Mendel's Genetics, 1865 – The Pea Experiments
In this article
In 1865, Gregor Mendel gave two lectures to a provincial natural history society, describing his pea experiments. Nobody in the room grasped the scale of the question. Thirty-five years later, those lectures would be regarded as the beginning of a science that would reshape all of biology. This article explains, with sources, what Mendel did and did not do, and the six most frequently repeated misconceptions about him, before tracing the path from peas to the cancer clinic.
What Mendel did was strikingly unfamiliar to the biology of his time: count living things. 19th-century botany was descriptive: plants were drawn, classified and compared. Mendel instead counted the offspring of hybrid plants by the thousands and examined ratios. This seemingly ordinary approach showed that inheritance was not continuous blending but rather the redistribution of discrete units . Everyone before him had likened inheritance to mixing paint; Mendel showed that it resembled reshuffling cards.
This distinction is not abstract in oncology. A mother's precisely fifty percent chance of passing a BRCA1 mutation to her child directly follows the segregation rule Mendel observed in peas. Counseling for hereditary cancer syndromes, pedigree construction and risk calculation all operate within a framework established in 1865.
Six Corrections First
Some statements about Mendel circulating in Turkish sources either misdate events or confuse different facts. One, the fourth item, is more than a wrong date: it is an entirely false account of the history of science.
| Circulating statement | The correct account |
|---|---|
| "Mendel's work was published in 1865" | 1865, the year of the lectures : February 8 and March 8, 1865, at the Natural History Society in Brünn (Brno). The paper was published in 1866 : Versuche über Pflanzen-Hybriden, Verhandlungen des naturforschenden Vereines in Brünn, vol. IV, pp. 3–47. |
| "More than 20,000 pea plants" | The total commonly given is approximately 28,000 plants. Figures in Mendel's own text, however, measure different things: More than 10,000 plants were carefully examined, 287 artificial fertilizations on 70 plants were performed, and in the seed-shape experiment, 7,324 seeds from 253 hybrid plants were counted. |
| "Mendel established three laws" | Mendel did not formulate numbered laws. The "first law/second law" framework was constructed by 20th-century textbooks; formalization largely came through William Bateson and Thomas Hunt Morgan (1916). Some textbooks count two, others three laws. |
| "Mendel's experiments disproved Lamarck's theory of acquired characteristics" | False. Mendel's paper never mentions Lamarck , nor Darwin. It cites hybridization researchers: Kölreuter, Gärtner, Herbert, Lecoq and Wichura. The person who experimentally challenged inheritance of acquired characteristics was August Weismann (germ-plasm theory, 1892; mouse-tail experiment, 1889). |
| "DNA's structure was discovered in the 1940s" | Two separate events have been confused. 1944: Avery, MacLeod and McCarty showed that the hereditary material was DNA , establishing its identity. 1953: Watson and Crick, using Franklin and Wilkins's X-ray data, worked out DNA's double-helical structure (Nature 171:737–738, April 25, 1953). |
| "Sutton and Boveri proved that hereditary factors are carried on chromosomes" | They proposed it; they did not prove it. Sutton in 1902–1903, Boveri in 1904. The theory faced resistance for more than a decade. Decisive experimental evidence came from the Morgan group's fruit fly studies ; to identify a single study, it was Calvin Bridges's 1916 paper "Non-disjunction as Proof of the Chromosome Theory of Heredity." |
The fourth row deserves attention, because this error is not harmless. "Mendel disproved Lamarck" compresses two separate debates in 19th-century biology into a single heroic narrative. Mendel's experimental design contained no variable called acquired characteristics: he studied fixed, preexisting varietal differences. An experiment cannot disprove something it does not test.
Weismann's work was not a clean refutation either. He cut the tails off 68 white mice and followed five generations; among 901 offspring, none had tail abnormalities. He reported the result in 1889 . But Lamarck's proposed mechanism was use and disuse; cutting off mice's tails is not "disuse." The correct statement is: Weismann's germ-plasm theory, rather than Mendel's peas, fundamentally challenged inheritance of acquired characteristics, and even that experiment remains debatable. Reducing scientific history to heroic narratives is the easiest way to teach it incorrectly.
Who Was Mendel?
| Birth | July 20, 1822, Heinzendorf bei Odrau, Austrian Silesia (now Hynčice, Czechia). Baptismal name: Johann Mendel. |
|---|---|
| Monastery | In Brünn (Brno), St. Thomas's Augustinian Abbey, 1843. He took the name "Gregor" here. |
| Priesthood | December 25, 1846 |
| Teaching examination | Failed more than once : documented attempts were in 1850 and 1856. He never obtained certification and worked as a substitute teacher at Brünn Modern School. |
| Experiments | 1856–1863 |
| Lectures | February 8 and March 8, 1865 |
| Publication | 1866 |
| Abbot | Elected in 1867, took office in 1868. Afterward, his scientific work largely ended because of administrative duties and a prolonged dispute with the government over taxation of monasteries. |
| Death | January 6, 1884, Brünn, aged 61, from chronic nephritis. |
| Recognition during his lifetime | No. His work's significance was understood sixteen years after his death. |
The man who failed his teaching examination wrote biology's foundational text. Beyond a pleasing irony, this tells us something: Mendel's work did not match the skills measured by biology education at the time. The examination demanded identification, classification and memorization; Mendel introduced statistics. A similar tension exists today: evaluating an oncology study requires being able to read the numbers as much as knowing the disease.
The Experiment: Seven Traits, Thousands of Seeds
Mendel did not choose peas at random. They self-pollinate, naturally producing pure lines; their flowers suit controlled manual crosses; they reproduce quickly and yield many seeds. Most importantly, their traits show clearly defined opposite forms: a seed is either round or wrinkled, with nothing in between. Had he studied a continuously varying trait such as human height, he would never have seen the ratios.
He obtained 34 pea varieties and selected 22 for experiments, focusing on seven pairs of characteristics.
| # | Characteristic | Dominant form | Recessive form |
|---|---|---|---|
| 1 | Seed shape | Round | Wrinkled, angular |
| 2 | Seed (cotyledon) color | Yellow | Green |
| 3 | Seed coat color | Gray-brown (with purple-red flowers) | White (with white flowers) |
| 4 | Mature pod shape | Inflated, smooth | Constricted, deeply grooved |
| 5 | Unripe pod color | Green | Yellow |
| 6 | Flower position | Along the stem (axial) | Clustered at the top (terminal) |
| 7 | Stem length | Tall (180–210 cm) | Short, dwarf (23–45 cm) |
Based on descriptions in Mendel's 1866 text. Note that flower color in the third row is not a separate characteristic but is observed alongside seed coat color, although Turkish sources often present it as an eighth trait.
The figures from two experiments show the scale of his method:
The Rules: Two or Three?
Generalizations drawn from Mendel's findings are formulated as follows today. But first, this numbering was not Mendel's; it was organized by later generations.
Segregation
An individual carries two alleles for each trait. As reproductive cells form, these alleles separate , and each gamete receives only one. We now explain this through homologous chromosomes separating during meiosis, but Mendel did not know chromosomes existed. He inferred the rule solely from ratios.
Where Does the 3:1 Ratio Come From? Crossing a pure purple-flowered plant (RR) with a pure white-flowered one (rr) produces an entirely Rr first generation (F1), all purple. When F1 plants self-pollinate, either parent can contribute R or r, producing four equally likely combinations: RR, Rr, Rr, rr. The first three appear purple because R is dominant; the fourth is white. The genotype ratio is 1:2:1, and the phenotype ratio is 3:1. A Punnett square displays these four possibilities visually. Named after Reginald Punnett, it was not a tool Mendel himself used.
Independent Assortment
Allele pairs for different traits assort independently during gamete formation. Following two traits together produces an F2 9:3:3:1 ratio.
Dominance
In a heterozygote (Rr), the dominant allele's phenotype appears and the recessive allele's is hidden.
The third rule is the weakest of the three , so many textbooks do not count it as a "law." Exceptions abound: incomplete dominance (crossing red and white four-o'clock flowers produces pink, neither parental color), codominance (both alleles are expressed in blood group AB), epistasis (one gene masks another gene's expression). In oncology, the most important exception to this "law" is penetrance , discussed separately below.
The Major Exception: Linkage
Independent assortment applies only to genes on different chromosomes or sufficiently far apart on the same chromosome. Nearby genes on one chromosome tend to be inherited together, a phenomenon called linkage .
The first to demonstrate linkage were William Bateson, Edith Rebecca Saunders and Reginald Punnett in 1905, in sweet peas (Lathyrus odoratus). When flower color and pollen shape were followed together, parental combinations appeared much more often than expected. The modern interpretation, that crossover frequency measures physical distance along a chromosome, was provided by Thomas Hunt Morgan ; Sturtevant drew the first genetic map in 1913.
"Mendel was lucky: his seven genes were on seven separate chromosomes" is another myth. This appealing explanation, because peas have seven chromosome pairs, was refuted in Stig Blixt's 1975 Nature article, "Why didn't Gregor Mendel find linkage?"
The documented distribution is seven loci across four chromosomes . LE (stem length) and V (pod shape) are on the same chromosome and are closely linked; A, I and R are together on chromosome 7; R (seed shape) and GP (pod color) are on the same chromosome but are weakly linked.
Why did Mendel not observe linkage? Three reasons combine: (1) He did not test all twenty-one possible pairwise combinations, reporting only a few two- and three-trait crosses. (2) In weakly linked pairs, deviations from 9:3:3:1 were probably too small to detect with his sample sizes. (3) The actual pod-shape locus he observed is uncertain. Instead of V , he may have measured the LE -unlinked P locus. This was neither pure luck nor a concealed flaw, but a combination of selected crosses, sample size and locus uncertainty.
Thirty-Five Years of Silence and the Myth of Being "Completely Forgotten"
Mendel's paper appeared in 1866 and had no scientific impact until 1900. Yet "nobody saw or read it" is also untrue. It appeared in Focke's Die Pflanzen-Mischlinge (1881) compilation, the 1881 edition of Encyclopaedia Britannica and the Royal Society's 1879 catalogue of scientific papers . Mendel sent approximately 40 offprints , and the journal itself reached more than a hundred institutions. Some sources report at least 14 citations before 1900. Since the origin of that figure is disputed, we present it here as a reported statement, not an established claim.
The reason for neglect was not access but the framework. Biology then viewed inheritance as blending, and Darwin's evolutionary debate dominated. Mendel's statistical approach was foreign to the language botanists knew. A correct answer remains incomprehensible in a world not asking the question.
1900: Three People, at the Same Time Three botanists independently reached Mendel's conclusions and rediscovered the paper: Hugo de Vries (Dutch, the first to publish), Carl Correns (German, explicitly acknowledging Mendel's priority), and Erich von Tschermak (Austrian; later historians debated how fully he understood the principles).
The field then quickly developed its own language: William Bateson proposed "genetics" in a letter in 1905 and used it publicly at the international plant hybridization conference in London in 1906, describing the field as "still nameless." Wilhelm Johannsen in 1909, in his book Elemente der exakten Erblichkeitslehre , coined "gene" . The same book also introduced genotype and phenotype . Johannsen deliberately chose a word "completely free of any hypothesis."
The Fisher Controversy: Were the Data Too Good?
In 1936, a founder of statistics, Ronald A. Fisher, published in Annals of Science a paper titled "Has Mendel's Work Been Rediscovered?" His claim was startling: Mendel's data fitted expectations too well.
Pooling Mendel's experiments, Fisher calculated a total chi-square statistic of 41.6056 with 84 degrees of freedom. He reported the probability of such a small deviation arising by chance as approximately one in 3,000 . Real experimental data fluctuate around expectations; Mendel's were abnormally close.
But Fisher did not accuse Mendel of fraud, a distinction repeatedly overlooked. Fisher's proposed explanation was: "Mendel had been deceived by an assistant who knew too well what was expected." Fisher accepted Mendel's theory as entirely valid; his objection concerned the numbers, not the conclusions.
What Is the Current Academic Position? Mendel has largely been exonerated, but the statistical peculiarity is still debated.
| Question | Position |
|---|---|
| Were Mendel's conclusions correct? | Yes. They were never seriously questioned and have been independently confirmed repeatedly. |
| Did Mendel deliberately fabricate data? | No. Researchers such as Hartl and Fairbanks rejected Fisher's analysis and concluded there was "no reason to suggest that Mendel fabricated his results" . |
| Is the Statistical Anomaly Real? | Partly. In the 2008 volume Ending the Mendel–Fisher Controversy , statisticians, physicists and biologists rejected all Fisher's allegations except the "too-good fit" statistic. |
| How Is the Anomaly Explained? | Unconscious experimenter bias when classifying ambiguous seeds; continuing counts until the ratio "looked right" (a stopping rule); excluding doubtful plants; and Fisher's pooling method being unsuitable for the data structure. Pires and Branco's model in Statistical Science in 2010 showed that without deliberate falsification , unconscious selection bias could produce Fisher's result. |
Why Discuss This on an Oncology Website? Because it contains the essence of today's clinical research concerns: selective reporting, stopping rules, classification of ambiguous cases and adjusting after seeing results. What Mendel did in the 1860s, deciding where to classify an ambiguous seed, is the same problem a radiologist faces when deciding whether an ambiguous lesion counts as a "response." The lasting legacy is not a verdict on Mendel, but the habit of examining how data were produced.
From Peas to DNA: The Correct Sequence
| Year | Who | What | Status |
|---|---|---|---|
| 1866 | Mendel | Inheritance operates through discrete units and predictable ratios | The unit's identity is unknown |
| 1902–1904 | Sutton, Boveri | Hereditary factors may lie on chromosomes | Proposal |
| 1905 | Nettie Stevens | An additional chromosome determining sex in mealworms | Supporting evidence |
| 1910–1913 | Morgan, Sturtevant | Sex-linked inheritance in fruit flies; the first genetic map | Strong evidence |
| 1916 | Calvin Bridges | Nondisjunction cases demonstrate chromosome theory | Proof |
| 1944 | Avery, MacLeod, McCarty | The hereditary material is DNA , not protein | Identity |
| 1953 | Watson, Crick; Franklin, Gosling; Wilkins | DNA's double-helical structure | Structure |
The 1953 paper depended critically on X-ray diffraction data obtained by Rosalind Franklin and Raymond Gosling, "Photograph 51." Their supporting paper and Wilkins's appeared in the same Nature issue. The 1962 Nobel Prize in Physiology or Medicine went to Crick, Watson and Wilkins.
The identity of the "invisible unit" Mendel described in 1866 was answered eighty-seven years later as the geometry of a molecule. We discussed how it is copied and amplified in the laboratory in our article on PCR, and how it can be edited in our article on CRISPR's origins .
Mendel in the Cancer Clinic: From Pedigrees to Cells
So far, this has been scientific history. Now to the central question: What use are the rules observed in peas in 1865 in an oncology clinic today?
The answer has two layers. The first is direct: hereditary cancer syndromes follow Mendelian inheritance, and a carrier's probability of passing a mutation to a child, fifty percent , follows precisely from segregation. The second, more interesting point is: cancer best demonstrates where Mendel's rules end.
First, the scale: According to the National Cancer Institute, approximately 5–10% of all cancers result from harmful genetic changes inherited from a parent. Most cancers are therefore not hereditary: they arise from somatic mutations accumulating during life. This is frequently confused in patient communication: "my tumor has a mutation" and "I carry an inherited mutation" are different statements. As NCI itself says, tumor (somatic) testing does not replace hereditary risk testing.
Knudson's Two-Hit Hypothesis: Resolving the Paradox
Applying Mendel to cancer reveals a puzzle. Hereditary cancer syndromes appear autosomal dominant in pedigrees: one defective copy suffices, with affected people in each generation. At the cellular level, however, these genes behave recessively : one intact copy protects the cell. How can both be true?
The answer came from Alfred Knudson in 1971 , studying retinoblastoma, a retinal tumor in children. He noticed two statistical patterns: hereditary retinoblastoma appears earlier than sporadic disease and hereditary cases often involve both eyes.
The Two-Hit Model: For cancer to begin, both copies of the gene must lose function. In hereditary disease, the first hit is present from birth : every cell already contains a defective copy. Only one hit remains, and across trillions of cells this is almost inevitable, producing early, multifocal disease. In sporadic cases, two independent hits in the same cell are required, a much lower probability resulting in later onset and a single focus.
| Hereditary form | Nonhereditary form | |
|---|---|---|
| Proportion | Approximately one third | Approximately two thirds |
| Mutation location | Germline: in every body cell | Somatic: only in eye cells |
| Involvement | Usually both eyes | Usually one eye |
| Family history | May be present | Absent |
| Additional cancer risk | Present : pineoblastoma, osteosarcoma, soft-tissue sarcomas and melanoma | No increased risk |
Responsible gene RB1, on chromosome 13 (13q14.1–q14.2). The hereditary form is autosomal dominant. Disease usually appears before age five. For the association with bone sarcomas, see our osteosarcoma article .
This resolves the paradox: mutant alleles of tumor-suppressor genes are recessive at the cellular level(both copies must be disrupted), but appear dominant at the family level (inheriting one defective copy greatly increases cancer probability, producing a dominant pedigree pattern). The same gene behaves differently depending on scale. Oncogenes are the opposite: mutant alleles are typically dominant at the cellular level too. In oncogenes such as KRAS , this is why one mutation can leave the cell permanently "switched on."
Major Hereditary Cancer Syndromes
| Syndrome | Gene(s) | Inheritance | Main cancers | Frequency |
|---|---|---|---|---|
| Hereditary breast–ovarian cancer | BRCA1, BRCA2 | Autosomal dominant | Breast, ovarian, pancreatic, prostate, male breast | — |
| Lynch syndrome | MLH1, MSH2, MSH6, PMS2, EPCAM | Autosomal dominant | Colorectal, endometrial, ovarian, stomach, small bowel, urinary tract, brain and skin | 1 in 279 |
| Familial adenomatous polyposis | APC | Autosomal dominant | Hundreds to thousands of colon polyps; without treatment, at least 90% by age 50 develop colorectal cancer | 1 in 8,500 |
| Li-Fraumeni | TP53 | Autosomal dominant | Breast cancer, osteosarcoma, soft-tissue sarcoma, brain tumors, leukemia and adrenocortical carcinoma, particularly in children and young adults | 1 in 5,000–20,000 |
| Retinoblastoma | RB1 | Autosomal dominant | Retinoblastoma; secondary sarcoma and melanoma risk | — |
| MEN type 1 | MEN1 | Autosomal dominant | Parathyroid (most common), pituitary, pancreatic islet cell tumors | 1 in 30,000 |
| MEN type 2A | RET | Autosomal dominant | Medullary thyroid carcinoma, pheochromocytoma, hyperparathyroidism | 1 in 35,000 |
| Von Hippel-Lindau | VHL | Autosomal dominant | Hemangioblastoma, clear cell renal carcinoma, pancreatic neuroendocrine tumors, pheochromocytoma | 1 in 36,000 |
| Hereditary diffuse gastric cancer | CDH1 (E-cadherin) | Autosomal dominant | Stomach: lifetime risk of 56% in women and 70% in men; in women, lobular breast cancer 40–50% | Less than 1% of stomach cancers |
A "—" in the frequency cells means no verified prevalence figure was found for this article. CDH1 mutations are detected in 20–40% of hereditary diffuse gastric cancer families; 60–70% have no CDH1 mutation.
Penetrance: Why "Dominant" Does Not Mean "Certain"
Dominance was absolute in Mendel's peas: every Rr plant was purple, without exception. Humans, particularly in cancer, are different. This is called penetrance.
Penetrance: the proportion of people carrying a particular genetic variant who show the disease's signs and symptoms. If some carriers never develop the disease, this is called reduced (incomplete) penetrance . The separate concept of variable expressivity describes the range of symptoms among people with the same genetic condition.
BRCA is a textbook example. Consider NCI's figures and recognize that each is actually a penetrance value :
| Cancer | BRCA1 | BRCA2 | General population |
|---|---|---|---|
| Breast cancer (women) | More than 60% of carriers develop breast cancer during their lifetime | ~13% | |
| Ovarian cancer | 39–58% | 13–29% | ~1.1% |
| Second cancer in the opposite breast (over 20 years) | 30–40% | 25% | ~8% |
| Pancreatic cancer | Up to 5% | 5–10% | — |
| Prostate cancer (by age 80) | 7–26% | 19–61% | — |
| Male breast cancer (by age 70) | 0.2–1.2% | 1.8–7.1% | — |
The other side of "more than 60%" is that a substantial proportion of carriers do not develop breast cancer. NCI states this explicitly: "Some people who inherit a harmful change in BRCA1 or BRCA2 never develop cancer." You would never encounter this in Mendel's peas, and it is the most difficult part of genetic counseling. What a female carrier needs to hear is not "you will get cancer," but "your risk is several times that of the general population, and there are ways to manage it."We discussed who chooses risk-reducing surgery and why in our article on preventive surgery in BRCA carriers .
The Recessive Side: Syndromes Requiring Two Defective Copies
Hereditary cancer brings dominant syndromes to mind, but predispositions following Mendel's recessive rule also exist, challenging textbook Mendelism in another direction.
| Syndrome | Gene | Main cancers | Frequency |
|---|---|---|---|
| MUTYH-associated polyposis | MUTYH | Colorectal cancer: without surveillance, 80–90% lifetime risk; also duodenal, breast, ovarian, endometrial and bladder cancer | 1 in 20,000–60,000; ~1% of all colorectal cancers |
| Fanconi anemia | More than 20 genes; 80–90% of cases involve FANCA, FANCC or FANCG | Acute myeloid leukemia; liver, digestive and genital tract tumors; head and neck squamous cell carcinoma. Probability of developing cancer: 10–30% | 1 in 100,000–160,000 |
| Ataxia-telangiectasia | ATM | Leukemia and lymphoma | 1 in 40,000–100,000 |
The ataxia-telangiectasia row shows where Mendel's binary framework is insufficient. The disease is recessive: two defective ATM copies are required for symptoms. Yet approximately 1% of the US population carries one copy and female carriers have an increased breast cancer risk compared with noncarriers. A "recessive" gene is therefore not silent in one copy: it creates an intermediate cancer risk. Mendel's Rr pea is purple or not; an ATM heterozygote is neither affected nor entirely risk-free. Cancer genetics operates in this gray area.
Beyond Mendel: Polygenic Risk
Most cancer risk does not come from one gene. It reflects small contributions from hundreds or thousands of common variants, called polygenic risk , outside Mendel's framework. Polygenic risk scores (PRS) try to estimate risk by summing disease-associated changes in an individual.
But what must be said about polygenic risk scores today is: in the National Human Genome Research Institute's words, these scores are "not yet in routine clinical use" and "there are no guidelines for implementation," although private providers and direct-to-consumer companies have begun offering them. A second, less discussed problem is ancestry bias: as the same institute says, "most genomic studies to date have examined people of European ancestry," so the scores "may be valid and useful only for European-ancestry populations." PRS also provide relative risk, not absolute risk : probability, not certainty. No cancer-specific polygenic risk score in clinical use could be verified for this article.
Who Should Be Tested, and Why a "Test in a Box" Is Not Enough
According to NCI, clinical guidelines recommend offering hereditary (germline) genetic testing for triple-negative breast cancer, ovarian cancer, pancreatic cancer, colorectal cancer diagnosed before age 50, metastatic prostate cancer or male breast cancer; young age at diagnosis; multiple cancers in one person; several affected relatives; cancer in both paired organs; rare cancers; associated congenital anomalies; or membership of an ethnic group at increased hereditary cancer risk.
For direct-to-consumer BRCA tests, the key figure is this: NCI says these tests miss approximately 80% of cancer-causing BRCA variants.
The reason appears in the FDA's own decision document: the approved direct-to-consumer test examines only three variants : BRCA1 185delAG, BRCA1 5382insC and BRCA2 6174delT. In the document's words: "The three variants included in this report are most common in people of Ashkenazi Jewish ancestry and do not represent the majority of BRCA1/BRCA2 variants in the general population." The document also states that absence of a tested variant "does not rule out the presence of other variants that may be associated with cancer" and that results "should be confirmed in a clinical setting before any medical action is taken." It states this explicitly.
The practical implication is clear: a "negative" home BRCA test does not provide reassurance for someone with a family history of cancer. This is the article's most direct clinical warning.
A Note for Turkey
Genetic testing and counseling for hereditary cancer syndromes in Turkey are provided through medical genetics departments, university hospitals and training and research hospitals. However, this article makes no Turkey-specific numerical claims: test access, waiting times, which tests are performed under which conditions, and SGK reimbursement coverage could not be verified for this article. Question the source whenever you encounter definitive claims about these matters.
Patients should, in practice, tell their doctor about any family cancer history in detail : who, which cancer and at what age. Genetic counseling begins with a pedigree, not a blood test, and the framework for drawing it is still Mendel's. For an overview of Turkey's cancer burden, see Turkey's cancer statistics, and for progress in cancer science, see our article on the AACR Cancer Progress Report 2026 .
Timeline
DROZDOGAN Academy Commentary
Mendel is usually portrayed as an "ignored genius." Updating this article revealed something more interesting: Mendel was misunderstood, rather than ignored , and much of what is said about him today is still wrong.
⚡ Why It Matters
Mendel is used daily in cancer clinics. The framework for drawing family cancer histories, giving a fifty percent transmission probability and recommending sibling testing was established in 1865. More importantly, hereditary cancer best demonstrates where Mendel's rules are insufficient. Penetrance, intermediate risk in heterozygous carriers of a recessive gene and polygenic predisposition are invisible in peas.
✅ The Strength
Mendel's method introduced numbers into biology. Without knowing chromosomes, genes or DNA, he inferred an invisible unit solely from ratios. Eighty-seven years later, its geometry confirmed his prediction. Few inferences in scientific history have been so far-reaching and accurate. Knudson's work in 1971 was similar: before molecular biology could answer, he used age distribution and involvement of one or both eyes to develop the two-hit model.
⚠️ The Weakness
Mendelian thinking often paints an overly tidy human picture. An Rr pea is purple; a human BRCA1 carrier is neither certain to develop cancer nor certain not to. An ataxia-telangiectasia carrier is neither affected nor risk-free. The real language of cancer genetics is probability, not certainty , and textbook Mendelism does not speak it. The Fisher controversy also reminds us that even foundational scientific texts must face the question: how were the data produced ?
🩺 What to Tell Patients
(1) Only 5–10% of cancers are hereditary; a tumor mutation does not mean "it runs in the family." (2) Genetic counseling begins with a pedigree , not a blood test: who, which cancer, at what age. (3) "Dominant" means high probability, not certainty; being a carrier does not mean being ill. (4) A negative direct-to-consumer home BRCA test is not reassurance: it misses approximately 80% of cancer-causing BRCA variants. (5) A result affects siblings and children as well as the individual; the decision is therefore a family process.
❓ Open Questions
What determines penetrance: modifier genes, environment or chance? We cannot yet explain why one sister with the same BRCA1 mutation develops cancer and another does not. Can polygenic risk scores become valid in non-European populations, or will genomic data inequality impose a permanent ceiling? Third, and most clinically urgent, optimal timing of screening and risk-reducing interventions in hereditary cancer syndrome carriers still rests largely on consensus rather than randomized evidence.
Conclusion: Knowing Mendel's rules is the beginning of understanding cancer genetics, but only the beginning. Clinical work starts where those rules fail: variable penetrance, carrier status without certainty, and a recessive gene carrying risk even in one copy. Of the six corrections needed in updating this article, the Lamarck point mattered most, because making scientific history appear tidier conceals how untidily science actually progresses. Mendel did not end a debate; he started one nobody had yet opened, and nobody heard it for thirty-five years.
Frequently Asked Questions
- When Was Mendel's Work Published: 1865 or 1866?
- Both dates are correct for different events. The lectures were delivered on February 8 and March 8, 1865 at the Brünn Natural History Society; the paper appeared in 1866 in the society's journal. "Published in 1865" is common but technically incorrect.
- Did Mendel Really Disprove Lamarck?
- No. Mendel's paper never mentions Lamarck, and his experimental design does not test acquired characteristics: it examines fixed varietal differences. August Weismann experimentally challenged inheritance of acquired traits (mouse experiment, 1889; germ-plasm theory, 1892). Modern assessments suggest even that experiment did not fully test Lamarck's "use and disuse" mechanism.
- How Many Mendelian Laws Are There: Two or Three?
- It depends on the textbook because Mendel did not write numbered laws. Segregation and independent assortment are usually counted as "laws," while dominance is an exception-rich "principle." Formalization came through Bateson and Morgan in the early 20th century.
- Did Mendel Fabricate His Data?
- There is no evidence of this. In 1936, Fisher showed that the data fitted expectations too well, but pointed to "an assistant who knew too well what was expected," rather than Mendel. A comprehensive 2008 review rejected all allegations except the too-good-fit statistic. Current explanations focus on unconscious experimenter bias and the tendency to stop counting at a particular point.
- What Proportion of Cancers Are Hereditary?
- According to NCI, approximately 5–10% of all cancers result from inherited genetic changes. The remainder arise from somatic mutations accumulating during life.
- My Tumor Has a Mutation. Will My Children Inherit It?
- Generally not. Tumor tissue testing shows somatic mutations, present only in tumor cells and not inherited. A separate test is needed to identify a hereditary (germline) mutation. NCI explicitly states that tumor testing does not replace hereditary risk testing. Do not hesitate to ask your doctor which test was performed.
- What Does "Autosomal Dominant" Mean? Will I Definitely Become Ill?
- It means one defective copy is sufficient and there is a 50% chance of transmission to each child. However, it does not mean you will become ill. Penetrance explains this: only some variant carriers develop disease. For example, NCI reports that more than 60% of BRCA carriers develop breast cancer, meaning a substantial proportion do not.
- How Can a Tumor-Suppressor Gene Be Both Dominant and Recessive?
- It depends on the scale. It is recessive at the cellular level: one intact copy protects the cell, and both must be disrupted for cancer. It appears dominant in pedigrees: inheriting one defective copy raises the likelihood of a second hit across trillions of cells so much that affected people appear in every generation. Knudson's 1971 two-hit model explains precisely this.
- Is a Home BRCA Test Reliable?
- It is limited. According to NCI, the approved direct-to-consumer test misses approximately 80% of cancer-causing BRCA variants because it examines only three variants. The FDA document says these "do not represent the majority of BRCA1/BRCA2 variants in the general population." A negative result should not be reassuring for someone with a family history of cancer.
- There Is Cancer in My Family. What Should I Do?
- The first step is not testing but gathering information: which relative, which cancer and at what age. Tell your doctor. Guidelines summarized by NCI recommend considering genetic testing for young age at diagnosis, multiple cancers in one person, several affected relatives, cancer in both paired organs and specific cancers: triple-negative breast, ovarian, pancreatic, colorectal before 50, metastatic prostate or male breast cancer.
- How Is Testing Done in Turkey?
- Hereditary cancer testing is provided through medical genetics departments. However, access conditions, waiting times and SGK reimbursement coverage could not be verified for this article; no claims about them are made here. Confirm current arrangements with your treating doctor.
Glossary
- Allele
- Each alternative form of a gene. Diploid organisms have two alleles for each gene.
- Genotype
- An individual's genetic constitution. Johannsen defined the term in 1909.
- Phenotype
- Observable characteristics produced jointly by genotype and environment.
- Homozygous / Heterozygous
- Having identical (RR, rr) or different (Rr) alleles for a gene.
- Dominant
- The allele expressed in the phenotype of a heterozygote.
- Recessive
- An allele expressed phenotypically only in the homozygous state.
- Segregation
- Separation of a gene's two alleles during gamete formation.
- Independent Assortment
- Independent distribution of allele pairs from different genes into gametes. It applies only to unlinked genes.
- Linkage
- The tendency of nearby genes on the same chromosome to be inherited together. The main exception to independent assortment.
- Punnett Square
- A visual table arranging possible genotypes from a cross. Named after Reginald Punnett; Mendel did not use it.
- Penetrance
- The proportion of variant carriers showing disease symptoms. When some never become ill, penetrance is described as "reduced."
- Expressivity
- The range of symptom severity and variety among people with the same genetic condition.
- Germline (Hereditary) Mutation
- A mutation present in an egg or sperm cell, carried in all body cells and transmissible to children.
- Somatic Mutation
- A mutation arising in body cells during life, confined to that cell lineage and not inherited.
- Tumor-Suppressor Gene
- A gene that restrains cell division or repairs damaged DNA. Mutant alleles are generally recessive at the cellular level.
- Oncogene
- A gene that continually stimulates cell division when activated. Mutant alleles are typically dominant at the cellular level.
- Two-Hit Hypothesis
- Knudson's 1971 model stating that both copies of a tumor-suppressor gene must lose function.
- Polygenic Risk Score
- A score summing small contributions from many common variants to estimate disease risk. Not yet in routine clinical use.
Transparency Note.
Verified: Lecture dates (February 8 and March 8, 1865), publication year (1866), journal title, volume, pages and the paper's German title; figures in Mendel's text (22 varieties, 287 fertilizations / 70 plants, 7,324 and 8,023 seeds and their ratios); definitions of seven characteristic pairs; that Mendel did not number his laws and their formalization came through Bateson and Morgan; the absence of Lamarck and Darwin from his citations and the names he did cite; Weismann's mouse experiment figures (68 mice, 901 offspring, five generations, 1889) and the date of germ-plasm theory; the distinction between Avery–MacLeod–McCarty (1944) and Watson–Crick (1953), and the 1953 paper's citation; the dates and contributions of Sutton, Boveri, Stevens, Morgan, Sturtevant and Bridges; the three rediscoverers in 1900 and their nationalities; when "genetics" and "gene" were coined; Fisher's 1936 citation, chi-square statistic (41.6056 / 84), approximately one in 3,000 and his not accusing Mendel; the 2008 volume and Pires–Branco 2010 model; Blixt's 1975 Nature article and the distribution of seven loci across four chromosomes; Mendel's life dates, monastery, abbacy and lack of recognition during life; Knudson in 1971 and the retinoblastoma observations; RB1 chromosome location and hereditary/sporadic proportions; the syndromes' genes, inheritance, cancers and frequencies in the table; NCI's 5–10% estimate and BRCA risks; definitions of penetrance and expressivity; MUTYH, Fanconi anemia and ataxia-telangiectasia data; clinical status and ancestry bias of polygenic risk scores; the three variants and limitations in the FDA direct-to-consumer BRCA test decision; NCI's "misses 80%" statement and testing indications.
Not Verified: The total number of Mendel's teaching-examination failures (only attempts in 1850 and 1856 could be documented, so no total is given); an independent source for the widely repeated "at least 14" pre-1900 citations; molecular identities of Mendel's seven genes reportedly completed in 2025; R–GP and LE–V map distances in centimorgans; exact years of Weismann's tail-cutting experiments; the year RB1 was cloned (therefore omitted); Cowden/PTEN genes and risks; constitutional mismatch repair deficiency (CMMRD); lifetime risk percentages for Lynch and Li-Fraumeni; whether testing indications can be attributed to NCCN (attributed to NCI here); the FDA's first authorization date for a direct-to-consumer BRCA test; a cancer-specific polygenic risk score in clinical use; genetic testing access, waiting times and SGK reimbursement coverage in Turkey.
Source of the Criticisms: The six-item correction table results from comparing common Turkish statements with primary and secondary sources. The judgment that "Mendel disproved Lamarck" is false, and the comment that reducing scientific history to heroic narratives is the easiest way to teach it incorrectly, belong to this article. Blixt's linkage finding, the present state of the Fisher debate and Knudson's model, by contrast, are the cited studies' own conclusions.
Disclosure: We have no affiliation with any test or company mentioned in this article.
References
- Mendel G. "Versuche über Pflanzen-Hybriden." Verhandlungen des naturforschenden Vereines in Brünn 1866;IV:3–47. (English translation: MendelWeb.)
- Blixt S. "Why didn't Gregor Mendel find linkage?" Nature 1975;256:206.
- "Mendel: from genes to genome." Plant Physiology 2022;190(4):2103.
- Fisher RA. "Has Mendel's Work Been Rediscovered?" Annals of Science 1936;1(2):115–137.
- Franklin A, Edwards AWF, Fairbanks DJ, Hartl DL, Seidenfeld T. Ending the Mendel–Fisher Controversy. University of Pittsburgh Press, 2008.
- Pires AM, Branco JA. "A Statistical Model to Explain the Mendel–Fisher Controversy." Statistical Science 2010;25(4):545–565.
- Weismann A. Das Keimplasma: eine Theorie der Vererbung. 1892. (English: The Germ-Plasm: A Theory of Heredity, 1893.)
- Avery OT, MacLeod CM, McCarty M. Journal of Experimental Medicine, February 1944.
- Watson JD, Crick FHC. "Molecular Structure of Nucleic Acids: A Structure for Deoxyribose Nucleic Acid." Nature 1953;171:737–738.
- Bridges CB. "Non-disjunction as Proof of the Chromosome Theory of Heredity." Genetics 1916;1:1–52 and 107–163.
- Knudson AG. Two-hit hypothesis, 1971.
- National Cancer Institute. "BRCA Gene Changes: Cancer Risk and Genetic Testing." cancer.gov
- National Cancer Institute. "Genetic Testing for Inherited Cancer Susceptibility Syndromes." cancer.gov
- MedlinePlus Genetics. Pages on Lynch syndrome, familial adenomatous polyposis, Li-Fraumeni syndrome, retinoblastoma, multiple endocrine neoplasia, von Hippel-Lindau syndrome, hereditary diffuse gastric cancer, MUTYH-associated polyposis, Fanconi anemia, ataxia-telangiectasia, penetrance and expressivity. medlineplus.gov
- National Human Genome Research Institute. "Polygenic Risk Scores." genome.gov
- U.S. Food and Drug Administration. De Novo decision document DEN170046 — 23andMe PGS Genetic Health Risk Report for BRCA1/BRCA2 (Selected Variants). accessdata.fda.gov
Sources
Was this article helpful?
Your feedback helps us improve our content.
Reader comments
Share your comment
Your name and comment are published only after administrator approval. Your email address is never shown to visitors.
Be the first to comment.