Experiments were conducted to learn whether the binding of microtubule-associated proteins (MAPs) to microtubules alters the flexibility of the microtubules. Flexibility was measured in vitro by two established techniques. The first employed measurement of the bending of the microtubule in a flow of buffer; the second involved repeated measurement of random thermal fluctuations in the microtubule's shape. Similar values were obtained from microtubules prepared from purified tubulin and those prepared from microtubule protein containing saturating concentrations of MAPs isolated from bovine brain. When measured by the flow technique at 37 degrees C and pH 6.9, the persistence length of pure tubulin microtubules was found to be 8.4 +/- 2.2 mm and that of MAP-containing microtubules was 9.4 +/- 2.7 mm, not significantly different from each other. When measured by the thermal fluctuation technique under identical conditions, values of 6.2 +/- 0.8 and 6.5 +/- 0.8 mm were obtained, again not significantly different from each other. The results show that the binding of MAPs to native microtubules in vitro has little or no effect on their flexibility. MAP-induced effects on the cytoskeleton observed in vivo are likely to be due to other causes, such as formation of microtubule bundles.