In this paper we discuss the likely future milk production scenarios and breeding and management strategies in the EU in general, and in Britain and Ireland in particular. EU markets for most dairy products are stagnant in volume terms. There is, however, scope for value growth which would emphasise quality and added value, not price. The background scenario is therefore one of sharp commercial focus heavily influenced by consumer demands for quality, not only in physical terms but also in ethical, welfare and aesthetic ones. Future systems of production will need to be in tune with future markets. Perhaps three main sectors can be identified: intensive high output, pasture based systems and niche systems (e.g. organic systems). In each of these options the same questions arise: (i) What kind of cow/breeding strategy is suitable for the sector?, and (ii) What are the management guidelines which will secure efficient, sustainable productivity? In the past, the majority of dairy cattle improvement programmes have focused primarily on improving returns by increasing milk or milk solids yields. Future breeding programmes are likely to pay much greater attention to reducing costs than they have in the past. In pasture-based systems, or niche markets, this may lead to renewed interest in cross breeding to reduce health, re-breeding and replacement costs. In all systems there is likely to be much greater emphasis on traits other than production in selection indexes. Customised indexes will help producers to tailor their selection decisions to their particular markets and production systems. If the differences between future production systems are extreme, it may be cost effective to produce bull evaluations for each of the main systems. New molecular techniques are beginning to assist conventional selection programmes. In the longer term the transfer of genes between strains, breeds or species may be used for agricultural applications. However, it is questionable whether or not this would be acceptable to consumers in the EU. Management issues which will be important in future are exactly the same as they have ever been, dealing with feeding, fertility, health, housing, milking practice, hygiene and pasture management. Nearly all of these interrelate with each other and with breeding strategy. Particular issues in future may include management of robotic milking, loose housing, deliberately extended lactations, organic production systems and extended grazing. Future ‘feeding’ challenges will include optimizing concentrate use for cows of different genetic merit, and finding alternatives to conserved grass. Direct genetic modification of grass and other forages to improve their qualities both as grazed and as conserved material would also be useful. There is also likely to be increased emphasis on feeding cows to improve control of nutrient partition, and on improved feeding of animals in the transition between lactations. Tailoring feeding and management policies to the genetic merit of cows will be a continuing challenge.

The financial implications of delays in conception at different stages of lactation in the average and the high yielding dairy cows were investigated. Variables included in the calculations were net loss in annual yield, lost income from a calf, cost of extended dry period, cost of slipping in calving pattern, cost of extra veterinary treatments and AI services, benefit of a delay in calving for reduced risk of production diseases, potential benefit of delay in conception on milk yield from the current lactation, and the value of quota leasing. For the average yielding cow, the net cost of one day of delay in conception was calculated at £2.41 when conception is delayed from 85 to 100 days post-calving, increasing to £5.02 per day if conception occurred at 146 to 175 days post-calving. After taking value of quota leasing into account, the net cost of a lost day was calculated at £1.73 and £3.55 per day for the two delay intervals respectively. For the high yielding cow, the net cost of one day of delay in conception was calculated at £2.48 when conception is delayed from 85 to 115 days post-calving, increasing to £6.52 per day if conception occurred between 206 and 235 days post-calving. The net costs after quota leasing being considered were calculated at £1.68 and £4.08 per day for the two delay intervals respectively. On the basis of the above estimations, and after considering the cost of culling for poor fertility, it was concluded that it is a cost-effective option to keep trying to get the typical average cow in calf until 266 days post-calving, whereas the breakeven point for the high yielding cow is at 290 days post-calving.

Following parturition, there is an early resumption of sequential but transient FSH increases of 2-3 days duration in dairy and beef cows. The first increase results in the emergence of the first postpartum follicle wave and the decline in FSH results in selection of a dominant follicle (DF). The ovulatory fate of this DF is dependent on LH pulse frequency and IGF-I concentrations. The energy status of the cows affects the degree of anoestrus. High yielding cows in prolonged negative energy balance (NEB) have a greater incidence of anoestrous, where the DF is smaller and fails to produce sufficient oestradiol to induce ovulation. Thus, there are sequential follicle waves in anoestrus prior to first ovulation. A small percent of DFs in dairy cows fail to ovulate and they continue to grow due to high LH pulse frequency and form follicular cysts. These cysts produce oestradiol for a variable period and then become physiologically defunct, despite their morphological presence for variable periods before final regression. During their physiological active state, new follicle wave emergence and ovulation are suppressed. A further problem in high yielding cows is the increased incidence of abnormal ovarian cycles after ovulation, and specifically, the high incidence of maintained corpora lutea (CL), probably related to uterine problems in the peri-parturient period. The ovulation of smaller DFs and the high liver metabolic rate may adversely affect oestradiol and progesterone concentrations. The long-term effects of prolonged NEB on oocyte competence, DF physiology and follicular and luteal steroidogenesis all may have detrimental effects on conception rates to AI. Hormonal methods to induce ovulation should be carried out in conjunction with corrective management problems, once diagnosed The pre-treatment of anoestrous cows with progesterone for 5-9 days is a prerequisite for the concomitant expression of oestrus at first ovulation, but whether or not further hormonal therapy is necessary to ensure the ovulation of the DF, is dependent on energy status, body condition score and postpartum interval, which regulate LH pulse frequency.

A number of measurable physiological events characteristically occur and cause changes during the perioestrual period including the classical, diagnostic sign of standing behaviour. The onset of oestrus coincides with peak titres of oestradiol-17β that subsequently induce the preovulatory surge of LH within 1 to 3 h and ovulation of a mature follicle some 24 to 32 h after the onset of oestrus. Although detection efficiencies are consistently greater in higher producing herds, oestrus-detection efficiency generally has declined in recent years as herd size and milk production have increased. New technologies have introduced some needed assistance for detecting cows in oestrus. These include various in expensive heat mount detectors to more sophisticated electronic gadgetry, such aspedometry and radiotelemetric sensors that detect temperature, tissue impedance, and pressure. Oestrus detection aids are usually more efficient but not necessarily more accurate than visual observation. Differences in housing and environmental conditions, in addition to labor inputs, costs, and efficacies, result in variable acceptance of such technologies. Detection efficiency and accuracy can be improved by simultaneous use of synergistic technologies; those that compliment each other and monitor different indicators of oestrus. Combining technologies for simultaneous measurements of several physiological events associated specifically with the onset of oestrus and their radiotelemetrically signaling to a central computer for subsequent analysis should provide greater efficiency ofoestrus detection with fewer false positives. The ultimate goal of determining the onset of oestrus or ovulation is to predict the optimal timing for insemination. Ultimately, herd personnel must interpret information gathered by these technologies and judge whether or not and when to inseminate cows based on their visual inspection of identified cows.

The abnormalities of the puerperium can be arbitrarily divided into two broad categories. Those which arise following a difficult calving or retained placenta are referred to as “common”, while abnormalities such as sub- and anoestrus, irregular cycles, cystic follicles, absence of ovarian activity, repeat breeding and increased number of inseminations per pregnancy are described as “special”. Since the “common” abnormalities are well understood, this paper deals in greater detail with the “special” abnormalities of the puerperium. Based on the axiom that many facets of the “special” abnormalities can be caused by dysfunction of the liver (as in the case of fatty liver), information is presented based on experiments conducted with the over-conditioned dairy cow model (AM) which suffers from both a severe negative energy balance (NEB) and fatty liver syndrome during early lactation. This cow model guarantees a severe NEB, with body weight loss and clear changes in the body condition score (BCS), lower milk production, a higher milk fat concentration, temporary decline in blood insulin and glucose concentrations, elevated blood concentrations of non esterified fatty acids (NEFAs) and beta-hydroxybutyrate (BHB) as well as elevated liver concentrations of tri-acyl-glycerol (TAG) during the first weeks post partum. Additionally, this model leads to a higher incidence of metabolic and infectious diseases. Particular attention was paid to LH pulse frequency and to its value in predicting the occurrence of first post partum ovulation would occur, to oestrous behaviour and to oocytes quality. Cows with fatty liver (> 50 mg TAG/g liver tissue) had longer intervals between parturition and 1st ovulation and showed fewer standing heats within the first 100 days post partum, but the proportion of detected heats, compared to endocrinological “heats” was higher in confirmed AM-cows. Moreover, when AM cows come in heat, their heats lasted longer and had a higher pregnancy rate following AI. However, the developmental capacity of oocytes in AM cows, and in particular oocytes that were destined to ovulate between 80 and 120 days pp, was adversely affected. We conclude that the fatty liver condition is a trigger for many problems, including metabolic and infectious diseases, and also of reduced fertility. Prevention of over fatness at calving (BCS>3.5) is the best guarantee for a normal puerperium.

A series of in vitro and in vivo experiments were conducted to characterise the dialogue between embryo and maternal units relative to the mechanisms controlling embryo survival in dairy cattle. Endometrial explants from pregnant cows had an attenuated PGF2α secretory response following treatment with melittin (stimulator of PLA2) and phorbol 12, 13 dibutyrate (PDBu). Thus previous exposure to the conceptus appears to regulate the endometrial synthetic pathway at a point coincident with or distal to PLA2 as well as inhibit PKC or PKC mediated events. Endometrial explants collected from cows receiving intrauterine infusions of rblFN-τ had a reduced secretory response following stimulation with PDBu indicating attenuation in PKC activity. Based upon tyrosine-phosphorylation of STAT-proteins and their translocation to the nucleus after treatment with rbIFN-τ, the JAK-STAT pathway is functional in immortalised bovine endometrial cells (BEND cells). Bend cells, exposed to rblFN-τ, reduced PDBu induction of PGF2α secretion and also decreased protein expression of Cox-2 and PLA. RblFN-τ clearly reduced PKC mediated events leading to an antiluteolytic response in endometrial cells. Feeding diets containing 2.6, 5.2 and 7.8% Menhaden fish meal to lactating dairy cows reduced uterine secretion of PGF2α following sequential injections of oestradiol and oxytocin. Thus antiluteolytic effects in early pregnancy may be amplified by feeding by-pass fats. Pregnancy rate to a timed insemination at first service post-partum is increased in association with injection of bST(500 mg; sc) given at insemination. Furthermore injection of bST at time of insemination in superovulated donor cows increased the number of blastocysts and reduced number of unfertilised embryos. Prospects of integrating novel strategies to improve embryo development and survival into reproductive management systems appear to be attainable in high producing dairy cows.

Over the past 30-40 years genetic improvement and better nutrition of dairy cows have led to a significant increase in milk production per cow but this is associated with an increase in cow reproductive wastage. Reproductive wastage in the dairy herd particularly in a seasonal calving system, results in a serious financial loss. Early embryo death accounts for a significant portion of cow reproductive wastage and information is becoming available on the extent and timing of early embryo loss and on aspects of embryo growth, development, metabolism and viability. Such information is necessary to facilitate objective investigation of factors that contribute to early embryo death. For heifers and moderate yielding dairy cows published estimates of fertilisation rate of about 90%, and of average calving rates of about 55% indicate an embryonic and foetal mortality rate of about 40%. Of this total loss, 70 to 80% is sustained between days 8 and 16 after insemination, a further 10% between days 16 and 42 and a further 5-8% between day 42 and term. In high yielding cows there is some evidence of a higher increment of late embryo loss. During the period of greatest embryo loss, between days 8 and 16 after fertilisation, there is a dramatic increase in the growth rate and protein content of embryos, particularly from day 13 to day 16 when the increase is exponential. There is evidence that from day 13 to 15 cattle embryos undergo time and developmental stage-dependent changes in the rate of de-novo protein synthesis and protein phosphorylation. It seems that by the time cattle embryos have elongated they have passed their maximal synthetic activity in terms of protein synthesis and phosphorylation, which seems to occur at day 13 or earlier. While there is little published information on the causes of embryo loss it is clear that even a short-term reduction in energy intake near the time of insemination can significantly reduce embryo survival rate. Abo and low post-ovulatory systemic progesterone has been associated with increased embryo loss. Recent evidence shows that elevated systemic concentrations of ammonia and urea per se do not reduce embryo survival rate but there may be other modifying factors, such as negative energy balance, operating in the high yielding dairy cow that lead to reduced fertility when the systemic concentrations of urea and or, ammonia are high.

Decreasing levels of fertility of dairy cows are occurring, associated with increased average annual milk yields, increased herd size and a decreased labour investment per animal. To-date, there has been no positive genetic selection in the UK for improved female fertility due to the lack of reliable recording of fertility traits. Selection is further limited by the low heritabilities of traditional fertility measures, which are subject to environmental influences and management decisions and biological restraints such as age and sex. Assessment of the hormone patterns of fertile cows and determination of the atypical patterns exhibited by subfertile animals provides an objective method of identifying the causes and assessing the impact of subfertility and for the development of remedial treatment strategies. This knowledge can then be used to identify potential physiological parameters associated with high fertility which, in future, may be used for sire breeding value estimations to select for more fertile offspring. Regular assessment of the progesterone concentrations in milk provides a non invasive method of determining progesterone patterns. The results of two studies of milk progesterone levels in cows taken between 1975/82 and 1995/98 involving over 3200 lactations have been analysed and compared. They indicate a major increase over this period in the proportion of animals showing atypical milk progesterone patterns before mating from 32 to 44% (P<0.001) associated with less animals inseminated, delays to conception and lower conception rates. There was a significant decrease (P<0.01) over this period in animals calving to first postpartum insemination from 57% to less than 40% a decrease of approximately l%per annum. This may indicate an increase in the level of early embryo mortality. An early post ovulatory progesterone rise to adequate luteal phase levels has been shown essential for normal embryo development with low post ovulatory levels occurring in some cows resulting in lower calving rates. Milk progesterone analysis provides a robust and reliable method of measuring progesterone patterns for identifying subfertile animals, for targeting remedial treatments to improve fertility and for investigations into heritable fertility parameters for future selection programmes.

The last decade of the 20th century has seen unprecedented change within the UK dairy industry, with consolidated of production quotas, payment schemes based on milk solids, deregulation of the milk market with the demise of the Milk Marketing Board and more recently a major downward pressure on milk prices. All of these have contributed to further intensification of a large part of the industry. However the largest change has been the introduction of improved genetics, with the accompanying increase in Holsteins (now the preferred breed) at the expense of Friesians. As a consequence individual cow yields have increased dramatically and such cows are becoming an increasing problem in respect of nutritional management, as it relates to optimisation of milk solids content and yield and maintenance of reproductive competence. This paper has attempted to examine some of these issues by first describing the changes in more detail and then from a theoretical consideration of the biochemical principles associated with milk solids production to provide information which may permit better decisions to be made in relation to the feeding of such cows. In this respect the importance of glucose is established as well as the potentially competing processes of energy use for body maintenance or milk fat synthesis. Two experimental studies from this laboratory with high genetic merit cows are used to illustrate how the overall utilisation of energy for lactation and body tissue metabolism may be affected, especially when energy intake in early lactation fails to meet whole body energy demands. Subsequently, the nutrient costs of pregnancy, an important but often forgotten part of the ‘annual’ life cycle of dairy cows, are also considered and it is concluded that increased emphasis needs to be put on the management of cows for lifetime performance rather than simply for lactational output. Finally, whilst recognising that high yielding cows represent a major challenge to all management systems, the paper concludes that existing feeding systems which rely solely on nutritional entities such as metabolisable energy will need to be abandoned in favour of those which recognise specific nutrients, if the full potential of high genetic merit cows is ever to be achieved

Increased genetic potential for milk production has been associated with a decline in fertility of lactating cows. Following parturition the nutritional requirements increase rapidly with milk production and result in negative energy balance (NEBAL). NEBAL delays the time of first ovulation thereby affecting ovarian cycles before and during the subsequent breeding period The effects of NEBAL on reinitiation of ovulation are manifested through inhibition of LH pulse frequency and low levels of glucose, insulin and IGF-I in blood that collectively restrain oestrogen production by dominant follicles. Upregulation of LH pulses and peripheral IGF-I in association with the NEBAL nadir increases the likelihood that emerging dominant follicles will ovulate. The legacy of NEBAL is reduced fertility after insemination in conjunction with reduced serum progesterone concentrations. Diets high in crude protein support high milk yield, but may be detrimental to reproductive performance. Depending upon protein quantity and composition, serum concentrations of progesterone may be lower and the uterine luminal environment is altered. High protein intake is correlated with plasma urea concentrations that are inversely related to uterine pH and fertility. The direct effects of high dietary protein and plasma urea on embryo quality and development in cattle are inconsistent. In conclusion, the poor fertility of high producing dairy cows reflects the combined effects of a uterine environment that is dependent on progesterone, but has been rendered suboptimal for embryo development by antecedent effects of negative energy balance and may be further compromised by the effects of urea resultingfrom intake of high dietary protein.

The reduced fertility that is becoming more evident in high yielding dairy cows may be related to many factors including changes in milk production, food intake and fluctuations in body condition. Metabolic and production markers have been studied as a way of predicting success to a particular artificial insemination. Successful conception to a particular service was not associated with milk production, body condition or plasma concentrations of several indicators of metabolic state around the time if insemination. This highlights the importance of time of information collection in fertility management programmes. Increased food intake may reduce systemic progesterone concentrations. This is more evident in sheep than cattle, but a positive relationship between systemic progesterone early post mating and establishment of pregnancy in cattle has been reported. However, progesterone concentrations in the ovarian vein and endometrium are not strongly correlated with systemic progesterone. Thus, the significance of modest changes in systemic progesterone in affecting oocyte and embryo development must be questioned. Blood urea concentrations can be altered by diet, and reduced pregnancy rates have been reported in cows with high urea concentrations. However, in other recent studies, no difference was reported in serum urea in cows that conceived and those that failed to conceive. Pregnancy rate was equally high in heifers when in-vitro produced embryos were transferred to heifers on high and low urea diets. When embryos were produced in sheep on high and low dietary urea, the effects on embryo development appear to occur early in the developmental process, suggesting a substantial effect on the development of the oocyte. The developmental capacity of oocytes and quality of embryos is reduced in cattle maintained on extremely high dietary intakes. Oocyte developmental capacity is reduced in cows of higher genetic merit and embryo quality can be substantially reduced in the early postpartum period. Collectively, these results suggest that high dietary intake or high metabolic load is deleterious to normal oocyte development and establishment of pregnancy. This highlights the importance of further studies on the effect of dietary intake on metabolic state and follicle, oocyte and embryo development. In a practical context, these results highlight the importance of nutritional management and avoiding changes in the amount or type of diet around the time of mating in high-production dairy cows.

To further identify factors which influence pregnancy rates, three experiments were conducted to determine the effect of insemination time on sperm transport, fertilization rate, and embryo quality. All cows were continuously monitored for behavioural oestrus by HeatWatch®, and received AI at heat onset (0 h after the first standing event), 12 h after onset, or received natural service at 0 hfrom one of three bulls (Exp. 1). In Exp. 2, cows received AI at 0 h, 12 h, or 24 h after the first standing event. On d 6 after insemination 115 embryos(ova) (Exp. 1) and 117 embryos(ova) (Exp. 2) were recovered from single-ovulating cows. For Exp. 1, median accessory sperm values were: 1 (0 h), 10 (12 h), 27 (natural service O h) (P < 0.05). For Exp. 2, median accessory sperm values were: 1 (0 h), 2 (12 h), 4 (24 h) (P < 0.05). Fertilization rates were: 67% (0 h), 79% (12 h), 98% (natural service O h) (P < 0.05)(Exp. 1); and did not differ in Exp. 2. Embryo quality was not different in Exp. 1. In Exp. 2, percentages of excellent and good fair and poor, and degenerate embryos were: 77, 15, 8 (0 h), 52, 38, 10 (12 h), 47, 19, 34 (24 h) (P < 0.05). In Exp. 3, 30 cows were superovulated and were inseminated once at either 0 h, 12 h, or 24 h after the first standing event. On d 6 after insemination, 529 embryos(ova) were recovered. Fertilization rates were: 29% (0 h); 60% (12 h); 81% (24 h)(P < 0.01). Percentages of embryos with accessory sperm were: 5 (0 h); 8 (12 h); and 41(24 h) (P < 0.01). Embryo quality was not affected by time of AI. We conclude that the time of insemination affects: 1) sperm transport as measured by median accessory sperm number (Exp. 1 and 2) and the percentage of embryos with accessory sperm (Exp. 3); 2) fertilization rate (Exp. 1 and 3); and embryo quality (Exp. 2).

The economic consequence of low heat detection rates is the main reason cattle reproduction research programmes continue to focus on developing practical controlled breeding systems for dairy cows. Three approaches can be taken to control the oestrous cycle in cattle: (i) Use of the luteolytic agent prostaglandin F2α alone or one of its potent analogues, (ii) Cycle regulation using short-term progestagen treatments and (Hi) Prior follicle wave synchrony followed by induced luteolysis. Administration of prostaglandin F2α (PGF2α) to cows after about Day 5 of the oestrous cycle causes immediate regression of the CL with progesterone concentrations declining rapidly to basal concentrations within 24 hours. Because PGF2α is only effective in animals between Days 5 and 17 of the oestrous cycle up to 40% of any group of randomly cyclic animals will not respond to a single administration. A number of PGF2α administration regimens have been developed for dairy cows. A two-injection regimen, with an interval of 9-13 days between successive administrations, elicits a higher response, as all animals possess a susceptible CL the time of the second injection. In general, cyclic heifers respond with good precision of heat onset to two PGF2α injections 11 days apart. Fixed-time AI at 72 and 96 hours after the second injection can result in acceptable pregnancy rates. In post partum dairy cows the slower turnover of follicle waves, contribute to the greater variation in oestrus onset, resulting in more variable and generally lower pregnancy rates to fixed-time AI at 72 and 96 hours. Consequently, intensive oestrous detection and AI on the basis of observed oestrus is more appropriate for lactating cows following PGF2α administration. It is generally accepted that fertility of the cyclic heifer after PGF2α treatment is not impaired or may even be higher when compared with untreated control animals inseminated at a natural oestrus. However, in lactating cows, conception rate following PGF2α treatment have been frequently lower (about l0%) than in cows bred following a natural heat. Recently, aprotocol to synchronise follicular development using GnRH and to induce luteal regression using PGF2α was developed In this protocol (Ovsynch), GnRH is administered to cows at random stages of their cycle 7 days before luteolysis is induced with PGF2α; a second GnRH injection at 36-48 hours later induces ovulation. All cows are inseminated once at 16-20 hours after the second GnRH injection. Ovsynch system is not recommended for use in heifers. The development of progestogen delivery devices, such as the PRID, CIDR and Crestar has facilitated the use ofexogenous progestogens for oestrous cycle synchronisation in cattle. Oestradiol is administered at the start of treatment to (i) shorten the life span of the corpus luteum (CL) and (ii) to induce the emergence a new follicle wave. Generally, a high proportion (up to 85%) of the cows that exhibit oestrus after removal of the progestogen do so between 36 and 60hours ofremoval. This allows for either one (54-56 hours) or two fixed-time AIs (48 & 72 hours) after progestogen withdrawal. Alternatively, cows can be observed and inseminated once at a detected oestrus. While fixed-time AI of all treated cows gives 100% submission rate conception rates tend to be lower than when AI is on the basis of observed heat. However, the overall number of cows becoming pregnant (submission x conception rate) is frequently the same, reflecting submission and conception rate differences for the two AI options. Choice of AI option will depend the herdsman 's ability to accurately detect heat, semen costs and labour availability.

A range of hormonal therapies has been evaluated to potentially improve the reproductive performance of lactating dairy cows. Early lactation treatments with gonadotrophin releasing hormone (GnRH) or prostaglandin F2∝ (PGF) may reduce the interval to first insemination or increase the conception rate to first insemination, but mainly in cows which have had a difficult pueperium or which are in herds with low conception rates. These two hormones, as well as progesterone and oestradiol benzoate (ODB) are commonly used either singly, or in combination (GnRH + PGF; progesterone + ODB + PGF) to synchronise the oestrus preceding first inseminations. None of these synchrony treatments is associated with increased conception rates. Extensive series of trials have been completed to identify post-oestrous or post-insemination hormonal therapies which could increase conception rates to the preceding insemination. The wide variation in results has precluded any being commonly regarded as sufficiently reliable for routine use. Nonetheless, meta-analyses have shown that GnRH treatment at insemination or in late dioestrus (11 to 13 day post-first insemination) can significantly increase “the risk of pregnancy”. Insemination treatments have been most effective with repeat breeders (+22.5%), whereas late dioestrous treatments (10%) may be dose and analogue specific (10 μg buserelin). Although metoestrous supplementation with progesterone can stimulate early embryonic development, the associated reduction in oestrous cycle length also reduces conception rates in heifers. Late dioestrous use of GnRH can prevent both of these negative effects. Early dioestrous supplementation with progesterone may enhance production of interferon tau, but this potentially beneficial effect has not been able to be reliably translated into increased conception rates. Many of these hormonal therapies are associated with altered patterns of ovarian follicle development which are similar to those in some synchrony treatments preceding first insemination. Recent studies have indicated that OBD and progesterone can be used to synchronise returns to service and increase the submission rate for second inseminations made about 3 weeks after first inseminations. This can make the non-return rate a more accurate measure of the response to a hormonal treatment and potentially overcome confusing impressions created when oestrous detection rates may be around 50%. Even if effective hormonal therapies are successfully developed, the results may be compromised by environmental factors such as heat stress, energy balance or energy partitioning for lactation. These factors may reduce oocyte quality, fertilization rates or normal uterine secretion patterns. Reduced conception rates associated with high daily milk yields in early lactation may not be able to be remedied simply with hormonal supplementation or by altering patterns of ovarian follicle development. Under these circumstances, controlling the inter-service interval could reduce the impact of the lowered conception rates.

There is concern that the substantial increases in production in dairy herds that can be achieved will bring greater health and fertility problems. Despite considerable advancements in the understanding of the biology of reproduction and its control, fertility apparently continues to decline. As many factors, including health and nutrition, influence reproductive performance it is generally agreed that a multidisciplinary approach is necessary if improvement in fertility is to be achieved Herd health schemes have evolved to improve health, fertility, productivity and profitability but the uptake of such schemes in the UK has been low. The veterinary input to many farms is at best routine fertility visits, confined to examination of the reproductive tract, and at worst occasional emergency calls. The Dairy Herd Health and Productivity Service (DHHPS) provides the opportunity for veterinary surgeons to lead a multidisciplinary team which can monitor health, fertility and production and can plan, when necessary, remedial action. Over a period of twenty years it has continued to identify infertility, mastitis and lameness as the main reasons for the involuntary disposal of dairy cows, although on some farms BSE has been a major factor. The trend in the last few years has been for an increase in culling for reasons of disease rather than for age or yield. The average disease rates have remained high with over 100 treatments per 100 cows each year for conditions which may directly or indirectly influence reproduction. Blood profiling and condition scoring demonstrated that at least a third of the cows sampled were mobilising excessive fat during transition from the dry period to early lactation. Improving both health and nutrition, before and after calving, would greatly improve reproductive performance in many herds. A team approach, with farmers, veterinarians, nutritionists and other advisors working together with well defined goals and objectives, is necessary if progress is to be made in improving reproductive performance. High yield can not always be the excuse for poor fertility.

Selection of dairy cattle for increased milk production has decreased some indices of reproductive efficiency. For example, days open are increased by one day for every 100 kg of increased milk yield per lactation. Some of the change in days open can be explained by delayed onset of oestrous cyclicity and lower conception rate to artificial insemination in cows with greater milk production. Despite these negative associations between milk production and reproduction, reproduction in herds of high producing dairy cattle is not necessarily compromised relative to reproduction in herds of low producing dairy cattle. This is because there is a large environmental effect on dairy reproduction. High producing herds generally have better management and better oestrous detection. Therefore, high producing dairy herds may partially overcome the antagonistic relationship between milk production and reproduction. Physiological mechanisms that lead to poorer reproduction in high producing cows are partially defined. Negative energy balance that occurs in high producing dairy cows can be associated with a delay in the initiation of ovarian cycles and the interval to first breeding. Many of the effects of negative energy balance on postpartum reproduction can be explained by decreased serum luteinizing hormone (LH) that is associated with negative energy balance. Serum LH increases as cows move toward positive energy balance and greater LH stimulates growth and ovulation of ovarian follicles. We have initiated studies to address physiological differences in high and low index dairy cows. The reproductive endocrinology of cows from a control line (5,900 kg milk/lactation) and a select line (10,900 kg milk/lactation) of dairy cows at the University of Minnesota was studied over a two-year period. Cows in Year 1 were similar for serum concentrations of LH, follicle stimulating hormone (FSH), and oestradiol (preovulatory period). In both years, serum concentrations of progesterone during luteal phases, however, were decreased in select cows. The Year 2 cows also had a delay in the return to oestrous cyclicity that was associated with reduced LH. The possibility that decreased progesterone causes infertility in dairy cows will require further study. Collectively, these data suggest that changes in blood progesterone concentrations may explain, partially, lower fertility in high index dairy cows.

In recent years there has been considerable genetic progress in milk production. Yet, increases in yield have been accompanied by an apparent lengthening of calving intervals, days open, days to first heat and a decline in conception rates, which appears to be both at the genetic and phenotypic level. Fertility has a high relative economic value compared to production traits such as protein, making it attractive to include in a breeding programme. To do this there needs to be genetic variance in fertility. Measures of fertility calculated from service dates have a small genetic compared to phenotypic variance, hence heritability estimates are small, typically less than 5%, although coefficients of genetic variance are comparable to those of production traits. Heritabilities of commencement of luteal activity determined using progesterone profiles are generally higher, and have been reported as being from 0.16 to 0.28, which could be because of a more precise quantification of genetic variance, as management influences such as delaying insemination and heat detection rates are excluded. However, it might not be the use of progesterone profiles alone, as days to first heat observed by farm staff has a heritability of 0.15. The most efficient way to breed for improved fertility is to construct a selection index using the genetic and phenotypic parameter estimates of all traits of interest in addition to their respective economic values. Index traits for fertility could include measures such as calving interval, days open, days to first service, or days to first heat but there may also be alternative measures. Examples include traits related to energy balance, such as live weight and condition score (change), both of which have higher heritabilities than fertility measures and have genetic correlations of sufficient magnitude to make genetic progress by using them feasible. To redress the balance between fertility and production, some countries already publish genetic evaluations of fertility including: Denmark, Finland, France, Germany, Israel, The Netherlands, Norway and Sweden.

Female fertility affects the culling rate as well as the direct reproduction costs and influences the calving interval and the calving season. It is therefore obvious that female fertility should be included in a Total Merit Index (TMI), which directs the genetic changes in dairy cattle selection. When the first TMI was introduced in Sweden in 1975 female fertility was included. A basic prerequisite for inclusion of fertility in the Scandinavian breeding programme was that milk recording, pedigree and AI-data were integrated in the same databases.

The paper deals exclusively with the additive genetic effects on female fertility. The heritabilities of the measures used for female fertility are low, usually below 0.05. Despite this, the additive genetic variance is high as demonstrated by large differences between progeny groups. Thus, selection must be based on progeny testing of bulls on daughter groups of 100-150 daughters or more, which is the case in all the Scandinavian countries. Different procedures for calculation of daughter fertility indices are practised in Scandinavia.

Research shows the importance of including measures of both the time interval from calving to insemination and a direct fertility trait such as number of inseminations or non-return rate (NR). It has also been shown that reproductive measures of different stages of life, i.e. heifer and lactating cow periods, must be considered. The genetic correlations between daughter fertility and different yield measures are negative (-0.2 to -0.4). Selection for yield without consideration of daughter fertility results in a deterioration of daughter fertility. A comparison of the flat genetic and phenotypic trends for daughter fertility along with strongly increased productivity in the SRB-breed, where Scandinavian bulls have dominated as sires of sons, and the negative trends in the SLB-breed where almost all sires of sons during a decade and a half have been imported from North America, illustrates the consequences of selection with and without consideration of daughter fertility in the sire-son path.

Daughter fertility will probably remain an important trait in the future, not only in countries with seasonal milk production connected to the pasture period. As dairy cattle breeding is quite international, calculation of international breeding values for daughter fertility in the same way as is practised by INTERBULL for production and conformation seems to be an important challenge for the future.