At the start of the twentieth century, gay and straight held no obvious semantic relation; in the middle of the twentieth century, the words remained unrelated; now, at the start of the twenty-first century, they would be readily identified as ‘opposites’. But why do certain pairs of words have antonymous status conferred upon them? What does this process involve? And where do new antonyms come from? This chapter seeks to address such questions by identifying embryonic antonyms: pairs of words which, corpus evidence suggests, might be developing into the ‘opposites’ of tomorrow.