The Most Beautiful Experiment in Biology: A Walkthrough of the Meselson-Stahl Experiment

The Mystery of the Copying Molecule

In 1953, James Watson and Francis Crick unveiled their model for the double helix structure of DNA, a discovery that fundamentally changed our understanding of life. In their landmark paper, they noted that their model “immediately suggests a possible copying mechanism for the genetic material.” While the structure hinted at how DNA might replicate, the exact method was still a profound open question.

The central mystery facing the scientific community was clear: How does a cell accurately copy its entire genetic blueprint before it divides? To solve this puzzle, scientists first had to imagine the possibilities, leading to three main hypotheses for how this essential process could occur.

The Three Proposed Models of DNA Replication

Before experiments could provide a definitive answer, scientists considered three main possibilities for how the strands of an original DNA molecule are distributed to its descendants during replication.

Semiconservative Replication

Proposed by Watson and Crick themselves, this model suggests the two original DNA strands separate, with each serving as a template for a new, complementary strand. The result is that each daughter DNA molecule is a “hybrid,” containing one original strand and one newly synthesized strand. This model made specific, testable predictions about the composition of DNA after successive rounds of replication:

• After one round of replication: Every single new DNA molecule would be a hybrid, consisting of one old strand and one new strand.

• After two rounds of replication: Half of the DNA molecules would be hybrids, and the other half would be composed of two entirely new strands.

Conservative Replication

This model proposed that the original DNA molecule could be copied while remaining completely intact. One proposed mechanism for this involved the entire parent molecule acting as a template, possibly by having histone proteins bind to the DNA and revolve the strands to temporarily expose the nucleotide bases for copying without permanently separating them. After replication, one daughter cell would receive the completely intact original DNA molecule, while the other would receive a completely new DNA molecule.

The predictions for this model were distinctly different from the semiconservative model:

• After one round of replication: Half of the DNA molecules would be the original, “old” DNA, and the other half would be completely new DNA.

• After two rounds of replication: One-quarter of the DNA molecules would be the original “old” DNA, and the remaining three-quarters would be completely new.

Dispersive Replication

The third hypothesis, exemplified by a model from scientist Max Delbrück, suggested a “mixing” mechanism to solve the challenge of unwinding the long double helix. This model proposed that the original DNA double helix breaks apart into fragments. One suggested mechanism involved breaking the DNA backbone every 10 nucleotides or so, allowing the molecule to untwist and attach the old fragments to newly synthesized ones. The outcome would be that every new DNA molecule is a mosaic, with both of its strands containing a mixture of old and new DNA segments.

This model’s predictions focused on the progressive dilution of the original material:

• After one round of replication: Every DNA molecule would be a hybrid, with old and new DNA fragments mixed together along both strands.

• After two rounds of replication: The hybrid molecules would persist, but they would contain a greater proportion of new DNA, becoming progressively less “old” with each cell division.

Each of these models offered a unique and testable prediction about the composition of DNA after replication, setting the stage for a decisive experiment to determine the truth.

At a Glance: Comparing the Predictions

To understand how scientists could solve this puzzle, the distinct, testable predictions of each model must be compared directly. The table below summarizes the expected results for each model after the first two rounds of cell division.

Replication Model	Result After ONE Round	Result After TWO Rounds
Semiconservative	All molecules are hybrids (1 old strand + 1 new strand).	50% are hybrids; 50% are entirely new.
Conservative	50% are original; 50% are entirely new.	25% are original; 75% are entirely new.
Dispersive	All molecules are hybrids (old and new pieces mixed on both strands).	All molecules are still hybrids, but contain less of the original DNA than the first round.

Setting the Stage for an Experiment

In the years following the discovery of the double helix, the scientific community was faced with three distinct and competing ideas for how DNA copies itself. The most crucial insight for any student of biology is that because each model—semiconservative, conservative, and dispersive—made a different and specific prediction about the physical makeup of the new DNA, it created a clear path forward. Scientists could now design an experiment capable of distinguishing between these outcomes to prove which hypothesis was correct.

The Experimental Design: A Stroke of Genius

The core challenge Meselson and Stahl faced was how to physically distinguish between the “old” parental DNA and the “new” daughter DNA. Their solution was remarkably clever, combining isotope labeling with a powerful separation technique.

1. Labeling with Nitrogen: The scientists chose to label DNA with nitrogen because it is a major component of the DNA bases. They used two nitrogen isotopes: the common, lighter form (¹⁴N) and a heavier, non-radioactive version (¹⁵N). By altering the mass of the nitrogen atoms, they could create “heavy” and “light” DNA that was chemically identical but physically different in density.

2. Growing the Bacteria: They began by growing E. coli bacteria for several generations in a growth medium where the only nitrogen source was the heavy isotope, ¹⁵N. This ensured that the entire DNA of the bacterial population was uniformly labeled as “heavy.”

3. The Switch: In the experiment’s critical step, the scientists transferred the bacteria with heavy ¹⁵N DNA into a new medium containing only the light isotope, ¹⁴N. From this point forward, any newly made DNA would be synthesized using ¹⁴N and would therefore be “light.”

4. Separation by Density: To separate the DNA molecules based on their density, they used a technique called buoyant density centrifugation. They would extract DNA from the bacteria, mix it with a cesium chloride (CsCl) solution, and spin it at very high speeds. During centrifugation, the CsCl forms a continuous gradient of density, like a spectrum of saltiness from top to bottom. The DNA molecules sink until they reach their “flotation” level—the point where their own density perfectly matches the density of the CsCl solution, causing them to form a distinct band. Heavy ¹⁵N DNA would sink further and form a band lower down the tube than light ¹⁴N DNA.

With their ingenious system in place, Meselson and Stahl were ready to ask the bacteria a simple question and watch for the answer written in the language of density. The story would unfold one generation at a time.

Following the Evidence: Generation by Generation

By taking samples of bacteria after each round of cell division, Meselson and Stahl could photograph the DNA bands and directly test the predictions of the three models.

Generation 0: The Starting Line

The first sample was taken from the bacteria grown exclusively in the heavy ¹⁵N medium. This was the baseline.

• Result: When centrifuged, the photographic plate showed a single, sharp, dark band in the “heavy” position. This confirmed that their labeling and separation techniques worked perfectly.

Generation 1: The First Big Clue

Next, they took a sample after the bacteria had been transferred to the light ¹⁴N medium and allowed to divide exactly once.

• Result: The photographic plate from the centrifuge revealed the answer. The original dark band at the “heavy” position had vanished entirely. In its place was a single, new, sharp band that had settled precisely in the middle of the tube, halfway between the heavy ¹⁵N and light ¹⁴N positions.

This finding provided an immediate and crucial insight.

This single intermediate band immediately ruled out the conservative model. If that model were correct, there would have been two separate bands: one heavy (the original DNA) and one light (the new DNA).

The dispersive model also predicted a single intermediate band, as the old ¹⁵N fragments would be mixed evenly with new ¹⁴N fragments throughout all the new DNA molecules. Thus, after one generation, both the semiconservative and dispersive models remained in play.

Generation 2: The Decisive Result

The final piece of the puzzle came from the sample taken after the bacteria had completed a second round of division in the ¹⁴N medium.

• Result: After two generations, the results were definitive. The photographic analysis showed two distinct bands of equal intensity. One was the intermediate hybrid band from the first generation. The other was a new band, positioned exactly where pure, light ¹⁴N DNA would be.

This two-band result definitively disproved the dispersive model. A dispersive mechanism would have distributed the original ¹⁵N atoms among all four granddaughter molecules, resulting in a single band that was lighter than the Generation 1 band, but still heavier than the pure ¹⁴N band. The appearance of a purely light band was impossible under the dispersive hypothesis. This outcome perfectly matched the prediction of the semiconservative model.

With this clear and unambiguous evidence, the long-standing question about DNA replication was finally answered.

The Verdict and Legacy

The Meselson-Stahl experiment synthesized the evidence from across the generations to provide powerful and conclusive proof that DNA replication is semiconservative. In their own words from the 1958 paper, the authors summarized their core findings with remarkable clarity: “The nitrogen of a DNA molecule is divided equally between two subunits which remain intact through many generations,” and “Following replication, each daughter molecule has received one parental subunit.”

This confirmed the elegant mechanism first hinted at by Watson and Crick: when a DNA double helix replicates, its two strands unwind, and each serves as a template for a new, complementary strand. The result is two new DNA molecules, each consisting of one strand from the original helix and one newly synthesized strand, ensuring the faithful copying of genetic information from one generation to the next.

For its brilliant simplicity, conceptual clarity, and definitive result, the experiment has since been called “the most beautiful experiment in biology.”

Image Summary

References

M. Meselson, & F.W. Stahl, The replication of DNA in Escherichia coli*, Proc. Natl. Acad. Sci. U.S.A. 44 (7) 671-682, https://doi.org/10.1073/pnas.44.7.671 (1958).
WATSON, J., CRICK, F. Molecular Structure of Nucleic Acids: A Structure for Deoxyribose Nucleic Acid. Nature 171, 737–738 (1953). https://doi.org/10.1038/171737a0