cess all the pipes now made at the Boston lead-works, and also at the New York works, are manufactured; and the public may rely on their producing a faultless water-pipe.

"It is also found that, by this improvement, the pipes can be made more rapidly than heretofore, and that a plug or compound cylinder for drawing may be cast thirteen inches long, which is a great advantage.

"A series of experiments were made to ascertain the action of distilled water and of Boston well-water on lead pipes, and on those made as first described, in which the tin had become alloyed with 22 per cent of lead; and it was found that a lead pipe in 24 hours yielded to distilled water 4.80 grs. of oxide of lead per gallon, and that a pipe lined with tin alloyed with 22 per cent of lead yielded to a gallon of the water 2.24 grs. of oxide of lead, while the pipe lined with pure tin yielded nothing to the water. When Boston well-water, containing 26 grs. of various saline matters per gallon, was substituted for distilled water, lead pipe, in 24 hours, yielded to it 2 grains of oxide of lead, and the pipe lined with tin alloyed with 22 per cent of lead yielded 1-865 grs. of lead to the water, the sulphates in the well-water protecting the lead to a considerable extent. The well-water of Waltham, which was very much less saline than Boston well-water, dissolved 0.8 gr. of oxide of lead from a pipe made of the tin alloyed with lead; this impregnation having taken place in a single night, or about 12 hours.

"It would seem from these researches, that the lead-encased tin pipe, as originally manufactured, is better than lead pipe, but is still objectionable as a water conduit, and much more so for more powerful solvents for lead, such as soda-water and beer; while the pipe, as now made, is as unobjectionable as pure block-tin pipe, and is actually cheaper than lead when purchased by the linear foot."

Five hundred and ninetieth Meeting.

January 29, 1868. STATUTE MEETING.

The PRESIDENT in the chair.

Professor August De La Rive was elected a Foreign Honorary Member in Class I. Section 3, in place of the late Michael Faraday.

Professor M. E. Chevreul was also elected a Foreign Honorary Member in Class I. Section 3.

Five hundred and ninety-first Meeting.

February 11, 1868. MONTHLY MEETING.

The PRESIDENT in the chair.

The following paper was presented and read by the author:

A Conjectural Solution of the Origin of the Classificatory System of Relationship.

By LEWIS H. MORGAN,

OF ROCHESTER, NEW YORK.

ABOUT twenty years ago I found among the Iroquois Indians of New York a system of relationship, for the designation and classification of kindred, both unique and extraordinary in its character, and wholly unlike any with which we are familiar. At the time I supposed it was a scheme devised by themselves, and confined to this particular stock of the American aborigines. Afterwards, in 1857, I had occasion to re-examine the subject, when the idea of its possible prevalence among other Indian nations suggested itself, together with its uses, in that event, for ethnological purposes. In the following summer I obtained the system of the Ojibwa Indians, of Lake Superior; and, although prepared in some measure for the result, it was with some degree of surprise that I found among them the same elaborate and complicated system which then existed among the Iroquois. Every term of relationship was radically different from the corresponding term in the Iroquois ; but the classification of kindred was the same. It was manifest that the two systems were identical in their radical characteristics. It seemed probable, also, that both were derived from a common source, since it was not suppo-able that two peoples, although speaking dialects of stock-languages, as widely separated as the Algonkin and Iroquois, could simultaneously have invented the same system, or derived it by borrowing one from the other.

From this fact of identity, several inferences at once presented

themselves. Its prevalence among these stocks rendered probable its prevalence among the remaining stocks of the American aborigines. If then it should, upon investigation, be found to be universal. among them, it would follow that the system was coeval, in point of time, with the commencement of their spread upon the American continent; and also, as a system transmitted with the blood, it might contain the necessary evidence to establish their unity of origin. And, in the next place, if the Indian family came in fact from Asia, that they must have brought the system with them from that continent, and have left it behind them among the people from whom they separated; and, further than this, that its perpetuation upon this continent would render probable its like perpetuation upon the Asiatic, where it might still be found; and, finally, that it might possibly furnish some evidence upon the question of the Asiatic origin of the Indian family.

Having found, before the close of 1859, that the system prevailed in the five principal Indian stock-languages east of the Rocky Mountains, as well as in several of the dialects of each, its universal spread through the Indian family had become extremely probable; and having also discovered traces of it.both in the Sandwich Islands and in South-India, it seemed advisable to prosecute the investigation upon a more extended scale, and to attempt to reach, as far as possible, all the families of mankind. This would require an extensive foreign correspondence, which a private individual could not hope to maintain successfully. I then applied to the Secretaries of the several American Boards of Foreign Missions for the co-operation of their respective missionaries in foreign fields, which was cordially promised, and the promise amply redeemed. I also applied to Professor Joseph Henry, Secretary of the Smithsonian Institution, for the use of the name of that institution to insure attention to the circular and schedule by means of which the system of relationship of the different nations was to be obtained. Professor Henry not only complied with this request, but also, at my suggestion, procured a circular to be issued by the Secretary of State of the United States to the diplomatic and consular representatives of the government in foreign countries, commending the investigation to their attention. From this time onward, the foreign correspondence, except with the missionaries, was conducted through the Smithsonian Institution and the Department of State.

In verification of the results it will be sufficient to state, that, by personal explorations, continued through several years, in the Lake

Superior region, in the Hudson's Bay Territory, and in the territories between the Missouri River and the Rocky Mountains, and by correspondence with government officials and private individuals in other parts of North America, I have been able to bring together the system of relationship of upwards of seventy Indian nations, speaking as many independent dialects. Beside these, and by means of the foreign correspondence referred to, the system of the principal nations of Europe and Asia, of a portion of those of Africa, of Central and South America, and of the Islands of the Pacific, have also been obtained. The tabulated schedules, now in course of publication by the Smithsonian Institution, will cover four hundred and fifty pages of the Smithsonian Contributions, and represent four fifths and upwards, numerically, of the entire human family. These strictly personal statements would be inappropriate in this connection, except as they become necessary to show that the solution about to be presented rests upon a wide basis of ascertained facts.

I propose to present, in a brief form, 1st. The system of relationship of the Aryan Family: using the Roman form as typical. 2d. That of the Malayan Family: using the Hawaiian form as typical. 3d. That of the Ganowanian Family: using the Seneca-Iroquois as typical. These are preliminary to the principal object, which is: 4th. To submit a conjectural solution of the origin of the classificatory system of relationship.

It may be premised that all of the systems of consanguinity and affinity, thus far ascertained, resolve themselves into two radically distinct forms, of which one will be called the descriptive, and the other the classificatory.

In the first, consanguinei are, in the main, described by a combination of the primary terms of relationship. There is a small amount of classification, by means of special or secondary terms introduced by civilians and scholars to relieve the burdensomeness of the system; but the great body of relatives, both by blood and marriage, are described. This is the system of the Aryan, Semitic, and Uralian families. In its origin, as the parent of the present form, it was purely descriptive, as is still exemplified by the Erse and Scandinavian, and by the condition of the Sanskritic, when this language ceased to be spoken. This system follows the streams of the blood, and is in

* Gä-no-wä'-ni-an: name proposed for the American Indian family. From Gä'-no, an arrow, and Wä-ă'-no, a bow; the family of the Bow and Arrow.

accordance with the nature of descents. It is, therefore, a natural system, for the reason that the relationships recognized are those which actually exist. But it assumes as its fundamental basis the antecedent existence of marriage between single pairs. Before this system could come into existence, mankind must have raised themselves to this state of marriage; after which this form of marriage, and not nature, teaches the descriptive system of relationship. It is important that this distinction should be noted.

In the second form, consanguinei are never described by a combination of the primary terms; but they are classified into categories, and the same term of relationship is applied, without distinction, to each of the members of the same category. This is the system of the Malayan, Ganowanian, and Turanian families. It suggests the probability that there might have been a state of society in the primitive ages in which marriage between single pairs was unknown, in which the family, in its modern sense, was unknown; but in which a system of relationship might have originated in compound marriages in a communal family, and thus be in strict accordance with the nature of descents, and, therefore a natural system because it recognized the relationships actually existing. This suggestion should also be noted.

1. System of Relationship of the Aryan Family.

A knowledge of the descriptive system became important for two principal reasons. First, it was necessary to find the limits of its spread to circumscribe the classificatory form: and, secondly, it was necessary to find the basis upon which it rested, to reach the instrumentalities by means of which the classificatory system, if it ever prevailed among the remote ancestors of the Aryan family, might possibly have been overthrown, and the descriptive substituted in its place.

comment.

As none of the characteristics of the former system are involved in the solution of the origin of the latter, it will be sufficient for my present purpose to present the substance of the Aryan form without The Roman, as found in the Pandects* and Institutes of Justinian, will be used as the typical system. Its completeness and perfection is due to the Roman civilians, and arose from a necessity for a code of descents, defining the relations of consanguinei to each other, to regulate the transmission of property by inheritance.

*Pand. Lib. XXXVIII. Tit. X. "De gradibus et adfinibus et nominibus eorum."

Just. Inst. Lib. III. Tit. VI. "De gradibus Cognationum."