Lipopolysaccharides (LPS, endotoxins) belong to the strongest elicitors of the mammalian immune system due to the induction of a series of cytokines such as tumor-necrosis-factor-α (TNFα) in immunocompetent cells like mononuclear cells. Since the effects of LPS on human health may be pathologically at too high concentrations (e.g., septic shock syndrome), it is of uttermost importance to have a reliable assay for measuring the concentrations of endotoxins in vitro and in vivo (human body fluids). The activation of the clotting cascade from the horseshoe crab (Limulus polyphemus), the Limulus amoebocyte lysate test (LAL), has been the standard and most sensitive assay to detect bacterial endotoxins. However, there are restrictions with this test. It was found in some clinical trials that the results from the LAL test did not correlate with the presence of bacteremia due to Gram-negative organisms or with the mortality but correlated with the presence of fungal bloodstream infections. This resulted from the fact that the LAL assay does not only respond to bacterial endotoxins but is activated also by (1→3)-β-D-glucan. Furthermore, in extensive studies the structural requirements for activation of the LAL test were analyzed, and it was found that the LAL activity correlated with pyrogenicity but not with activation of the complement cascade. Furthermore, there was no correlation of the LAL activity with cytokine expression (for example tumor-necrosis-factor-α and interleulkins-1 and 6) in mononuclear cells when the 4/2 acyl chain pattern of enterobacterial lipid A was changed, or when the cytokine production induced by LPS from various different species in the whole blood assay was compared with the response from the LAL test. To clarify the questions raised by the different experimental findings, data from literature are summarized to get a more closer insight where the Limulus test confidentially monitors the endotoxicity of LPS and other compounds and where this is not the case, and which are the decisive epitopes for recognition of the LPS molecules. These data are very crucial for example in clinical tests, whether the LAL assay can reliably describe the effectivity of an antibacterial therapy.