In the dry season when the brisk east wind is blowing, and the sharpness of outline of hills, woods, and sandy beaches, give a great charm to this spot.

While resting in the shade during the great heat of the early hours of afternoon, I used to find amusement in watching the proceedings of the sand wasps. A small pale green kind of Bembex (Bembex ciliata) was plentiful near the bay of Mapirí. When they are at work a number of little jets of sand are seen shooting over the surface of the sloping banks. The little miners excavate with their fore feet, which are strongly built and furnished with a fringe of stiff bristles; they work with wonderful rapidity, and the sand thrown out beneath their bodies issues in continuous streams. They are solitary wasps, each female working on her own account. After making a gallery two or three inches in length, in a slanting direction from the surface, the owner backs out and takes a few turns round the orifice, apparently to see whether it is well made, but in reality, I believe, to take note of the locality, that she may find it again. This done the busy workwoman flies away; but returns, after an absence varying in different cases from a few minutes to an hour or more, with a fly in her grasp, with which she re-enters her mine. On again emerging the entrance is carefully closed with sand. During this interval she has laid an egg on the body of the fly, which she had previously benumbed with her sting, and which is to serve as food for the soft footless grub soon to be hatched from the egg. From what I could make out the Bembex makes a fresh excavation for every egg to be deposited; at least, in two or three of the galleries which I opened there was only one fly inclosed.

I have said that the Bembe on leaving her mine took note of the locality: this seemed to be the explanation of the short delay previous to her taking flight; on rising in the air, also, the insects generally flew round over the place before making straight off. Another nearly allied but much larger species, the Monedula signata, whose habits I observed on the banks of the Upper Amazons, sometimes excavates its mine solitarily on sand-banks recently laid bare in the middle of the river, and closes the orifice before going in search of prey. In these cases the insect has to inake a journey of at least half a mile to procure the kind of fly, the Motúca (Madrus lepidotus), with which it provisions its cell. I often noticed it to take a few turns in the air round the place before starting; on its return it made without hesitation straight for the closed mouth of the mine. I was convinced that the insects noted the bearings of their nests, and the direction they took in flying from them. The proceeding in this and similar cases (I have read of something analogous having been noticed in hive bees) seems to be a mental act of the same nature as that which takes place in ourselves

when recognizing a locality. The senses, however, must be immeasurably more keen, and the mental operation much more certain, in them than they are in man; for to my eye there was absolutely no land-mark on the even surface of sand which could serve as guide, and the borders of the forest were not nearer than half a mile. The action of the wasp would be said to be instinctive; but it seems plain that the instinct is no mysterious and unintelligible agent, but a mental process in each individual, differing from the same in man only by its unerring certainty. The mind of the insect appears to be so constituted that the impression of external objects, or the want felt, causes it to act with a precision which seems to us like that of a machine constructed to move in a certain given away. I have noticed in Indian boys a sense of locality almost as keen as that possessed by the sand-wasp. An old Portuguese and myself, accompanied by a young lad about ten years of age, were once lost in the forest in a most solitary place on the banks of the main river. Our case seemed hopeless, and it did not for some time occur to us to consult our little companion, who had been playing with his bow and arrow all the way while we were hunting, apparently taking no note of the route. When asked, however, he pointed out, in a moment, the right direction of our canoe. He could not explain how he knew; I believe he had noted the course we had taken almost unconsciously. The sense of locality in his case seemed instinctive. The Monedula signata is a good friend to travellers in those parts of the Amazons which are infested by the bloodthirsty Motùca. I first noticed its habit of preying on this fly one day when we landed to make our fire and dine on the borders of the forest adjoining a saud-bank. The insect is as large as a hornet, and has a most waspish appearance. I was rather startled when one out of the flock which was hovering about us flew straight at my face: it had espied a Motúca on my neck, and was thus pouncing upon it. It seizes the fly not with its jaws, but with its fore and middle feet, and carries it off tightly held to its breast. Wherever the traveller lands on the Upper Amazons in the neighborhood of a sand-bank he is sure to be attended by one or more of these useful vermin-killers.

The buy of Mapirí was the limit of my day excursions by the river-side, to the west of Santarem. A person may travel, however, on foot, as Indians frequently do, in the dry season for fifty or sixty miles along the broad clean sandy beaches of the Tapajos. The only obstacles are the rivulets, most of which are fordable when the waters are low To the east my rambles extended to the banks of the Mahicá inlet. This enters the AmazoDS. about three miles below Santarem, where the clear stream of the Tapajos begins to be discolored by the turbid waters of the main river. The Mahicá has a broad margin of rich level pasture, limited on each side by the

Its mouth with a lens while it was lying on the mortar. Every fresh pellet was brought in with a triumphant song, which changed to a cheerful busy hum when it alighted and began to work. The little ball of moist clay was laid on the edge of the cell, and then spread out around the circular rim, by means of the lower lip guided by the mandibles. The insect placed itself astride over the rim to work, and, on finishing each addition to the structure, took a turn round, patting the sides with its feet inside and out before fly. ing off to gather a fresh pellet. It worked only in sunny weather, and the previous layer was sometimes not quite dry when the new coating was added. The whole structure takes about a week to complete. I left the place before the gay little builder had quite finished her task; she did not accompany the canoe, although we moved along the bank of the river very slowly. On opening closed nests of this species, which are common in the neighborhood of Mahicá, I always found them to be stocked with small spiders of the genus Gastracantha, in the usual half-dead state to which the mother wasps reduce the insects which are to serve as food for their progeny.

Besides the Pelopæus there were three or four kinds of Trypoxylon, a genus, also found in Europe, and which some naturalists have supposed to be parasitic, because the legs are not furnished with the usual row of strong bristles for digging, characteristic of the family to which it belongs. The species of Trypoxylon, however, are all building wasps; two of them which I observed (T. albitarse and an undescribed species) provision their nests with spiders, a third (T. aurifrons) with small caterpillars. Their habits are similar to those of the Pelopeus: namely, they carry off the clay in their mandibles, and have a different song when they hasten away with the burden from that which they sing while at work. Trypoxylon albitarse, which is a large black kind, three quarters of an inch in length, makes a tremendous fuss while building its cell. It often chooses the walls or doors of chambers for this purpose, and when two or three are at work in the same place their loud humming keeps the house in an uproar. The cell is a tubular structure about three inches in length. T. aurifrons, a much smaller species, makes a neat little nest shaped like a carafe; building rows of them together in the corners of verandas.

But the most numerous and interesting of the clay artificers are the workers of a species of social bee, the Melipona fasciculata. The Meliponæ in tropical America take the place of the true Apides to which the European hive-bee belongs, and which are here unknown; they are generally much smaller insects than the hive-bees and have no sting. The M. fasciculata is about a third shorter than the Apis mellifica; its colonies are composed of an iminense number of individuals. The workers are generally seen collecting pollen in the same way as other bees, but

great numbers are employed gathering clay. The rapidity and precision of their movements while thus engaged are wonderful. They first scrape the clay with their jaws; the small portions gathered are then cleared by the anterior paws and passed to the second pair of feet, which in their turn, convey them to the large foliated expansions of the hind shanks, which are adapted normally in bees, as every one knows, for the collection of pollen. The middle feet pat the growing pellets of mortar on the hind legs to keep them in a compact shape as the particles are successively added. The little hodsmen soon have as much as they can carry, and they soon fly off. I was for some time puzzled to know what the bees did with the clay; but I had afterward plenty of opportunity for ascertaining. They construct their combs in any suitable crevice in trunks of trees or perpendicular banks, and the clay is required to build up a wall so as to close the gap, with the exception of a small orifice for their own entrance and exit. Most kinds of Meliponæ are in this way masons as well as workers in wax and pollen-gatherers. One little species (undescribed), not more than two lines long, builds a neat tubular gallery of clay, kneaded with some viscid substance, outside the entrance to its hive, besides blocking up the crevice in the tree within which it is situated. The mouth of the tube is trumpet-shaped, and at the entrance a number of the pigmy bees are always stationed, apparently acting as sentinels.

A hive of the Mellipona fasciculata which I saw opened, contained about two quarts of pleasantly-tasted liquid honey. The bees, as already remarked, have no sting, but they bite furiously when their colonies are disturbed. The Indian who plundered the hive was completely covered by them, they took a particular fancy to the hair of his head, and fastened on it by hundreds. I found fortyfive species of these bees in different parts of the country; the largest was half an inch in length; the smallest were extremely miuute, some kinds being not more than one twelfth of an inch in size. These tiny fellows are often very troublesome in the woods. on account of their familiarity; for they settle on one's face and hands, and, in crawling about, get into the eyes and mouth or up the nostrils.

The broad expansion of the hind shanks of bees is applied in some species to other uses besides the conveyance of clay and pollen. The female of the handsome golden and black Euglossa Surinamensis has this palette of very large size. This species builds its solitary nest also in crevices of walls or trees; but it closes up the chink with fragments of dried leaves and sticks, cemented together, instead of clay. It visits the cajú-trees, and gathers with its hind legs a small quantity of the gum which exudes from their trunks. To this it adds the other materials required from the neighboring bushes, and when laden flies off to its nest.

To the south my rambles never extended further than the banks of the Irurá, a stream which rises among the hills already spoken of, and running through a broad valley, wooded along the margins of the watercourses, falls into the Tapajos at the head of the bay of Mapirí. All beyond, as before remarked, is terra incognita to the inhabitants of Santarem. The Brazilian settlers on the banks of the Amazons seem to have no taste for explorations by land, and I could find no person willing to accompany me on an excursion further toward the interior. Such a journey would be exceedingly difficult in this country, even if men could be obtained willing to undertake it. Besides, there were reports of a settlement of fierce runaway negroes on the Serra de Mururarú, and it was considered unsafe to go far in that direction, except with a large armed party. I visited the banks of the Irurá and the rich woods accompanying it, and two other streams in the same neighborhood, one called the Panéma, and the other the Urumarí, once or twice a week during the whole time of my residence in Santarem, and made large collections of their natural productions. These forest brooks, with their clear cold waters brawling over their sandy er pebbly beds, through wild tropical glens, always had a great charm for me. The beauty of the moist, cool, and luxuriant glades was heightened by the contrast they afforded to the sterile country around them. The bare or scantily wooded hills which surround the valley are parched by the rays of the vertical sun. One of them, the Pico do Irurá, forms a nearly perfect cone, rising from a small grassy plain to a height of 500 or 600 feet, and its ascent is excessively fatiguing after the long walk from Santarem over the campos. I tried it one day, but did not reach the summit. A dense growth of coarse grasses clothed the steep sides of the hill, with here and there a stunted tree of kinds found in the plain beneath. In bared places, a red crumbly soil is exposed; and in one part a mass of rock, which appeared to me, from its compact texture and the absence of stratification, to be porphyritic; but I am not geologist sufficient to pronounce on such questions. Mr. Wallace states that he found fragments of scoriæ, and believes the hill to be a volcanic cone. To the south and east of this isolated peak the elongated ridges or table-topped hills attain a somewhat greater elevation.

The forest in the valley is limite to a tract a few hundred yards in width on each side of the different streams; in places where these run along the bases of the hills, the hill-sides facing the water are also richly wooded, although their opposite declivities are bare, or nearly so. The trees are lofty and of great variety; among them are colossal examples of the Brazil-nut-tree (Bertholletia excelsa), and the Pikiá. This latter bears a large eatable fruit, curious in having a hollow chamber between the pulp and the kernel, beset with hard spines, which oro

duce serious wounds if they enter the skin. The eatable part appeared to me not much more palatable than a raw potato; but the inhabitants of Santarem are very fond of it, and undertake the most toilsome journeys on foot to gather a basketful. The tree which yields the tonka bean (Dipteryx odorata), used in Europe for scenting snuff, is also of frequent occurrence here. It grows to an immense height, and the fruit, which, although a legume, is of a rounded shape, and has but one seed, can be gathered only when it falls to the ground. A considerable quantity (from 1000 to 3000 pounds) is exported annually from Santarem, the produce of the whole region of the Tapajos. An endless diversity of trees and shrubs, some beautiful in flower and foliage, others bearing curious fruits, grow in this matted wil. derness. It would be tedious to enumerate many of them. I was much struck with the variety of trees with large and diversely shaped fruits growing out of the trunk and branches, some within a few inches of the ground, like the cacao. Most of them are called by the natives Cupú, and the trees are of inconsiderable height. One of them, called Cupú-aï, bears a fruit of elliptical shape and of a dingy earthen color, six or seven inches long, the shell of which is woody and thin, and contains a small number of seeds loosely enveloped in a juicy pulp of very pleasant flavor. The fruits hang like clayey ants' nests from the branches. Another kind more nearly resembles the cacao; this is shaped something like the cucumber, and has a green ribbed husk. It bears the name of Cacao de macaco, or monkey's chocolate, but the seeds are smaller than those of the common cacao. I tried once or twice to make chocolate from them. They contain plenty of oil of similar fragrance to that of the ordinary cacao-nut, and make up very well into paste; but the beverage has a repulsive clayey color and an inferior flavor.

My excursions to the Irur ad always a picnic character. A few rude huts are scattered through the valley, but they are tenanted only for a few days in the year, when their owners come to gather and roast the mandioca of their smali clearings. We used generally to take with us two boys-one negro the other Indian-to carry our provisions for the day; a few pounds of beef or fried fish, farinha, and bananas, with plates, and a kettle for cooking. José carried the guns, ammunition, and game-bags, and I the apparatus for entomologizing-the insect net, a large leathern bag with compartments for corked boxes, phials, glass tubes, and so forth. It was our custom to start soon after sunrise, when the walk over the campos was cool and pleasant, the sky without a cloud, and the grass wet with dew. The paths are mere faint tracks; in our early excursions it was difficult to avoid missing our way. We were once completely lost, and wandered about for several hours over the scorching soil without recovering the road. A fiue

view is obtained of the country, from the rising ground about half way across the waste. Thence to the bottom of the valley is a long, gentle, grassy slope, bare of trees. The strangely-shaped hills; the forest at their feet, richly varied with palms; the bay of Mapirí on the right, with the dark waters of the Tapajos and its white glistening shores, are all spread out before one, as if depicted on canvas. The extreme transparency of the atmosphere gives to all parts of the landscape such clearness of outline that the idea of distance is destroyed, and one fancies the whole to be almost within reach of the hand. Descending into the valley, a small brook has to be crossed, and then half a mile of sandy plain, whose vegetation wears a peculiar aspect, owing to the predominance of a stemless palm, the Curuá (Attalea spectabilis), whose large, beautifully pinnated, rigid leaves rise directly from the soil. The fruit of this species is similar to the cocoanut, containing milk in the interior of the kernel, but it is much inferior to it in size. Here, and indeed all along the road, we saw on most days in the wet season tracks of the jaguar. We never, however, met with the animal, although we sometimes heard his loud 'hough" in the night, while lying in our hammocks at home, in Santarem, and knew he must be lurking somewhere near us.

My best hunting ground was a part of the valley sheltered on one side by a steep hill, whose declivity, like the swampy valley be neath, was clothed with magnificent forest. We used to make our halt in a small cleared place, tolerably free from ants and close to the water. Here we assembled after our toilsome morning's hunt in different directions through the woods, took our well-earned meal on the ground-two broad leaves of the wild banana serving us for a tablecloth—and rested for a couple of hours during the great heat of the afternoon. The diversity of animal productions was as wonderful as that of the vegetable forms in this rich locality. It was pleasant to lie down during the hottest part of the day, when my people lay asleep, and watch the movements of animals. Sometimes a troop of Aanús (Crotophaga), a glossy black-plumaged bird, which lives in small societies in grassy places, would come in from the campus, one by one, calling to each other as they moved from iree to tree. Or a Toucan (Rhamphastos ariel) silently hopped or ran along and up the branches, peeping into chinks and crevices. Notes of solitary birds resounded from a distance through the wilderness. Occasionally a sulky Trogon would be seen, with its brilliant green back and rose-colored breast, perched for an hour without moving on a low branch. A number of large fat lizards, two feet long, of a kind called by the natives Jacuarú (Teius teguexim) were always observed in the still hours of mid-day scampering with great clatter over the dead leaves, apparently in chase of each other. The fat of this bulky lizard is much prized by the natives, who apply it as a poultice to draw palm spines or even

grains of shot from the flesh. Other lizards of repulsive aspect, about three feet in length when full grown, splashed about and swam in the water: sometimes emerging to crawl into hollow trees on the banks of the stream, where I once found a female and a nest of eggs. The lazy flapping flight of large blue and black morpho butterflies high in the air, the hum of insects, and many inanimate sounds, contributed their share to the total impression this strange solitude produced. Heavy fruits from the crowns of trees which were mingled together at a giddy height overhead, fell now and then with a startling 'plop" into the water. The breeze, not felt below, stirred in the topmost branches, setting the twisted and looped sipós in motion, which creaked and groaned in a great variety of notes. To these noises was added the monotonous ripple of the brook, which had its little cascade at every score or two yards of its course.

64

I seldom met with any of the larger ani. mals in these excursions. We never saw a mammal of any kind on the campos; but tracks of three species were seen occasionally besides those of the jaguar; these belonged to a small tiger-cat, a deer, and an opossum ; all of which animals must have been very rare, and probably nocturnal in their habits, with the exception of the deer. I saw in the woods, on one occasion, a small flock of monkeys, and once had an opportunity of watching the movements of a sloth. The latter was of the kind called by Cuvier Bradypus tridactylus, which is clothed with shaggy gray hair. The natives call it, in the Tupí language. Aï ybyreté (in Portuguese, Preguiça da terra firme), or sloth of the mainland, to distinguish it from the Bradypus infuscatus, which has a long black and tawny stripe between the shoulders, and is called Aïygapó (Preguiça das vargens), or sloth of the flooded lands. Some travellers in South America have described the sloth as very nimble in its native woods, and have disputed the justness of the name which has been bestowed upon it. The inhabitants of the Amazons region, however, both Indians and descendants of the Portuguese, hold to the common opinion, and consider the sloth as the type of laziness. It is very common for one native to call another, in reproaching him for idleness, "bicho de Embaüba" (beast of the Cecropia tree); the leaves of the Cecropia being the food of the sloth. It is a strange sight to watch the uncouth creature, fit production of these silent shades, lazily moving from branch to branch. Every movement betrays, not_indolence exactly, but extreme caution. He never loses his hold from one branch without first securing himself to the next, and when he does not immediately find a bough to grasp with the rigid hooks into which his paws are so curiously transformed, he raises his body, supported on his hind legs, and claws around in search of a fresh foothold. After watching the animal for about half an hour I gave him a charge of shot; he fell with a terrific crash, but