[Truncated abstract] Mastitis during lactation is a severe illness and approximately one in five breastfeeding women in Australia suffer at least one episode in the first 3-6 months post partum. Despite this, there is little understanding of the physiological and pathological processes occurring before, during and following mastitis. In this study 26 women who, on the basis of my previous research, were assessed to be at risk for developing mastitis, were recruited during the first week post partum and followed prospectively throughout the course of their lactation. Breastmilk, and 24 hour urine samples were collected at Days 5, 14, 30, 60 and 90 post partum and blood was collected at Days 5 and 14 post partum. If participants experienced inflammation of the breast at any time, either during the 90 day reference sampling period or later in their lactation, samples were then collected daily for the duration of symptoms and then, as a follow up, again one week following resolution of symptoms. Breastmilk samples were analysed for a range of biochemical components that reflect immunological (sIgA, lactoferrin) and acute phase (C-reactive protein) response, synthetic activity (lactose, glucose), and permeability of the paracellular pathway (sodium, chloride, lactose and serum albumin) within the breast. Blood and 24 hour urine samples were analysed for lactose, and blood was also analysed for C-reactive protein (CRP). Bacteriological examination of milk samples was undertaken where clinical mastitis was present. Results from these analyses were compared to the severity of breast and systemic symptoms experienced. Twenty-two episodes of mastitis and 13 episodes of blocked duct(s) were identified during the study period. When adjusted for co-existing breast pathology milk composition in the breast affected by blocked duct(s) was generally not different from that of healthy breasts. One mother, who was IgA deficient, experienced six of the 13 episodes of blocked duct(s). It is possible that the absence of sIgA in the milk of this mother increased her susceptibility to inflammation of the breast. During mastitis there was a significant increase in sodium, chloride, and serum albumin to a median concentration of 23 mmol⁄l; (p<0.001); 30 mmol⁄l; (p<0.001) and 0.8g⁄l; (p<0.001) respectively, and a decrease in the median concentration of lactose in milk to 152mmol⁄l; (p<0.016) from the mastitis breast when compared to the contralateral asymptomatic breast. Increased permeability of the paracellular pathway was confirmed by a significant increase in the median daily excretion of lactose in urine to 7.5 mmol⁄24hour (p<0.001). The rate of excretion of lactose in urine over a 24 hour period proved to be, not only a reliable means of assessing breast permeability, but also allowed the researcher to discern whether milk sampled from the breast affected was representative of milk at the site of the inflammation. The changes in lactose in urine were generally consistent with the changes in sodium, chloride and lactose in milk confirming milk expressed for sample analysis was representative of milk from the site of the inflammation