The role of miscarriage and sororal birth order in male same-sex orientation: Theorical predictions and empirical data
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (Peer Community in Evolutionary Biology)
Abstract
This study explores the proximal and biological mechanisms underlying male same-sex orientation, with a focus on the Fraternal Birth Order Effect (FBOE), a robust phenomenon whereby same-sex oriented men tend to have more older brothers than the population average, and its relationship with the Sororal Birth Order Effect (SBOE), whereby older sisters also appear to influence sexual orientation, albeit less consistently. The Maternal Immune Hypothesis (MIH), which posits that maternal immune responses to male-specific antigens accumulate across successive male pregnancies, provides a compelling proximal explanation for the FBOE, but it fails to fully account for the SBOE and other birth order patterns, such as the elevated prevalence of same-sex orientation among only-children compared to firstborns in larger sibships. Through explicit modelling of the MIH, our simulations reveal that the correlation between the number of older brothers and sisters generates a spurious SBOE, which disappears when controlling for older brothers. However, this control becomes insufficient when miscarriages are included in the simulations. Additionally, the increased prevalence of same-sex orientation among only-children, relative to firstborns with siblings, only emerges when miscarriages are incorporated into the model. Empirical analyses across eight diverse populations (Indonesia, France, French Polynesia, Greece, Canada, Czech Republic, Samoa, Iran) confirm the presence of an overall significant FBOE and, critically, an overall significant SBOE even after controlling for the number of older brothers. The higher frequency of same-sex orientation men among only-children, compared to firstborns in larger sibships, supports a possible role of miscarriage for this SBOE. However, the estimated miscarriage rates (37% - 58%) explaining the observed SBOE slightly exceeds reported rates (30% - 45%) which conservatively include early pregnancy losses. This possibly suggests that additional mechanisms contribute to a spurious SBOE or that a genuine SBOE coexists with the FBOE. Limitations of this study are discussed, as well as whether the MIH framework can be extended to accommodate these findings, or if alternative explanations are needed to resolve these discrepancies.
Article activity feed
-
-
Dear authors,
I have now received four reviews of your ms. Based on their comments, as well as my own evaluation of the paper, I am happy to invite you to revise your manuscript according to the comments of the reviewers and my own. Some of the writing could be clarified, there are a few mistakes, and this should all be fairly easy to address.
There are two reviewer comments I would like to highlight. The first is the comment of reviewer 3, who points out immunological details which are relevant for the interpretation and discussion of your results, and I think you should take this into account. The second is a suggestion by reviewer 2, to also look at the effect of numbers of younger brothers and sisters. I think this is an interesting comment in the sense that we would not expect this to have an effect if immunology is the mechanism …
Dear authors,
I have now received four reviews of your ms. Based on their comments, as well as my own evaluation of the paper, I am happy to invite you to revise your manuscript according to the comments of the reviewers and my own. Some of the writing could be clarified, there are a few mistakes, and this should all be fairly easy to address.
There are two reviewer comments I would like to highlight. The first is the comment of reviewer 3, who points out immunological details which are relevant for the interpretation and discussion of your results, and I think you should take this into account. The second is a suggestion by reviewer 2, to also look at the effect of numbers of younger brothers and sisters. I think this is an interesting comment in the sense that we would not expect this to have an effect if immunology is the mechanism of interest here. But it could therefore be an valuable negative control to test for that effect - if no effects are found, it supports the immunology angle. But if effects of younger siblings are found, it must mean that something else is also in play. These effects could/should then be statistically controlled for when testing for effects of older siblings, as to get cleaner estimates, even if we don´t understand its mechanism. I invite you to consider these additional analyses, as it seems to increase the value and usefulness of your paper.
Finally, reviewer 4 has many useful comments on testing and interpretation, and I invite you to consider all of these with care.
You should have access to the four reviews throught the system to look at the comments in detail.
I also add a few comments of my own here:
L 185: why assume a linear effect
L 187: why assume a logistic effect?
L 225: presence of a spurious SBOE?
L 270/272: remove duplication
L 484: controlling for FBOE
I am looking forward to your resubmission.
-
Title and abstract - Yes, yes
Introduction - Yes, yes
Materials and methods - Yes, yes, yes
Results - N/A, yes
Discussion - Yes, yes
-
*Review of the manuscript: “Increased birth rank of homosexual males:
disentangling the older brother effect and sexual antagonism hypothesis”*This manuscript addresses an important and long-standing question in the
study of male sexual orientation: whether the increased birth rank observed
among homosexual males is best explained by a genuine fraternal birth order
effect, by increased female fecundity in maternal-line relatives, or by a
statistical overlap between these two processes. The topic is theoretically
relevant because it touches on the relationship between proximate
developmental mechanisms, such as prenatal maternal immune responses, and
ultimate evolutionary explanations, such as sexually antagonistic selection.Overall, the manuscript is valuable, ambitious, and methodologically
explicit. It makes a serious attempt to clarify a …*Review of the manuscript: “Increased birth rank of homosexual males:
disentangling the older brother effect and sexual antagonism hypothesis”*This manuscript addresses an important and long-standing question in the
study of male sexual orientation: whether the increased birth rank observed
among homosexual males is best explained by a genuine fraternal birth order
effect, by increased female fecundity in maternal-line relatives, or by a
statistical overlap between these two processes. The topic is theoretically
relevant because it touches on the relationship between proximate
developmental mechanisms, such as prenatal maternal immune responses, and
ultimate evolutionary explanations, such as sexually antagonistic selection.Overall, the manuscript is valuable, ambitious, and methodologically
explicit. It makes a serious attempt to clarify a literature in which birth
order, sibship size, number of older brothers, number of older sisters, and
maternal fecundity have often been difficult to disentangle. The authors
should be commended for addressing this statistical complexity directly,
for proposing formal models, and for testing alternative interpretations
across different datasets. The effort to distinguish the fraternal birth
order effect from an antagonistic fecundity effect is particularly useful,
and the paper represents a significant contribution to the ongoing debate.However, I believe that several aspects of the statistical and conceptual
framework require further clarification and, in some cases, a more cautious
interpretation. My main concern is not that the analyses are irrelevant or
poorly conceived. Rather, it is that the inferential structure of the paper
may be too strongly tied to a “separatist” model in which the fraternal
birth order effect and sexual antagonistic selection are treated as
mechanisms that can be cleanly partitioned. This may not exhaust the
biologically plausible interpretations of the relationship between these
phenomena.*General assessment*
The central strength of the manuscript is that it recognizes a genuine
statistical problem. Families with higher fertility naturally produce
children with higher average birth rank, and this can generate apparent
associations between sexual orientation and birth order even in the absence
of a specific older-brother mechanism. Conversely, a true older-brother
effect may generate patterns that resemble increased family fecundity if
not properly modeled. This problem is real, and the authors’ attempt to
formalize it is both timely and useful.The paper is also strong in its effort to separate older brothers from
older sisters and to consider the possible contribution of maternal-line
female fecundity. This is preferable to simpler analyses that use total
sibship size or birth order without distinguishing the sex and genealogical
position of relatives. The inclusion of individual-level datasets is
another important strength, as it allows the authors to move beyond purely
aggregate comparisons and to test more refined hypotheses.Nevertheless, the manuscript would benefit from a clearer distinction
between statistical separability and biological separability. The fact that
two predictors can be entered separately in a model does not necessarily
imply that they correspond to independent biological mechanisms. This issue
is especially important for the interpretation of sexually antagonistic
selection and the fraternal birth order effect.*Major comments*
*1. The operationalization of the antagonistic effect appears narrower than
the broader sexually antagonistic selection model*The manuscript uses the antagonistic effect largely in terms of increased
fecundity among mothers and, in the extended maternal family, through the
number of cousins produced by maternal aunts. This is a reasonable and
useful operationalization, but it is not equivalent to the full range of
possible sexually antagonistic selection models.A sexually antagonistic selection model does not necessarily predict only
that mothers of homosexual males should have more children, nor only that
maternal aunts should produce more offspring. It may predict a broader
pattern of increased female reproductive advantage distributed across the
maternal line, potentially involving mothers, maternal aunts, maternal
grandmothers, sisters, and other female maternal-line relatives, depending
on age, parity, reproductive completion, cultural conditions, and
genealogical structure.Thus, the current tests appear to evaluate a restricted version of the
antagonistic effect: direct maternal fecundity and a proxy for
maternal-aunt fecundity. This is informative, but it should not be
presented as a definitive test of the entire sexually antagonistic
selection framework. A more cautious formulation would be that the study
finds little or no evidence for an antagonistic fecundity component as
operationalized in the available datasets, particularly after controlling
for fraternal birth order.*2. The proxy for extended maternal-line fecundity is useful but limited*
The use of the number of cousins from maternal aunts is an interesting
attempt to extend the analysis beyond the mother. However, this variable is
a relatively coarse proxy for female-line fecundity. It depends not only on
the fecundity of maternal aunts, but also on the number of maternal aunts,
their age, reproductive completion, marital or partnership status, cultural
fertility norms, mortality, contraceptive practices, and possibly the
fertility of their male partners.The authors control for the number of maternal aunts, which is appropriate,
but this does not fully solve the problem. For example, the same number of
cousins may have different meanings depending on whether the aunts are
young or old, whether they have completed reproduction, whether the proband
belongs to an older or younger cohort, and whether fertility is culturally
constrained.It would therefore be helpful for the authors to describe more explicitly
what this variable can and cannot test. It is not a complete measure of
extended maternal-line fecundity; it is a partial measure of offspring
produced by maternal aunts. This distinction matters because the
manuscript’s conclusions regarding the antagonistic effect may otherwise
appear broader than the data allow.*3. Possible overcontrol: FBOE may not be merely a confounder under an
integrated SAS framework*The most important conceptual issue concerns the treatment of the fraternal
birth order effect as something to be controlled for when testing the
antagonistic effect. This is appropriate under one model: namely, a model
in which FBOE and sexually antagonistic selection are separate and
potentially competing mechanisms. In that framework, controlling for male
birth rank or older brothers estimates whether any female-fecundity effect
remains after removing the older-brother-related component.However, this is not the only plausible biological model. Under a broader
sexually antagonistic selection framework, the FBOE may represent one
developmental pathway through which a maternally mediated or familial
genetic liability is expressed, rather than an external confound to be
removed. In such an integrated model, sexually antagonistic selection would
operate at the genetic or evolutionary level, while the fraternal birth
order effect could be one ontogenetic, prenatal, or maternal-fetal
expression of the same broader system.If this interpretation is plausible, then controlling for FBOE does not
test the full SAS model. It tests only the residual female-fecundity effect
that is independent of older-brother-related developmental processes. This
is an important and valid question, but it is narrower than the question of
whether sexually antagonistic selection contributes to male homosexuality.I suggest that the authors explicitly distinguish between two causal
frameworks:1. A separatist framework, in which FBOE and the antagonistic effect are
independent or competing explanations. Under this framework, the authors’
statistical approach is appropriate.
2. An integrated multilevel framework, in which SAS represents an ultimate
genetic or evolutionary process, while FBOE represents one possible
proximate developmental pathway through which that broader process is
expressed. Under this framework, controlling for FBOE may remove part of
the phenomenon that the SAS model seeks to explain.This distinction would make the interpretation of the results more precise
and would prevent an overly strong conclusion against sexually antagonistic
selection.A possible formulation for the Discussion could be:
“Although our analyses test whether a female-fecundity effect remains after
controlling for fraternal birth order, this approach assumes that FBOE and
the antagonistic effect are at least partly separable causal processes. An
alternative interpretation is that FBOE may represent one developmental
expression of a broader maternally mediated or sexually antagonistic
system. Under this interpretation, our models estimate the residual
fecundity effect independent of older brother related mechanisms, rather
than excluding all possible forms of sexually antagonistic selection.”This addition would not weaken the paper. On the contrary, it would make
its theoretical contribution more rigorous.*4. Statistical disentangling does not necessarily imply causal
disentangling*The manuscript’s statistical strategy is designed to separate predictors
that are naturally correlated: sibship size, birth rank, older brothers,
older sisters, and family fecundity. This is necessary, but the
interpretation requires caution. These variables are not merely independent
measurements; they are structurally linked by the reproductive history of
the family.In particular, maternal fecundity, total sibship size, and male birth rank
are mathematically and biologically intertwined. A woman with more children
creates higher birth ranks, more older siblings, and a greater chance that
at least one child will be born after one or more older brothers. This
means that statistical adjustment can be informative, but it can also
obscure causal pathways if the adjusted variable lies within the causal
chain.The manuscript should therefore be careful in moving from “the antagonistic
effect is not significant after controlling for FBOE” to “the antagonistic
effect is not supported.” A more cautious interpretation would be that
there is no clear evidence for an antagonistic effect independent of the
older-brother-related pathway in the available datasets.*5. More emphasis should be placed on effect sizes, uncertainty, and model
robustness*The manuscript appears to rely substantially on significance testing to
evaluate whether support exists for different hypotheses. However, in this
literature, effect sizes and confidence intervals are especially important.
Several expected effects are likely to be small, and the genealogical
variables are noisy. A non-significant result should not automatically be
interpreted as strong evidence of absence.I recommend that the authors emphasize effect sizes and uncertainty
intervals for the antagonistic-effect tests, particularly those involving
maternal aunts and cousins. Where possible, they should also present
robustness checks using alternative count models, such as Poisson,
quasi-Poisson, negative binomial, or zero-inflated models if appropriate.
The number of cousins is a count variable that may be overdispersed, and
model choice could affect inference.It would also be useful to report whether results change when using cousins
per maternal aunt, total cousins with number of aunts as a covariate, or
models with an offset for the number of maternal aunts. These alternatives
test subtly different questions and could clarify how robust the
conclusions are.*6. The “extended maternal family” terminology should be used more
cautiously*The phrase “extended maternal family” may overstate the genealogical
coverage of the available data. If the analysis includes mothers and
maternal aunts, but not maternal grandmothers, sisters, adult female
cousins, or a broader maternal-line pedigree, then the expression should be
qualified.I would suggest using a more precise formulation, such as “maternal-aunt
fecundity” or “offspring of maternal aunts” when referring to the cousin
analysis. The broader term “extended maternal-line fecundity” should be
reserved for analyses that include a more complete reconstruction of female
reproductive success across multiple maternal-line relatives.*7. The relationship between older sisters, older brothers, and maternal
fecundity deserves further clarification*The manuscript’s treatment of older sisters is useful because apparent
sororal birth-order effects may arise as artifacts of sibship size or
maternal fecundity. However, the interpretation of older sisters should be
carefully separated from the interpretation of female-line fecundity. Older
sisters are siblings of the proband and therefore part of the mother’s
reproductive output; they are not equivalent to the reproductive success of
maternal-line female relatives. This distinction should be made explicit.A sister’s existence contributes to sibship composition and birth order,
but her adult fecundity would be a different variable relevant to sexually
antagonistic selection. If the datasets do not include the reproductive
output of sisters, then the manuscript should avoid implying that
female-line reproductive advantage has been broadly tested.*Minor comments*
1. The manuscript would benefit from a clearer glossary distinguishing
FBOE, birth rank, male birth rank, older-brother number, older-sister
number, maternal fecundity, maternal-aunt fecundity, and the broader
sexually antagonistic selection hypothesis.
2. The authors should clarify whether “cousins from maternal aunts” refersto all cousins or only male cousins. If all cousins are included, this
should be explicitly stated.
3. It would be useful to specify whether maternal aunts’ reproductive
output is complete or potentially censored by age.
4. The discussion should avoid wording that may be read as a broad
rejection of sexually antagonistic selection if the available data only
test a restricted operationalization of the antagonistic effect.
5. The statistical supplement should clearly indicate how overdispersion
was assessed for count variables.
6. The authors should consider adding a directed acyclic graph or conceptual diagram showing alternative causal interpretations: FBOE as independent confounder versus FBOE as mediator or developmental expression of a broader SAS process.
7. The manuscript should distinguish more explicitly between ultimate and
proximate explanations. SAS is an evolutionary-genetic explanation; FBOE is
a developmental or prenatal mechanism. These levels are not necessarily
mutually exclusive.*Suggested reframing of the main conclusion*
The manuscript’s conclusion would be stronger if it were framed more
narrowly and precisely. Instead of concluding that the analyses do not
support the antagonistic effect in general, I suggest a formulation closer
to the following:“Our results provide support for a fraternal birth order effect and do not
provide clear evidence for an additional female-fecundity effect
independent of FBOE in the available datasets. However, these analyses test
a specific operationalization of the antagonistic effect, mainly involving
maternal and maternal-aunt fecundity, and do not exclude broader integrated
models in which FBOE may be one developmental pathway associated with
sexually antagonistic selection.”This wording would preserve the authors’ empirical conclusion while
avoiding an overextension of the statistical result.*Final evaluation*
This is a valuable and thoughtful manuscript that addresses an important
problem in the literature on male homosexuality, birth order, and familial
fecundity. Its major contribution is to show that some apparent
associations between sexual orientation, birth rank, and family
reproductive structure can be statistically confounded, and that FBOE
deserves careful treatment as a specific effect rather than as a generic
birth-order pattern.At the same time, the manuscript should be revised to clarify the scope of
its conclusions regarding the antagonistic effect and sexually antagonistic
selection. The analyses appear well suited to testing whether a
female-fecundity effect remains after controlling for fraternal birth
order. They are less suited to excluding an integrated multilevel model in
which sexually antagonistic selection operates at the genetic/evolutionary
level and FBOE represents one developmental expression of the same broader
phenomenon.For this reason, I would recommend revision rather than rejection. The
paper is important and should be published, but its theoretical claims
should be made more cautious. In particular, the authors should distinguish
between rejecting a residual antagonistic effect independent of FBOE and
rejecting the broader possibility that SAS and FBOE are linked components
of a multilevel biological system -
Review text not available.
-
Review text not available.
-
Dear authors,
Your resubmission has now been evaluated by the same three reviewers as before and myself, and all are very pleased with the changes. Two reviewers are fully satisfied with the new version, whereas one reviewer still has some minor comments. I invite you to address these carefully before the final decision is made.
Best wishes,
Pim.
-
Review text not available.
-
As mentioned in my original review, this is an interesting and worthwhile study. The authors have addressed the issues I raised in my first review. For example, the authors have now included the role of younger siblings in their analyses.
-
The authors have incorporated all my major and minor suggestions. I especially appreaciate the elaborate discussion of rates of early pregnancy loss and of the timing when the immunization is able to occur. I am now satisfied with the manuscript.
-
Data Editor's report
Date: 23/10/2026
1. Data and metadata must be archived and adhere to FAIR guiding principles
[x] Data are in a public repository
[x] Data repository has a persistent identifier (e.g., a DOI)
[x] Data are cited in the manuscript (in data availability statement or similar, as well as in the Literature Cited)
[x] Data repository has a license
[x] All necessary data files are present in the repository
[] All raw data present
If not:
[] Information on where raw data are located
[x] Information about how to access the raw data
[] Sufficient information about database version/filtering/processing/exclusion criteria
[x] All processed data present
[x] Data are contained in an interoperable format
[x] Metadata present (including README file)
[x] Metadata adequate (including …
Data Editor's report
Date: 23/10/2026
1. Data and metadata must be archived and adhere to FAIR guiding principles
[x] Data are in a public repository
[x] Data repository has a persistent identifier (e.g., a DOI)
[x] Data are cited in the manuscript (in data availability statement or similar, as well as in the Literature Cited)
[x] Data repository has a license
[x] All necessary data files are present in the repository
[] All raw data present
If not:
[] Information on where raw data are located
[x] Information about how to access the raw data
[] Sufficient information about database version/filtering/processing/exclusion criteria
[x] All processed data present
[x] Data are contained in an interoperable format
[x] Metadata present (including README file)
[x] Metadata adequate (including README file)
2. Archived data corresponds with the data reported in the manuscript
[x] Variables used in analysis present in the data
[x] The structure of the data presented matches the manuscript (e.g., it is the right size)
3. Code and metadata must be archived and adhere to FAIR guiding principles
[x] Code has a repository
[x] Code repository has a DOI
[x] Code is cited in manuscript
[x] Code repository has a license
[x] Code files are present in the repository
[x] Code is contained in an interoperable format
[x] Metadata present (README file and annotations in code)
[x] Metadata adequate
4. Archived code matches the manuscript
[x] Code is present for all analyses in the manuscript, along with code used to produce figures/tables where appropriate
5. Archived code runs with the archived data-NOT ASSESSED
[] Runs without errors
6. Results can be computationally reproduced by running the archived code -NOT ASSESSED
[] Numeric results (in table or text)
[] Figures
—--------
Comments from the data editor: The authors implemented all the requested changes (mainly adding readme files, and some missing scripts) and it is now clear where how each result presented in the manuscript and its supplement were produced. Note that I did not assess points 5 and 6 due to limited time.
-
Same-sex orientation has been studied a lot, including in the context of evolutionary biology, because of the potential consequences for reduction in reproductive success and therefore for natural selection against same-sex orientation. Unknown to me before accepting the role of recommender for this article, there is also really interesting work on the causes of same-sex orientation. One robust pattern is that the number of older brothers a man has is correlated with his probability to develop same-sex orientation. What might be the mechanism(s) behind this pattern?
One hypothesis is the maternal immune hypothesis. Under this hypothesis, mothers would develop antibodies against proteins involved in male brain development such as Y-linked proteins, with increasingly stronger effects with each male pregnancy, affecting the brain …
Same-sex orientation has been studied a lot, including in the context of evolutionary biology, because of the potential consequences for reduction in reproductive success and therefore for natural selection against same-sex orientation. Unknown to me before accepting the role of recommender for this article, there is also really interesting work on the causes of same-sex orientation. One robust pattern is that the number of older brothers a man has is correlated with his probability to develop same-sex orientation. What might be the mechanism(s) behind this pattern?
One hypothesis is the maternal immune hypothesis. Under this hypothesis, mothers would develop antibodies against proteins involved in male brain development such as Y-linked proteins, with increasingly stronger effects with each male pregnancy, affecting the brain structures underlying sexual partner preference in later-born sons (Blanchard et al., 2001; Bogaert & Skorska, 2011). This hypothesis has received quite some support. However, it has also been challenged. First, there also seems to be an effect of the number of older sisters. This seems to be at least in part a spurious effect of the correlation with having more older brother too, however the effect remains even when statistically controlling for this. Second, first-born males may still develop same-sex orientation, suggesting other factors are also at play. Third, male same-sex orientation is also higher in only-child men (i.e., without brothers or sisters).
A potential explanation for these challenges of the maternal immune hypothesis is the effect of miscarriages. Male foetuses that are not carried full term might still cause a response in the mother’s immune system, and thereby affect later born males. Yet miscarriages are typically not included in the available datasets to test for the effect of the number of older brothers and sisters on male same-sex orientation, so this possibility is hard to test.
In this paper, Raymond et al. (2026) tackle this problem by first simulating datasets under different scenarios, and then analysing these with the traditional statistical models. As commented by all reviewers, this is a powerful approach, because since it is exactly known which processes have generated the simulated data, one can exclude many confounding factors. Most importantly, they manipulated the correlation between the number of older brothers and older sisters, and assessed the potential for miscarriage to explain the effect of the number of older sisters on same-sex orientation and to explain the prevalence of same-sex orientation in firstborn males. In response to reviewer comments, the effect of the number of younger brothers and sisters was also estimated, since these should not contribute to the development of same-sex orientation according to the hypothesis (and this was indeed confirmed).
In addition, the authors analysed 8 datasets from different countries, and estimated the effect of the number of older brothers and the effect of the number of older sisters on male same-sex orientation. If an effect of the number of older sisters was found, they then estimated the frequency of miscarriage that could explain this effect.
What did they find? First, for the simulated data, indeed a spurious effect of number of older sisters was found when there was a correlation between the number of older brothers and older sisters. Next, the presence of miscarriage indeed resulted in an effect of the number of older sisters (since these are correlated). Miscarriage also contributed to the greater prevalence of male same-sex orientation in single male children. For the empirical datasets, the effect of number of older sisters could be explained by miscarriage, but the required miscarriage rates were rather high compared to rates reported in the literature.
What can we conclude from this study? Indeed, miscarriage is a key piece of information when we want to test the maternal immune hypothesis. It can explain why single male children have a higher prevalence of same-sex orientation, and it can explain why the number of older sisters seems to have an effect on this too. However, it seems that rather high rates of miscarriage have to be assumed to understand these effects (although one reviewer pointed to literature that suggests that perhaps very young foetuses could already trigger an immune response, and miscarriage rates go up if younger foetuses are included). This suggests that these rates are underreported, or that an as-of-yet unknown mechanism is behind a real effect of the number of older sisters on male same-sex orientation, presenting a critical avenue for further research.
References
Blanchard, R., & Ellis, L. (2001). Birth weight, sexual orientation and the sex of the preceding siblings. Journal of Biosocial Science, 33(3), 451-467. https://doi.org/10.1017/S0021932001004515
Bogaert, A. F., & Skorska, M. (2011). Sexual orientation, fraternal birth order, and the maternal immune hypothesis : A review. Frontiers in Neuroendocrinology, 32, 247-254. https://doi.org/10.1016/j.yfrne.2011.02.004
Michel Raymond, Ana Aguerre, Valerie Durand, Menelaos Apostolou, Julien Barthes, Sarah Nila, Bambang Suryobroto, Mostafa Sadr-Bazzaz, Paul L. Vasey, Daniel Turek, Pierre-André Crochet (2026) The role of miscarriage and sororal birth order in male same-sex orientation: Theoretical predictions and empirical data . BioRxiv, ver.3 peer-reviewed and recommended by PCI Evolutionary Biology https://doi.org/10.64898/2026.03.09.710348
-
-
-