Cellular Respiration Diagram: Where Each Stage Happens, How Much ATP to Label, and Why 38 Is the Wrong Number

What does a cellular respiration diagram show?

It shows a pathway placed in compartments. That is the part worth being careful about, because it is the part a diagram can get wrong in a way a paragraph cannot.

Two things on the diagram are claims. The order of the stages, and the compartment each stage is drawn inside. Both are checkable, both are stated in the sources at the bottom of this page, and a diagram that misplaces either one is teaching an error no amount of good drawing will fix.

Everything else is a drawing decision. The colours are conventional. The Krebs cycle is drawn as a circle because it returns to oxaloacetate, not because anything in the matrix is arranged in a ring. The electron transport chain is drawn left to right because the page runs that way. None of that is evidence, and none of it is worth arguing about.

There is one more thing a respiration diagram carries that a cell diagram does not: arithmetic. Yields, in ATP. Those numbers are printed on the figure as though they were measurements, and the section after next is about what they actually are.

The three stages, and where each one happens

Alberts is explicit about the compartments. The matrix enzymes are the ones that "metabolize pyruvate and fatty acids to produce acetyl CoA" and the ones that "oxidize acetyl CoA in the citric acid cycle", while the respiratory chain lives in the inner membrane. Glycolysis is not in that list because it is not in the mitochondrion at all.

Two rows deserve a second look.

The citric acid cycle runs twice per glucose. Alberts gives the per-turn yield: each turn produces two CO₂, three NADH, one FADH₂ and one GTP. One glucose gives two pyruvate, so two acetyl CoA, so two turns. A diagram that draws the cycle once and labels it with the per-turn numbers has halved the pathway without saying so, and it is a common enough error that the figure below marks the multiplier on the circle itself.

The chain's ATP figure is a derived number, not a measured one. It is what you get after deciding how many ATP a NADH is worth, which is the next section.

How much ATP should the diagram say?

Here is the honest version, which almost no page on this topic gives you: no authority publishes a canonical total, and two standard textbooks that are both free and both online right now disagree by eight ATP.

Alberts, in Molecular Biology of the Cell, says the complete oxidation of one glucose to H₂O and CO₂ "is used by the cell to produce about 30 molecules of ATP". Cooper, in The Cell, counts 4 ATP made directly plus "an additional 32 to 34 ATP molecules by oxidative phosphorylation", which totals 36 to 38. Both are cited in the sources below. Neither is a typo.

The disagreement is about how many ATP one NADH is worth, and that number was not settled by counting molecules. It comes from the machine.

Walker's group solved the structure of the bovine ATP synthase rotor and found a ring of eight c-subunits. Each c-subunit carries one proton per rotation, and one full 360° rotation makes three ATP, so the enzyme's own cost is 8 ÷ 3 = 2.7 protons per ATP. Add one more proton to import phosphate and swap ADP for ATP across the inner membrane, and the mitochondrion pays 3.7 protons per ATP. With ten protons pumped per NADH and six per FADH₂, that gives P/O ratios of 10 ÷ 3.7 and 6 ÷ 3.7, or 2.7 and 1.6 — which the paper notes are "close to experimental values of 2.5 and 1.5".

Run the pathway on those numbers:

The 36–38 figure comes from the older whole-number convention of 3 ATP per NADH and 2 per FADH₂. That convention was a reasonable estimate before the rotor was counted. It is not a reasonable thing to print on a diagram as a fact today.

So label one of three things, and be explicit about which: the range 30–32, the single figure your syllabus examines on, or both with a line saying why they differ. What a good diagram should never do is print 38 with the confidence of a measurement.

Where the electron transport chain actually sits

Not on the outer membrane. On the inner one, and more precisely on the cristae, which are the infoldings of the inner membrane that project into the matrix.

The reason to fold is surface area, and Alberts puts a number on how far it goes: in a liver cell the inner mitochondrial membrane accounts for about a third of the total membrane in the whole cell. That is worth holding next to our labelled animal cell diagram, which works through how much of a cell mitochondria really occupy: rather more than the two small ovals most drawings give them. The folding also tracks demand — a mitochondrion in a cardiac muscle cell has three times as many cristae as one in a liver cell.

Almost every diagram draws those folds as smooth shelves running most of the way across, with ATP synthase spaced evenly along them like fence posts. Two things are wrong with that, and both are now well enough established to draw differently.

A crista is not a shelf, and its opening is a neck. Cristae connect to the boundary membrane through narrow constricted openings called crista junctions, held in shape by the MICOS complex and OPA1. The constriction is functional, not incidental: it keeps the crista lumen as a separate compartment that can hold its own proton concentration. When those junctions widen, cytochrome c escapes the lumen and the cell commits to apoptosis, which is a fairly emphatic demonstration that the neck matters.

The synthase is not evenly spaced. ATP synthase assembles into V-shaped dimers that line up into long rows, and those rows sit at the highly curved rims of the cristae — they are what bends the membrane into a rim in the first place. The proton pumps, complexes I, III and IV, sit on the flat slopes. The arrangement puts source and sink close together, so protons cross a short distance from pump to synthase rather than diffusing across the whole membrane.

There is a third simplification worth knowing about, even though most diagrams are right to keep it. Schägger and Pfeiffer showed in 2000 that the respiratory complexes are not floating independently in the membrane: in bovine heart mitochondria all of complex I is bound up in supercomplexes with complex III and up to four copies of complex IV, assembled into what they named a respirasome. Drawing four separate complexes in a row is a schematic of the electron path, not a picture of the structure. That is a defensible choice for a teaching figure. It stops being defensible the moment the figure is captioned as showing the arrangement of the complexes.

How simple can the diagram get and still be right?

Very simple. The thing to hold onto is not the number of stages, it is the boundary.

Strip a respiration diagram down and it survives losing almost everything: the individual NADH counts, the names of the intermediates, the distinction between pyruvate oxidation and the cycle proper, the complexes as separate objects. A GCSE-level figure with three arrows and two compartments is not a worse diagram, it is a diagram aimed at a different question.

What it cannot lose is the line between cytosol and mitochondrion. Drop that and the figure stops making its only structural claim, and it becomes very easy to draw glucose entering the mitochondrion — which is the single most common error in the simplified versions.

A working order for cutting detail, from safe to reckless:

• Drop the intermediate names. Glucose, pyruvate, CO₂, H₂O and ATP carry the story on their own.
• Merge pyruvate oxidation into the cycle. Two stages instead of three, and no claim is lost as long as both are inside the matrix.
• Draw the chain as one band on the inner membrane rather than as numbered complexes. The chain's job is a gradient, and the gradient is what the arrow should show.
• Drop the per-stage ATP counts and keep only the total. Better an honest total than four sourceless subtotals.
• Do not drop the compartments, do not drop the CO₂ exit, and do not merge the O₂ and CO₂ arrows into one.

The general rules for keeping a figure legible once it is this stripped down — type sizes, line weights, what survives reduction — are the same ones that apply to any figure, and they are covered in our guide to making scientific figures.

The same diagram without the labels

A blank version is a different tool, not the same figure with the text deleted, and the difference is which labels you remove.

Remove the compartment names and the stage names. Keep the arrows, keep the membranes, keep the boxes. What you are testing is whether a student can say where a stage happens, and the arrows are the scaffold that makes the question answerable rather than a guessing game.

Two ways to set it up, depending on what you are marking:

• Numbered blanks. Replace each label with a number and give a separate answer list. Faster to mark, and the numbers keep the layout stable between the blank and the labelled version, so a student can hold both side by side.
• Empty leader lines. No numbers, just lines ending in blank space. Harder, because the student has to produce the term rather than match it, and it makes an unlabelled diagram usable as an exam-practice sheet.

One warning that applies more to blanks than to labelled figures. A worksheet propagates whatever it asserts, and it does so with the teacher's authority behind it. If the blank diagram puts glycolysis inside the mitochondrion, every student who fills it in correctly according to the answer key has learned something false, and they will have learned it in their own handwriting. Check the compartments before you check anything else.

Photosynthesis and cellular respiration in one diagram

The two equations are mirror images, which is why they are taught together and why so many diagrams pair them. Respiration takes glucose and O₂ and gives back CO₂ and H₂O; photosynthesis does the reverse.

The pairing is genuinely useful, and it has one trap in it. Photosynthesis is not cellular respiration run backwards. The equations reverse, the pathways do not: respiration dismantles glucose through glycolysis and the citric acid cycle, while photosynthesis builds it through the Calvin cycle, a different set of reactions with different enzymes in a different organelle. Writing one equation as the other reversed is fine. Drawing one pathway as the other reversed is not.

What the two do share is the mechanism at the end. Both run an electron transport chain across a membrane, both use it to pump protons into a closed compartment, and both let those protons back through an ATP synthase. That shared mechanism is chemiosmosis, and it is the most interesting thing a combined diagram can show.

This is also where the popular Venn treatment does badly. A Venn diagram asserts overlap between two sets, and the honest relationship here is a flow: the products of one are the reactants of the other, in a loop that runs through the whole biosphere. A cycle diagram with two organelles and four arrows says that. Two overlapping circles say something vaguer. If a Venn is what the assignment asks for, put chemiosmosis, ATP synthase, electron transport chains and membrane compartments in the intersection, and keep the organelles, starting materials and end products in the two outer fields.

Six things cellular respiration diagrams get wrong

38 ATP printed as a fact. The measured proton cost puts the total nearer 30 to 32. Print a range, or print a convention and name it.

Glycolysis drawn inside the mitochondrion. It is a cytosolic pathway. This error is almost always a side effect of dropping the compartment boundary to save space.

An arrow from O₂ to CO₂. The carbons in the CO₂ come from glucose, released in the matrix. The O₂ goes to the end of the chain and comes out as water. Two separate arrows, two separate exits.

The chain on the outer membrane. The outer membrane is permeable and holds no gradient. Everything about oxidative phosphorylation depends on the inner membrane being the tight one.

Evenly spaced ATP synthase on a smooth shelf. The dimers sit in rows at the curved rims and the pumps sit on the slopes. If the figure is a schematic of the electron path, say so in the caption rather than implying it is a map.

One turn of the citric acid cycle per glucose. Two acetyl CoA, two turns. Label the multiplier on the circle so the per-turn yields cannot be mistaken for per-glucose totals.

Frequently asked questions

What does a cellular respiration diagram show? One glucose taken apart in three stages, each drawn in the compartment where it happens: glycolysis in the cytosol, pyruvate oxidation and the citric acid cycle in the matrix, the electron transport chain in the inner membrane. The order and the compartments are the claims; everything else is drawing.

How many ATP does cellular respiration produce? About 30 to 32 by the current accounting, 36 to 38 by the older convention. Alberts prints the first, Cooper prints the second, and both are in circulation. The gap comes from counting protons rather than assuming whole numbers.

Where does each stage take place? Glycolysis in the cytosol; pyruvate oxidation and the citric acid cycle in the mitochondrial matrix; the electron transport chain and ATP synthase in the inner mitochondrial membrane, on the cristae.

Does oxygen become carbon dioxide? No. The CO₂ carbons come from glucose; the O₂ is the final electron acceptor and ends up as water.

How many turns of the Krebs cycle per glucose? Two — one per acetyl CoA, and one glucose yields two. Per turn: two CO₂, three NADH, one FADH₂, one GTP.

What is the difference between aerobic and anaerobic respiration? The terminal electron acceptor. With oxygen, roughly 30 to 32 ATP per glucose. Without it, glycolysis alone at a net 2, with fermentation to lactate or ethanol regenerating the NAD⁺ that keeps glycolysis running.

How do you draw a simple cellular respiration diagram? Boundary first, then three arrows: glucose to pyruvate outside the mitochondrion, pyruvate into the matrix and around the cycle, electrons down to the inner membrane. Mark the CO₂ leaving and the H₂O forming.

Where to go next

• The labelled animal cell diagram — where the mitochondrion sits among the other organelles, and how much of a cell it actually occupies, which is more than most diagrams suggest.
• The plant cell diagram — the chloroplast drawn the same way, for anyone building the paired photosynthesis figure.
• How to make scientific figures — type sizes, line weights and the reduction test, for turning any of this into a figure a journal will print.