only life of the islands. The sea-birds range away from the land, in pursuit of their finny prey, to a distance varying from sixty to eighty and sometimes even to a hundred miles; and so unerring are they in their return to their breeding and roosting places, that we were told the captains of the small schooners, who seek the islands to harvest their crop of cocoanuts and to look about for guano, find the objects of their search by laying to, when they imagine themselves somewhere in their vicinity, until evening, when the birds wing their way homeward, and then shaping their course by the direction of the flight of the birds.

The other terrestrial animals of Palmyra are a minute shell, a Tornatellina, that clings to the under surface of the fronds of the Polypodium aureum, and a land leech that fastens itself to the eyelids of the young birds.

A soldier-crab (Canobita Glivieri) quits the water and lives on the land at certain seasons of the year, and it may therefore be properly considered terrestrial. They are ubiquitous; they climb the trees and bushes, dragging after them the heavy shells of the Turbo argyrostoma.

Washington Island is a home of the gigantic Burgus latro. "In the Pacific this species, or one with closely allied habits, is said to inhabit a single coral island north of the Society group." Since this statement of Darwin's was made, we have gained a much better knowledge of its distribution, so that now it may be said there is scarcely a group of islands in the Pacific Ocean where it is not found. The stories about this crab climbing trees after cocoanuts are entirely unfounded. It lives in burrows in the ground, and feeds on the cocoanuts as they fall from the trees. It first strips off the husk, shred by shred, and then, with its strong pincers, breaks through the shell at the extremity that holds the eyes. The strength of their claws is sufficient to crush a lath in twain, and so tenaciously do they hold on to anything when once they have obtained a grip, that I have known them to hang suspended from a tree for an hour or more, holding on by their claws. Sometimes the unwary native, in searching their burrows with his hand for the fine cocoanut husk which forms their bed, is surprised to find his fingers in the. vice-like grasp grasp of the crab; and it may be interesting to know that in such a dilemma a gentle titillation of the under soft parts of the body with any light material will cause the crab to loose his hold. They are said to visit the sea at night for the purpose of wetting their branchiæ.

EXPLORATIONS MADE IN COLORADO UNDER THE DIRECTION OF PROF. F. V. HAYDEN IN 1876.

FOR

OR reasons beyond the control of the geologist-in-charge, the various parties composing the United States Geological and Geographical Survey of the Territories did not commence their field work until August. Owing to the evidences of hostility among the northern tribes of Indians, it was deemed most prudent to confine the labors of the survey to the completion of the atlas of Colorado. Therefore the work of the season of 1876 was a continuation of the labors of the three preceding years, westward, finishing the entire mountainous portion of Colorado with a belt fifteen miles in width of Northern New Mexico and a belt twenty-five miles in breadth of Eastern Utah. Six sheets of the physical atlas are now nearly ready to be issued from the press. Each sheet embraces an area of over 11,500 square miles or a total of 70,000 square miles. The maps are constructed on a scale of four miles to one inch with contours of two hundred feet which will form the basis on which will be represented the geology, mineral, grass, and timber lands, and all lands that can be rendered available for agriculture by irrigation. The areas of exploration are located in the interior of the continent, far remote from settlements, and among the hostile bands of Ute Indians that attacked two of the parties the previous year. The point of departure the past season was Cheyenne, Wyoming Territory. The primary triangulation party was placed in charge of A. D. Wilson, and took the field from Trinidad, the southern terminus of the Denver and Rio Grande Railway, August 18th, making the first station on Fisher's Peak. From this point the party marched up the valley of the Purgatoire, crossed the Sangre de Cristo Range by way of Costilla Pass, followed the west base of the range northward as far as Fort Garland, making a station on Culebra Peak.

About six miles north of Fort Garland is located one of the highest and most rugged mountain peaks in the West, called Blanca Peak, the principal summit of the Sierra Blanca Group. On the morning of August 28th the party, with a pack mule to transport the large theodolite, followed up a long spur which juts out to the south. They found no difficulty in riding to the timber line, which is here about twelve thousand feet above the sea-level. At this point they were compelled to leave the animals, and, distributing the instruments among the different

members of the party, proceeded on foot up the loose rocky slope to the first outstanding point, from which a view could be obtained of the main peak of the range. Although this first point is only six hundred feet lower than the main summit, yet the most arduous portion of the task was yet to come. The main summit is about two miles north of the first point, in a straight line, and connected with it by a very sharp-toothed zigzag ridge over which it is most difficult to travel on account of the very loose rocks and the constant danger of being precipitated down, on either side, several hundred feet into the amphitheatre below. After some two hours of this difficult climbing, they came to the base of the main point, which though very steep was soon ascended, and at eleven o'clock, A. M. they found themselves on the very summit. From this point one of the most magnificent views in all Colorado was spread out before them. The greater portion of Colorado and New Mexico was embraced in this field of vision. This point is the highest in the Sierra Blanca group and so far as is known at the present time is the highest in Colorado. The elevation of this point was determined by Mr. Wilson in the following manner: first, by a mean of eight barometric readings taken synchronously with those at Fort Garland, which gave a difference between the two points of 6466 feet; secondly, by fore and back angles of elevation and depression, which gave a difference of 6468 feet. The elevation at the fort was determined by a series of barometric readings, which compared with those of the signal-service barometer at Colorado Springs gave it an elevation of 7997 feet, making the Blanca Peak 14,464 feet above sea-level. This peak may be regarded therefore as the highest or at least next to the highest yet known in the United States. A comparison with some of the first-class peaks in Colorado will show the relative height:

The foregoing table will afford some conception of the difficulty encountered in determining the highest peak when there are so many that are of nearly the same elevation. About fifty peaks are found within the limits of Colorado that exceed fourteen thousand feet above sea-level.

From this point the party proceeded westward across the San Luis valley and up the Rio Grande to its source, making two primary stations on the way, one near the summit district and the other on the Rio Grande Pyramid. From the head of the Rio Grande the party crossed the continental divide, striking the Animas Park, thence west by trail to Parrott City.

After making a station on La Plata Peak, the party marched northwest across the broken mesa country west of the Dolores, making three stations on the route to complete a small piece of topography that had been omitted the previous year on account of the hostility of the Ute Indians. After making a primary station on the highest point of the Abajo Mountains, the party turned westward to Lone Cone, where another station was made; thence, crossing the Gunnison and Grand rivers, they proceeded to the great volcanic plateau at the head of White River. The final station was made between the White and Yampah rivers in the northwestern corner of Colorado. During this brief season, Mr. Wilson finished about one thousand square miles of topography and made eleven geodetic stations, thus connecting together by a system of primary triangles the whole of Southern and Western Colorado.

In company with the triangulation party, Mr. Holmes made a hurried trip through Colorado, touching also portions of New Mexico and Utah. He was unable to pay much attention to detailed work, but had an excellent opportunity of taking a general view of the two great plain-belts that lie, the one along the eastern, the other along the western base of the Rocky Mountains. For nearly two thousand miles of travel he had constantly in view the Cretaceous and Tertiary formations, with which are involved some of the most interesting geological questions. He observed among other things the great persistency of the various groups of rocks throughout the east, west, and north, and especially in the west, for from Northern New Mexico to Southern Wyoming the various members of the Cretaceous formation lie in almost unbroken belts, while the Tertiaries are hardly less easily followed.

Between the east and the west there is only one great incongruity. Along the eastern base of the mountains the upper Cretaceous rocks, including numbers four and five, are almost wanting, consisting at most of a few hundred feet of shales and laminated sandstones. Along the western base this group becomes a prominent and important topographical as well as geological feature.

In the southwest, where it forms the "mesa verde" and the cap of the Dolores plateau, it comprises upwards of two thousand feet of coal-bearing strata, chiefly sandstone, while in the north it reaches a thickness of 3500 feet and forms the gigantic hogback of the Grand River Valley.

While in the southwest he visited the Sierra Abajo, a small group of mountains which lie in Eastern Utah, and found, as he had previously surmised, that the structure was identical with that of the other four isolated groups that lie in the same region. A mass of trachyte has been forced up through fissures in the sedimentary rocks and now rests chiefly upon the sandstones and shales of the lower Cretaceous. There is a considerable amount of arching of the sedimentary rocks, probably caused by the intrusion of wedge-like sheets of trachyte, while the broken edges of the beds are frequently bent abruptly as if by the upward or lateral pressure of the rising mass. He was able to make many additional observations on the geology of the San Juan region and secured much valuable material for the coloring of the final map.

He states that the northern limit of ancient cliff builders in Colorado and Eastern Utah is hardly above latitude 37° 45'.

The Grand River Division was directed by Henry Gannett, topographer, with Dr. A. C. Peale as geologist. James Stevenson, executive officer of the survey, accompanied this division for the purpose of assisting in the management of the Indians, who last year prevented the completion of the work in this locality by their hostility.

The work assigned this division consisted in part of a small area, containing about one thousand square miles, lying south of the Sierra la Sal. The greater portion of the work of this division lay north of the Grand River, limited on the north by the parallel of 39° 30′ and included between the meridians of 108° and 109° 30'.

This division took the field at Cañon City, Colorado, about the middle of August. The party traveled nearly west, up the Arkansas River, over Marshall's Pass and down the Pomichi and Gunnison rivers to the Uncompahgre (Ute) Indian agency. Here they secured the services of several Indians as escort in the somewhat dangerous country which they were first to survey. This area lying south of the Sierra la Sal was worked without difficulty. It is a broken plateau country and presents many curious pieces of topography. Eleven days were occupied in this work.