>I agree that the cause of these focus shifts and halos is often chromatic
>aberration but I don't believe that lateral aspect of the problem has
>anything to do with misalignment.
>
>Chromatic abberation refers to tendency of the focal length of any optical
>system to vary with wavelength. (axial chromatic aberration: hence the
>focus-plane shift problem).
>
>However, this exact phenomenon is also the cause of "lateral chromatical
>aberration" in beam-scanning systems.  This is because the distance that a
>point in the image appears to be off the axis (read: x-y mirror position)
>depends on the magnification of the optical system between the source (or
>pinhole) and the focus plane in the specimen.  This, optical magnification,
>in turn, depends inversely on the focal length of the objective.
> Therefore, if the focal length is shorter for the exciting wavelength than
>for the emitted wavelength, the returning spot will be magnified less and
>hence appear to be closer to the axis.  If a small pinhole is used, the
>returning ray may miss the pinhole entirely and the image will appear dark.
>If a larger pinhole is used, then the majority of the signal will be
>recorded when, in fact an adjacent pixel is being illuminated. If this
>effect is present, then the image will match will be worse as one moves
>away from the axis.
>
 
 
One more comment on this:
 
        Many microscopes allow you to shift the scanned area around on
the specimen.  If you do this then lateral chromatic aberration
(chromatic difference of magnification) will be manifested as an
overall displacement of one image relative to the other.  You are, in
effect, misaligning the scan head relative to the optic axis.  Furthermore,
if the difference between exciting and emission wavelengths is about 60nm
in each channel, you could, even with a single pinhole closed well down,
be 'correctly' aligned yet be collecting each channel from a different
point on the specimen!!
 
        If you don't displace the scan in this way you will instead get
one image a different size from the other - a point which has already
been noted by another contributor to this discussion.  But if you zoom
up taking care to stay on axis it should be minimised, since the displacement
will be less than the size of the Airy disk.
 
        Of course none of this should ever happen since lateral CA is
in principle correctable by the projection eyepiece or tube lens.  In
practice things don't work out quite so perfectly!  A good check is
to do a two or three colour reflection image.  Since each wavelength
will now retrace its own path back through the lenses, pinhole alignment
will not be a problem - but if lateral CA is present the images will
not match!
 
                                        Guy Cox