Early in the spring of 1901 the writer was ordered into the field to conduct ethnological and archeological investigations in northeastern Arizona. (See Plate 1.) The plan settled upon embraced two distinct explorations, the first during the month of May, for the United States National Museum alone, and the second from June 1 to August 30, for the Museum in conjunction with Mr. Peter Goddard Gates, of Pasadena, California, whose interest in the exploration of the Southwest has been productive of excellent results for science.
Field work began on May 3, and making Holbrook, Arizona, the base, the McDonalds Canyon ruins to the southeast of that place were visited and explored. The remainder of the month was spent at the Canyon Butte ruins east of Holbrook in a thorough reconnoissance of the Petrified Forest Reserve and a visit to the ruins north of Holbrook. These groups of ruins are new to science, and the results of the explorations are very satisfactory.
On June 1 the Museum-Gates expedition took the field, selecting for exploration a large ruin a few miles east of the Petrified Forest. On the completion of this work the party returned to Holbrook and went south into the White Mountains, reaching, on June 19, the ruin at Linden. On June 26 the party camped on the great Forestdale ruin on the White Mountain Apache Reservation. On July 9 a small ruin at Interior Sawmill was investigated, and after a visit to Fort Apache the expedition returned to Showlow, working for a day or two a large ruin on the ranch of Mr. Henry Huning. Returning north, ruins at Shumway, Snowflake, and Woodruff were examined, Holbrook being reached on July 17.
July 29 found the party engaged in excavating a large ruin called Kokopnyama, on the Jettyto Wash, 2 miles east of Keams Canyon. On August 11 a ruin near Jettyto Spring called Kawaiokuh was worked for a week, when the party closed excavation and proceeded to the Hopi pueblos for ethnological studies, remaining there till the 28th, when the Museum-Gates expedition disbanded.
The writer returned to Holbrook to complete the packing and shipping of the specimens secured. During September 12–14 the groups of ruins on the Le Roux and Cottonwood washes were carefully platted and plans of the sites made. This closed up the season, and on September 23 the writer returned to Washington.
In addition to the avowed objects of the expedition, collections of plants, fossils, minerals, etc., were made, Mrs. Gates aiding materially in the botanical work. A large series of photographs was made by Messrs. Vroman, Gates, and the writer.
The groups of ruins described in this paper are treated geographically, beginning, for convenience, with the southernmost, at Forestdale. Taking the more important sites in order to the northward, we have Forestdale, Linden, Showlow, Shumway, McDonalds Canyon, Scorse Ranch, Petrified Forest Reserve, Biddahoochee, and Jettyto Valley. This line of archeological reconnoissance shows in an interesting way the prevalence of red and gray pottery south of the Little Colorado and Puerco rivers, with exceptions at Shumway and Stone Axe, gray ware in the Little Colorado Valley, and yellow ware at Biddahoochee and Jettyto Valley. Thus we may divide the field explored into three regions, namely: (1) Region of the White Mountains, red and gray ware; (2) region of the Little Colorado Valley, gray and red ware, and (3) region of the Hopi buttes and mesas, yellow and little red and gray ware. In detail the ruins examined in the region of the White Mountains are Forestdale, Interior Sawmill, Linden, Showlow, Shumway (yellow and red), Snowflake, Woodruff Butte, Canyon Butte, Petrified Forest, Metate ruin, Stone Axe ruin (yellow ware), and Adamana. Those of the Little Colorado Valley are McDonalds Canyon and Scorse Ranch, and those of the Hopi buttes and mesas are Biddahoochee and Jettyto Valley.
The environment of the three regions is semiarid. The White Mountain region, however, from the height and mass of the range, especially the Mount Thomas condensing focus, has greater rainfall than the other regions. For this reason there is here abundant vegetation, and in the radius of this influence and in this respect the environment seems more favorable for human habitation. On the other hand, geological causes have determined the lack of springs on the north side of the range, and dependence must be put on fluviatile waters. South of the Mogollon Rim springs are abundant, and here were located important pueblos like those of Forestdale and others in the Apache Reservation.
The conditions in the valley of the Little Colorado are similar in regard to available water supply to that of the White Mountains, but the region is more arid and the vegetation is of desert types, the cottonwood along the stream beds being the only tree.
The region of the Hopi buttes and mesas has an elevation of about 6,500 feet, 1,500 feet above the Little Colorado Valley. Geological Causes here also determine the numerous springs in this region, the rainfalls being stored in sandhills or in the heavy strata of porous sand rock underlaid by shales, which brings the water to the surface. This region is practically uninhabitable without corn, which is grown in the beds of the washes and depends on local rains for irrigation. The same remark is true of the second region, while in the White Mountain region hunting tribes could exist.