Richthofen, Ferdinand Freiherr von. "Ueber den Gebirgsbau an der Nordküste von Formosa" [On the geological composition of Formosa's north coast]. Zeitschrift der Deutschen geologischen Gesellschaft 2 (1860): 532-545. Translated by Tina Schneider; edited by Douglas Fix.

[P. 532] Formosa consists of one high mountain range that seems to rise up to 12,000 feet above sea level (1). On the east coast the mountains fall precipitously into the sea; in some places, such as near the village of Chockeday, they drop sharply from a height of 6,000 to 7,000 feet. The descent continues so abruptly that all soundings along the coast show no bottom, with the exception of the few places where a valley penetrates the mountains and a creek has accumulated a cone of debris with a steep slope on the bottom of the sea. To the west, the mountain slope declines and extends gradually; this is the inhabited and cultivated region of the island. To the north, the range ends in hill country, from which only isolated peaks emerge. Until recently, there has been hardly any information on the geological composition of the island, the surface of which covers 600-700 geographic square miles, apart from the old report by De Mailla that the country shows traces of former volcanic activity, in spite of the lack of verified documentation on the existence of active volcanoes(2), [p. 533] and the more recent reiterated fact that there is coal near Kelung. When the Royal Prussian frigate Thetis approached the Tamsui harbor on the northern coast, I had the opportunity of looking at some rocks on the island. The ship could not enter the harbor because it drew too large a draft, and, therefore, our stay was limited to one day. Because the site of inspection was not unfavorable and since every statement of fact about such an unknown land is worthwhile, I hope that even a mere outline of a report such as the following one will be accepted with the reader's indulgence.

1. Surface formation.

The northern tip of Formosa carries the name Syauki Point; it is connected to the northwest and northeast coasts, both of which bend back towards the land in a flat curve. The innermost portion of the former contains Tamsui harbor, and the harbor of Kelung is situated in the innermost part of the northeast coast. Tamsui sits on the broadened estuary of a great river, while Kelung harbor is an indentation on the coast.

If you approach the northwest coast from the sea, you see two high, isolated mountain massifs, between which the Tamsui River discharges; to the left and to the right of these peaks there is a seemingly level plateau, 400-500 feet high. The northern mountains reach a height of 2800 English feet, according to the English marine charts of the island and the separate chart of Tamsui harbor; the southern mountains rise to 1720 feet. The former peak seems to be the beginning of a mountain range that extends further to the east, while the southern massif is smaller, more isolated and more abruptly precipitated. The forms of both are reminiscent of trachyte mountains. Not long before reaching Formosa, we had been in Nagasaki on Kyushu, which is surrounded by trachyte mountains. Everyone was surprised by the similarities in the mountain formations near Tamsui and those at the Japanese harbor. [p. 534] Only the arrangement of the group is completely different on Formosa, especially because of the isolated appearance of both mountain massifs. The northern mountain range falls towards the coast with a slope rich in ridges and ravines onto a flat country which gradually declines towards the sea. We could make out many villages and much cultivation on the slopes from the deck of our ship. The southern mountain range has no such level land declining toward the sea, but the plateau (400-500 feet in height) continues towards the southwest. It consists of completely horizontally layered strata, whose intersecting lines continue horizontally along the coast and are visible for a long distance because of the sparse vegetation. The slope towards the sea cuts them off quite abruptly, but on closer inspection one can see a large number of small gorges and valleys, which interrupt the continuity of the slope. The seemingly level plateau dissolves into a hill country, whose irregular formation but regular height makes the profile seem horizontal. Between the two mountain massifs the continuation of the plateau is only visible in two slopes, which are somewhat less steep than their upper counterparts while embedding the broad canal of the Tamsui River. Further upcountry, one can see mountain ranges in the distance.

The northeast coast is described as beautiful hill country; only in the distance can one supposedly see a few higher mountains. The harbor of Kelung, the only one on Formosa that is suitable for larger ships, is a small indentation in this hill country and is protected from the north by an island situated in the harbor.

Except for the region surrounding Tamsui and Kelung harbors, there is almost no information on the surface formation and the composition of the northern part of Formosa. Since the time of the Dutch settlement, which ended after rather short duration in 1662, European ships have only occasionally visited some villages on the coast in recent times. Kelung has been visited occasionally and was recently opened to foreigners for very limited trade. Tamsui has only rarely been visited. For years only a few small merchant vessels have been trading illegally between this port and officially closed harbors in China, and therefore one can usually see some European ships in the Tamsui harbor; [p. 535] we found three of them. However, because visitors in both places have been either merchants or officers charting the coast, no one was interested in continuing into the interior. The only one who has seen the land somewhat thoroughly is the Jesuit priest De Mailla, who in April and May of 1714 made an astronomic chart of the western side of the island for the Chinese government. Of the more recent travelers, only Mr. Swinhoe from Amoy has undertaken a short excursion (in 1858, of 55 hours), starting from Kelung (3). In addition, the captain of an English trade brig told us that he had sailed 40 sea miles upstream on the Tamsui River on his ship. From the coinciding reports of these two last investigators it becomes clear, as is already stated on the newest maps, that the Tamsui River is created from two rivers four sea miles above its discharge into the ocean (which is beyond the effect of the tides), of which the smaller river runs from the east and passes closely by Kelung, but, as is indicated, does not constitute a canal to that port. Swinhoe followed this river upstream, and there were great difficulties because of the numerous cataracts. In the end he arrived in a broadened basin that is only fed by mountain streams. The surrounding area was wild and rocky. From that basin, Kelung was only two English miles across the hills. According to the indications on newer maps, the Tamsui River has a very long course, and is navigable for small sea ships as long as its course in a southeastern direction, which amounts to not quite 40 sea miles. Thirteen miles upstream there lies the city of Mangka with its 20,000 inhabitants. The land there is supposedly extraordinarily fertile and therefore fuels the prohibited trade from the Tamsui harbor with China with the value of its products.

[p. 536]

2. Mining in the Tamsui harbor.

The Tamsui harbor is at the estuary of the Tamsui River, where it broadens into a basin and is still fully affected by the tides. Our ships landed not far from the small town of Ho-bi [Hoowei], a short distance from the intersection of the two mountain massifs mentioned above. The land between these peaks and the level region that stretches toward the sea (in which the harbor is set) is part of the hilly slope that gradually stretches toward the sea from the mountains. My investigation had to be limited to this part of the Tamsui region because the shortness of the stay precluded a more thorough inspection of the massifs above. There was a lot of data available in this foreland, which helped me ascertain some geological facts. The investigated rock specimens belonged to two different series, of which one is connected to trachyte outcroppings, while the other belongs to later formations.

A. Rocks from the trachyte period.

1. Trachyte. If the very formation of the two massifs already indicated their composition as being from trachyte and trachytic friction conglomerates, this was certain for the northern blocks, which have come down in masses. There can be almost no doubt that the southern counterpart consists of the same rock types. The blocks belong to two different kinds of trachyte. The first kind, which dominates, is a hornblende-oligoclase trachyte without sanidine or augite, a fairly common middle-level rock. The fine-grained grey basic material recedes against the mass of enclosed crystals. The hornblende is dark red-brown, extremely petalled, and forms small pillars of 2-6 lines in length. Their arrangement is very strange. When the rock is broken, on some of the exposed surfaces one can see the shining breaking points of the crystals weaving into each other from all directions, similar to what is seen in an actinolite slate. In contrast, on the exposed broken surfaces arranged vertically to these, there are hardly any broken crystals. Therefore, the axis-surfaces a b have to be more or less parallel throughout the whole rock, whereas of single axes only c has the same direction throughout the stone. [p. 537] The oligoclase is greenish-white; its crystals are smaller than those of the hornblende. Regarding their arrangement, they follow a similar pattern. On the same broken surface, on which the shining breaking points appear on the hornblende, one sees evenly rounded, dimly shining surfaces of the oligoclase, while on the horizontal cleavages, there are oblong intersections, for the most part not shiny. Both minerals give this trachyte an extremely characteristic imprint. Next to these there are also green, hard mineral inclusions that become visible only after closer inspection. The entire rock has an irregular fragmentation; it breaks easily and in the direction in which the breaking surfaces of the crystals lie, but it breaks with difficulty, usually splintering in the other direction. The weathered layer is double: the outer layer is rust-brown and paper-thin, while the inner one is dark reddish-brown and as thick as a line; both are clearly distinguishable.

The second trachyte is basalt-like, very brittle; it breaks into flat bowls and sharp-edged pieces and only consists of a grey-black, fine-grained mineral mixture, in which are found irregular brittle crystals of dark leek-green, translucent augite. The weathered layer is the same as in the above type, but the outer cover is more yellowish-brown.

I could not discover in which combinations the two trachytes appear. It seems to me that there are additional variations, but all of them are more subordinate.

2. Rough trachytic conglomerate. In several places along the coast near Ho-bi one can see a rough conglomerate that constitutes the basis for all further sediment formations of the hill country. It includes jagged fragments of different trachytes in a sturdy trachytic mass. This conglomerate is possibly an eruptive mass that melted and then solidified into one layer and would then be included into the category of formations that are most fittingly called eruptive tuffs [whine stones]. The base rock is too sturdy to be merely sedimentary, and the spread is too even to be only eruptive. The level on which the rocks become visible is overall the same; one finds them only on the coast, which is then covered by blocks because of the weathering that occurs. Such a sea of stones lies at the end of a huge sand barrier, which the Tamsui River has accumulated in its estuary.

[p. 538]

3. Trachytic tuffs, weathered to a reddish-brown, earthy mass. Nowhere did I find these tuffs in their original form, even though they formed the whole hill country over the level of the above-mentioned conglomerates. However, one can still clearly see the structure of the rocks and the horizontal layering. Some layers are filled with trachyte blocks that are weathered to an equally earthy mass as the binding agent, but are distinguishable because of their yellowish-white coloring and the denser bonding of the decomposed mineral mixture. The huge size has to coincide more or less with the height of the hills, first of the slope of the trachyte mountains, which rise about 400-500 feet. These decomposed tuffs are the reason for the fertility of the area around the harbor of Tamsui. The uneven plain of the slope is covered with fields; flat valleys with soft slopes are sunken into the hill country. The ground is leveled by humans, and terraced into rice fields. In addition to this main staple crop farmers also grow all kinds of grain, cotton, sweet potatoes, yams, sugar cane and other useful products, and only the limited area precludes great harvests.

The three formations from the trachyte period almost exclusively compose the area around Tamsui harbor. The tuffs seem to spread out very widely. The regularity of the mountain massif formation and the unhindered horizontal layering make it more than probable that the entire expansive plain, which stretches lengthwise along the coast southwest of Ho-bi, is formed of tuff layers. As noted above, its horizontal layering is visible in the layering surfaces, which cut through the sea in horizontal lines.

B. Recent formations.

Recent formations play a rather subordinate role, but give some indication of the current geological processes taking place along the coast and of the geognostic construction of areas farther away. These formations are also threefold: gravel, dolomite banks and sand.

[p. 539]

4. Gravel. This consists of fully rounded tills that are cemented together through a chalk and clay-like binding agent and form rough terraces of up to 30 feet in height situated at the bank below Ho-bi. The main constituents of this formation are the same boulders that the river still carries with it: trachytes of different kinds; rough-grained granite from grey quartz, reddish orthoclase, black glimmer in deep boards and very sparingly scattered yellowish-green oligoclase; but especially rough-grained white, yellow and light-red quartz sandstone. This latter stone is found in great blocks, which indicate that it originates from not too far away. I did not find crystalline slate and chalk.

5. Shell breccias are located on the slopes of the tuff hills everywhere up to a height of a little more than 100 feet. The binding agent of the loose rocks is the ocher-colored sandy earth. Most of the graves in this area are situated on this bank.

6. Sand seems to be the youngest formation. It constitutes the anchorage in the harbor, the banks in the river, the barrier that separates the harbor from the sea, and it covers the trachytic conglomerate on the right-hand bank of the river estuary in the form of infertile hills that are often only covered with pandanus.

[Here the original text includes a rough sketch of the typical profile on the Tamsui River. It shows sand and trachytic conglomerate near the seashore, gravel further up the slope, followed by trachytic tuffs and shell breccias higher on the hills, with the mountain top covered with trachyte.]

The mutual stratification relationship of the rocks described here can be seen in the above typical profile of the geological formation of the hills on the Tamsui River. [p. 540] One can see the complete chronological order in almost any ascent from the river to the heights; only the valleys usually lack the recent formations.

The developmental history of the small area we are concerned with is clearly visible in the described composition. The trachytic mountains, the trachytic friction conglomerates and the tuffs belong to an earlier, probably tertiary period, after which the land rose and the river bed was formed. The depression was only formed in a fairly recent period, but the terrain must have at least reached up to the surrounding heights, to which the recent formations extend on the slopes. The broadly eroded riverbed was filled with the boulders that the river carried down, and in the last periods of the lowering process (when the brackish waters, moved by the tides, washed around the slopes of what now are the tuff hills), the river sand was deposited with the earthy destructive products of the tuffs and mixed with the fragments of the shells on a higher level than the gravel banks. At this moment there was another upheaval in this region. Again the river dug its bed deeper into the material that it had deposited itself. Then the atmospheric waters further eroded the slopes and removed the cover lying on the tuffs in many places, until the only thing left were the recent formations that we can see today on the slopes. The river now covers the flatter portions of the exposed areas with boulders and sand. The sand dunes have already partly become flat hills above the level of the sea, and they are rising further and further above it.

3. Massif formation in the harbor of Kelung.

We are in possession of a valuable report about Kelung written by Mr. Jones, ship chaplain who accompanied the Macedonian, one of the ships that Commodore Perry sent to Kelung in 1854 in order to investigate Formosa's coal. Unfortunately, Mr. Jones focused completely on the description of the coal seams, their strength, stroke and fall and only mentions in passing the fact that the entire hill country near Kelung consists of sandstone, in which the coal is embedded. [p. 541] One could think that this was a continuation of the considerable accumulations of quartz sandstone, whose fragments can be found in boulders in Tamsui. However, it is more likely that this is tuff sandstone. Jones describes a place called Image Point, a protrusion of the land that received its name because of the weathering of the layered stone into the shape of baroque figures. The base rock formation of these figures consists of a brown sandstone topped with rounded large black stones. A sketch [by Jones] shows that they are similar in shape to the earth pyramids of Botzen: thin long cones, upon each of which sits one of these black stones. These latter ones are probably darker trachyte in this area so marked by trachyte, and the whole formation seems to belong to the tuff mountain massif.

This is unfortunately all that Mr. Jones says about the geognostic composition. We owe two interesting facts to the Kelung harbor sailing rules compiled by Lieutenant Preble of the Macedonian. These rules state that Kelung-khid, or Kelung Island, is a small volcanic isle, 500-600 feet in height, located in the entry to the harbor. Jones says that the island consists of syenite. Whether this is accurate remains to be seen. Furthermore, Preble writes that the flat island called Tong-fung-si (Collinson’s Bush Island) is covered with a layer of old corals; it is located on the east end of the harbor, which was supposedly raised up from below sea level along with the underlying sandstone.

Coal bed of Kelung. For a long time, there had been uncertain accounts of fossil fuel in the northernmost part of Formosa. As much as the exploitation of coal in a place so much closer than England was desirable, especially when taking into account the growing steamship traffic in the East Asian seas, there had been no trustworthy investigation of Formosan coal. Perry gave the Macedonian certain instructions regarding this question, and we owe the following description of the Kelung coal bed to Mr. Jones's report. In July, 1854, the coal beds east of the harbor were investigated, and a small map was made. In spite of the obstructions to travel beyond the harbor, Jones arrived at his destination. [p. 542] According to him the beds begin close to the harbor and stretch from there eastwards, where one can find them in the small bays of Qua-se-ku and Kea-lau. The identified seams begin on the surface and were mined from there on inwards; a large number have turned out to be extractable. Whether it is hard or brown coal is not reported, but the latter is more probable. According to tests that Commodore Perry ordered, the Formosan coal turned out to be of a lesser quality than the "best" English coal, which was the only [English] type it was compared to, but much better than the Japanese product, which is not a great achievement. It burns quickly, but there is surprisingly little residue. The costs of exploitation for a ton of twenty hundredweights would be one and one-quarter dollars; the conveyance from coal bed to ship would hardly add any expense at all because of the proximity of the island's best harbor. Therefore, the price comparison to English coal would seem extraordinarily advantageous, since English coal is traded in Shanghai at 69 shilling or 15 dollars a ton. Everything said in the American report about Formosan coal should lead one to think that it has a very bright future. Until the date of the report (1854), this coal was only extracted to a very limited extent, according to the needs of the inhabitants. It was never exported or even sold locally to ships, and even the Macedonian could organize a sale (100 hundredweights for 16 dollars) only through tricks. Since that time, seven years have passed. At first, one knew of the existence of this Formosan coal; now one has more certain knowledge, and yet we are still dealing in Kelung coal from Chinese port cities. However, Formosan coal has still not been able to compete with English coal even in East China. She shares a fate with Nagasaki coal, which has similar traits, and yet is better than Yokohama coal (and this latter type was the one that Perry used in his comparisons). From Nagasaki and from Kelung, Chinese junks bring coal to Shanghai and sell it to the Chinese. However, steamships prefer the English coal in spite of the price that is four to five times higher, and only sailboats that also use a propeller make use of the brown coal of both harbors. [p. 543] It is the same as the coal of Labuan near Borneo and other Sunda Islands. The brown coal can't compete with the most expensive hard coal.

While Jones only investigated the coal beds to the east of the harbor, Swinhoe, who visited the harbor four years later (in 1858), writes only of coal beds west of the harbor in a bay:

These mines are worked by Chinese, who live at their entrance in huts built of straw and wood. There are eleven or twelve excavations, their mouths opening out, at different heights, on the side of a hill facing the sea. I went to the end of one, guided by a man bearing a lighted piece of twisted paper. The excavation, which ran in a horizontal direction, varied from about four and a half to three feet in height and three to ten feet or more in breadth. The strata of coal run along on both sides in parallel lines from one to three feet in thickness. The roof above, and the floor underfoot, were composed of sandstone. Water was constantly dropping from the roof, and this mixing with the sand formed a slimy mud. The hole ran in pretty nearly in a straight line for two hundred and forty paces; at the end it took a sudden turn to the right. Small wicks in saucers of oil lighted the way along, and we found five or six men at work in a state of nudity with pick-axes, blunt at one end and sharp at the other. The coal which they obtained was very small and bituminous, and burns fast but with a great heat and flame. It is very certain that they get the best there is in that locality. They asked twenty cents a pecul for it, and declared that five men, at work in a mine for twenty-four hours, did not procure more than thirty baskets, containing a pecul each(4).

When one takes these reports together, it seems that coal is common around Kelung, and forms several seams between one and three feet wide that are worth exploiting, but those coal beds are also included in tuff sandstone and it is a tertiary brown coal.

4. The Sulfur mines between Tamsui and Kelung.

Sulfur is one of the most important export products of the northern harbors of Formosa. Until a few years ago, it was unknown where this sulfur was collected; now it has become known that there are sulfur mines on the northern tip of the island, [p. 544] exactly on the north side of the higher of the two mountains near Tamsui. Swinhoe is probably the only European who has visited the mines (in 1858). He found them deserted, even though they had been heavily worked shortly before. The mandarins of Fokien had sent soldiers to prevent their exploitation, and now the mines can only be worked secretly. Swinhoe wrote the following description:

The sulphur was produced in a chasm, appearing as if the green hills, covered with coarse grass, had been riven in sunder, thus forming a deep valley of limestone tinted with yellow and red; in some parts of this chasm the hot steam was gushing out in jets, with tremendous noise and force, like the steam from the escape-pipe of a high pressure engine; in other spots small pools of pure sulphur were bubbling, and merely wanted the ladling out and cooling to form the sulphur of commerce. At the bottom of the barren ravine rippled a foul rivulet, carrying off the sulphureous oozings from the ground. Standing on the top of a hill, I looked down upon the scene: one sulphur pool was not more than fifteen feet below me, and its stench was intolerable; the earth under foot crumbled and groaned, as if it were going to give way; pieces of limestone, covered with the crystals of sulphur, lay strewn about; while beetles and butterflies, unhappy victims to the noisome exhalations, were scattered wingless and legless ove r the ground(5) .

At the same location of these sulfur mines, Klaproth located one of the four active volcanoes on Formosa. From Swinhoe's description it seems to become clear, however, that there are only remainders of volcanic activity here, and more precisely a solfatara instead of an active volcano.

The most general conclusions can be summarized as follows: The northernmost part of Formosa in its interior consists of older rocks (of which numerous fragments are washed away in the Tamsui River) and of trachyte. Near the coast the land seems to be exclusively composed of tertiary eruption products, which in part loom in high trachytic mountains, and in part spread out further from these mountains as eruptive tuffs, [p. 545] and form a mighty system of sedimentary tuffs. The latter constitute a far reaching hill country that surrounds the trachyte-mountain range near the Tamsui border, stretches to the southwest as a low-lying plain to Namkan Point and Paksa Point, and probably forms the entire area around the Kelung harbor, including the coal beds there. There are traces of former volcanic activity that continue in their last stages in the sulfur mines near the northern tip of the island.

The recent formations point to the fact that the island is currently in the process of a slow upheaval, similar to that of the Liu Kiu Islands and Kyushu.


1. On the maps one finds in the middle of the island a mountain indicated as Mount Morrison, 10,800 feet high, and to the north a mountain with no further designation at "about 12,000 feet." No European has ever been near these mountains. The latter height estimate stems from Mr. von Humboldt, who concluded that the absolute elevation was at least 11,400 feet (1900 toises [1 toise = 1 949 m]) from the fact that the highest mountains in Formosa are covered with snow for most of the summer. I could not find out where the height estimate for Mount Morrison came from.

2. Usually, four active volcanoes are specified on Formosa, because of a statement made by Klaproth, who received this information from Chinese reports on Formosa. But until this statement has been affirmed by travelers, one should treat it very seriously, since mistakes are very likely in the interpretation of Chinese sources, especially regarding the "Fire Mountains." I will try to show that for one of these cases on the northern tip of the island there certainly is a wrong interpretation.

3. This was reported in the Journal of the North-China branch of the Asiatic Society, Volume 1, Number 2, May 1859.

4. Translator's note: This section is quoted directly from Robert Swinhoe, "Narrative of a visit to the island of Formosa," Journal of the North-China Branch of the Royal Asiatic Society 1, ii (1859): 157.

5. Translator's note: This section is quoted from Swinhoe, 1859, pp. 159-160.