and a wee forked tongue protruded with lightning speed to probe the inmost recesses of its nectar-bearing tube. Honey and insects the humming-birds suck hastily from the blossoms they visit; for they are really by origin insect-eating birds, and their nearest relatives in the northern hemisphere are our own dingy and inconspicuous little churchhaunting and fly-catching English swifts.

I never killed a humming-bird myself I never had the heart to do it: but a friend who used to come and stay with me, often brought his murderous scientific air-gun, and, sitting leisurely in his bamboo chair under the verandah, used to shoot my pretty visitors with an infinitesimal charge of fine sand, "in the interests of science.' When once they were really dead, however killed, there was no cause or just impediment why one should not dissect them: and dissection very soon showed what were the reasons which had induced modern ornithologists, like Mr. A. R. Wallace, to separate the humming-birds in their systematic classifications from the closely similar and brilliant sun-birds, and to place them next in order to the very unlike and commonplace northern swifts. The story of the strange cross-relationship between these three groups of birds and the English swallows is so curious a one, and it throws so much unexpected side-light on the queerly deceptive results often produced by evolution in the animal world, that we cannot do amiss in devoting half an hour to clearing up this odd bit of natural genealogical mystification.

Very few people in England, I suppose, really know the difference between swallows and swifts. We have some three kinds of true swallows in Britain, more or less familiar in our fields and meadows, from the common fork-tailed swallow that skims and curves along the ground in damp weather in pursuit of its flitting insect prey, to the rarer and prettier tiny sand-martins whose deep gallery nests honeycomb the soft sandy banks along the roadsides of our southern counties. But besides these familiar longwinged birds there is another common creature in England, extremely like them in shape, length of wing, and habit of flight; and that is the equally migratory European swift. The swifts come later than the true swallows to this country, and they go away earlier yet hardly anybody except the regular naturalists ever notices, in all probability, the underlying difference between the two creatures. Most people, as they see the dainty agile birds swooping in long curves around the church steeples where

:

they love to make their airy home, take them merely for large swallows, though they are really hardly any bigger in the body, and owe their deceptive appearance of great size entirely to the immense and disproportionate length of the wing pinions. Yet the swifts, in spite of their close resemblance to the swallow type, must in fact be reckoned as members of a totally different order of birds, and the similarity between them is due, not to any original identity of ancestral descent, but to like modifications of outer shape in adaptation to a like mode. of life.

On the other hand, though swallows and swifts, which so closely imitate one another in every external particular, are not at all related in genealogical order; humming-birds. and swifts, which differ so immensely from one another in every external particular, are very nearly related indeed. But, as if to complete this queer natural family puzzle, there are in India and other Asiatic countries, certain pretty little flower-haunting thrushes, accurately known as the sun-birds, which are so ridiculously like the true humming-birds in shape, size, colour, decoration, and flitting habits that they are universally described as humming-birds by all old Anglo-Indians and Eastern travellers outside the restricted naturalist interest. Let us

see if we can discover how this odd set of cross-relationships and imitative resemblances has arisen in the simple course of the evolutionary process.

The birds with the history of whose development we have here to deal, all belong to one of two great natural orders, with whose technical peculiarities and distinctions I don't think we need trouble ourselves in any detail to-day. One of these is the order of the perchers or sparrow-like birds, in which also our own thrushes, blackbirds, crows, larks, finches, and wagtails are included. The other is the order of the woodpeckers, in which must be reckoned our English cuckoos and nightjars, as well as the parrots, toucans, hornbills, trogons, and plantaineaters of the tropics. The swallows and sun birds belong to the first of these two great natural divisions: the swifts and hummingbirds belong to the second. They may be said to represent one another in the two orders, or, in other words, to be similar modifications of unlike ancestral types, specialised for filling the same places in the economy of nature.

And now first let us look at the two duskier and more northern groups, the swifts and swallows. Among the finch-like

perching birds there is one considerable division which has taken entirely to living upon the smaller winged insects, and to the habit of catching them while actually on the wing, instead of hunting for them as they settle lightly upon the ground or the foliage of trees. This division is that of the common

European swallows and martins. As a consequence, these birds have acquired certain noticeable general peculiarities of form and structure which fit them for their chosen mode of life-peculiarities partly due to increased use of special limbs or muscles, and partly to the constant survival of the best adapted individuals in the constant struggle for existence which everywhere goes on between the members of the same species. From time immemorial, ever since the ancestors of the swallows first took to hawking after flies in the open sunshine, the longestwinged and most powerful swallow must always have secured the largest amount of food for himself and his callow nestlings. As a consequence, such longest-winged birds have always best survived in their own persons and those of their descendants, while their shorter-winged rivals have constantly died out by being distanced in the continuous competition for food and offspring. All the other peculiarities of the swallows follow as a matter of course from these prime features of the primitive race. The body is small and very buoyant; the bones of the pinions are specially shaped for extended flight; the

beak is broad and short, so as to snap easily at the prey which the bird pursues openmouthed through the air; the whole system is in every respect designed for that rapid skimming motion over the grass of meadows which is so characteristic of the entire family. As the swallow hardly ever perches on the ground, its feet are naturally small, and its walking powers extremely feeble. Its insect food being abundant in the north during the summer only, it is necessarily also a migratory bird, and it passes the winter months in Algeria or Central Africa, returning to England, like other winter tourists, about April or May. Altogether it may be looked upon as the very type of a rapidly-flying, insect-hunting modification of the original sparrow form.

Now, among the totally unlike woodpecker group of birds, there is also a family which has taken quite independently to just the very selfsame mode of life, by catching midges and small mayflies as they hover lightly poised above grass or water. This family is that of the beautiful and graceful English swifts. There are other members of the woodpecker group which have adopted to some extent similar habits, especially the family of the goatsuckers or mosquitohawks, including our own nightjars and the American whip-poor-wills-nocturnal birds, which prowl after the insects that love the dusk, and have so acquired comparatively

long wings and considerable power of pinion. But in these respects the swifts far surpass them, excelling even the swallows themselves in their marvellous breadth of wing and sustained endurance in the air. The same bones which have grown long in the swallow have for the same reason grown still longer in the swift: there is the same broad short beak, the same smallness and feebleness of the feet, the same habit of living almost entirely on the wing, the same migratory instinct, necessitated by the similar annual deficiency of food. Even the practice of building round about habitations is common to both races, for though the swift is the better ecclesiologist of the two kinds, loving to perch his nest under the tall pinnacles of some cathedral steeple, yet the swallow, too, is fond of churches, and neither bird wholly despises the shelter of overhanging cottage eaves. So like, in fact, have the two kinds become by this independent modification to similar circumstances, that they were long counted by earlier naturalists as members of a single great family. As our modern evolutionary biologists put it, the two groups, though originally quite distinct, have converged in external adaptive characters. How, then, are we able now to discover their primitive ancestral unlikeness? Simply by certain deep-seated unlikenesses in certain unobtrusive but structurally important anatomical points. Into the details of these points (which are by no means attractive to the untechnical mind) it is not necessary here to enter fully: it must suffice to say that in five main particulars the

swift differs essentially from the swallows, and agrees with the woodpecker group, and more especially with its own near relations, the humming-birds: and these five particulars are the shape and outline of the breastbone, the distribution of the feathers on the skin, the arrangement of the muscles of the toes, the number of the tail feathers, and the want

of a vocal organ on the windpipe. Such deep-seated points of structure serve as clues to the real ancestral relationships of animals: the mere external likenesses often depend only (as in this case) on similarity of accidental circumstances and mode of life.

When we pass on to consider the respective development of the sun-birds and

have taken to the habit of eating the honey of tropical flowers, which they extract from the nectaries at the base of the long funnel-shaped tubes so common in big showy southern blossoms. It is easy enough to see how a family of insect-eating birds, inhabiting the forests and jungles of Africa and India, might readily undergo this simple change of taste and habit. Large numbers of insects are always to be found searching for honey around the mouths of brilliant scarlet or crimson flowers; and it was the insects, not the nectar, in the first place, that the ancestral sun-birds originally took to collecting from the base of the blossoms. Naturally enough, they would learn to hunt for them where they were most abundant: and it is a significant fact that the nearest relatives of the sun-birds, such as our own English creepers, feed entirely off an unmixed insect diet. Indeed, the sun-birds themselves are still to a great extent insectivorous: for

though they now feed largely on honey, they take their honey a good deal mixed with casual weevils or flies, and in confinement they do not refuse small worms, ants' eggs, and bits of bread sweetened to their taste in sugar and water. Still, they have naturally undergone a great deal of modification in adaptation to their acquired habits of flowerhaunting and honey-eating. The bill has become very long and slender with a curved point, so as to probe the inmost recesses of the long trumpet-shaped Indian flowers; the tongue has grown extremely long and extensile, so that it can be rapidly darted out to ransack the deep nectary and catch the insects that flutter in its honeyed depths: the body and wings have been fitted not for swooping