PCVAD is a disease of finishing pigs that is controlled by vaccination. It is effective in prevention of PCVAD and reducing the level of PCV2 in serum, but does not eliminate infection. Since nearly all of the finishing herds in the United States are vaccinated, we are inadvertently providing a large-scale selective pressure on PCV2 for new strains that grow better in the presence of an anti-PCV2 vaccine response. At the same time, we are maintaining a continuous source of infection from shedding pigs, and raising the fixed cost of pork production. Our goal is to understand mechanisms of immunity that will help make vaccines that completely eliminate PCV2 infection. To achieve this goal we need a challenge model that mimics infection in the field. Contrary to the widespread belief that infection occurs at 10-15 weeks of age, we thought that infection might occur early in life since nearly all market-age pigs appeared to be infected. Here, we examined PCV2 and anti-PCV2 immune status in sows, piglets, and the farrowing environment from 6 sow farms in Minnesota, Iowa, and Indiana. The vast majority of sows are viremic for PCV2 even though they have high levels of PCV2-specific antibodies. The virus is everywhere; colostrum, oral fluids, feces, skin, crate bars, and floor. Cleaning with disinfectants reduces viral load, but does not eliminate it. Seventy-eight percent (78%) of piglets were born viremic, and the rest are infected soon after. Our findings, that sows of all parities are persistently infected in the presence of an anti-PCV2 immune response and that piglets are infected in utero or during nursing, show that PCV2 infection occurs near the beginning of life and is lifelong. Thus, a relevant challenge model for vaccine development should use infected pigs.