capital handed on to them by the mother-plant. But what will feed a seedling will feed an animal as well; and it is just these rich little beans in the clover-pod which give it all its dangerous value as a fodder for cattle. Hence, in the wild state those clovers which have their seeds least protected are most likely to be eaten off and killed down by birds or animals, while those which have them most protected are most likely to survive and become the parents of future generations. Here, then, we have the basis upon which natural selection can act in differentiating the primitive ancestral clover into various divergent species. Whatever accidental variation happens to give any particular clover protection for its seeds in any special habitat will certainly be preserved and increased, while all opposite variations will be cut off and demolished at once. So far as their foliage and their flowers are concerned, the clovers as a body are practically in a state of stable equilibrium; so far as their fruit and seeds are concerned, they are still undergoing modification by natural selection.
Clearly to illustrate this fundamental point, let us first look at some neighboring and closely allied plants, which are not exactly clovers, but which resemble them in almost all important particulars. These also show the same devices for specially protecting their seeds and pods from birds or animals. Take, for example, the genus of the medics. These are mostly small greensward plants, with trefoil leaflets like the clovers, but with the flowers in rather tall, one-sided spikes or loose bunches. Their pods are usually long and many-seeded, but they have this curious peculiarity, that instead of growing straight like that of a pea or bean, they coil up spirally like a snail-shell. When ripe they fall off the plant entire, and thus defeat the hopes of birds and other creatures which wait patiently for the opening of the pods. The simpler medics, such as the agricultural lucern, have smooth, spiral pods alone, and therefore they can be employed successfully as fodder for cattle. But this, which proves an advantage from the point of view of the farmer, is naturally a disadvantage from the point of view of the plant in a wild state, because it insures the seeds being eaten; and hence the more developed and weedy medics have acquired stout protective prickles, fringing their globular spirals, and making them very distasteful morsels to cows or horses. We have two such prickly medics in England, one closely coiled and rolled round like a ball, and thickly set with curved hooks; the other loose like a corkscrew, with two rows of sharp bristles at the adjacent edges; and both these, as I learn from farmers, are extremely objectionable weeds in meadows, rendering the bay almost uneatable. Indeed, I am assured that cattle will never touch even fresh meadow-grass containing them except when absolutely driven by hunger. It is noteworthy that our two doubtfully native smooth medics (lucern and nonesuch) both grow naturally in rough, dry places, and are only largely found as "artificial grasses"—that is to say, were introduced and maintained
