The biology and significance of micrometastases remain poorly understood. Whether such deposits represent hypothetical "dormant" metastases, simply the earliest metastases recognizable, or both has not been resolved. Attempting to answer the latter question among many others, we carried out studies on the rates of proliferation, and microvessel counts in melanoma micrometastases taken from sentinel lymph nodes as compared to conventional melanoma lymph node macrometastases. We found that these micrometastases were not vascularized and had low (and comparable) rates of both proliferation and apoptosis, suggesting a steady or dormant state. On the other hand, the macrometastases were fully vascularized and more proliferative, i.e., they had rates of proliferation that were significantly higher compared to the micrometastases and rate of apoptosis. How micrometastases develop is also poorly understood. Tumor cells are thought to arrive in lymph nodes and other sites through the blood and lymphatic circulation and to extravasate. However, in addition to the latter explanation, another mechanism may be operable which we have proposed as extravascular migratory metastasis. In studies of metastatic melanoma we have identified melanoma cells positioned on the surface of endothelial cells both by light and electron microscopy. We have also identified ultrastructurally the presence of an amorphous matrix interposed between the latter two cell types that shows immunostaining for laminin and, recently, the beta-2 chain of laminin. Thus, we currently believe that the latter form of "free" laminin may play a role in this proposed mechanism of metastasis and thus in the formation of micrometastases.