INTRODUCTION: There is spatial variability of selenium (Se) in soil and crops in Ethiopia. We assessed the Se content of food items, breast milk, and urine among infants in Ethiopia from two areas with contrasting Se concentrations in soils. METHODS: Dietary Se intakes among children (6-23 months) were evaluated using a weighed food record on two non-consecutive days. Also, spot urine samples from children and breast milk samples from their mothers were collected to determine Se concentration. Selenium concentrations in the samples were analyzed using an inductively coupled plasma mass spectrometer (ICP-MS). RESULTS: Injera (prepared from teff and mixtures of other cereals) with a legume-based stew were the most frequently consumed foods by the children in both areas, followed by pasta. Overall, the Se concentration (mean ± SD) of food items, breast milk (12.2 ± 3.9 μg/L vs. 3.39 ± 1.5 μg/L), and urine samples (22.5 ± 11.5 μg/L vs. 3.0 ± 1.9 μg/L) from East Amhara were significantly higher than the corresponding samples from West Amhara (p < 0.001). The total Se intakes by the study children from East Amhara and West Amhara were 30.2 [IQ 25%, 14.2; IQ 75%, 54.1] and 7.4 [IQR 25%, 4.2; IQ 75%, 10.6] μg day-1, respectively; 31.5% of children from East Amhara and 92% of children from West Amhara were at risk of inadequate Se intakes. Urinary Se excretion accounted for 53 and 39% of daily dietary Se intake in East Amhara and West Amhara, respectively. Dietary Se intake was positively correlated with urinary Se excretion in East Amhara (r = 0.56; p < 0.001) but not among samples from West Amhara (r = 0.16; p ≥ 0.05), suggesting greater physiological Se conservation in a state of deficiency. CONCLUSION: There is spatial variability of Se in foods, breast milk, and urine in Ethiopia, suggesting the need for implementation of targeted agronomic interventions that enhance Se concentrations in the edible portion of plant foods.