When you write a handler for a message, you specify a sequence of statements for HyperCard to run. Each message handler has the following form, with the italicized words as placeholders:

```
on <i>messageName</i>
    <i>statements</i>
end <i>messageName</i>
```

When it runs the message handler, HyperCard sends each line of the handler as a message itself (so handlers can call other handlers).

<b>Important</b>: The message name does <i>not</i> have to be one of HyperCard’s built-in system messages or commands.

For example, if you wanted a new command called<code> doubleBeep </code>that would beep twice, you would write a handler for it as follows:

```
on doubleBeep
   beep
   beep
end doubleBeep
```

Of course, HyperCard doesn’t know about the [[c:doubleBeep]] command, so it will never be sent automatically in response to an event. But you can send [[c:doubleBeep]] from the Message box or use it as a statement in other handlers.

For example, the script of a card button might contain the following handler for the system message [[c:mouseUp]]:

```
on mouseUp
  doubleBeep
end mouseUp
```

If the script also contains the [[c:doubleBeep]] handler (or if the script of any object later in the message-passing path contains it), the [[c:mouseUp]] message will send the [[c:doubleBeep]] message, and [[c:doubleBeep]] will send two [[c:beep]] commands. Because [[c:beep]] is a built-in command, HyperCard beeps twice.

If HyperCard can’t find the [[c:doubleBeep]] message, it will complain (with a dialog box) that it “can’t understand” the message.