Two experiments were conducted in order to determine whether children are able to make analogies in learning to read. In the first experiment, 24 children from a primary school were taught three types of word pairs--only one pair of which was analogies--and then tested. Results showed not only that children are aware that consistency of spelling predicts consistency of pronunciation in the way required for an analogy but also that they are able to apply this knowledge selectively when given conflicting information about spelling-sound sequences. In the second experiment, primary school children were given a "clue" word from which analogies could be made, and then asked to read analogous and nonanalogous test words. The subjects were tested under three different conditions with three different types of test words. If children were able to make analogies in reading without any training on the relevant orthographic sequence, they should have been better at reading the target words than the control words. This pattern of results seems to be supported. These results suggest that the ability to make analogies is not a developmental ability but a fundamental strategy in learning to read. (DF)