al., 1985;Evelhoch et al., 1986;Tozer et al., 1987; Vaupel et al., 1989a, b;Steen & Graham, 1991 (Bremner et al., 1993). The reduction in the oxygenation status caused by hydralazine is insufficient for detection by 31P-NMR in human xenografted tumours, whereas in RIF-1, SCCVII/Ha and KHT murine tumours large increases in the Pi/total phosphate ratio are found (Bremner et al., 1991;Adams et al., 1992). In the light of these discrepancies, a number of relevant issues need to be clarified regarding the relationship between energy status and tumour blood flow. In the present study, the key questions were as follows:(1) Do nected to a ratemeter (FHT 1100 FAG Kugelfischer, Erlangen, Germany). The method of evaluation of TBF was identical to that described earlier (Kluge et al., 1992). Measurements were performed on tumours of varying sizes at 20 min intervals before application of TNF-x and at 30 min intervals thereafter over a total time period of 2 h post treatment. In all experiments performed in this study, animals were allowed to stabilise following the surgical procedures. Measurements commenced once constant baseline readings for MABP and flow were obtained for at least 20 min.Laser Dopplerflowmetry A Periflux model PF 3 dual-channel laser Doppler flowmeter was used for this study (2 mW He-Ne laser, wavelength 632.8 nm; Perimed, Stockholm, Sweden , 1989a, 1991). Measurements were made either on tumours of varying sizes or before and 120 min after acute flow changes upon TNF-a application.Measurement ofglobal concentrations of adenylate phosphates in perchloric acid extracts In order to obtain mean (global) levels of adenylate phosphates, the tumour-bearing hindfoot was rapidly frozen and the tumours (n = 12) were prepared under liquid nitrogen and stored at -80°C for further processing. In a first series of experiments, tumours of varying sizes were analysed. In another series, tumours were assayed before or 120 min after administration of TNF-o. Each deep-frozen tumour was ground to a fine powder and freeze dried. For determination of ATP, ADP and AMP levels, aliquots of freeze-dried tissue were extracted with 0.66 M perchloric acid, centrifuged and the supernatant neutralised with 2 M potassium hydroxide. The concentrations of the adenylate phosphates were then determined using reversed-phase high-performance liquid chromatography (HPLC) techniques at 254 nm (for more details see Kruger et al., 1991;Schaefer et al., 1993). Concentrations were expressed as j.mol per g tissue wet weight.Determination of microregional A TP distribution Before preparing the rapidly frozen tumours for HPLC analysis, approximately 30% of the tumour mass was separated, cut at -25°C in a cryostat into 5-nm sections and used for ATP bioluminescence measurements to assess the microregional ATP distribution using single-photon imaging and quantitative bioluminescence (for methodological details see Walenta et al., 1992;Schaefer et al., 1993). The spatial resolution gained by this method is about 50 jam, thus revealing informatio...