Or keep the original graph and add the Y variable a second time, then adjust the scales
... of the original graph to focus on the main distribution
... of the additional graph to indicate that there are outliers:

Names Default to Here(1);
dt = Open( "$SAMPLE_DATA/Big Class.jmp" );
:sex[18]="";
:sex[40]="";
:weight[18]=1500;
:weight[40]=1000;
Graph Builder(
Size( 462, 468 ),
Show Control Panel( 0 ),
Variables( X( :height ), Y( :weight ), Y( :weight ), Overlay( :sex ) ),
Relative Sizes( "Y", [77 341] ),
Elements( Position( 1, 1 ), Points( X, Y, Legend( 37 ) ) ),
Elements( Position( 1, 2 ), Points( X, Y, Legend( 39 ) ) ),
SendToReport(
Dispatch(
{},
"weight",
ScaleBox,
{Min( 200 ), Max( 2000 ), Inc( 200 ),
Minor Ticks( 1 )}
),
Dispatch(
{},
"weight",
ScaleBox( 2 ),
{Min( 0 ), Max( 200 ), Inc( 50 )}
)
)
)