Search this article.   Use 'Enter' to find next.  Use 'Home' to search again.

A Dutch, early 18th century, Captain's watch with Multiple Tide table.

igned: G.J. Nauta, Leuwaerden



Kees Grimbergen
& Cees Peeters.

Table of contents:

More about the construction.
Tide table.
Execution of the multiple tide table.
  Konst der Stuerlieden. (Art of navigation)
Operation and use.
  A typical complication.
  Captain's watch?
  Digital assistent (PDA).

Translated by W. A. van Klaveren.


Recently (spring 2008) an unusual watch was discussed in the television programme "Tussen Kunst & Kitsch" (the Dutch equivalent of The Antiques Road Show). It concerned a golden² pocket watch, signed "G.J. Nauta, Leuwaerden", made c. 1720.

click here to enlarge

fig 1. Captain's Watch.
G.J. Nauta Leuwaerden, ca 1720.

Gijsbert (Gijsbartus) Jacobus Nauta is known as a ‘celebrated’ clockmaker who worked around 1720. According to Ottema (Lit. 8) he is mentioned in deeds of sale from 1718 and 1725, in which he is the purchaser of such property as houses on the Oranje-Eewal. He died in or before 1757, as an advertisement in the Leeuwarder Courant of that year concerns the sale of a "curious silver pocket watch, made by the late Gijsbert J. Nauta". In 1946 Ottema saw a gold watch in a shagreen leather case, signed G. J. Nauta, in the Fries Museum. According to him it was a watch that showed the hour, the month, the date, the phases of the moon and the tides in Dutch and foreign ports on the European continent. It was thus a different watch than the specimen described in this article. Longcase and table clocks by Nauta are also known. (Morpurgo, Lit. 3)

The watch, described here in unrestored condition, is a relatively simple one. It has a Roman chapter ring with half hour divisions (French lilies) and quarter divisions. Along the outer edge of the chapter ring the names of the months with the corresponding number of days are indicated. Originally it had one hand, which probably had a 'tail'

click here to enlarge

fig 2. Back plate.

The verge watch has a duration of c. 27 hours. It has a balance with balance spring and is from a constructional point of view simple. The spring feet, pillars and balance bridge are beautifully shaped, pierced and engraved, which was common practice in those days.

click here to enlarge

fig 3. Side view movement.

click here to enlarge

fig 4. G.J. Nauta's signature.


The movement is somewhat different from most watches of the same period. The motion work is not driven from the centre, but via a 16-leaf pinion, which is engaged with the great wheel (fusee arbor). For this reason the usual centre arbor, running between the plates, is not used here.

click here to enlarge

fig 5. Front plate.

This 16-leaf pinion, which turns fully once every four hours, drives the central part in the dial, which revolves fully in exactly twelve hours.

This unusual central part carries the hand
³. Engraved in it is a tide table, which indicates the times of high water at nine different locations. It consists of two superimposed discs turning around the same arbor. The two discs are the following, with their characteristics described respectively:

  1) A gilt brass disc with a semi circular aperture; along this aperture from right to left the Arabic numerals 6-12 and 1-6 are engraved. At one end of the aperture on the left the word "Hoogh" (High) is engraved and at the other on the right the word "Water". Opposite nine of the thirteen numerals a place name is engraved. The numerals indicate the times of high water when the moon is new (or full) at the corresponding locations. In addition the disc has a moon phase aperture and on the outer rim a moon date ring 1-29½. (Figs 6, 7 and 10)

  2)  A silver moon disc with the phases of the moon and a tide table with twice the division 1-12.

click here to enlarge

fig 6. Chapters

Both discs are toothed and engaged with each other via an 8-leaf pinion. The silver moon disc and 2) the probably originally tailed hand³ are fixed on the hour arbor (Figs 6 and 10).

This arbor is supported in the front plate and is driven via a frictional connection with the hour wheel by the eccentrically placed great wheel pinion.


The phenomenon of tides (high and low water) is related to the position of the moon. The sun also plays a role but to a much lesser extent. The sun has for instance a reinforcing effect when it finds itself on the same side of the earth as the moon.
(spring tide and neap tide twice per lunar month).

 More about Tides

Therefore a tide table is always put on a disc which also indicates the phases of the moon and/or the day of the lunar month.

The moon revolves around the earth in 27.32 days, which during this period (1/13 = 7.5%) runs a part of its orbit around the sun. It takes therefore 29.53 days until the moon has the same phase (position relative to the earth and sun).

On a moon disc of clocks and watches the month is approximated by 29½ days. Therefore after about two years and nine months an adjustment of one day is required.

In this process the earth revolves once around it axis every day (in 23 hours and 56 minutes). If the motion of the moon in relation to the earth is disregarded, then high water and low water occur twice a day with a 12-hour period in between. (˜12.4 hours when this motion is not disregarded)

The times of high and low water vary from day to day, so that after one month a 24-hour cycle is completed. The tide tables consist therefore of a scale of twice twelve hours, usually round the moon date ring(1-29½).

The only thing which is not yet established is the relative movement (position) of both scales in relation to one another. This movement is dependent on the place, for which the tides are given.

Amsterdam (Pampus) highwater, moonphase and age aperture on a 18th century Dutch longcase clock.
(Source: Bernard Meier)

For example, for Amsterdam (Pampus, the sand bank in front of the Amsterdam port) it is high water at 3 o'clock when the moon is new or full. For Harlingen high water occurs at 9 o'clock, when the moon is new or full. Therefore, many tide table on clocks from Harlingen and Leeuwarden have this indication.
A complication is caused by the closing off of the Zuyderzee, which put an end to the tidal movement near Amsterdam and in the IJsselmeer (the name of lake that was created). So for antique clocks and watches tide tables from the 17th and 18th centuries should be consulted.

 End of this section, click here to continue.

2 Restaurations to the case, and case-movement assembly points.  (Click here to view case details)
3 To be discussed:

a) The use of the second hand, which was probably added later, is for the moment unclear. See also Fig. 7.
The fixed angle with the probably original main hand is ≈9.5 lunar days. Is it a coincidence that this is the difference between new moon and neap tide or between half moon and spring tide? The main hand may originally have had a 'tail' (Click here to view hands)

b) The centre of the circular shaped aperture in the gilt brass disc 1) shows signs of a part now missing. (Click here to view details)

c) For what other uses could this watch have been made??
Anyone who can supply information about the aspects mentioned above is invited to share his knowledge with the editors.

4  'Clockwise' or 'anti clockwise' on a twelve-hour dial.

back to text. Literatuur:


1) Schat-kamer ofte konst der Stuurlieden",
compiled by Klaas the Vries.
Published "By Joannes van Keulen and Zoonen" 1777.

2) Gietermaker, C.H. 't Vergulde licht der zeevaart, ofte konst der stuurluyden. Zijnde een volkomen and klare onderwijsinge der navigatie, bestaende in 't geen een stuurman hoognoodig behoorde te weten. 1677.

3) Morpurgo, E. Nederlandse klokken- and horlogemakers vanaf 1300. Amsterdam 1970.

4) C.A. Davids. Zeewezen and wetenschap. De wetenschap and the ontwikkeling van the navigatietechniek in Nederland tussen 1585 and 1815, De Bataafsche Leeuw, Amsterdam ISBN 9067071137

5) Maritime Digital. Collection search tool of the
 Dutch Maritieme Museums.

7) Antiquarian Horology #3 vol.30 p.546-547


- Courtesy Maritiem Museum Rotterdam.

- THF project webmaster: Fred Kats.

   Back to previous section.      


The silver moon disc 2) is - like the hand³ - fixed to the hour hand arbor and therefore revolves fully once every twelve hours.

The superimposed gilt disc 1) is not fixed to the hour hand arbor but revolves around it. (see also Fig. 10)

The silver tide table is visible through the semi circular aperture in this gilt disc. 1)

click here to enlarge

Fig. 7.
Noon in Ostende on the day when it is new moon.
The forecast of the high water times in other places can be read simultaneously.
The second, probably later hand has been removed (simulation)

When the moon is new (or full) the hour numerals on the silver tide table 2) correspond with the ones engraved opposite the locations along the semi circular aperture. (Fig. 7)

Konst der Stuerlieden (Art of Navigation).

To find out which locations are actually meant, two versions of the "Kunst der Stuurlieden" can be consulted (Lit. 1 & 2): the first "'t Vergulde Licht der Zeevaart" by Claas Hendriksz Gietermaker dated 1677, the second "Schat-kamer ofte konst der Stuurlieden" by Klaas the Vries dated 1777. They were published before and after the Nauta watch was made.

click here to enlarge

Fig. 8. Two manuals for navigating sailors. They were published before and after the Nauta watch was made.

(Maritiem Museum Rotterdam)

Both books include a "Tafel" (table), in which 16 times of high water are mentioned (every 45 minutes) and a number of "plaatsen and kusten" (places and coasts) where these high tides occur.
(Fig. 9) In addition the times of high water for al these locations on the days after a new or full moon are mentioned, namely every day 48 minutes later (See also Fig.12).

click the pages below to enlarge
    1777    1677

Page: 14-15   Page: 16
Page: 16-17   Page: 17
Page: 18-19   Page: 18
 Page: 20-21                 
 Page: 22-23                 

fig 9.  1777 en 1677 book tables.

In the table below, the high water data of the Nauta watch are compared with the relevant data in the book tables.

Click here to view the relevant North Sea map section.



High waters
as in book tables





6 uur 0 min. "Voor Breemen en Hamburg"

Bremen and Hamburg

6 ure 0 min.
"Voor Hamborgh"




6 uur 45 min.
“In de gaten van Texel”



6 uur 45 min.
“In de gaten van Texel”







9 uur 0 min. “Voor het Vlie en het Amelander Gat” 


Waterways between Vlieland and Terschelling (to Harlingen and Amsterdam), and between Terschelling and Ameland.

9 uur 0 min. “Voor't Vlie"


Vliestream (waterway to Harlingen)



9 uur 45 min. 
o.a. “Voor Calais”






12 uur 0 min. “Langs de kust van Vlaanderen”


Along the Flemish coast

altijdt 12 uren. "Aen alle Kusten van Vlaenderen"



At all Flemish coasts




1 uur 30 min. 
“Tot Vlissingen”




12 uur 45 min.
“Tot Vlissingen”







3 uur 0 min.  Amsterdam (Pampus)

Amsterdam (Pampus)




5 uur 15 min.
“Voor Dordrecht”







6 uur 0 min. "Voor ter Goes en Antwerpen"


Goes and Antwerp

6 ure 0 min.
"Voor Antwerpen"



Click here to enlarge
 fig 9a. A relevant North Sea map section.

The times indicated in the tables in the books are regional, as at the time there were no time zones. The local times in Calais and Hamburg for instance differ 32 minutes (8°). This and other factors cause the difference between the times from the "time zone times" (click here) in present-day nautical almanacs.

The changes of waterways and shallows in the past centuries also contribute to these differences.


It is clear that on this Dutch (Friesland) watch of nine locations the time of high water is indicated simultaneously, varying from Calais to Hamburg, in other words in the region of the Netherlands.

click here to enlarge

fig 10. Under-dial work.

How the watch operates is quite simple. As the silver moon disc 2) which has 58 teeth, engages with the same pinion as the superimposed moon date disc 1), which has 59 teeth, the upper disc will lag behind at every 12-hour turn by 1/59. (Fig. 10)

After two turns the hand³ will indicate the moon date a day later and the times of high water will be indicated 24/29½ hour = ˜ 49 minutes later.

Fig. 11.
The upper disc with moon date 1) revolves more slowly than the silver moon disc 2), so that the reading of high water opposite the location will gradually indicate later times.
3) Time of high water (9 o'clock) at this location (Vliestroom) when the moon is new or full.

The months and the corresponding number of days, engraved along the outer rim of the chapter ring, can serve as an aid when converting lunar days to calendar days.

A typical complication.

For the tide table to work correctly, it is strictly speaking necessary to set the hand on the right starting time when the moon is full or new, in other words when the hour numerals on the scales correspond perfectly (fig. 7) without turning the silver moon disc.

The maximum error that might occur owing to an incorrect setting of the time, is six hours4, which would result in a deviation of the tide table by ˜12 minutes (0.2 uur) (6/24 x 49 minutes).

This required resetting of the time is a typical complication of a multiple time table.
The right time set varies from location to location, for instance 12 o'clock for Ostende (Fig. 7) or 3 o'clock for Amsterdam etc. This 'norm location' could be called the 'home port' in this context.

For setting purposes there is an angle of 45° between the two "squares" on the hour hand arbor, to which the hand and the silver moon disc are attached. (Fig. 10) In this way the hand can be set in eight positions (times) which differ by 1.5 hours. This cannot really be done without the help of a qualified clockmaker.

Setting the watch for stationary use in one particular location is in fact feasible. For mobile use (onboard), however, it would be necessary to set the watch according to the local time of the 'home port' to maintain a correct time table.


The scales which turn gradually (constantly) in relation to each other provide a correct reading of the phases and  age of the moon at any moment.

Moonphase aperture (waxing crescent).
(click here for more on moonphases)

However, for predicting the next time of high water, this time must be shown by a jumping action comparable to calendar indication mechanisms in clocks, and is to remain fixed during a set period of time. Therefore, on this permanently turning scale, which does not jump, the reading is correct only once per period between two "high waters".

Because of this 'imperfection' this repeating error is largest shortly after the previous time of high water, viz. ≈25 minutes (0.4 hours) on average too early. The times of high water established when the moon is new (such as 12 o'clock for Ostend, 3 o'clock for Amsterdam etc.) have an error of ≈1 hour, so +0.5 and -0.5 hours. The accumulated error can therefore be ≈0.9 hours (0.5+0.4).


Captain's watch?

Whether this pocket watch in this case could be worn on deck at rough seas with lots of "spray" is doubtful. After all, it is not a waterproof "Omega Seamaster". A dry spot in the captain's cabin seems to be a minimum condition for longer use. The movement does not show traces of excessive corrosion as a result of wet conditions.
(Click here to view case details)

As not only the tides do cause high and low waters but also the direction of the current and its strength, accurate tidal data are extremely important for safe and successful navigation at sea.

The doubtful seaworthiness and possible inaccuracy of ≈0.9 hours, perhaps justifies the question of whether this watch was not mainly used on land (by a ship owner or a trader?) to be able to estimate the times of arrival and departure of a ship in various ports (?). Moreover it was for example important for the market price of fish to know whether the fishermen were be able to go out fishing once or twice a day. (Lit.7)

click here to enlarge

Fig. 12.
"to find the time of high water,
on the basis of the age of the moon."

It is likely that the captain or the mate himself calculated the times of high water with the 'Konst Der Stuurlieden' (Fig. 12), or used the tables in this book. His data on currents and tides of the Dutch waters, improved by his own observations were possibly also onboard?

Digital assistant. (PDA)

Interesting about this "Captain's or Ship owner's watch" is that Jacobus Nauta made it possible to read the times of high water at nine different locations in the Dutch waters at the same time with a very simple construction.

In this way a number of data in "Konst der Stuurlieden" was directly available to the user; something that in this age is realised by a Digital Assistant. This watch shows that this was done mechanically almost 300 years ago.

November 2007, The Horological Foundation.

This page is subject to ongoing revisions.