OF THE HUMAN FAMILY. 489 



111 the coinmun:il family, consisting of several brothers and sisters, and their 

 children, the family in its first stage is recognized. 



IV. The Hawaiian Custom. 



The existence of this custom is not necessary to an explanation of the origin of 

 the Malayan system. All it contains bearing upon this question is found in the 

 intermarriage of brothers and sisters, where the brothers live in polygynia, and the 

 sisters in polyandria ; but it holds a material position in the series, for the reason 

 that it was an existing and still ])revalent custom in the Sandwich Islands at the 

 epoch of their discovery. It finds its type in the previous custom out of which it 

 naturally arose, and for Avhich reason it may be expected that it will yet be found 

 in other barbarous nations. So far as it brought unrelated persons into the house- 

 its mechanism. Mr. Caleb Swan, who visited the Creeks in 1190, thus describes their houses : 

 "These houses stand in clusters of four, five, six, seven, and eight together, . . . each cluster of 

 houses containing a clan or family of relatives, who eat and live in common" (Schoolcraft, Hist. 

 Cond. and Pros. Ind. Tribes, 5, 262.) Lewis and Clarke thus speak of a village of the Chopunnish 

 (Xez Perces) in the valley of the Columbia. (Travels, Lond. ed., 1814, p. 548.) "The village of 

 Tumacheraootool is in fact only a single house one hundred and fifty feet long. ... It contains 

 twenty-four fires, about double that number of families, and might, perhaps, muster one hundred 

 fighting men." In like manner the Dirt Lodge of the Mandans and Minnitares is a communal 

 house, about forty feet in diameter, and polygonal in form, and capable of accommodating seven or 

 eight families. It is comparted with willow screens ; each apartment being open towards the fire- 

 pit in the centre. These specimens illustrate the principle. If we now turn to the architecture of 

 the Village Indians of New Mexico, Mexico, Chiapa, and Yucatan, it will be found that their houses 

 were great communal edifices, constructed of adobe brick, or of rubble stone and mud mortar, or of 

 slate stone, or of stone fractured or cut, and laid with mortar, possibly in some cases of lime and 

 sand. The pueblo of Taos, in New Mexico, consists of two such houses, one of which is 260 feet 

 long, 100 feet deep, and five stories high, the stories being in the retreating or terrace form ; and 

 the second is 140 feet long, 220 feet deep, and six stories high. They are built of adobe brick, and 

 each capable of accommodating about four hundred persons. They are now occupied by 361 Taos 

 Indians. In the canon of the Rio de Chaco, /about one hundred and forty miles northwest of Santa 

 Fe, there is a remarkable group of some seven pueblos, now in ruins (they answer very well to the 

 seven cities of Cibola), constructed of stone, a thin tabular limestone. That of Ilungo Pavie is 

 built on three sides of a court, is 300 feet long, by 130 deep on the two sides, and three stories high. 

 It contained 144 chambers, each about 15 by 18 feet, and would accommodate seven or eight hundred 

 persons. It was built in the terraced form, the stories retreating from the court backward, and the 

 court was protected by a low stone wall. If this communal edifice is compared with the so-called 

 palaces of Mexico, as they are imperfectly described by the early Spanish writers, a very satisfactory 

 explanation of the latter will be found in the former, and the reason why the communal houses of 

 Mexico were mistaken for palaces will also be made apparent. By the light of the same testimony 

 the so-called palaces of Palenque, Uxmal, and Chi-Chen-Itza fade away into communal houses, 

 crowded with Indians throughout all their apartments.* 



» In an article upon the " Seven Cities of Cibola," published in the April number of the Xorth American Review 

 for 1869, I pointed out, with some minuteness of detail, the characteristics of the architecture of the Village 

 Indians ; and in two subsequent articles in the same Beview, published in the October number, 18C9, and in the 

 January number, 1870, I treated at length the subject of " Indian Migrations." The latter was considered under 

 three principal divisions: First, the influence of physical causes, including the geographical features of North 

 America, and the natural subsistence afiforded by its different areas ; second, the influence of Indian agriculture ; 

 and third, their known migrations, together with such as might be inferred to have occurred from the relations in 

 which the several Indian stocks were found. These articles form a proper supplement to Part II., and this reference 

 is made to them as such. 

 62 April, 1870. 



