Minimum sperm dose for optimal fertility after artificial insemination in ostriches.

Artificial insemination (AI) in ostriches may present potential solutions to high proportions of infertile eggs commonly recorded on commercial farms and assist in reducing the number of males for breeding purposes thereby leading to a more economical and efficient farming system. Although non-invasive methods to collect semen and to artificially inseminate female ostriches have been developed, the insemination dose for maximum fertility of eggs remains unknown. This study was thus conducted to determine the minimum sperm dose that would promote fertility of eggs following AI in female ostriches. A total of 22 South African black ostriches (7 males and 15 females) aged between 2 and 9 years old were used. Semen samples were collected using the dummy female method and diluted 1:4 (semen: diluent) with an ostrich specific semen diluent. Females were inseminated with various sperm doses of diluted semen from the same male three times a week, every second day resulting in a total sperm dose of A: <2.5 × 109, B: 2.5-5 × 109, C: 5-7.5 × 109 and D: 7.5-9.6 × 109 sperm/week. Eggs produced after insemination were opened to determine the fertilization status of the germinal disc (GD) with an unaided eye. The fertile period was then calculated as the number of days fertilized eggs were laid after the last AI. Furthermore, the number of sperm trapped in the outer perivitelline membrane (spermOPVL) above the GD region was counted under fluorescent light, following staining with 4',6-diamidino-2-phenylindole to determine the rate of sperm loss and the number of days up to when the last egg containing sperm was laid. On average, a mean (±sd) of 35.34 ± 25.72% eggs produced after AI were fertilized. Fertility was lower (mean ± se) when sperm dose A was used (6.71 ± 9.40%), as compared to sperm dose B (46.01 ± 6.71%), C (37.34 ± 6.60%) and D (37.75 ± 8.36%) (P < 0.05). No significant difference was recorded in the latter three doses (P > 0.05). Furthermore, the length of the fertile period and the rate of sperm loss did not differ significantly between sperm doses (P > 0.05). Hence, a sperm dose of between 2.5 and 5 × 109 sperm/week is recommended to optimize fertility after AI in ostriches, as increasing the sperm dose would not benefit fertility. Further studies are, however, needed to determine the frequency of insemination that would maintain fertility throughout the breeding season as well as hatchability of eggs laid after AI.

Mass Sperm Motility Is Correlated to Sperm Motility as Measured by Computer-Aided Sperm Analysis (CASA) Technology in Farmed Ostriches.

Semen analyses have gained momentum in various livestock industries. However, in farmed ostriches, semen analysis is still in its experimental stage, and males are not screened for sperm quality before breeding. This study investigated the correlations between computer-aided sperm analysis (CASA) technology and the traditional, yet affordable, mass sperm motility score. Semen was collected from nine South African Black ostrich males (mean age ± SD: 5.25 ± 1.21 years), using the dummy female method for 5 consecutive days monthly, for 8 months. Mass sperm motility scores were recorded on a scale of 1−5 (1: little to no sperm movement; 5: rapid sperm movement). The CASA traits recorded were: total motility (MOT), progressive motility (PMOT), curve−linear velocity (VCL), straight-line velocity (VSL), average path velocity (VAP), amplitude of lateral head displacement (ALH), linearity (LIN), straightness (STR), wobble (WOB), and beat-cross frequency (BCF). The results revealed positive correlations between mass sperm motility and PMOT, MOT, VCL, and VAP ranging from 0.34 to 0.59 (p < 0.0001). In contrast, negative correlations were recorded between mass sperm motility and LIN, STR, and BCF, with correlations ranging from −0.20 to −0.39 (p < 0.0001). VSL, ALH, and WOB were not correlated to mass sperm motility (p > 0.05). Ostrich farmers may thus be able to evaluate sperm motility reliably and potentially select breeding males by using the affordable mass sperm motility scoring method. Determining the correlation between these methods and fertility after artificial insemination or natural mating is however needed.

Mobility of Japanese quail spermatozoa and its relationship to egg fertility.

Sperm mobility (SM) appears to be primary determinant of fertility in chicken and turkey. The aims of this study were to extend the concept to the Japanese quail by developing an assay to quantify SM, explaining the basis of SM using motility properties measured by CASA, and exploring the relationship between SM and egg fertility. The study was carried out in three stages: i) males (n = 20) and females (n = 20) were mated individually; ii) ejaculates were collected from 20 males, and SM was measured; iii) males (n = 20) and females (n = 20) were mated individually. In Stages I and III, data were collected for egg fertility, SpermOPVL and HolesIPVL . In Stage II, SM assay was developed and assay conditions were defined: effect of sperm numbers on absorbance in Accudenz solution; effect of Accudenz concentration on sperm motility and mobility; effect of quail proctodeal gland foam extract and incubation temperature on SM at 37 and 41°C. The recorded absorbance of sperm movement was dependent on sperm numbers in the sperm suspension overlaying the Accudenz (p < .001). At 41°C, SM, progressively motile sperm, VCL, VSL and VAP were negatively affected by Accudenz concentration (p < .05). The effect of foam on SM and motility depended on an interaction between the concentration of foam extract and incubating temperature. Males were categorized into low, average and high SM phenotypes. These categories differed significantly (p < .001), but sperm motility and SM were not related to egg fertility. In conclusion, SM assay can be used to identify mobility phenotypes, but the poor relationship between SM and egg fertility indicates a need for further studies on interaction between the concentration of foam extract, incubating temperature, and in vivo sperm movement and egg fertilization success.

The effect of dilution rate and successive semen collections on sperm quality and sexual motivation of sexually mature South African Merino rams.

This study investigated whether dilution rate and successive semen collections influenced sperm viability, morphology, motility and male sexual motivation in sexually mature South African Merino rams (SAMR). Semen was collected from 11 rams up to either sperm or behavioural exhaustion. Semen was then immediately serial diluted to make 0, 2, 4, 8 and 16× dilutions with Hams F10 diluent. Following dilution, sperm motility was evaluated using computer-assisted sperm analysis (SCA®), while sperm morphology and viability were assessed using nigrosin-eosin staining and SYBR14/PI, respectively. Male sexual motivation was recorded by reaction time to first mount, while male dexterity was calculated as the ratio of mounts to ejaculations. An increase of dilution rate did not affect sperm motility (P > 0.05) but resulted in a significant decrease in the percentage of live normal sperm (P < 0.05). Furthermore, while sperm concentration and number decreased with semen collection number (P < 0.05), no effect was detected on sperm viability, morphology and motility (P > 0.05), except for average curvilinear velocity which showed a biphasic trend (P < 0.05). Mating success and reaction time were negatively affected by successive semen collections (P < 0.05), while male dexterity was unaffected (P > 0.05). Nevertheless, relatively high numbers of motile sperm (>2 billion) were collected up to the 4th successive semen collection, with a short initial reaction period (<25 s) and good success rate (>65%). SAMR thus withstood frequent semen collections without affecting their sperm reserves or sexual motivation. Further studies are required to investigate optimal conditions for semen collection and artificial insemination in this breed.

Extensive human presence and regular gentle handling improve growth, survival and immune competence in ostrich chicks.

A total of 416 day-old ostrich chicks were randomly allocated to one of the three different husbandry practices for 3 months after hatch; HP1 (extensive human presence with gentle human voice, visual and gentle physical stimuli), HP2 (similar to HP1 but without physical stimuli) and S (human presence limited to supply of feed and water). Chick weight (kg) was measured at 6 and 12 weeks of age, while mortalities were recorded daily to calculate the survival rate. Finally, chicks' antibody responses to vaccination against Newcastle disease (NCD) was measured using the Hemagglutination-Inhibition (HI) test at 20 weeks of age. While HP1 chicks were heavier and survived better to 6 weeks of age than HP2 and S chicks (p < .05), no difference was observed thereafter (p > .05). Furthermore, HP1 chicks had an improved immune competence, as illustrated by their lower percentage of positive HI titers, compared to HP2 and S chicks (p < .05). Hence, integrating extensive human presence with positive human-chick interactions may assist in alleviating challenges related to chick rearing in the ostrich industry.

Age-related declines in ejaculate quality and sperm kinematics vary among strains of Japanese Quail (Coturnix japonica).

For successful breeding programs, it is important to quantify the useful period of a male's reproductive life and it is often done simply by measurement of semen quality. This information is lacking for Japanese quail so we tested whether there is a decline in ejaculate quality and sperm kinematics with age, and whether the decline varies among strains. Nine males (n = 9) from each of 5 strains (A, B, C, D and E) were subjected to 4 semen collections (n = 16 per male) at 8, 16, 26 and 36 weeks of age. Ejaculate volume, sperm concentration and total sperm per ejaculate were measured, and sperm kinematics were analysed using a Sperm Class Analyser (SCA® ). There was a significant effect of age for ejaculate volume, total sperm per ejaculate and per cent medium sperm. The effect of the interaction between age and strain was significant for percent progressive motile sperm, percent rapid sperm, velocity curvilinear, velocity straight line, velocity average path, linearity, straightness and beat cross frequency. Ejaculate volume peaked at Week 26 in all strains, while peak values for sperm concentration and total sperm per ejaculate were observed at Week 16 for most strains. There were declines in percent motile sperm, progressive motile sperm and rapid sperm, and in velocity curvilinear velocity, velocity straight line and velocity average path, by Week 16 for most strains. Linearity declined by Week 26 in some strains, and all strains showed a significant decline in beat cross frequency by that age. In conclusion, the ability of CASA to detect age-related changes in sperm kinematics makes it a valuable tool for identifying the best males and thus improving quail flock fertility. It is essential that breeders understand that age affects both sperm production and sperm kinematics, and that the changes vary with strain.

The Effect of Extensive Human Presence at an Early Age on Stress Responses and Reactivity of Juvenile Ostriches towards Humans.

The effect of extensive human presence and regular gentle handling performed at an early age (0⁻3 months old) on stress responses and reactivity of juvenile ostriches towards humans was investigated. A total of 416 ostrich chicks over two years were exposed to one of three treatments for three months after hatching; namely, Human Presence 1 (HP1, N = 144): extensive/prolonged human presence with physical contact (touch, stroking), gentle human voice, and visual stimuli; Human Presence 2 (HP2, N = 136): extensive/prolonged human presence without physical contact, but with gentle human voice and visual stimuli; and the Standard treatment (S, N = 136): human presence limited to routine feed and water supply as a control. At 7.5 months of age, the plasma heterophil/lymphocyte (H/L) ratio was measured before and 72 h after feather harvesting and feather clipping to determine acute stress responses, while chronic stress was measured by quantification of corticosterone (CORT) concentrations in the floss feathers of the birds. Birds' behavioural response towards a familiar or an unfamiliar handler was evaluated at 12 months using docility and fear tests, and through behavioural observations conducted on random days between the ages of 8⁻13 months. Willingness to approach, and to allow touch interactions, aggressiveness, and exhibition of sexual display towards the handler, was recorded. No difference in the H/L ratios before and after feather harvesting and clipping was observed in HP1 birds, whereas H/L ratios showed a significant increase 72 h post feather harvesting and clipping in HP2 and S birds (p < 0.05). Birds from the S treatment exhibited a significantly (p < 0.05) higher feather CORT concentration compared with HP1 birds, while HP2 birds had intermediate responses. Birds' reactivity towards humans and temperament as evaluated using behavioural observations, docility, and fear tests was not affected by treatment (p > 0.05). However, HP1 and HP2 birds were more inclined (p < 0.05) to approach a familiar rather than an unfamiliar handler during the behavioural observations, indicating an ability to distinguish between a familiar and an unfamiliar handler. Overall, the results indicate that early gentle human interactions with ostrich chicks can be beneficial in reducing physiological stress sensitivity later in life and facilitate the ability of ostriches to distinguish between familiar and unfamiliar handlers.

Correlation between objective semen analysis and fertility in Japanese quail.

This study investigated the relationship between sperm kinematics and egg fertility in Japanese quail in an attempt to identify a semen trait that could be used to predict male fertility. Males (n=45) and females (n=180) from five strains (A, B, C, D, E) were used. Ejaculates (n=720) were collected from 8 to 38 weeks of male age. Semen volume and sperm concentration were recorded and sperm motility was analyzed using Sperm Class Analyzer (SCA®-CASA). At the time of ejaculate collection, males were allowed to mate with females in order to obtain egg fertility data: percent fertile eggs and the numbers of sperm (SpermOPVL) and sperm-holes (HolesIPVL) present on the egg perivitelline membranes. Sperm concentration was positively correlated with ejaculate volume (r=0.35; P < 0.01) and percent fertile eggs (r=0.08; P < 0.05). There were high correlations (P < 0.01) among the sperm kinematic parameters: percent motile (MOT%), percent rapid (Rapid%), percent progressive (PROG%), percent medium (Medium%), velocity curvilinear (VCL), velocity straight line (VSL), and velocity average path (VAP), linearity (LIN%), straightness of trajecotry (STR%) and beat cross frequency (BCF). The strains differed with respect to the correlations for sperm kinematics and egg fertility: SpermOPVL was correlated (P < 0.05) with VSL (r=0.18), VAP (r=0.18), and BCF (r=0.23) for Strain A, with PROG% (r=0.17) for Strain B, and with Medium% (r=0.18) for Strain C. When the data were adjusted for the effects of strain and age, Medium% was correlated with HolesIPVL (r=0.36; P < 0.01) and percent fertile eggs (r=0.31; P < 0.01), whereas PROG% was correlated with SpermOPVL (r=0.30; P < 0.01) and HolesIPVL (r=0.30; P < 0.01). Males could be ranked into high and low fertility categories based on initial (i.e. Week 8) and the life-time (i.e. Weeks 8-38) data for sperm kinematics and egg fertility. Of the males classified as poorly fertile by Week 8 sperm kinematics, 30% were also confirmed as poor on the basis of life-time sperm kinematics. Of the males classified as poorly fertile by Week 8 egg fertility, 47% were confirmed as poor on the basis of life time-data of egg fertility. We concluded that sperm concentration, Medium%, PROG%, VAP, VSL and BCF are important determinants of egg fertility in quail, and that these relationships depend on genotype.

Developing a female-only flock for artificial insemination purposes in ostriches: Progress and future directions.

The development of a flock of females that can produce eggs and maintain egg production rate without the presence of males is a prerogative for a viable artificial insemination protocol in ostriches. Over six consecutive breeding seasons (May-December, 2009-2014), we recorded the egg production performance of 40 single-penned (ART) South African Black ostrich females (2-9 years of age), and compared these records with the egg production of 162 pair-mated females of comparable age from the breeding flock (BP). ART females laid significantly fewer eggs than BP females (mean±SEM: 3.49±0.13 eggs per month vs. 4.64±0.09 eggs per month respectively; P<0.001). Both groups showed a similar pattern of laying, with a peak production in July to September. The mean egg weight of ART females was significantly lower than those of BP females (1367±2.25g vs. 1423±1.1g, respectively; P<0.001). Furthermore, female age significantly affected egg production and egg weight whereby BP females reached a peak egg production at 3 years of age, while in ART females, egg production was the highest at 5 years of age. Interestingly, the number of eggs produced, clutches and eggs per clutch of ART females were independent of visual stimulation from the males. These results indicate that male presence is not needed to ensure egg production. Continuous recruitment of young females based on human-friendly behaviour to breeding by artificial insemination from high egg production performance parents could improve egg production of the ART flock. Studies are also needed to gain a better understanding of underlying physiological mechanisms promoting spontaneous ovulation in this species.

Roles of small RNAs in the effects of nutrition on apoptosis and spermatogenesis in the adult testis.

We tested whether reductions in spermatozoal quality induced by under-nutrition are associated with increased germ cell apoptosis and disrupted spermatogenesis, and whether these effects are mediated by small RNAs. Groups of 8 male sheep were fed for a 10% increase or 10% decrease in body mass over 65 days. Underfeeding increased the number of apoptotic germ cells (P < 0.05) and increased the expression of apoptosis-related genes (P < 0.05) in testicular tissue. We identified 44 miRNAs and 35 putative piRNAs that were differentially expressed in well-fed and underfed males (FDR < 0.05). Some were related to reproductive system development, apoptosis (miRNAs), and sperm production and quality (piRNAs). Novel-miR-144 (miR-98), was found to target three apoptotic genes (TP53, CASP3, FASL). The proportion of miRNAs as a total of small RNAs was greater in well-fed males than in underfed males (P < 0.05) and was correlated (r = 0.8, P < 0.05) with the proportion of piRNAs in well-fed and underfed males. In conclusion, the reductions in spermatozoal quality induced by under-nutrition are caused, at least partly, by disruptions to Sertoli cell function and increased germ cell apoptosis, mediated by changes in the expression of miRNAs and piRNAs.

Nutrition affects Sertoli cell function but not Sertoli cell numbers in sexually mature male sheep.

We tested whether the reversible effects of nutrition on spermatogenesis in sexually mature sheep were mediated by Sertoli cells. Rams were fed with diets designed to achieve a 10% increase (High), no change (Maintenance) or a 10% decrease (Low) in body mass after 65 days. At the end of treatment, testes were lighter in the Low than the High group (PP<0.05) in the expression of seven Sertoli cell-specific genes. Under-nutrition appeared to reverse cellular differentiation leading to disruption of tight-junction morphology. In conclusion, in sexually mature sheep, reversible reductions in testis mass and spermatogenesis caused by under-nutrition were associated with impairment of basic aspects of Sertoli cell function but not with changes in the number of Sertoli cells.

Predicting ejaculate quality and libido in male ostriches: effect of season and age.

The success of artificial breeding program depends largely on the reproductive performance of males. Male performance can vary with season and age impacting on quality and quantity of semen collected for artificial insemination purposes and therefore fertility of inseminated females. We examined variation in semen output and male libido of seven male ostriches (aged 2-5 years) over a period of 24 months. We collected ejaculates using a dummy female and measured semen characteristics (ejaculate volume, sperm concentration, number of spermatozoa per ejaculate, sperm motility and morphology) and male libido (willingness to mount the dummy). A total of 1006 ejaculates were collected. Across months, the volume of semen (mean ± SEM) ranged from 1.03 ± 0.12 mL to 1.85 ± 0.07 mL, the sperm concentration from 3.21 ± 0.12 × 10(9)/mL to 4.16 ± 0.74 × 10(9)/mL, and the number of spermatozoa from 3.42 ± 0.28 × 10(9) to 7.66 ± 0.47 × 10(9). The largest volume of ejaculates and the highest number of sperm were collected in spring. Ejaculates with higher number of normal sperm were also collected in spring-early summer, whereas ejaculates with higher numbers of live abnormal and dead sperm were collected in winter. Sperm motility was relatively constant over months, despite a reduction in summer (January-February), while male libido peaked in winter (June-July) and spring (October-November). Furthermore, we observed high individual variation between males for all variables tested, except for motility. These results indicate that collections conducted in spring yield higher number of spermatozoa, when the libido of males is also at a maximum. Therefore in this species seasonal variation in semen quality should be considered in breeding programmes by artificial insemination to maximise fertility.

Under-nutrition reduces spermatogenic efficiency and sperm velocity, and increases sperm DNA damage in sexually mature male sheep.

We tested whether the quality of spermatozoa from mature male sheep would be affected during nutrition-induced changes in testicular mass. Merino rams were fed for 65 days with diets that increased, maintained or decreased body and testis mass (n=8 per group). In semen collected on Days 56 and 63, underfed rams had less sperms per ejaculate than well-fed rams (P<0.05) and a lower sperm velocity (computer-assisted semen analysis) than well-fed or maintenance-fed rams (P<0.05). Sperm chromatin structure assay revealed more sperm DNA damage in underfed rams than in well-fed rams (P<0.05). The amount of sperm DNA damage was inversely correlated with change in scrotal circumference (r=-0.6, P<0.05), the percentages of progressive motile sperm (r=-0.8; P<0.01) and motile sperm (r=-0.6, P<0.05), and the numbers of sperms per gram of testis (r=-0.55, P<0.05). In testicular tissue collected on Day 65, underfed rams had fewer sperm per gram of testis than rams in the other two groups (P<0.001). We conclude that, in adult rams, underfeeding reduces spermatogenic efficiency and that this response is associated with a reduction in spermatozoal quality.

The effect of temperature and pH on the motility and viability of ostrich sperm.

As the chemical environment of semen diluents can have a profound effect on sperm quality, we examined the effect of temperature and pH on the motility and viability of sperm in the ostrich. Semen was collected from four males, each male being replicated three times. Ejaculates were diluted and incubated for 10 min at 20°C and 40°C in four different buffers, temperature adjusted at pH 6, 7, 8 and 9 respectively. Average path velocity (VAP), curvilinear velocity (VCL), straight-line velocity (VSL), linearity (LIN), beat cross frequency (BCF) and amplitude of lateral displacement (ALH) were then recorded for each sample using CASA. The viability of sperm was assessed using nigrosin-eosin staining. Sperm incubated at 40°C had higher motility parameters, except for ALH. At 40°C, VAP, VSL and LIN increased with pH while VCL, BCF and ALH were higher for lower pHs. The viability of sperm was not affected by temperature but decreased at pH values>7. A pH in the neutral range appeared to yield higher quality sperm after in vitro storage at 20°C. However, the effect of different pH levels and temperatures on sperm longevity needs to be investigated further to develop viable ostrich specific diluents.

Artificial insemination technology for the emu--improving sperm survival.

For the emu, where monogamous mating is normal, artificial insemination (AI) promises much faster genetic improvement and a considerable reduction in production costs by reducing the number of male birds needed for mating. Semen collection is now a routine procedure so the next step is to develop successful protocols for sperm storage. In this paper, we briefly overview our recent progress on the development of protocols for liquid storage and cryopreservation of emu spermatozoa. We have shown that emu semen can be stored at 10 °C for up to 48 h with a minimal loss of viability, and that cryopreservation with dimethylacetamide (DMA) as a cryoprotectant is feasible because we have observed no adverse effects of this cryoprotectant on the emu sperm membrane integrity, morphology and motility. We now need to establish the predictability of the various tests in vivo, but the proportions of live normal and motile sperm with good egg membrane penetration potential suggest that acceptable numbers of competent sperm are preserved and that this will be sufficient for AI.

Twice daily collection yields greater semen output and does not affect male libido in the ostrich.

The success of an artificial insemination program in ostriches is highly dependent on the yield of viable semen. We, therefore, tested how semen output is affected by three different collection frequencies: once every 2d (48h interval), daily (24h interval), and twice a day (6h interval). Ejaculates were collected from seven male ostriches (aged 2-4 years) for 10 consecutive days using the dummy female method. We assessed semen characteristics (sperm motility, volume, concentration, number of sperm per ejaculate and sperm viability) and male libido (the delay between the presentation of the dummy and ejaculation, and the willingness to mount the dummy). The total daily output of semen and the number of sperm were greater at the 6h collection interval than at the 24h or 48h interval while sperm motility and viability were not affected. At the 6h interval, the number of live normal sperm increased over the treatment period while the number of live abnormal sperm was reduced. Furthermore, the time that males took to mount the dummy and their willingness to copulate with the dummy were unaffected by collection frequency. Across males we observed great individual variation in both semen characteristics and libido suggesting there is the potential to increase the efficiency of semen collection by selecting superior males. These results indicate not only that two collections per day yield maximum semen output and may improve semen viability, but also that quantifying variation between males may help further increase semen collection efficiency.

Distribution of spermatozoa in the outer perivitelline layer from above the germinal disc of emu and ostrich eggs.

In the present study, we determined the distribution of spermatozoa in the perivitelline layer above the germinal disc (GD) of emu and ostrich eggs that had been laid at random intervals after mating. Eggs were opened, the perivitelline layer overlying the GD region was collected and sperm were visualized with 4',6'-diamidino-2-phenylindole under a fluorescence microscope. To map the distribution of sperm, the GD was divided into six areas (A-F), with A being the centre of the GD and F the area furthest from the centre. In both species, more spermatozoa were found in areas B, C and D than in areas A, E and F. More than half the GD spermatozoa were found in areas B, C and D. The pattern of distribution of spermatozoa across the GD depended on the total number of sperm in the GD. In the emu, the pattern was related to delay since last copulation and time of laying, whereas in the ostrich the pattern was related to the month of the season and the sex ratio of the mating system. When the total number of spermatozoa in the GD increased, the number of spermatozoa increased in every area of the GD, but the centre and the outer areas were the least affected. We conclude that sperm numbers are highest in a band immediately around the centre of the GD and then decline with increasing distance from the centre. The low numbers in the centre of the GD may be due to either low attractiveness of the centre for sperm or high attractiveness of the area immediately adjacent to the centre.

Fertility of male and female emus (Dromaius novaehollandiae) as determined by spermatozoa trapped in eggs.

Changes in the fertility status of 10 pairs of emus were investigated using egg break-out and numbers of sperm in the perivitelline membrane of the germinal disc (GD) region. After the sexes were separated, sperm in consecutive eggs declined approximately logarithmically at a mean (+/-SEM, n = 10 females) rate of -0.148 +/- 0.021 per log day. Sperm continued to be detected in eggs for 16.5 +/- 1.7 days during which 5.6 +/- 0.6 fertilized eggs were laid. Fertilized eggs that did not contain detectable sperm were laid by five females for a further 2.2 +/- 0.9 days. Based on break-out fertility, the fertile period continued for up to 18.7 +/- 2.1 days, for which the mean number of laid eggs was 6.3 +/- 0.8. An egg with a 50:50 chance of being fertilized would contain 3.5 sperm mm(-2) of GD. Based on the sperm decline model, an egg containing that many sperm would be laid 21 days after the last copulation. In emus that were not separated and allowed to incubate their eggs (n = 3 pairs), the number of sperm in eggs laid before and during incubation declined in a manner similar to that after the last copulation and egg-laying stopped after the females had laid 3.3 +/- 0.3 eggs. After incubation was terminated, females resumed laying within 8.3 +/- 1.2 days and the number of sperm in eggs gradually increased but it did not return to pre-incubation levels. In non-incubating emus (three pairs), the number of sperm in eggs declined as laying progressed, although lit was higher during the period when the first seven eggs were laid than during the period when the rest of eggs were laid (214 +/- 39 v.100 +/- 16 sperm mm(-2) of GD). Sperm numbers varied between successive eggs but a sharp increase followed by a decrease acted as an indicator of recent copulation. There were 8.7 +/- 0.3 such increases per laying period (one per 2.8 +/- 0.2 eggs), a frequency that suggests that emus copulate once weekly. In conclusion, as long as a female emu is supplied with sperm on a weekly basis, she will be fertile but, when copulations stop, she will stop laying soon after. Male fertility appears to fall towards the end of the laying season and it can be affected by egg incubation at any time of the season.