Summary Eighteen mature Jersey cows were divided into two groups of nine cows eash. One group was fed five million units of vitamin D in the form of Viosterol daily for 2 wk. prepartum and for 3 days postpartum. The other group served as controls. All cows were fed a high mineral ration containing 5 per cent steamed bonemeal at the rate of 6lb. daily beginning 8 wk. before the due date and continuing for 1 wk. postpartum. Blood samples were drawn for analyses two weeks prepartum, within 12 hr. both pre and postpartum and 1 wk. postpartum. An increase in serum Ca and P resulted prior to parturition which was greater than when one and two million units had been fed previously. However, the increase was nullified within 12 hr. postpartum or before the usual time for milk fever to occur, which was found to average 21.7±7.1 hr. postpartum in 36 eases. The upward trend of serum Mg at parturition was similar to changes previously observed. Four of the nine vitamin D-fed cows and three of the controls developed milk fever. The vitamin D content of the blood of cows fed five million units daily was found to be about ten times higher than that of the control cows. The changes in plasma carotene, vitamin A and ascorbic acid were somewhat similar to those previously reported. Data are presented showing that milk-fever cows secrete no more colostrum than normally freshening cows, nor does the colostrum of milk-fever cows have a higher percentage of ash or Ca than normally freshening cows. Thus, no greater drain on blood Ca due to mammary secretion is experienced by milk-fever cows than cows that freshen normally. It also was shown that the average time between the first treatment and the relapse, when one occurred, was approximately the same as that between parturition and the first attack. These observations are considered evidence that milk fever is primarily caused by the failure of the blood Ca regulatory mechanism (probably the parathyroid glands) to mobilize blood Ca rapidly enough to meet even the normal drain on blood Ca at parturition. Thus recovery is characterized by the return to normal of the blood Ca regulatory mechanism and may or may not be influenced by the injected Ca which serves to eliminate the symptoms by elevating the blood Ca. It is reasoned that the failure to obtain sufficient calcemic effect at parturition from vitamin D feeding to prevent milk fever may have been due, at least in part, to the suppression of the cow's own parathyroid activity resulting from too long a feeding period with its associated prolonged elevated blood Ca. Also, the dosage may have been too low. Experiments are now in progress in which larger dosages of vitamin D are being fed for shorter periods of time prepartum with the aim of supplementing without suppressing parathyroid activity.