properly implemented, the lines will connect from one character to the next, and be centered on the whole group:

Markings can be cascaded, with examples below. However, the risk of incorrect rendering increases with the number of markings.

Cascading to the point of abuse leads to Zalgo text.