We had been two months in the canoes; from the 27th of February to the 26th of April. We had gone over 750 kilometres. The river from its source, near the thirteenth degree, to where it became navigable and we entered it, had a course of some 200 kilometres—probably more, perhaps 300 kilometres. Therefore we had now put on the map a river nearly 1,000 kilometres in length of which the existence was not merely unknown but impossible if the standard maps were correct. But this was not all. It seemed that this river of 1,000 kilometres in length was really the true upper course of the Aripuanan proper, in which case the total length was nearly 1,500 kilometres. Pyrineus had been waiting for us over a month, at the junction of what the rubbermen called the Castanho and of what they called the upper Aripuanan. (He had no idea as to which stream we would appear upon, or whether we would appear upon either.) On March 26 he had measured the volume of the two, and found that the Castanho, although the narrower, was the deeper and swifter, and that in volume it surpassed the other by 84 cubic metres a second. Since then the Castanho had fallen; our measurements showed it to be slightly smaller than the other; the volume of the river after the junction was about 4,500 cubic metres a second. This was in 7 degrees 34 minutes.

We were glad indeed to see Pyrineus and be at his attractive camp. We were only four hours above the little river hamlet of Sao Joao, a port of call for rubber-steamers, from which the larger ones go to Manaos in two days. These steamers mostly belong to Senhor Caripe. From Pyrineus we learned that Lauriado and Fiala had reached Manaos on March 26. On the swift water in the gorge of the Papagaio Fiala's boat had been upset and all his belongings lost, while he himself had narrowly escaped with his life. I was glad indeed that the fine and gallant fellow had escaped. The Canadian canoe had done very well. We were no less rejoiced to learn that Amilcar, the head of the party that went down the Gy-Parana, was also all right, although his canoe too had been upset in the rapids, and his instruments and all his notes lost. He had reached Manaos on April 10. Fiala had gone home. Miller was collecting near Manaos. He had been doing capital work.

The piranhas were bad here, and no one could bathe. Cherrie, while standing in the water close to the shore, was attacked and bitten; but with one bound he was on the bank before any damage could be done.

We spent a last night under canvas, at Pyrineus' encampment. It rained heavily. Next morning we all gathered at the monument which Colonel Rondon had erected, and he read the orders of the day. These recited just what had been accomplished: set forth the fact that we had now by actual exploration and investigation discovered that the river whose upper portion had been called the Duvida on the maps of the Telegraphic Commission and the unknown major part of which we had just traversed, and the river known to a few rubbermen, but to no one else, as the Castanho, and the lower part of the river known to the rubbermen as the Aripuanan (which did not appear on the maps save as its mouth was sometimes indicated, with no hint of its size) were all parts of one and the same river; and that by order of the Brazilian Government this river, the largest affluent of the Madeira, with its source near the 13th degree and its mouth a little south of the 5th degree, hitherto utterly unknown to cartographers and in large part utterly unknown to any save the local tribes of Indians, had been named the Rio Roosevelt.

We left Rondon, Lyra, and Pyrineus to take observations, and the rest of us embarked for the last time on the canoes, and, borne swiftly on the rapid current, we passed over one set of not very important rapids and ran down to Senhor Caripe's little hamlet of Sao Joao, which we reached about one o'clock on April 27, just before a heavy afternoon rain set in. We had run nearly eight hundred kilometres during the sixty days we had spent in the canoes. Here we found and boarded Pyrineus's river steamer, which seemed in our eyes extremely comfortable. In the senhor's pleasant house we were greeted by the senhora, and they were both more than thoughtful and generous in their hospitality. Ahead of us lay merely thirty-six hours by steamer to Manaos. Such a trip as that we had taken tries men as if by fire. Cherrie had more than stood every test; and in him Kermit and I had come to recognize a friend with whom our friendship would never falter or grow less.

Early the following afternoon our whole party, together with Senhor Caripe, started on the steamer. It took us a little over twelve hours' swift steaming to run down to the mouth of the river on the upper course of which our progress had been so slow and painful; from source to mouth, according to our itinerary and to Lyra's calculations, the course of the stream down which we had thus come was about 1,500 kilometres in length—about 900 miles, perhaps nearly 1,000 miles— from its source near the 13th degree in the highlands to its mouth in the Madeira, near the 5th degree. Next morning we were on the broad sluggish current of the lower Madeira, a beautiful tropical river. There were heavy rainstorms, as usual, although this is supposed to be the very end of the rainy season. In the afternoon we finally entered the wonderful Amazon itself, the mighty river which contains one tenth of all the running water of the globe. It was miles across, where we entered it; and indeed we could not tell whether the farther bank, which we saw, was that of the mainland or an island. We went up it until about midnight, then steamed up the Rio Negro for a short distance, and at one in the morning of April 30 reached Manaos.

Manaos is a remarkable city. It is only three degrees south of the equator. Sixty years ago it was a nameless little collection of hovels, tenanted by a few Indians and a few of the poorest class of Brazilian peasants. Now it is a big, handsome modern city, with Opera house, tramways, good hotels, fine squares and public buildings, and attractive private houses. The brilliant coloring and odd architecture give the place a very foreign and attractive flavor in northern eyes. Its rapid growth to prosperity was due to the rubber trade. This is now far less remunerative than formerly. It will undoubtedly in some degree recover; and in any event the development of the immensely rich and fertile Amazonian valley is sure to go on, and it will be immensely quickened when closer connections are made with the Brazilian highland country lying south of it.

Here we found Miller, and glad indeed we were to see him. He had made good collections of mammals and birds on the Gy-Parana, the Madeira, and in the neighborhood of Manaos; his entire collection of mammals was really noteworthy. Among them was the only sloth any of us had seen on the trip. The most interesting of the birds he had seen was the hoatzin. This is a most curious bird of very archaic type. Its flight is feeble, and the naked young have spurs on their wings, by the help of which they crawl actively among the branches before their feathers grow. They swim no less easily, at the same early age. Miller got one or two nests, and preserved specimens of the surroundings of the nests; and he made exhaustive records of the habits of the birds. Near Megasso a jaguar had killed one of the bullocks that were being driven along for food. The big cat had not seized the ox with its claws by the head, but had torn open its throat and neck.

Every one was most courteous at Manaos, especially the governor of the state and the mayor of the city. Mr. Robiliard, the British consular representative, and also the representative of the Booth line of steamers, was particularly kind. He secured for us passages on one of the cargo boats of the line to Para, and thence on one of the regular cargo-and-passenger steamers to Barbados and New York. The Booth people were most courteous to us.

I said good-by to the camaradas with real friendship and regret. The parting gift I gave to each was in gold sovereigns; and I was rather touched to learn later that they had agreed among themselves each to keep one sovereign as a medal of honor and token that the owner had been on the trip. They were a fine set, brave, patient, obedient, and enduring. Now they had forgotten their hard times; they were fat from eating, at leisure, all they wished; they were to see Rio Janeiro, always an object of ambition with men of their stamp; and they were very proud of their membership in the expedition.

Later, at Belen, I said good-by to Colonel Rondon, Doctor Cajazeira, and Lieutenant Lyra. Together with my admiration for their hardihood, courage, and resolution, I had grown to feel a strong and affectionate friendship for them. I had become very fond of them; and I was glad to feel that I had been their companion in the performance of a feat which possessed a certain lasting importance.

On May 1 we left Manaos for Belen-Para, as until recently it was called. The trip was interesting. We steamed down through tempest and sunshine; and the towering forest was dwarfed by the giant river it fringed. Sunrise and sunset turned the sky to an unearthly flame of many colors above the vast water. It all seemed the embodiment of loneliness and wild majesty. Yet everywhere man was conquering the loneliness and wresting the majesty to his own uses. We passed many thriving, growing towns; at one we stopped to take on cargo. Everywhere there was growth and development. The change since the days when Bates and Wallace came to this then poor and utterly primitive region is marvellous. One of its accompaniments has been a large European, chiefly south European, immigration. The blood is everywhere mixed; there is no color line, as in most English-speaking countries, and the negro and Indian strains are very strong; but the dominant blood, the blood already dominant in quantity, and that is steadily increasing its dominance, is the olive-white.

Only rarely did the river show its full width. Generally we were in channels or among islands. The surface of the water was dotted with little islands of floating vegetation. Miller said that much of this came from the lagoons such as those where he had been hunting, beside the Solimoens—lagoons filled with the huge and splendid Victoria lily, and with masses of water hyacinths. Miller, who was very fond of animals and always took much care of them, had a small collection which he was bringing back for the Bronx Zoo. An agouti was so bad- tempered that he had to be kept solitary; but three monkeys, big, middle-sized, and little, and a young peccary formed a happy family. The largest monkey cried, shedding real tears, when taken in the arms and pitied. The middle-sized monkey was stupid and kindly, and all the rest of the company imposed on it; the little monkey invariably rode on its back, and the peccary used it as a head pillow when it felt sleepy.

Belen, the capital of the state of Para, was an admirable illustration of the genuine and almost startling progress which Brazil has been making of recent years. It is a beautiful city, nearly under the equator. But it is not merely beautiful. The docks, the dredging operations, the warehouses, the stores and shops, all tell of energy and success in commercial life. It is as clean, healthy, and well policed a city as any of the size in the north temperate zone. The public buildings are handsome, the private dwellings attractive; there are a fine opera-house, an excellent tramway system, and a good museum and botanical gardens. There are cavalry stables, where lights burn all night long to protect the horses from the vampire bats. The parks, the rows of palms and mango-trees, the open-air restaurants, the gay life under the lights at night, all give the city its own special quality and charm. Belen and Manaos are very striking examples of what can be done in the mid-tropics. The governor of Para and his charming wife were more than kind.

Cherrie and Miller spent the day at the really capital zoological gardens, with the curator, Miss Snethlage. Miss Snethlage, a German lady, is a first rate field and closet naturalist, and an explorer of note, who has gone on foot from the Xingu to the Tapajos. Most wisely she has confined the Belen zoo to the animals of the lower Amazon valley, and in consequence I know of no better local zoological gardens. She has an invaluable collection of birds and mammals of the region; and it was a privilege to meet her and talk with her.

We also met Professor Farrabee, of the University of Pennsylvania, the ethnologist. He had just finished a very difficult and important trip, from Manaos by the Rio Branco to the highlands of Guiana, across them on foot, and down to the seacoast of British Guiana. He is an admirable representative of the men who are now opening South America to scientific knowledge.

On May 7 we bade good-by to our kind Brazilian friends and sailed northward for Barbados and New York.

Zoologically the trip had been a thorough success. Cherrie and Miller had collected over twenty-five hundred birds, about five hundred mammals, and a few reptiles, batrachians, and fishes. Many of them were new to science; for much of the region traversed had never previously been worked by any scientific collector.

Of course, the most important work we did was the geographic work, the exploration of the unknown river, undertaken at the suggestion of the Brazilian Government, and in conjunction with its representatives. No piece of work of this kind is ever achieved save as it is based on long continued previous work. As I have before said, what we did was to put the cap on the pyramid that had been built by Colonel Rondon and his associates of the Telegraphic Commission during the six previous years. It was their scientific exploration of the chapadao, their mapping the basin of the Juruena, and their descent of the Gy- Parana that rendered it possible for us to solve the mystery of the River of Doubt.

The work of the commission, much the greatest work of the kind ever done in South America, is one of the many, many achievements which the republican government of Brazil has to its credit. Brazil has been blessed beyond the average of her Spanish-American sisters because she won her way to republicanism by evolution rather than revolution. They plunged into the extremely difficult experiment of democratic, of popular, self-government, after enduring the atrophy of every quality of self-control, self-reliance, and initiative throughout three withering centuries of existence under the worst and most foolish form of colonial government, both from the civil and the religious standpoint, that has ever existed. The marvel is not that some of them failed, but that some of them have eventually succeeded in such striking fashion. Brazil, on the contrary, when she achieved independence, first exercised it under the form of an authoritative empire, then under the form of a liberal empire. When the republic came, the people were reasonably ripe for it. The great progress of Brazil—and it has been an astonishing progress—has been made under the republic. I could give innumerable examples and illustrations of this. The change that has converted Rio Janeiro from a picturesque pest-hole into a singularly beautiful, healthy, clean, and efficient modern great city is one of these. Another is the work of the Telegraphic Commission.

We put upon the map a river some fifteen hundred kilometres in length, of which the upper course was not merely utterly unknown to, but unguessed at by, anybody; while the lower course, although known for years to a few rubbermen, was utterly unknown to cartographers. It is the chief affluent of the Madeira, which is itself the chief affluent of the Amazon.

The source of this river is between the 12th and 13th parallels of latitude south and the 59th and 60th degrees of longitude west from Greenwich. We embarked on it at about latitude 12 degrees 1 minute south, and about longitude 60 degrees 15 minutes west. After that its entire course lay between the 60th and 61st degrees of longitude, approaching the latter most closely about latitude 8 degrees 15 minutes. The first rapids we encountered were in latitude 11 degrees 44 minutes, and in uninterrupted succession they continued for about a degree, without a day's complete journey between any two of them. At 11 degrees 23 minutes the Rio Kermit entered from the left, at 11 degrees 22 minutes the Rio Marciano Avila from the right, at 11 degrees 18 minutes the Taunay from the left, at 10 degrees 58 minutes the Cardozo from the right. In 10 degrees 24 minutes we encountered the first rubbermen. The Rio Branco entered from the left at 9 degrees 38 minutes. Our camp at 8 degrees 49 minutes was nearly on the boundary between Matto Grosso and Amazonas. The confluence with the Aripuanan, which joined from the right, took place at 7 degrees 34 minutes. The entrance into the Madeira was at about 5 degrees 20 minutes (this point we did not determine by observation, as it is already on the maps). The stream we had followed down was from the river's highest sources; we had followed its longest course.



APPENDIX A.

The Work of the Field Zoologist and Field Geographer in South America

Portions of South America are now entering on a career of great social and industrial development. Much remains to be known, so far as the outside world is concerned, of the social and industrial condition in the long-settled interior regions. More remains to be done, in the way of pioneer exploring and of scientific work, in the great stretches of virgin wilderness. The only two other continents where such work, of like volume and value, remains to be done are Africa and Asia; and neither Africa nor Asia offers a more inviting field for the best kind of field worker in geographical exploration and in zoological, geological, and paleontological investigation. The explorer is merely the most adventurous kind of field geographer; and there are two or three points worth keeping in mind in dealing with the South American work of the field geographer and field zoologist.

Roughly, the travellers who now visit (like those who for the past century have visited) South America come in three categories— although, of course, these categories are not divided by hard-and-fast lines.

First, there are the travellers who skirt the continent in comfortable steamers, going from one great seaport to another, and occasionally taking a short railway journey to some big interior city not too far from the coast. This is a trip well worth taking by all intelligent men and women who can afford it; and it is being taken by such men and women with increasing frequency. It entails no more difficulty than a similar trip to the Mediterranean—than such a trip which to a learned and broad-minded observer offers the same chance for acquiring knowledge and, if he is himself gifted with wisdom, the same chance of imparting his knowledge to others that is offered by a trip of similar length through the larger cities of Europe or the United States. Probably the best instance of the excellent use to which such an observer can put his experience is afforded by the volume of Mr. Bryce. Of course, such a trip represents travelling of essentially the same kind as travelling by railroad from Atlanta to Calgary or from Madrid to Moscow.

Next there are the travellers who visit the long-settled districts and colonial cities of the interior, travelling over land or river highways which have been traversed for centuries but which are still primitive as regards the inns and the modes of conveyance. Such travelling is difficult in the sense that travelling in parts of Spain or southern Italy or the Balkan states is difficult. Men and women who have a taste for travel in out-of-way places and who, therefore, do not mind slight discomforts and inconveniences have the chance themselves to enjoy, and to make others profit by, travels of this kind in South America. In economic, social, and political matters the studies and observations of these travellers are essential in order to supplement, and sometimes to correct, those of travellers of the first category; for it is not safe to generalize overmuch about any country merely from a visit to its capital or its chief seaport. These travellers of the second category can give us most interesting and valuable information about quaint little belated cities; about backward country folk, kindly or the reverse, who show a mixture of the ideas of savagery with the ideas of an ancient peasantry; and about rough old highways of travel which in comfort do not differ much from those of mediaeval Europe. The travellers who go up or down the highway rivers that have been travelled for from one to four hundred years—rivers like the Paraguay and Parana, the Amazon, the Tapajos, the Madeira, the lower Orinoco—come in this category. They can add little to our geographical knowledge; but if they are competent zoologists or archaeologists, especially if they live or sojourn long in a locality, their work may be invaluable from the scientific standpoint. The work of the archaeologists among the immeasurably ancient ruins of the low-land forests and the Andean plateaux is of this kind. What Agassiz did for the fishes of the Amazon and what Hudson did for the birds of the Argentine are other instances of the work that can thus be done. Burton's writings on the interior of Brazil offer an excellent instance of the value of a sojourn or trip of this type, even without any especial scientific object.

Of course travellers of this kind need to remember that their experiences in themselves do not qualify them to speak as wilderness explorers. Exactly as a good archaeologist may not be competent to speak of current social or political problems, so a man who has done capital work as a tourist observer in little-visited cities and along remote highways must beware of regarding himself as being thereby rendered fit for genuine wilderness work or competent to pass judgment on the men who do such work. To cross the Andes on mule-back along the regular routes is a feat comparable to the feats of the energetic tourists who by thousands traverse the mule trails in out-of-the-way nooks of Switzerland. An ordinary trip on the highway portions of the Amazon, Paraguay, or Orinoco in itself no more qualifies a man to speak of or to take part in exploring unknown South American rivers than a trip on the lower Saint Lawrence qualifies a man to regard himself as an expert in a canoe voyage across Labrador or the Barren Grounds west of Hudson Bay.

A hundred years ago, even seventy or eighty years ago, before the age of steamboats and railroads, it was more difficult than at present to define the limits between this class and the next; and, moreover, in defining these limits I emphatically disclaim any intention of thereby attempting to establish a single standard of value for books of travel. Darwin's "Voyage of the Beagle" is to me the best book of the kind ever written; it is one of those classics which decline to go into artificial categories, and which stand by themselves; and yet Darwin, with his usual modesty, spoke of it as in effect a yachting voyage. Humboldt's work had a profound effect on the thought of the civilized world; his trip was one of adventure and danger; and yet it can hardly be called exploration proper. He visited places which had been settled and inhabited for centuries and traversed places which had been travelled by civilized men for years before he followed in their footsteps. But these places were in Spanish colonies, and access to them had been forbidden by the mischievous and intolerant tyranny— ecclesiastical, political, and economic—which then rendered Spain the most backward of European nations; and Humboldt was the first scientific man of intellectual independence who had permission to visit them. To this day many of his scientific observations are of real value. Bates came to the Amazon just before the era of Amazonian steamboats. He never went off the native routes of ordinary travel. But he was a devoted and able naturalist. He lived an exceedingly isolated, primitive, and laborious life for eleven years. Now, half a century after it was written, his "Naturalist on the Amazon" is as interesting and valuable as it ever was, and no book since written has in any way supplanted it.

Travel of the third category includes the work of the true wilderness explorers who add to our sum of geographical knowledge and of the scientific men who, following their several bents, also work in the untrodden wilds. Colonel Rondon and his associates have done much in the geographical exploration of unknown country, and Cherrie and Miller have penetrated and lived for months and years in the wastes, on their own resources, as incidents to their mammalogical and ornithological work. Professor Farrabee, the anthropologist, is a capital example of the man who does this hard and valuable type of work.

An immense amount of this true wilderness work, geographical and zoological, remains to be done in South America. It can be accomplished with reasonable thoroughness only by the efforts of very many different workers, each in his own special field. It is desirable that here and there a part of the work should be done in outline by such a geographic and zoological reconnaissance as ours; we would, for example, be very grateful for such work in portions of the interior of the Guianas, on the headwaters of the Xingu, and here and there along the eastern base of the Andes.

But as a rule the work must be specialized; and in its final shape it must be specialized everywhere. The first geographical explorers of the untrodden wilderness, the first wanderers who penetrate the wastes where they are confronted with starvation, disease, and danger and death in every from, cannot take with them the elaborate equipment necessary in order to do the thorough scientific work demanded by modern scientific requirements. This is true even of exploration done along the courses of unknown rivers; it is more true of the exploration, which must in South America become increasingly necessary, done across country, away from the rivers.

The scientific work proper of these early explorers must be of a somewhat preliminary nature; in other words the most difficult and therefore ordinarily the most important pieces of first-hand exploration are precisely those where the scientific work of the accompanying cartographer, geologist, botanist, and zoologist must be furthest removed from finality. The zoologist who works to most advantage in the wilderness must take his time, and therefore he must normally follow in the footsteps of, and not accompany, the first explorers. The man who wishes to do the best scientific work in the wilderness must not try to combine incompatible types of work nor to cover too much ground in too short a time.

There is no better example of the kind of zoologist who does first- class field-work in the wilderness than John D. Haseman, who spent from 1907 to 1910 in painstaking and thorough scientific investigation over a large extent of South American territory hitherto only partially known or quite unexplored. Haseman's primary object was to study the characteristics and distribution of South American fishes, but as a matter of fact he studied at first hand many other more or less kindred subjects, as may be seen in his remarks on the Indians and in his excellent pamphlet on "Some Factors of Geographical Distribution in South America."

Haseman made his long journey with a very slender equipment, his extraordinarily successful field-work being due to his bodily health and vigor and his resourcefulness, self-reliance, and resolution. His writings are rendered valuable by his accuracy and common sense. The need of the former of these two attributes will be appreciated by whoever has studied the really scandalous fictions which have been published as genuine by some modern "explorers" and adventurers in South America; and the need of the latter by whoever has studied some of the wild theories propounded in the name of science concerning the history of life on the South American continent. There is, however, one serious criticism to be made on Haseman: the extreme obscurity of his style—an obscurity mixed with occasional bits of scientific pedantry, which makes it difficult to tell whether or not on some points his thought is obscure also. Modern scientists, like modern historians and, above all, scientific and historical educators, should ever keep in mind that clearness of speech and writing is essential to clearness of thought and that a simple, clear, and, if possible, vivid style is vital to the production of the best work in either science or history. Darwin and Huxley are classics, and they would not have been if they had not written good English. The thought is essential, but ability to give it clear expression is only less essential. Ability to write well, if the writer has nothing to write about, entitles him to mere derision. But the greatest thought is robbed of an immense proportion of its value if expressed in a mean or obscure manner. Mr. Haseman has such excellent thought that it is a pity to make it a work of irritating labor to find out just what the thought is. Surely, if he will take as much pains with his writing as he has with the far more difficult business of exploring and collecting, he will become able to express his thought clearly and forcefully. At least he can, if he chooses, go over his sentences until he is reasonably sure that they can be parsed. He can take pains to see that his whole thought is expressed, instead of leaving vacancies which must be filled by the puzzled and groping reader. His own views and his quotations from the views of others about the static and dynamic theories of distribution are examples of an important principle so imperfectly expressed as to make us doubtful whether it is perfectly apprehended by the writer. He can avoid the use of those pedantic terms which are really nothing but offensive and, fortunately, ephemeral scientific slang. There has been, for instance, a recent vogue for the extensive misuse, usually tautological misuse, of the word "complexus"—an excellent word if used rarely and for definite purposes. Mr. Haseman drags it in continually when its use is either pointless and redundant or else serves purely to darken wisdom. He speaks of the "Antillean complex" when he means the Antilles, of the "organic complex" instead of the characteristic or bodily characteristics of an animal or species, and of the "environmental complex" when he means nothing whatever but the environment. In short, Mr. Haseman and those whose bad example he in this instance follows use "complexus" in much the same spirit as that displayed by the famous old lady who derived religious—instead of scientific— consolation from the use of "the blessed word Mesopotamia."

The reason that it is worth while to enter this protest against Mr. Haseman's style is because his work is of such real and marked value. The pamphlet on the distribution of South American species shows that to exceptional ability as a field worker he adds a rare power to draw, with both caution and originality, the necessary general conclusions from the results of his own observations and from the recorded studies of other men; and there is nothing more needed at the present moment among our scientific men than the development of a school of men who, while industrious and minute observers and collectors and cautious generalizers, yet do not permit the faculty of wise generalization to be atrophied by excessive devotion to labyrinthine detail.

Haseman upholds with strong reasoning the theory that since the appearance of all but the lowest forms of life on this globe there have always been three great continental masses, sometimes solid sometimes broken, extending southward from the northern hemisphere, and from time to time connected in the north, but not in the middle regions or the south since the carboniferous epoch. He holds that life has been intermittently distributed southward along these continental masses when there were no breaks in their southward connection, and intermittently exchanged between them when they were connected in the north; and he also upholds the view that from a common ancestral form the same species has been often developed in entirely disconnected localities when in these localities the conditions of environment were the same.

The opposite view is that there have been frequent connections between the great land masses, alike in the tropics, in the south temperate zone, and in the Antarctic region. The upholders of this theory base it almost exclusively on the distribution of living and fossil forms of life; that is, it is based almost exclusively on biological and not geological considerations. Unquestionably, the distribution of many forms of life, past and present, offers problems which with our present paleontological knowledge we are wholly unable to solve. If we consider only the biological facts concerning some one group of animals it is not only easy but inevitable to conclude that its distribution must be accounted for by the existence of some former direct land bridge extending, for instance, between Patagonia and Australia, or between Brazil and South Africa, or between the West Indies and the Mediterranean, or between a part of the Andean region and northeastern Asia. The trouble is that as more groups of animals are studied from the standpoint of this hypothesis the number of such land bridges demanded to account for the existing facts of animal distribution is constantly and indefinitely extended. A recent book by one of the most learned advocates of this hypothesis calls for at least ten such land bridges between South America and all the other continents, present and past, of the world since a period geologically not very remote. These land bridges, moreover, must, many of them, have been literally bridges; long, narrow tongues of land thrust in every direction across the broad oceans. According to this view the continental land masses have been in a fairly fluid condition of instability. By parity of reasoning, the land bridges could be made a hundred instead of merely ten in number. The facts of distribution are in many cases inexplicable with our present knowledge; yet if the existence of widely separated but closely allied forms is habitually to be explained in accordance with the views of the extremists of this school we could, from the exclusive study of certain groups of animals, conclude that at different periods the United States and almost every other portion of the earth were connected by land and severed from all other regions by water—and, from the study of certain other groups of animals, arrive at directly opposite and incompatible conclusions.

The most brilliant and unsafe exponent of this school was Ameghino, who possessed and abused two gifts, both essential to the highest type of scientist, and both mischievous unless this scientist possess a rare and accurate habit of thought joined to industry and mastery of detail:—namely, the gift of clear and interesting writing, and the gift of generalization. Ameghino rendered marked services to paleontology. But he generalized with complete recklessness from the slenderest data; and even these data he often completely misunderstood or misinterpreted. His favorite thesis included the origin of mammalian life and of man himself in southernmost South America, with, as incidents, the belief that the mammalian-bearing strata of South America were of much greater age than the strata with corresponding remains elsewhere; that in South America various species and genera of men existed in tertiary times, some of them at least as advanced as fairly well advanced modern savages; that there existed various land bridges between South America and other southern continents, including Africa; and that the ancestral types of modern mammals and of man himself wandered across one of these bridges to the old world, and that thence their remote descendants, after ages of time, returned to the new. In addition to valuable investigations of fossil-bearing beds in the Argentine, he made some excellent general suggestions, such as that the pithecoid apes, like the baboons, do not stand in the line of man's ancestral stem but represent a divergence from it away from humanity and toward a retrogressive bestialization. But of his main theses he proves none, and what evidence we have tells against them. At the Museum of La Plata I found that the authorities were practically a unit in regarding his remains of tertiary men and proto- men as being either the remains of tertiary American monkeys or of American Indians from strata that were long post-tertiary. The extraordinary discovery, due to that eminent scientist and public servant Doctor Moreno, of the remains of man associated with the remains of the great extinct South American fauna, of the mylodon, of a giant ungulate, of a huge cat like the lion, and of an extraordinary aberrant horse (of a wholly different genus from the modern horse) conclusively shows that in its later stages the South American fauna consisted largely of types that elsewhere had already disappeared and that these types persisted into what was geologically a very recent period only some tens of thousands of years ago, when savage man of practically a modern type had already appeared in South America. The evidence we have, so far as it goes, tends to show that the South American fauna always has been more archaic in type than the arctogeal fauna of the same chronological level.

To loose generalizations, and to elaborate misinterpretations of paleontological records, the kind of work done by Mr. Haseman furnishes an invaluable antiscorbutic. To my mind, he has established a stronger presumption in favor of the theory he champions than has been established in favor of the theories of any of the learned and able scientific men from whose conclusions he dissents. Further research, careful, accurate, and long extended, can alone enable us to decide definitely in the matter; and this research, to be effective, must be undertaken by many men, each of whom shall in large measure possess Mr. Haseman's exceptional power of laborious work both in the field and in the study, his insight and accuracy of observation, and his determination to follow truth with inflexible rectitude wherever it may lead—one of the greatest among the many great qualities which lifted Huxley and Darwin above their fellows.



APPENDIX B.

The Outfit for Travelling in the South American Wilderness

South America includes so many different kinds of country that it is impossible to devise a scheme of equipment which shall suit all. A hunting-trip in the pantanals, in the swamp country of the upper Paraguay, offers a simple problem. An exploring trip through an unknown tropical forest region, even if the work is chiefly done by river, offers a very difficult problem. All that I can pretend to do is to give a few hints as the results of our own experience.

For bedding there should be a hammock, mosquito-net, and light blanket. These can be obtained in Brazil. For tent a light fly is ample; ours were brought with us from New York. In exploring only the open fly should be taken; but on trips where weight of luggage is no objection, there can be walls to the tent and even a canvas floor- cloth. Camp-chairs and a camp table should be brought—any good outfitter in the United States will supply them—and not thrown away until it becomes imperative to cut everything down. On a river trip, first-class pulleys and ropes—preferably steel, and at any rate very strong—should be taken. Unless the difficulties of transportation are insuperable, canvas-and-cement canoes, such as can be obtained from various firms in Canada and the United States, should by all means be taken. They are incomparably superior to the dugouts. But on different rivers wholly different canoes, of wholly different sizes, will be needed; on some steam or electric launches may be used; it is not possible to lay down a general rule.

As regards arms, a good plain 12-bore shotgun with a 30-30 rifle- barrel underneath the others is the best weapon to have constantly in one's hand in the South American forests, where big game is rare and yet may at any time come in one's path. When specially hunting the jaguar, marsh-deer, tapir, or big peccary, an ordinary light repeating rifle—the 30-30, 30-40, or 256—is preferable. No heavy rifle is necessary for South America. Tin boxes or trunks are the best in which to carry one's spare things. A good medicine-chest is indispensable. Nowadays doctors know so much of tropical diseases that there is no difficulty in fitting one out. It is better not to make the trip at all than to fail to take an ample supply of quinine pills. Cholera pills and cathartic pills come next in importance. In liquid shape there should be serum to inject for the stoppage of amoebic dysentery, and anti-snake-venom serum. Fly-dope should be taken in quantities.

For clothing Kermit and I used what was left over from our African trip. Sun helmets are best in the open; slouch-hats are infinitely preferable in the woods. There should be hobnailed shoes—the nails many and small, not few and large; and also moccasins or rubber-soled shoes; and light, flexible leggings. Tastes differ in socks; I like mine of thick wool. A khaki-colored shirt should be worn, or, as a better substitute, a khaki jacket with many pockets. Very light underclothes are good. If one's knees and legs are unfortunately tender, knickerbockers with long stockings and leggings should be worn; ordinary trousers tend to bind the knee. Better still, if one's legs will stand the exposure, are shorts, not coming down to the knee. A kilt would probably be best of all. Kermit wore shorts in the Brazilian forest, as he had already worn them in Africa, in Mexico, and in the New Brunswick woods. Some of the best modern hunters always wear shorts; as for example, that first-class sportsman the Duke of Alva.

Mr. Fiala, after the experience of his trip down the Papagaio, the Juruena, and the Tapajos, gives his judgment about equipment and provisions as follows:

The history of South American exploration has been full of the losses of canoes and cargoes and lives. The native canoe made from the single trunk of a forest giant is the craft that has been used. It is durable and if lost can be readily replaced from the forest by good men with axes and adzes. But, because of its great weight and low free-board, it is unsuitable as a freight carrier and by reason of the limitations of its construction is not of the correct form to successfully run the rapid and bad waters of many of the South American rivers. The North American Indian has undoubtedly developed a vastly superior craft in the birch-bark canoe and with it will run rapids that a South American Indian with his log canoe would not think of attempting, though, as a general thing, the South American Indian is a wonderful waterman, the equal and, in some ways, the superior of his northern contemporary. At the many carries or portages the light birch-bark canoe or its modern representative, the canvas-covered canoe, can be picked up bodily and carried by from two to four men for several miles, if necessary, while the log canoe has to be hauled by ropes and back-breaking labor over rollers that have first to be cut from trees in the forest, or at great risk led along the edge of the rapids with ropes and hooks and poles, the men often up to their shoulders in the rushing waters, guiding the craft to a place of safety.

The native canoe is so long and heavy that it is difficult to navigate without some bumps on the rocks. In fact, it is usually dragged over the rocks in the shallow water near shore in preference to taking the risk of a plunge through the rushing volume of deeper water, for reasons stated above. The North American canoe can be turned with greater facility in critical moments in bad water. Many a time I heard my steersman exclaim with delight as we took a difficult passage between two rocks with our loaded Canadian canoe. In making the same passage the dugout would go sideways toward the rapid until by a supreme effort her three powerful paddlers and steersman would right her just in time. The native canoe would ship great quantities of water in places the Canadian canoe came through without taking any water on board. We did bump a few rocks under water, but the canoe was so elastic that no damage was done.

Our nineteen-foot canvas-covered freight canoe, a type especially built for the purpose on deep, full lines with high free-board, weighed about one hundred and sixty pounds and would carry a ton of cargo with ease—and also take it safely where the same cargo distributed among two or three native thirty or thirty-five foot canoes would be lost. The native canoes weigh from about nine hundred to two thousand five hundred pounds and more.

In view of the above facts the explorer-traveller is advised to take with him the North American canoe if he intends serious work. Two canoes would be a good arrangement for from five to seven men, with at least one steersman and two paddlers to each canoe. The canoes can be purchased in two sizes and nested for transportation, an arrangement which would save considerable expense in freight bills. At least six paddles should be packed with each boat, in length four and one half, four and three fourths, and five feet. Other paddles from six and one half feet to eight and one half feet should be provided for steering oars. The native paddler, after he has used the light Canadian paddle, prefers it to the best native make. My own paddlers lost or broke all of their own paddles so as to get the North American ones, which they marked with their initials and used most carefully.

To each canoe it would be well to have two copper air tanks, one fore, one aft, a hand-hole in each with a water-tight screw cover on hatch. In these tanks could be kept a small supply of matches, the chronometer or watch which is used for position, and the scientific records and diary. Of course, the fact should be kept in mind that these are air tanks, not to be used so as to appreciably diminish their buoyancy. Each canoe should also carry a small repair kit attached to one of the thwarts, containing cement, a piece of canvas same as cover of canoe, copper tacks, rivets, and some galvanized nails; a good hatchet and a hammer; a small can of canoe paint, spar varnish, and copper paint for worn places would be a protection against termites and torrential downpours. In concluding the subject of canoes I can state that the traveller in South America will find no difficulty in disposing of his craft at the end of his trip.

MOTORS—We had with us a three and one half horse-power motor which could be attached to stern or gunwale of canoe or boat. It was made by the Evinrude Motor Company, who had a magneto placed in the flywheel of the engine so that we never had to resort to the battery to run the motor. Though the motor was left out in the rain and sun, often without a cover, by careless native help, it never failed us. We found it particularly valuable in going against the strong current of the Sepotuba River where several all-night trips were made up-stream, the motor attached to a heavy boat. For exploration up-stream it would be valuable, particularly as it is easily portable, weighing for the two horse-power motor fifty pounds, for three and one half horse-power one hundred pounds. If a carburetor could be attached so that kerosene could be used it would add to its value many times, for kerosene can be purchased almost anywhere in South America.

TENTS—There is nothing better for material than the light waterproof Sea Island cotton of American manufacture, made under the trade name of waterproof silk. It keeps out the heaviest rain and is very light. Canvas becomes water-soaked, and cravenetted material lets the water through. A waterproof canvas floor is a luxury, and, though it adds to the weight, it may with advantage be taken on ordinary trips. The tent should be eight by eight or eight by nine feet, large enough to swing a comfortable hammock. A waterproof canvas bag, a loose-fitting envelope for the tent should be provided. Native help is, as a rule, careless, and the bag would save wear and tear.

HAMMOCKS—The hammock is the South American bed, and the traveller will find it exceedingly comfortable. After leaving the larger cities and settlements a bed is a rare object. All the houses are provided with extra hammock hooks. The traveller will be entertained hospitably and after dinner will be given two hooks upon which to hang his hammock, for he will be expected to have his hammock and, in insect time, his net, if he has nothing else. As a rule, a native hammock and net can be procured in the field. But it is best to take a comfortable one along, arranged with a fine-meshed net.

In regard to the folding cot: It is heavy and its numerous legs form a sort of highway system over which all sorts of insects can crawl up to the sleeper. The ants are special pests and some of them can bite with the enthusiastic vigor of beasts many times their size. The canvas floor in a tent obviates to a degree the insect annoyance.

The headwaters of the rivers are usually reached by pack-trains of mules and oxen. The primitive ox-cart also comes in where the trail is not too bad. One hundred and sixty to one hundred and eighty pounds is a good load for the pack-animals, and none of the cases should weigh more than fifty or sixty pounds. Each case should be marked with its contents and gross and net weight in kilos.

For personal baggage the light fibre sample case used by travelling men in the United States does admirably. The regulation fibre case with its metal binding sold for the purpose is too heavy and has the bad feature of swelling up under the influence of rain and dampness, often necessitating the use of an axe or heavy hammer to remove cover.

The ordinary fibre trunk is good for rail and steamer travel, but it is absolutely unpractical for mule-back or canoe. The fibre sample case could be developed into a container particularly fitted for exploration. The fibre should be soaked in hot paraffin and then hot- calendered or hot-pressed. This case could then be covered with waterproof canvas with throat opening like a duffel-bag.

The waterproof duffel-bags usually sold are too light in texture and wear through. A heavier grade should be used. The small duffel-bag is very convenient for hammock and clothing, but generally the thing wanted will be at the bottom of the bag! We took with us a number of small cotton bags. As cotton is very absorbent, I had them paraffined. Each bag was tagged and all were placed in the large duffel-bag. The light fibre case described above, made just the right size for mule pack, divided by partitions, and covered with a duffel-bag, would prove a great convenience.

The light steel boxes made in England for travellers in India and Africa would prove of value in South American exploration. They have the advantage of being insect and water proof and the disadvantage of being expensive.

It would be well if the traveller measured each case for personal equipment and computed the limit of weight that it could carry and still float. By careful distribution of light and heavy articles in the different containers, he could be sure of his belongings floating if accidentally thrown into the water.

It is not always possible to get comfortable native saddles. They are all constructed on heavy lines with thick padding which becomes water- soaked in the rainy season. A United States military saddle, with Whitman or McClellan tree, would be a positive luxury. Neither of them is padded, so would be the correct thing for all kinds of weather. The regulation army saddle-blanket is also advised as a protection for the mule's back. The muleteer should wash the saddle-blanket often. For a long mule-back trip through a game country, it would be well to have a carbine boot on the saddle (United States Army) and saddle-bags with canteen and cup. In a large pack-train much time and labor are lost every morning collecting the mules which strayed while grazing. It would pay in the long run to feed a little corn at a certain hour every morning in camp, always ringing a bell or blowing a horn at the time. The mules would get accustomed to receiving the feed and would come to camp for it at the signal.

All the rope that came to my attention in South America was three- strand hemp, a hard material, good for standing rigging but not good for tackle or for use aboard canoes. A four-ply bolt rope of best manilla, made in New Bedford, Mass., should be taken. It is the finest and most pliable line in the world, as any old whaler will tell you. Get a sailor of the old school to relay the coils before you go into the field so that the rope will be ready for use. Five eighths to seven eighths inch diameter is large enough. A few balls of marline come in conveniently as also does heavy linen fish-line.

A small-sized duffel-bag should be provided for each of the men as a container for hammock and net, spare clothing, and mess-kit. A very small waterproof pouch or bag should be furnished also for matches, tobacco, etc.

The men should be limited to one duffel-bag each. These bags should be numbered consecutively. In fact, every piece in the entire equipment should be thus numbered and a list kept in detail in a book.

The explorer should personally see that each of his men has a hammock, net, and poncho; for the native, if left unsupervised, will go into the field with only the clothing he has on.

FOOD—Though South America is rich in food and food possibilities, she has not solved the problem of living economically on her frontiers. The prices asked for food in the rubber districts we passed through were amazing. Five milreis (one dollar and fifty cents) was cheap for a chicken, and eggs at five hundred reis (fifteen cents) apiece were a rarity. Sugar was bought at the rate of one to two milreis a kilo—in a country where sugar-cane grows luxuriantly. The main dependence is the mandioc, or farina, as it is called. It is the bread of the country and is served at every meal. The native puts it on his meat and in his soup and mixes it with his rice and beans. When he has nothing else he eats the farina, as it is called, by the handful. It is seldom cooked. The small mandioc tubers when boiled are very good and are used instead of potatoes. Native beans are nutritious and form one of the chief foods.

In the field the native cook wastes much time. Generally provided with an inadequate cooking equipment, hours are spent cooking beans after the day's work, and then, of course, they are often only partially cooked. A kettle or aluminum Dutch oven should be taken along, large enough to cook enough beans for both breakfast and dinner. The beans should be cooked all night, a fire kept burning for the purpose. It would only be necessary then to warm the beans for breakfast and dinner, the two South American meals.

For meat the rubber hunter and explorer depends upon his rifle and fish-hook. The rivers are full of fish which can readily be caught, and, in Brazil, the tapir, capybara, paca, agouti, two or three varieties of deer, and two varieties of wild pig can occasionally be shot; and most of the monkeys are used for food. Turtles and turtle eggs can be had in season and a great variety of birds, some of them delicious in flavor and heavy in meat. In the hot, moist climate fresh meat will not keep and even salted meat has been known to spoil. For use on the Roosevelt expedition I arranged a ration for five men for one day packed in a tin box; the party which went down the Duvida made each ration do for six men for a day and a half, and in addition gave over half the bread or hardtack to the camaradas. By placing the day's allowance of bread in this same box, it was lightened sufficiently to float if dropped into water. There were seven variations in the arrangement of food in these boxes and they were numbered from 1 to 7, so that a different box could be used every day of the week. In addition to the food, each box contained a cake of soap, a piece of cheese-cloth, two boxes of matches, and a box of table salt. These tin boxes were lacquered to protect from rust and enclosed in wooden cases for transportation. A number in large type was printed on each. No. 1 was cased separately; Nos. 2 and 3, 4 and 5, 6 and 7 were cased together. For canoe travel the idea was to take these wooden cases off. I did not have an opportunity personally to experience the management of these food cases. We had sent them all ahead by pack- train for the explorers of the Duvida River. The exploration of the Papagaio was decided upon during the march over the plateau of Matto Grosso and was accomplished with dependence upon native food only.

DAILY RATION FOR FIVE MEN

SUN. MON. TUES. WED. THUR. FRI. SAT. Rice 16 16 16 Oatmeal 13 13 13 Bread 100 100 100 100 100 100 100 Tea-biscuits 18 18 18 Gingersnaps 21 21 21 21 Dehydrated potatoes 11 11 11 11 11 11 Dehydrated onions 5 5 5 5 5 5 Erbswurst 8 8 8 Evaporated soups 6 6 6 Baked beans 25 25 Condensed milk 17 17 17 17 17 17 17 Bacon 44 44 44 44 44 44 44 Roast beef 56 Braised beef 56 56 Corned beef 70 Ox tongue 78 Curry and chicken 72 Boned chicken 61 Fruits: evaporated berries 5 5 5 5 Figs 20 20 Dates 16 Sugar 32 32 32 32 32 32 32 Coffee 10 10 10 10 10 10 10 Tea 5 5 5 5 5 5 5 Salt 4 4 4 4 4 4 4 Sweet chocolate 16

EACH BOX ALSO CONTAINED

Muslin, one yard 1 1 1 1 1 1 1 Matches, boxes 2 2 2 2 2 2 2 Soap, one cake 1 1 1 1 1 1 1

Above weights of food are net in avoirdupois ounces. Each complete ration with its tin container weighed nearly twenty-seven pounds. The five pounds over net weight of daily ration was taken up in tin necessary for protection of food. The weight of component parts of daily ration had to be governed to some extent by the size of the commercial package in which the food could be purchased on short notice. Austin, Nichols & Co., of New York, who supplied the food stores for my polar expedition, worked day and night to complete the packing of the rations on time.

The food cases described above were used on Colonel Roosevelt's descent of the Rio da Duvida and also by the party who journeyed down the Gy-Parana and Madeira Rivers. Leo Miller, the naturalist, who was a member of the last-named party, arrived in Manaos, Brazil, while I was there and, in answer to my question, told me that the food served admirably and was good, but that the native cooks had a habit of opening a number of cases at a time to satisfy their personal desire for special delicacies. Bacon was the article most sought for. Speaking critically, for a strenuous piece of work like the exploration of the Duvida, the food was somewhat bulky. A ration arrangement such as I used on my sledge trips North would have contained more nutritious elements in a smaller space. We could have done without many of the luxuries. But the exploration of the Duvida had not been contemplated and had no place in the itinerary mapped out in New York. The change of plan and the decision to explore the Duvida River came about in Rio Janeiro, long after our rations had been made out and shipped.

"Matte" the tea of Brazil and Paraguay, used in most of the states of South America, should not be forgotten. It is a valuable beverage. With it a native can do a wonderful amount of work on little food. Upon the tired traveller it has a very refreshing effect.

Doctor Peckolt, celebrated chemist of Rio de Janeiro, has compared the analysis of matte with those of green tea, black tea, and coffee and obtained the following result:

IN 1,000 PARTS OF GREEN TEA BLACK TEA COFFEE MATTE Natural oil 7.90 0.06 0.41 0.01 Chlorophyl 22.20 18.14 13.66 62.00 Resin 22.20 34.40 13.66 20.69 Tannin 178.09 128.80 16.39 12.28 Alkaloids: Mateina 4.50 4.30 2.66 2.50 Extractive substances 464.00 390.00 270.67 238.83 Cellulose and fibres 175.80 283.20 178.83 180.00 Ashes 85.60 25.61 25.61 38.11

Manner of preparation: The matte tea is prepared in the same manner as the Indian tea, that is to say, by pouring upon it boiling water during ten to fifteen minutes before using. To obtain a good infusion five spoonfuls of matte are sufficient for a litre of water.

Some experiments have been made lately with the use of matte in the German army, and probably it would be a valuable beverage for the use of our own troops. Two plates and a cup, knife, fork, and spoon should be provided for each member of the party. The United States Army mess- kit would serve admirably. Each man's mess-kit should be numbered to correspond with the number on his duffel-bag.

An aluminum (for lightness) cooking outfit, or the Dutch oven mentioned, with three or four kettles nested within, a coffee pot or a teapot would suffice. The necessary large spoons and forks for the cook, a small meat grinder, and a half dozen skinning knives could all be included in the fibre case. These outfits are usually sold with the cups, plates, etc., for the table. As before suggested, each member of the party should have his own mess-kit. It should not be carried with the general cooking outfit. By separating the eating equipments thus, one of the problems of hygiene and cleanliness is simplified.

RIFLES—AMMUNITION—A heavy rifle is not advised. The only animals that can be classed as dangerous are the jaguar and white-jawed peccary, and a 30-30 or 44 calibre is heavy enough for such game. The 44-calibre Winchester or Remington carbine is the arm generally used throughout South America, and 44 calibre is the only ammunition that one can depend upon securing in the field. Every man has his own preference for an arm. However, there is no need of carrying a nine or ten pound weapon when a rifle weighing only from six and three fourths to seven and one half pounds will do all that is necessary. I, personally, prefer the small-calibre rifle, as it can be used for birds also. The three-barrelled gun, combining a double shotgun and a rifle, is an excellent weapon, and it is particularly valuable for the collector of natural-history specimens. A new gun has just come on the market which may prove valuable in South America where there is such a variety of game, a four-barrel gun, weighing only eight and one fourth pounds. It has two shotgun barrels, one 30 to 44 calibre rifle and the rib separating the shotgun barrels is bored for a 22-calibre rifle cartridge. The latter is particularly adapted for the large food birds, which a heavy rifle bullet might tear. Twenty-two calibre ammunition is also very light and the long 22 calibre exceedingly powerful. Unless in practice it proves too complicated, it would seem to be a good arm for all-round use—sixteen to twenty gauge is large enough for the shotgun barrels. Too much emphasis cannot be placed upon the need of being provided with good weapons. After the loss of all our arms in the rapids we secured four poor, rusty rifles which proved of no value. We lost three deer, a tapir, and other game, and finally gave up the use of the rifles, depending upon hook and line. A 25 or 30 calibre high power automatic pistol with six or seven inch barrel would prove a valuable arm to carry always on the person. It could be used for large game and yet would not be too large for food birds. It is to be regretted that there is nothing in the market of this character.

We had our rifle ammunition packed by the U. M. C. Co. in zinc cases of one hundred rounds each, a metallic strip with pull ring closing the two halves of the box. Shot-cartridge, sixteen gauge, were packed the same way, twenty-five to the box.

The explorer would do well always to have on his person a compass, a light waterproof bag containing matches, a waterproof box of salt, and a strong, light, linen or silk fish-line with several hooks, a knife, and an automatic at his belt, with several loaded magazines for the latter in his pocket. Thus provided, if accidentally lost for several days in the forest (which often happens to the rubber hunters in Brazil), he will be provided with the possibility of getting game and making himself shelter and fire at night.

FISH—For small fish like the pacu and piranha an ordinary bass hook will do. For the latter, because of its sharp teeth, a hook with a long shank and phosphor-bronze leader is the best; the same character of leader is best on the hook to be used for the big fish. A tarpon hook will hold most of the great fish of the rivers. A light rod and reel would be a convenience in catching the pacu. We used to fish for the latter variety in the quiet pools while allowing the canoe to drift, and always saved some of the fish as bait for the big fellows. We fished for the pacu as the native does, kneading a ball of mandioc farina with water and placing it on the hook as bait. I should not be surprised, though, if it were possible, with carefully chosen flies, to catch some of the fish that every once in a while we saw rise to the surface and drag some luckless insect under.

CLOTHING—Even the experienced traveller when going into a new field will commit the crime of carrying too much luggage. Articles which he thought to be camp necessities become camp nuisances which worry his men and kill his mules. The lighter one can travel the better. In the matter of clothing, before the actual wilderness is reached the costume one would wear to business in New York in summer is practical for most of South America, except, of course, the high mountain regions, where a warm wrap is necessary. A white or natural linen suit is a very comfortable garment. A light blue unlined serge is desirable as a change and for wear in rainy weather.

Strange to relate, the South American seems to have a fondness for stiff collars. Even in Corumba, the hottest place I have ever been in, the native does not think he is dressed unless he wears one of these stiff abominations around his throat. A light negligee shirt with interchangeable or attached soft collars is vastly preferable. In the frontier regions and along the rivers the pajama seems to be the conventional garment for day as well as night wear. Several such suits of light material should be carried—the more ornamented and beautifully colored the greater favor will they find along the way. A light cravenetted mackintosh is necessary for occasional cool evenings and as a protection against the rain. It should have no cemented rubber seams to open up in the warm, moist climate. Yachting oxfords and a light pair of leather slippers complete the outfit for steamer travel. For the field, two or three light woollen khaki-colored shirts, made with two breast pockets with buttoned flaps, two pairs of long khaki trousers, two pairs of riding breeches, a khaki coat cut military fashion with four pockets with buttoned flaps, two suits of pajamas, handkerchiefs, socks, etc., would be necessary. The poncho should extend to below the knees and should be provided with a hood large enough to cover the helmet. It should have no cemented seams; the material recently adopted by the United States Army for ponchos seems to be the best. For footgear the traveller needs two pairs of stout, high hunting shoes, built on the moccasin form with soles. Hob nails should be taken along to insert if the going is over rocky places. It is also advisable to provide a pair of very light leather slipper boots to reach to just under the knee for wear in camp. They protect the legs and ankles from insect stings and bites. The traveller who enters tropical South America should protect his head with a wide-brimmed soft felt hat with ventilated headband, or the best and lightest pith helmet that can be secured, one large enough to shade the face and back of neck. There should be a ventilating space all around the head-band; the wider the space the better. These helmets can be secured in Rio and Buenos Aires. Head-nets with face plates of horsehair are the best protection against small insect pests. They are generally made too small and the purchaser should be careful to get one large enough to go over his helmet and come down to the breast. Several pairs of loose gloves rather long in the wrist will be needed as protection against the flies, piums and boroshudas which draw blood with every bite and are numerous in many parts of South America. A waterproof sun umbrella, with a jointed handle about six feet long terminating in a point, would be a decided help to the scientist at work in the field. A fine-meshed net fitting around the edge of the umbrella would make it insect proof. When folded it would not be bulky and its weight would be negligible. Such an umbrella could also be attached, with a special clamp, to the thwart of a canoe and so prove a protection from both sun and rain.

There are little personal conveniences which sometimes grow into necessities. One of these in my own case was a little electric flash- light taken for the purpose of reading the verniers of a theodolite or sextant in star observations. It was used every night and for many purposes. As a matter of necessity, where insects are numerous one turns to the protection of his hammock and net immediately after the evening meal. It was at such times that I found the electric lamp so helpful. Reclining in the hammock, I held the stock of the light under my left arm and with diary in my lap wrote up my records for the day. I sometimes read by its soft, steady light. One charge of battery, to my surprise, lasted nearly a month. When forced to pick out a camping spot after dark, an experience which comes to every traveller in the tropics in the rainy season, we found its light very helpful. Neither rain nor wind could put it out and the light could be directed wherever needed. The charges should be calculated on the plan of one for every three weeks. The acetylene lamp for camp illumination is an advance over the kerosene lantern. It has been found that for equal weight the carbide will give more light than kerosene or candle. The carbide should be put in small containers, for each time a box is opened some of the contents turns into gas from contact with the moist air.

TOOLS—Three or four good axes, several bill-hooks, a good hatchet with hammer head and nail-puller should be in the tool kit. In addition, each man should be provided with a belt knife and a machete with sheath. Collins makes the best machetes. His axes, too, are excellent. The bill-hook, called foice in Brazil, is a most valuable tool for clearing away small trees, vines, and under-growths. It is marvellous how quickly an experienced hand can clear the ground in a forest with one of these instruments. All of these tools should have handles of second-growth American hickory of first quality; and several extra handles should be taken along. The list of tools should be completed with a small outfit of pliers, tweezers, files, etc.—the character, of course, depending upon the mechanical ability of the traveller and the scientific instruments he has with him that might need repairs.

SURVEY INSTRUMENTS—The choice of instruments will depend largely upon the character of the work intended. If a compass survey will suffice, there is nothing better than the cavalry sketching board used in the United States Army for reconnaissance. With a careful hand it approaches the high degree of perfection attained by the plane-table method. It is particularly adapted for river survey and, after one gets accustomed to its use, it is very simple. If the prismatic compass is preferred, nothing smaller than two and one half inches in diameter should be used. In the smaller sizes the magnet is not powerful enough to move the dial quickly or accurately.

Several good pocket compasses must be provided. They should all have good-sized needles with the north end well marked and degrees engraved in metal. If the floating dial is preferred it should be of aluminum and nothing smaller than two and one half inches, for the same reason as mentioned above regarding the prismatic compass.

Expense should not be spared if it is necessary to secure good compasses. Avoid paper dials and leather cases which absorb moisture. The compass case should allow taking apart for cleaning and drying.

The regular chronometer movement, because of its delicacy, is out of the question for rough land or water travel. We had with us a small- sized half-chronometer movement recently brought out by the Waltham Company as a yacht chronometer. It gave a surprisingly even rate under the most adverse conditions. I was sorry to lose it in the rapids of the Papagaio when our canoes went down.

The watches should be waterproof with strong cases, and several should be taken. It would be well to have a dozen cheap but good watches and the same number of compasses for use around camp and for gifts or trade along the line of travel. Money is of no value after one leaves the settlements. I was surprised to find that many of the rubber hunters were not provided with compasses, and I listened to an American who told of having been lost in the depths of the great forest where for days he lived on monkey meat secured with his rifle until he found his way to the river. He had no compass and could not get one. I was sorry I had none to give; I had lost mine in the rapids.

For the determination of latitude and longitude there is nothing better than a small four or five inch theodolite not over fifteen pounds in weight. It should have a good prism eyepiece with an angle tube attached so it would not be necessary to break one's neck in reading high altitudes. For days we travelled in the direction the sun was going, with altitudes varying from 88 deg. to 90 deg.. Because of these high altitudes of the sun the sextant with artificial horizon could not be used unless one depended upon star observations altogether, an uncertain dependence because of the many cloudy nights.

BAROMETERS—The Goldsmith form of direct-reading aneroid is the most accurate portable instrument and, of course, should be compared with a standard mercurial at the last weather-bureau station.

THERMOMETERS—A swing thermometer, with wet and dry bulbs for determination of the amount of moisture in the air, and the maximum and minimum thermometer of the signal-service or weather-bureau type should be provided, with a case to protect them from injury.

A tape measure with metric scale of measurements on one side and feet and inches on the other is most important. Two small, light waterproof cases could be constructed and packed with scientific instruments, data, and spare clothing and yet not exceed the weight limit of flotation. In transit by pack-train these two cases would form but one mule load.

PHOTOGRAPHIC—From the experience gained in several fields of exploration it seems to me that the voyager should limit himself to one small-sized camera, which he can always have with him, and then carry a duplicate of it, soldered in tin, in the baggage. The duplicate need not be equipped with as expensive a lens and shutter as the camera carried for work; 31/4 x 41/4 is a good size. Nothing larger than 3 1/4 x 5 1/2 is advised. We carried the 3A special Kodak and found it a light, strong, and effective instrument. It seems to me that the ideal form of instrument would be one with a front board large enough to contain an adapter fitted for three lenses. For the 3 1/4 x 4 1/4:

One lens 4 or 4 1/2 focus One lens 6 or 7 focus One lens telephoto or telecentric 9 to 12 focus

The camera should be made of metal and fitted with focal-plane shutter and direct view-finder.

A sole leather case with shoulder-strap should contain the camera and lenses, with an extra roll of films, all within instant reach, so that a lens could be changed without any loss of time.

Plates, of course, are the best, but their weight and frailty, with difficulty of handling, rule them out of the question. The roll film is the best, as the film pack sticks together and the stubs pull off in the moist, hot climate. The films should be purchased in rolls of six exposures, each roll in a tin, the cover sealed with surgical tape. Twelve of these tubes should be soldered in a tin box. In places where the air is charged with moisture a roll of films should not be left in a camera over twenty-four hours.

Tank development is best for the field. The tanks provided for developing by the Kodak Company are best for fixing also. A nest of tanks would be a convenience; one tank should be kept separate for the fixing-bath. As suggested in the Kodak circular, for tropical development a large-size tank can be used for holding the freezing mixture of hypo. This same tank would become the fixing tank after development. In the rainy season it is a difficult matter to dry films. Development in the field, with washing water at 80 degrees F., is a patience-trying operation. It has occurred to me that a small air-pump with a supply of chloride of calcium in small tubes might solve the problem of preserving films in the tropics. The air-pump and supply of chloride of calcium would not be as heavy or bulky as the tanks and powders needed for development. By means of the air-pump the films could be sealed in tin tubes free from moisture and kept thus until arrival at home or at a city where the air was fairly dry and cold water for washing could be had.

While I cordially agree with most of the views expressed by Mr. Fiala, there are some as to which I disagree; for instance, we came very strongly to the conclusion, in descending the Duvida, where bulk was of great consequence, that the films should be in rolls of ten or twelve exposures. I doubt whether the four-barrel gun would be practical; but this is a matter of personal taste.



APPENDIX C.

My Letter of May 1 to General Lauro Muller

The first report on the expedition, made by me immediately after my arrival at Manaos, and published in Rio Janeiro upon its receipt, is as follows:

MAY 1st, 1914.

TO HIS EXCELLENCY THE MINISTER OF FOREIGN AFFAIRS, RIO-DE-JANEIRO. MY DEAR GENERAL LAURO MULLER:

I wish first to express my profound acknowledgments to you personally and to the other members of the Brazilian Government whose generous courtesy alone rendered possible the Expedicao Scientifica Roosevelt- Rondon. I wish also to express my high admiration and regard for Colonel Rondon and his associates who have been my colleagues in this work of exploration. In the third place I wish to point out that what we have just done was rendered possible only by the hard and perilous labor of the Brazilian Telegraphic Commission in the unexplored western wilderness of Matto Grosso during the last seven years. We have had a hard and somewhat dangerous but very successful trip. No less than six weeks were spent in slowly and with peril and exhausting labor forcing our way down through what seemed a literally endless succession of rapids and cataracts. For forty-eight days we saw no human being. In passing these rapids we lost five of the seven canoes with which we started and had to build others. One of our best men lost his life in the rapids. Under the strain one of the men went completely bad, shirked all his work, stole his comrades' food and when punished by the sergeant he with cold-blooded deliberation murdered the sergeant and fled into the wilderness. Colonel Rondon's dog running ahead of him while hunting, was shot by two Indians; by his death he in all probability saved the life of his master. We have put on the map a river about 1500 kilometres in length running from just south of the 13th degree to north of the 5th degree and the biggest affluent of the Madeira. Until now its upper course has been utterly unknown to every one, and its lower course although known for years to the rubbermen utterly unknown to all cartographers. Its source is between the 12th and 13th parallels of latitude south, and between longitude 59 degrees and longitude 60 degrees west from Greenwich. We embarked on it about at latitude 12 degrees 1 minute south and longitude 60 degrees 18 west. After that its entire course was between the 60th and 61st degrees of longitude approaching the latter most closely about in latitude 8 degrees 15 minutes. The first rapids were at Navaite in 11 degrees 44 minutes and after that they were continuous and very difficult and dangerous until the rapids named after the murdered sergeant Paishon in 11 degrees 12 minutes. At 11 degrees 23 minutes the river received the Rio Kermit from the left. At 11 degrees 22 minutes the Marciano Avila entered it from the right. At 11 degrees 18 minutes the Taunay entered from the left. At 10 degrees 58 minutes the Cardozo entered from the right. At 10 degrees 24 minutes we encountered the first rubberman. The Rio Branco entered from the left at 9 degrees 38 minutes. We camped at 8 degrees 49 minutes or approximately the boundary line between Matto Grosso and Amazonas. The confluence with the upper Aripuanan, which entered from the right, was in 7 degrees 34 minutes. The mouth where it entered the Madeira was in about 5 degrees 30 minutes. The stream we have followed down is that which rises farthest away from the mouth and its general course is almost due north.

My dear Sir, I thank you from my heart for the chance to take part in this great work of exploration.

With high regard and respect, believe me

Very sincerely yours, THEODORE ROOSEVELT.

THE END