Starting with something perhaps more familiar, like a diatom, here I wanted to show an example of what different optical sections can reveal about an object in the microscope. Experience with beat-up student scopes (or none at all) often leads you to think the world in the microscope as being two dimensional — depth somehow simply vanishes on that scale. Usually you go as far as something being in focus or out of focus, and that’s it. Some microscopy techniques don’t allow for optical sectioning much, and can be particularly thrown off by thick specimens (everything fuses into one big blurry mess). Differential Interference Contrast (DIC) microscopy is the mode I use most often — largely because I lack access to phase at the moment — which is arguably less pretty, but allows to see fine structures in much greater contrast. Basically, it uses polarised light voodoo whose physics I still only vaguely understand (as do most biological microscopists, it seems…) to generate a pseudo-3D effect. I emphasise ‘pseudo’ because you’re not actually seeing the shape of the specimen, but rather the pattern generated by different materials having different light polarising properties. This can be misleading in some cases: for example, folds can appear as thickenings, or highly refractile granules of some polarising-happy material can distort the appearance and shape of the object they’re in. But overall, it’s a popular technique, and is amenable to sexy images.
Another major benefit of DIC, in addition to an appealing relief-style contrast, is that it allows for optical sectioning — in my understanding, a lot more of the out-of-focus light gets excluded than in other techniques (eg. brightfield or phase), which allows you to view different focal planes with only a relatively slight disturbance from the stuff out of focus. This really allows one to appreciate the three dimensional nature of the microbial world, and once again, for a different reason than the relief and shadows would first suggest!
So first off, a classic valve view of a pennate diatom (don’t know the species). Along the middle you can see the raphe, which is a slit used in gliding motility (yes, these diatoms can glide about, and sometimes quickly too!).
This was just at the surface of the diatom, and close to the coverslip itself (where the best image quality generally resides; the further below a coverslip you have to go, the crappier the image gets). Just below the frustule (glass house of the diatom), you can see the surface of its plastid, here screaming quite loudly, and accurately, about its brown algal identity (‘brown algae’ can also be all sorts of funky shades of not brown, like green, yellow and red).
Let’s scroll a bit further down. Here you can see the edges of the same plastid we saw above — it forms a contour along the frustule. This might make sense, as that way most of the plastid is exposed to light: there’s not much point in burying photosynthetic organelles deep inside a cell. The brown you see is now coming from both the top and the bottom ends of the plastid — there’s a layer on the bottom too. The rather prominent small blob near the middle is a lipid vesicle of some sort — lipids tend to be quite refractile and happy to polarise anything in sight: you can infer a bit about the material based on how it behaves under DIC and other lighting conditions. Now for the subtly granular stuff in the middle — that’s the nucleus! You can make out the clumpiness of chromatin, as well as a nucleolus — the smooth round thing just above the centre of the nucleus. Probably thought one needed to go all the way to electron microscopy to see the nucleolus, eh? Granted, this is at about the limit of light microscopy’s resolving power. This was using a 100x objective, but you won’t see more detail if you went 200x or 500x (I don’t think either of those even exist). The theoretical limit of visible light microscopy is at 200nm at best, but realistically you’ll probably only get to resolve details at about 400nm, probably much worse if your lens are dirty and you get bored too easily to deal with aligning the light path (yes, that happens in the lab too).
NB: There are new developments allowing us to surpass that resolution limit (just look up superresolution microscopy), but those usually require a rather sophisticated preparation of the specimen (fluorescence is a must) and, in my opinion, a good understanding of the physics behind it to ensure you’re still seeing some reality there. I don’t get to play with such toys yet…
This is it for the views intended, but I figured I can show you what happens as you go further down, and what happens to the image quality. Usually I try to get the whole stack because as the crappier bottom sections may still hold valuable information, since not everything is conveniently symmetrical like a diatom! Note how the bottom valve view of the frustule didn’t really work out at all.