NCERT grounding
NCERT Class XII Biology, Chapter 5, Section 5.4 introduces replication by recalling that Watson and Crick, "while proposing the double helical structure for DNA, had immediately proposed a scheme for replication of DNA." The textbook then quotes their 1953 paper directly and names the scheme it describes — separation of the two strands, each acting as a template, so that "each DNA molecule would have one parental and one newly synthesised strand." This is the definition of semi-conservative DNA replication, and it is the single most examined idea in this section.
"It has not escaped our notice that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material."
— Watson and Crick, 1953 (quoted in NCERT, Section 5.4)
Section 5.4.1, "The Experimental Proof," carries the verification: replication "is now proven" to be semi-conservative, "shown first in Escherichia coli and subsequently in higher organisms, such as plants and human cells." It then describes the 1958 experiment of Matthew Meselson and Franklin Stahl, and the parallel experiment by Taylor and colleagues on Vicia faba. Every claim in this article is anchored in that section; the NIOS supplement (Chapter 23.6) adds that "the semiconservative mode of DNA replication was experimentally proven by Meselson and Stahl."
Three models & the Meselson–Stahl proof
Once the double helix was on the table, the question was not whether DNA copies itself but how the parental strands are distributed into the two daughter molecules. Three logically possible schemes were proposed. The Meselson–Stahl experiment was designed with surgical precision to make these three schemes give different, measurable outcomes — and so to eliminate two of them.
The Watson–Crick prediction
The double helix is held together by complementary base pairing: adenine with thymine, guanine with cytosine. Because the sequence of one strand fixes the sequence of the other, each strand carries the complete information needed to rebuild its partner. Watson and Crick reasoned that during replication the two strands simply unwind and separate, and each old strand serves as a template on which a new complementary strand is laid down. The result is two daughter duplexes, each made of one conserved parental strand and one newly synthesised strand — hence the name semi-conservative. The parental molecule is "half-kept" in each product.
The defining ratio
Every daughter duplex retains exactly one of the two original strands. The other strand is built fresh by template-directed polymerisation. "Semi" means half — not the whole molecule, not none of it.
The three competing models
Before 1958, three hypotheses were each consistent with the structure of DNA. They differ entirely in what happens to the parental strands.
Read each model by one question: after copying, where have the two original parental strands ended up?
Conservative
The parental duplex stays fully intact. Both new strands pair only with each other, forming an entirely new duplex.
Product: one all-old + one all-new molecule.
Semi-conservative
The strands separate; each templates a new partner. Every daughter has one old, one new strand.
Product: two identical hybrid (old + new) molecules.
Dispersive
Old and new DNA are interspersed along both strands of both daughters — patches of parental and patches of new.
Product: two molecules, each a patchwork of old and new.
All three models duplicate the genetic information faithfully. They cannot be told apart by simply asking whether the daughter cells are normal. They can be told apart by physically labelling the parental strands and watching where the label goes — which is exactly what Meselson and Stahl did.
Figure 1. The three models trace different fates for the parental strands. Conservative keeps the old duplex whole; dispersive scatters old DNA in fragments; semi-conservative cleanly splits the strands, giving two hybrid daughters.
Designing the experiment: heavy nitrogen and the density gradient
Meselson and Stahl needed a way to tell a "parental" strand from a "new" strand without disturbing the chemistry of replication. Their solution was to make the two kinds of strand differ in density. They grew E. coli for many generations in a medium where the only nitrogen source was 15NH4Cl. 15N is the heavy isotope of nitrogen; it is non-radioactive but heavier than ordinary 14N. Over many generations the heavy nitrogen was built into the nitrogenous bases of every DNA molecule, producing uniformly "heavy" DNA.
They then transferred the bacteria abruptly into a medium containing only ordinary 14NH4Cl. From that moment, every newly synthesised strand would be "light," because the cell could only draw on 14N to build new bases. Samples were withdrawn at intervals matched to the bacterial generation time, the DNA extracted as intact double-stranded helices, and each sample spun in a caesium chloride (CsCl) density gradient. Under prolonged centrifugation CsCl forms a smooth density gradient in the tube; a DNA molecule migrates to the position where its own density matches the surrounding solution, settling as a discrete band.
15N is not radioactive
Students routinely call the Meselson–Stahl label "radioactive nitrogen." NCERT explicitly cautions otherwise: "15N is not a radioactive isotope, and it can be separated from 14N only based on densities." The detection method is density, not radioactivity.
Rule: 15N = heavy, stable, non-radioactive → separated by CsCl density-gradient centrifugation. The radioactive label belongs to Hershey–Chase (32P, 35S) and to Taylor's thymidine.
The result: three bands tell the whole story
The pattern of bands at successive generations is the heart of the experiment. E. coli divides roughly every twenty minutes, so each sampling time corresponds to a clean generation.
Figure 2. Generation 0 is a single heavy band. After one round of replication (Generation I) a single hybrid band appears — and only the semi-conservative model predicts this. With each later generation the hybrid band persists but light DNA accumulates, shifting the hybrid : light ratio toward more light DNA.
Meselson–Stahl: generation by generation
-
Gen 0
All heavy
Cells grown long-term in 15N. Every DNA molecule sits as one dense band.
100% heavy -
Gen I
One hybrid band
After 20 min in 14N. Each duplex = one 15N + one 14N strand.
100% hybrid -
Gen II
Hybrid + light
After 40 min. Hybrid duplexes split; half stay hybrid, half become light.
50% : 50% -
Gen III+
Hybrid fades
Hybrid fraction halves each round but never vanishes — two 15N strands persist forever.
25% hybrid
Why the result eliminates the conservative and dispersive models
The decisive observation is the single hybrid band at Generation I. Run each model forward and check it against that fact.
Conservative — rejected
- Parental heavy duplex stays whole; new duplex is fully light.
- Predicts two bands after Gen I — one heavy, one light.
- No intermediate band is possible at any stage.
- Observed result is a single hybrid band → model fails.
Dispersive — rejected
- Old and new DNA mixed within both strands of every duplex.
- Predicts one hybrid band at Gen I — matches so far.
- But predicts a single band of steadily decreasing density thereafter.
- It can never resolve into separate hybrid and light bands → model fails at Gen II.
The conservative model is killed at Generation I: it demands two bands where only one appeared. The dispersive model survives Generation I but is killed at Generation II: it predicts a single, ever-lightening band, whereas the experiment showed two distinct bands — a hybrid band and a light band — coexisting. Only the semi-conservative model fits both observations: a hybrid-only first generation, followed by the splitting of hybrid duplexes into one hybrid and one light daughter, so that hybrid and light DNA are seen side by side from Generation II onward.
A single hybrid band after one generation, then hybrid and light bands together — a fingerprint no model but semi-conservative replication can produce.
Meselson & Stahl, 1958
Taylor's confirmation in Vicia faba
Meselson and Stahl proved the point for a prokaryote. NCERT notes that "very similar experiments involving use of radioactive thymidine to detect distribution of newly synthesised DNA in the chromosomes was performed on Vicia faba (faba beans) by Taylor and colleagues in 1958." Thymidine is incorporated specifically into DNA, so tritiated (radioactive) thymidine labels newly made strands. By following which daughter chromosomes carried the label through successive divisions, Taylor showed that the chromosomes of a higher plant also distribute their DNA semi-conservatively. The principle therefore holds from bacteria to plants to human cells — exactly as NCERT states.
| Feature | Meselson & Stahl (1958) | Taylor et al. (1958) |
|---|---|---|
| Organism | E. coli (prokaryote) | Vicia faba — faba bean (eukaryote) |
| Label used | 15N — heavy, non-radioactive isotope | Radioactive (tritiated) thymidine |
| Detection method | CsCl density-gradient centrifugation | Autoradiography of chromosomes |
| What it tracked | Density of whole DNA molecules | Distribution of new DNA among chromosomes |
| Conclusion | Replication is semi-conservative in bacteria | Chromosomes of higher organisms also replicate semi-conservatively |
Taken together, these experiments converted a theoretical prediction into established fact. The semi-conservative model is also the reason replication can be so accurate: each new strand is checked against an intact, error-free parental template, and the chemistry of complementary base pairing — A with T, G with C — guides synthesis at every step. The mechanism predicted by the structure was confirmed by the structure's own logic.
Worked examples
A culture of E. coli with fully 15N-labelled DNA is shifted to 14N medium and allowed to grow for 80 minutes. What proportion of the DNA molecules will be of hybrid density, and what proportion fully light?
E. coli divides every 20 minutes, so 80 minutes is four generations. The DNA count doubles each round: 1 → 2 → 4 → 8 → 16 duplexes. Only the two original parental strands carry 15N, and a semi-conservative strand is never destroyed — so exactly 2 of the 16 duplexes are hybrid and the remaining 14 are fully light. Hybrid = 2/16 = 1/8; light = 14/16 = 7/8. The hybrid fraction halves each generation but never reaches zero.
After one generation in 14N medium, a Meselson–Stahl-style experiment shows a single band of intermediate density. A classmate claims this alone proves semi-conservative replication. Is the claim complete?
No. A single hybrid band after one generation is consistent with two models — semi-conservative and dispersive — because both place equal old and new DNA in every duplex. It rules out only the conservative model, which would have given two separate bands. To exclude dispersive replication you must examine the second generation: semi-conservative predicts two distinct bands (hybrid + light), while dispersive predicts a single band of reduced density. The first-generation band is necessary but not sufficient.
Which property of the DNA double helix led Watson and Crick to hypothesise semi-conservative replication, and how does it work?
The decisive property is complementary base pairing between the two antiparallel strands — A always pairs with T, G always with C. Because the base sequence of one strand uniquely specifies the sequence of the other, each strand carries enough information to direct the synthesis of its missing partner. Watson and Crick therefore proposed that the strands separate and each acts as a template, producing two duplexes that each retain one parental and one new strand. This is NCERT Exercise Q.5 of Chapter 5, almost verbatim.
Common confusion & NEET traps
This subtopic is a reliable source of one-mark traps. The errors cluster around three points: confusing the label, confusing the organisms, and over-claiming what the first-generation band proves.
First proof was in a bacterium — not a plant or fungus
The experimental proof for semi-conservative replication was first shown in E. coli by Meselson and Stahl. Taylor's Vicia faba work confirmed it for chromosomes of higher organisms, but it came after, not first. NEET 2018 asked precisely this and the answer was "Bacterium."
Rule: First proof → E. coli (bacterium). Extension to eukaryotic chromosomes → Vicia faba via Taylor. Never Pisum sativum — that is Mendel's pea, a distractor seen in NEET 2018.
Meselson & Stahl
¹⁵N
heavy, non-radioactive isotope
- Proves replication is semi-conservative.
- Organism: E. coli.
- Separation by CsCl density gradient.
- Read-out: density bands.
Hershey & Chase
³²P / ³⁵S
radioactive isotopes
- Proves DNA is the genetic material.
- Organism: bacteriophage infecting E. coli.
- Separation by blender + centrifuge.
- Read-out: radioactivity inside cells.
"Semi-conservative" means one strand kept, not the whole molecule
A common slip is to read "semi-conservative" as "the parental DNA molecule is conserved." It is not — the parental duplex is broken up. What is conserved is one strand per daughter. Conservative replication is the model where the whole parental duplex stays intact, and that model was disproved.
Rule: Semi-conservative → each daughter duplex = 1 old strand + 1 new strand. The original two-stranded molecule is never kept whole.