Lowe at eee EO oreciee a ah sé e Re ht Bah be Ve a AA 2 3 : TA ARAB RIND A wat AManennannnntn ANN naa ananianivanenan: c Mtns anne MAM An | manne AAAAAABAAT IAA an'ss Ai ARAna aRaasicaaie enters tino ar anee Anan aannannnncann\nannA te ‘ala AMAA AARAASAARS Taeatennmnna Ba nt , SAARAAARANA AAA RAAA aaa ASAE RAARAR | nmAAAAan WM AANA Ana aa aana, ka a8 Nahe RRAAPARAMANAA NA RAE AAA serene vcc as, seen AWy NAAR . Ane Ae ; Ninn vy f Rages nnnnnnyynynnn Wirt versveeterteareewe wo wea ARIANA RA AAR ves ALC uaa nny lel AA AN AAAS Pe We ida WY : ; aan aA na ANA AA RANIAAABARANA AAA A set an snp Aantannr en WANA ap AAp hs a An re BRAN: Recaacccreceenac AARAAARAAAARAR AA AAAs aaehp . “Op ES Be Ss SD > S : $3 : 3: > > ee > Anan wan A rn ene AR ARN iciiae => D> DP > 2 ad ; ae ees ae Sey cite ‘a AA ar nea ARAR Al CARBAAR A lac Nanna’ ‘A AAAAAAAARBARBARACAT A ar ve A'S o's = 5o>) Ss 33-3 Aralfaltear TARR ye a aaa / meron wi AnaA Ap Aeaaeanstantsnaainaa ARAAARINA « Aaah p «TARA OAAAAAAAR AA ag ahr Wan annie MAY Biparaaiarens AAAS AAAr Seton nnn : PR. ala cha a Baal Anan ee ALE ia - " a ie FEW Walt: Pes ay.v ae ia Raaatonnn AAs AB AAGE aR Alt Ty | AaAAa a | ees a ao mann, annnnAnnnn nner | ne naanannAn ARAARA nA vn anny nan : ttn cca se Naan, ect , ear £ See See T\AAP a\.\e ie ya VAT a ; AAA Sanne maa IY Ar-naA AAA elt Ree ttals Ag LAA 4 t AY ve bag: Oe Lan OS eR AAARAG: d Ds 2. Ss D> »> > > “2 2») 2 ? » > BY ON x I 4 mi a: nie alae: a hx + LANE. Paiegs AAahé vy , 4 =e yy a Ail» & MORGAN EXPEDITIONS, ’70-’71, Cu. FRED. HARTT, 7z Charge. XI1.—Contributions to the Geology and Physical Geology of the Lower Amazonas. The Ereré—Monte-Alegre Dis- trict and the Table- Topped Hills. BY CH. FRED. HARTT. XXIII.—Onx the Devonian Brachtopoda of Eréxé, Province of Pard, Brasil. By RICHARD RATHBUN. [From the Bulletin of the Buffalo Society of Natural Science, January, 1874. ] 201 XXII. Contributions to the Geology and Physical Geography of the Lower Amazonas BY CH. FRED. HARTT, Prof. of Geology in Cornell University. [Read before this Society, January 2, 1874.] THE ERERE-MONTE-ALEGRE DISTRICT AND THE TABLE-TOPPED ® HILLS. WOODED PLAINS Re INY) TN | VICINITY OF MONTE ALEGRE DEVONIAN AND ERERE. : i WOODED PLAINS \y) ‘Sulphur 4 PLAIN * t Saudoso é . | sou ” “MONTE ALEGRE EE Upper town e rrsanten teens ASCENDING the Amazonas from Para, the topographical features observable from the river for the first 300 miles, are very monoto- nous. With the exception of the immediate vicinity of Para, Breves and Gurupa, where the land rises to a height of twenty to thirty feet above tide-level, the country is perfectly flat, scarcely above water.even in the dry season, and of recent origin. Where BUL, BUF. S80C. NAT. SCT. (26) JANUARY, 1874. 202 the land is perennially wet, as along the fwros* connecting the main river and the Para estuary, it is so densely forest-clothed that, from the water, one sees nothing but foliage, and the land-effect is pro- duced not by terra firma, but by the forest-wall that at once borders and limits the channels. Were the vegetation removed from the region just mentioned, the vision of the traveler, instead of being shut in everywhere by the forest, would range over a tract as level as the sea. Enormous mud flats, partially covered by every tide, nowhere more than a very few feet out of water, traversed by a network of deep channels, and diversified by lakes, would be seen stretching away to the horizon on every side, only here,and there a ¢orrao, like that of Breves, rising above the general dead level. Such would be the appearance of the Breves district during the dry season if deprived of trees; but, during the rains, the Amazonas deluges the whole region and pours over it in one broad sheet into the bay of Marajé. To rightly appreciate the topography of the lower Amazonas, we must eliminate the effect produced upon us by the vegetation. True it is that the alluvial lands, just described, depend upon the forest both for their origin and existence, but one is apt to mistake forest topography, if I may use such a term, for land topography, and count for more than its real geographical value, a district whose height and limits are intensified or defined by forest. After having made six voyages between the bay of Marajé and the main river, I am satisfied that, one reason why voyagers have so much doubted whether the, so called, Para river should be considered a mouth of the Amazonas, is largely due to the fact, that the forest prevents a just appreciation of the magnitude of the united channels of the Breyes district, while, at the same time, the size of the Tocantins has been much over-esti- mated. Above Trocard this river is, during the dry months, only a small, narrow stream, while, in the lower course, it is not a true river, but a wide, extremely shallow, tidal estuary, the upper part of which is in process of filling up with sand, brought down by the river. The enormously wide, lower reaches, that open broadly into the bay of Marajo, are swept by very strong tides, and are being silted up by Amazonian mud. Travelers who hastily pass through * A furo on the Amazonas is a channel that connects two different streams-and it differs from a parana-merim, which is a side channel that leaves a river and joins it again lower down. 203 the Breves district, and trust to maps and the glimpses they get of the mouth of the Tocantins, may set down the Para as simply the extension of that river, but they are not correct. The Tapajos and Tocantins are rivers of very nearly the same size, but the waters of the former river, on issuing from its mouth, are crowded by the mighty torrent of the Amazonas against its bank, as if they were a mere brook. ‘To attribute the fresh waters of the Para to the To- cantins, is like referring a giant’s work toa pigmy. The Tocantins, Moju, Acara and all the true rivers emptying into the Para, taken together, would not, during the dry season, furnish enough water to make more than a respectable Amazonian parana-merim, and they would be utterly insignificant, in comparison with the united Breves furos. Of course the rivers just enumerated must be enormously increased in volume during flood time, but even at that time they cannot compare with the wide Amazonian flood which then pours through channel and forest over the Breves lowlands. It must not be forgotten that these lowlands are bordered on the east by the higher lands of Maraj6 and on the south-west by those of the southern side of the Amazonian valley, and the traveler on the lower Amazonas should remember that the flat, alluvial banks, which so monotonously accompany the river, do not extend very far into the interior. If we ascend the Tocantins, we shall encounter the higher grounds at Cameta, and the town of Gurupa is built on, What appears to be a low spur of these same lands. They reap- pear again at the mouth of the Xingu, to the westward of which, at a greater or less distance from the river, they stretch in a line of bluffs to the Tapajos. Ascending the Amazonas by the ordinary route, one sees no high lands on the northern side of the river, until, having passed the mouth of the Xingu, the table-topped serras of Part rise before one, stretching along the river in patches nearly to Prainha, beyond which soon come into view the highlands of the Monte-Alegre dis- trict. It is to the Geology and Physical Geography of these north- ern Highlands and their vicinity that I now invite the attention of the reader. The villa of Monte-Alegre is situated a few miles above the mouth of the rio Curupatuba,* one of the northern affluents of the Ama- * From the Lingoa geral Kurupd, a port, and ty'ua, a place of. The name appears to have . been primarily applied to the village, because of its convenient landing place. Rio Curupatuba 204 zonas, and is distant 350-360 miles nearly directly west of the city of Belém or Para. On the maps, the Curupattiba is usually represented as a large river, taking its rise in the highlands of Guiana, to the north-west- ward of Monte-Alegre, and which, shortly before entering the Ama- zonas, receives by a short outlet the waters of a large lake. Accord- ing to Sr. Ferreira Penna* this is inexact. The river that descends from the interior is called the Maecurt + (or Maycurt) and it empties directly into the lake. This river has never been explored and noth- ing is known of its upper course. The lower part is bordered by rich grazing grounds and is inhabited. The lake, commonly known as the Lago Grande de Monte-Alegre and celebrated for its fishery of the piraruca (Sudis grandis) is situated in the alluvial bottom about midway-between Monte-Alegre and Santarem, and to the south-west of the former villa. Sr. Penna says that it is about twenty-five miles long, and from three to five in width. It is most probably an old channel of the Amazonas. The same author states that the lake empties by two channels which soon unite in one called the Cururuhy'.{ This presently receives on the left the Tgarapeé-apara,§ when the stream takes the name Curupatiba. The course of the latter is a5 first north or north-east, but, just before reaching the villa of Monte-Alegre, it makes a bend to the east, and, hugging the higher lands on the northern side of the valley, emp- ties into the Amazonas, a few miles east of Monte-Alegre, just below which town, it communicates with the main river by a navigable parand-merim.|| It is interesting to observe that the Amazonas runs obliquely across the valley, in a north-easterly direction, from the highlands, a few miles east of Santarem, to those of Monte- Alegre, leaying a very broad strip of alluvial campos on the north- ern side, which narrows towards the east, running out near the then corresponds to Rio de Monte-Alegre, which one sometimes hearsused. On some maps we find the spelling, Gurupatiba. Gurupa, the name of a little town a few hours east of the mouth of the Xingt, is a corruption of Kurupd. * A Regido occidental da Prov. do Para, p. 125. +From the reports of the vaqueiros and some fragments of a fine sharp sandstone I have seen, I am led to believe that the geology of the river would prove interesting. + Toad river, from Kururt, a toad, and y'g, water or river. § Apdra means crooked, || More properly a furo or cross-cut. 205 mouth of the Curupatuba; these plains having been formed by the growth and fusion of islands in the silting up of the valley. The villa of Monte-Alegre* is divided into two parts, the upper or principal town, and the lower town or port. The latter is situated on the left bank of the river, while the upper town, distant about « mile to the north, and reached by a steep, weary, sandy ascent, is built on the edge of a high, broad, flattened ridge or plateau, extend- ing northward from the river to the serra of Tauajuri, distant some eighteen miles.t This ridge, which has a height of five or six hundred feet, more or less, is composed of horizontal beds of clays and sands, of proba- ble Tertiary age, and is, as I suppose, a degraded outlier of the once extensive formation of the Serras de Pari. On top it is very flat, but the surface is gently rounded, descending to the plains, both to the east and west, by gradual slopes, abrupt descents being infre- quent, except on the southern side, which, having been encroached upon by the Amazonas, is steep, sometimes precipitous along the base, and gullied by many ravines. The upper town of Monte-Alegre is composed of some fifty, for the most part shabby, tumble-down houses and vendas, together with a handsome new church and a curious, little, old, barn-lhke chapel, surrounding an immense, shadeless, sandy, Sahara of a square. The inhabitants are principally of Indian descent, but among the white families there are a few of education and refinement. The town has been ruined by the rubber trade, and is fast going to decay. The people are chiefly engaged in grazing, fishing and trade. From the villa there is a magnificent view over the Amazonian valley. Below is the Curupatuba, which one may trace far to the south-westward, winding, tree-fringed, over the verdant, grassy, alluvial plain, which, levelas the sea and variegated by forest patches and mirror-like lagoons, stretches southward for miles to the turbid flood of the mighty river, while away beyond in the south-west, are * Happy mount. The name sounds strangely to the traveler who has enjoyed its delectable nights, the cheerful serenades of its carapands and the moon worship of its numerous canine population ! +I regret very much that I shall be obliged to estimate all the distances given in this paper, and that I can furnish nothing more than a rough sketch-map of the district examined. The region has never been surveyed and mapped, and I have hence labored under a very great dis- advantage. All my work on land was done on foot, many days often being spent ina fruitless ‘search for rock exposures. 206 the white cliffs of Cucary, and the blue, level highlands of the vicinity of Santarem. Seen from Monte-Alegre, the Amazonas does not resemble a river. It comes mysteriously from the west, stretches a broad, reddish belt across the landscape, and disappears in the east, with a wide water-horizon, as if, an arm of the sea, it opened out to the ocean. When the annual flood comes, all the green campos and clear-water lakes are whelmed beneath the turbid current, and even from the heights of Monte-Alegre, the southern shore is but dimly discernible. No wonder that the Indian fisher- man calls it parand, the sea! Looking westward from the village, one sees distinctly the high, rocky, irregularly flat-topped serra of Paituina, with a curious mushroom-like pillar standing on its south- ern extremity, and called the mao de pilio, or Indud ména. A few miles to the north, is the rugged serra of Ereré, breaking down precipitously towards the north. From the top of the ridge behind the town, the beautiful serra of Tauajuré comes into view, while, to the eastward, lie broad plains and campos, with the level-topped mass of Parauaquara lying low down on the horizon. After this recon- noissance of the region we are to explore, let us descend to the lower town and go by water to EHrere. The descent to the river is at first down a long, sandy incline, showing very few exposures, but the upper part, which is very steep, appears to be composed of reddish, clayey sands, much cut up by rain-courses, the clay being washed out and carried away, while the coarse sand is left lying loose on the surface, swpporting a sparse vegetation, consisting mainly of small trees and shrubs, with here and there a giant cactus, caja trees (Anacardiwm occidentale) being abundant, as we shall find them elsewhere on similar ground. Following the sandy path, and directing our steps to the ravine leading to the lower town, we presently reach a sort of terrace that runs out into a high, bluff, projecting point, extending to the river side just west of the village. This point is formed by a heavy bed of more or less sandy, and variegated feldspathic clay, which, tougher than the overlying beds, has resisted denudation. A little stream of water issuing from above the clay, falls into a ravine, that extends down to the river, and in a steep bank by the side of the road near where the inhabitants resort for water, the clays are well exposed. ‘They vary in character from a pure feldspathic 207 tabatinga to a clayey sand, and are usually more or less deeply tinted, some of the layers being of a rich, purplish red. This bed of clay appears to be the lowest member of the formation of the ridge of Monte-Alegre. If we descend the ravine cut by the above stream, we presently strike a sloping, fan-shaped deposit of loose, white sand occupying the mouth of the ravine and forming a praia or beach along the river. On this sand, the lower town, consisting of a few houses and stores, is built. It is not a flourishing place; everything speaks of decay, and but little business is done in it. I found the people, however, very hospitable, and Senhor Onetti and his partner did everything in their power to aid me. In Mr. Rath- bun’s paper, annexed, I shall have an opportunity of recognizing the kindness of Sr. Valente, of the upper town. Ascending the Curupatuba in a montaria, we find the stream to have a width of 400-500 feet,* and a depth during the dry season of 7-8 fathoms, the current of course varying with the season. The bluffs, 60-100 feet in height, and covered with woods under- grown with curua palms, continue for a short distance above the town, where they cease, and the highlands trend away from the river. The southern side of the ridge is high, abrupt, and with a steep slope. In the valley of Surubiju,t+ just west of the town, are swampy grounds, supporting a luxuriant forest with miritif (W/aw- ritia flecuosa) and assai§ palms (Zuterpe oleracea), but the vegeta- tion of the sandy slope is very meagre. In the valley is an isolated hill, on which beds of a white, sandy tabatinga are exposed, and near by, were obtained the irregular, concretionary masses of iron-stone, used in building the new church in the upper village. On the opposite side of the Curupatuba are the alluvial campos of the river-bottom, covered with coarse grasses and bordered along the water’s edge by a thin line of trees. We soon leave the Curu- patuba, which bends round to the south-westward, and enter the igarape de Paituna, a little river, that flows eastward past the * 260 metres, Penna. A + Von Martius derives this name from Sorubim (Platystoma, a genus of fishes) and y'g, water, or river. Glossarios, p. 475. A A A A A A t Ymyrd. tree, eté, true. Ymyrd was originally ymbyrd, whence the form bu7 itz (Port) used in Eastern Brazil. A § Uasai, lingoa geral, very likely from yué, fruit, and sé or seé, sweet. 208 serra of the same name, and which, like all the streams of the allu- vial bottom of the Amazonas, has a deep, narrow channel, with very steep, muddy banks. In the dry season, the water of the igar- apé is almost stagnant, simply rising and falling with the tide, and the stream literally swarms with alligators of large size. Porpoises gambol in its waters, and its banks abound in game, uacara and mauari cranes, piasdcas, corta-agoas, alencérnos and other birds being exceedingly common. Capibaras are also very abundant in the vicinity. After following the Paituna for some distance, we turn off north- ward into a still smaller stream, called the igarapé de Erere, and now enter a sort of alluvial bay, bounded by the Monte-Alegre plateau on one side, and on the other by the serra of Paituna and the swelling sandy highlands stretching thence to, and east of, the serra of Hreré. The little igarapé is exceedingly tortuous, bending hither and thither in a manner most bewildering to the voyager. Its banks are in part open river bottom, in part margined by a thin line of small trees, palms, as Prof. Agassiz has already remarked, being rare. The water of the stream is very turbid during the dry season, and the narrow channel is often interrupted by floating balsas of canna- rdna. As one ascends the igarapé the valley grows narrower, and at the cattle-fazenda of Sta. Maria, the higher lands of this Ereré plateau come down to the stream, and, in a bluff, obliquely lamina- ted beds of tinted sands and clays are exposed, The alluvial cam- pos of the lower course of the igarapé de Ereré and of the vicinity of Monte-Alegre, are used during the dry season as a pasturage for cattle, and there are several cwrraes along the route we have just followed. Cattle raising is indeed the chief branch of industry followed in this part of the Amazonas. The lands in the Ereré— Monte-Alegre district—are for the most part unfit for cultivation, and agriculture is practiced on a very small scale. The proprietor of the fazenda of Sta. Maria informed me that the satba ant (Oecodoma) was so very abundant on his farm that it was next to impossible to raise a crop. It was even necessary to place the house plants upon a staging erected over the igarapé to protect them, and there they were not always safe. 209 On the left bank of the stream, above the fazenda, begins a very extensive and beautiful grove of miriti palms, which occupies a marshy tract, that seems to be quite dry during several months of the year. A little farther on we meet with higher lands on the left bank, and on the same side, between the upper and lower ports of the village of Ereré, there is a narrow ridge of sandstone, rising about twenty feet above the general level of the campos, and which runs off eastward, perpendicular to the river. This ridge is very much broken, the sandstone lying in huge masses, overgrown with trees and spiny shrubbery, so that I found it very difficult to exam- ine it, and I could not satisfactorily determine the direction of the strata. The rock is, for the most part, a very hard sandstone with a clayey cement, but some of the beds are very argillaceous and _ beautifully striped with brilliant colors. We have now emerged from a sort of pass between the Ereré and Monte-Alegre highlands, and have entered a vast, low plain, sur- rounded by hills and high ground on all sides. From north to south this plain probably measures not less than fifteen miles, while its width from east to west must be over ten miles. It lies a little higher than the alluvial plains of the Amazonas, and is drained by the igarapé by which we have just ascended. It is composed of nearly horizontal strata of Devonian age, through which the igarapé has cut a little valley, now partially filled in with alluvial deposits, lying at a lower level than the plain, the Devonian strata forming low bluffs bordering them. The valley narrows to the northward, and, in the upper part, the igarapée flows directly through, and over the Devonian rocks, a clear water stream. In a little bluff by the side of the road leading from the igarapé to Ereré, and just as one ascends from the alluvial flat, there is an exposure of about fifteen feet in thickness of the Devonian beds. The lower part of the bluff is composed of soft, well-laminated, fine- grained shale, dark gray in color, alternating with white or red layers, and consisting of a fine, more or less sandy silt, with an abundance of little flakes of mica. This locality was discovered in 1870 by my assistants, Messrs. T. B. Comstock, Herbert Smith, and Phineas Staunton, who collected from the variegated shales a pretty little Discina, with which are associated two species of Lingula. BUL. BUF. SOC. NAT. SCI. (27) JANUARY, 1874. 210 The only other fossils yet found in the shales consist of obscure, flattened casts, probably of some marine plant, together with a number of minute, discoid bodies, sometimes arranged in little chains, but of which I can make nothing. Above the shales, just described, is a heavy bed of a not well lami- nated clay-rock, white, mottled with red, in which I have found nothing except some very obscure fucoid-like markings. All these beds have a very slight inclination to the south-eastward. Going northward, the bluffs gradually increase in elevation, but are proba- bly nowhere more than fifty feet in height. The inclination of the beds of the Ereré plain is quite variable, and, over large areas on both sides of the igarapé, they are almost perfectly horizontal, often forming open campos of large extent, which are sometimes so ex- ceedingly stony as to appear as if macadamized, the soil not being sufficient to support even a growth of grass. The lowest beds of the series, that I have examined, are exposed in the north-western part of the campo at the cachoeirinhas of Parica* and Cumamirit situated on branches of the igarapé de Ereré. At the former locality the rock varies from a very hard, dark-colored, silicious shale, to a well bedded, dark gray, compact, cherty rock, breaking with a conchoidal fracture. The strike of these beds, taken along a water-line, is N. 10-15° W., the dip being westward and exceedingly slight. Leaving this locality and going eastward, the surface of the plain rises noticeably for about a mile, the dip being towards the west, continuing with but few elevations to the cachoeirinha do Igarapé do Cumamiri, where similar cherty rocks, with the same very slight westward dip, are seen in the bed of the stream, forming, during the dry season, a little cascade, which at the time of my visit was not more than two feet in height. The cherty beds have afforded no fossils, except a few fragments found in the more shaly portions. Between the cachoeirinhas, above named, the beds are traversed by two dykes, which crop out, much decomposed on the surface; one forming a low ridge running nearly north-south, while the direc- tion of the other is nearly east-west. On the right bank of the igarapé de Ereré, and some distance above the trail to Monte- * A tree, furnishing a seed out of which the Indians make snuff. + This appears to mean Little Milk. 211 Alegre, a sulphur-spring bubbles up through the Devonian shales. The water is limpid, of a greenish tinge, and with a strong sulphu- reous odor and taste; notwithstanding which, the basin in which the water collects is inhabited by little fishes and a species of Ampulla- ria. I regret that I failed in an attempt to bring away some of the water for analysis, especially since at Monte-Alegre it has consider- able repute for its medicinal qualities. Going eastward from the igarapé along the Monte-Alegre trail, one rises by an ascent of a few feet from the alluvial flat to the De- vonian plain, that, almost as level as a floor, stretches to the foot of the Monte-Alegre highlands, beneath which the Palaeozoic beds dis- appear. ‘The surface is quite destitute of soil and is strewn with little nodules of iron-stone, so that large areas are quite barren both of wood and herbage. Just before reaching the Monte-Alegre highlands, several slight elevations, only a few feet high, are met with, that show, in place, light-colored shales, with thin bands of a reddish sandstone, some of which are full of fossils, Streptorhynchus Agassizti, nob., being especially abundant. At this locality I obtained a single glabella of what appears to be a new species of Homalonotus. If we now retrace our steps to the igarapé, and follow the path to the village of Ereré, we shall find the Devonian beds forming a flat or rolling, open campo, with long, gentle ascents and descents, in the rain-courses of which are indifferent exposures of whitish shales, apparently nonfossiliferous. On this campos-land there is very little soil, what there is being baked hard and strewn with small, angular fragments of red sandstone, that occasionally fur- nish fossils. The surface is often covered with little, rounded iron- stone nodules, scarcely larger than beans, sometimes forming a continuous layer. The campo is sparingly clothed with coarse grass, trees being few, scattered, stunted and disfigured by campos fires. Occasional large, arborescent cactuses heighten the dry, bar- ren appearance of the landscape. The low places are covered with woods densely filled in, on the drier grounds, with Curud palms. Between the igarapé and the village of Hreré are several large dykes that project above the surface like ruined walls, but the vein- rock is always badly decomposed, so that it is difficult to say what it originally was. Similar dykes occur in all parts of the plain. 212 The strata, for a few feet on each side of a dyke, are usually consid- erably altered, being hard and flinty, while at the same time they are tilted upward at a more or less strong angle, as if the rent had been widened, not by a horizontal movement of the beds, but by the bending upwards of the strata on both sides of the fissure, through the force of the extruding matter. Sometimes in the denu- dation of the surface, these dykes, as just remarked, project like ruined walls, while at others, with the hardened strata on each side, they form low ridges, that run, sometimes for long distances, on the surface of the campo. In the village and immediate vicinity, there are no good rock exposures. ‘The most interesting locality, and by far the best col- lecting ground for fossils, lies at a distance of about two miles to the northward, in a large, open, treeless, grassy campo. The surface here is quite undulating, and strewn with angular frag- ments of a red or whitish sandstone, rarely ever seen in place. In the rain-courses the rock exposed is usually a fine, soft, well laminated, whitish or yellowish shale, usually quite unproductive in fossils. J'rom the yellow shale I have obtained only a large Lingula, fragments of Vitulina pustulosa Hall, nob., showing the imprints of the little spines and a single ventral valve of a Spir- ifer. This shale, which I know only in a somewhat decomposed state, is largely made up of minute silicious particles and little mica flakes. It takes excellent casts of fossils, and would proba- bly repay more careful examination, but I was unsuccessful in my search for a good exposure. The great repository of fossils is the sandstone, which, as on the eastern side of the igarapé de Ereré, appears to form bands, a few inches in thickness, interstratified with the shales in their upper part. On the washing out of the shales by water the sandstone has cracked up and been left lying in fragments on the surface. Fossils were collected from the loose fragments, but, on the summit of a low ridge, to the north of a deserted house, I discovered on my last visit a layer of the sandstone, which, with great labor, Mr. Derby and I succeeded in uncovering; and this yielded us a magnificent lot of fossils. The layer is only about four inches in thickness, but it is completely filled with fossils which are usually in the shape of moulds, the organic matter having been entirely removed. The rock is com- 213 posed of fine, sharp, quartz-sand, with a sight admixture of clay, and occasionally a tiny, silvery flake of mica. The fragments of sandstone lying on the surface are usually more or less decomposed, and are apt to be stained with iron oxide, which makes them very hard on the outside, while sometimes the surface is covered with a thin layer of the same material. When unaltered the rock appears to be white, or slightly reddish in color. The fossils most abundant in the sandstone are the Brachiopoda, which are represented by twenty species belonging to the following genera: Terebratula, Vitulina, Tropidoleptus, Spirifera, Crytina (?) Retzia, Streptorhynchus, Chonetes, Orthis, Rhynchonella, and Lin- gula, all of which are described in the paper of Mr. Rathbun, annexed. The only other Articulates are the trilobites which are represented by a beautiful Dalmania that occurs in abundance, and a species of Homalonotus, of which last only a fragment is known. Several species of Lamellibranchs occur in the sandstone, belong- ing to Nuculites, Palaeoneilo, Grammysia (?), Ldmondia, and one or two other genera. The Gasteropods number about eight species, representing the genera Bellerophon, Platyceras, Holopea, Pleuroto- maria and Tentaculites. A few fragments of crinoid stems have been found, together with a number of obscure markings which may be of plants. Serra of Ereré from the North. This fauna has an unmistakable Devonian facies, but it is diffi- eult to determine its exact equivalency. In some features, as for instance in Spirifer Pedroana, which closely resembles S. varicosa, the fauna recalls that of the Corniferous, while in the occurrence of Tropidoleptus and Vitulina it approaches the Hamilton.* * See concluding remarks to Mr. Rathbun’s paper. 214 The serra of Ereré is a high, narrow, rugged, irregular ridge, four or five miles long, trending about east-north-east and west-south- west, and with abrupt and often precipitous sides. The upper part of the serra is formed of very heavy beds of sandstone, that dip to the south-south-east at an angle varying from 5°-20°. The top of the ridge is very irregular, ragged and picturesque, the sandstone being often exposed, in situ, in bare ledges or ridges, or lying strewn about in enormous blocks over the surface, which is so rough that it is no easy task to traverse the mountain from one end to the other. Along the northern side of the serra the sandstone forms a broken line of bluffs, varying in height from a few feet to several hundred; and just opposite the little village, and shown in the cut, there is a splendid precipice, remarkable for being rent by fissures from top to bottom. Below these bluffs the side of the serra slopes very steeply, presenting the appearance of a talus, the surface being covered with loose fragments of sandstone. At both ends the serra is cut squarely off, but on the east the sandstone extends downwards, with a strong dip, disappearing under the more modern clays and sands of a swelling ridge like that of Monte-Alegre, that stretches eastward to the igarapé, covered with the characteristic vegetation of the high, sandy campos. hye Wel pith ¢ Serra of Ereré from the East. On the southern side of the serra, and near the eastern end, these sandy campos rise by a gentle incline nearly to the summit, so that 215 one may ascend the serra on horseback. To the westward of this incline, the sides of the serra are exceedingly rough and picturesque. On this side there is hollowed out of the sandstone a large and curi- ous grotto, called Ita-tupa-dka.* This is situated at some little height above the base of the mountain, and is reached by a steep ascent, encumbered by blocks of sandstone, and overgrown with cacti and stiff bushes. The cavern forms ‘a large, irregular, bat- inhabited chamber 50-60 feet long, and with a sandy floor. Wallace had already described the entrance, which is 10-15 feet high, and divided into two parts by a layer of sandstone that runs horizon- tally across the opening about five feet from the floor. This layer is harder than the rock above-or below, much of which is very friable. Immediately west of the serra of Ereré, and separated from it by a deep notch, is a short, angular ridge, with the same trend and geological structure, called Aroxi. In this mountain, which is a little lower than Ereré, the inclination of the sandstone is very marked. On the southern side a broad belt of large cactuses ex- tends from top to bottom. Serras of Ereré and Aroxi from the South-west. To the westward of Aroxi, at a little distance, is another short, high, conical ridge, called Aracuri, while beyond appear to be sev- eral other hills, in a line with those just enumerated, and apparently part of the same outcrop. The sandy campos decline towards the southward from the serra for several miles, when they rise gradually to the rocky plateau of the serra of Paittina. This serra I did not visit, but in 1870, Mr. Phineas Staunton examined it for me, reporting it to be composed of horizontal beds of the same kind of sandstone as that of the serra of Ereré, so that the two serras probably form part of a syn- clinal fold. Paittina is flattened on top, and very broken and pre- cipitous on all sides. Wallace, who visited it, says that the curious, mushroom-like pillar on the southern end is composed “ of friable * Literally, God’s stone house. J/d, stone; Tupa, or Tupana, God; and 6ka, house. stone in horizontal layers and is constantly decaying away by the action of the weather. The top.is formed by a stratum of hard, crystalline rock, which resists the rain and sun,” ete. This upper crystalline rock is probably like the excessively hard sand-stone of the serra of Hreré. The pillar bears the name Jadud ména* in Lingoa Geral, or Mao de pildo in Portuguese, and, together with another similar column in the vicinity, figures in the legend of the Paittina, a mythological personage from whom the Indians say that the serra has derived its name. The sandstone of Ereré is, for the most part, composed of fine, rounded grains of clear quartz, with a silicious cement, the rock being so excessively hard that a fracture passes directly through the sand grains. ‘The rock has a slight brownish tint, and a saccharine look, sometimes being almost translucent in thin flakes. On the surface the cement decomposes, becoming milk white, and the hard beds scale away in concentric coats, giving rise to rounded surfaces. This is the general character of the Ereré sandstone, but there are some very fine-grained layers like quartzite, while others are soft and friable. The rock is never very coarse, and pebbles are rare. The bedding is massive, and oblique lamination is everywhere observable. Underneath the sandstone at the notch of Aroxi there is a thick band of hardened, variegated clay. Being well jointed and of une- qual hardness, the Kreré sandstones, have, under denudation, given rise to a multitude of curious pillars and imitative forms. 'To the latter class belongs a large rock on the east extremity of the serra, called Pirayaudrat or porpoise, because of its resemblance to that animal, while near by, on the brink of a precipice, is a projecting, bird- like rock, called yuwrutaui. On the summit of the mountain, and overlooking the lofty precipice facing the village, is an immense, iso- lated rock, about fifty feet high, which, from afar, looks like a huge boulder perched upon the top of the serra. This mass, which is rep- resented in the cut on page 213, is composed of a very hard, white sandstone, obliquely laminated and rounded by decomposition. Its western side is covered with rude Indian drawings in red paint. *Pestle. Sometimes it is called Zadud, the mortar. J/éna means husband. By some the pillar is called yapona, the oven. +t Pird, fish, and yaudra, dog. 217 Standing just in front of the cliff at the upper part of the serra, on the northern side near the Aroxi notch, is a large, high pillar, coy- ered with similar rude paintings, and apparently at one time an object of superstitious regard. Similar figures are drawn on the cliffs near by and in the notch. These so-called hieroglyphics of Ereré were examined and copied by Wallace, but the sketches were unfortunately lost. I have reproduced some of the more important forms in the American Naturalist.* Mr. J. B. Steere, on a visit with me to the mountain, had the good fortune to find a large fragment of silicified wood, imbedded in the sandstone, near the great painted rock on top of the serra. ‘This is clearly coniferous in structure, but Dr. Dawson, to whom it has been referred, has not been able to determine it. Mr. Steere also found what appears to be the impression of the trunk of a large tree on the surface of a bed of sandstone, on a ridge about a quarter of a mile to the south- westward of the painted rock. One point in the geology of the Ereré District is settled upon the best of palaeontological evidence, and that is, the age of the beds forming the great plain to the north of the serra. These are cer- tainly Devonian. But what is the age of the rocks forming the serra itself? I have already expressed the opinion that the strata of the serra were disturbed before the beds forming the plains were laid down, since these strata are highly inclined, while the Devonian rocks bordering the base are quite horizontal, presenting nowhere more than an exceedingly slight inclination. There is no reason why coniferous wood should not occur in strata of Devonian, or even Upper Silurian age under the Equator; but I must freely confess, that, after carefully considering the whole subject, it seems to me quite probable that the Ereré sandstones are really newer than the fossiliferous beds of the plains, and that these last may dip under the serra; but, if this is the case, it is extraordinary that the sandstones, if once continuous over the plains, should have been so completely worn away and that the plains should have been so very evenly denuded. It is also somewhat strange that the structure of the serra of Tajuri should differ so markedly from that of Ereré. I have made a long and careful search for exposures along the base * Brazilian Rock Inscriptions, Amer. Nat., May, 1871. BUL. BUF. SOC. NAT. SCI. (28) JANUARY, 1874. 218 of the serra of Ereré, but I have been unable to determine from stratigraphical evidence the relative age of the beds of the moun- tains and plains. There can be no doubt that the serra of Ereré is older than the true table-topped hills, and the question of its glacial origin needs no further discussion. In the sandstone of the serra are occasional veins, partly com- posed of iron oxide. The original vein-rock appears to have been traversed by a perfect network of delicate veinlets of hematite, forming interlacing laminae often not more than one or two mil- limetres in thickness, which, on the decomposition and removal of the vein-rock, form masses presenting the appearance of honey- combed wood. In 1870, I made barometrical measurements on the summit of Ereré, which gave me the height as 970 feet. Since the observations were taken, I have noticed that a point to the west- ward of those I had chosen appears considerably higher, so that the serra is not far from 1,000 feet in height. The vegetation of the serra resembles that of the high, sandy campos of the vicinity, and is very scanty. The sandy tracts are sparsely sown with tufts of long, coarse grass. Trees are as usual very small, rough-barked, gnarly-branched, stunted and scorched by campos fires. Caju trees grow all over the serra, and the visitor will always gratefully remember their thirst-assuaging, acid fruit. The cajuciros of the serra are all very small, and the fruit is dwarfed and rather sour. On the sandy campos the tree is everywhere met with, and the fruit is sometimes very large and delicious. I have never seen a caja tree on the Devonian plain. It is a true campos species, and, as elsewhere in Brazil, it appears to be confined to dry, sandy soils. It flourishes also on the campos in the vicinity of Santarem, where, as well as at Monte-Alegre, a very delicious wine is made from its juice, some of the brands being not inferior to good grape wine. The manufacture of this beverage was known to the old Tupis, who called the liquor akayt kauim. The fruit has an extended reputation in Brazil for its anti-syphilitic properties, and it is supposed that the wine also possesses medicinal virtues. Two palms are common on the serra, the Sacurt and Jatd. The former appears to be allied to the Curud, but the leaves are much more stiff and erect. It is rarely seen elsewhere in the vicinity. The Jata grows to a height of about fifteen feet, and is a very con- 219 spicuous element in the vegetation of the serra. It also occurs on the campos. Armadillos and jabuti-tortoises abound in the serra, and a pretty little species of deer occurs, but I could never succeed in getting a shot at one. Before we leave the serra let us take a survey of the landscape. The eye follows the sandy campo, with its scattered trees and patches of bare sand, southward to the flat, insignificant-looking, rocky serra of Paituna, which, tied by the high campos to the serra of Ereré, forms a point projecting southward into the alluvial bottom of the Amazonas. On the right, or west of Paittina, the alluvial lands form a sort of bay, bordered by sandy campos-land. Into this region I made an excursion in 1870. From the Aroxi notch the sandy and sparsely-wooded plain slopes gradually from the moun- tains to the southward, for a few miles, to a little igarapé, called, I believe, Maxira; but this name I have also heard applied to the serra of Aroxi. Crossing the stream, one finds on the opposite side a line of terraces rising about 10-15 feet, if I rightly remem- ber, above the general level, but considerably more above the Ama- zonas. ‘These terraces are composed of beds of variegated sands and clays, in which I made an unsuccessful search for fossils. This formation appears to occupy a large area to the westward, and the terraces mark an old shore-line when the land stood at a somewhat lower level than at present, and the Amazonas, still a broad arm of the sea, had not yet passed into the riverine condition. Between the terraces and Paittina is the alluvial bay just alluded to, in which is a small lake and a magnificent grove of miritis. The lake, I suspect, disappears during the dry months, as I do not find it repre- sented on one of my sketches. Eastward of the serra of Ereré, a high, rounded, sandy plateau stretches off to the igarapé, on the opposite side of which the Monte-Alegre highlands run off obliquely to the villa, in a line of steep slopes. Between these highlands and Paituna is the alluvial bay traversed by the igarapé of Ereré. Across its mouth stretches the Curupattiba, and southward lie the beautiful, smiling plains, beyond which is the Amazonas, with the long, level line of smoke of a descending steamer. We trace to the northward the ridge of Monte-Alegre, at first level-topped, then more and more irregular, to the splendid, blue, mountain mass of Tauajuri, which, with pre- 220 cipitous front, heaves its back against the horizon, like a giant wave ready to break upon the level plains of Hreré, that lie spread out before us, flecked with open, bright, grassy campos, dark woodland, and coursing cloud-shadows from the glorious sky above. Below us, and beyond a little strip of woodland, is the little village of Ereré, with its white church and scattered, thatched houses. In the west are the tops of Aroxi and Aracuri, with low lands beyond on the horizon, while, northward from the hills, stretches a belt of low, wooded ridges, skirting the campos on the west and north, and bending round to close the circuit with Tauajuri. And away beyond them, on the far-off northern horizon, are table-topped hills, evidently of the same formation as the serras of Part. To give a clearer idea of the topographical features of the highlands west of the campos and of the distant table-topped hills, I have introduced the following little outline sketch taken without alteration from my note-book. Sketch looking Northward from Serra of Ereré. A mile or more west of the village, a very narrow, angular ridge extends northward from the northern side of the serra of Ereré, in a straight line for perhaps a mile, presenting a very even height of about 200 feet, as nearly as I could judge. On the eastern side this ridge is very steep, and near the top there is a line of exposures of a rather compact, not well laminated clay-rock, mottled red and white, and apparently without fossils. This has a decided dip to the westward, and the western slope of the ridge is consequently less steep than the eastern. The ridge is unfortunately covered with small trees, abominable “ Devil’s fish-hooks” and cactuses, so that it is very difficult to study it. After running along for a con- siderable distance, it breaks down abruptly, or perhaps more prop- erly speaking, it is cut through by a broad gap, through which runs the road to Maecurt. In the gap, the lower part of the ridge to a considerable height, is seen to be composed of a heavy mass of diorite; but whether this 221 rock forms a dyke, or a bed interstratified with the clay-rock, I could not determine. . In the rain-courses of the Maecurt road the diorite has given rise to a great number of well-rounded boulders of decomposition, imbedded in a dark soil of decomposed trap ; and, at a hasty glance, they might be taken for erratics. On the northern side of the gap the ridge appears to be continued for some distance. Looking from the top of Ereré there appears to be a ridge running northward from the Serra de Aroxi like that just de- scribed. I made an attempt to reach it, but lost myself in the thick woods. An attempt to explore the zone of highlands to the west of the campos proved very unsatisfactory. I made a very long excursion among these hills, but I cannot give an intelligible account of their structure, because of the want of exposures and the difficulty of making and recording observations in the dense undergrowth, and in the beds of the exceedingly tortuous igarapés. The prevailing rock appears to be similar to that exposed in the ridge extending northward from Ereré, but I found also a few wretched exposures of a firmly laminated, dead-black shale without fossils. I know nothing of the relation borne by these beds to the undoubted Devonian beds of the plains. Trap dykes are very nu- merous, and some are very heavy. The whole region seems to have been much disturbed. At Matarupi and elsewhere in the vicinity there are superficial deposits of impure haematitic iron ore. Cam- pos, apparently composed of Devonian rock, extend from the ridge running north from the serra of Ereré to the serra of Aroxi. Almost directly north of Monte-Alegre is an isolated, precipitous hill several hundred feet in height, which, in company with Messrs. Smith and Staunton and my guide Sr. Liberato, I tried to reach from the campo on the southern side. All I was able to do was to climb a sort of high platform, in front of the hill, which was so covered with spiny plants, yurupart pindé and underbrush, that I was obliged to turn back. I should have persisted, but that I had sey- eral hours’ march before me over the stony plain to Ereré that evening. I could only determine that the platform above spoken of was composed of diorite like that of the ridge just west of Ereré. The little hamlet of Ereré is situated on the Devonian plain, a little more than a mile to the north of the eastern extremity of the 222 serra of Ereré, and consists of some twenty to thirty miserable thatched houses and a neat little chapel. The inhabitants are civil- ized Indians, of more or less mixed blood, but it is not known from what tribe or tribes they are descended. The old people still speak the Tupi language, but it is becoming so rapidly superseded by the Portuguese that it is only rarely used for conversational purposes. The people are quiet, orderly, and clean, and I came to have a real respect for them. Sr. Liberato, my host, is a fine, intelligent, trust- worthy fellow, to whom I am under deep obligations for the faith- ful way in which he served me on both visits to Hreré, and I take pleasure in recommending him as a guide to future visitors. The men of Ereré are fishers, hunters, vaqueiros, and, like other Indians, work well when they must. Of the industry of the women I cannot speak in too high praise. On them falls all the labor of the field and household; from morning to night they are steadily at work, and I never think of Ereré without fancying that I still hear the measured rhythmic beat of the carana wand, in beating cotton for spinning. The sandy ridge or plateau east of Ereré shows but few superficial, and not very interesting exposures. Like the Monte-Alegre high- lands, it appears to consist of soft Tertiary beds, horizontally strati- fied, which have been much denuded down and superficially worked over, the clayey particles haying been washed out, leaving the sand lying loose on the surface. On the northern side of the ridge, at some distance east of the serra, is a small, isolated hill composed of fine clayey sands, white, variegated with purple, together with white sands, sufficiently compacted to form a low bluff, that runs round the eastern side of the hill. The ridge behind is composed of the same materials, as is seen In several deep rain-courses. On the hill just described, and in its immediate vicinity, I picked up several loose fragments of a very curious rock which I was unable to find in place. It consists of iron-oxide and is filled with little, empty cell-like cavities separated by very thin walls, and consequently spongy and very light. Each cavity corresponds to a sand-grain which has been dissolyed out, leaving only the iron oxide that cemented the whole together. The grains were probably calcareous, but I have no clue to the origin of these very interesting fragments. 223 The serra of Tauajuri,* though in plain sight from Ereré and from the vicinity of Monte-Alegre, is quite unknown to the white inhabitants of these places, and I found none except Indians who had visited it. Failing to reach the mountain in 1870, I made an excur- sion thither the following year, in company with Messrs. Derby and J. B. Steere. We left Monte-Alegre on foot at day-break, accompa- nied by four Indians, striking off northward over the highlands, following the road to Saudoso, a little agricultural settlement, situ- ated on the low grounds east of the ridge. The Monte-Alegre plateau is noted for its flat, rounded outlines, its long, gentle slopes, rarely gullied by rains, its superficial coating of coarse sand, and its peculiar campos vegetation, in all which features it agrees with the similar elevated, sandy campos of the vicinity of Ereré and Paituna, and also with those of Santarem, which last I shall not attempt to describe here. The covering of loose, coarse sand completely masks the geological structure of the plateau, except along its southern border and in a few localities where the underlying beds come to the surface in knolls. Here and there on the road, across the plateau, from Ereré to Monte-Alegre, one meets with slight knolls composed of small, ferruginous concre- tions, cemented together and resembling a conglomerate. The sur- face sands are so coarse and loose that it is very fatiguing to walk over them. The vegetation they support to-day is that of the high, sandy campos districts everywhere in northern Brazil, modified by campos fires. The sandy campos of the Ereré-Monte-Alegre district closely resemble those of Piauhy, Pernambuco and Bahia. Trees are sparsely sown, and, having been singed by fire, are small, rough- barked, stout and gnarly-branched, and thick-leaved. A large pro- portion of the trees are cajus, with whose grateful acid fruit the traveler may refresh himself. Grass grows only in widely separated tufts, and the surface is yearly burned over. The effect of these campos fires is most disastrous, and if kept up they must inevitably convert the ridge into a desert. *T am not sure that this is the correct form of the name of the serra. The pronunciation varies from Tajuri to Tayuri, Tauayuri, Tauajuri, and I have even heard Tauacuri. Penna uses Tauajury, and this appears more nearly right, but it would still be a Portuguese form. In all this uncertainty it seems scarcely worth while to inquire into the origin of thename. The first point to be settled is, whether the first part of the word, in lingoa geral, is itd, stone, or taud, a kind of clay. 224 The Monte-Alegre campos are quite unfit for agricultural pur- poses, but according to Sr. Valente, who accompanied us for a part of the way to Tauajuri, beans and even corn may be grown during the wet months; but mandioca cannot be raised on these lands, because it requires at least six months to mature, and, during the rains, the roots are apt to be washed out of the soil. The climate of the Ereré-Monte-Alegre district, during the dry season, is very pleasant. Day after day, and week after week passes without astorm. The days are hot, the thermometer in the shade ranging about 90° in the middle of the day; but the air is so dry and there is so constantly a stiff sea breeze blowing, that the temperature in-doors is very agreeable. On the plains, I have found the heat oppressive while in exercise, though much more endurable than in the interior of New York in the summer months; but the moment one stands still, even on the open plains, he is apt to be chilled by the breeze. The nights are very cool, and one is obliged to sleep wrapped in a blanket and with closed doors. Late in the dry season and in the rainy months, the mosquitoes are a veritable plague. Of the wet season on the Amazonas I can say nothing from my own personal acquaintance. As the plateau approaches Tauajuri it becomes more broken, and better wooded, but it soon gives way to hills, probably of a differ- ent geological structure. The lowlands east of the ridge are well wooded, but, except in marshy places, the forest is not luxuriant, and the same seems to be the case with the higher plains of the vicinity. We reached Jacaré at the foot of the serra at 3 o’clock Pp. M., hay- ing rested for dinner at Saudoso for perhaps a couple of hours, so that the distance from Monte-Alegre to the base of the mountain must be about 18 miles. At Jacaré we found a ruined house, and as we had outwalked our guides and were obliged to wait until late in the afternoon for them to come up, we here spent the night, as well as the carapanas and the white ants, that swarmed from the rotten timbers of the house, would permit. On the banks of a little, clear-water igarapé that runs through the forest, bordered by beautiful palms, we found sandstones, and I discovered a bed of dark-bluish limestone, that looked as though it ought to contain fossils, but afforded us nothing recognizable. Its strike was N. 8., and the dip 30° to the eastward. 225 Early the next morning we climbed the serra by a very rough, steep ascent through the woods over loose rocks, and worked our way with much difficulty nearly to the western end of the moun- tain. The serra is a sharp-crested monoclinal ridge, trending ap- proximately E. 8. E., W. N. W., and much longer than Ereré. The southern side is exceedingly steep, almost precipitous, and wooded nearly to the top, along which runs a line of low bluffs. The north- ern side slopes off at an angle of 10°-15° in a series of beautiful campos interspersed here and there with trees. This side of the serra is scored deeply with deep parallel gorges that extend in many cases up to and through the crest of the serra, which consequently presents a notched appearance when seen from the south. The uppermost stratum observed near the crest of the serra was a light bluish, nearly white, tough, not well laminated clay-rock, with a large percentage of very fine sand in its composition. Beneath this are beds of fine, clayey sandstone, white, mottled with purplish, and with fucoid (?) casts, alternating with which beds are shaley bands and layers of sandstone, the whole not well exposed. Then follow about 4 inches of red shaley iron-stone, overlying a bed of rather coarse sandstone about 10-15 feet in thickness, which forms a bluff running along the upper part of the southern side of the serra, while underneath are light purplish brown, fine-grained sand- stones poorly exposed. The dip of the Tauajuri beds in the serra is about 10°-15° towards the N. N. E. or N. E. I found the elevation of the serra at its highest point to be 8590 feet above the level of the sea.* T'auajuri appears to differ entirely from Ereré in its geological structure. It is, indeed, true that I examined only the upper beds of the series, but if the Ereré sand- stone were represented lower down, it is hardly possible that it should not have shown itself in bluffs on the mountain side.t *]I made butasingle observation, and as the mountain looks much higher than Ereré, 1 suspect that the observation may be unreliable. + Tuajuri is resorted to by the Indians of Monte-Alegre for the purpose of gathering the bark of the cumaté or cumati tree (Apocynea vel Asclepidea follicularis ? v. Mart. Glossarios, p. 393, sub voce cumat?), the sap extracted from which is used to varnish the drinking gourds, for the manufacture of which Monte-Alegre has been so long famed. The name of the tree appears to be derived from kamy’g, milk, sap, and e¢é, true. Cumaté probably more nearly preserves the original form than cwmat2, but I suspect it is still a corrupt form. The sap is obtained from the bark, I believe, by pounding and squeezing. The: cvias are prepared as follows: The BUL. BUF. SOC. NAT. SCI. (29) JANUARY, 1874. 226 From the summit there is a magnificent view over an immense area of country, the whole Monte-Alegre—Ereré highlands and the great Devonian plain being distinctly seen. I have reproduced from my note-book a little sketch of the Ereré hills taken from the top of Tauajuri, because it shows a line of hills extending westward be- yond Aracuri, apparently forming parts of the same outcrop. a " win writ —s—td —_ The Devonian plain and serras of Ereré from the Serra of Tauajuri. To the northward of the zone of highlands bordering the Ereré plain on the north and west, the country is low, somewhat irregu- lar, though with but few hills, and uniformly covered with forest. Along the horizon, on the north-west, high, table-topped hills stretch along for many miles like a cordilheira. To the east of Tauajuri the country is low, but still considerably higher than the Amazonian bottom. Just east of the Monte-Alegre highlands these higher grounds do not come down to the river, but their margin, once an old shore-line, describes a strong curve forming a sort of bay which has been silted up and converted into alluvial grassy campos, while, skirting the old shore, is a long, narrow, crescent-shaped lake, once a side-channel of the river. This alluvial bay and lake put one in mind of the campos and parana-merins of Taperinha, of which I hope to speak in another paper. From what I have seen of the Amazonian valley in the province of Para, Iam of the opinion that the greater part of the country gourd, or fruit of the Crescentia Cuyeté (kuia-eté=cuia par excellence) is cut in two and the inside pulp removed. When the rind is dry it is carefully scraped, both inside and out, and polished with the sandpaper-like leaves of the caimbé tree (Curatelia). Cmts tar a a a } ea eee aa yO Og MIA G66 oe Stay uN UY YUL YSU i SY veun: MOUS Wy Wi al filthy SO, staan phebe baccaace : wi GAN aawoa ce “Wed \ A'S iN WWE. WO NG IN) Wy yee ¥ Mov NN v VEY \ y | Wy Nyy = v vy SACI, Yu dy AU IOR Yl WUYY UY Suoy oak Jy at ve Ne We ude uelys WY vit Wvwwvuvev" Wei WIM Wc eve r SNe Uy Si CLARY : wit saath Wi Wie The PIN Wildy raat ) Weel V ie iv ‘ sy Mis SIS eed cet vd Vy, Veue Sl wre “SUR awe ee uate Nita Su Ga N Pes very a We SUV Mi WW UUM “VS WW J sd ove UN WEUY | WW WW es anise aociga pays wn ty weppnicnonnci cuca Weeesscoa sitio neo ude wily nn ' Wy saliva V Tee Gl Kl wee TOW AL SERGI yee MIAME Sov i ne WivwW. : yy i iiayi om BUN WW ean Ie Werte MULTE NEM My: WW Gee ue it wy WWeWiiiyuW¥ vnc Wanktinnnne vic WW Wve wesc sitet te WW iccuceerevivc soe ti ¥S.Y Wii sug! “Wve Sc Minne A Ye uu eo A GET WAM Md rt ee wy ois Wuyi levine J WIV “ wy wurst wy it Sec NG AM ' WW . ay wai " V V BUYS y y9e yi} ; Wasa’ “eouse’, we Ws Th: ~ Keine sa Sy EBs se VINYLS UG yygyyy ues rehash A Bt ep tia E EN a Nal Voter th on Woy U ier Ube wage ae agg u hy ny u ees Skee ey seeee Wi since ¥ ‘ slain ar Wade We ow i a i Nt Teg Mate = ae ae?