How to type Reference Mark
To type Reference Mark (※), press Alt + 8251 on Windows, holding Alt while typing the digits on the numeric keypad, or write ※ in HTML. On any system you can copy it from this page, or enter its codepoint U+203B directly: Ctrl+Shift+U then 203b on Linux, or 203b then Alt+X in Microsoft Word.
Reference Mark
- Copy it
- Unicode
- U+203B
- Windows
- Alt + 8251 needs regedit
| Where | Shortcut | How |
|---|---|---|
| Windows | Alt + 8251 | Above 255, so Windows will not accept it as typed. Copy the character instead. |
| Mac | N/A | Press the keys together. |
| HTML | ※ | Paste this into HTML source. |
| Unicode | U+203B | REFERENCE MARK |
Other punctuation symbols
※ lives in the General Punctuation block (U+2000-U+206F), dashes, quotation marks, spaces and reference marks used across languages. Within Unicode it is classified as Po, a punctuation mark.
How to type ※ in Microsoft Word
In Microsoft Word, type 203b and press Alt + X. Word reads the four characters before the cursor as a hex codepoint and replaces them, and pressing Alt + X again turns ※ back into 203b, which is the quickest way to check what a character really is.
What Unicode says about Reference Mark
The standard recognises ※ under other names as well: Japanese kome, Urdu paragraph separator. Those are a formal alias, not a nickname, which is why they turn up in other people's documentation for the same codepoint.
The code chart points from ※ to tibetan ku ru kha bzhi mig can (U+0FBF) and U+200AD (U+200AD). These are the characters the standard thinks you might have meant instead, which makes them the ones worth checking when Reference Mark looks nearly right but not quite.
How Reference Mark behaves in software
※ is East Asian Ambiguous: one column in a Western locale and two in a CJK one. That is why a table that lines up on your machine can break on somebody else's, with no change to the text at all. It carries no direction of its own, so in a line that mixes English with Arabic or Hebrew it takes the direction of the text around it rather than forcing one.
Every way to write Reference Mark
| Form | Value |
|---|---|
| Codepoint | U+203B |
| Decimal | 8251 |
| UTF-8 bytes | E2 80 BB |
| UTF-16 bytes | 20 3B |
| HTML entity | ※ |
| HTML numeric | ※ or ※ |
| CSS | \203B |
| JavaScript | \u203B |
| Python | \u203B |
| URL-encoded | %E2%80%BB |
| SVG | 203B.svg · (SIL Open Font License) |
If ※ arrives as ※
One character has become three, which is what a three-byte UTF-8 sequence looks like when each byte is read separately: E2 80 BB read as Windows-1252 gives exactly ※. The text is intact and the damage is in the reading, so setting the reader to UTF-8 restores it and retyping does not. A rectangle instead of the mojibake means the opposite: the encoding is right and the font has no glyph at U+203B, which for a character this far outside Latin-1 is common in older office fonts.