OK, so let's get into this in (minimal) depth.

The reason an oil, or water, objective has a higher resolution than a water one is that the wavelength of light in oil, or water, is shorter than in air so resolution is automatically better.  However the homogeneous medium has to go all the way to the objective, otherwise the acceptance angle is reduced by the refractive index change - and since both effects depend directly on the refractive index the loss exactly cancels out the gain.  (See Chapter 1 of Optical Imaging Techniques in Cell Biology).  

Now if we have an oil immersion objective of (say) NA 1.0, it will still have that NA even imaging into water because the change in refractive index means that the lens can collect a wider angle and that will compensate exactly for the longer wavelength (as Steffen says).  However this cosy relationship breaks down once we reach the critical angle, since rays leaving the sample in a downward direction are obviously never going to reach the objective.  Likewise, in confocal illumination, rays beyond the critical angle will experience total internal reflection and will never reach the sample.  Therefore an oil immersion lens cannot have an NA of more than 1.33 when imaging into water, and every oil immersion lens of NA greater than 1.33 will have an  effective NA of  1.33 when imaging into water.  

I don't want to write pages on this, but I think if you just draw it out with a pencil and paper it will be quite clear.  There is neither magic nor deep physics involved!

                                                               Guy

-----Original Message-----
From: Confocal Microscopy List [mailto:[log in to unmask]] On Behalf Of Steffen Dietzel
Sent: Friday, 5 October 2012 10:29 PM
To: [log in to unmask]
Subject: Re: Oil vs water objectives

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

On 04.10.2012 18:20, Cammer, Michael wrote:
> Also, while people have generated various test samples of
collagen and latex and acrylic and fat globules and glass beads and carrageen gum and dissolved Scotch tape glue etc., the only test sample that really answers the question is the sample itself.

Again, I agree, in particular if you look at tissues with more or less unpredictable optical properties. Still, bead preparations allow to demonstrate the general effect very clearly and thus convincingly. They also allow quantification and sometimes hard numbers help to convince people. Some person might argue that they don't care about resolution since s/he is planning to use a 1 µm pixel size anyway. But you still can convince these people if you can show them that they will also loose a lot of intensity.
I use such a demonstration to argue for the usage of 170 µm coverslips and a good embedding medium.

On 05.10.2012 02:40, Guy Cox wrote:
> Nobody seems to have mentioned so far that the NA of an oil
objective will NOT be 1.4 if it is imaging a sample in water.
The maximum it can be is 1.33 - the refractive index of water.
...
> So the oil objective has little or no advantage in NA

Is that really true? If there is oil between the coverslip and the oil objective, but water immersion for the water objective, this should make a noticeable difference in effective acceptance angle, should it not? 
(Assuming the object is rather close to the coverslip).

Steffen


--
------------------------------------------------------------
Steffen Dietzel, PD Dr. rer. nat
Ludwig-Maximilians-Universität München
Walter-Brendel-Zentrum für experimentelle Medizin (WBex) Head of light microscopy

Mail room:
Marchioninistr. 15, D-81377 München

Building location:
Marchioninistr. 27,  München-Großhadern