Guest blog: ‘Unfamiliar items’ on maps – Legal drugs

By Heleen Hayes

While georeferencing Indonesian maps of the late 19t till early 20th century, I came across some unfamiliar, but surprisingly interesting buildings and area descriptions. This made me look around the internet to find out more about these subjects.

A striking one is an ‘Opiumverkoopplaats’. Are we looking at an official sales point of opium? Legal drugs? Yes, that’s right.[1] Even though the detrimental effects of opium use had been clear for centuries, the government allowed trade and use of the product. A nice cashflow was apparently more important than ethics and the government monopoly continued. [2] I will not go further into politics and ethics in this blog, but if you google ‘opium’ or ‘amphioen/amfioen’ there are lots of (Dutch) articles to be found for those who are interested in the subject.

Opium was manufactured on Java. In 1894 the first opium factory was built on the Struiswijk estate. In 1901 it had become too small and a larger, new opium factory was built on the surrounding Keramat area. A special railway was built to transport the necessary raw materials from the harbours.

A traveller tells us the following (in Dutch): “Wie het voorrecht heeft, die inrichting te kunnen bezoeken, zal zich niet beklagen over den er doorgebrachten tijd. Die fabriek, waar men het vooral door chineezen zoo geliefde, phantastische droombeelden opwekkende bedwelmingsmiddel vervaardigt, is gelegen aan de fraaie laan, die naar Buitenzorg voert. [3]”,

“Whoever has the privilege to visit these facilities, won’t be complaining about the time spent there. This factory, where one manufactures the intoxicant that is especially loved by the Chinese, lies on the attractive lane leading to Buitenzorg”.

The first opium factory was built on the estate of Landhuis Struiswijk in 1894[4], though we only see the manor (Landhuis Struiswijk) on this 1904 map issued by an Amsterdam book shop:


To enable a higher production, a new factory was built on the surrounding area. At the same time a specially designed railway was built to enable swift transport of the tons of ‘heulsap’, the fluid harvested from the poppy seed bolster, from the harbours to the factory.
This map shows the Batavia Gouvernements Opiumfabriek and the railway on a map by the official tourist bureau of Batavia, dated ca. 1910. The manor house has disappeared.


Click here for the map in the KITLV media library

The local book shop apparently proudly stocked a postcard featuring the factory:


Click here for the image in the KITLV media library

Points of sale (marked ‘opiumverkoopplaats’ on the maps) were to be found all over the islands. One of them is pictured here. Some sources say they were painted light blue and that close to where a person lived they would be situated about 5 km apart, in larger kampongs (villages), near the local passar (market).[4] I have found some of those points of sale on maps, but not as many to verify this observation. Actually I have been searching the database looking for the indication ‘opiumverkoopplaats’ to show in this blog, but I have not been successful yet.



Click here for the image in the KITLV media library


[2] and more pages.

[3] ‘Nou … tabé dan!’: De ‘bootreis’ naar Indië met de Rotterdamsche Lloyd en de ‘Nederland’ tussen 1899 en 1949. Bert L.T. van der Linden


A milestone: 90% of the maps has been georeferenced!


Today our Maps in the Crowd project reached a milestone, with currently 90% of the 6678 maps georeferenced. With only 10% left, the project draws to a close. When all maps have been done, probably sometime in November, we will organise a meeting in the University Library to thank our volunteers. All top-10 finishers will be invited to attend a ‘Maps in the Crowd afternoon’, including lectures, an award ceremony for the overall winner and the runners-up, a tour of our special collections and a reception. We kindly ask our top-10 volunteers to contact us, since we don’t have participants’ personal details. We will then personally invite you to our event. More information about the closing event will follow as soon as possible.

A reservoir of maps: topographical maps of Java before the building of embankment dams

The maps from the KITLV collection that are selected for georeferencing consist for a large part of sheets of topographical map series. The Topographisch Bureau in Batavia was the largest colonial mapping agency in Asia. Only for Java, various series on scales 1:25,000 and 1:50,000 were published in the first half of the twentieth century, as well as series of the residences on scale 1:20,000 starting from the late nineteenth century. Furthermore, topographical map series on scale 1:50,000, 1:100,000 and 1:250,000 were published by the Americans during World War II.

For this blog we concentrate on the 1:25,000 map series and the region between Jakarta and Bandung. In this area three large reservoirs are created on the Citarum River: Waduk Jatiluhur, Waduk Cirata and Waduk Saguling. When we compare the topographical maps of the early twentieth century with the modern situation on Google Satellite, we see the topography changed dramatically.

The first example shows the northern part of Waduk Jatiluhur. The building of this first embankment dam started in 1957. The Jatiluhur Dam became operational in 1967. The dam, built for hydroelectric power generation, water supply, flood control, irrigation and aquaculture, is located in the upper right corner. The map sheet is published in 1918. We see a relatively small meandering river Citarum dominating the map image. Today half of the area of this map sheet is covered by the artificial lake.

1 Jatiluhur
Overlay of the 1918 topographical map of the Waduk Jatiluhur area on Google Maps
Click here for the georeferenced 1918 sheet of Waduk Jatiluhur

In the 1980’s the two other dams were built further upstream, resulting in the involuntary resettlement of over 100,000 people. In the second example we see the northeastern part of Waduk Cirata. The Cirata Dam was primarily constructed for hydroelectric power generation. The topographical map sheet was published in 1920.  Apart from the fact that the water body before the dam is larger than the earlier situation, the width of the river behind the dam clearly decreased. Despite the changed topography, the map was easily georeferenced by using control points along the – still existing – major roads.

2 Cirata
Overlay of the 1920 toographical map of the Waduk Cirata area on Google Satellite
Click here for the georeferenced 1920 sheet of Waduk Cirata

The third example shows the southern part of Waduk Saguling, the most upstream reservoir of the Citarum River. The map sheet was published in 1923. In comparison with the other two reservoirs, the outline of this lake is more erratic in shape because of the hilly nature of the terrain. It is striking how accurate this outline matches the contours, indicating the altitude, on the topographical map.

3 Saguling
Overlay of the 1923 topographical map of the Waduk Saguling area on Google Maps (detail)
Click here for the georeferenced 1923 sheet of Waduk Saguling

The three examples of locations were the landscape changed dramatically, show how useful historical maps can be to understand the history of the landscape and the impact of human interference, in this case the building of embankment dams, on the spatial environment. Accurate map series, like the early twentieth century topographical map of Java 1:25,000, are extremely useful for such analyses.

Kotagede revisited

Earlier we reported about a manuscript plan of the Royal Tombs at Kotagede, the historic neighbourhood in Yogyakarta where the the first capital of the Mataram sultanate is located. In the ground plan for types of graves where distinguished by colour. Because the map was in Javanese script we couldn’t explain more about it at that time. Later, one of our volunteers noticed that there was a copy of the map in Dutch in the collection too.

Original plan of the Royal Tombs in Javanese script

As we suspected, there is a kind of hierarchy in the subdivision of the graves. The purple graves are those of kings and queens that have been canonized by the people. There are only twelve purple graves in the map. The green tombs are of princes and princesses. The red graves are those of further royal relatives. Together these three categories count 295 graves, these are numbered and listed on the map, with the name of the person who has been buried. Of the yellow graves, uncoloured in the copy, it is unknown who they are. These are another c. 250 graves, so in total the cemetery count over 500 graves.

Dutch copy of the plan

Other elements that are indicated in the map’s legend are the wells and basins for cleaning (blue), houses for the keyholders (brown, yellow in the copy) and ‘sawo and other trees’ (point symbol). The lines in red and black indicate, respectively, the stone and wooden walls of the grave houses. Moreover, the Dutch copy of the map gives more information than the Javanese original. On the Dutch map the names (and functions) of the buildings and courtyards are given.

Legend in Javanese (left) and Dutch (right)

These manuscript plans are those in the KITLV collection that are drawn on the largest scale, of about 1 : 150. The complex with mosque and walled courtyards are still visible on a modern satellite image.

kotagede luchtfoto
Satellite image of the complex

Rouffaer and the sketch maps of Adonara

The island of Adonara in the East Nusa Tenggara province is one of the Lesser Sunda Islands, squeezed between Flores, Lembata and Solor. With an area of less than 500 km2 and a population of about 100,000 inhabitants it is one of the smaller islands of Indonesia. Due to its remoteness it took until 1911 before the Dutch published a printed map of the island. In November 1911 the Topographische Inrichting (Topographical Service) of the Dutch East Indies in Batavia published a sketch map of the island (Schetskaart van het Eiland Adonara), on a scale of 1:100,000. On the printed sheet the sources were listed with which the map was compiled: a hydrographic chart, a map of the island published in the journal of the Royal Dutch Geographical Society and ‘other sketches’.

Adonara D_E_44_7
The map of the Topographische Inrichting, November 1911 (UBL, KITLV D E 44,7).
Schetskaart van het Eiland Adonara / Topographische Inrichting Batavia, 1911 (november)

The sketch map in the journal of the Royal Dutch Geographical Society (KNAG: Koninklijk Nederlands Aardrijkskundig Genootschap) was published earlier in the same year of 1911, but on a smaller scale of 1:200,000. The map titled Solor-Eilanden: Schetskaart van het Eiland Adonara by J.D.H. Beckering. The map belongs to an article by Beckering that described the islands of Adonara and Lembata. Both printed maps show strong resemblance, especially in the outline of the island’s coasts (based on the hydrographic chart), the borders of the various ‘landscapes’ or ‘principalities’ (‘rijkjes’ in Dutch), footpaths and river courses (for which supposedly the Topographische Inrichting used the KNAG-map as source). The depiction of the elevation of the terrain is very different. On the KNAG-map the heights are indicated with contours and on the map of the Topographische Inrichting the heights are mapped surprisingly less accurate with a more indicative use of symbols.

Adonara KIT KNAG
The map in the KNAG journal, 1911 (UBL, KIT, KK 165-04-04).
Solor-eilanden : schetskaart van het eiland Adonara / J.D.H. Beckering, 1911

Adonara Chart
The hydrographic chart, 1911 (UBL, KIT, KK 021-05-10).
Kleine Soenda-eilanden : straten tusschen Flores en Lomblem / Ministerie van marine. Afdeeling hydrographie, 1911

However, in the KITLV-collection two manuscript sketch maps of the island are kept, one dated 8 February 1910, the other undated but in all probability from the same year. Could these two hand-drawn maps be the “other sketch maps”, used as source for the map published by the Topographische Inrichting? Two answer this question we have to look to Gerret Pieter Rouffaer (1860-1928), explorer and collector and map cataloguer for KITLV, in that time a colonial institute in The Hague. In the first months of 1910 he joined the expedition of the Topographische Inrichting, led by Major P.J. Spruijt, to survey the smaller islands near Timor, including Adonara. In the same months, Captain Beckering led a military expedition to Adonara and Lembata in order to disarm and register the local population. For eight days Rouffaer stood under Beckering’s convoy on the island of Adonara (and later for another eight days on Lembata). It was Rouffaer, who had been editorial secretary of the KNAG journal from 1906 to 1908, who asked Beckering to write a description of the two islands for that journal. By the way, the mapmaker of the KNAG journal, Coenraad Craandijk, was appointed by Rouffaer in 1906.

Adonara D_E_44_6
The dated manuscript map, 8 February 1910 (UBL, KITLV, D E 44,6).
Schetskaart van het eiland Adonara, 1910

Both manuscript maps are not in the handwriting of Rouffaer, apart from the pencil dating. Nevertheless, they both have to be related to the expeditions to the island in February 1910. When we compare the dated manuscript sketch map with the printed maps, it is clear that this sketch is less accurate. Especially the outline of the island is inaccurate, but for example also the location of the Ile Boleng volcano. The manuscript map has also contours for heights but these do not match with the KNAG-map or the other manuscript map. The dated manuscript map probably drawn a little earlier than the other manuscript map, since the place names in the southwestern part of the island are missing.

Adonara D_E_44_5
The undated manuscript map, [1910] (UBL, KITLV, 44,5).
Schetskaart van het eiland Adonara / [G.P. Rouffaer], [1910]

The undated manuscript map, drawn on tracing paper, is more accurate. The coastline is probably derived from the hydrographic chart. Again, the contours for heights show a different pattern. The borders of the ‘landscapes’ are more or less comparable on both manuscript maps, but differ from the pattern on the printed maps, especially in the southern districts of Trong, Lama Hala and Lonik Boerik. With only a few exceptions, all place names on this map match the names of the two printed maps. Therefore, it can be assumed that the manuscript maps could have served as a source for these place names in particular.

J.D.H. Beckering, ‘Beschrijving der eilanden Adonara en Lomblem, behoorende tot den Solor-groep.Tijdschrift van het Koninklijk Nederlandsch Aardrijkskundig Genootschap, Tweede Serie 28 (1911), pp. 167-202.
Paul van den Brink, Dienstbare kaarten. Een cartografische geschiedenis van het Koninklijk Nederlands Aardrijkskundig Genootschap en het Tijdschrift 1873-1966. Houten, 2010.
Frank Okker, Rouffaer. De laatste indische ontdekkingsreiziger. Amsterdam, 2015.

Executive decisions

Sometimes when you are georeferencing, you have to take executive decisions. Last week one of our volunteers stumbled upon this map of ‘Poelau Kisser’, according to the catalog record an island in the Maluku Tenggara area. Soon the small island of Kisar was found. It is located just northeast of Timor.


Now the georeferener looked at the island and the map and noticed the island on Google Maps looked much more like the pencil sketch than like the printed map. She was in doubt which map to georeference, but made the ‘executive decision’ to georeference the printed map, because it is interesting to see how difficult it must have been in the 1910’s to map a tiny island. Looking at the Google terrain map it seems highly unlikely that the island had a different shape back then.

Kisar Google Maps and Terrain
Kisar in Google Maps and Google Terrain

One wonders when the penciled map was drawn. It seems the user started by correcting the printed map, but soon found out the map was so far from reality, it was better to sketch a new map on a blank part of the page. The pencil corrections are probably made by Gerret Pieter Rouffaer, map librarian at KITLV and considered the last explorer of Indonesia (in his recently published biography). In 1910 he traveled to Timor from where he explored the surrounding islands.

Click here to compare the georeferenced map with Google Satellite: Poelau Kisser

This blog is written by Heleen Hayes, on of our enthusiastic georeferencing volunteers.

How to map coral atolls?

Ajoe-EilandenMaps in the Crowd brings us to remote and exotic places. An interesting map of a small group of islands is the ‘voorlopige uitgave’ (provisional edition) of the map of the Ajoe-Eilanden (Ayu Islands) from 1958, as part of the 1:100.000 topographical map series of Dutch New Guinea. These islands are located just north of Waigeo, northwest of New Guinea. This small archipelago is formed by two coral atolls with five islands, of which two are uninhabited. The Ayu Islands are known for its good snorkeling and scuba diving sites and its beaches as breeding grounds for the leatherback turtle. When we compare the georeferenced map with Google Satellite we can see how accurate and detailed the intertidal reefs are surveyed.

Ajoe-Eilanden legenda It is interesting to have a look at the legend below the map. Alongside common symbols for borders, settlements, roads, waterways and vegetation there is a special signature to indicate coral reefs. These reefs are the most prominent features on this sheet. You don’t see them in Google Maps but in Google Satellite they are prominently visible.

Click here to compare the georeferenced map with Google Satellite:Ajoe-Eilanden

Mapping an volcano eruption

One of the largest recorded volcano eruptions ever is the eruption of the Krakatoa in 1883. Two-third of the original Krakatoa island, situated in the Sunda Strait between Sumatra and Java, vanished after the eruption. Tsunamis caused by the eruption killed over 36,000 people in the wider area. By order of the Governor General of the Dutch East Indies, the National Geographical Institute in Brussels (surprisingly not the Topographical Bureau in either Batavia or The Hague) published a series of 43 maps, plans and graphs on 12 sheets in relation to this natural disaster. This map shows the terrain where ashes were distributed, as far as the Cocos Islands, 1200 kilometers away. The red numbers indicate the thickness of the ash layer. In the direct environment of the volcano the ash layer was more than one meter. Other maps in this series show, among other aspects, the parts of the original island that were vanished, the distribution of the tsunami waves and the flooded coasts in the Sunda Strait area.


View the georeferenced map here: Kaart van het westelijk gedeelte van den Indischen Archipel

Map of the royal tombs at Kotagede

One of the most striking maps that is georeferenced recently, is this manuscript plan of the royal tombs at Kotagede. In this historic neighbourhood in Yogyakarta the remains of the first capital of the Mataram sultanate, including its royal cemetery, are still visible. The large plan, measuring 81 x 93 cm., is in Javanese script and drawn around 1890. In the legend 295 graves are listed and in the map itself four types of graves are distinguished by colour: purple, green, red and yellow.


The image can be viewed in detail in the KITLV digital image library:

Urban planning from the nineteenth century

From time to time we will post some background story in this blog about an object in our collection.

It’s always nice to see a direct result from your work in the form of a reward or a goal you reached. With our Maps in the Crowd project we can see the result bar progressing to our end goal with each map you, the reader, georeferenced. At the moment we are close to having half of our maps done, thanks to you. Not all things in life show a direct consequence, but studying the past can give us a view how some plans can show their influence ten, twenty and maybe even a hundred years later.

Soerabaja 1898

Our story begins with a map of Surabaya from 1898. The map, made as an attachment for a research about the water supply of the city, is marked with a few red stripes. It was made for a Dutch engineer and architect named Hidde Petrus Nicolaas Halbertsma. While Halbertsma’s report is not easily available, the map clearly shows his plan. To get more clean water to the growing city, water sources from the nearby mountains needed to be used.

Soerabaja 2 1898

In a report from 1917 about the status of the water supply, Halbertsma’s work is mentioned again. The plans he worked on, almost twenty years before, serve as a base for the current urban planners. The report from 1917 states that the city is in dire need of more clean water and carefully explains the cost and processes needed to accomplish this. Haste was needed according to the writers, before 1920 the supply needed to rise with nearly 50% and even more water was needed before 1930.  Aside from a lot of interesting stories about urban planning and water supplies, the main question that rises is: Did they eventually build those reservoirs that our Dutch architect planned? The KITLV collection gives us the option to view the city over the years. Soerabaja 1925

The map above from 1925 does not show any waterworks yet at the spots planned by Halbertsma. Maybe the building did not go as planned, or there wasn’t enough funding, we don’t know. The map from 1943 however…

Soerabaja 1943

This map clearly shows the water supplies of the city. The reservoir or tower with the number 17 beneath it is almost on the same spot where our Dutch architect planned it. The reservoir is build closer to the city then Halbertsma planned, but it is there. Even today the Surabaya water works, build somewhere between 1925 and 1943, are still there: View the water works on Google Maps

If you are interested you can view the maps yourself in the links below. For more colonial city planning or architecture you can go to

Halbertsma’s map from 1898

Map of Surabaya from 1925

Map of Surabaya from 1943