Rapid Responses to:

PAPERS: Luc J M Smits, Rob A de Bie, Gerard G Essed, and Piet A van den Brandt Time to pregnancy and sex of offspring: cohort study BMJ 2005; 331: 1437-1438 [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Time to pregnancy and sex of offspring

Jørn Olsen, Weijin Zhou, Jin Liang Zhu   (21 December 2005)

Miscarriage terminology

David J R Hutchon   (22 December 2005)

Sex ratio and infertility related factors

Michael J Davies   (23 December 2005)

Curvilinear relationship ignored

Johannes C van der Wouden   (24 December 2005)

Are subfertile couples more likely to have boys? A comprehensive nationwide study

Rémy Slama, Jean Bouyer, Béatrice Ducot, Alfred Spira and Béatrice Blondel.   (6 January 2006)

Time to pregnancy and sex of offspring

William H James   (17 January 2006)

Entrenched misinformation about X- and Y-sperm

Valerie J. Grant   (16 March 2006)

Time to pregnancy and sex of offspring. Authors’ reply to J Olsen et al, DJR Hutchon et al, MJ Davies, JC van der Wouden, R Slama et al, WH James, and VJ Grant

Luc Smits, Rob de Bie, Gerard Essed, and Piet van den Brandt   (27 June 2006)

Time to pregnancy and sex of offspring 21 December 2005
Jørn Olsen, Professor Department of Epidemiology, School of Public Health, UCLA, Los Angeles, USA, Weijin Zhou, Jin Liang Zhu Send response to journal: Re: Time to pregnancy and sex of offspring	Dear Sir, It is an old idea that fecundity may correlate with the sex ratio and many of the theories related to frequency of males to females (sex ratio) would imply such an association. In Smits et al's paper,1 they put forward the idea that cervical mucous viscosity correlate with both the penetration of Y bearing spermatozoa and fecundity. They find a striking empirical support for the hypothesis using a Dutch data source with almost 5300 births. We expect that many have looked into this hypothesis but found no association and did not or could not publish their negative results. They refer to our previous negative study2 and suggest that our data were confounded by infertility treatment. We therefore report a replication of their study using the Danish National Birth Cohort that has data on infertility treatment and time to pregnancy. The cohort includes data from 100,000 pregnancies and has been described in detail elsewhere.3 In brief, the data on infertility treatment and time to pregnancy were reported early in pregnancy, and the data on time to pregnancy were used the following categories: 0, 1-2, 3-5, 6-12, and >12 months. Data on the sex of the newborns came from the National Birth Register. We identified all planned first pregnancies for each woman in the cohort. In total, 61249 singletons and 3156 twins were available for analysis. Our replication of Smits et al's calculations is shown in the table. Table Percentages of male infants and odds ratios of being male gender according to time to pregnancy and infertility treatment Time to pregnancy or infertility treatment No. of births Male infants Crude OR Adjusted OR (95% CI) n % Among singletons 0 months 13639 7024 51.5 1.00 reference 1-2 months 14400 7425 51.6 1.00 1.00 (0.96-1.05) 3-5 months 13096 6657 50.8 0.97 0.97 (0.93-1.02) 6-12 months 9762 4950 50.7 0.97 0.97 (0.92-1.02) >12 months 5764 2943 51.1 0.98 0.98 (0.92-1.04) With infertility treatment 4588 2362 51.5 1.00 0.99 (0.92-1.06) Among twins 0 months 415 220 53.0 1.00 reference 1-2 months 396 214 54.0 1.04 1.04 (0.78-1.37) 3-5 months 362 187 51.7 0.95 0.91 (0.68-1.21) 6-12 months 193 106 54.9 1.08 1.04 (0.74-1.48) >12 months 100 48 48.0 0.82 0.77 (0.49-1.20) With infertility treatment 1690 873 51.7 0.95 0.87 (0.69-1.11) Logistic regression; OR: odds ratio; CI: confidence interval; Adjusted for maternal age, parity, pre-pregnancy body mass index, smoking, alcohol use, month of conception, and regularity of the menstrual cycle. We found no association between time to pregnancy and sex ratio in singletons or in twins. Infertility treatment was also not associated with the sex ratio and is an unlikely explanation for our previous results. Jørn Olsen, MD, PhD, Professor The Danish Epidemiology Science Centre, University of Aarhus, Vennelyst Boulevard 6, DK 8000 Aarhus C, Denmark; and Department of Epidemiology, School of Public Health, UCLA, Los Angeles, USA. Email: jo@ucla.edu or jo@soci.au.dk. Weijin Zhou, MD, PhD Shanghai Institute of Planned Parenthood Research, 20032 Shanghai, China. Email: zw0822@yahoo.com. Jin Liang Zhu, MD, PhD The Danish Epidemiology Science Centre, University of Aarhus, Vennelyst Boulevard 6, DK 8000 Aarhus C, Denmark. Email: zjl@soci.au.dk. References 1. Smits LJ, de Bie RA, Essed GG, van den Brandt PA. Time to pregnancy and sex of offspring: cohort study. BMJ 2005;331:1437-1438. 2. Weijin Z, Olsen J. Offspring sex ratio as an indicator of reproductive hazards. Occup Environ Med 1996;53:503-504. 3. Olsen J, Melbye M, Olsen SF, Sorensen TI, Aaby P, Andersen AM, et al. The Danish National Birth Cohort - its background, structure and aim. Scand J Public Health 2001;29:300-307. Competing interests: None declared
Miscarriage terminology 22 December 2005
David J R Hutchon, Locum consultatnt obstetrician and gynaecologist Grey Base Hospital, Greymouth, New Zealand Send response to journal: Re: Miscarriage terminology	Sir, It is now accepted by many major english medical journals that the miscarriage terminology adopted by the Royal College of Obsetricians and Gynaecologists should be used. "Spontaneous abortion" should be "miscarriage". I appreciate that the Dutch researchers may not be aware of this but the BMJ editorial should be.(1) As the terminology is used more widely it will become second nature to us. Reference 1. David J R Hutchon and Sandra Cooper Terminology for early pregnancy loss must be changed BMJ, Oct 1998; 317: 1081 Competing interests: None declared
Sex ratio and infertility related factors 23 December 2005
Michael J Davies, Senior Research Fellow Rsearch Centre in Reproductive Health, University of Adelaide, Australia Send response to journal: Re: Sex ratio and infertility related factors	Editor – The reported association between time to pregnancy and sex ratio of the offspring was argued to be due to a maternal factor that correlates negatively with fecundability 1. Specifically, it was hypothesised that viscous cervical mucus differentially impairs fertilisization of X and Y bearing sperm. Without direct measurement of mucosal viscosity this relationship remains conjectural. It’s of note that that sex of the offspring of couples who had received medical help in getting pregnant did not show any relation with time to pregnancy, which does not support the hypothesis. Indeed, as time to pregnancy is used as a definition of infertility, time to pregnancy in this study must be a proxy measure of some other phenomenon. A further hypothesis may be offered in addition to that proposed. Time to pregnancy also reflects active contraception, and is therefore a proxy for planned fertility and child spacing. It has been reported that families in Denmark with boys are more likely to continue to have children, due presumably to some social desirability factor 2. It has also been reported that inter-birth interval is greater following a male baby due to factors such as longer breast feeding for male babies 3. Together these could create a higher sex ratio in families where there is greater than average number of boys, and longer spacing. This doesn’t negate the potential role of fertility related factors or treatments altering sex ratios. Indeed, using a national registry of all births from assisted reproduction in Australia it has been reported that the sex ratio in live births after assisted conception (n=4,253) was 105.0 male infants per 100 females in 2000 4 . Sex ratio at birth, however, varied widely by treatment modality and infertility aetiology. Intracellular sperm injection (ICSI) conception had the lowest sex ratio of 95.2 male infants per 100 females. Higher sex ratios were reported for gamete intrafallopian transfer (GIFT) (112.3) and for in-vitro fertilisation (IVF) conception (115.3). In conclusion, the proposed mechanism is best examined within a study population in which there is a capacity for better individual level observation and control for potential confounders. 1. Smits LJ, de Bie RA, Essed GG, van den Brandt PA. Time to pregnancy and sex of offspring: cohort study. Bmj 2005;331(7530):1437-8. 2. Jacobsen R, Moller H, Engholm G. Fertility rates in Denmark in relation to the sexes of preceding children in the family. Hum Reprod 1999;14(4):1127-30. 3. Koziel S, Ulijaszek SJ. Waiting for Trivers and Willard: do the rich really favor sons? Am J Phys Anthropol 2001;115(1):71-9. 4. Dean J, Sullivan E. Assisted conception Australia and New Zealand 2000 and 2001. Sydney: Australian Institute of Health and Welfare National Perinatal Statistics Unit, 2003. Competing interests: None declared
Curvilinear relationship ignored 24 December 2005
Johannes C van der Wouden, senior lecturer Erasmus MC, Dept of General Practice PO Box 1738 3000 DR Rotterdam, the Netherlands Send response to journal: Re: Curvilinear relationship ignored	It is surprising that the authors do not comment on the apparent curvilinear relationship that is shown in the graph. Around the logistic regression line winds an almost perfect sinusoid curve. The only outlier seems the datapoint before the last one, which can be explained by the small number of data contributing to this point. It would be interesting to ask whether any readers can think of an explanation for this curvilinearity. Competing interests: None declared
Are subfertile couples more likely to have boys? A comprehensive nationwide study 6 January 2006
Rémy Slama, Researcher Inserm, National Institute of Health and Medical Research, U569, F-94276 Le Kremlin-Bicêtre, France., Jean Bouyer, Béatrice Ducot, Alfred Spira and Béatrice Blondel. Send response to journal: Re: Are subfertile couples more likely to have boys? A comprehensive nationwide study	Are subfertile couples more likely to have boys? A comprehensive nationwide study Rémy Slama1, researcher, Jean Bouyer1, senior researcher, Béatrice Ducot1, researcher, Alfred Spira1, professor, Béatrice Blondel2, senior researcher. 1: Inserm, National Institute of Health and Medical Research, U569, "Epidemiology, Demography and Social Sciences", IFR 69, 82 rue du Général Leclerc, F-94276 Le Kremlin-Bicêtre Cedex, France; INED, National Institute for Demographic Studies; University Paris 11, Faculté de Médecine, Le Kremlin-Bicêtre, France. 2: Inserm, U149, Epidemiological Research Unit on Perinatal Health and Women’s Health, IFR 69, F-94807 Villejuif Cedex, France; Université Pierre et Marie Curie-Paris6, F- 75012 Paris. Correspondence to: Rémy Slama, slama@vjf.inserm.fr Sir, Smits et al.1 reported that couples with a long time to pregnancy (TTP) were more likely to have boys. Such an association would make sex-ratio an interesting marker of a couple's fecundity, but this finding conflicts with a previous report.2 We studied the association between TTP and sex ratio in the National Perinatal Survey, including all births in France during one week of October 2003.3 Midwives interviewed mothers before discharge from hospital. Of the 14,482 mothers, 94% answered questions on TTP. Time from discontinuation of contraception until pregnancy (TTP) was defined for 9,395 (74%) of the 12,673 singleton live births not conceived after sterility treatment. The cumulative probability of pregnancy within 12 months of discontinuation of contraception was 87.0%. The proportion of male births was 50.9% for TTPs of up to 12 months (8,247 births), and 51.1% for TTPs of 13 months or more (1,148 births, p=0.9). No association was found between sex and TTP modelled as a continuous variable in a log-binomial regression model4 (p=0.7). If TTP was treated as a categorical variable, the proportion of male births tended to increase with TTP until 14-19 months, decreasing thereafter (table). The relative annual increase in this proportion between TTPs of 1 and 18 months was 5.1% (95% confidence interval, -1 to 11%, log-binomial regression4), but this was mostly due to a low proportion of boys born to couples who conceived during the first month. This trend disappeared if we restricted the analysis to primiparous women (table). This restriction was designed to limit bias due to the over-representation of couples with high fertility rates in cross-sectional samples of pregnant women. In conclusion, in a large population representative of all births occurring in France, a TTP above one year was not associated with an increase in the proportion of male births. Table: Proportions and relative risks of birth of a male child according to time to pregnancy for 13,245 singleton live births. [if !supportMisalignedColumns] [endif] Crude analyses Adjusted analyses‡ % boys (boys/total) RR* 95% Cl† RR* 95% Cl† All women Time to pregnancy 1 month 48.9 (1,288 /2,633) 1 1 2-4 months 51.6 (1,536 /2,975) 1.06 1.00-1.11 1.05 0.99-1.11 5-7 months 52.0 (761 /1,465) 1.06 1.00-1.13 1.06 0.99-1.13 8-10 months 52.4 (279 /533) 1.07 0.98-1.17 1.09 0.99-1.19 11-13 months 51.9 (351 /676) 1.06 0.98-1.15 1.06 0.98-1.16 14-19 months 53.5 (220 /411) 1.09 0.99-1.21 1.08 0.97-1.19 20-24 months 49.0 (152 /310) 1.00 0.89-1.13 1.02 0.91-1.15 ≥ 25 months 50.0 (196 /392) 1.02 0.92-1.14 1.03 0.92-1.15 All births with defined TTP 50.9 (4,783 /9,395) No TTP defined§ 51.1 (1,676 /3,278) 1.05 0.99-1.10 1.05 0.99-1.10 Infertility treatment 50.4 (288 /572) 1.03 0.94-1.13 1.07 0.96-1.19 Primiparous women only Time to pregnancy 1 month 50.7 (569 /1,122) 1 1 2-4 months 50.4 (640 /1,270) 0.99 0.92-1.08 0.98 0.90-1.06 5-7 months 49.2 (322 /654) 0.97 0.88-1.07 0.96 0.87-1.06 8-10 months 55.6 (148 /266) 1.10 0.97-1.24 1.10 0.97-1.24 11-13 months 52.2 (165 /316) 1.03 0.91-1.16 1.02 0.90-1.15 14-19 months 51.4 (110 /214) 1.01 0.88-1.17 0.98 0.84-1.13 20-24 months 52.7 (77 /146) 1.04 0.88-1.23 1.06 0.90-1.25 ≥ 25 months 48.8 (83 /170) 0.96 0.82-1.13 0.99 0.84-1.17 All births with defined TTP 50.8 (2,114 /4,158) No TTP defined§ 51.5 (656 /1,273) 1.02 0.94-1.10 1.02 0.94-1.10 Infertility treatment 49.9 (171 /343) 0.98 0.87-1.11 0.99 0.86-1.14 * Relative risk of giving birth to a male child, as estimated by a binomial regression model with a logarithmic link. This corresponds to the ratio of the proportions of boys in the groups compared. † Confidence interval. ‡ Adjusted for maternal and paternal ages at the time of discontinuation of contraception (age and age squared), maternal body mass index (6 categories) and tobacco use before pregnancy (3 categories). § Couples using a contraceptive method at the time of conception, couples who had not used any contraceptive method since the previous pregnancy, and women who could not report a TTP. References: 1. Smits LJ, de Bie RA, Essed GG, van den Brandt PA. Time to pregnancy and sex of offspring: cohort study. Bmj 2005;331(7530):1437-8. 2. Weijin Z, Olsen J. Offspring sex ratio as an indicator of reproductive hazards. Occup Environ Med 1996;53(7):503-4. 3. Blondel B, Supernant K, du Mazaubrun C, Bréart G. Enquête Nationale Périnatale 2003 - Situation en 2003 et évolution depuis 1998. Ministère des Solidarités, de la Santé et de la Famille, 2005. http://www.sante.gouv.fr/htm/dossiers/perinat03/enquete.pdf 4. Greenland S. Model-based estimation of relative risks and other epidemiologic measures in studies of common outcomes and in case-control studies. Am J Epidemiol 2004;160(4):301-5. Funding: French Ministry of Health. Competing interests: None declared
Time to pregnancy and sex of offspring 17 January 2006
William H James, Honorary Research Fellow Galton Laboratory, University college, 4 Stephenson Way, London W1 2HE Send response to journal: Re: Time to pregnancy and sex of offspring	Smits et al(1)reported on the sexes and conception waits of 5283 singletons. There was a highly significant regression in their data such that the longer the wait to conception, the higher the probability that it was male. This finding is in marked contrast with previous work on this topic which has consistently suggested that sons are more frequent after short conception waits(2-4). Moreover a substantial body of direct evidence has been cited to substantiate the thesis that coital rate (which is inversely related to conception wait) is positively associated with sex ratio (proportion male) in a number of other mammalian species (horse, rabbit, rat, mouse, seal)(5). According to Smits et al1, their result is unrelated to assisted reproduction techniques, so the cause of the difference between their result and those of previous authors is unclear. This work is urgently in need of replication. References 1. Smits LJM, De Bie RA, Essed GG, Van den Brandt PA. Time to pregnancy and sex of offspring. Br Med J 2005;331:1437-8 2. Renkonen KO. Heterogeneity among first postnuptial deliveries. Ann Hum Genet 1970;33:319-321 3. Brewis AA. Sex ratios at birth in a Micronesian atoll population. Soc Biol 1993;40:207-14 4. Weijin Z, Olsen J. Offspring sex ratio as an indicator of reproductive hazards. Occup Environ Med 1996;53:503-4 5. James WH. Further evidence that mammalian sex ratios at birth are partially controlled by parental hormone levels around the time of conception. Hum Reprod 2004;19:1250-6 Competing interests: None declared
Entrenched misinformation about X- and Y-sperm 16 March 2006
Valerie J. Grant, senior lecturer University of Auckland, Private Bag 92019, Auckland, 1. New Zealand, Send response to journal: Re: Entrenched misinformation about X- and Y-sperm	Entrenched misinformation about X- and Y- sperm In a recent paper, Smits et al1 found that the longer it took a woman to become pregnant the more likely it was that she would have a male infant. While the methodology and data are impressive, the authors’ explanation for their findings is less so, resting as it does on the claim that Y-chromosome-bearing spermatozoa (Y-sperm) swim faster than X- chromosome-bearing spermatozoa (X-sperm). This widely-held idea appears to have originated with Dr Landrum B Shettles. In 1960, using his new phase-contrast microscope, Shettles claimed to have observed “two distinct populations”2 of spermatozoa. After attempting to count the chromosomes, he concluded that “the smaller heads contain the Y- and the larger the X-chromosome.” “There were no intermediate types.”2. The following year he reiterated these findings, adding that “… smaller headed spermatozoa can migrate more rapidly and fertilize the ovum more often in the distal part of the tube”.3 Reading Shettles’ reports in Nature and other peer-reviewed journals, many researchers thereafter believed that Y-sperm swim faster than X- sperm. The finding particularly influenced research on sperm separation. In an article also published in Nature, researchers noted that the basis of their procedure was “the superior swimming ability of Y sperm as a function of time”. 4 Although there were several attempts to correct this impression 5,6 it was not until the development of computer-assisted sperm analysis (CASA)7 that reliable observations could be made. So far, researchers have found no morphological differences between human X- and Y- sperm 8,. “Neither mature sperm nor their precursors possess significant morphological differences between X and Y genotypes”8 and “Y bull sperm do not swim faster than X sperm” 9. References 1. Smits LJM, de Bie RA, Essed GG, van den Brandt PA. Time to pregnancy and sex of offspring: cohort study. BMJ 2005; 331:1437-8. 2. Shettles LB. Nuclear morphology of human spermatozoa. Nature 1960;186:648-9. 3. Shettles LB. After office hours: Conception and birth sex ratios: A review. Obstet Gynec 1961;18: 122-130. 4. Ericsson R J, Langevin C N, Nishino M. Isolation of fractions rich in human Y sperm. Nature 1973; 246: 421-4. 5. Rothschild. X and Y spermatozoa. Nature 1960; 187:253-4. 6. Gledhill BL. Selection and separation of X- and Y-chromosome- bearing mammalian sperm. Gamete Res 1988; 20:377-95. 7. Moruzzi JF, Wyrobek AJ, Mayall BH, Gledhill BL. Quantification and classification of human sperm morphology by computer-assisted image analysis. Fertil Steril 1988; 50:142-52. 8. Hossain AM, Barik S, Kulkarni PM. Lack of significant morphological differences between human X and Y spermatozoa and their precursor cells (spermatids) exposed to different prehybridization treatments. J Androl 2001; 22:119-23. 9. Penfold LM, Holt C, Holt WV, Welch DG, Cran DG, Johnson LA. Comparative motility of X and Y chromosome-bearing bovine sperm separated on the basis of DNA content by flow sorting. Mol Reprod Dev 1998; 50:323- 7. Competing interests: None declared
Time to pregnancy and sex of offspring. Authors’ reply to J Olsen et al, DJR Hutchon et al, MJ Davies, JC van der Wouden, R Slama et al, WH James, and VJ Grant 27 June 2006
Luc Smits, lecturer Maastricht University, Dept. Epidemiology, PO Box 616, 6200MD Maastricht, The Netherlands, Rob de Bie, Gerard Essed, and Piet van den Brandt Send response to journal: Re: Time to pregnancy and sex of offspring. Authors’ reply to J Olsen et al, DJR Hutchon et al, MJ Davies, JC van der Wouden, R Slama et al, WH James, and VJ Grant	We would like to thank Drs Olsen et al, Hutchon et al, Davies, Van der Wouden, Slama et al, James and Grant for their responses to our paper. The replications of our study by Olsen et al and Slama et al (refs) do not confirm our observation of an association between time to pregnancy (TTP) and sex of the offspring. Although chance cannot be discarded as an explanation for the heterogeneity of the results, other explanations are worth examining. There are several differences between the studies that might be considered. First, the analysis by Olsen et al included only first deliveries, while our analysis and that of Slama et al included both first and later deliveries (although only one per woman). In the analysis of Slama et al, a rising trend of sex ratio with longer TTP was visible among TTPs below 18 months. The trend disappeared after restricting the analysis to primiparous women. In our data, there was a suggestion of effect modification by parity. In primiparous women, the slope of the logistic function was smaller than among secundiparas or higher-order parity (β=0.0110 for primiparas, β=0.0138 for secundiparas, β=0.0339 for higher order parity). Possibly, couples struggling to conceive for the first time may generally have other reproductive characteristics than subfertile couples who already managed to establish pregnancy and birth before. Some causes of reduced fertility may be more prevalent among first-time conceivers than among the rest. If these causes also correlate with the likelihood of male offspring, this might explain a parity-specific effect. Second, in the two negative analyses, women with any miscarriage before the current pregnancy were not explicitly asked to start counting TTP from the date of the most recent miscarriage, while in our study we did request the women to do so. We are, however, unable to check the impact of this difference, nor do we have a hypothesis about the direction of the bias if any. Third, our study population, in contrast to those of Olsen et al and Slama et al, was not recruited nationwide, but from practices of midwives and gynecologists throughout the southeastern part of The Netherlands. It is however unlikely that our population was any more homogeneous with respect to factors that potentially influence both TTP and sex ratio, such as environmental exposures. Our population, on the other hand, included relatively low numbers of women of ethnic minorities (ref); we do not know, however, how this lack of ethnic heterogeneity could have caused the observed association. Fourth, the range of potential confounders evaluated in our study and that of Olsen et al was greater than in the study of Slama et al. In all studies, however, adjusted and unadjusted results did not appear to differ much, and therefore this is not likely to be an important factor. Fifth, we included women with unintended pregnancies in the analysis, while the other authors did not. In order to check whether this difference could have contributed to the heterogeneity of study results, we reanalyzed the data with exclusion of unintended pregnancies. This caused virtually no change in the results. In summary, apart from chance, some differences between the studies may have contributed to the heterogeneity of the results. Of these differences, parity and in/exclusion of time to miscarriage are worth further examination. Dr James argues that our study is in need of replication and that its results are not consistent with those of earlier work. We think that it is obvious that every new finding should be tested for consistency in other, comparable, populations. We are now planning a replication by use of data from a cohort study among mothers and children from Amsterdam. Since pregnancy and birth cohort studies usually measure TTP and sex of the offspring, it should be possible for many groups to copy or approach our design without much effort, as exemplified by the studies of Olsen et al and Slama et al. We agree with Drs Hutchon et al that we should have used the word ”miscarriage” where we used the term “spontaneous abortion”. Dr Van der Wouden draws the attention to a possible curvilinear relationship in the data. We agree that visually a sinusoid curve is discernible, but we cannot think of any plausible biological basis for such a phenomenon. We agree with Dr Davies that prolonged TTP only reflects disorders in a couple’s reproductive capacity, and that our study cannot prove (or disprove) that mucosal viscosity is the key factor in the association. To our knowledge, no studies have been published that described the reproductive characteristics of couples naturally conceiving after different TTPs (in terms of sperm quality, mucus quality, hormonal characteristics etc). Such studies would however be very worthwhile because they could provide clues with respect to the conditions possibly underlying an association between TTP and offspring sex ratio (as well as other outcomes that have been found to vary with TTP). These conditions might not be related to differential velocity of X and Y sperm, as Dr Grant argues. The alternative hypothesis offered by Dr Davies is interesting, but can only hold if couples with a greater number of male offspring are more likely to have male offspring in the future (irrespective of their fecundity); this implicit assumption however lacks any underpinning. References 1. Jørn Olsen, Weijin Zhou, Jin Liang Zhu: Time to pregnancy and sex of offspring (Rapid Response, 21 December 2005) 2. Rémy Slama, Jean Bouyer, Béatrice Ducot, Alfred Spira and Béatrice Blondel: Are subfertile couples more likely to have boys? A comprehensive nationwide study (Rapid Response, 6 January 2006) 3. William H James: Time to pregnancy and sex of offspring (Rapid Response, 17 January 2006) 4. www.abcd-study.nl 5. David J R Hutchon: Miscarriage terminology (Rapid Response, 22 December 2005) 6. Johannes C van der Wouden: Curvilinear relationship ignored (Rapid Response, 24 December 2005) 7. Michael J Davies: Sex ratio and infertility related factors (Rapid Response, 23 December 2005) 8. Valerie J. Grant: Entrenched misinformation about X- and Y-sperm (Rapid Response, 16 March 2006) Competing interests: None declared