Search

Fertilization rates were similar for normal and subfertile mares, and much of the difference in fertility between normal and subfertile mares was due to embryonic loss. Fertilization rate estimates for mares ranged from 71 to 96 per cent. The incidence of embryonic loss detected by ultrasonography between Days 11 and 50 was approximately 9 per cent for normal mares, and the estimated incidence of embryonic loss before Day 14 was also 9 per cent. Therefore, the estimated incidence of embryonic loss in normal mares between fertilization and Day 50 is approximately 18 per cent (Fig. 1). In subfertile mares, the corresponding estimate for embryonic loss between fertilization and Day 50 is 80 per cent, with most embryonic losses occurring before Day 14 in subfertile mares (Fig. 1). The high rate of early embryonic loss in subfertile mares could be related to embryonic defects, oviductal environment, or uterine environment. Oviductal embryos from subfertile mares were less viable than embryos from normal mares; therefore, embryonic defects were important in early embryonic losses in subfertile mares. These defects might be inherent within the embryo or might arise from the early oviductal environment. The uterine environment of subfertile mares was adequate to support normal embryos in early gestation; however, the relationship between the uterine environment and the increased metabolic demands of the conceptus in the late embryonic or early fetal periods requires further study. The uterine environment is also altered in mares with endometritis; therefore, endometritis may also be an important factor in embryonic loss in some mares. Uterine-induced luteolysis, as well as the effect of the pathogen or the resulting inflammation, may lead to embryonic loss. An increased susceptibility of some subfertile mares to endometritis could result in embryonic loss secondary to a postcoital endometritis that persists until the embryo reaches the uterus at Days 5 or 6 postovulation. Although progesterone is critical to embryonic survival, the cause-and-effect relationship between progesterone and spontaneous embryonic loss remains unclear. Reduced progesterone concentrations could be related to endometritis, failure of maternal pregnancy recognition, or luteal insufficiency. Progesterone supplementation may be indicated for some mares, but the value of exogenous progesterone for prevention of spontaneous embryonic loss has not been critically tested. A number of other factors have been associated with embryonic loss in mares.(ABSTRACT TRUNCATED AT 400 WORDS)

Pregnancy rates at Days 2 and 14 postovulation were determined for 15 normal mares and 15 subfertile mares. Embryonic loss rates were estimated by the difference in the Day 2 and Day 14 pregnancy rates. Mares were artificially inseminated with the pooled ejaculates from three stallions, and the embryonic vesicle was detected with ultrasonography at Days 9, 10, 12 and 14. Mares were short-cycled with prostaglandin F(2) alpha (PGF(2alpha)) and rebred to the same stallions, and the Day 2 pregnancy rates were determined by recovery of cleaved ova (embryos) from the surgically excised oviducts. Significantly more (P0.01) normal versus subfertile mares were pregnant at Day 14 (12 15 vs 3 15 ). There was no significant difference in the Day 2 pregnancy rate for normal versus subfertile mares (10 14 vs 11 14 ). There were no significant differences (P0.5) in the mean number of blastomeres per embryo or in the mean diameter of embryos recovered at Day 2 from normal or subfertile mares. The estimated embryonic loss rate was significantly lower (P0.01) for normal verusus subfertile mares (0 10 vs 8 11 ). Fertilization rates were similar for normal and subfertile mares; however, subfertile mares had a higher embryonic loss rate prior to Day 14 postovulation.

Pregnancy rates, length of interovulatory intervals, and progesterone concentrations were examined in mares which had ultrasonically detected collections of fluid in the uterine lumen and in mares which lost the embryonic vesicle during Days 11 to 15 and did not become pseudopregnant. In mares with embryonic loss, the loss rate for mares with re-established pregnancies (9 18 ) was greater (P0.05) than the loss rate for all pregnancies (38 154 ), indicating repeatability. Pregnancy rates were higher (P0.01) in controls (100 177 ) than in mares with intrauterine fluid collections (2 34 ) or mares with embryonic loss (10 33 ), excluding the pregnancy associated with embryonic loss. The mean length (days) of the interovulatory interval was reduced (P0.05) in mares with intrauterine fluid collections (20.4 +/-0.9) and in mares with embryonic loss both for the intervals in which loss occurred (19.6 +/-0.7) and for intervals in which pregnancy was not detected (21.0 +/-1.0; controls, 23.5 +/-0.6). Mean progesterone concentration (ng/ml) on Day 7 was lower (P0.05) in mares with intrauterine fluid collections (8.8 +/-1.8) and in mares with embryonic loss (12.1 +/-1.1) than in pregnant controls (17.2 +/-0.9) and nonpregnant controls (17.5 +/-0.1). The embryonic loss seemed attributable to uterine-induced luteolysis in association with uterine inflammation, but the possibility of involvement of a primary luteal inadequacy or other factors in at least some of the mares was not eliminated.

The fertilization rate and incidence of embryonic loss in virgin Brahman crossbred heifers between d 3 and 35 postinsemination (d 0 = first day of estrus) were determined. One hundred fifty-five virgin heifers, maintained in a dry lot, received approximately 22 Mcal.head-1.d-1 and were allotted to one of three groups (d 3, 16 or 35) according to body weight at the time of estrous detection. Seven empty ruptured zona pellucidae (ERZP) were recovered from five heifers in the d 3 group. It was impossible to determine whether the ERZP originated from fertilized or unfertilized ova and whether they were ovulated at the estrus immediately preceding or during a previous cycle. Consequently, the fertilization rate (d 3) was calculated to be 80 or 93% depending upon whether the ERZP were included or deleted from the calculations, respectively. Because of the large number of ERZP recovered, a second study was conducted with an additional 21 virgin Brahman crossbred heifers from which ova were recovered on d 3 postinsemination. The results of the two studies were similar. The percentage of heifers with an embryo on d 16 was 78 (10% had degenerating embryos and 12% no embryos), and the percentage pregnant at d 35 was 72. The conclusions suggested from this study depend upon the classification of the ERZP. If the ERZP are designated as ova ovulated during a previous cycle or ova damage in the collection process and are deleted from the calculations, the fertilization rate is high (93%), and embryonic loss apparently occurs between d 3 and 35 (P less than .95). However, if the ERZP are classified as ova ovulated at the immediately preceding estrus, unavailable for further embryonic development, and are included in the calculations, the fertilization rate is comparatively low (80%). In the latter case, the primary loss of potential embryos occurs before d 3, and the loss after d 3 is negligible (P greater than .05).

The chromosome complement and anatomical development of 10–16-gday rabbit embryos were studied following delayed fertilization. Preimplantation loss was 40% in the experimental group compared with 17% in control animals. Also, a greater number of embryos retarded in growth or with structural anomalies was seen in the delayed-mated group. Cytogenetic analysis revealed a trisomy for a small metacentric chromosome in 104 control embryos. Only 1 triploid embryo was found in 84 implantation sites from rabbits that had been delayed mated. The incidence of triploidy postimplantation was much less than the 13% found previously in 6-day blastocysts. Increased embryonic loss at the time of or shortly after implantation may be responsible for the discrepancy.

Synopsis An experiment is reported in which hatching eggs were stored in a room at 12.7° ± 1.1° C. for up to 3 weeks after storage on the farm for 2 to 5 days. The eggs were stored in the room on keyes trays in part open‐sided wooden boxes or in sealed Cryovac bags or in sealed Cryovac bags flushed out with nitrogen before sealing. It was found that the normal decline in hatchability with storage time was reduced by storage in bags. There was no difference in hatchability between bags with or without nitrogen though there was an indication that the proportion of cull chicks at hatching was reduced by storage in bags flushed out with nitrogen for eggs stored more than 8 days. The effect on hatchability of the storage in bags is pinpointed as resulting from reduced embryonic loss between fertilisation and the 2nd day of incubation. This disagrees with a previous result and it is suggested that differences in genetic origin of eggs may account for differences in time of effect of storage treatment.

1. Twenty-four Merino ewes acclimatized to a hot environment were compared with twenty-four kept at normal temperatures, with respect to ovulation rate and embryonic loss by 25 days' pregnancy. Possible modifying effects of the three subsidiary factors, thyroxine level, vitamin A intake and progesterone level, were sought. The design of the experiment was a balanced factorial one.2. While the differences between the hot-room and yard groups as a whole were not significant, in the former there were fewer ovulations, a smaller proportion of the potential embryos actually began to develop and, of these, a larger proportion had died by 25 days' pregnancy. Consequently only thirteen hot-room ewes contained live embryos compared to twenty in the yard; the total numbers of live embryos were fourteen and twenty-seven, respectively.3. Of the subsidiary factors only thyroxine produced any marked effect. Both in the hot-room and in the yard the proportion of potential embryos developing and the proportion of actual embryos surviving at 25 days' pregnancy were greater where thyroxine injections were given. There appeared, however, to be no effect on ovulation rate. There was a significant increase in the proportion of hot-room ewes with live embryos due to thyroxine. Grouping hot-room and yard ewes together, thyroxine injections were associated with a significant increase in the proportion of potential embryos developing and in the proportion of actual embryos surviving; also, consequently, in the proportion of ewes with live embryos.4. The results are discussed in relation to published work on the effects of high temperatures on reproduction.

A single oral dose of mestranol (l7ct-ethynylestradiol 3-methyl ether) was administered to two groups of eight adult female beagle dogs each on Day 6 or 21 after first acceptance of the male. Based on corpora lutea counts, embryonic loss was 95.5 and 67.3%, respectively, for the two treatment groups, and only 34.5% for the control. In a second study a group of 12 beagles were used to determine the effect of mestranol on ova transport. Nine bitches were dosed on Day 6 and the reproductive tracts were removed between 1 and 5 days later and three controls were examined on Day 11. Thirty-nine of a possible 56 ova were flushed from the tracts of the treated females and all but one were from the oviducts. Eighteen (46%) of the ova were degenerating. The controls produced 17 normal zygotes out of a possible 24, and all were flushed from the uteri. Mestranol apparently did not accelerate ova transport, and the possibility of both a direct effect on the ova and an indirect effect via a delay in ova transport is discussed.

Normal pregnant guinea pigs were killed on gestation days 12–20 and examined for implantation sites and resorbing embryos. Their ovaries were removed and serially sectioned, and the functioning corpora lutea counted for comparison with the number of conceptuses. It was found that a single corpus luteum was associated with two implantation sites with a frequency of 9.55%. Embryonic loss prior to implantation was 12.6% and the value for resorption following implantation varied from 5.8 to 7.2% depending on the time of killing and the method employed for determining resportion. A count of functioning corpora lutea facilitated the determination of embryonic loss prior to implantation, but offered little advantage over simpler techniques for determining resorption after implantation.

Summary A total of 837 services were analyzed for the effects of sire and system of mating on estimated embryonic mortality by 150 days of gestation. The sire had a highly significant effect on the rate of embryonic death in the females to which he was bred. Outbred embryos had an estimated death rate of 19.8% by 150 days as compared-to 22.3% for inbred embryos. Outbred dams had 19.2% embryonic death by 150 days, whereas embryonic mortality in inbred dams was 28.4%. Although the difference between outbreds and inbreds was not statistically significant in either case, the trend was for inbreeding of either embryos or dams, particularly the latter, to increase embryonic mortality.

1. Embryonic mortality and reproductive performance were studied in 150 mature Merino ewes, groups of which were restricted to approximately 15% of their maintenance requirement from days 0to7 (group 1), 6to13 (group 3) and 13to20 (group 4) respectively. Groups 2and5 were controls. Ovulation point counts, returns to service, pregnancy test by laparotomy at day 40 and lambing performance were used to assess and apportion the embryonic loss.

To elucidate the mechanism of and to study the method preventing the high embryonic loss in swine, the authors are investigating to obtain some basic informations on this aspect using rats.The embryonic loss in swine is estimated to be 30 to 50% on full term of pregnancy1). Approximately 70% of the total embryonic loss occurs prior to day 25 of pregnancy2). High energy intake in swine has usually led to an increased ovulation rate and a higher percentage of embryonic loss3, 4). Approximately 70% of the total loss (32%) in gilts fed the high energy diet occurred prior to day 25 of pregnancy5). The authors found that most of the total embryonic loss in the rats fed the high energy diet occurred in the first third of pregnancy6) and the loss during this stage was affected by dietary calorie-protein ratios7).This study was designed to determine the strain difference in the embryonic viability in the rats fed the high energy diet.

Methyl chloride (MeCl) is a testicular and epididymal toxicant in the Fischer 344 rat that induces significant decreases in sperm quality in males and significant increases in preimplantation embryonic loss in females mated to exposed males. This study examined the possibility that these losses are due to failure of fertilization and not to preimplantation embryonic death. Groups of males either were exposed to 1000 or 3000 ppm MeCl 6 hr/day for 5 days or received a single ip injection of 0.2 mg/kg triethylenemelamine (TEM) on Day 5 and were bred to two females weekly for up to 8 weeks. Females were killed 12 hr postmating; embryos and ova were isolated and scored as fertilized or unfertilized. Nearly 90% (602/683) of the normal ova recovered from females bred to control males were fertilized, and fertilization rates in the 1000-ppm MeCl and TEM groups were not significantly depressed from that level. The percentage of fertilized ova in the 3000-ppm MeCl group was significantly decreased, ranging from 3 to 72% over the 8 weeks. The frequency of unfertilized ova in this group always equaled or exceeded the frequency of preimplantation loss recorded in an earlier dominant lethal assay. In contrast, only a fraction of the preimplantation losses caused by the genotoxicant TEM could be accounted for by failure of fertilization. After culture of fertilized ova for 24 hr, only the TEM group exhibited a significant decrease in the number of fertilized ova which cleaved (85 vs 96% in the control). We conclude that the preimplantation loss caused by MeCl exposure is due to failure of fertilization and not to a genotoxic effect of MeCl. The methods used here permit a more accurate assessment of the reproductive toxicity of a chemical by discriminating between its cytotoxic and genotoxic effects.

Numerous environmental risks damage the reproductive system, causing various negative outcomes. Research in this area has emphasized the study of congenital malformations. Other negative outcomes, however, such as embryonic loss, pre- or post-implantation, might be considered significant, because of their severity from the biological point of view and their more frequent occurrence. It is estimated, for example, that 60–70% of all conceptions terminate in spotaneous abortion (Roberts and Lowe, 1975).

The objective of this study was to investigate an hypothesis that chromosome anomalies are an important cause of prenatal loss in the mare. An attempt was made to analyse, cytogenetically, a series of 26 equine abortuses. Cell cultures were prepared from a range of tissues, but failed to grow, and chromosome analysis was therefore not possible for any of these specimens. Consequently, a study was made of the metaphase chromosomes prepared from 22 equine embryos after their surgical removal from mares' uteri. The karyotypes prepared for each specimen were normal. The current findings are discussed in the light of similar studies of induced abortions in women and embryonic loss in animals, and the potential for further investigation in this field is emphasized.

The origin of the early embryonic loss that occurs during the first 4 days after mating in Californian doe-rabbits has been studied at 24 h or 60 h post coitum (p.c.). For comparison, 11 New Zealand doe-rabbits were slaughtered at the same times. The development stage of the embryo was determined after examination in toto at 24 h or by counting the morula nuclei at 60 h on histological sections. At 24 h p.c., 5.3% of the New Zealand eggs had not divided vs 37% of the Californian ones; this difference was highly significant. At 60 h p.c. frequency histograms of the number of morula nuclei showed that development was more variable in Californian does than in the New Zealand ones. However, the mean number of morula nuclei of each genotype was not significantly different. The higher proportion of Californian eggs (compared to New Zealand eggs) that were unfertilized or showed retarded segmentation is the main cause of the unexpectedly lower prolificacy of Californian rabbits.

In science, we are called upon to recognize and assimilate “the facts” as interpretations, based on data samples that we trust statistically to represent a larger body of conditions or responses, but that are not practically available for examination as a whole. Unlike the performing arts, we cannot produce useful results through creative expression alone; rather, imaginative ideas must be tested with minimal bias to ascertain elements of truth, even if they prove only partially valid. With maturity, these truths are seen to interconnect, producing the dogma of current understanding. In this context, the evidence available to offer secure interpretation on the frequency and causes of early embryonic loss among primates is severely limited. Frequently, requisite data are either incomplete or nonexistent even for humans and the more commonly studied laboratory primates. Despite such restrictions, I hope to present a few persuasive interpretations on the incidence and origins of early embryonic loss in both human and nonhuman primates, with emphasis on their relationship to in vitro fertilization and embryo transfer (IVF/ET).

The prolificacy of two rabbit strains (Californian and New Zealand) has been studied in parallel with ovulation rate and embryonic development. The number of mean ovulations was 11.6 in Californians and only 9.6 in New Zealand. In spite of the higher rate in the former, the number of young born was about 8 in both strains. Californian and New Zealand blastocysts were transferred at 96 h post-coitum by injecting them into the uterus of recipient does of the same strain or of the other strain (crossed transfer). The percentage of unfertilized and delayed eggs was significantly higher (20% vs 12%) in Californians. The size of blastocysts recovered at 96 h p.c. was significantly higher in New Zealand than in Californian does (370 vs 319 micron). The difference was due to the larger proportion of small Californian blastocysts and of large New Zealand blastocysts. The larger blastocysts were placed in the right horn of recipients and the small ones in the left horn. Number of foetuses at D27 showed that the survival percentage was always higher when large blastocysts were transferred. However the survival difference with small transferred blastocysts was not significant. When transfer was within the same strain or crossed the results were not different. Since the New Zealand blastocysts were larger at D4 p.c. (one day after they arrived in the uterus), it could not be determined if this improved development was due to the conditions in the fallopian tubes or to the quality of the uterine secretion.(ABSTRACT TRUNCATED AT 250 WORDS)

The estimated embryonic loss rate between Days 4 and 14 after ovulation for young, normal mares (9%) was significantly lower (P less than 0.01) than the estimated embryonic loss rate for aged subfertile mares (62%). Fertilization rates, which were based on the recovery of embryos at Day 4 after ovulation, were 96% and 81% (P less than 0.1) for normal and subfertile mares, respectively. Day-4 embryos were collected from the oviducts of normal and subfertile donors mares. These embryos were transferred to the uteri of synchronized, normal recipient mares to test the hypothesis that the high incidence of embryonic loss in subfertile mares was related to embryonic defects. The hypothesis was supported because embryo survival rates were significantly higher (P less than 0.05) for Day-4 embryos from normal compared to subfertile mares. These defects may have been intrinsic to the embryo or might have arisen due to the influence of the oviducal environment before Day 4 after ovulation.

Twenty-one pregnant pony mares were assigned to one of the following groups: 1) controls, 2) ovariectomy at Day 12, 3) ovariectomy at Day 12 plus daily progesterone treatment on Days 12 to 40, 4) PGF(2alpha) on Day 12, 5) PGF(2alpha) on Day 21, and 6) PGF(2alpha) on Day 30. Based on daily examinations by ultrasound, the embryonic vesicle was maintained to Day 40 in all control mares and in mares that were ovariectomized on Day 12 and given progesterone. The embryonic vesicle was lost in all mares of the other four groups. Administration of progesterone prevented the embryonic loss associated with ovariectomy at Day 12, indicating that progesterone may be the only ovarian substance required for survival of the early embryo. The mean number of days to embryonic loss was greater for mares treated with PGF(2alpha) on Day 12 (6.8 days) than for mares ovariectomized on Day 12 (3.0 days). In the PGF(2alpha)-treated group, the vesicles did not become fixed at the expected time (Day 15), and mobility continued until the day of loss. In the mares treated with PGF(2alpha) on Day 21 and in one of the mares treated on Day 30, the vesicle was lost within one to three days without prior indication. Loss may have occurred by expulsion through the cervix, since the cervix was patent on the day of loss in these mares and in the mares ovariectomized or treated with PGF(2alpha) on Day 12. In the remaining mares treated on Day 30, the intact embryonic vesicle was dislodged on Day 31 or 32. The dislodged vesicle was mobile within the uterus and was frequently found in the uterine body. The fluid volume of the dislodged vesicle gradually decreased, and the fluid was no longer detected by Day 38 to 42. Some of the placental fluids may have been eliminated by resorption since the cervix remained closed while the fluid volume decreased.

The association between endometriosis and infertility is widely recognized, but several studies demonstrate that endometriosis also affects pregnancy rates in different ways other than anatomical distortion of internal reproductive organs; however the evoking mechanisms are complex. In mild to advanced stages of disease, peritoneal adhesions, tubal obstruction, destruction of ovarian tissue or distortion of the uterine wall are evident originating factors of infertility or miscarriage.

White Leghorn pullets were fed standard diets supplemented with 0.0%, 0.1%, 0.5%, and 1.0% levels of dietary caffeine. The two highest levels of caffeine caused the production of soft-shelled eggs and a cessation of egg production within 3 days. Although 0.1% caffeine did not effect fertility, it significantly increased embryonic mortality over a 14-day period following a single insemination.

In our previous experiment1), the authors found that high embryonic loss rate in rats fed the high energy diet could be reduced by decreasing of calorie-protein ratio (CPR).On the other hand, DEAN et al.2), GOSSETT and SORENSEN3) reported that negative correlation was observed between backfat thickness and embryonic survival rate in gilts.This experiment was designed to determine a favorable CPR for minimizing the embryonic loss rate in female rats fed the high energy diet and to investigate the relationship between energy or nitrogen retention and reproductive performance in rats.

White Leghorn pullets were fed standard laying rations supplemented with 0.0%, 0.05%, and 0.1% levels of dietary caffeine. There was no statistical difference in fertility among the three groups. Embryonic loss in eggs from pullets fed 0.05% caffeine was between values previously reported for 0.0% and 0.1% dietary caffeine. This suggests that embryonic mortality increases as levels of caffeine are increased in the diet up to 0.1%.

Five theoretical models of embryonic loss based on certain biological concepts concerning the ewe or the embryo are developed. It is shown that these models lead to only three distinguishable distributions of embryonic survivals in single and twin ovulating ewes, emphasizing the difficulties in interpreting such data. Two published sets of data concerning experiments involving embryonic mortality are used to illustrate the use of distribution data in interpreting the results of such experiments. The importance of isolating fertilization failure from embryonic mortality is demonstrated and the difficulties of determining causal relationships are discussed.

A series of studies was conducted in 1968 and 1969 to compare fertilization rates, return rate, and sperm distribution in the genital tract in Bungaree and Peppin Merino ewes. Differences in lambing rates of the two ewe strains were demonstrated, but there was no major difference in fertilization rate in either year. A significantly higher proportion of Bungaree ewes had less than 1000 sperm in the fallopian tubes 24 hours after service, but there were no significant differences in any other character examined. Contrary to an earlier report, fertilization failure is unlikely to have been a major cause of a difference in the fertility of the two strains in the two years, and embryonic loss is likely to have made a significant contribution to the difference in their fertility.

Ewes which fail to lamb can be classified as failing to conceive (ENC) or conceiving but failing to carry at least one embryo to full term (ENE). Estimates of ENE have been made indirectly from mating and lambing records involving 2509 observations over 3 years in a flock of medium Peppin Merinos. Estimates of ENE and of individual embryo loss have been made directly from autopsies of 419 ewes in the same Peppin flock and in one of mixed strains. The indirect estimate of ENE (ewes served and expected to have conceived but failing to lamb, as a percentage of ewes joined) averaged 4.3% for ewes aged 1½–9½ years at joining. Two measures of early embryonic loss were obtained from the autopsies of ewes expected to have been pregnant for approximately 28 days: (i) Ewes predicted not to lamb, as a percentage of ewes expected to have conceived, gave an estimate of 2.8–3.9 % at 1½ years, and 10.1–12.0% at 8½–11½ years. (ii) By relating the number of active corpora lutea to the number of healthy embryos, the estimates of loss obtained were 4.7–5.8% for maidens and 18.6–20.6% for aged ewes. From approximately 28 days to term no significant foetal loss from resorption or abortion could be demonstrated by comparing autopsy predictions with observed performance in ewes proceeding to term, though a very small loss of one of a pair of twins may have occurred in ewes aged 1½ years. Ewes with two active corpora lutea suffered a significantly greater loss of embryos than did ewes with one. With the corpus luteum as the unit, the loss in maiden ewes was 20.0% with two corpora lutea and 2.9–4.1% with one, the corresponding figures for aged ewes being 34.7–36.1 and 9.0–11.5%. These differences are not evident with the ewe as the unit, as a ewe with two corpora lutea which had lost one embryo would remain in the group predicted to lamb. Estimates of ENE from autopsy were made on ewes selected from the complete flock because they had not returned to service. Estimates for the same ages on the complete flock were 3.6–4.5% for maiden and 8.7–10.2% for aged ewes. Calculations of ENC have been based on the estimate of ENE made here and on previously reported figures of the numbers of dry ewes per ewe joined and present at lambing. Ranges of ENC were 13.6–14.5% at 2 years, and 3.3–4.8% respectively. No identical twins were observed at autopsy, and there is no reason to believe that any would have been present at any earlier stage of pregnancy.