represents a large population of indolent colored people), the values of agricultural products per acre of improved land cohere with the death-rates from nervous diseases and the values of manufactured products and oppose the birth-rate. On the other hand, the agricultural values per capita, with the exception of the South Atlantic division, cohere with the birth-rates and oppose the death-rates from nervous diseases and the values of manufactured products per capita.

In order still further to verify the conclusion that the birth-rate and the death-rate from nervous diseases are usually opposed, comparisons have been made from other available statistics. A study of the State of Massachusetts by counties shows that in 1885, in ten counties, the birthrate per thousand women between the ages of fourteen and forty-nine opposed the death-rate from nervous diseases (Table X.). This is practically the same result as that obtained from the United States census figures.

A comparison of the birth-rate with the density of population per square mile gives a result very different from that of the United States census statistics. In eight of the fourteen counties, the birth-rate and the density cohere, in only six do they oppose each other; and in nine of the fourteen counties the density and the deaths from nervous diseases are opposed. Another unexpected result is found in Table XI. The birth-rate in the cities of Massachusetts since 1870 has been higher than in the rest of the State. These facts, which seem to be contrary to the results obtained for the United States as a whole, probably may be accounted for (1) by the peculiar race conditions in Massachusetts; and (2) because in the cities there is a large proportion of population between the ages of fourteen and forty-nine.

The rural population in Massachusetts consists of the old New England stock which is slowly dying out; the cities have a large Irish and French Canadian population, which is very prolific and, as statistics prove, less subject to nervous diseases than the native population. These peculiar

conditions in Massachusetts are anomalous and deserve to be the subject of a separate investigation. The larger proportion of population between the ages of fourteen and fortynine, which probably is the cause of the higher marriagerate in the cities (Table XI.), must be an important factor in increasing the birth-rate.

TABLE X.

BIRTH-RATES IN MASSACHUSETTS WITH COMPARISONS, 1885.

BIRTH-RATES AND MARRIAGE-RATES IN MASSACHUSETTS, 1890.

Taken from the Registration Report of Massachusetts for 1890 (pp. 372-373).

Density.

The following conclusions may therefore be drawn from the preceding study:

1. Whether or not it be true that the means spoken of by Dr. Billings, M. Dumont, M. Levasseur, and Dr. Edson has become an important factor in the diminishing birth-rate of civilized countries, it is evident that it is not the only factor, and that, quite apart from voluntary prevention, there is a distinct problem to be investigated. This is shown by the fact that the white and the colored birth-rate vary together.

2. Mr. Spencer's generalization that the birth-rate diminishes as the rate of individual evolution increases is confirmed by a comparison of the birth-rates with the deathrates from nervous diseases, and also with the density of population, the values of agricultural and manufactured products, and the mortgage indebtedness.

3. The Malthusian theory in general, that population tends to increase faster than the means of subsistence, is not true of the United States at the present time. In the regions where wealth increases most rapidly, the population increases most slowly.

It is hoped that this study may be continued when the full statistics for 1890 are published, unless the work is done by the census office, and that ultimately a more complete investigation, on a different basis, may be made by taking statistics from the registration reports of several States and making the comparisons by counties and townships.

Bryn Mawr College.

J. L. BROWNELL.

RENT AND PROFIT.

Not a little of the confusion in recent economic literature would seem to be due to the attempt to force the new wine of more modern concepts into the old bottles of Ricardian dicta. This is nowhere more evident than in the varied and conflicting duties which of late we have imposed upon the term "rent."

To the Ricardian school of economics the price of every product contained two elements, "cost" and "rent." By the former, they understood the cost at the margin of production; by the latter, the surplus obtained by those enjoying special advantages in the production of any commodity-the differential surplus-as it is sometimes called. They also held, that while the first enters into the determination of price, the second is a surplus that is determined by price.

To these two concepts recent literature has added a third, namely, a surplus which does enter into the determination of price; or, as it is usually stated, "there is a marginal surplus."

Some foreshadowing of this new concept may be found as early as 1829. And yet not a little confusion may still be found in the writings of its strongest advocates, due to the fact that they continue to include both the "price-determined" and the "price-determining surplus," under the one term "rent."

The old contention, that cost determines exchange value, seems to involve the assumption that there is no surplus at the margin of production; or, as it is sometimes stated, "there is always some no-rent land." This assumption, however, was not accepted without protest, even by the adherents of the Ricardian school.

J. S. Mill files exceptions to it, all through chapters 4, 5 and 6, Book III, and in the latter sums up as follows: "Rent

is not an element in the cost of production of the commodity which yields it, except in the case (rather conceivable than actually existing) in which it results from and represents a scarcity value. But when land capable of yielding rent in agriculture is applied to some other purpose, the rent which it would have yielded is an element in the cost of production of the commodity which it is employed to produce." It is manifest that we have here a recognition of the "marginal" or "price-determining surplus." *

Professor S. N. Patten, in his "Premises of Political Economy," gives us possibly the fullest statement of this phase of the question, and holds that the contention of a no-rent land fails on five different counts: "First, to obtain uncultivated land for tillage, farmers must compete with those who can afford to pay rent for uncultivated land by using it for pasture, for wood and many similar purposes. For this reason the poorest land in cultivation must pay rent, since if the farmers will not pay rent, the landlords would let it to herders and others who could afford to give much for the use of uncultivated land."

It will hardly be necessary to follow Professor Patten through his other four counts; the fundamental thought running through these, as through all the protests against the contention of a no-rent land, is in simple, as follows: If certain lands or farms, however much they vary among themselves in fertility and distance from market, are yet all of them distinctly superior to all other lands for the production of a certain brand of wine, and the supply of such land is relatively limited, two forms of surplus may arise. The variations in fertility and distance from market, within the the group, will give rise to a "price-determined surplus "— the old Ricardian rent. So long as all this land is specially efficient in the growing of this wine, and the supply of this.

*The German literature on this point is quite interesting, especially Nebenius, 1829, and Herman, 1832. For a fuller treatment, see the writer's "History of the General Doctrine of Rent in German Economics."