570 TRANSACTIONS OF SECTION F. 
I should be supposed to mean that what you pay for a certain piece of land 
as rent represents the superiority of that piece of land to another that you can 
get for nothing. In this use of the word everything depends upon the different 
quality of the things compared. But what I want is a word which shall always 
carry the underlying assumption that we are considering the expansion and 
contraction of a homogeneous supply, the ‘differential’ value of that supply 
being a function of its breadth or magnitude. 
Again, the same theory of rent which regards it as a differential charge, in 
the sense of a charge due to an inherent difference of quality in the things 
charged for, assumes that there is some land which bears no rent at all. This 
is the land on the ‘margin’ of cultivation. Hence ‘marginal’ has come to be 
used in economic literature to signify the lowest grade or quality of any com- 
modity, or service, or the least favourable set of conditions, that just hold their 
footing in any industry. Thus the marginal land would mean the worst land 
under cultivation, the marginal workman the least efficient man in actual 
employment, the marginal conditions of an industry the least advantageous 
conditions under which it is actually conducted, and, I suppose, the marginal 
grade of potatoes or wheat the worst quality actually in the market; or to 
the hungry individual the marginal mouthful of beef would be the one just 
not rejected and left on the plate because too largely composed of ‘ veins’ to 
be eaten, even if no more of any kind were to be had. 
Now attempts have been made to erect a theory of distribution upon the 
consideration of ‘margins’ in this sense. The ‘ marginal’ man, working on the 
‘marginal’ land, under the ‘marginal’ conditions, and with the ‘ marginal ’ 
appliances, is taken as the ultimate basis of the pile, and wages, rent and 
interest are explained as ‘ differential’ in their nature; that is to say, as due 
to the superiority in quality, position, or point of application, of such-and- 
such work, land, or apparatus, over the ‘marginal’ specimens. 
I do not stay to examine this theory on its merits; but it is necessary to 
insist on the almost incredible fact that there is constant confusion between 
it and what I have tried to expound as the ‘differential’ theory of distribu- 
tion, simply because they can both be described as ‘ marginal,’ and the term 
‘differential,’ though in quite divergent senses, may be introduced in the 
exposition of either. 
Once again, then, if I speak of the differential or marginal significance of 
my supply of bread and milk, and say that it depends, ceteris paribus, upon how 
many loaves of bread and how many pints of milk I take, I am supposing all 
the bread and milk to be of the same quality. And if I speak of the differen- 
tial or marginal significance of labour in a particular industry, I am either 
speaking of a uniform grade of labour or of different grades reduced to some 
common measure and expressed in one and the same unit, and I mean the 
significance which such a unit has when it is one out of so many others like 
itself. Thus, in my use of the word, there is no earmarked marginal unit, 
which is such in virtue of its special quality. Any one of 100 units has 
exactly the same marginal value; but as soon as one unit is withdrawn, all 
the remaining 99 have a higher marginal value; and when one is added, all 
the 101 a lower. 
The only word I can think of free from misleading associations would be 
“quotal’; for guotus means (amongst other things) ‘ one out of how many,’ and 
so quotal significance might mean the significance which a unit has when 
associated with such-and-such a number of others homogeneous with itself. 
And now let us turn again to the further consideration of the changes of 
method and the purgings which seem inevitably to follow upon the full recog- 
nition of the differential principle in economics. Severe selection and limitation 
is, of course, necessary, and I think we cannot do better than take up a few of 
the current phrases or conceptions and diagrammatic illustrations connected 
with the phenomenon of rent. Antecedently we must expect that as there 
is no theoretical difference between the part played by land and that played 
by other factors of production (or more direct ministrants to enjoyment), so 
there can be no general assertion about rent and land which is at once true 
and distinctive; for if true it must be based on that aspect of land which 
expresses its function in a unit common, say, to capital, and which brings its 
