Natural Killer (NK) cells take part in protective immune responses against infections or malignant transformations. They distinguish healthy from abnormal cells by assessing expression levels of MHC class I, that provide inhibitory input, and expression of stress-induced proteins, providing activating input. These stress-induced proteins are ligands for activating NK cell receptors. NK cells variably express a number of activating receptor. Amongst these, NKG2D is a well-characterized receptor expressed by all NK cells. NKG2D ligands are expressed by a number of tumour cells. However, as an immune evasion strategy, NKG2D ligands may be shed from the tumour cells as soluble ligands as a way to evade NK cell recognition. We hypothesized that these soluble ligands could act to evade tumour recognition by preventing migration of NK cells to the tumour site. This hypothesis was based on previous work demonstrating that activating signals may reduce lymphocyte migration. We produced soluble NKG2D ligands in HEK293T cells, and investigated their effect on NK cell migration towards CXCL12 induced via the chemokine receptor CXCR4, by culturing NK cells in transfection supernatants. We were able to demonstrate that soluble NKG2D ligands, the mouse Rae-1δ, Rae-1ε, and MULT1 and the human MICA, reduced NK cell migration. This was assessed with both primary NK cells and with cell lines. Furthermore, we introduced a more potent stimulation of NKG2D by cross-linking of NKG2D with antibodies, and found that the negative effect on migration was further amplified. NK cells are a heterogeneous cell population, and NK cells in distinct stages of differentiation co-exist at steady-state in the periphery. Primarily they are distinguished as CD56bright and CD56dim subsets in the human, with further sub-divisions of CD56dim NK cells are distinguished based on CD57 and NKG2A expression. We tested whether subsets of NK cells displayed different migration behaviour towards CXCL12, and found higher response with the less mature CD56bright subset but similar responses within the CD56dim subsets. To further pinpoint the molecular mechanism for the putative crosstalk between NKG2D and CXCR4, a pilot experiment for identifying intracellular signalling molecules of cross-talk was conducted, however the results were too preliminary to reach a conclusion. In conclusion, we found that NKG2D ligands may negatively affect migration. An influence on migration by soluble ligands shed from tumour cells might be an escape mechanism of tumour cells. Enlightening this field, might open for new strategies in cancer therapy.