the little streamlet whose tiny ranges and stickles form the middle thread of this green combe in the dorset downs is just at present richly clad with varied foliage. tall spikes of the yellow flag rise above the slow-flowing pools, while purple loose-strife overhangs the bank, and bunches of the arrowhead stand high out of their watery home, just unfolding their pretty waxen white flowers to the air. in the rapids, on the other hand, i find the curious water crowfoot, a spray of which i have this moment pulled out of the stream and am now holding in my hand as i sit on the little stone bridge, with my legs dangling over the pool below, known to me as the undoubted residence of a pair of trout. it is a queer plant, this crowfoot, with its two distinct types of leaves, much cleft below and broad above; and i often wonder why so strange a phenomenon has attracted such very scant attention. but then we knew so little of life in any form till the day before yesterday that perhaps it is not surprising we should still have left so many odd problems quite untouched.

this problem of the shape of leaves certainly seems to me a most important one; and yet it has hardly been even recognised by our scientific pastors and masters. at best, mr. herbert spencer devotes to it a passing short chapter, or mr. darwin a stray sentence. the practice of classifying plants mainly by means of their flowers has given the flower a wholly factitious and overwrought importance. besides, flowers are so pretty, and we cultivate them so largely, with little regard to the leaves, that they have come to usurp almost the entire interest of botanists and horticulturists alike. darwinism itself has only heightened this exclusive interest by calling attention to the reciprocal relations which exist between the honey-bearing blossom and the fertilising insect, the bright-coloured petals and the myriad facets of the butterfly's eye. yet the leaf is after all the real plant, and the flower is but a sort of afterthought, an embryo colony set apart for the propagation of like plants in future. each leaf is in truth a separate individual organism, united with many others into a compound community, but possessing in full its own mouths and digestive organs, and carrying on its own life to a great extent independently of the rest. it may die without detriment to them; it may be lopped off with a few others as a cutting, and it continues its life-cycle quite unconcerned. an oak tree in full foliage is a magnificent group of such separate individuals—a whole nation in miniature: it may be compared to a branched coral polypedom covered with a thousand little insect workers, while each leaf answers rather to the separate polypes themselves. the leaves are even capable of producing new individuals by what they contribute to the buds on every branch; and the seeds which the tree as a whole produces are to be looked upon rather as the founders of fresh colonies, like the swarms of bees, than as fresh individuals alone. every plant community, in short, both adds new members to its own commonwealth, and sends off totally distinct germs to form new commonwealths elsewhere. thus the leaf is, in truth, the central reality of the whole plant, while the flower exists only for the sake of sending out a shipload of young emigrants every now and then to try their fortunes in some unknown soil.

the whole life-business of a leaf is, of course, to eat and grow, just as these same functions form the whole life-business of a caterpillar or a tadpole. but the way a plant eats, we all know, is by taking carbon and hydrogen from air and water under the influence of sunlight, and building them up into appropriate compounds in its own body. certain little green worms or convoluta have the same habit, and live for the most part cheaply off sunlight, making starch out of carbonic acid and water by means of their enclosed chlorophyll, exactly as if they were leaves. now, as this is what a leaf has to do, its form will almost entirely depend upon the way it is affected by sunlight and the elements around it—except, indeed, in so far as it may be called upon to perform other functions, such as those of defence or defiance. this crowfoot is a good example of the results produced by such agents. its lower leaves, which grow under water, are minutely subdivided into little branching lance-like segments; while its upper ones, which raise their heads above the surface, are broad and united, like the common crowfoot type. how am i to account for these peculiarities? i fancy somehow thus:—

plants which live habitually under water almost always have thin, long, pointed leaves, often thread-like or mere waving filaments. the reason for this is plain enough. gases are not very abundant in water, as it only holds in solution a limited quantity of oxygen and carbonic acid. both of these the plant needs, though in varying quantities: the carbon to build up its starch, and the oxygen to use up in its growth. accordingly, broad and large leaves would starve under water: there is not material enough diffused through it for them to make a living from. but small, long, waving leaves which can move up and down in the stream would manage to catch almost every passing particle of gaseous matter, and to utilise it under the influence of sunlight. hence all plants which live in fresh water, and especially all plants of higher rank, have necessarily acquired such a type of leaf. it is the only form in which growth can possibly take place under their circumstances. of course, however, the particular pattern of leaf depends largely upon the ancestral form. thus this crowfoot, even in its submerged leaves, preserves the general arrangement of ribs and leaflets common to the whole buttercup tribe. for the crowfoot family is a large and eminently adaptable race. some of them are larkspurs and similar queerly-shaped blossoms; others are columbines which hang their complicated bells on dry and rocky hillsides; but the larger part are buttercups or marsh marigolds which have simple cup-shaped flowers, and mostly frequent low and marshy ground. one of these typical crowfoots under stress of circumstances—inundation, or the like—took once upon a time to living pretty permanently in the water. as its native meadows grew deeper and deeper in flood it managed from year to year to assume a more nautical life. so, while its leaf necessarily remained in general structure a true crowfoot leaf, it was naturally compelled to split itself up into thinner and narrower segments, each of which grew out in the direction where it could find most stray carbon atoms, and most sunlight, without interference from its neighbours. this, i take it, was the origin of the much-divided lower leaves.

but a crowfoot could never live permanently under water. seaweeds and their like, which propagate by a kind of spores, may remain below the surface for ever; but flowering plants for the most part must come up to the open air to blossom. the sea-weeds are in the same position as fish, originally developed in the water and wholly adapted to it, whereas flowering plants are rather analogous to seals and whales, air-breathing creatures, whose ancestors lived on land, and who can themselves manage an aquatic existence only by frequent visits to the surface. so some flowering water-plants actually detach their male blossoms altogether, and let them float loose on the top of the water; while they send up their female flowers by means of a spiral coil, and draw them down again as soon as the wind or the fertilising insects have carried the pollen to its proper receptacle, so as to ripen their seeds at leisure beneath the pond. similarly, you may see the arrowhead and the water-lilies sending up their buds to open freely in the air, or loll at ease upon the surface of the stream. thus the crowfoot, too, cannot blossom to any purpose below the water; and as such among its ancestors as at first tried to do so must of course have failed in producing any seed, they and their kind have died out for ever; while only those lucky individuals whose chance lot it was to grow a little taller and weedier than the rest, and so overtop the stream, have handed down their race to our own time.

but as soon as the crowfoot finds itself above the level of the river, all the causes which made its leaf like those of other aquatic plants have ceased to operate. the new leaves which sprout in the air meet with abundance of carbon and sunlight on every side; and we know that plants grow fast just in proportion to the supply of carbon. they have pushed their way into an unoccupied field, and they may thrive apace without let or hindrance. so, instead of splitting up into little lance-like leaflets, they loll on the surface, and spread out broader and fuller, like the rest of their race. the leaf becomes at once a broad type of crowfoot leaf. even the ends of the submerged leaves, when any fall of the water in time of drought raises them above the level, have a tendency (as i have often noticed) to grow broader and fatter, with increased facilities for food; but when the whole leaf rises from the first to the top the inherited family instinct finds full play for its genius, and the blades fill out as naturally as well-bred pigs. the two types of leaf remind one much of gills and lungs respectively.

but above water, as below it, the crowfoot remains in principle a crowfoot still. the traditions of its race, acquired in damp marshy meadows, not actually under water, cling to it yet in spite of every change. born river and pond plants which rise to the surface, like the water-lily or the duck-weed, have broad floating leaves that contrast strongly with the waving filaments of wholly submerged species. they can find plenty of food everywhere, and as the sunlight falls flat upon them, they may as well spread out flat to catch the sunlight. no other elbowing plants overtop them and appropriate the rays, so compelling them to run up a useless waste of stem in order to pocket their fair share of the golden flood. moreover, they thus save the needless expense of a stout leaf-stalk, as the water supports their lolling leaves and blossoms; while the broad shade which they cast on the bottom below prevents the undue competition of other species. but the crowfoot, being by descent a kind of buttercup, has taken to the water for a few hundred generations only, while the water-lily's ancestors have been to the manner born for millions of years; and therefore it happens that the crowfoot is at heart but a meadow buttercup still. one glance at its simple little flower will show you that in a moment.