Inheritance and DNA · Investigation
How we worked out DNA's structure
Nobody has ever seen a DNA molecule. Its structure was deduced from a photograph of scattered X-rays, a table of chemical ratios, and a set of metal models built on a bench in Cambridge.
Start here
A molecule far too small to see, worked out in 1953.
DNA is a hundred thousand times thinner than a hair. No microscope of the 1950s could resolve it, and no microscope today can show you the shape of a single molecule directly. Yet by April 1953 the structure was published, and it was right.
What kind of evidence could possibly show the shape?
X-ray diffraction. Fire X-rays at a crystal or fibre and they scatter off the atoms in a pattern; the pattern is not a picture of the molecule, but its geometry can be worked backwards to the arrangement that produced it. Rosalind Franklin was one of the best in the world at it, and the image known as Photo 51 is the reason this lesson exists.
At the bench · the evidence available in 1952
Build a model that fits all of it
0 of 4 tests passed
Three decisions to make, and four pieces of evidence that each rule something out. Exactly one combination survives all four — which is what Watson and Crick were doing with pieces of cardboard and bent metal.
How many strands
Where the bases sit
How the bases pair
Three strands, bases on the outside, any base with any base
Photo 51 · Franklin and Gosling, 1952
consistent
rules this model out
An X-ray diffraction image of DNA fibres. The cross-shaped pattern of spots is the signature of a helix, and the spacing of the spots gives its width and the distance per turn.
The measured width does not fit this. A single strand is too narrow and three strands are too wide — the pattern says two.
Franklin’s water measurements
consistent
rules this model out
DNA takes up a great deal of water, which means its water-attracting parts — the phosphate groups — must be on the outside of the molecule, in contact with it.
If the bases are on the outside then the phosphates are on the inside, which contradicts the water measurements — and the negative phosphates would repel each other.
Chargaff’s ratios, 1950
consistent
rules this model out
In DNA from any organism, the amount of A always equals the amount of T, and the amount of C always equals the amount of G — while the ratio of A to C varies from species to species.
If any base could pair with any other there would be no reason for A and T to come in equal amounts in every organism ever measured.
Pauling’s triple helix, early 1953
consistent
rules this model out
The most respected chemist alive published a model with three strands and the bases facing outwards. It was wrong, and seeing why it was wrong narrowed the field for everyone else.
This is Pauling’s model, and it has already been ruled out. Its phosphates were crowded into the centre where their negative charges would push the molecule apart.
4 tests still failing
Each failing test is telling you which decision to change, and none of them requires you to know the answer in advance — which is exactly the position everyone was in during 1952.
Who did what
Four names, two laboratories, one structure.
Rosalind Franklin
King’s College London · X-ray crystallography
Produced the diffraction images, including Photo 51 with her student Raymond Gosling, and calculated from them that the molecule was a helix of a particular width with the phosphates on the outside. Hers is the evidence the model had to fit.
Maurice Wilkins
King’s College London · X-ray crystallography
Began the X-ray work on DNA and continued it alongside Franklin, in a working relationship that was difficult from the start. He showed Photo 51 to Watson.
Erwin Chargaff
Columbia University · biochemistry
Measured the amounts of the four bases in DNA from many species and found the equalities that carry his name — without proposing what they meant. The pairing rule was sitting in his tables for three years.
Watson and Crick
Cambridge · model building
Built physical models and tested them against everyone else’s measurements until one fitted all of them. They did no experiments on DNA themselves, which is not a criticism — model building is a legitimate method, and it was the right one here.
Key fact
DNA is a double helix: two strands twisted round each other, with the bases paired on the inside — A always with T, C always with G. The structure was deduced in 1953 from X-ray diffraction images taken by Rosalind Franklin and Maurice Wilkins, and from Erwin Chargaff's base ratios, by James Watson and Francis Crick.
Think again
“Watson and Crick discovered DNA.”
DNA was discovered in 1869 by Friedrich Miescher, who isolated it from the pus on used surgical bandages and called it nuclein — eighty-four years before 1953. By the 1940s Avery, MacLeod and McCarty had shown it was the material carrying inherited information, which was the genuinely surprising result, since most biochemists had assumed the genetic material must be protein. What Watson and Crick worked out was the structure, and even that was model-building from other people's measurements: they did no experiments on DNA themselves. Getting this right is not pedantry about credit. It is the difference between imagining science as a series of individual discoveries and understanding it as what it is — a slow accumulation where the person who assembles the final piece is standing on twenty years of other people's work.
“A great discovery is one person's flash of insight.”
Look at what 1953 actually required. Franklin and Wilkins produced the diffraction images at King's College London and Franklin calculated the measurements from them. Chargaff, in New York, had published the base ratios years earlier without seeing what they meant. Linus Pauling in California published a triple-helix model that was wrong, which mattered, because ruling out a serious rival narrows the field. Watson and Crick built physical models in Cambridge and tested them against everyone else's numbers until one fitted. Five laboratories, three countries, two decades. The flash-of-insight story survives because it is easier to tell, and it gives students a false picture of what doing science looks like: mostly it looks like checking whether your idea is consistent with someone else's data.
Mastery ladder
Not started yet.
Rungs 3 and 4 you mark yourself.
Rung 1 · What the evidence showed
What did Chargaff’s ratios contribute to the model?
Rung 2 · The one that catches people
What did Watson and Crick actually do in 1953?
Rung 3 · Match evidence to conclusion
For each of the three features of the DNA model — two strands, bases on the inside, and A pairing with T — say which piece of evidence established it and how.
Rung 4 · Take it somewhere new
The 1953 paper acknowledged only that the authors had been "stimulated by" unpublished results from King’s College. Using what you know about how the structure was worked out, write what you think a fair acknowledgement would have said — and explain what rule about using other people’s data you would draw from this episode.
Key note
DNA's structure — a double helix with paired bases inside — was deduced in 1953 from evidence nobody could see directly. Franklin and Wilkins produced the X-ray diffraction images; Franklin's measurements showed the helix's dimensions and that the phosphates lie outside; Chargaff's ratios showed A pairs with T and C with G; Watson and Crick built the model that fitted all of it. The structure immediately suggested how DNA could be copied.
Going further
The paper announcing the structure ran to about a page, and it closed with a single sentence, phrased with deliberate understatement, noting that the pairing they had proposed immediately suggested a way the material could copy itself. That sentence is the reason the structure mattered so much. If A always pairs with T and C always with G, then each strand carries the full information: unzip the double helix and each half specifies exactly what must be built alongside it. A structure that explains, in its own geometry, how inheritance can work is a rare thing in science — most structures have to have their function worked out afterwards. Five years later Meselson and Stahl showed experimentally that copying does happen that way, and the model stopped being a proposal.
A note on the credit
Photo 51 was taken in Franklin's laboratory by her research student Raymond Gosling. It was shown to Watson by Wilkins without her knowledge, and a report containing her unpublished measurements reached Crick through a research council committee, also without her being asked. Watson and Crick's 1953 paper acknowledged only that they had been "stimulated by" unpublished results from King's. Franklin died of ovarian cancer in 1958, aged 37, probably from her exposure to X-rays; the Nobel Prize was awarded to Watson, Crick and Wilkins in 1962, and the rules do not allow it to be given posthumously. Historians still argue about how much of this was ordinary scientific rivalry, how much was the treatment of a woman in a hostile department, and how much was the Nobel's own rules. What is not in dispute is that the data were used without permission and that the acknowledgement understated what was owed.
Before this lesson
Connects to
At GCSE this becomes
- The structure of DNA in detail, the history of the model as an example of how science works, and DNA replication.
Where to next
Ask Mr Badmus AI
Want to talk through what Photo 51 actually shows?
The bench is a simplification of a two-year argument into three decisions and four tests. The real work involved crystallography mathematics well beyond this level, several intermediate wrong models by everyone concerned, and evidence not listed here. The historical account in the credit note follows the mainstream of what historians of science have established; several details remain contested.
Lesson content © MrBadmusAI.