Now, plants are good shopkeepers; they lay themselves out strictly to attract their customers. Hence the character of the flowers on this beeless belt of mountain side is entirely determined by the character of the butterfly fertilisers. Only those plants which laid themselves out from time immemorial to suit the butterflies, in other words, have succeeded in the long run in the struggle for existence. So the butterfly-plants of the butterfly-zone are all strictly adapted to butterfly tastes and butterfly fancies. They are, for the most part, individually large and brilliantly coloured: they have lots of honey, often stored at the base of a deep and open bell which the long proboscis of the insect can easily penetrate: and they habitually grow close together in broad belts or patches, so that the colour of each reinforces and aids the colour of the others. It is this cumulative habit that accounts for the marked flowerbed or jam-tart character which everybody must have noticed in the high Alpine flora.

Aristocracies usually pride themselves on their antiquity: and the high life of the mountains is undeniably ancient. The plants and animals of the butterfly-zone belong to a special group which appears everywhere in Europe and America about the limit of snow, whether northward or upward. For example, I was pleased to note near the summit of Mount Washington (the highest peak in New Hampshire) that a large number of the flowers belonged to species well known on the open plains of Lapland and Finland. The plants of the High Alps are found also, as a rule, not only on the High Pyrenees, the Carpathians, the Scotch Grampians, and the Norwegian fjelds, but also round the Arctic Circle in Europe and America. They reappear at long distances where suitable conditions recur: they follow the snow-line as the snow-line recedes ever in summer higher north toward the pole or higher vertically toward the mountain summits. And this bespeaks in one way to the reasoning mind a very ancient ancestry. It shows they date back to a very old and cold epoch.

Let me give a single instance which strikingly illustrates the general principle. Near the top of Mount Washington, as aforesaid, lives to this day a little colony of very cold-loving and mountainous butterflies, which never descend below a couple of thousand feet from the wind-swept summit. Except just there, there are no more of there sort anywhere about: and as far as the butterflies themselves are aware, no others of their species exist on earth: they never have seen a single one of their kind, save of their own little colony. One might compare them with the Pitcairn Islanders in the South Seas—an isolated group of English origin, cut off by a vast distance from all their congeners in Europe or America. But if you go north some eight or nine hundred miles from New Hampshire to Labrador, at a certain point the same butterfly reappears, and spreads northward toward the pole in great abundance. Now, how did this little colony of chilly insects get separated from the main body, and islanded, as it were, on a remote mountain-top in far warmer New Hampshire?

The answer is, they were stranded there at the end of the Glacial epoch.

A couple of hundred thousand years ago or thereabouts—don't let us haggle, I beg of you, over a few casual centuries—the whole of northern Europe and America was covered from end to end, as everybody knows, by a sheet of solid ice, like the one which Frithiof Nansen crossed from sea to sea on his own account in Greenland. For many thousand years, with occasional warmer spells, that vast ice-sheet brooded, silent and grim, over the face of the two continents. Life was extinct as far south as the latitude of New York and London. No plant or animal survived the general freezing. Not a creature broke the monotony of that endless glacial desert. At last, as the celestial cycle came round in due season, fresh conditions supervened. Warmer weather set in, and the ice began to melt. Then the plants and animals of the sub-glacial district were pushed slowly northward by the warmth after the retreating ice-cap. As time went on, the climate of the plains got too hot to hold them. The summer was too much for the glacial types to endure. They remained only on the highest mountain peaks or close to the southern limit of eternal snow. In this way, every isolated range in either continent has its own little colony of arctic or glacial plants and animals, which still survive by themselves, unaffected by intercourse with their unknown and unsuspected fellow-creatures elsewhere.

Not only has the Glacial epoch left these organic traces of its existence, however; in some parts of New Hampshire, where the glaciers were unusually thick and deep, fragments of the primæval ice itself still remain on the spots where they were originally stranded. Among the shady glens of the white mountains there occur here and there great masses of ancient ice, the unmelted remnant of primæval glaciers; and one of these is so large that an artificial cave has been cleverly excavated in it, as an attraction for tourists, by the canny Yankee proprietor. Elsewhere the old ice-blocks are buried under the débris of moraine-stuff and alluvium, and are only accidentally discovered by the sinking of what are locally known as ice-wells. No existing conditions can account for the formation of such solid rocks of ice at such a depth in the soil. They are essentially glacier-like in origin and character: they result from the pressure of snow into a crystalline mass in a mountain valley: and they must have remained there unmelted ever since the close of the Glacial epoch, which, by Dr. Croll's calculations, must most probably have ceased to plague our earth some eighty thousand years ago. Modern America, however, has no respect for antiquity: and it is at present engaged in using up this palæocrystic deposit—this belated storehouse of prehistoric ice—in the manufacture of gin slings and brandy cocktails.

As one scales a mountain of moderate height—say seven or eight thousand feet—in a temperate climate, one is sure to be struck by the gradual diminution as one goes in the size of the trees, till at last they tail off into mere shrubs and bushes. This diminution—an old commonplace of tourists—is a marked characteristic of mountain plants, and it depends, of course, in the main upon the effect of cold, and of the wind in winter. Cold, however, is by far the more potent factor of the two, though it is the least often insisted upon: and this can be seen in a moment by anyone who remembers that trees shade off in just the self-same manner near the southern limit of permanent snow in the Arctic regions. And the way the cold acts is simply this: it nips off the young buds in spring in exposed situations, as the chilly sea-breeze does with coast plants, which, as we commonly but incorrectly say, are "blown sideways" from seaward.

Of course, the lower down one gets, and the nearer to the soil, the warmer the layer of air becomes, both because there is greater radiation, and because one can secure a little more shelter. So, very far north, and very near the snow-line on mountains, you always find the vegetation runs low and stunted. It takes advantage of every crack, every cranny in the rocks, every sunny little nook, every jutting point or wee promontory of shelter. And as the mountain plants have been accustomed for ages to the strenuous conditions of such cold and wind-swept situations, they have ended, of course, by adapting themselves to that station in life to which it has pleased the powers that be to call them. They grow quite naturally low and stumpy and rosette-shaped: they are compact of form and very hard of fibre: they present no surface of resistance to the wind in any way; rounded and boss-like, they seldom rise above the level of the rooks and stones, whose interstices they occupy. It is this combination of characters that makes mountain plants such favourites with florists: for they possess of themselves that close-grown habit and that rich profusion of clustered flowers which it is the grand object of the gardener by artificial selection to produce and encourage.

When one talks of the 'the limit of trees' on a mountain side, however, it must be remembered that the phrase is used in a strictly human or Pickwickian sense, and that it is only the size, not the type, of the vegetation that is really in question. For trees exist even on the highest hill-tops: only they have accommodated themselves to the exigencies of the situation. Smaller and ever smaller species have been developed by natural selection to suit the peculiarities of these inclement spots. Take, for example, the willow and poplar group. Nobody would deny that a weeping willow by an English river, or a Lombardy poplar in an Italian avenue, was as much of a true tree as an oak or a chestnut. But as one mounts towards the bare and wind-swept mountain heights one finds that the willows begin to grow downward gradually. The 'netted willow' of the Alps and Pyrenees, which shelters itself under the lee of little jutting rocks, attains the height of only a few inches; while the 'herbaceous willow,' common on all very high mountains in Western Europe, is a tiny creeping weed, which nobody would ever take for a forest tree by origin at all, unless he happened to see it in the catkin-bearing stage, when its true nature and history would become at once apparent to him.

Yet this little herb-like willow, one of the most northerly and hardy of European plants, is a true tree at heart none the less for all that. Soft and succulent as it looks in branch and leaf, you may yet count on it sometimes as many rings of annual growth as on a lordly Scotch fir-tree. But where? Why, underground. For see how cunning it is, this little stunted descendant of proud forest lords: hard-pressed by nature, it has learnt to make the best of its difficult and precarious position. It has a woody trunk at core, like all other trees; but this trunk never appears above the level of the soil: it creeps and roots underground in tortuous zigzags between the crags and boulders that lie strewn through its thin sheet of upland leaf-mould. By this simple plan the willow manages to get protection in winter, on the same principle as when we human gardeners lay down the stems of vines: only the willow remains laid down all the year and always. But in summer it sends up its short-lived herbaceous branches, covered with tiny green leaves, and ending at last in a single silky catkin. Yet between the great weeping willow and this last degraded mountain representative of the same primitive type, you can trace in Europe alone at least a dozen distinct intermediate forms, all well marked in their differences, and all progressively dwarfed by long stress of unfavourable conditions.