These labels identifying each type of convention can include a character, a word, a group of words, a whole verse, several verses, a whole paragraph, or any fraction of these units.

The usual rule is to insert the diacritic label within the label containing it, respecting the hierarchy:

<l><secl>arma</secl> virumque cano Troiae qui primus ab oris</l>

<l><secl>arma virumque cano Troiae qui primus ab oris</secl></l>

But if the text we need to highlight doesn’t conform to the nesting hierarchy of tags, (for example if the marked text starts in the middle of one line and ends in the middle of the next), we’ll need to use a non-hierarchical structuring system to indicate where each mark begins and ends.

To do this we use the <milestone/> tag which connects to an <anchor/> placed at the point where that condition ends. Each type of diacritical mark will carry within the <milestone/> the @unit attribute that corresponds to its value.

An example with a spurious text:

<p>Párrafo que está solo parcialmente borrado. Hasta aquí el texto se reconoce como auténtico, pero
	<milestone unit="secl" spanto="#anclajefinal"/>ésta es la parte a borrar del párrafo.</p>
	<p>Este párrafo tampoco es original.</p>
	<p>Segundo párrafo que no es original.</p>
	<p>Párrafo parcialmente borrado, porque a la mitad del párrafo acaba el texto espurio marcado por este anclaje
	<anchor xml:id="anclajefinal"/>
y desde aquí ya sigue ya el texto válido.</p>

Thus, the necessary attributes of <milestone/> are:

• @unit: each diacritical mark will have its own, which will always correspond to the name of its base tag. For example, if additions are marked with <supplied>, the non-hierarchical nesting with <milestone> will be called <milestone unit="supplied"/>. And so on.
• @spanTo: this attribute points to the @xml:id of the <anchor/> always placing the #symbol first), which we have placed at the point where it finishes the text indicated in the <milestone>. Cf. <anchor> to learn how to write this tag.