Thermodynamic feasibility analysis (TFA) has been used as a toot capable of providing additional constraints to the mass balance-based methods of analysis of Metabolic networks (e.g., flux balance analysis). Several publications have recently appeared in which TFA of different metabolic pathways from relatively simple to the genome-scale networks was described as a means of detecting the possible metabolic control steps. However, in order to perform TFA, many simplifying assumptions were necessary. On the other hand, it has been shown by applying TFA to the well-known pathway of glycolysis that erroneous simplifying assumptions may seriously bias the results of the analysis. A quantitative analysis of the influence of non-ideality of the biochemical system, pH, temperature, and complexation of the metabolites with Mg2+ ions as well as a number of other factors on the TFA is reported. It is shown that the feasibility of glycolysis is very seriously limited by the reaction of oxidative phosphorylation of glyceraldehyde phosphate, and that the intracellular concentration of the main product of this reaction, biphosphoglycerate, must be anywhere from 10 to 100 times lower than published values. In addition, the driving force for this reaction, and consequently the feasibility of the entire pathway depend strongly on the intracellular pH and ionic strength and to a lesser extent on pMg and temperature. The analysis may also be influenced by uncertainties of the dissociation and magnesium complexation constants of glyceraldehyde phosphate. The analysis demonstrates the crucial importance of taking such factors into account when performing TFA. It also suggests an urgent need for experimental determinations of such factors as a prerequisite for sensible thermodynamic analysis of metabolism on a genome-wide scale. Biotechnol. Bioeng. 2009;103: 780-795. (c) 2009 Wiley Periodicals, Inc.