even a bent of grass, upright, leafless or all but leafless, with heads of small blue or yellow flowers, and carrying, in one species, a few very minute bladders about the roots, in another none at all.  A strange variation from the normal type of the family; yet not so strange, after all, as that of another variety in the high mountain woods, which, finding neither ponds to float in nor swamp to root in, has taken to lodging as a parasite among the wet moss on tree-trunks; not so strange, either, as that of yet another, which floats, but in the most unexpected spots, namely, in the water which lodges between the leaf-sheaths of the wild pines, perched on the tree-boughs, a parasite on parasites; and sends out long runners, as it grows, along the bough, in search of the next wild pine and its tiny reservoirs.

In the face of such strange facts, is it very absurd to guess that these Utricularias, so like each other in their singular and highly specialised flowers, so unlike each other in the habit of the rest of the plant, have started from some one original type perhaps long since extinct; and that, carried by birds into quite new situations, they have adapted themselves, by natural selection, to new circumstances, changing the parts which required change—­the leaves and stalks; but keeping comparatively unchanged those which needed no change—­the flowers?

But I was not prepared, as I should have been had I studied my Griesbach’s West Indian Flora carefully enough beforehand, for the next proof of the wide distribution of water-plants.  For as I scratched and stumbled among the tussocks, ’larding the lean earth as I stalked along,’ my kind guide put into my hand, with something of an air of triumph, a little plant, which was—­there was no denying it—­none other than the long-leaved Sundew, {260a} with its clammy-haired paws full of dead flies, just as they would have been in any bog in Devonshire or in Hampshire, in Wales or in Scotland.  But how came it here?  And more, how has it spread, not only over the whole of Northern Europe, Canada, and the United States, but even as far south as Brazil?  Its being common to North America and Europe is not surprising.  It may belong to that comparatively ancient Flora which existed when there was land way between the two continents by way of Greenland, and the bison ranged from Russia to the Rocky Mountains.  But its presence within the Tropics is more probably explained by supposing that it, like the Bladder-worts, has been carried on the feet or in the crop of birds.

The Savanna itself, like those of Caroni and Piarco, offers, I suspect, a fresh proof that a branch of the Orinoco once ran along the foot of the northern mountains of Trinidad.