I expect this would be readily learned from observation (of humans [changes in menstruation relating to puberty and during pregnancy would give a faster-acting

I expect this would be readily learned from observation (of humans [changes in menstruation relating to puberty and during pregnancy would give a faster-acting conceptual connection] and/or of livestock). The exact mechanism might easily be misunderstood and subject to all kinds of wrong theories, but the causal linkage would be hard to avoid learning. Is it not the general consensus among anthropologists that this practical knowledge would long predate writing? Surely, if ancient humans could understand and then predict the seasonal movement of the stars to plan agriculture, they'd be able to work out that there is a linkage between sex and pregnancy.