Yet it is certain that the supply of the winter pole cap must come from the evaporation of water in some region or other. And here is another oversight of the artificial canal theory. The canals are supposed to be necessary for the conveyance of water from the pole towards the equator; although, as this was "uphill," vast pumping stations at short intervals had to be predicated. But it is not supposed that the water needed to travel by way of the canals to the poles. If, however, the moisture is conveyed as vapour through the atmosphere to the pole as winter approaches, it cannot be impossible that it should be conveyed in the same manner from the pole as summer draws on, and in that case the artificial canals would not be needed. If the canals are necessary for conveying the water in one direction, they would be necessary for the opposite direction. But there would be something too farcical in the idea of the careful Martians dispatching their water first to the pole to be frozen there, and then, after it had been duly frozen and melted again, fetching it back along thousands of miles and through numerous pumping stations for use in irrigating their fields. Of all the many hundreds of canals only a few actually touch the polar caps. But on the theory that the entire canal system is fed by the polar cap in summer, the carrying capacity of the polar canals should be equal to, if not greater than, that of the entire system outside the polar circle. A glance at the charts of the planet shows that the polar canals could not supply a twentieth part of the water needed for those in the equatorial regions. Another oversight is that of the significance of the alleged uniformity and breadth of the canals. Prof. Lowell repeatedly insists that the canals are of even breadth from end to end, and spring into existence at once throughout their whole length. This statement is in itself a proof that the canals cannot be what he supposes them to be. An irrigation system could not have these characteristics; the region fertilized would take time to develop; we should see the canal extending itself gradually across the continent, and its breadth would not be uniform from end to end, but the region fertilized would grow narrower with increase of distance from the fountain head of the canal.