Identical twins, also called monozygotic or MZ twins, are less common than fraternal twins, also called dizygotic or DZ twins. A widely used rule of thumb is that about one-third of twins are identical and about two-thirds are fraternal, but those proportions are not fixed worldwide. The underlying rate of identical twinning is relatively stable, while fraternal twinning varies much more by population, maternal age, family history and fertility treatment.
This page separates the percentage of twins that are identical or fraternal from the population rate of twinning. Those are different measurements, and confusing them is one of the main reasons twin statistics from different sources can appear to disagree.
Quick Twin Statistics
- Identical twins: broadly around one-third of twin pairs in commonly cited estimates.
- Fraternal twins: broadly around two-thirds of twin pairs, although the proportion varies substantially between populations.
- Identical twin frequency: approximately 3 to 4 per 1,000 births worldwide.
- Fraternal twin frequency: more variable than the identical-twin rate and strongly influenced by population and reproductive factors.
- Latest U.S. overall twin birth rate: 30.1 twin births per 1,000 total births in 2024.
- U.S. infants born in twin deliveries in 2024: 109,195.
The U.S. figures above describe all twin births and do not separately identify monozygotic and dizygotic twins. Source: CDC/NCHS, final 2024 birth data.
Data reviewed: September 1, 2026. The latest final U.S. birth data used on this page are for 2024.
Identical vs Fraternal Twin Statistics at a Glance
The most useful way to compare identical and fraternal twins is to separate their biological classification from their statistical frequency. Monozygotic twinning is comparatively stable across populations, while much of the variation in overall twin rates comes from changes in dizygotic twinning.
| Measure | Identical Twins | Fraternal Twins |
|---|---|---|
| Scientific term | Monozygotic (MZ) | Dizygotic (DZ) |
| How they form | One fertilized egg develops and splits into two embryos | Two eggs are fertilized separately during the same cycle |
| Broad share of twin pairs | About one-third as a rule of thumb | About two-thirds as a rule of thumb |
| Population frequency | Approximately 3 to 4 per 1,000 births worldwide | No single universal rate; varies widely between populations |
| Geographic variation | Relatively limited | Substantial |
| Maternal-age association | Much less pronounced | Established association |
| Family-history association | Most cases are not clearly inherited | Can run in families through a tendency toward multiple ovulation |
| Fertility-treatment influence | ART can also increase MZ twinning in some settings | Ovulation stimulation and transfer of multiple embryos can substantially increase multiple gestation |
| Can a pair be opposite sex? | Typically no | Yes |
The one-third versus two-thirds split is a broad population-level rule rather than a biological constant. Because the DZ rate changes much more across populations, the proportion of all twins that are MZ also changes.
What Percentage of Twins Are Identical vs Fraternal?
The commonly quoted answer is that roughly one-third of twins are identical and two-thirds are fraternal. This is the simplified proportion used by sources such as the UK National Health Service.
It is useful for a quick answer, but it should not be interpreted as an exact worldwide percentage. The proportion depends on the relative number of monozygotic and dizygotic twin pairs in the population being studied.
The One-Third vs Two-Thirds Rule
The rule works reasonably well as a general summary because MZ twinning occurs at a fairly stable background rate, while DZ twins are usually more common overall. MedlinePlus Genetics describes DZ twins as about twice as common as MZ twins in broad population terms.
That does not mean every country or every dataset will show exactly 33.3% identical and 66.7% fraternal twins. A population with a low DZ twinning rate can have a much larger share of identical twins even if its underlying MZ rate is similar to that of a population where fraternal twinning is common.
Why the Percentage Is Not the Same Everywhere
Consider two hypothetical populations with the same MZ rate of 4 twin pairs per 1,000 deliveries.
- If Population A also has 4 DZ pairs per 1,000 deliveries, identical twins would represent about half of all twin pairs.
- If Population B has 12 DZ pairs per 1,000 deliveries, identical twins would represent only about one-quarter of all twin pairs.
The identical-twin rate did not change in that example. What changed was the number of fraternal twins. This is why a statement such as “one-third of twins are identical” describes a proportion among twins, not the underlying population frequency of monozygotic twinning.
How Common Are Identical Twins?
Monozygotic twins occur when a single egg is fertilized and the resulting zygote divides early in development, producing two embryos. MedlinePlus Genetics reports that MZ twins occur at approximately 3 to 4 per 1,000 births worldwide.
Unlike fraternal twinning, the MZ rate is relatively similar across many populations and is much less strongly associated with maternal age, parity or ancestry. Most cases of identical twinning are not known to be caused by inherited genetic factors.
Identical Twin Rate per 1,000 Births
| Measure | Approximate Figure | Interpretation |
|---|---|---|
| MZ twin frequency | 3 to 4 per 1,000 births | Broad worldwide estimate reported by MedlinePlus Genetics |
| Common pregnancy-based description | About 1 in 250 pregnancies | Approximately 4 per 1,000 pregnancies; the denominator differs from a birth-based rate |
The figures are numerically similar, but they should not be treated as technically identical measurements. One is expressed relative to births, while the other is framed relative to pregnancies.
Why Identical Twin Rates Are Relatively Stable
The biological event behind MZ twinning is the splitting of one early embryo. The reason this occurs remains incompletely understood. Unlike dizygotic twinning, it does not depend on a person releasing two eggs during the same cycle.
Research reviews therefore tend to find a much more stable background MZ frequency across populations and over time. This relative stability is useful when interpreting overall twin statistics: large geographic or historical changes in twinning are usually driven mainly by changes in dizygotic twinning.
How Common Are Fraternal Twins?
Fraternal twins form when two different eggs are fertilized by two different sperm cells during the same menstrual cycle. Each embryo begins as a separate zygote, which is why DZ twins are genetically comparable to other siblings rather than genetically near-identical in the way MZ twins are.
There is no equally useful single worldwide DZ rate. MedlinePlus describes fraternal twins as about twice as common as identical twins in broad terms, but spontaneous DZ rates vary dramatically among populations.
That variability means a global statement such as “fraternal twins occur X times per 1,000 births” can be misleading unless the geography, period, conception setting and denominator are specified.
Why Fraternal Twin Rates Vary More
Dizygotic twinning requires more than one egg to be available for fertilization. The release of multiple eggs in one cycle is often described as hyperovulation or multiple ovulation.
DZ twinning is associated with factors including maternal age, family tendency, parity, population background and fertility treatment. Because the prevalence and influence of these factors differ between populations and have changed over time, DZ twinning accounts for much of the observed variation in total twin rates.
Identical vs Fraternal Twin Rates Around the World
Worldwide data show large differences in overall twinning rates. A major analysis published in Human Reproduction assembled data for 165 countries covering more than 99% of the world’s population. It estimated that the global twinning rate increased from 9.1 twin deliveries per 1,000 deliveries in 1980-1985 to 12.0 in 2010-2015.
The study also concluded that most geographic and historical variation reflects DZ twinning because MZ twinning remains around 4 per 1,000 deliveries across much of the world.
| Region | 1980-1985 | 2010-2015 | Change in Pattern |
|---|---|---|---|
| Africa | 16.5 per 1,000 deliveries | 17.1 per 1,000 deliveries | High rate, relatively little change |
| Asia | 7.0 | 9.2 | Substantial increase |
| Europe | 9.1 | 14.4 | Large increase |
| North America | 9.9 | 16.9 | Large increase |
| Oceania | 10.1 | 14.8 | Large increase |
| South America | 8.7 | 9.3 | Relatively small change |
| World | 9.1 | 12.0 | Global rate increased by roughly one-third |
These figures are overall twin deliveries per 1,000 deliveries, not MZ-specific or DZ-specific birth rates. The 2010-2015 period is the latest period covered by this particular worldwide harmonized study, so the table should not be interpreted as a 2026 global snapshot. Source: Monden, Pison and Smits, Twin Peaks: more twinning in humans than ever before, Human Reproduction, 2021.
More type-specific data demonstrate why DZ statistics need geographic context. The American Society for Reproductive Medicine cites spontaneous dizygotic twinning estimates of about 1.3 per 1,000 in Japan, 8 per 1,000 in the United States and Europe, and 50 per 1,000 in Nigeria, while citing an MZ rate near 4 per 1,000 live births.
Those DZ figures are useful for illustrating the scale of population variation, but they should not be treated as a synchronized set of current national rates. They are literature-based estimates summarized in professional guidance and may represent different underlying periods and populations.
U.S. Twin Birth Statistics
For the United States, the National Center for Health Statistics provides much more recent and standardized data. Final 2024 data show that 109,195 infants were born in twin deliveries and that the twin birth rate was 30.1 per 1,000 total births.
This statistic covers all twins. U.S. birth-certificate statistics do not provide a simple nationwide MZ-versus-DZ breakdown comparable to the overall twin-birth series, so it would be inappropriate to divide the 30.1 rate mechanically into one-third and two-thirds and label the results official MZ and DZ rates.
Latest U.S. Twin Birth Rate
| U.S. Statistic | 2024 Figure |
|---|---|
| Infants born in twin deliveries | 109,195 |
| Twin birth rate | 30.1 per 1,000 total births |
| Change from 2023 twin birth rate | Down from 30.7 per 1,000 |
| Change from 2014 | 11% decline in the twin birth rate |
Source: CDC/NCHS, Births: Final Data for 2024 and A Decade of Decline in Twin Childbearing in the United States, 2014-2024.
How the U.S. Twin Birth Rate Has Changed
The modern U.S. pattern has two distinct phases. The rate rose sharply over several decades, from 18.9 per 1,000 births in 1980 to 33.9 in 2014. It then declined to 30.1 in 2024.
| Year | Twin Birth Rate per 1,000 Total Births | Number of Twin Births |
|---|---|---|
| 1980 | 18.9 | Not shown in this comparison series |
| 2014 | 33.9 | 135,336 |
| 2015 | 33.5 | 133,155 |
| 2016 | 33.4 | 131,723 |
| 2017 | 33.3 | 128,310 |
| 2018 | 32.6 | 123,536 |
| 2019 | 32.1 | 120,291 |
| 2020 | 31.1 | 112,437 |
| 2021 | 31.2 | 114,161 |
| 2022 | 31.2 | 114,483 |
| 2023 | 30.7 | 110,393 |
| 2024 | 30.1 | 109,195 |
The 2024 rate was the lowest U.S. twin birth rate reported in more than 20 years. The decline does not mean MZ twinning itself fell by the same percentage. This national series combines all twin births, and changes in fertility practices, maternal-age patterns and other factors can affect DZ and overall twinning differently.
Twin Birth Rates by Maternal Age
U.S. data also show a strong relationship between maternal age and the overall twin birth rate. In 2024, the rate generally increased across age groups before leveling off among the oldest groups.
| Maternal Age | 2024 Twin Birth Rate per 1,000 Births |
|---|---|
| Younger than 20 | 16.1 |
| 20-24 | 22.7 |
| 25-29 | 28.9 |
| 30-34 | 33.2 |
| 35-39 | 35.4 |
| 40 and older | 35.6 |
These are overall twin birth rates, not DZ-only rates. Source: CDC/NCHS, 2024 natality data.
The age pattern is consistent with evidence that maternal age influences dizygotic twinning. It should not be interpreted as proof that every age-related difference in the national twin rate is caused by spontaneous DZ twinning, because fertility treatment and other population-level factors can also differ by age.
What Changes Fraternal Twin Rates More Than Identical Twin Rates?
Many familiar “twin factors” primarily affect the probability of producing two separately fertilized eggs and therefore have a stronger relationship with fraternal twinning than with spontaneous identical twinning.
| Factor | More Relevant To | Why It Matters Statistically |
|---|---|---|
| Maternal age | Primarily DZ | DZ twinning is associated with maternal age and multiple ovulation |
| Family history | Primarily DZ | A familial tendency toward hyperovulation can increase the likelihood of DZ twinning |
| Fertility treatment | Overall multiples and especially DZ in many treatment settings | Ovarian stimulation can produce multiple follicles, while transferring more than one embryo can produce multiple implantation |
| Population and geography | Primarily DZ | Spontaneous DZ rates vary markedly between populations |
Maternal Age
Maternal age is a well-established correlate of dizygotic twinning. The relationship is associated with reproductive biology, including the probability of multiple ovulation. This contributes to differences in twin rates between age groups, although observed national rates may also reflect fertility treatment and other demographic factors.
Family History and Hyperovulation
Fraternal twinning is more likely to run in families than identical twinning. MedlinePlus reports that having DZ twins is about twice as likely when a close relative, such as a sister, has had DZ twins compared with the general population.
The relevant mechanism is hyperovulation, the release of more than one egg during a menstrual cycle. Genetic influences on this tendency have been studied, but the genetics are complex and should not be reduced to a single “twin gene.”
Fertility Treatment
Fertility treatment can change twin statistics through several mechanisms. Ovulation induction and ovarian stimulation can lead to development of multiple follicles, increasing the possibility of dizygotic multiple gestation. In assisted reproductive technology, transferring more than one embryo also increases the risk of multiple pregnancy.
The relationship is not exclusively about fraternal twins. The American Society for Reproductive Medicine notes that MZ twinning may also be more frequent in ART cycles than in unassisted conception. Modern single-embryo-transfer practices are designed in part to reduce treatment-associated multiple gestation.
Population and Geography
Population differences are particularly important when interpreting DZ statistics. Research and professional guidance consistently report substantial geographic variation in spontaneous fraternal twinning, while the MZ rate remains comparatively stable.
This is why total twin rates cannot be interpreted as though every population has the same MZ-to-DZ ratio. Two countries can have similar MZ rates but very different overall twin rates because their DZ frequencies differ.
Why Different Sources Report Different Twin Statistics
Two reputable sources can report different twin numbers without either source being wrong. The most common reason is that they are measuring different things.
Percentage vs Rate
A percentage of twins answers a question such as:
Of all twin pairs, what proportion are identical?
A population rate answers a different question:
How often does a particular type of twinning occur among all births, deliveries or pregnancies?
For example, “about one-third of twins are identical” is a proportional statement. “MZ twins occur 3 to 4 times per 1,000 births” is a frequency statement. They should not be treated as interchangeable.
Pregnancy vs Delivery vs Birth
The denominator matters. Twin statistics can be expressed relative to:
- pregnancies,
- deliveries,
- total births,
- live births, or
- twin pairs.
A claim of “1 in 250 pregnancies” is therefore not automatically identical to “4 twin births per 1,000 births,” even though both may look like 0.4% when converted mathematically. The populations being counted are different.
Twin Pair vs Individual Babies
This distinction is particularly important when comparing U.S. statistics with international demographic studies.
The CDC’s U.S. twin birth rate counts births in twin deliveries relative to total births. A global demographic study may instead report twin deliveries per 1,000 total deliveries. One twin delivery usually produces two infants, so these denominators and numerators cannot be compared directly as though they measure the same event.
This also explains why a “twinning rate” can look much lower than a “twin birth rate” even when the figures describe broadly similar populations.
Geography and Data Year
Twin rates vary by both place and time. A figure from Japan, Nigeria, Europe or the United States may describe a very different underlying DZ frequency. Likewise, a U.S. figure from 1980 should not be used as though it represents 2024.
Every major twin statistic is therefore best interpreted with at least four labels:
- what is being counted,
- the denominator,
- the population or geography, and
- the data period.
Zygosity Classification
Another limitation is how twins are classified as monozygotic or dizygotic. Zygosity is not always genetically confirmed in large population datasets.
Opposite-sex twins can be classified as dizygotic under ordinary circumstances, but same-sex twins can be either MZ or DZ. Placental findings can provide information about twin type, yet separate placentas do not automatically prove that twins are fraternal. DNA testing after birth provides the most direct way to confirm whether a same-sex pair is genetically identical.
Studies that estimate zygosity rather than confirming it genetically should therefore be interpreted with that methodological limitation in mind.
Data Sources, Methodology and Limitations
This page prioritizes official statistics, authoritative medical and genetics resources, professional guidance and peer-reviewed population research. Statistics are not combined unless their units and denominators are sufficiently compatible.
How the Data Are Interpreted
- MZ frequency: worldwide MZ estimates are based primarily on authoritative genetics and reproductive-medicine sources reporting a relatively stable rate of approximately 3 to 4 per 1,000.
- MZ vs DZ proportions: the one-third versus two-thirds split is treated as a broad rule of thumb, not a fixed global constant.
- U.S. data: current overall twin-birth statistics use CDC/NCHS final birth data for 2024.
- Global trends: regional figures use a harmonized international study covering 1980-1985 and 2010-2015 and are labeled with those periods rather than described as current 2026 rates.
- DZ geographic examples: spontaneous DZ estimates cited by professional guidance are used to illustrate variability, not as a synchronized ranking of current national rates.
Primary Evidence Used
- CDC/National Center for Health Statistics, Births: Final Data for 2024.
- Martin JA, Osterman MJK. A Decade of Decline in Twin Childbearing in the United States, 2014-2024. NCHS Health E-Stat 117, 2026.
- MedlinePlus Genetics, Is the probability of having twins determined by genetics?
- Monden C, Pison G, Smits J. Twin Peaks: more twinning in humans than ever before. Human Reproduction, 2021.
- American Society for Reproductive Medicine. Multiple gestation associated with infertility therapy: a committee opinion.
- National Health Service, Pregnant with twins.
Important Limitations
There is no single contemporary worldwide database that reports genetically confirmed MZ and DZ rates for every country using one standardized method. A global MZ-versus-DZ table with apparent precision would therefore risk combining incompatible datasets.
Some statistics refer to spontaneous conception, while others include assisted reproductive technology. Some use deliveries as the denominator and others use births or pregnancies. Historical and geographic comparisons should be interpreted only after checking those definitions.
Population averages also do not predict an individual’s personal probability of having twins. Factors such as age, reproductive history, family history and fertility treatment need to be considered separately when estimating individual odds.
Frequently Asked Questions
Does “1 in 250 identical twin pregnancies” mean one in every 250 babies is an identical twin?
No. The commonly cited 1-in-250 figure is framed as a chance per pregnancy. A statistic per pregnancy should not be interpreted as though its denominator were individual babies or total live births.
Can the percentage of identical twins change even if the identical twin rate stays stable?
Yes. If the MZ rate remains close to its usual background frequency but the DZ rate rises or falls, the percentage of all twin pairs that are identical will change. This is one reason the one-third figure is not universal.
Can same-sex twins be fraternal?
Yes. Fraternal twins can be two boys, two girls, or a boy and a girl. Same-sex appearance alone does not determine zygosity.
Are twin rates from different countries directly comparable?
Only when the definitions are compatible. Check whether each source measures twin pregnancies, twin deliveries, individual twin births or twin pairs, and whether it includes fertility-treatment pregnancies. The data year also matters.
Does IVF only increase fraternal twins?
No. Fertility treatment is strongly associated with multiple gestation through mechanisms such as multiple follicle development and transfer of more than one embryo, but ART can also be associated with an increased rate of monozygotic twinning. The effect depends on the treatment and clinical practice.
Why is there no official current U.S. identical-versus-fraternal percentage on this page?
The CDC provides excellent nationwide statistics for total twin births, but that series does not provide a simple national MZ/DZ split with the same coverage and methodology. Applying the one-third versus two-thirds rule directly to the 2024 U.S. total would create an estimate, not an official CDC statistic, so this page does not present it as one.