The prognosis of malignant disease is essentially determined by the metastatic potential of the primary tumor. In the past, scientific attention was chiefly directed to systemic metastasis. A multitude of biological and molecular tumor markers and mechanisms has been uncovered enabling a better contemporary understanding of the process of hematogenic metastasis. This is in contrast with our knowledge of the mechanisms and pathways of lymphatic tumor spread, which is rather limited. We do know, however, that adequate surgical clearance of the regional lymphatics improves treatment results of many tumors. How far this lymph node dissection is directly therapeutic is a source of controversy. While in some instances, a stage-adjusted survival benefit was demonstrated, this may very well be attributable at least in part to the phenomenon of stage migration (Will Rodgers phenomenon) through better staging. However, it is uncontested that an established diagnosis of regional lymphatic spread is prognostically significant and should influence the indication for additional therapy and eventually for an intensive follow-up. For many tumors, the indication for adjuvant chemotherapy depends on the nodal status. On the other hand, it is equally well known that aggressive lymphatic dissection increases perioperative morbidity and even mortality. Long-term sequellae from regional lymphatic dissection are common and the effect on the local, maybe even the systemic immunological response to the malignant disease, remains unclear. To incur such risk seems especially problematic in those patients without any lymphatic spread at the time of the pathologist's work-up. Thus, there is ongoing debate about the rationale, value, extent, advantage, or disadvantage of regional surgical lymph node dissection or even radiotherapy of the regional lymphatic drainage area for many different tumors. A considerable step forward could be made if there was any diagnostic modality enabling a reliable preoperative lymph node staging. However, there is none. General criteria like size, shape, structure, or texture in variable imaging modalities are unreliable. While it is still too early to definitely evaluate in this context new diagnostic modalities like PET, immunoscintigraphy, or contrast-enhanced MRI, the initial results do not provoke clear enthusiasm toward the development of a sensitive and specific staging tool with regard to the nodal status. Adequate specificity may be obtained by external or endoluminal ultrasound-guided fine needle biopsies. However, uncertainty arises from eventually unrepresentative tissue sampling. The sentinel lymphonodectomy technique may remedy the dilemma, enabling a risk-adapted, individual indication for regional lymphatic dissection. This concept, first introduced in 1977 by Cabanas into the treatment of penis carcinoma, is based on the evidence of orderly and predictable lymphatic drainage pathways. Tumor cell progression within the lymphatic system seems to follow a sequential pattern. Primary draining lymph nodes possess the structural and functional capability to retain and to fight tumor cells efficiently. The 'sentinel node' is defined as the first tumor draining filter, and, if uninvolved, should thus adequately predict the nodal status of the disease. Skip metastases beyond an uninvolved sentinel node are supposed to be a very rare event. The reliability of the 'Cabanas approach', however, was limited by its relatively poor localization technique, and therefore failed to gain widespread acceptance. Unfortunately, the significance of the concept was not fully appreciated at the time. It is to Morton's credit that the procedure was reinstituted in malignant melanoma through a dye injection technique at the primary tumor site. This led to a rapid development and refinement of intraoperative lymphatic mapping. One major step in this process was to use radiolabeled colloids in conjunction with gamma-camera imaging or gamma probe-guided detection of the sentinel node. At present, a multitude of studies are conducted in a variety of tumors and sites, aiming at further refinements of the technique or at clinical evaluation in comparison with established lympadenectomy. The results may well change many aspects of our operative strategy in the near future. However, assuming a technically optimized procedure, will this solve the underlying tumor biological and clinical problem with respect to the necessity and efficacy of a regional lymph node dissection in node-positive cases? This is not the case; moreover, there are additional questions raised and left unanswered so far. Without any doubt, the rate of unnecessary diagnostic lymph node dissections can be considerably reduced as soon as the sentinel node concept is sufficiently validated for general use outside clinical trials. This would be a clear step forward. It is undetermined, however, how far a cancer patient with a positive sentinel node-thus already proven lymphatic metastases-would still profit from a more or less extensive lymph node dissection. It might be sufficient to use the staging information obtained through the sentinel node's status alone to decide upon adjuvant therapies. A further aspect arises from the possibility for investigating this single and supposedly most representative lymph node in far more detail than it would be possible for the large number of nodes previously sampled in conventional lymphatic dissections. This more extensive work-up may include serial sectioning, immunological and molecular techniques to enhance the sensitivity for micrometastases detection. However, very little is known about the true prognostic significance of such conventionally occult micrometastases, and even less experience exists as to the value of adjuvant therapies in those cases. Thus, while the sentinel node procedure will probably enable a more precise though less invasive lymphatic staging of malignant disease, it raises a number of important questions, as well. The general principles of multimodal treatment will have to be redefined with regard to the new diagnostic tool, which will require extensive prospective and randomized testing before a safe and reliable advantage for the patients may be established.