Well, in the 3rd century BCE, Eratothenes calculated a fairly accurate circumference of the earth. I don't know how widely accepted this measurement was, but, given that measurement, and the general distribution of land vs ocean observed by the ancients, it's not a huge stretch to assume that there must be something out there on the other side of the world. Some of the early world maps were remarkably accurate, at least for the areas that were well-known to the geographers, even given the very limited tools at their disposal[1].
There's also the curious case of the apparent Roman-era shipwreck that was discovered off the coast of Brazil[2]. It's a little tinfoil-hatty, and possibly a hoax, but it is interesting.