Microscopy resolution is not magnification

Magnification enlarges a disc of blur. Resolution decides whether two discs were ever separate. The decision this explainer supports is whether an objective and a camera can carry a spatial claim, or whether the image only looks more magnified. Counting fluorescent cells without caring where the light sits inside them is the cytometer's job, described in how a flow cytometer reads a cell. The controls that stop a bright picture becoming a false label are in fluorescence microscopy without the myths.

Objective classes are in the scientific instruments catalogue. Imaging as part of a larger experiment is the molecular biology pathway. Numerical aperture and camera pixel size belong in a quote request.

Numerical aperture does the work

Numerical aperture, engraved as NA, describes the cone of light an objective can gather. A wider cone collects more of the diffracted light that carries fine detail, and it collects more fluorescence photons as well. Magnification, engraved beside it, tells you how large the intermediate image is. A 100× objective with a modest aperture can resolve less than a 60× or 40× objective with a high aperture. The barrel is the specification. A habit of always clicking to the highest magnification click-stop is how empty enlargement becomes the default.

Wavelength sits in the same relationship. Shorter wavelengths can carry finer detail than longer ones, which is why a blue fluorophore can, in principle, be resolved more finely than a far-red one through the same aperture. In fluorescence you often do not get to choose the wavelength the biology needs. You do get to choose an objective whose aperture is high at that wavelength, and whose transmission does not die in the red if the dye emits there.

The Abbe relation as a concept

Ernst Abbe's account of the microscope ties the smallest resolvable distance to wavelength divided by numerical aperture. A familiar simple form says that distance scales as the wavelength over twice the numerical aperture. The Rayleigh criterion, used for the separation of two point sources, has a different numerical factor and the same dependencies. You do not need to memorise a constant to make the decision. If you halve the aperture you roughly double the blur. If you move from green emission toward deep red you lose some resolving power. Immersion that the objective was designed for is what lets a high aperture exist at all. An oil objective used dry is not a slightly worse version of itself. It is a different, poorer optical system.

These relations describe a lateral distance in the focal plane under good conditions. Thick specimens, refractive-index mismatch, a wrong coverslip and a closed condenser or a stopped-down aperture diaphragm all make the real image worse than the engraving promised. The formula is an upper bound on hope, not a measurement of your slide.

Empty magnification and the camera

Once the objective has set the blur disc, the eyepiece or the camera magnifies that disc. If the disc already covers many pixels, further magnification spreads it over still more pixels and the eye thinks it sees texture. That is empty magnification. The practical check is sampling: the pixel size at the specimen, after objective and any relay lens, should be small enough to put more than one pixel across the resolved distance, and it need not be tiny beyond that. A stage micrometer or the microscope's calibrated pixel size tells you the number. A software zoom applied to a saved image does not create pixels the camera never collected.

Binning combines pixels and makes each sample larger. It helps a dim image and can push you into undersampling, where the pixel, not the aperture, sets the detail. Gain does not change resolution. It changes the numbers attached to the same blur. Confocal zoom that reduces the scanned field can change sampling. It still cannot outrun the objective's aperture by very much. The MicroscopyU fluorescence section and the Evident fluorescence primer discuss these optical ideas at primer length. They are not a warrant to claim a distance smaller than the objective allows.

Claims that resolution does not support

Two colours that overlap inside one blur disc are not proven to occupy the same molecule. They are unresolved. A brighter spot is not a smaller spot. Intensity and resolution are different axes, and saturation makes intensity unmeasurable without fixing resolution. A cropped image at high display magnification still has the pixel size of the acquisition. Write the objective magnification, the numerical aperture, the immersion, and the micrometres per pixel. A reader can then judge the claim. "High magnification" cannot.

A cytometer plot that cleanly separates two markers does not add spatial resolution. It never looked inside the cell. Quote it as a count. Quote the micrograph as a location, and only down to the distance the aperture and the sampling allow.

Sampling belongs in the same note as the aperture. If the resolved distance covers fewer than about two pixels, the sensor is limiting you, and a different magnification or relay may be the honest next step. If it covers a long run of pixels, you are enlarging blur and a lower power may give a wider field without losing the claim. Count pixels across a small object you believe is at the limit, using the calibrated pixel size, and write that count next to the objective you used.

Safety and research limits

Resolution is not a biosafety control. The specimen on the stage may be unfixed human or infectious material, which the institution assesses using the WHO Laboratory biosafety manual and the CDC BMBL. This explainer does not approve that placement. It also does not approve a diagnostic reading of a research image. Choosing an objective is a methods decision.

What to send with an enquiry

State the smallest distance you need to separate, the emission wavelengths, whether the sample is live in medium or fixed under a coverslip, and the camera pixel size if you already have a camera. Ask for numerical aperture, immersion class and coverslip specification, not only magnification. Use the scientific instruments catalogue and the quote request. Ask whether a quotation is possible. A 100× label without an aperture is not a resolution specification.