Elliott Roberts

Lieutenant, United States Coast and Geodetic Survey

(Originally published in The Military Engineer,

Vol. XXIV, No. 135, May-June 1932, pp. 245-249.)

It was said, by a certain writer that a particular strip of swampy jungle land in the tropics was utterly impenetrable. "That is," he qualified, "except to the hardiest of explorers, or a civil engineer!" Without bothering to examine into the sincerity of this remark, it must be admitted that civil engineers have a great tradition to uphold - the tradition of success in any endeavor, however hard.

The writer had occasion to recall the foregoing quotation when ordered by the United States Coast and Geodetic Survey to lay out and execute second-order triangulation through the interior of Luzon. What this meant, in effect, was to measure off the miles, meters, even the millimeters, separating a series of stations which should be placed in orderly arrangement on the summits of jungle-clad peaks nearly 10,000 feet high, in the remote interiors of vast tropical forests, and through the swampy bottom lands of the great Cagayan Valley. Covering a strip of territory over 200 miles in length, in places 40 miles from side to side, these stations were to compose a triangulation net reaching from the Central Valley, at the foot of which lies Manila, through the little-known lands of northern Luzon to Aparri, on the north, the seaport of the Cagayan Valley.

The nautical charts of the Philippines, like those of all other well-surveyed countries, are based on a single well-chosen datum. The three elements of this datum (latitude, longitude, and azimuth) are made available to all sections of the archipelago by a country-wide geodetic survey. Throughout large parts of the island group, with its seven-thousand-odd separate land areas, adjacent islands are within sight of one another, so that the triangulation nets used to carry the geodetic datum may span the water areas, and instrument stations can thus be placed along the shores, within easy reach of the parties from the survey ships. Only in a few places are the over-water lengths of lines, or the requirements as to their directions, so exacting as to require the scaling of the high peaks far inland on the islands.

Northern Luzon is a large land mass, having almost no off-lying islands. Triangulation around its coast, therefore, must be almost wholly upon the mainland. The mountains are high and precipitous, almost all the way from Manila Bay, around by way of the west, north, and east coasts, to Casiguran Sound. Coastal triangulation surveys were made in the early days of the American occupancy, from Manila to Aparri, but no further progress in such fashion could be made. The country was so rugged, and in the early days the natives so hostile, that no great distance inland could be reached, and the triangulation nets were long in proportion to their width. They were very weak geometrically in this and other ways.

While such work, as far as it went, was adequate for the control of local charting operations, it failed utterly as a means of carrying the elements of the datum, accurately, to the great areas of northern Luzon and the island groups that reach still farther north toward the Japanese territory of Formosa.

When the surveys of these regions came due for completion, it became necessary to plan means of extending accurate triangulation northward from the Central Valley - work which should reach the northernmost of the Philippine group, controlling, along the way, all the local third-and fourth-order projects upon which the hydrographic parties base their charting work. Only in this way could the fundamental plan of a single and strong datum, basic for all the charts, be followed. The work was essential to insure that isolated and separate surveys, when brought together, would join without error, and that all the charts, laid out in far-flung array to cover all the country, would be consistent and harmonious.

As it is impossible to lay out satisfactory figures along the coast of this rugged island, the interior must naturally be utilized. Despite whatever obstacles might be encountered in a little-known land, a well-proportioned network of triangulation figures must be planned and worked over, spanning the whole distance. A backbone, a continental main scheme, as it were, would first be erected, then other links connecting with local coastal areas, especially on the east coast, could follow. The performance of this first through link between the Manila area and Luzon's north coast was the task assigned the writer.

The Cordillera Central, domain of the hill tribes, so-called head-hunters, occupies the western third of the island. It is rugged, and approaches 10,000 feet in elevation. Along the east coast lies the Sierra Madre, a range of precipices and virgin jungle, unexplored and feared. Between lie the broad reaches of the Cagayan Valley 3,000 square miles of rolling hills and rich bottom lands. Sparsely settled it yields, nevertheless, a great part of the crop of fine Philippine tobaccos.

Down this valley lay our route, but first must come the crossing of the southeast spur of the Cordillera, and the traversing of the narrow and tortuous valley of the Magat River. These high mountains are trailless and forbidding, but there is one road, of sorts, through Balete, the lowest pass. The road, culmination of years of effort, straggles northward over mountains and through swamps, the scene of our operations, affording access by truck to the Cagayan country and Aparri. All plans for transport were necessarily made dependent upon our natural determination to make all possible use that we could of this road.

The Bureau of Lands, the cadastral survey unit of the local Government, wanted to make extensive surveys of the Cagayan Valley in order to open more land to homesteaders. They wanted control points for the surveys, as well as instruction in the art of extending geodetic survey work on their own account. These desires were made the basis of an arrangement whereby the field expenses were undertaken by the Philippine Government. Six Filipino surveyors were assigned as students and potential observers. The final outcome of this arrangement was highly satisfactory to all concerned; the students nobly did their part and bore their full share of discouragement, and, later, of triumphant accomplishment.

Little was known of the country, and nothing of the working conditions to be met. The first part of the job consisted in finding out. With a tentative first organization of our resources, the work of studying the situation and beginning the reconnaissance for the layout of figures was undertaken.

The crossing of the Cordillera was the greatest difficulty. Progress to the mountain peaks was necessarily over trails of our own cutting, in a land where trail cutting is a fine art. Often days were required to reach eminences from which the surrounding terrain could be examined. The time and effort required for even the most casual excursion made necessary a nice judgment in deciding how to see most with the least travel. At last a layout, geometrically excellent, was adopted, depending only on a number of choice "if's": "If that ridge, by good fortune, sags low enough to pass a line of sight from peak to peak; if that peak is the same one we can see from over the Isabela country; if we can get there at all with our instruments." It is a tribute to the luck that persistently followed us that all these "if's" came out "yes." Various questions relating to the procurement of cargadores (carriers) and the selection of routes were left for solution when the peaks were to be occupied for the actual measurements.

The Cagayan Valley presented an easier aspect in many ways. Here the country is partly open, alternating between the low flood plains and ranges of grassy hills, intersected in all directions by deep ravines filled with almost impenetrable vegetation. These, the foothills of the high mountains farther back on either side, rise, one range above another, gradually giving way to the forested slopes of the mighty eminences that rim the valley. The problem was to make a proper balance against the desire, the one hand to utilize the lower more accessible open summits, and, as opposed to this easier country the necessity of placing the stations far enough back from the river to provide width and geometrical strength to the layout.

Many of the stations on this part of the work were selected without visit, their locations, height and intervisibility being determined through the use of an alt-azimuth instrument. Only in a few cases was there grave doubt, and such cases resolved themselves after visits to the proposed stations.

Travel along the one highway which penetrated the northern part of Luzon was accomplished by motor-car, this route being a sort of main stem from which all travel radiated. Expeditions into the hills were mainly afoot, although in a few places horses could penetrate short distances. Natives always accompanied the party as packers and guides. In some places along the river it was found expedient to travel by launch; this form of transportation was obtainable in Aparri.

During the course of the work two airplane flights were made over the Sierra Madre, through the courtesy of the Army Air Corps station in Manila. Aside from making a study of this unknown region searching for routes of access, and getting the lay of the land - it was determined that many of the high peaks of the range are visible alike from the river valley and from the surface of the Pacific Ocean to the eastward. This assured us that such peaks which would be located by the use of intersecting cuts from the main stations in the valley, would be directly available as control stations for the impending survey of the Pacific area east of Luzon. One great step was thus made in solving the difficult problem of how to get geodetic control on the east and northeast coast of the island. A good tentative layout for triangulation extensions to that forlorn stretch of coast was also made available by these flights.

The civil engineer, accustomed as he must be to all eventualities on a job such as this, found a real thrill in the unwonted stopping of his airplane motor, over wild country an hour's flight from the nearest possible landing field. The detection of a frozen fuel pump, and the substitution of an auxiliary hand affair, were the work of but a few moments for the able pilot, yet there was a quick vision of a parachute landing in the summits of 200-foot trees in trailless and precipitous jungle land far from food and shelter, in country containing great pythons and wild headhunters.

Traveling light and as fast as possible, the whole region was thus studied, and a tentative plan evolved for the work. The advent of a rainy season closed up all fieldwork for nearly six months. The country, during such a time, is torn by cloudburst and gale, being in the mean track of the feared East Indies hurricanes, or typhoons. As an instance of the violence of such storms, the Cagayan River, midway of its course sometimes rises 30 and even 40 feet within the space of a night's time, flooding great areas of its valley. Such a storm destroys practically all crops. The native planters contemplate the loss of at least one of three years' efforts.

The time was spent perfecting our organization and outfitting a field party to undertake observing work. Six units took the field the following dry season each in charge of a Filipino assistant, and occupied the assigned stations. Camping there, each of these parties operated heliotrope and electric signaling apparatus for the benefit of the observing unit. Work started, when all was in readiness, with the writer as observer.

A frail-appearing but resolute Model-T truck began wending its way up and down the muddy lane that threaded the Cagayan Valley, carrying its regular overload of men and equipment. In many a case the services of the plodding water-buffalo or carabao, were required to move it from seemingly bottomless quagmires, but somehow it always got there. It clung precariously to the ribbon of trail that skirted cliff-like mountain sides steaming mightily with the strain of it, and it forded rushing streams hopping gaily from boulder to boulder. It disturbed the slumber of torpid iguanas along the roadside, and sent many a band of rollicking monkeys into wild flight.

When the truck reached a town or outlying barrio that lay nearest the station to be visited, there was always a band of gaping natives to watch the unloading of instruments and camping gear, and to offer the services of two-wheeled bull carts or, more often, carabao sledges, under whose jolting mercies the trip into the wilds would begin. The sledges could travel over unbelievably rough country. Nothing but dense forest really stopped them. At times it seemed to be preferable to pack the equipment on the backs of the carabaos or the ridiculously small native horses. The theodolites and other delicate instruments were always carried on the backs of trusted Cargadores, and when animals could go no further, everything was so handled.

Some twenty or more dialects were available in the mouth of the linguistic surveyor-assistants. Whatever babel prevailed in a remote barrio up under the brow of the mountain some one of them could always address the head man, expressing our need for labor, which was always forthcoming. The men assigned by the village ruler dropped all personal matters for the prospect of earning 30 to 40 cents a day apiece, and came, dressed in cotton undershirts, drawers, and wide-brimmed woven hats, ready for their trip into the forest. They welcomed this excuse to make a holiday, and anticipated a return laden with rattan and other prized forest products. There would always be certain strong ones, proud of the great loads they could carry, who would jeer at the others, thus spurring on everybody.

These men subsisted principally upon rice, using a little dried or fermented fish, or native fruits, as seasoning. When working they ate without stint, and, for the longer trips, the packs consisted more of rice than anything else. Indeed, it was frequently the case that a man could carry little more than the rice he would consume during his absence from home, making it necessary to hire as many as thirty or forty carriers, with elaborate plans for regular trains of supply to follow. At mealtime each man filled, with rice and water, a section of bamboo, 4 or 5 inches in diameter, and 18 inches long. The contents of this, boiled over a small fire, set him eager for two or three hours of packing his load through hip-deep swamp mud, or up sixty-degree mountain slopes. At night, a few strokes of the ever-present bolo would provide the sticks and palm leaves which suddenly and mysteriously would form themselves into quite adequate sleeping shacks.

The men engaged far in the back country were usually expected to act as guides, although their limited intelligence often made it difficult for them to understand where we wanted to go. They were most peaceable, Christianized Filipinos - illiterate farmers who worked more or less diligently on their lands, and who made excellent laborers when needed. In the surrounding mountain lands lived a variety of pagan races. The negritos of the Sierra Madre ... roam the forest in search of wild pig and deer, which they shoot unerringly with bow and poisoned arrow. They wear nothing but the gee string, or breech clout. Timid and diffident, they were of little use to the party, except through their expert knowledge of woodcraft. As guides or packers they were useless ....

The extensive mountain lands west and south of the valley harbor the various tribes of the so-called head hunters, robust and picturesque savages, who have hardly risen above their ancient traditions of warfare. They still believe the gods of the harvest will be propitiated by offerings consisting of the severed heads of enemy tribesmen. They are of various races and characteristics, but are uniformly fond of bright color in their dress and personal ornaments, and are all of proud and disdainful mien. They wear the gee string, eat dog meat whenever possible, and hunt with spear and head-axe. Centuries of rice culture on the steep mountain slopes of their home land have seen the construction, by these people, of thousands of miles of stone retaining walls, holding in place almost endless levels of terraced rice paddies. This system of terraces must be ranked with the foremost engineering feats of the world. These pagans, forced by the local government, have done extensive work in the building of graded trails through certain of the more accessible areas of the mountains, yet they persist, almost unanimously, in making their way direct toward their objective, over thousands of feet of rise and fall, rather than utilize the excellent trails laid out through their country.

Under show of authority they would submit to the indignity of acting as cargadores. At this work the women, the habitual doers of all manual labor excelled. They were far more useful than the men who were accustomed to traveling light in their business of warfare. They could often be made to transport 60-pound loads up 7,000-foot mountains, all within a day, for which they received 25 cents.

Each of the assistants was trained as his station was occupied, in the use of the instruments, with the object of making him an observer. They were rapid learners, meticulous followers of instructions, and developed into expert instrument men, but they were slow to acquire ingenuity and resourcefulness in meeting with daily obstacles. The chief of party as a consequence, found it necessary at all times to exercise close supervision over their work, and he traveled extensively to this end.

Because of the need of spending his time in this manner, and also because of the limited opportunities afforded by adverse weather conditions, it became desirable to turn over the actual observing to the assistants as soon as possible, thus permitting simultaneous observations at several stations. By far the larger part of the entire volume of the observing work was performed by them.

Communication between the various units had to be close. Only in this way could things be coordinated, supplies and laborers be sent where needed, and instructions be transmitted. The results of all observations had to be in the hands of the chief of party before orders could be given for the transfer of parties from station to station. The various parties were in communication, during favorable weather, by heliotroping. The instrument is difficult to manipulate, as the reflecting mirror must continually be adjusted for the changing direction of the sun while shading and unshading the ray of light to form the elements of the telegraph code. Under insistence, the observers became so expert at this signaling that they found delight in exchanging bits of Spanish poetry in leisure moments. The writer traveling from station to station, or along the way on eminences from which some station could be "called," kept in close touch with all operations.

The weather, even during the so-called "dry" season, was extremely uncertain. Frequent protracted spells of northeast monsoon brought, in this region, fine mist which reduced visibility to almost nothing. Perhaps 15 per cent of the time consisted of sunny weather. Such times are those chosen, of course, by the farmers for the burning over of coarse stands of grass on the hills. This is done to promote the growth of tender new shoots for the young carabaos. To the heavy natural haze was thus added the smoke from countless grass fires. The observers waited, during rain, for the fair days to follow, and when they arrived they must perforce wait again for the occasional windy day to clear the atmosphere. The writer waited nine days on one occasion for observing weather, confined to a little clearing in the jungle, large enough to contain two tents and a rustic tower. This had been erected to elevate the instruments and signals above the level of the surrounding trees. The forest was so dense that the most expert native trail cutters could back out in an hour no more than a kilometer of trail barely passable to an agile man.

Clearer atmosphere and a larger proportion of fair weather were found in the high mountains, but a new hazard, the worst encountered, was met. The hot air of the tropics rolling up the mountainsides into coolness of higher elevations, condensed into heavy cloud mantles which seldom left the summits. Days of waiting would finally find a few clear moments on a mountain peak, only to be offset by the fact that other stations were obscured. At least three stations must be clear to permit of an angle measurement, and the most dogged kind of patience was worn nearly threadbare before such an opportunity presented itself. Indeed, what must be nearly a record of such frustration was experienced as four observing parties waited, ever on the alert, at four stations marking a quadrilateral of the network, during a period of almost two months, in which time the only real work accomplished consisted of just one completed angle measurement at a single station.

It is hard to give adequate credit to the patience of the Filipino assistants who waited during months, their campsites boggy morasses of mud, while trees, tents, and equipment dripped forever with the condensation of the clouds which surrounded them with bleak curtains of obscurity. Patience only, of the most determined kind, was the answer to this bugbear of obstinate elements.

Observations of horizontal angles were made with Fennel direction theodolites of about 7 ½-inch diameter circles, using eight positions of the circle for direct and reversed telescope pointings. All observed values of a direction were averaged, all values within 4 seconds of the average being retained. Vertical angles were measured by two sets of double-zenith-distance observations, from which could be computed the elevations of all stations within a meter or so. In addition to the necessity of knowing these elevations in reducing the triangulation position computations to sea level, a valuable contribution was thus made to any future topographic surveys of this land. The horizontal angle measurements, when applied to the triangles of the network, gave triangle closure errors averaging 2 seconds of arc, thus qualifying the work under the standards set for second-order triangulation. In addition to the observations of main-scheme angles, a host of other objects were sighted upon, notably mountain peaks, the buildings visible in the river towns, and supplementary stations throughout the arable lands of the valley, where cadastral surveys were contemplated.

Signals were hard to see. At night, the powerful electric searchlights could seldom be seen more than a few miles. Banners or flags of any description could be seen only over short distances. The main reliance was on the heliotropes, which would, under exceptional circumstances, show as faint pin-pricks of light at distances of 30 or 40 miles, often hovering apparently unsupported in the haze that completely obscured the distant station hills. In the highest of the mountains, on the few occasions when clouds were absent, the atmosphere was far clearer, and large black banners were sometimes observed upon at distances as great as 50 miles.

The matter of food was a difficult one for the chief of party, although there was no lack of rice to satisfy the palates of the Filipino personnel. When the inevitable supply of canned goods became unbearable, resource was necessarily had to the supremely hard and tough chicken common to the locality, and to certain sorts of native vegetables with strange names and even stranger flavors. The only cooks obtainable for such expeditions were Llocano boys, whose half-hearted efforts were harder to bear than the work of preparing one's food. Water, except when the rain was caught direct, had necessarily to be boiled because of the danger of pollution by the tribesmen in the mountains. The boiling undoubtedly killed germs, but it didn't remove the slime. Campsites were generally in dense jungle, and living in them was a continual fight against the attacks of gnats, leeches, and great mosquitos. The opportunity to stop a time in the thatched houses of even the poorest of Filipino farmers was, by comparison, a treat.

The days on the trail were spent in plowing, hip deep, through swamp mud, always with a weather eye out for the lurking semi-savage water buffalo, or in climbing thousands of feet over the most precipitous slopes ever seen by the writer, short of rocky cliffs, through vegetation whose every fiber seemed designed to ensnare the unwary traveler. This life induced a vigorous state of health that undoubtedly did much to preserve one from the hovering agencies of malaria and other scourges.

Reference was made to a factor of luck in the working out of this project. It did not fail, even at the very end, when there came the day of reckoning - the day when the final closures were made, and the ultimate criterion of excellence was applied to the work. The base lines, measured out on the beaches at the two ends of the work, agreed with each other in length, as computed through the triangulation, with an error of only about one part in a million. Almost no revision or reoccupation of the work had been made necessary by abnormal triangle closures, nor had any of the assumptions regarding intervisibilities or identities of peaks, made during the days of the reconnaissance, been found wanting. A veritable backbone for survey purposes in northern Luzon had been erected, with, as offshoots, numbers of fixed and located peaks in the mountain ranges at the sides of the valley. Despite the fact that no banners or flags had been used - that pointings upon these distant peaks had been made upon whatever aspects had been given by their tree-clad summits -- the computations showed their probable error of position to be only three or four meters.

The valiant little Ford truck chugged wearily into Manila, one day just two years after the start of the first work, its last gesture in a great adventure.