--- a/Float.st	Tue Jan 21 11:13:19 1997 +0100
+++ b/Float.st	Tue Jan 21 11:20:33 1997 +0100
@@ -85,6 +85,29 @@
     [author:]
         Claus Gittinger
 "
+!
+
+errorHandling
+"
+    Floating point error handling and signalling depends on the systems
+    (actually: the C-runtime systems) ability to handle floating point errors.
+    Most systems report errors by raising an OS floatingPoint exception,
+    which is mapped to a fpExceptionInterrupt in ST/X.
+    However, some systems do return NaN as result.
+
+    Currently, ST/X does not care specially for these systems - it maybe added 
+    easily, if there is sufficient customer interest, though.
+
+    Try:
+        |f1 f2|
+
+        f1 := 1.0.
+        f2 := 0.0.
+        f1 / f2
+
+    or:
+        2 arcSin
+"
 ! !
 
 !Float class methodsFor:'instance creation'!
@@ -1235,5 +1258,5 @@
 !Float class methodsFor:'documentation'!
 
 version
-    ^ '$Header: /cvs/stx/stx/libbasic/Float.st,v 1.46 1996-11-05 18:40:32 cg Exp $'
+    ^ '$Header: /cvs/stx/stx/libbasic/Float.st,v 1.47 1997-01-21 10:20:33 cg Exp $'
 ! !