Religion starts with a commitment to a number of basic beliefs. The details of these beliefs vary from faith to faith, and I do not think that they can be subjected to experimental test. They may conflict with current scientific ideas and thinking, in which case they merit re-examination. The classic conflict between geocentricity and heliocentricity led most serious religious thinkers to the conclusion that geocentricity was not an essential item of belief. But the determinism of nineteenth-century Newtonian mechanics with its negation of divine providence and free will could not be accepted. This time it was science which had to retract; twentieth-century theories have abandoned the deterministic picture and talk instead in terms of probability. These examples indicate that the interaction between science and religion exercises a healthy influence on both.