[Footnote 1: _Principles of Political Economy_, by John Stuart Mill.]
§3. _The Differential Aspect_. Most ordinary commodities are not produced on a single, uniform pattern. As a rule there are many variations of grade and quality, and consequently of price. But these variations are usually designed to meet the differences of taste among the purchasers, and we do not expect to find that any variety of an ordinary commodity will be produced, which is so poor in quality as to be entirely valueless. But since it is nature which has produced the land, without any assistance or guidance from man, there are many pieces of land which are so unfertile, or are otherwise so unsuitable for productive purposes, as to be quite valueless from the economic standpoint. Even in a densely populated country like Great Britain, there are considerable tracts of land which it is unprofitable to employ for any economic purpose whatsoever, and which possess no further value than what the mere pride of ownership may give them. This fact makes it possible to apply the conception of the margin to the case of land with particularly illuminating results.
In the first place, however, it should be observed that the value of any piece of land does not depend solely on the intrinsic fertility of the soil. The fact that land is an immobile thing makes its _situation_ a factor of great importance. In the case of urban land, situation is, of course, the only thing that counts. The value of a site in Bond Street or the City is entirely unaffected by its capacity or incapacity for potato-growing purposes. But even for agricultural land, situation is a most important matter. A farm, which is so remote that considerable transport charges must be incurred to bring its produce to market, will be less sought after, and less valuable, than one which is much better situated though somewhat less fertile. In what follows, therefore, we must speak of the "quality" of a piece of land in a broad sense to include advantages of situation, as well as of fertility. Let us now, imagine the different pieces of land in Great Britain to be arranged in order of quality, so that we have a long series, with land of the best quality at one end, and of the poorest quality at the other. At the latter end, we will have such land as is found near the top of Snowden or Ben Nevis, which it clearly does not pay to cultivate at all. Somewhere, then, between these two extremes, we shall come to a point where the land is just, but only just, worth cultivating, or where, to revert to a form of words we previously employed, it is a matter of _doubt_, whether the land is really worth using for a productive purpose. Such land we can regard as the "marginal land"; and since the variety of nature is at once infinite and fairly minutely graduated we shall probably find that on one side of this margin there is much land which is only slightly superior, and on the other, much which is only slightly inferior, to the marginal land itself. What, then, is likely to be the value and the rent of this marginal land, this land which is just on the "margin of cultivation"? Some readers may find the answer startling. The rent of the marginal land will be nil, because it will not pay to cultivate it, if any appreciable rent is charged. A piece of land for which it is worth a tenant's while to pay an appreciable rent, will not be the marginal land, because there will be land just slightly inferior to it which it will also pay to cultivate if a somewhat lower rent is charged. And so we can pass to poorer and poorer qualities of land, with an ever diminishing rent, until at the margin of cultivation the derived utility of the land is negligible and the rent vanishes.