Plasmodium vivax is responsible for 70-80 million cases of malaria annually and has the widest geographical distribution of the human malaria parasites. Although not as deadly as the human malaria parasite, Plasmodium falciparum, its widespread distribution and the lack of sustained naturally acquired immunity cause it to be a very heavy public health and economic burden worldwide. Development of an effective vaccine against this parasite is a public health imperative. Clinical symptoms of malaria are caused by the blood stage of the disease in which the parasites invade red blood cells, replicate inside them and then are released into the circulation to invade new red blood cells. This process results in the destruction of large numbers of red blood cells causing fever and chills and severe anemia among other symptoms. The invasion process is reliant upon specific interactions between parasite ligands and host cell receptors. In the P. vivax, the interaction between the parasite Duffy binding protein (PvDBP) and the host Duffy antigen/receptor for chemokines (DARC) is absolutely required for invasion. Individuals who lack DARC are completely protected from blood stage infection and therefore, clinical disease. PvDBP is, therefore, a very attractive vaccine candidate. PvDBP is a highly polymorphic protein, complicating vaccine development. Understanding of the role of these polymorphisms in determining strain-specific immune responses and determination of residues that form conserved epitopes will be vital for successful development of a strain-transcending vaccine. We examined the basis for a limited polymorphisms N417K and found that a number of amino acid substitutions are excluded from natural isolates because of their poor binding phenotypes and that the naturally occurring residues N and K are the most antigenically distinct. We also created a panel of PvDBP variants representative of global genetic diversity and examined their responses to immune inhibition, demonstrating that they show strain-specific responses. Furthermore, we compared PvDBP to a DBP from Plasmodium knowlesi, PkÌ_å_, to determine the molecular basis for their host-specific invasion phenotypes. We also validated a primate model for use in vaccine testing, demonstrating that invasion of Aotus nancymai erythrocytes by PvDBP is Duffy antigen dependent.