Our objective was to investigate the effects of intravenous (IV) or intrauterine (IU) lipopolysaccharide (LPS) challenge at 5 or 40 d postpartum (DPP) on clinical signs, systemic and uterine inflammation, dry matter intake (DMI), and milk yield (MY). Holstein cows at 5 DPP (n = 23) or at 40 DPP (n = 24) were blocked by parity and randomly assigned to one of 3 treatments: 1) IV-LPS [0.0625 μg/kg BW (5 DPP) or 0.1 μg/kg BW (40 DPP) over 1h], 2) IU-LPS [100 μg (5 DPP) or 300 μg (40 DPP) in 20 mL saline], or 3) 20 mL saline IU (IU-SAL; same for 5 and 40 DPP). The proportion of polymorphonuclear (PMN) cells was measured by endometrial cytology at d -1, 1, 4, and 7 relative to treatment. Blood haptoglobin (Hp), serum-amyloid A (SAA), and LPS-binding protein (LBP), DMI, and MY were measured from d -1 through 7. Data were analyzed separately for each DPP group in multivariable linear regression models accounting for repeated measures. For both DPP groups, there were increases in rectal temperature, heart and respiratory rates, and a decrease in rumination rate following IV-LPS, but not following IU-LPS. At 5 DPP, endometrial PMN proportion was similar in IU-LPS and IU-SAL. Serum Hp was unaffected by LPS challenge, SAA was greater in IV-LPS from 12 to 24 h after challenge, and LBP was greater in IV-LPS from 8 to 24 h. At 40 DPP, PMN was greater in IU-LPS (37 ± 4%) than in IU-SAL (15 ± 4%) 1 d after LPS challenge. Serum Hp was greater from 24 to 72 h after challenge in IV-LPS than in the other groups, SAA was greater in IV-LPS from 6 to 48 h, and LBP was greater in IV-LPS from 8 to 24 h. At both 5 and 40 DPP, treatment did not affect DMI, but MY was lesser in IV-LPS cows at 12 and 24 h than in IU-SAL or IU-LPS. The IV-LPS challenge resulted in more pronounced changes in clinical signs and acute phase protein (APP) concentrations than IU-LPS or IU-SAL at 40 DPP, but more subtle or inconsistent changes at 5 DPP. These may be due to the different doses of LPS used at 5 and 40 DPP or possibly due to the high variation in baseline clinical signs and APP observed in all groups at 5 DPP. The IU-LPS increased uterine PMN 1 d after challenge at 40 DPP, but not at 5 DPP. At each time, IU-LPS did not produce changes in clinical signs or markers of systemic inflammation.