Castoe suggests that you only have to look at the evolution of snakes to see a possible answer. Snake ancestors were lizard-like animals that took to a burrowing way of life. Their limbs were a hindrance and were gradually lost. Over time, their bodies elongated, and their right lung followed suit while their left lung became small and useless. They may have been rather a lot like the blind snakes that still live in the tropics today.

Or they may not have been.
The other hypothesis is that the snakes came from the sea, that their closest relatives are the dolichosaurs and the mighty mosasaurs; at least the mosasaurs routinely swallowed large prey whole, and almost all of the earliest known snakes (from the beginning of the Late Cretaceous) were marine. The burrowing adaptations of the blind-, thread- and wormsnakes are interpreted as a separate package of innovations under this scenario, which is bolstered by recent papers (2004, 2006) on the fossil madtsoiid snakes Wonambi and Yurlunggur.
The basic venom inventory is older than the snakes, too: it is shared with the anguimorphs (monitors, slow-worms etc.) and the iguanians (iguanas, agamas, chamaeleons etc.). Google “Toxicofera”.