Paratuberculosis is a chronic and progressive disease of the intestine in ruminants caused by Mycobacterium avium subsp. paratuberculosis (Map). The bacterium is transmitted to young animals, becomes manifest in adulthood and leads to economic losses. The aim of this study is to investigate if cows shedding Map possess oocytes and embryos that are carriers of the bacterium. New genetical material can enter the dairy farm using embryo transfer but the question as to whether this technique is safe with respect to transmission of paratuberculosis has yet to be addressed. We selected and bought 16 cows, all proven to be moderate shedders of the bacterium in the faeces immediately prior to the experiment but none were clinically sick. One sample of uterine content was collected from each animal by flushing the uterus on the day of heat and five samples of homogenised uterine tissue were collected on the eighth day of the same cycle by biopsy. In addition, 217 cumulus-oocyte complexes (COCs), ranging from 3 to 35 COCs per animal, were collected using ultrasound guided transvaginal puncture of the ovarian follicles (OPU). On the seventh day of the subsequent cycle 31 embryos were obtained using the classic technique of super ovulation induction, artificial insemination (AI), followed by flushing of the uterus. These embryos have been washed and trypsinised. Fourteen of the 16 cows were treated again for super ovulation in the subsequent cycle and 19 foetuses were collected by opening of the uterus after euthanasia on Days 35-49 of the cycle. All samples were cultured for presence of Map and checked every 2 months during 1 year for bacterial growth. None of the samples showed growth of Map after 12 months of culture. Pathological examination of the cows revealed different degrees of severity of pathological alterations of the intestinal tract and mesenteric lymph nodes. However, the results suggest that neither in vivo embryo's nor oocytes are carriers of the bacteria and do not form an extra risk at transfer. However, due to the limited size of the experiment (sample size of 16 cows), a certain margin for error remains.