Visual walkthrough — Chirality — chiral centres, enantiomers, diastereomers, meso compounds
Step 1 — What "four different groups" looks like in 3-D
WHAT. Start with the simplest chiral object: one carbon at the centre of a tetrahedron, with four different coloured balls at its corners.
WHY start here. Everything about chirality is born from this single shape. Before we count four stereoisomers we must be certain that one chiral centre already gives two — a left form and a right form. If we can't see the "2", we can never explain the "".
PICTURE. Look at 
Step 2 — Naming the two forms so we can count them: R and S
WHAT. Give the left form and the right form labels. We call one configuration and the other .
WHY labels. To count combinations we need a symbol per choice. Saying "left/right" is vague once a molecule has two centres; the standardised tags and come from the CIP priority rules and let us write things like "(R,S)" unambiguously.
PICTURE. In 
- clockwise (orange arrow) (Latin rectus, right),
- anticlockwise (blue arrow) (Latin sinister, left).
Each symbol in that line: the numbers are priority ranks, the arrow direction is the only thing that flips between the two mirror forms of Step 1. So one centre = one binary choice: or .
Step 3 — Two centres: the product rule builds
WHAT. Now chain two chiral carbons together, as in tartaric acid . Each centre is independently or .
WHY the product rule. If choice A has 2 options and choice B has 2 options, and they don't influence each other, the total is — the same reason two coin flips give 4 outcomes (HH, HT, TH, TT). This is the origin of the famous formula.
PICTURE. 
Here is how many independent handed carbons the molecule has; the base is the R-or-S choice per carbon. So the upper bound is 4 labels: (R,R), (S,S), (R,S), (S,R). Hold onto these four — the drama is about to happen to two of them.
Step 4 — Sorting the four into relationships (mirror or not)
WHAT. Compare the four labels pairwise. Which pairs are mirror images? Which are not?
WHY compare. The label alone doesn't tell you if two molecules are enantiomers, diastereomers, or secretly identical — you must reflect one and see what it becomes. Reflecting flips every centre: .
PICTURE. In 
- (R,R) and (S,S) are genuine mirror twins that can't be stacked → enantiomers (green link in the figure).
- (R,R) vs (R,S) are not mirror images (only one centre differs) → diastereomers (red link).
- (R,S) and (S,R) — the figure marks these with a question mark, because their reflection relationship is about to collapse. That is Step 5.
Step 5 — The trick: (R,S) is its own mirror image
WHAT. Take the (R,S) molecule, reflect it to get (S,R), then rotate it 180° in the plane of the page. Watch it land exactly on the original (R,S).
WHY this is the whole point. For the (R,R)/(S,S) pair, reflection made a new object you could never rotate back. But (R,S) has a special build: its top half is the mirror of its bottom half. So reflecting the whole molecule and flipping it end-over-end returns the same molecule. When a mirror image is superimposable on the original, the object is achiral.
PICTURE. 
So (R,S) and (S,R) are not two things — they are one thing seen twice. This molecule is a meso compound: it has two chiral centres yet is achiral overall. (The plane here is a genuine mirror plane σ.)
Step 6 — Why the meso molecule twists no light
WHAT. Explain physically why the (R,S) meso form has zero optical rotation, tying the picture back to polarimetry.
WHY it matters. A chiral centre normally rotates plane-polarised light. Meso has two centres — so why does the meter read ?
PICTURE. 
Step 7 — The corrected count
WHAT. Repair the formula so it gives the true number for tartaric acid.
WHY. We proved is only an upper bound: it double-counts each meso molecule (once as (R,S), once as (S,R)). Each meso form was counted twice, so we subtract one label per meso form.
PICTURE. 
For tartaric acid: , meso forms , so The three survivors are: (R,R) and (S,S) — a pair of enantiomers (the optically active and forms) — plus one meso (R,S). Mystery solved: the "missing fourth" was never a separate molecule; it folded onto its own reflection.
The one-picture summary
This final figure stacks the whole journey: one centre → 2 forms (R/S), two centres → 4 labels by the product rule, a mirror sorting them into an enantiomer pair (green) and a diastereomer relationship (red), and finally the meso merge (grey tile) that turns 4 into 3, with a little polarimeter reading beside it.
Recall Feynman retelling — say it to a friend who's never seen this
Picture a carbon holding four different-coloured balls at the corners of a pyramid. Its mirror image is a "left-hand / right-hand" pair you can't stack — that's two forms, and we tag them and . Bolt two such carbons together (tartaric acid) and each can be or on its own, so you expect four combos, like two coin flips: RR, SS, RS, SR. But here's the sneaky part — the RS molecule has a top half that is the perfect mirror of its bottom half. Fold it and it lands on itself, so RS and SR aren't two molecules, they're one, and it's called meso. It even twists no light: its top half spins the light right by exactly as much as its bottom half spins it left, and they cancel inside the one molecule. So the real count isn't , it's : a left twin, a right twin, and one self-cancelling meso.