The dose of vitamin D that some researchers recommend as optimally therapeutic exceeds that officially recognized as safe by a factor of two; it is therefore important to determine the precise mechanism by which excessive doses of vitamin D exert toxicity so that physicians and other health care practitioners may understand how to use optimally therapeutic doses of this vitamin without the risk of adverse effects. Although the toxicity of vitamin D has conventionally been attributed to its induction of hypercalcemia, animal studies show that the toxic endpoints observed in response to hypervitaminosis D such as anorexia, lethargy, growth retardation, bone resorption, soft tissue calcification, and death can be dissociated from the hypercalcemia that usually accompanies them, demanding that an alternative explanation for the mechanism of vitamin D toxicity be developed. The hypothesis presented in this paper proposes the novel understanding that vitamin D exerts toxicity by inducing a deficiency of vitamin K. According to this model, vitamin D increases the expression of proteins whose activation depends on vitamin K-mediated carboxylation; as the demand for carboxylation increases, the pool of vitamin K is depleted. Since vitamin K is essential to the nervous system and plays important roles in protecting against bone loss and calcification of the peripheral soft tissues, its deficiency results in the symptoms associated with hypervitaminosis D. This hypothesis is circumstantially supported by the observation that animals deficient in vitamin K or vitamin K-dependent proteins exhibit remarkable similarities to animals fed toxic doses of vitamin D, and the observation that vitamin D and the vitamin K-inhibitor Warfarin have similar toxicity profiles and exert toxicity synergistically when combined. The hypothesis further proposes that vitamin A protects against the toxicity of vitamin D by decreasing the expression of vitamin K-dependent proteins and thereby exerting a vitamin K-sparing effect. If animal experiments can confirm this hypothesis, the models by which the maximum safe dose is determined would need to be revised. Physicians and other health care practitioners would be able to treat patients with doses of vitamin D that possess greater therapeutic value than those currently being used while avoiding the risk of adverse effects by administering vitamin D together with vitamins A and K.