The ability to read well is essential for individuals to function effectively in modern societies. However, numerous studies suggest this ability changes across the adult lifespan, and that older people are slower and less good readers. Measures of eye movements in reading provide an excellent method for studying these effects, as this is informative about both basic oculomotor behaviour and cognitive processes involved in word recognition in reading. I therefore report two studies looking at adult age differences in eye movements during reading. The first is an experiment that used the boundary paradigm to investigate age differences in the use of contextual information to parafoveally pre-process words. In this paradigm, an invisible boundary is placed in front of a specific target word in the text. Prior to the reader’s eyes crossing this boundary, the word is shown either as normal or masked (e.g., by replacing letters with visually similar letters). This word was also either highly predictable or less predictable from the prior context, allowing an examination of parafoveal preview effects for words that differ in predictability. Data collection was disrupted by the COVID-19 pandemic, although preliminary analyses point to age differences in predictability effects. A second study used meta-analysis to examine age differences in basic eye movement parameters and both word frequency and word predictability effects across studies conducted in alphabetic scripts and Chinese. The findings show a larger age difference for Chinese than alphabetic scripts (with slower and more careful reading by older adults). The findings also confirm age differences in word frequency and word predictability effects across both alphabetic scripts and Chinese, consistent with slower word recognition processes for older adults, especially when words are used less frequently or are less predictable from context. I discuss these findings in relation to theories of ageing effects in reading.