I think there is more information being transmitted than just that the soil is moist. The soil is lightly moist + highly porous + has high levels of microbiolog

I think there is more information being transmitted than just that the soil is moist. The soil is lightly moist + highly porous + has high levels of microbiological activity which are the optimal conditions for seed germination. Consider that inedible seeds could act as a preserved food that could be germinated and edible in only a few days with no energy input by hunter gatherers who were known to gather and eat seeds - giving more like 70-80k years for Petrichor to be interpreted by humans as an attractive smell. The plant growth/food hypothesis doesn't make sense to me because this signal is timing specific and the timing isn't related fruiting periods. Also growing plants have their own smells so why would an indirect smell be preferred? Also, consider the evolutionary path of associating a smell doesn't require any protein production or chemical changes - only neurological ones. This shows that mice can inherit the ability fear certain smells from their parents. It's possible that it could happen epigenetically in humans as well considering the significant changes in taste and smell sense experienced in ones lifetime. [0]https://www.smithsonianmag.com/smart-news/baby-mice-can-inherit-fear-of-certain-smells-from-their-parents-180948096/ https://www.smithsonianmag.com/smart-news/baby-mice-can-inhe... [edit: added source]