Cow Embryos Research Articles

Yield of embryos in PMSG-superovulated cows treated with anti-PMSG six or 18 hours after the peak of luteinising hormone.

One hundred and forty-six Dutch cross Friesian cows were selected from a local slaughterhouse and synchronised with norgestomet. The 134 cows with a normal progesterone pattern after the removal of the norgestomet implant were treated intramuscularly with 3000 iu pregnant mare's serum gonadotrophin (PMSG) on day 10 followed by 22.5 mg prostaglandin 48 hours later. Blood samples were collected daily and at hourly intervals from 30 to 54 hours after the prostaglandin. The 113 cows with a pre-ovulatory peak of luteinising hormone (LH) were divided into three groups: 37 control cows (group 1) received a placebo six hours after the LH peak; 42 cows (group 2) received anti-PMSG six hours after the LH peak and 34 cows (group 3) received anti-PMSG 18 hours after the LH peak. All the cows were inseminated 10 hours after the LH peak. Six or seven days after insemination the cows were slaughtered and the embryos were evaluated after flushing the ovaries, and the numbers of corpora lutea, cysts and follicles on the donor ovaries were counted. Treatment with anti-PMSG had no significant effect on the numbers of corpora lutea or the numbers of embryos compared with the control group. The mean (+/- sem) numbers of corpora lutea were 14.7 +/- 1.4, 16.3 +/- 1.4 and 16.6 +/- 1.4 for groups 1, 2 and 3, respectively. The numbers of transferable embryos were 3.5 +/- 0.6, 4.1 +/- 0.7 and 5.0 +/- 0.7 for groups 1, 2 and 3, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Sexing of mouse and cow embryos with freemartin serum

Gonadal morphology and steroidogenic patterns of female bovine fetuses in twin pregnancies were compared with those of single male and female fetuses to determine the morphogenetic and functional correlates of ovarian alteration in freemartins. In male fetuses, seminiferous cords were detected by 40 days and Leydig cells by 45 days. The ovaries of 50-day fetuses showed clusters of steroidogenic cells forming medullary cords, which appeared to be the precursors of the follicular cells detected in the ovarian cortex of fetuses over 90 days old. Among the steroids produced by the testis of single male fetuses, testosterone was the most prominent, androstenedione was intermediate and estrone and estradiol were in trace amounts, whereas ovaries secreted low levels of testosterone and moderate amounts of androstenedione, estradiol and estrone. Ovarian estradiol secretion in the fetus peaked during 50 to 70 days of development while estrone was produced at elevated levels during 75 to 95 days. During the peak period for estradiol, steroidogenesis in the ovary appeared to follow the androstenedione-to-estrone-to-estradiol pathway. From the relative concentrations of steroids detected, it would appear that steroid synthesis declines in fetal ovaries, probably due to the inhibition of 17β-hydroxysteroid dehydrogenase, prior to that of aromatase. In contrast, in female fetuses of heterosexual twins (freemartins) an interruption of ovarian differentiation was evident as early as 55 days whereas testicular cords and cells resembling Leydig cells were apparent only in fetuses over 70 days old. The morphologic evidence of ovarian alteration was preceded by functional alterations in that the levels of estrogen were significantly reduced and androstenedione was elevated in freemartins as early as 40 days of fetal development. These initial changes were followed by elevated levels of testosterone and androstenedione after 75 days in virilized freemartin gonads coincident with the induction of Leydig cells in the ovary under the influence of the male co-twin. The higher levels of androstenedione secreted by freemartin gonads in early stages seem to further confirm that the androstenedione-to-testosterone pathway may be inherently inactive in normal bovine ovaries.

Observation by ultrasonography of embryonic loss following the transfer of two or three embryos in beef cows

Ultrasonographic observations were carried out at 10-day intervals from 27 to 107 days of gestation, to monitor embryonic losses in Japanese Black cows which received two or three embryos that had been transferred nonsurgically on Day 7 (Day 0 = estrus). Group I cows (n=60) received two embryos, one in each uterine horn. Group II cows (n=31) received three embryos, two embryos in the uterine horn ipsilateral to the corpus luteum and one embryo in the uterine horn contralateral to the corpus luteum. The Group II pregnancy rate was maintained above 83.9% until Day 37, but it decreased significantly (P<0.05) to 58.1% by Day 107. The pregnancy rates on Day 27 and Day 37 in the three-embryo group (Group II) were significantly (P<0.05) higher than those of the two-embryo group (Group I). The embryonic survival rates of both groups showed almost the same values at Days 27 to 107. The triplet pregnancy rate in pregnant cows was 26.9% on Day 37, but fell to 11.1% on Day 107. The twinning rate in both groups also decreased with the advance of the gestation period. Most of the embryonic losses (46%) in triplet and twin pregnancies in both groups occurred between Day 38 and Day 57. Thirty-four and 22 live healthy calves were born, respectively, to 30 cows in the two-embryo group (Group I) and to 18 cows in the three-embryo group (Group II). There were no statistically significant differences between the two groups in the calf crop or in the number of embryos.

Immunoelectron microscopic localization of small nuclear ribonucleoproteins during bovine early embryogenesis

Small nuclear ribonucleoproteins (snRNPs) were localized using human autoimmune or monoclonal anti-snRNP antibodies and ultrastructural immunocytochemistry in early preimplantation bovine embryos before and at transcription onset. Bovine cleavage stages up to 16-cell embryos were obtained either by culture of in vitro-fertilized ovarian oocytes or by isolation at slaughter of embryos fertilized and developed in vivo. Before transcription onset, up to the four-cell stage, diffuse labeling of nucleoplasm was detected, whereas cytoplasm labeling remained low. At the transcription onset, labeling of all eight-cell embryo nuclei was markedly concentrated at the borderline of already formed, condensed chromatin aggregates, where it was associated mainly with perichromatin fibrils. The condensed chromatin blocks revealed by several different staining methods were more prominent than is the case in most somatic cells. The degree of chromatin condensation and the pattern of its distribution differed between in vivo- and in vitro-produced eight-cell embryos: the in vitro embryos showed a higher degree of chromatin condensation and a peripheral distribution of chromatin blocks. A relation of this observation to the developmental potential of cow embryos is suggested. In two- and four-cell embryos, intensive labeling was seen in interchromatin granules, which, in their turn, were often seen in close proximity to the nucleolus precursor bodies, or in the four-cell stage were interconnected to them. No labeling was ever seen, using antibodies specific for the snRNP Sm antigen, in nucleolar precursor bodies during embryonic nucleologenesis nor in the resulting nucleoli. There was some incidental labeling of the large central vacuole appearing at the beginning of the nucleolus precursor body transformation, testifying the nucleoplasmic origin of this entity.

Numbers of small follicles in ovaries affect superovulation response in cattle

The global demand for in vitro–produced (IVP) embryos of determined sex has greatly increased over the last decade. Efficient protocols for the direct transfer of IVP embryos are lacking. This study aimed to compare the pregnancy rates for fresh, vitrified, or frozen/directly transferred IVP dairy cow embryos. Oocytes (n = 3171) recovered by ovum pickup (n = 112) from Girolando (Holstein–Gir) females (n = 36) were selected and submitted to IVM for 24 hours at 38.5 °C with 5% CO2 in air with saturated humidity. In vitro fertilization was performed with the thawed, sexed semen from 5 Holstein bulls. After IVF, presumptive zygotes were denuded and cultured for 7 days under the same IVM and IVF conditions of temperature and humidity, except with 5% CO2 and 5% O2. Grade I blastocysts were randomly assigned for either the transferred fresh, vitrified/thawing, or frozen/directly embryo transfer into previously synchronized recipient females. Conception rates were analyzed by binomial logistic regression, and a probability level of P < 0.05 was considered significant. The conception rates were 51.35 ± 1.87% (133/259) for the fresh embryos, 35.89 ± 3.87% (84/234) for the vitrified embryos, and 40.19 ± 4.65% (125/311) for the frozen directly transferred embryos. These data demonstrate that IVP embryos with sexed semen could be directly transferred into recipient cows with similar conception rates to vitrified embryos. The comparison found that the use of frozen embryos in direct transfer provides easier logistics and a more practical approach for the transfer of IVP embryos on dairy farms.

Ultrastructural localization of silver‐staining nuclear proteins at the onset of transcription in early bovine embryos

Argyrophilic nuclear proteins, known to be functionally associated with ribosomal genes, were localized, in four-, eight-, and 16-cell bovine embryo blastomere nuclei using two different silver-staining procedures. Within the eight-cell cleavage stage by the process of embryonal nucleologenesis in the cow embryo the full-capacity ribosome-producing machinery is established. In the four-cell embryo, many patches and islands of argyrophilic (Ag+) material were detected in the nucleoplasm. The nucleolus-precursor bodies (NPBs), composed uniformly of a homogeneous compact mass, were completely devoid of any silver staining. On the other hand, clear-cut localization of argyrophilic proteins was detected during the eight-cell stage either inside the transforming NPBs or in the close vicinity, or in the already differentiated nucleolus. In compact, nonvacuolated NPB, an intensive Ag+ area was detected, in the form of a lenticle, at the periphery of the NPB. During and following vacuolation of the NPB, no Ag+ was detected inside these vacuoles. It was seen, however, in the dense fibrillar nucleolar component surrounding the smaller vacuoles formed at the time of the establishment of nucleolar structure. Ag+ areas were seen repeatedly in the vicinity of NPBs, probably a part of the nucleolus-associated chromatin or, alternatively, representing the extranucleolar bodies. In blastomere nuclei of 16-cell embryos, already possessing reticulated nucleoli known from intensively synthesizing somatic cells, the silver-staining pattern corresponded to the usual situation in differentiated cells: slight staining of fibrillar centers, heavy labelling in the dense fibrillar component, and absence of silver deposits in the granular component.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of oxygen concentration on in-vitro development of preimplantation sheep and cattle embryos

Two-cell sheep embryos and 2-4-cell and 8-cell cow embryos were cultured for 5 days in stoppered test-tubes in Synthetic Oviduct Fluid supplemented with 32 mg BSA/ml. The medium had been previously equilibrated with one of the following O2 concentrations (sheep: 0, 2, 4, 6, 8, 10, 12, 17, 20%; cow: 0, 4, 8, 12, 17, 20%). At the end of culture embryos were examined for morphology and stained to assess numbers of nuclei. Mean (+/- s.e.m.) nuclei/embryo was highest at 8% O2 for sheep embryos (23.6 +/- 3.1), 4% for 2-4-cell cow embryos (23.2 +/- 6.1) and 8% for 8-cell cow embryos (29.6 +/- 5.2). The minimum number of nuclei/embryo occurred at 20% O2 in each case (10.3 +/- 0.9, 10.3 +/- 2.7, 14.5 +/- 2.4, respectively) with similar values also recorded at 0% O2 (10.8 +/- 1.9, 16.5 +/- 6.0, 14.6 +/- 2.4, respectively). Analysis of the proportion of embryos reaching at least the morula stage demonstrated a significant quadratic component for the different oxygen concentrations for sheep (P less than 0.01) and cow (P less than 0.05) embryos. A number of sheep and cow embryos showed abnormalities, suggesting that the culture conditions require further refinement. The results confirm that, under lowered oxygen levels, development of sheep and cattle embryos can occur through the 8- to 16-cell block in a simple defined medium without somatic cell support.

Development and survival after transfer of cow embryos cultured from 1-2-cells to morulae or blastocysts in rabbit oviducts or in a simple medium with bovine oviduct epithelial cells

This study compares development of bovine 1-2-cell embryos in bovine oviduct epithelial cell co-culture (Group EC) with a glucose- and serum-free simple medium (CZB), or after surgical transfer to ligated oviducts of rabbits (Group RO). Embryos were surgically collected from superovulated donor cows 40-48 h after the beginning of oestrus and randomly distributed between the two groups. Embryos were cultured or incubated for 5 days. In Exp. 1, embryo quality scores and total numbers of cells in the two groups were compared. In Exp. 2, pairs of similarly treated morulae were transferred to each of 10 or 12 recipients in the Groups RO and EC, respectively. Total cell counts per embryo in both groups averaged 52 (P greater than 0.05), and the in-vitro culture system was equivalent to the rabbit oviducts in promoting embryo development for all characteristics measured. Embryo survival, as determined by ultrasound between Days 39 and 43 after oestrus, in 13 ideal recipients was 57% for embryos in Group EC and 58% for embryos Group RO. None of the 9 less desirable recipients was pregnant for either group. These results establish that cattle zygotes can develop to morulae in culture with bovine oviduct epithelial cells in a simple medium and can produce normal pregnancy rates.

Effect of Variability in Response to Superovulation on Donor Cow Selection Differentials in Nucleus Breeding Schemes

The effectiveness of nucleus breeding schemes based on multiple ovulation and embryo transfer in practice is highly dependent upon the achieved reproductive rate. This study quantifies the effects of donor cow variability in response to superovulation (random and phenotypically correlated with milk yield) and failure rates (proportion of selected cows producing no transferable embryos) on the mean and variance of realized female selection differentials as affected by herd size, average embryo yield, and the coefficient of variation for embryo yield. Results show that differences in failure rates (0 to 40%) reduced realized female selection differentials up to 64%. The existence of a negative phenotypic correlation (–.3) between donor cow embryo yield and milk yield caused smaller reductions (up to 19%). Variability in realized female selection differentials is largest for small nucleus units and should not be neglected when planning the establishment of a nucleus breeding unit. The use of multiple flushes rather than a single flush per cow to produce the same average number of transferable embryos can reduce the variability in realized selection differentials but increases generation interval. Variability in response to superovulation will also affect inbreeding rates and the average accuracy of selection. Simulation studies are needed to quantify the total effect of variability in response to superovulation on the genetic responses possible in nucleus breeding schemes relying on multiple ovulation and embryo transfer.

Survival of DNA-injected cow embryos temporarily cultured in rabbit oviducts

Holstein or Angus cows were superovulated, inseminated with fresh bull semen, and necropsied about 12 h after estimated time of ovulation. Ova were centrifuged at 15,600 G for 3 to 8 min to reveal pronuclei. In Experiment 1, pronuclear bovine embryos were transferred to ligated or unligated oviducts of 1-d pseudopregnant rabbits for 7 d; 30 of 32 embryos were recovered from ligated oviducts but only 2 of 26 from oviducts and uterine horns of unligated oviducts. In Experiment 2, a Rous sarcoma virus-chloramphenicol acetyl transferase fusion gene was injected into one pronucleus of about half of 404 fertilized bovine ova, using a micromanipulator and interference contrast optics. Injected and noninjected embryos were then transferred to opposite ligated rabbit oviducts. Embryos were recovered after 7, 8 or 9 d. Of 120 centrifuged but ininjected embryos recovered from rabbit oviducts, 66 (55%) were in the morula to hatching blastocyst stage of development. Of 105 embryos centrifuged and injected with foreign DNA, 55 (52%) were in the morula to hatching blastocyst stage. In Experiment 3, centrifuged bovine embryos, noninjected or DNA-injected, were cultured in rabbit oviducts for 7 d then transferred nonsurgically to the uterus of recipient cows. Embryos were also flushed from superovulated cows 8 d after estrus and transferred directly to recipient cows. After 7 d, the uterus of recipient cows was flushed nonsurgically to recover embryos. The proportion of transferred embryos recovered with normally elongated trophoblastic membranes and the proportion of recipient cows with developing embryos were 14 of 25 DNA-injected embryos, 5 of 8 cows; 6 of 15 centrifuged but noninjected embryos, 4 of 6 cows; and 11 of 29 embryos transferred directly, 5 of 8 cows. Results indicate that bovine embryos can be cultured in rabbit oviducts and survive after transfer to cow uteri and that injection of foreign DNA may not increase embryonic loss within the first 2 wk after injection.

Cell loss in bisected mouse, sheep and cow embryos

The purpose of this experiment was to determine the loss of cells during the splitting of embryos. The experiment was carried out on early, middle and late stage mouse blastocysts as well as on late stage sheep and cow blastocysts. The embryos were bisected with a metal mioroscalpel. Manipulation of the embryos was done in phosphate buffered saline (PBS) solution with 20% fetal calf serum (FCS) at +37°C or in PBS solution free of Ca ++ and Mg ++ ions. To determine the number of cells present, intact embryos and demi-embryos were placed in acridine orange solution and examined under a fluorescent microscope. The average number of cells in early, middle and late stage mouse blastocysts was 39.4, 54.1 and 71.9, respectively. The loss of cells caused by splitting was 6.4, 7.6 and 9.9, respectively. The average number of cells in late stage sheep blastocysts was 115.3 and in late stage cow blastocysts it was 138.5. In sheep and cows, the loss of cells due to splitting the embryos in a PBS solution free of Ca ++ and Mg ++ ions did not differ from that observed in a PBS solution with 20% FCS. Similarly, bisection carried out at +37°C did not result in lower cell losses, than that done at room temperature.

Cloning and expression of cDNA encoding ovine trophoblastin: its identity with a class-II alpha interferon

The cDNAs encoding ovine trophoblastin (oTP) were isolated from an ovine embryo cDNA λgt11 library by screening with a synthetic 29-mer oligodeoxynucleotide corresponding to amino acid (aa) residues 34 to 43 of oTP. The cDNA contained an open reading frame of 595 bp and the deduced amino acid sequence indicates a protein precursor of 195 aa. Nucleotide and amino acid sequence comparisons establish that oTP shares extensive homology with a-interferon (IFN-α) but is more closely related to the IFN-αsII subfamily. When the oTP cDNA was cloned into an eukaryotic expression vector and transfected in monkey COS cells, a high level of antiviral activity was detected. RNA blot analyses of total RNA reveal that the oTP-coding gene is expressed during a relatively short period (eleven to 21 days). The abundant expression of oTP mRNA corresponds closely to the time at which the embryo acts to extend luteal lifespan. RNAs homologous to oTP were also detected in goat and cow embryos at equivalent periods of their development, but not in the pig.

Cloning of sheep and cow embryos.

The results of experiments aimed at cloning of sheep and cattle embryos are described. Two experimental approaches were used to study the developmental potential of blastomeres from sheep and cow embryos: (i) blastomere separation followed by culture and (ii) fusion of isolated blastomeres with enucleated eggs followed by culture. Both approaches allow embryos to be cloned, but whereas blastomere separation allows only a relatively small number of genetically identical animals to be produced, nuclear transplantation will probably open the way for large-scale cloning of livestock.

Localization of replicated DNA‐containing sites in preimplantation bovine embryo in relation to the onset of RNA synthesis

The distribution of embryonic DNA in nuclei of blastomeres of early-preimplantation cow embryos was studied by autoradiography. Two-, 4-, 8-, and 16-cell embryos were cultured for 18-20 h in the presence of [methyl-3H]thymidine to ensure that all replicated embryonic DNA was labeled. In nuclei of blastomeres before the onset of transcription, taking place in the progressed 8-cell cleavage division of the cow embryo, labeled DNA was distributed excentrically at the nuclear periphery. After transcription started, the distribution of the labeled DNA was uniform throughout the nucleus. The progressive association of DNA with the nucleolus-precursor body (NPB) also correlated with the expected onset of rRNA synthesis. The penetration of the labeled DNA into NPB was seen to start contemporaneously with the formation of a big central vacuole in NPB, but remained restricted, at this stage, to a few points of contact with the adjacent nucleolus-associated chromatin. No labeled DNA was detected inside this big vacuole, still showing a uniform nucleoplasmic texture. The inside of the nucleolus was penetrated by DNA in the next step of nucleologenesis when secondary small vacuoles were formed and when the first signs of nucleolar transcription were detected. These small vacuoles contained a coarse fibrillar component which was a frequent site of labeled DNA detection. This study of shifts of DNA-containing sites during transition from maternal to embryonic genome expression correlates closely with previous evidence obtained by a study of genome reactivation which was situated in the late 8-cell cleavage division of cattle early embryogenesis.(ABSTRACT TRUNCATED AT 250 WORDS)

Simultaneous measurement of the metabolism of two energy substrates in individual horse and cow embryos

The influence of different maternal plasma progesterone concentrations on embryonic glucose metabolism was studied. Uterine flushes were obtained after treating ovariectomized mares (n = 3) with 0 (control), 100 or 200 mg progesterone daily for 7 d. A group of progesterone-induced proteins (PIP) of Mr~20,000 were identified in flushes from progesterone treatments by SDS-PAGE but were not observed in control flushes. Progesterone-induced proteins were removed from half the pooled flush in each treatment group by Sepharose blue CL-6B. In a 3 × 2 factorial (progesterone treatments, progesterone-induced proteins) experiment, 6 groups of Day 7 equine embryos (n = 6 per group) were incubated in culture media (MEM:DPBS; 1:3) containing radioactively-labeled glucose. Contributions of the Embden-Meyerhof pathway (EMP) and the pentose-phosphate pathway (PPP) to the total metabolism of glucose in early equine embryos were assessed separately. In the 200 mg progesterone treatment group, the presence of progesterone-induced proteins in the culture medium resulted in a 4-to 5-fold increase in the activities of the Embden-Meyerhof pathway and the pentose-phosphate pathway. These results lead to the following conclusions: 1)Addition of progesterone-induced uterine proteins from mares with high levels of circulating progesterone enhance the metabolic activities of the Embden-Meyerhof pathway and the pentose-phosphate pathway in Day 7.5 equine embryos in culture. 2)Uterine secretion of progesterone-induced proteins which is quantitatively and/or qualitatively adequate to modify embryonic glucose metabolism in vitro is dependent on a minimal concentration of maternal plasma progesterone.

Nucleologenesis and the onset of transcription in the eight-cell bovine embryo: fine-structural autoradiographic study.

Eight-cell cow embryos were isolated and cultured in vitro in a medium enriched with 200 microCi of [5-3H]uridine for 20 min. Epon ultrathin sections of the embryos were investigated for the nucleolar morphology and for the appearance and localization of the sites of [5-3H]uridine incorporation by means of electron microscopic autoradiography. In addition to this, a general pattern of replicated embryonal DNA distribution was revealed by [methyl-3H]thymidine incorporation and light microscopic autoradiography. The essential phases of the transformation of the small nucleolus precursor body (NPB) into a vast, functionally fully active nucleolus, characterized by typical nucleolar substructural components, are taking place within the eight-cell stage. This process differed in its morphology from the nucleologenetic process in early embryogenesis of other mammals, especially of that in the mouse. The first sign of NPB transformation was the appearance of a large central vacuole followed later on by perinucleolar chromatin penetration into NPB, documented by both morphology and [3H]thymidine autoradiography. In some cases, concentration of dense fibrillar material forming clumps or stalks was seen in the central vacuole. The following rapid nucleolar development was characterized by the formation of secondary vacuoles concomitant with the onset of [5-3H]uridine incorporation into the dense fibrillar component and with the appearance of the first granules in the otherwise fibrillar structure of the nucleolus. During the late eight-cell stage, the still-rounded nucleolus developed features of a reticulated nucleolus known from somatic cells intensively synthesizing rRNA: a dense fibrillar component with associated labeling encircling fibrillar centers and a well-developed granular component. The labeled dense fibrillar component was observed mostly in the central area of the nucleolus; early embryonic NPB dense fibrous material not involved in transcription was disappearing rapidly. At the transition to the 16-cell stage the nucleoli lost their rounded shape because of the accumulation of a large amount of granular component, and they occupied a considerable part of the nucleus. In conclusion, the appearance of the nucleolar vacuole in eight-cell cow embryo is the starting point for following morphogenetic events linked with the onset of transcription.

Hartmann's solution as a medium for collection and transfer of rat and cow embryos

Hartmann's solution as a medium for collection and transfer of rat and cow embryos

Recent progress in cryopreservation of cattle embryos

Two cryopreservation methods adapted to field use were investigated with cow embryos. Firstly, a freezing method combining 1.36 M glycerol and 0.25 M sucrose in PBS. The use of sucrose, by its osmotic properties, predehydrates the embryo allowing slow cooling at 0.3°C/min to be interrupted at −25°C. Direct transfer, after thawing from liquid nitrogen, gave a pregnancy rate of 51.8% ( 14 27 ). Secondly, a vitrification method involving a mixture of glycerol and 1,2 propanediol in PBS at concentrations giving an amorphous state on plunging in liquid nitrogen. Pregnancy rate after transfer of vitrified late morulae very early blastocysts was 39.1% ( 9 23 ).

Effects of LH-RH Analogue on the Ovarian Response and Quality of Embryos in the Japanese Black Cows Superovulated with PMSG and PGF2&alpha;

Effects of LH-RH Analogue on the Ovarian Response and Quality of Embryos in the Japanese Black Cows Superovulated with PMSG and PGF<SUB>2</SUB>&alpha;

Autoradiographic detection of the earliest stage of [3H]‐uridine incorporation into the cow embryo

Cow embryos, obtained from superovulated heifers on days 3 and 4 after oestrus, were cultured for 20 min in Ménézo's complete culture medium (B2), enriched with 200 microCi/ml of 5-[3H]-uridine. Semi-thin Epon sections of this material were investigated by autoradiography for sites of RNA synthesis. It was found that 5-[3H]-uridine was incorporated into the nucleoplasm and nucleoli only at the end of the 8-cell stage. This suggested that synthesis of hnRNA and rRNA occurred from this stage onwards. Ultrastructural studies were performed on these embryos as well as on other non-incubated 4-cell embryos recovered on day 2. The transformation of dense fibrillar primary nucleoli into functional reticulated nucleoli appeared sooner in the development of cow embryos than in other mammalian species hitherto studied and took place generally during the 8-cell stage. An unusual step in this transformation was represented by the development of a single vacuole in nucleoli at the beginning of this stage (day 3 post-oestrus).