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

This page is under construction.
A Royal

Severijn Oosterwijck Haghe met privilege

 Table of contents:

Type Ctrl+F to find any word on this page

Huygens' authorities.
    Collections and Exhibitions
    Who was Severijn Oosterwijck.
General observations.
     The Inspection    
Unique Features.
     Plomp's Charateristics Properties


     The velvet fronted brass dial plate.
     Signature Plate
Chapter Ring
Coster Hands
Dial Latch
Pendulum Holdfast

(Type Ctrl+F to find any word on this page)


Severijn Oosterwijck’s, earliest known, spring-driven pendulum clock was first published by clockmaker-restorer Paul Shrouder Hon.FBHI, ("A Mantle Clock", Horological Journal, BHI, Sept. 2008). He recorded his restoration and alluded to its history. I recognised its real significance and contacted the BHI. Subsequently, at Mr. Shrouder’s workshop, I met the owner with his rare Hague-Clock (NL.Haagseklok).

First I established there were no commercial interests to serve, only horological and historical ones. I was told that, by family tradition, cited in Wills, and by descent, this little Hague clock has been in their family since its gift to their ancestor, with a Knighthood, from Charles the Second on His Accession to the Crown in June 1660, in gratitude for financial support in exile during the long interregnum (1649-1660).  

No antiquarian could want for a more tantalizing provenance, nor a more dynamic period in the early pendulum history. I was not disappointed. It is one of the earliest known Hague pendulum clocks, and a 'Striker'.  

Fig. 1 (click to enlarge)
Oosterwijck's Aristocratic Haagse Klok
(view high res picture)

First Impressions [Fig.2].

Some works of art, also clocks, have the power to hold the viewer; this is one such. At first sight, the regularity and quality of the movement and the dial presented features and components I had not seen in a Hague clock. I also observed the evidence of an ancient accident involving both the case and its movement.

Its workmanship is outstanding, superior to Coster’s pendulums, notwithstanding the latter's superb watches and table clocks  in the balance-era. Now I better understand why, in March 1662, Alexander Bruce (Earl of Kincardine), then Christiaan Huygens (Monsieur Zulichem)  in 1663, each chose Oosterwijck to make their Longitude ‘sea-clocks’, (inherently flawed by pendulum control in first instance), then to add Huygens’ too complex (also flawed) weight-remontoir in 1664.

Oosterwijck’s 'Haagseklok' or 'Haagseklokje' ("little Hague clock") deserves the fullest appreciation. Its privileged owners would remain anonymous, yet know more of  their heritage and its context; the restorer would publish via the BHI, a technical audience not versed in Huygens; and I would bring this rare Hague clock into the antiquarian fold, via the Dutch Horological Foundation website, (, with meaningful images also vital dimensions to promote new research. Knowing that I cannot succeed equally, I offer my findings, (PART I, "HOROLOGIUM"), also my historical perspectives and new hypotheses, (PART II, "OSCILLATORIUM").

Fig. O2V
King Charles the Second of England (1630–1685)


Readers not familiar with Christiaan Huygens (1629-1695) and the early pendulum-era will find the Horological Foundation website invaluable. It includes a Compilation on the Hague Contract of September 3rd, 1657, between young John Fromanteel (1638-1692) and established Hague clockmaker Salomon Coster (1623-1659) the pendulum patentee appointed by Huygens as inventor. One long misread, still misunderstood clause alludes to a ‘secreet’, that cannot be the pendulum or crutch which John already had seen and made. It taxes us yet. This clock may have great relevance.

Collections and Exhibitions.
G.B.Shaw once wrote, “if I had had more time, I should have written a shorter letter!” But gone are the days when Drummond Robertson could review two neglected Coster clocks in just two paragraphs, (“Robertson J.D., "The Evolution of Clockwork”, Chap.VI, pp.76-81). Scholarship has moved on apace, with specialist articles, new reference works; magnificent Dutch exhibitions, “Octrooi op de Tijd” (Museum Boerhaave, 1979), “Huygens Legacy” (Paleis Het Loo, Apeldoorn, 2004); also great private collections. Mr H.M. Vehmeyer’s astonishing catalogue, Hans van den Ende’s museum at Edam -- I was privileged to attend its opening, I stayed three days! My own study of Hague clocks was helped by many, especially by the late Willem Hana and the Dutch restorer Mr. L.H.J. ‘Berry’ van Lieshout; many rely on his wide knowledge and unique archives.

 Didactic Scholarship.
Professor Dr. Ir. Reinier Plomp has long been popularising Hague clocks, by numerous erudite articles and his standard reference work, “Spring-driven Dutch pendulum clocks, 1657-1710 (Interbook International BV, Schiedam, 1979). For the privilege extended to me, I presented copies of Dr.Plomp's book both to the owner also to Paul Shrouder. Lately, Dr.Plomp has identified “The Prototypes of Hague Clocks and Pendules Religieuse”, (Antiquarian Horology, June 2007); he also defined their significant characteristics, “The Earliest Dutch and French Pendulum Clocks”, (HF website, Op. Cit.). His matrix is based on 25 clocks, to determine both craft lineages also chronologies (‘D1’, etc., for Dutch pendulums; ‘F1’, etc., for French pendulums). Dr.Plomp's academic and horological credentials place him at the fore, currently he is about to publish his long awaited tome on "Pendules Religieuses", the French derivatives of Coster's pendulum clocks. Here, I follow his line; he may have to revise his chronology.

 Dr.Plomp's characteristics’ are not the only ones to be observed in a Hague clock. Several are too general to be useful; others too infrequent to compare yet still important. Among these is, or should be, the unidentified ‘secreet’’ construction– although many have made diverse attributions.

 In a paper for the Dutch  Horological Foundation, I cited research by Berry van Lieshout and myself, (HF website, Op.Cit. 2005). Implicitly, on Mayday 1658, a 'secreet' was to be shared between Fromanteel’s and Coster’s clocks. I pointed out, ‘secreet’ is not a Dutch word at all; “its etymon seems entirely English; if so what then?” I believe that overlooked etymon confirms a conspicuous and even important linguistic clue. Might that contractual secret also be found here in young Severijn Oosterwijck’s Royal Hague clock?

Who was Severijn Oosterwijck .
Notable authorities, (Robertson, Morpurgo, Edwardes, Dobson, and Plomp), have cited his life and work. Berry Van Lieshout records, “Severijn Oosterwijck was born before 1637, son of Adam Oosterwijck clockmaker of Middleburg (employer of Pieter Visbagh from 1649, after completing apprenticeship with Salomon Coster); He died around 1694. In 1657, he married a Sara Jans van Dueren at Rotterdam”; [KP. Did he then know Mr Simon Douw, ingenious Clockmaker of Rotterdam? Already he was a fine clockmaker, see Huygens Legacy, exhibit 07, balance controlled and dated to 1655, but curiously bears a Hague signature]. Whereas, “Severijn is first mentioned in the Hague in 1658, when his first son Adam was born; he registered there in 1659, first renting near the Spui [river]; in 1660 he bought ‘De Drie Vergulde Mollen‘ and then took Pieter de Roo as apprentice”.  In 1662, he made two copies of Alexander Bruce's (Earl of Kincardine) first Longitude pendulum clock fitted with a double-fork F-crutch, shown to Huygens in London in 1661, both tested by Captain Holmes. He then constructed Huygens’ Longitude design and, by August 1664, he also had incorporated Huygens' ingenious but flawed weight-remontoir, for which Huygens would obtain a Dutch patent; but chided by Sir Robert Moray, that priority for the remontoir was [Ahasuerus] Fromanteel’s, he assigned his English Patent (3rd March 1665) to the Royal Society. Robert Hooke had scoffed; he understood that a pendulum is inappropriate for any sea-clock; as is the weight remontoir; but inherent defects do not reflect on Oosterwijck’s craftsmanship. In 1664 Lord Brouncker, first President of the Royal Society, had one of his seconds’ regulators; Sir Robert Moray in Maastricht had one too. “Severijn had four, sons, all were clockmakers; in 1687/8, he and Adam (1658-1695) petitioned the Hague Magistrates for a Clockmakers’ Guild; and upon its incorporation Severijn became first Master. Around 1690 he made a year spring movement for Jean Brisson’s monumental case (modelled on the more lavish Breghtel-van den Bergh case of 1665-1670, now in the V&A London). Later, with third son Jacobus (1662-1711), he adapted it into a musical clock, which they signed jointly”. [KP. It is one of Holland’s horological icons, (see P.C.Spaans Collection, Lot 421, Christie's Amsterdam, 19/12/07). I saw it with Eugene Stender in 1976].  

Any clock by this particular maker is of interest, and for several reasons; his part in the birth of the Dutch pendulum clock; his abilities as a craftsman; also his early part in experimental maritime navigation to determine longitude by pendulum time-keeping (see Appendix Three). The subject clock ticks the first two boxes, hence this Royal patronage; Bruce and Christiaan Huygens himself ticked the third box. His name never disappoints. 


The Inspection.
Paul Shrouder took down, measured parts, and counted teeth, as I made notes and shot images (of variable quality). The minutia of our record is essential towards a better understand of “the evolution of clockwork”. For researchers, the dimensions and wheel-counts are recorded at Appendix One; Appendix Two touches on conservation needed to preserve this unique case; Appendix Three offers new lines for study, and proposes an "open-research" Web project.

Antiquarian catalogues rarely reveal the extent of any ‘restorations'. For the benefit of researchers, my examination found Oosterwijck’s movement to be very original. The few exceptions are; the mainspring is replaced; the original four-spoke escape-wheel's collet and pinion are newly made; the original four-spoke contrate wheel now has a new collet; the original pendulum-rod has a new bob, it now also has a suspension hook - (the original pulley mount sadly removed); door glass is replaced. A restorer’s defacing scribbles infer he moved the hammer and adjusted its clapper, the strike-lever is untouched One function has been lost, a strange ‘cam’ on the barrel arbor and vacant pivot holes offer cryptic clues of a unique feature, (see 'Wind-Me' - Part.II, "Oscillatorium").

Fig. 2 (click to enlarge)
Oosterwijck's Aristocratic Haagse Klok
(view high res picture)

Oosterwijck's spring clock has the typical strike-gates and central countwheel, with early internal bell, but also unique features like the pendulum holdfast (Fig.2).

The Unique Features:
Among Hague clocks, Oosterwijck’s movement has features that I believe to be unique, namely;

1 Octagonal pillars.
2 Fromanteel type strap-potence.
3 Watch-stop work hidden under ratchet work
4 Its four-spoke escape and contrate crossings are exceptional. One lost feature, an Up-Down mechanism? certainly would be unique among Hague clocks.
5 Dial: a folding pendulum-holdfast; an obelisk bell-stand; and rare sector cut-out. [His early 'Lieberge' clock has a date-sector, (Sothebys Amsterdam 21-02-1995, lot 324), as does his later clock, (Plomp R, Op.Cit. nrs.84)].
6 Box-case: hardwood carcass and backboard is constructed from an expensive show-wood, used in the solid. My initial recognition of  ‘Kingwood’  (Dalbergia Cearencis) has gained expert support.

Having early strike work and a split-barrel* (one spring serving multiple trains) also mark it out as special. [*Tandem-barrel misleads, being two or more barrels serving just one train].

Notwithstanding the cited changes, this is a most original Hague clock and whilst not virgo intacta, undoubtedly it is of huge academic significance. But how does it square with Dr Plomp's earliest characteristic properties?

Plomp’s Characteristic Properties, vis a vis Oosterwijck’s Clock.

Windows P1. Earliest simple box, no side windows, but sound-holes in base and side.
 Door Frame P2. Early plain section; Hinge plates are set into 45 deg. angled mortices, beneath veneers.
 Aperture P3. No backplate aperture for escapewheel, (higher escape-wheel).
 Pillar Shape P4.

Uniquely Octagonal!

 Holes P5. 2 Steady holes but for a  strap-potence, (higher verge).
 Key P6. Winding key also locks the door, (no special key).
 Chapter Ring P7. Pinned to dial , (not riveted).

To pre-empt conjecture, although the subject clock possesses some of the very earliest characteristics, it probably falls outside the first experimental year, 1657. Significant negatives include: Dial plate is not made of Iron (unlike extant Coster timepiece, provisionally ‘D1’, Plomp 34); Dial plate is not fixed, and the case has no rear door nor removable panel, (unlike Coster timepiece ‘D2’, Plomp 35); Hinges are not combined for the dial and door (unlike Coster ‘D3’, also Fromanteel's English box-cases); Spandrels were never fitted.

Oosterwijck’s 30-hour spring clock with hour strike is most directly comparable with Salomon Coster’s two known striking clocks; ‘D8’ and 'D10" in Plomp's chronology; <Compare Oblique> (Plomp R, Op.Cit., Nr.38; “Prototypes”, Op.Cit., Figs.6,7,8). Oosterwijck's clock even has ’Coster hands’, rarely seen with another maker.  

Coster's split-barrels have a significantly larger diameter, and have larger centre pinions, than Oosterwijck's Royal clock, but all have a similar duration around 30-hours, (see Appendix One, Table 4). Other than a tortoiseshell frame, the box case of Coster D8, is remarkably similar to Oosterwijck's case, also having sound-holes. Probably these two cases were made by the same furniture maker.

Whereas, Coster's next striker 'D10', and also Oosterwijck’s next, 'D9' in Plomp's chronology, (Huygens Legacy, Op.Cit., Nr.11), both possess more decorative hands, signature plates, cases, mouldings, and movements; all of which signify later dates. Internally, both D9 and D10 are laid out like the earlier Coster D8, and the subject Royal Haagseklok. All are 30-hours, on 4-wheel trains; but dispel prejudice, Huygens himself preferred fewer wheels for less friction.

Coster 'D8' is presently regarded as being the first Hague clock to have strike. However, I shall advance a new hypothesis - that Oosterwijck's Royal clock is the predecessor and is also the pattern for Coster's also Visbagh's striking clocks. I shall also propose an 'open research' project - to accept and collate trains and other technical data that may eventually reveal train evolutions, chronologies, and test new hypotheses. (see Appendix Three).


The velvet dial plate

Oosterwijck’s velvet on brass dial plate, (21.2 x 16.5 cm), is smaller than all but one of Coster’s - a sure indicator of an early Hague clock. His superb dial even creates the illusion of a Coster clock, it even has the earliest steel-tipped minute hand and lobed hour hand, set upon later velvet of Indian-red. Swivel pins (L) allow the dial to swing outwards. (Fig.3).

Fig. 3 (click to enlarge)
Oosterwijck's Dial
(view high res picture)

The dial retains Coster’s decorative brass winder-collet, that preserves the velvet. The cannon opening is oddly irregular, but I could not find evidence for separate alarm work like Coster 'D5', nor integral like Coster' D8' added retrospectively. The typical engraved Lambrequin signature plate covers the dial's access hole, needed to restart the pendulum, but the old red velvet covering is not cut out. While it  is not the original velvet, it is fragile  so was not removed to examine the plate. 

 Co-operation between these two earliest Dutch pendulum makers is writ large. Even Coster’s immediate successors, Frenchman Claude Pascal and first Dutch apprentice Pieter Visbagh, rarely replicated Coster hands but, instead, each introduced decorative piercings.

Early Dutch pendulum movements are rarely signed, so only its original signature cartouche denies Salomon Coster the credit for Severijn Oosterwijck’s rare clock and outstanding quality.

Just as the iconic square pillars were initially adopted, the velvet dials saved the great expense of engraving or matting, and also speeded up the fabrication of these new clocks to meet demand. Nevertheless, velvet became the reigning fashion for decades in Holland, Flanders and France, but not elsewhere in Europe nor in England.

Purple velvet, from the Purpura lapillus mollusc, was probably used originally; like another Royal patron’s baroque gilded console-clock by Johannes Van Ceulen, its case is now attributed to Daniel Marot, (see Turpin A, “A table for Queen Mary’s Water Gallery at Hampton Court”, fig.14, p.11 p.14, Apollo Magazine, Jan.,1999). Matching silk replicates the original appearance, while also protecting the rare original velvet, much worn and faded where not protected by the mounts to its most unusual Limewood dial in the French style.

Johannes van Ceulen.

Fig 05v

Purple velvet on a cartel by
Joh. van Ceulen and Daniel Marot.

Black, or scarlet, velvet cannot be ruled out, but I reason that the Prince Charles Stuart (soon to be crowned King Charles II) already possessed his executed father's French tastes and still held to the doctrine of "Divine right of Kings”. Therefore, an immediate visual impact, showing His personal clock’s Royal status, would have been irresistable.

Signature Plate. [Fig.4]
The typical wrought and gilded brass lambrequin plate, now with a pinned repair to one hanger, is finely engraved (not etched), and bears a full signature, also Huygens’ license;  Seueryn Oosterwijck Haghe met privilege. Note the phonetic spelling of 'Severijn'.

Severyn Oosterwijck  
met privilege

Early  signature plates hang on wire loops, over a dial access hole for restarting the pendulum. The hole is present, but the later velvet is uncut.

Fig. 4 (click to enlarge)
The signature plate.

Fig. 4a (click to enlarge)
Velvet is uncut.

Though it bears no date, here I do not doubt that this is the original signature plate, probably even sourced from Coster’s engraver; unlike the repoussé plate of his next clock, (Plomp, “Prototypes”, Op.Cit. D9). [Van Lieshout privately suggested that Huygens should not have granted “met privilege” while Coster lived, he died in December 1659. It is a telling point, since the assigned Octrooi was granted to Coster for a term of 21 years, so it would require both to consent. Even Pascal's earliest Hague clocks do not bear the legend.
But, also in 1657, Huygens did grant a second privilege, to Jan van Call in
Nijmegen, (Morpurgo E., "Nederlandse klokken en horlogemakers vanaf 1300", p.30, Scheltma & Holkema, 1970, Amsterdam); Berry himself privately records a dated clock by Pieter Visbagh, bearing the  legend “Met privilege 1659". Even Dr Plomp's chronology puts Oosterwijck 'D9', bearing the legend, before Coster's similar striker 'D10'. Was Coster, perhaps, incapacitated? Did Huygens anticipate his decease  by granting the coveted privilege to other Dutch clockmakers?]

Herein, therefore, I shall interpret and assess the subject Hague clock only against the evidence of extant comparables and their more authoritative dating. However, in the light of discovery of Oosterwijck's clock and my new evidence, I anticipate lively contributions on this point.

Chapter Ring [Fig.05].
The gilt brass ring is typically narrow (2.0 cm), of small diameter, (14.3 > 10.3 cm). It is finely engraved and very well finished. Among early Hague clocks, its design represents what I putatively identify as the 'Fourth-state'.*

Roman Chapters, I-XII, mark the ordinal hours. Half-hours have become stylised spring-flowers, now with the Quarters scribed within the narrow inner band; still with Coster's enclosed ordinal Minutes shown in fine Arabic ciphers, 1-60, in a wider outer band; each Minute is scored through in early manner. Seconds’ are not shown, though Huygens showed how in Horologium’,  and Philipp Treffler’s 1657-8 copy does, (see concluding Perspective 6, A Seconds’ Hiatus).

Fig. 5 (click to enlarge)
Chapter Ring
(view entire chapter ring in high resolution)

[*I observe, Coster's 'First-state' chapter-rings have every minute scored through, 'arrow-heads' mark the half-hours, there is no inner quarter-line; the  'Second-state' adds an inner line, yet without quarter marks; a 'Third-state' adds spring flower half-hour markers, still without quarter-marks; a 'Fifth-state' has quarters with arrow-heads now sprouting from base flowers.  Another variation, is where the single minutes (1-9) are not scored through, like Coster D3, D8, and D10; D3 anyway has an untypical heavy chapter-ring. However, any particular 'state' (design) might also depend upon the engraver chosen, or clients' wishes. Chapter ring styles soon proliferated as pendulum workshops sprouted across Europe- I understand Hans van den Ende is preparing a paper on this subject].

The reverse side, too, is also well finished but shows tool marks, also an indistinct cipher and a Roman XII  'scribed at the top stud. This is the original chapter ring, fixed by integral round studs pinned at the dialplate.

Coster Hands.
Oosterwijck’s ‘Coster-hands’ are finely wrought and sculpted in gilded brass. These are the original hands; despite evidence of maltreatment each retains most of their original fire-gilding, probably the steel tip of the minute hand was blued originally. 

Fig. 6 (click to enlarge)
The 'Coster' hands.

The rare moon-pierced minute hand, with early steel tipped pointer (7.05 cm), is held by a domed collet, having a collar and pin-slot The lobed hour hand (5.1 cm) is secured to the hour-cannon by two transverse pins, (Fig.6a). It is instructive to compare the subject hands directly with Coster's, which are rarely seen on another maker’s clocks.

Fig. 6a.
Transverse fixing point, to secure the
hour-hand onto the hour-cannon.

Oosterwijck’s next extant clock (Plomp D9) keeps Coster’s minute hand, but its broad hour hand has teardrop* piercing, silvered to match the silvered chapter ring in contrast to its black velvet dial. *Teardrop piercing is also found in his Lieberge clock, (Appendix Three). Interestingly, when discovered, Coster D3 possessed an untypically carved and teardrop-pierced hour hand, with a short trident tail, more reminiscent of early table clocks. Is Oosterwijck  recalling Coster's original pendulum hour hand? <Refer D3hand>

Dial Sector [Fig.7a]
On seeing the dialplate, already with its movement demounted, I noticed an unusual opening, an inverted-keyhole below XII, in the centre zone. The facing velvet is not cut out. If for display, like Oosterwijck's Lieberge clock (see Appendix Three), it would be a rare exception to early practice; I first took it to be for Huygens’ Seconds’ window, alternatively for the Weekdays and Deities. It proved a 'red herring’; when the movement was returned to its dial, the vertical motion-cock recessed flush into the sector. (Coster D8 has a canted cock, D10 has none). Had extra depth for a strike-lever been overlooked? Such oversights might be significant. I had not seen dial cut-outs for motion work in any Hague clock, except a French inspired wooden dial plate by Van Ceulen-Marot (see above)*.

Fig. 7a
Dial sector?

The dialplate also has a typical rectangular access hole, to restart the pendulum, faced by later velvet, uncut like the 'keyhole-sector' above. Tiny vestigial dial studs might be changes of mind, clockmaker's 'pentimenti'.  The unique holdfast is shown pivoted to the 'rest' position, held by a sprung geometric lock sited next to the obelisk bell stand.

Fig. 7 (click to enlarge)
Dial opens outwards on swivel-pins.
(view high res picture)

* An early ‘pendule religieuse’, by Isaac Thuret of Paris, has a similar cut out dial sector, but only through the brass false-plate which is riveted onto its fixed iron dial. [Thuret's dial is reminiscent of the earliest extant Costers; ‘D1’ having a pivoted iron dial; 'D2’ having fixed brass dial]. Note the octagonal steel work to a typical vertical hammer post, here unusually mounted on the dialplate.

Dial Latch [Fig.8]
The typical Coster pattern fabricated brass latch, having a broad flattened spring reversing on itself and turning into a single scroll foot, being riveted behind the brass dial plate. The latch has a brass slider protruding through a horizontal dial slot. Coster's many acolytes adopted it. The surprising exception is Coster 'D10' which has a brass latch but a straight spring. [Fromanteel's English box-clocks have brass latches with longer ’S shaped springs, yet are recognisably derived from Coster]

Fig. 8 (click to enlarge)
The 'Coster' Dial Latch.
(view high res picture)

Pendulum Holdfast [Fig.9]
The exceptional, robust, folding, pendulum-holdfast, pivoting in dial brackets; its lobed base forms a geometric lock against a broad spring with “T” shaped foot fixed to the dial, (with iron rivets, like the repair to signature plate). The long cranked folding arm ends in a square fork having a delicate swivel-hook, to enclose and hold the pendulum rod. All edges are chamfered, it is masterly, and also unique on a Hague clock.

Fig. 9 (click to enlarge)
Holdfast extended, swivel hook open.

Evidently, Oosterwijck foresaw that his Royal clock would be moved about, regularly. [Based on photographs, Van Lieshout suggests this device is not original, saying if any holdfast ever was fitted to this Hague clock it may have been pegged into the backboard, in one of the holes. KP. I do not share his view. The quality, colour, also the patina of this device matches the dialplate too closely].

 Why is the holdfast not seen in other Huygens-Coster spring clocks sent to Italy, Germany, France and England? It seems a practical idea, even for a seldom moved spring wall clock. I suggest it was then envisaged that, once delivered, those clocks would all live out sedentary existences.

 Whereas this pendulum clock was made for a Royal Court which, in the changing fashion of the time, still was often in progress around the realm to enjoy the hospitality of liege families. Evidently, this clock was intended to be included in every Royal baggage train.

Fig. 9a (click to enlarge)
Oosterwijck's Aristocratic Haagse Klok
(view high res picture)


The Plates [Fig.10]
The rectangular plates (11.5 x 9.4 cm) are conventional, any original gilding is now fugitive. Notches (V - \ /)  in the top edges reveal intended orientations which were ignored when being drilled, (see fig.15b). That simple oversight had no further consequences.

Fig. 9b
Notches in the plate edges.

The one-piece front plate is rebated at the left edge for the return of the drop-hammer; whereas the striking Coster frontplate (D8) has no cut-out because spacing is greater and its hammer stem is straighter and its steel clapper is itself rebated. Oosterwijck's front plate has two brass studs to locate the trapezoid bridge of motion work. [Note. Coster D8 has integral studs on the bridge itself].

 A central steel hammer stop-pin has been made redundant by the hammer move. There is a steel steady-pin to the left, possibly for a cock to a lost device (see "Wind-me").  In the lower right corner is a, now misformed, "L" shaped, brass hammer spring - to assist the gravity drop-hammer.

Fig. 10a (click to enlarge)
Frontplate obverse.
(view high res picture)

The thick brass 'L' shape hammer-spring is held by integral steady-pins being pressed into the plate, its long tail is now reshaped, curled oddly upwards, again caused by the hammer being moved. Between the central pivots at the lower edge is an internal post, now reduced.

Fig. 10b (click to enlarge)
Frontplate reverse.
(view high res picture)

Four, unique, octagonal pillars are riveted to the plate, giving 3.8cm separation; 1.5mm wider than Coster's D1-D5 plates, but exactly according with Huygens's spacing, "one and one-half inches apart", ('Horologium Oscillatorium',  p.15).

Fig. 10c (click to enlarge)
Backplate obverse.
(view high res picture)

The single back plate is undecorated and also  is unsigned, usual for all first-period Hague clocks, (compare his next clock, see Plomp D9, 'Huygens Legacy' Nr.11). I should want to planish out both of the 1970 restorer's accusatory inscriptions, this  (under the count wheel) defaces the plate.

Fig. 10d (click to enlarge)
Backplate reverse.
(view high res picture)

The back plate has a central steel stud for the count-wheel; the brass detent-spring is held by integral studs. The two small holes adjacent to the detent position are probably for the spring's first plant which Oosterwijck then changed.

Riveted inside the back plate is a long brass post of the lower potence, having dovetail jaws holding a steel wedge - to bear the vertical escape pivot.

Both the verge cock and the upper strap potence, have integral steady-pins. Individual pendulum suspension ‘cheeks’ are screwed high on the plate. Original tool marks abound, some finishes remain; Paul Shrouder has conserved these all.

 Plomp suggests all the extant Coster timepieces, all having regular plates (109/110 x 58/59 mm) and square pillars, were in fact all made by young John Fromanteel. Whereas Coster’s first striking-clock ‘D8’ required larger plates (120 x 98 mm), like Oosterwijck’s similar ‘D9’,  (117 x 95 mm); both similar to the subject plates (115 x 94 mm).  Is that mere co-incidence or had Oosterwijck the access to Coster’s workshop, therefore to young John Fromanteel, and thus to new developments in England, even to shared ‘secreet‘ construction? Suddenly, attention to every detail in this clock became paramount.

Octagonal Pillars [Fig.12]
Four octagonal brass pillars (3.8 cm ‘tween plates), are proudly riveted at the front plate and pinned at the back plate, without flourishes. Note the early strike-detent 'gates'.

Fig. 11 (click to enlarge)
Octagonal Pillars
(view more on high res picture)

These recall square pillars, used by Coster, also Ahasuerus Fromanteel (1607-1693) in England, to reduce time and costs in bringing their pendulums more quickly to ready markets. A Coster balance-wheel table-clock has hexagonal but twisted pillars, also square dial-feet, (Vehmeyer, Op.Cit, Pt.II, Chapt.2,LC7, pp.274-5).  Square pillars are uncommon, but several earlier ones are known,  (Vehmeyer, Op.Cit, Pt.I, pp.140-161; by Johann Sayler of Ulm, (G23, G24, G25); by Andreas Raeb of Hamburg, (G29); and lastly in 1651, by Jacub Gierke of Vilnius, (G33). On the evidence of Coster's table-clock (LC7), I am inclined to give him the credit for deciding on time saving square pillars for Hague pendulum clocks

Whereas Coster acolytes around Europe all adopted round pillars, here, Oosterwijck seems to take his lead from ‘Fromanteel-Coster’ square pillars. Curiously, Oosterwijck's ‘Huygens-Thuret’ style long-pendulum regulator has heavy square pillars. (Sotheby's New York, 13/10/04, "Time Museum", Part.4, Vol.1, lot 518); now in a Dutch private collection. <Refer_Regulator>

Early Ahasuerus Fromanteels often have octagonal dial-feet, like his roller-cage clock at the Museum of the History of Science, Oxford. Whereas, Coster's Hague clock dial feet are mostly round; except ‘D8’ having hexagonal dial feet, and 'D5' having square dial feet like the pillars to its movement and separate alarm. However, I do not know of another Hague clock having octagonal pillars; octagonal steelwork is more common. <Refer_Thuret>

Oosterwijck’s octagonal pillars probably gave him subtle savings in costs and time, to fulfil his Royal client’s order for a new Huygens' pendulum clock, but having a visibly superior movement and the additional novelty of hour striking - but no alarm. King Charles left the Hague in May 1660, for England and His Coronation, we may now presume, with His new striking Hague clock in His luggage.

The Motionwork [Fig.13]
The wrought brass bridge, notched with a ‘V’ matching the front plate, its trapezoid feet planted on brass studs set into the front plate and fixed by screws. (Fromanteel’s early bridges are also set on plate-studs, later ones have integral steady pins). Coster D8 has a rectangular bridge, Coster D10 has none. <Compare Frontplates>.

Fig. 13 (click to enlarge)
The motion work.

The up-stands to the wide bridge are slotted for the reverse-minute wheel and the strike lever. The cannon is plug-riveted into the bridge. The wrought brass vertical cock of the reverse minute wheel, with integral steady-pins is fixed by a dome-screw, it recesses flush into the dial sector. The frontplate motion work gearing (32_32/6_72) is identical to all the extant Coster’s, but Oosterwijck's ten strike pins and countwheel pinion are unlike.

 A hooked barb to a original long, steel, strike lever is tripped hourly by a pin on the centre minute-wheel. The weighted, strike-lever (Coster D8 is thinner steel, D10 is brass with a return spring) goes diagonally across the front-plate to the lower of two plain arbors, to pivot the arbor of the scroll pinwheel gates, all crafted in steel.

Shown here the hammer arbor is wrongly planted, its brass clapper has been moved along its own integral dovetail shaft to extend its reach, the long tail of the hammer spring has also been reshaped. (Coster D8 and D10 have flat steel hammer shafts, with steel stop and assist springs  mounted above and below the hammer pivot).

This form of simple hour strike, now regarded as English striking, is also the model for two striking clocks signed by Salomon Coster, D8 also D10; also for Pieter Visbagh's D18. While their details vary, their general layouts are the same.


Oosterwijck’s movement breaks new ground, firstly in having a new ‘experimental’ split-going-barrel driving its two trains. I have yet to see this on device any earlier European going train, where the former practice was to reserve barrels for striking work, and to always apply a fusee for the going.

Fig. 14 (click to enlarge)
The four-wheel trains.
(view high res picture)

Other wheels have similar evidence of marking-out by punch or radial scoring, by which I  was able to show that the abused contrate is the original. Domed brass collets to the centre-wheel, crutch, warning-wheel and the last strike wheel are all originals. Dutch trains are usually ‘scribed, Gaan or ‘G’ (going) and Slag or ‘S’ (strike); here a single ‘S’ is 'scribed, oddly enough, on the unmistakable warning wheel. The rare four spoke contrate and escape wheels are exceptional in any Hague clock, (see Figures 13 and 15d); the Hague clock standard being just three spokes.

 Several teeth exhibit normal deformations and wear, (see Figure 13a), but scarred edges to the centre teeth, also the scarred pin-wheel, warning wheel and escape wheel crossings probably, are all due to the trains running-on from plate derangement in the accident cited. None detract from the Royal clock's normal functioning.


However, I dismiss these as 'age-marks' which do not detract from originality nor academic interest. Horologists will find the wheel-counts and useful dimensions at Appendix One, also two comparable Coster trains of  timepiece D3 and striker D8.


Overview: [Fig.15]
The four-wheel going-train is planted vertically above its spring-barrel, like the Coster’s timepieces and his comparable striking clocks (Plomp R, “Prototypes”, Op.Cit. p.202, Fig.8).

Fig. 15 (click to enlarge)
Going train.
(view high res picture)

Oosterwijck's centre-pinion is driven from the rear going wheel, like Coster's trains, but his is now detached from its centre-wheel which is planted at the front plate and fixed to the centre arbor with a double-domed brass collet; Déjŕ vu, Fromanteel!

Fig. 15a (click to enlarge)
Centre wheel.

The tapered centre arbor, with centre wheel displaced from the rear pinion, is a new feature for Hague clocks. The deep filed (not turned) relief suggests an ad hoc revision by the clockmaker to correct  too small a clearance for the split-barrel strike-wheel (S1).

 Probably, in that distant accident the front pivot of the verge was damaged, re-cut and given an extended bushing to accommodate the shorter, now re-cut, front pivot (see Fig.15b). The open screw holes are for individual cocks to Huygens' cheeks (to eradicate any circular error and to hold the silk suspension).

Fig. 15c (click to enlarge)
Verge frontplate bearing.

Both the contrate collet and the escape collet are replacements; the escape pinion of five replaces a badly worn original. The potences, verge-cock also two suspension-cocks are all the originals. The deformed four-spoke contrate wheel was painstakingly reshaped by Mr.Shrouder. Though abused, and recently repaired, nevertheless it is the original. My proof of originality being, its hand-cut teeth, individually marked out by punch, and having double-ringed rim.

Fig. 15d (click to enlarge)
The double ringed contrate wheel with hand-cut theet marked out by punch.

Accidental damage also caused the escape wheel’s thick toothed-rim to be re-cut. Extensive scarring to the rare original four-spoke crossing and potence are clearly evident in Figure 15d.

Fig. 15b (click to enlarge)
The escape wheels' four spoke crossing.

Again I stress, in any Hague clock, having four spokes to the upper going train is exceptional, the ‘standard’ is just three. Dr Plomp has privately confirmed that he recalls one other, by Claude Pascal, (Vehmeyer H.M., “Clocks, Their Origin and Development 1320-1880, Vol.I. Pt II.,Chap.2, LC14, pp.288-9).

 Conversely, in Ahasuerus Fromanteel’s early pendulum oeuvre three spokes are infrequent, although the original strike train of his most famous pre-pendulum Chef d'Oeuvre, (a Solar-Zodiac-Musical clock, made in 1649 for Dudley Palmer of Grey's Inn, the most famous clock of its time), includes three-spoke wheels on high domed collets.  [Original cross-beat and remontoir trains were the subject of early conversion to a pivoted verge pendulum with remarkably small, uncrossed, contrate (7/28) and escape (8/11) wheels].

 We may assume, therefore, that John Fromanteel’s use of three spoke going wheels, in Holland, was copying Coster’s pattern, under contract. Oosterwijck’s independent use of four-spokes, in his upper going train, together with his novel centre arbor layout, and strap potence, might all infer an unknown and earlier connection to the Fromanteels in London, or a knowledge of their clocks.

Early Hague clocks did evolve very quickly; a minor consequence of early hand cutting is that the upper wheels are often larger, with fewer teeth, having coarser pitch, than later engine-cut equivalents. Trained watchmakers, like Coster, had their own standards - but the early conversion of Fromanteel's Zodiac clock has 7/28 and 8/11.

Another consequence of their rapid evolution is even Coster’s going trains are not uniform. (see Appendix One, Tables 4, 5, also Appendix Three, matrix). Oosterwijck’s train relates directly to Coster’s timepiece ‘D3’ and striking clock ‘D8’; their escape wheels are 5/27 and 5/29, Severijn’s is 5/27; Coster’s contrates are 5/64 and 5/60, Severijn’s 5/60. Such similarities are not by chance; they even hint at Severijn’s going train being a transition, ie. between Coster ‘D3’ and ‘D8’. It is compelling evidence, but of what; fraternal co-operation or industrial espionage? I prefer the first.

Future researchers may give greater weight to the evidence of wheel-counts and train designs, to determine origins and chronology. Here I give credit to Dr.Jeff Darken and the late John Hooper who recorded all wheel trains in their book, “English 30 Hour Clocks”, (Penita Books, 1997). Those typical English trains have upper pinions of 6 or 7, even 9, whereas Renaissance pinions of 5 are typically found in early Hague clocks. Berry van Lieshout has long recorded the details of their trains, in his unique database, notes, images, also AutoCad layouts; but he also adds a caveat, "the demands of any train may well give identical and yet independent solutions".

 Nevertheless, it is possible to deduce much from wheel-counts, also wheel forms. Arbors and collets too can reveal lineages. Coster ‘D1’, ‘D2’, ‘D3’ , ‘D4’  also 'D8' all share integral steel collets at the front of the centre arbor; whereas 'D5' does not; only the barrel-arbors of  'D3' and ‘D4’ extend beyond the back plate, only 'D4' has a turned flourish (like Reijnaert's pillars). Coster ‘D1’ and ‘D3’, have square collets, behind the contrate; Coster ‘D4’ has a shapely French collet at the front. And Coster D3, alone, has a dovetailed barrel-cap; also thinner and unrounded ends to its pillars; and when discovered an unique hour hand with an open teardrop trident pointer - not out of place on a Renaissance table clock. There is an abundance of similar unpublished data with Van Lieshout and others, therefore I propose an Open-Research project under aegis of the Horological Foundation from data inputs by owners, curators or restorers of early clocks, (see Appendix Three and 'open-research' matrix).

Astronomy's Pendulum.
Huygens was fully aware that the real kernel of his intellectual property, which he assigned in Coster’s June 16th 1657 Patent, was his crutched verge, that regulates his clock directly from astronomy’s freely suspended pendulum, loosely held within the jaws of his patent crutch. It was a simple construction, yet a profound insight. He deserved his glory.

 Yet Huygens’ way was very unlike Simon Douw’s spring-remontoir with a beam balance, (cross-beat?), which he and Coster had failed to quash in their ill-motivated 1658 litigation. Huygens’ way differed radically from, and was far superior to, Galileo’s pivoted pendulums (1635 and 1642); better too than Fromanteel’s pivoted pendulum. However, by 1664, even Huygens had acknowledged Fromanteel’s “new way of long-pendulum, which then founded the long English ascendancy. [Fromanteel’s equation clock at Belmont reveals an early conversion from the original verge long-pendulum circa 1662, firstly to cross-beat pallets on a saw-wheel (Burgi’s third cross-beat) circa 1664, before its third incarnation as, so-called, original anchor' circa 1670 - now with an even later overhead regulation].

 Oosterwijck’s original Huygens-type pendulum rod, in Mr.Shrouder’s images (HJ, Op.Cit. p.382, Photo2), then still retained a small, flattened, bracket plate at the top. He made a new brass bob, but he also replaced that tiny bracket with a hook; more appropriate for later Hague clocks and French variants. Yet that tiny vestigial bracket, certainly, had once held a small shrouded pulley, for the silk-suspension. Few Hague clocks now retain it, now known only to Huygens' scholars and Hague clock specialists, I have not seen it in Pendules Religieuses. But it can be seen in Huygens’ original 1657 Patent design, (ie. Figs.I, II, Horologium Oscillatorium,1673); also in Perelli's 1770 drawing of J.P.Treffler’s 1657/8 copy of Coster’s clock (Plomp. Op.Cit. Fig.9); also in Isaac Thuret's regulator (Vehmeyer, Op. Cit., Vol.II., Part II., Chap.5. pp.810-811, F20). This seemingly inconsequential part, actually serves an empirical purpose, it spreads and flattens the suspension chords to squarely meet the face of the cheeks. Lack of it gives a twisted strand prone to banking. [Its ultimate expression surely was in Huygens' triangular pendulum]. Happily, when I first saw the clock, Paul had kept the removed part so, eventually, it might be reconstructed.

Oosterwijck uses Huygens’ patented crutched-verge. The accident-reduced steel verge is pinned to its brass crutch through the collar of a domed brass collet unlike the steel block of Fromanteel's pivoted pendulum. Refaced pallets are set in boxes on the verge; similar pallets are to be seen in other Hague clocks, so this appears to be the original verge. The long stem of the thick brass crutch is bent outward and forked, to accept the round iron pendulum rod.

Fig. 16 (click to enlarge)
Hyugens' crutched verge.

The unusual open fork has two transverse pin-holes, unique in my experience. Their purpose must be to contain the pendulum, like Coster’s loop, but probably easier for making, also for attaching the simple rod pendulum - without the later flat section.

The crutch was the true intellectual core of Huygens' pendulum invention, patented by Coster,  and used here by Oosterwijck with both their consents, as evidenced by the legend, 'met privilege'. [Van Lieshout suggests these may be English book-pallets, inferring their replacement post accident? KP. I suggest any competent repairer would not repair a broken pivot, add a new form of pallets, only to then extend the front bushing to accept his shorter verge].

The Verge-Cock. 
The wrought brass, single-foot, verge cock has integral steady-pins, fixed by a single screw. It is planted at left, opposite the internal strap potence. It is the original cock, but its form, size and position is unlike any Coster clock, (or Hanet, Pascal or Visbagh); with the significant exception of Coster's first striker D8 which also is on the left, but has a trefoil foot. (Coster's timepiece alarum D5 has a double-foot, back cock like Fromanteel's, but it appears to be a French reconstruction). The subject clock's recent conservation and corrected verge alignment has left an extra pivot hole from a poor repair when broken in the cited accident.

Fig. 17 (click to enlarge)
The verge cock with integral steady-pins..

Like Hague clocks of all periods, this verge pivot has not the benefit of Fromanteel’s earliest steel-shim, roller-cage, or steel knife-edge; being his trademark attention to details towards perfection.

End of this section, click here to continue.


(Table of contents continued)
The movement proper.
     The Plates
Octagonal Pillars
The Motionwork
The four-wheel trains.
The watch (going) part.
     An Astronomer’s Pendulum.
The Verge-Cock
Suspension Cheeks
Strap Potence

The strike part.
     Strike ‘Gates’
     The Fly
     Count wheel, Detent and Spring
Bell on Dial
Split (Going and Strike) Barrel  (the strike part??
Ratchet-work  (the strike part??
Stop work  (the strike part??
Concealment  (the strike part??
The ‘Wind-Me’ Option?  (the strike part??
The unique box case
First Assessments.

Dimensions and construction

PART II  OSCILLATORIUM  Perspectives & Hypotheses
Concluding Perspectives.
     Coster’s Other Contracts?
Makers of Coster Striking Clocks?
Fromanteel Connections?
Hidden ‘Secreet’ Constructions?
Personal Associations?
A Seconds’ Hiatus?
Claims to Priority.

Sebastian Whitestone's discovery.
     Reflecting on the 1657 contract

General Conservation Issues & Valuation?
     Oosterwijck's Box Case    
Open research projects.
     Early pendulum clocks matrix
     Significant early makers
     Comparable Wheel Trains
Oosterwijck Bibliography.

Picture Gallery  (the author's original untouched pictures)

Reviewed by:
Keith Piggott

Back to end of previous section.

Suspension Cheeks.
uygens' famous ‘Cheeks’ were not part of his design, he claimed for Christmas Day 1656, (Plomp R, Op.Cit. Fig.4). However, he quickly identified circular-error (periods of oscillations varying, due to the changing amplitudes of fixed-radius pendulums) that Galileo and Wendelin had observed, but had not resolved. Seeking to cure that significant defect by empiric means, early in 1657, Huygens first arrived at cheeks to change the radii of a suspended pendulum with changing amplitudes, to minimise circular error.

Huygens' privilege also extended to Oosterwijck using pre-cycloid cheeks, evidently without any allegation by Coster or Huygens of plagiarism or litigation: which indicates to me their full cooperation in this English Royal commission.

Oosterwijck's two laminae (Cheeks) are set higher on the back plate than Coster's. Each separate cheek has a round cock (foot), fixed by elongated ball-head screws, joined at their contact point by a single screw clamping the thread. These original unmodified cheeks are an incredible rarity. [Van Lieshout identified the baroque Van Ceulen repeating movement as another, no less rare in that later period].

Fig. 18 (click to enlarge)
The pre-cycloid suspension cheeks.

Each cheek has two pinholes at its lower end, I had not seen that before although Coster ‘D1’ does have single larger holes at the top. Paul Shrouder neatly resolved these pinholes, by threading the silk suspension so as wear occurs new thread may be pulled through.

 Oosterwijck’s suspension cheeks already display his empiric form but are, technically, still ‘pre-cycloid', like Huygens' original 1657 timepiece (see 'DŘ1' below) and, oddly, still depicted in Horologium Oscillatorium (Part I, Fig.II); a true cycloid form is also shown, (Pt.I, p.21, Fig.2).  [Note. Huygens' original and patented Seconds' clock had his suspension cheeks, but he abandoned cheeks in subsequent 'OP' design, intended to reduce its half-second pendulum arc to eliminate circular error (Horologium, 1658, Op.Cit.). By 1660, he had determined its ideal shape was a Cycloid. Probably, some cheeks were reshaped retrospectively; or even replaced (like D8) but his geometrician’s, incremental evolutes, proof was only published in 1673; he credited Christopher Wren as the first to determine the cycloid's arc-length, ('Horologium Oscillatorium', Op.Cit. Part II, Proposition xv, and Part III, Propositions VII, IX)].

Strap Potence. [Fig.19]
Dr.Plomp regards the simple verge-cock as an English tradition (Plomp R, “Earliest”, Op.Cit. Conclusions, pf.3). He says nothing about the typical Dutch escapement 'potence block' (riveted or screwed) used since Gerardus Vibrandi's time - derived from earlier German watches. Pertinently, Coster was firstly a trained watchmaker, who later took to making table clocks before Huygens fortuitously offered him his pendulum Patent and rights.

 Oosterwijck's four-wheel train’s high escape wheel and extra long verge does not follow Dutch practice. He has resolved his higher geometry with a strap-potence, of wrought brass, mounted on the going side, its foot fixed by integral steady-pins and a screw; its curved pivot-arm bends under the verge to the escape wheel’s top pivot. This part, too, appears to follow an entirely English tradition.

 I believe this strap form is unique among Hague clocks. His novel strap potence has an external screw, which enables the escape wheel to be detached without needing to separate the plates. It seems much closer to the English form adopted by the Fromanteels for their horizontal verge with pivoted-pendulum, ie. suggesting a contemporary English pendulum existed. Déjŕ vu, Fromanteel!

Fig. 19 (click to enlarge)
The cast strap potence.

Post Potence.
The lower potence is a long, tapering, brass post, riveted into the back plate, having inferior dovetail jaws holding a steel wedge to bear the escape-arbor's pivot. [Coster's lower strap-potences all have similar dovetails, D1’ and ‘D4’ still have steel pivot wedges, ‘D3’ now has a brass rivet through the dovetail].

Fig. 19a (click to enlarge)
Escape Wheel and 'English' Potences.

Oosterwijck's might be mistaken for English potences; that might prove significant. Among Hague clocks this is an isolated even unique, appearance of the upper strap-potence Ahasuerus Fromanteel favoured, (Lee, Ronald A, “The First Twelve Years of the English Pendulum Clock” 1658-1670, Exhibits 3,13,30).

Whereas all ‘Costers’ and his acolytes' clocks have the typical Dutch riveted escapement block, with a lower strap-potence, Fromanteel's distinctive potences had originated from the English vernacular lantern clock. [Note. as happens in research, I have learned of a second Oosterwijck movement with a strap-potence, and a fusee, (exceptional in early examples). After 1680 some Hague clocks were fitted with fusees, as, by then, several English makers, Joseph Norris, the Fromanteel brothers, and Steven Tracy had set-up in Amsterdam and Rotterdam; they first adopted Hague going-barrels but later introduced fusees as standard, even with Cheeks - "when in Rome"].

 However, Coster's form is also seen in the rebuilt dated Fromanteel, (Plomp R, Op.Cit. Fig.22), yet Simon Bartram’s pendulum watch-clock now has Fromanteel’s strap-potence, (see “Horological Master-works”, Exhibit 6, AHS, Oxford, 2003). However, the authors cite evidence of a “Dutch type top stud [potence] as illustrated by Huygens in Horologium Oscillatorium which carried the crown wheel and front verge pivots". A caveat to those who would impose their own agenda.

 Among Huygens' many sketches and diagrams, only one depicts a strap-form top potence, (see Horologium, Fig.1, Sept.1658). Uniquely intended for Huygens' verge-pendulum, was that strap form known to him from an English model, or from Oosterwijck’s extant clock? The latter might suggest that his design was only prepared after seeing the subject Royal clock; the former implies that Fromanteel’s pendulum, in fact, was contemporary with Coster’s, or earlier?

 When Oosterwijck made his Royal striking-clock it was ‘state of the art’, and approved by Huygens, certainly not a pirated copy. His next striking clock (Plomp D9) has similar cheeks but these (now) mimic the true cycloid shape. (see Huygens’ Legacy, Op.Cit. nr.11, p.35); the authors date that lavishly decorated clock, “c.1660”. Here I stress, it is exceptional for any extant Hague clocks to have original also unmodified cheeks. So many cheeks have been modified, improved or even added, mostly during commercial restorations. When discovered, even Coster D8 had Fromanteel's imperfect pivoted verge-pendulum escapement; Dutch pride has required its present Huygens' escapement to be reconstructed. [Conversely, superb English turntable clocks by East, Matcham, Ebsworth, all recently found in Europe, had all been converted to Huygens'  improved escapement]. <Refer Ebsworth>


Huygens had always disclaimed striking systems as being already well known and not part of his invention, (Horologium, 1657, p.15). Here I can find no evidence for any case mounted alarm work, unlike Coster 'D5', nor an integral alarm such as was added to the plates of Coster 'D8'' by a different yet contemporary hand. But, this is one of the first Hague clocks to have striking, if not the first!

 Like Coster’s extant striking clocks, Oosterwijck’s clock strikes the full hours only, not the half-hours; now called ‘English striking’. Like Coster’s, his 4-wheel strike-train runs diagonally across the plates, (see Fig.13 above); Dr.Plomp suggests that Coster 'D10' is “actually a timepiece with a striking train added”, (Plomp R, "Prototypes", Op.Cit. p.202, Fig.8). Curiously, for that decorated, therefore later clock, D10's plates (109 x 84 mm) are smaller than the earlier D8 plates (120 x 98 mm), and are also smaller than the subject Oosterwijck's plates (115 x 94 mm). It needs considering.

 Is there an anomaly here? It would seem so, generally smaller is earlier. Only the plates of Pieter Visbagh’s first known Hague clock, circa 1659/1660, are smaller ('D18’ at 100 x 78 mm). Might Coster 'D10' be using early plates, set aside then used later? Or is it new evidence of sparing expensive materials? We may expect anomalies. Here I suggest, if Oosterwijck's subject clock had lost its signature plate, then its dimensions, plain gate arbors and steelwork would now see it lauded, indisputably, as ‘earliest’  in Coster's striking clock chronology. [One Johannes Van Ceulen clock suffered just such a misattribution].

 Strike ‘Gates’.
Hague clock strike-gates, (or warning and pinwheel detents), are relics of their Renaissance table-clock antecedents. A long weighted steel strike-lever goes, from its hooked barb at the minute-wheel, down and across the front-plate to the lower of two plain arbors bearing the original scroll gates crafted in steel, mounted across the plates.

 Oosterwijck's gates are of scroll form, on simple round arbors, and are comparable to Coster's - but being on plain arbors are earlier. Among comparables, Coster's 'D8' has more central reversed 'C' scrolls, set upon decoratively turned arbors, with plain locking detent and steel spring - now with replaced fancy detent and spring, like subsequent striking clocks by Oosterwijck, Hanet, Pascal, and Visbagh. (Plomp, “Pendulums”, Op.Cit. nr. 38). Coster 'D10' has even more ornate gates, and bolder baluster arbors, also an untypical (if original) rudimentary steel spring to a stubby angular locking detent. Details differ, but layouts comply.

Fig. 20 (click to enlarge)
The strike 'Gates'.
(view high res picture)

Evidently, in their pendulum clocks, the English makers omitted these strike gates. Even Claude Pascal’s first gates are rudimentary; but subsequently, Dutch and French gates soon became most elaborate, later becoming vestigial, finally discontinued after 1700. 

The Fly. [Fig.21] 
A heavy, cast-brass, lozenge (Rhombus) section fly is held by a narrow leaf-spring, slotted across its edges at the rear pivot. Its arbor has a 5-leaf pinion. This heavy fly's inertia, also momentum, must be considerably more than any thin-vaned fly.

Fig. 21 (click to enlarge)
The heavy cast brass fly.

This same lozenge fly is also found in the seventh Coster ‘D10’.  Coster ‘D8’ has an unreduced profile with a wider open section, to further reduce the fly rotation speed with its 6-leaf pinion? Coster fly-chronology seems reversed but Dr.van Grimbergen, director of the Museum of the Dutch Clock, Zaandam, suggests that Coster 'D8' had originally possessed a 5-leaf pinion. Was its consequently slower fly then a lozenge? Van Lieshout suggests the subject fly-pinion, of 5 off 48 teeth, is unusual; he also suggests its heavy lozenge fly was probably to counter a stronger, second, mainspring post-accident. KP. In view of the clock's  long English provenance, why does its Rhombus fly exactly match Coster's lozenge on ‘D10’, and be so similar to Coster 'D8' too? I regard these facts as clear evidence of an uninterrupted single-path evolution].

Count wheel, Detent and Spring. [Fig.22] 
The long front plate strike-lever, with a large barbed drop, resembles English lantern clock practice. Like early Fromanteel also Knibbs, the lever is directly tripped at the minute-wheel, not from the reverse-minute wheel.

 The large central count-wheel (7.2 cm diam.) is set on the back plate, mounted onto a central steel stud secured by a circular brass key-plate. It has a flat profile, it is un-numbered and ringed, being controlled by a simple steel detent having a looped brass spring. Vacant holes below the detent were possibly for a trial or prototype spring that obstructed the operation.

Fig. 22 (click to enlarge)
Count wheel, detend and spring.

A thick round brass spring, is fixed to the back plate, held by integral steady pins, its reversed tail tensions the spur of the vertical locking-detent; like Coster D8. Whereas, D10 has much smaller detent and spring components. <Compare Backplates>

Fig. 22b (click to enlarge)
Riveted to the reverse of the count wheel is a large (5.0 cm) four-spoke wheel with 78 teeth, driven by a small brass pinion of 10* leaves set on a squared pinwheel arbor,
(view high res picture)

Riveted to the reverse of the large count wheel is a four-spoke driving wheel of 5.0cm diameter. having the standard 78 teeth, but driven by a small brass pinion fitted onto the squared end of the pinwheel arbor and having only 10* leaves (see Figure 22a). The pin-wheel too has only 10* steel pins. The warning wheel has two brass pins, one pin being re-sited by 18mm, probably due to the cited accident. I suggest the use of tens is evidence towards an earlier chronology.

*Evidence for chronology? 
Coster ‘D8’ and ‘D10” each have 12 pins and 12-leaf pinions, not the Royal clock's 10' pins and 10-leaf pinion; otherwise all three have identical gearings.

Therefore the change from 10s, henceforth to the new standard using 12s, again points to Oosterwijck's Royal striking clock as being earlier in Dutch chronology than Coster's D8. It is not the derivative I first assumed.

This new evidence, infers that Hague clocks having 12 pins also 12 leafs in the ubiquitous new standard, are actually all derivatives of Oosterwijck's new pattern.

Anomalies such as these, I put it no higher, do make the case for a well supported open-research project to assemble and collate the physical data for custodians and researchers to access, (See Appendix Three, open-research matrix).

 Hammer. [Fig.23] 
The brass drop-hammer, with steel striker, is dovetailed to a long steel stem, which pivots along the plate like early horizontal table clocks. The hammer arbor is now mis-planted in secondary pivots, (Figure 23), requiring the brass clapper to be extended along its dovetail stem, to clear the frontplate cut-out.  Note the original stop-pin, and former shape of the hammer-spring. At first we took these vacant pivots to be for a half-hour passing strike, but that would not require the unique barrel-cam probably associated with vacant pivots, also screw holes and posts for cocks, (see 'Wind-me'). Fig.24 shows the hammer in correct pivots.

Fig. 23 (click to enlarge)
The hammer now wrongly planted.

The Royal clock's nearest comparables, Coster 'D8', has the same general layout and very similar proud rivets to the four pillars. Its diagonal strike lever is of thinner steel, its hammer is all steel having a thick flat-stem to a rebated pear shape clapper -needing no front plate cut-out, and also pivots along the plate from right. Its long diagonal hammer spring is set high on the left with a stubby stop-spring below. Coster D10 is very similar but with a brass strike-lever. Whereas the winder squares of both D8 and D10 have unusual cross-pins at the front plate. <Compare Frontplates>

[Vertical hammer-posts first appeared in 1659-1660, probably introduced by Hanet or Pascal, and soon became the Franco-Dutch standard, even having a short lived English manifestation with Fromanteel and his acolytes, including the famous Samuel Knibb and renowned nephew Joseph Knibb, even Tompion].

Dial Mounted Bell. [Fig.24]
The undisturbed bell stand with its original high-domed heavy bell, is mounted on the rear of the dial plate, on a beautifully crafted obelisk, with the dome facing out like the two known striking clocks which bear Salomon Coster's name.

Fig. 24 (click to enlarge)
Bell on dial.

Split-barrel (Going and Strike).

he single advantage of the “going-barrel”, over the fusee system, is uninterrupted power to its going-train on rewinding; it needs no maintaining power but its varying force is uncompensated. Therefore, historically, it was solely used for subsidiary trains, not for timekeeping. The ancient spring-barrel evolved to drive two trains, I give it nomenclature of ‘Split-Barrel’, or pedantically, Split-Subsidiary-Barrel’.  Certainly, this did not first appear during the mid-1650's, coincident with Coster's new 'Haagseklok'.

Fig. 25  The split-barrel

Jost Burgi (1552-1632) of Kassel, (later Prague), is now credited with inventing the first known 'split-barrel', driving separate Quarter and Hour trains in his famous series of Globes, circa 1582.

fig. 25a (click to enlarge)
Split-barrel in Burgi's (Anton Eisenhoit)
Armillarsphäre, 1585.
Nordiska Museet Stockholm.

Jobst Burgi 1552-1632

Within Burgi's Armillarsphäre, for Anton Eisenhoit circa 1585, is the 'split-barrel' he had developed to drive two trains, striking Hours and Quarters. (Acknowledgement to Nordiska Museet, Stockholm).

 Subsequently, for decades, German clock makers used the split-barrel exclusively for that purpose; never for going trains; the Fusee alone reigned supreme in their dominant City Guilds; In 1657/8, Philip Treffler of Augsburg even added a fusee to Coster's early going-barrel, as did Bruce's sea-clocks (see 'Hollar' 1667).

 Exactly when, where and who adapted Burgi’s split-striking-barrel for Going and Striking is not known. Significantly, the going-barrel, also its variant the  split-going-barrel, were only possible assuming  an erroneous premise that Galileo’s new pendulum, used in Huygens’ way, had no need of regulated power. All Dutch and French makers relied on that; but German, and English makers ultimately, did not. Fromanteel was one of the first to use the going-barrel, then one of the first to re-divide his trains, and barring Treffler was probably first to re-introduce the fusee (with his spring maintaining power; being derived from his 1649 spring-remontoir).

It is almost written in stone, Salomon Coster made the first Dutch pendulum clock, also the first to have strike work. His strikers, D8’ and ‘D10’, share Oosterwijck’s split-barrel. It became ubiquitous in Holland, French makers adopted it too but soon re-adopted multi-barrel formats, like the English had before them, and both before the Dutch. Dr.Plomp has the "tandem-barrel" as Coster’s contribution (Plomp R, “Prototypes”, Op.Cit. p.202). That hypothesis would depend on the nature of his supposed contribution.

Having studied Oosterwijck's work, I now suggest that this seemingly mundane device is actually fundamental to any understanding of ‘Hague-clocks’, and very probably the Hague Contract of September 3rd,1657, between Coster and Fromanteel, or Fromanteel and Coster. In the light of Oosterwijck's construction, Coster's priority must now be re-examined. I shall examine and consider the device as, potentially, the secret device, and I will review the circumstantial evidence to discover whose intellectual property it might be, and whose it is not, citing 'priorities' with all consequences. (See concluding Perspective, Nr.4)

Pendulum Applications.
Oosterwijck’s split-barrel (diameter 4.3 cm, 2.5 cm long), drives separate first-wheels for his going and strike trains. Its weaker new mainspring (the clock’s third) has thirteen turns but just six are useable. That gives it a duration of 30 hours; or much longer without strike. (G1 has 72 teeth, S1 has 80 teeth).

 Oosterwijck’s new split-barrel has a deceptively simple appearance, showing bold ratchet-work at the front. But, it is much more complex than is apparent. Here I describe its parts, so that its underlying characteristics and unique intellectual property is made clear.

Having by this time already formed a considered view, that both of Coster’s extant striking clocks are later than the subject clock, and on the basis of features I shall disclose, this may well be the earliest "split-going-barrel" yet observed in any pendulum clock.

The going-wheel,
(G1, 72 teeth, 4.94 cm diameter), is riveted onto the rear of the barrel. It drives the centre pinion (6 leaves) with a shaped arbor having a centre-wheel (70 teeth) at the front; unlike any of Coster's movements. The barrel's rearmost cap has been left roughly filed.

The strike-wheel,
(S1, 80 teeth, 4.95 cm diameter), forms a front cap, being pinned onto the squared arbor. A now internal set-up ratchet is affixed to the wheel, unlike Coster’s timepieces with ratchets set on the front plate. (Coster 'G5', early Pascal timepieces and later Hague clocks set the ratchet-work onto the back-plate). Oosterwijck’s 'split-barrel' appears to be conventional. It is not!

[NL.Palrad, Ger.Sperrad, Fr.Rochel,]. The purpose of ratchet-work is to set up a minimum spring tension, to get a more equal mid-range force. Further, if a strike train is fitted, to always reserve sufficient spring-power to operate the strike to the full duration. Comparable Coster strikers, D8 and D10, have ratchets on the front cap, behind the front plate. (Both have cross-pinned winder squares - which at present I cannot fully explain without dismantling - although one Van Ceulen has  its winder squares pinned to round barrel arbors].

Fig. 26 (click to enlarge)
Ratchet work
(view high res picture)

The solid and seemingly ordinary strike-wheel (S1) forms the front barrel-cap. Set proud, it mounts a thick circumferential brass-spring to a steel click, engaging with the domed steel ratchet (3.44cm, 21 teeth), secured by a stubby steel collar, having an unusual 'spur-cam', at the lower of the two stepped squares below the winder.  This appears to be a typical split-barrel, having a typical set-up ratchet and click pegged onto the barrel, like Coster D8 and D10, but having no stop work.  It is not!

Stop work.
[NL.Opwindbegrenzing, Ger.Stellung, Fr.Arretage,]. The purpose of stop-work is to limit spring tension and to prevent over-winding, that might bind the spring or damage its attachments.  (Britten, FJ.  “Watch and Clock- makers’ Hand- book Dictionary and Guide”,  SPON, 1938, p.415). If ratchet work were not fitted, the stop alone might also 'fix' a range of travel for the going, and maintain minimum power for strike if fitted.

Stop work is not invariably found in all Hague clocks, although the earliest extant Coster-Fromanteel ‘D1’ dated 1657 has it. Oddly enough, it is most uncommon in French  derivatives. Yet this unremarkable device warrants a special survey, it appears in Hague and London clocks c.1657, but has a long German ancestry.

German Origins.
top-wheels, early ones often shaped like a 'Maltese Cross', form an integral part of an ancient German craft device to modify the force delivered by a spring barrel. the ‘Stackfreed’. That combines a rotating barrel-cam, being limited by a stop-wheel, and having a sprung roller-follower that manages spring forces delivered to the train. (Baillie G.H., “Watches”, Chap.5, Plate XIV, pp.84-85; also Bassermann-Jordan, “Book of Old Clocks and Watches”, 4th Ed. p.180, Fig.40b).

Much later, Hans Keining adopts it to regulate his verge pallet depths and incidence angles with spring power, (Lloyd H.A., “Old Clocks”, p.52 Fig.9). But Dr.Hans Von Bertele ascribes that clock to 1595, and he attributes the original ‘adaptation’ to Keining’s master Jost Burgi (again), first observed in his ‘aequating’  book-clock, circa 1591. (Bertele, Dr.H von, “Precision Timekeeping in the pre-Huygens Era”, Horological Journal, BHI, Dec.1953).

 Sometime later, either in the Hague or possibly in London, now without the cam and its follower, the stopwheel is itself re-adapted to a new purpose; not to manage power but to cut-off top-end torque and to prevent spring derangements at clients’ hands; a desirable feature in any clock, but especially in any destined for export.

When present, early Dutch stop-work usually takes the typical form, with a visible stop wheel set onto the barrel-cap, being turned by a pin or pinion on its arbor (see Plomp R, "Pendulums" Op. Cit. Fig.55). Coster's successor, Pieter Visbagh, continued to use it.

A curious variant may be seen in clocks by Christiaen Reijnaert, at Leiden, Coster's last apprentice who in 1660 was taken over by Pieter Visbagh, later to be his brother-in-law. Reijnaert's stop wheel, often occupies the full radius of large split-barrels. In this case 34+2 teeth engage a cross-drilled barrel-arbor to a massive spring-barrel, as an integral pinion of report, allowing eight turns for longer duration and more power. Reijnaert's known examples have 'twin-click' ratchet work, sited at the opposite end of his split-barrel.

Stopwork on Reijnaert Haagse Klok.

Fig. 26a
Stopwork by "Christiaen Reijnaert Leijdae Fecit".

 [To digress, I am only surprised whenever this form of arbor-pinion survives intact, but several have. Relevantly, Coster's only alarum timepiece, 'D5,' with an external bell, probably had a form of 'Reijnaert' stopwork, under Coster's aegis, to maximise its duration; although it failed and was replaced with a plain arbor, that did not replace the integral pinion nor stop-wheel. Happily, it corroborates its Coster workshop pedigree, (see below, "Coster's Clockmakers")].

Hidden Stp Work
Oosterwijck’s clock has already presented surprises; did it surprise Coster too? On removing its front ring-collar, with the spur-cam, its combined barrel-cap strike-wheel and ratchet, a hidden inner-cap is revealed. His split-barrel's concealed inner-cap is mounted with diminutive 'watch' stop-work, being unique in my experience, at least in Hague clocks. But why hidden in this fashion? Its construction is perplexing, it raises challenging questions.

Fig. 27(click to enlarge)
Split barrel, ratchet and 'hidden stopwork'

This stacked, hidden, arrangement is doubly unique, as, whenever ratchet work and stop work are fitted, they 'always' occupy the opposite ends of their barrel, or the ratchet work moves onto either of the plates. I should like to know of another such example, in any reader's experience.

 Whatever Oosterwijck’s intent, it was not for convenience of making. Might it be a purloined secret, or have been constructed around Mayday 1658 when the famous Contract terminated? Might it even be the actual model being there described?

A diminutive stop wheel, just 10.3mm, with six teeth cut (8 scribed), permits six windings to the stop-sector, being driven by a pin set in the barrel arbor, below stepped squares for the ratchet wheel and winder key; Hague practice uses little taper to the winder square.

 Riveted to the inner-cap is a thin brass spring, possibly an afterthought, which engages the stop as a simple click. Oosterwijck’s barrel bears an indistinct cipher. If original, whose? Are comparables  similarly marked?

Fig. 27a (click to enlarge)
The concealed stopwork.

[Coster's timepiece 'D3' also has a pierced winding arbor, above its rare dovetail barrel-cap, but that piercing is for the pin to secure the ratchet on the front-plate, in common with his other timepieces].

English Variants.
Few early English pendulum clocks with going or split-barrels are extant; excepting later clocks by English makers trading in Holland, (Joseph Norris, Steven Tracy, the Fromanteels, etc.). At Appendix Three I identify several examples, but whenever striking is added twin going-barrels are more typical of English work, (Fromanteels, East, Bartram, etc.).

England's hiatus of split-barrels might well reflect their abject failure as timekeepers, especially when used for Fromanteel's pivoted pendulums, which probably accounts for his much earlier readoption of the fusee and his evident preference for weight clocks - as did Huygens. Both are indicative of his 'test-discard' approach, striving for better accuracy. Whereas, the Dutch makers stayed with this basic '1657' pattern even into the 1690s, consequently Holland fell far behind England, (see 'Concluding Perspectives', nr. 4).

Watch-stop work, like Oosterwijck’s but lacking his spring, is seen on the front barrel-cap of Fromanteel’s dated timepiece (c.1658); Simon Bartram’s watch-clock has two similar, but on the back plate, for his separate spring-barrels for going and strike, (c.1659). Both these have the new five-wheel trains, for longer duration, probably a Fromanteel innovation. Oosterwijck’s subject stop work relates more closely to these than to Coster’s. Have these any relevance to dating Oosterwijck’s clock, or to the Contractual ‘secreet’? Possibly. This should be investigated and didactic arguments developed. (see, 'Secreet Constructions').

But who invented, or first adapted, simple stop work? And how did Oosterwijck, Coster, also Ahasuerus Fromanteel in London come to it? Is John Fromanteel the link for a London origin? That invention's history is vital to our understanding, (see Concluding Perspectives, 4). This technical aspect has never been properly addressed.

 Taking a new account of this unique simple stop work, and its split-going-barrel, admittedly first observed in the subject Oosterwijck clock, we should now seek out comparable split-going-barrels  also stop-work in pre-Contract clocks. Publication might well bring other unknown early clocks back into the antiquarian domain, or, as already appears to be the case, cause owners and custodians to revisit their charges to inspect them with new keenness.

Wind-Me Mechanism? 
The curious much worn ‘spur-cam’, pinned to the barrel arbor at the first stepped square, sets the most intractable puzzle. Paul Shrouder and I considered its possible function. This barrel-cam appears to relate to central pivots below the barrel, (now wrongly occupied by the hammer), also relating to a reduced internal steel post, between the front plate pivots. The evidence for the hammer being mis-planted; a hammer stop-pin at its original pivot; the clapper now ‘bumped’ along its stem dovetail to extend its reach; the angle of the front plate relief is for a more upright hammer; the long ‘L’ spring is now awkwardly bent upwards, over the hammer-spur; whereas formerly it extended along the plate to beneath the hammer spur. Correcting the hammer position would clear those secondary pivots for the lost feature. We considered a passing strike at half-hours, but the well worn barrel-cam suggests otherwise.

 The stop work of this 30-hour striking clock allows six-windings of the spring, unwinding once every five hours on average, but at irregular rates according to the hours being struck. Such an irregularity precludes any feature that marks time, but would not preclude the barrel spur tripping a detent to show the state of winding also called “up-down", often flagged on dials by "Wind Me".

Cannot be shown on this dial; we dismissed the sector or even the centre openings. And any device like Fromanteel’s globe-moon set upon the case, at the two extant peg-holes, is not viable mechanically, (see Lee, Ronald A. Op.Cit, Pl.92-97). The back plate offers no evidence.

This is not resolved, but I proposed an hypothetical, also fully demountable solution, which might be suspended below the movement, fixed to long cocks from vacant sites on the front plate. It has one obvious drawback; such a 'dial' would only be visible by lifting the dial signature plate.

My fully demountable construction would be set on two long cocks at the front plate on vacant studs and screw holes; one set just below the winder, one above the ‘L’ spring. Investigation continues, better solutions are invited. For the record, there are no plans to 'reconstruct' this hypothetical system.


These early pendulums were popularly regarded as being the panacea of timekeeping and the zenith of scientific instruments. They were being rushed into production to meet eager new domestic and international markets. I suggest, for that reason alone, not any Calvinist or Puritan dogma, earlier fripperies were stripped away. But an emerging French taste soon adorned these imported Dutch cases, and their own derivatives, with new pediments, crestings, mouldings, and even fine metal mounts. The two suspension eyelets (nl. ophangers) are missing; later strips set across the base steady the 'Mantle Clock' - as this Royal clock recently was addressed.

Externally, Oosterwijck’s veneered ebony case looks exactly like Coster’s. (Fig.1 above). However, at just 25.4 x 21.0 cm, and 9.25 cm deep, with a dial reveal of 20.7 x 16.2 cm, actually, it is smaller than all known Costers, excepting 'D10'.  Oosterwijck's dial at 21.2 x 16.5 cm, is one of the smallest known; and, in Hague clocks, smaller infers earlier. In point of fact, might its true chronology antedate some Coster-Fromanteel timepieces? Technically, I can see no technical reason why not.

Show Wood Carcass. Whereas all of Coster’s box cases are veneered onto deal, like the comparables and other Hague clocks, (Fromanteel uses oak), Oosterwijck’s rare box appears to be constructed from an expensive show-wood used in the solid.

Even before opening the box, the most unusual, ‘full-span’, backboard glows with its exotic hardwood carcass. Typically, Hague clock carcasses are deal, veneered inside with Indian Rosewood (Dalbergia latifolia) or Padouk (Pterocarpus spp) with Ebony veneers outside; their softwood backboards being set within thick frames formed by the sides of the box.

Fig. 28 (click to enlarge)
The solid show wood back-board.

I knew this vivid hardwood as Kingwood, (Dalbergia Cearencis); several furniture experts have since concurred; but Howard Page suggests Cocobolo–Dalbergia Retusa, sourced from South America.

Inside and out, the backboard figuring is identical. However the sides might require a more invasive examination, but where cut out for the sound holes and pendulum arc they do seem to be solid, revealing Dalbergia’s typical vivid ochre sapwood.

Probably, Oosterwijck’s show-wood carcass of solid Kingwood is unique. Recalling its Royal association, might his choice of ‘a Royal’ timber be a visual and metaphorical pun? The opportunity to make invasive tests, including timber density and species, is left to an appointed conservator who can properly investigate.

Normally, such expensive woods are only ever seen as veneers; as in Ahasuerus Fromanteel’s only slightly later three-month long-case timepiece, (ex-Rous Lench, now in the British Museum; see Appendix Three, Hollar 1667); also Samuel Knibb’s Royal longcase with Kingwood veneered on oak. (Lee, R.A., Op.Cit., Pl. 91)*.


Fig. 28a
Samuel Knibb's Royal longcase clock

[* Formerly at Windsor Castle. Ronald Lee repatriated this gem from America. It has Fromanteel's own maintaining power and his pivoted short pendulum to a verge escapement, with a rare early quarter-strike train. I shot these photographs at Mr Lee's galleries in Bruton Place, just after this magnificent Kingwood case had been beautifully restored by Mathew King. To Ronald's chagrin, it soon returned to America, unhindered by petitions or legal barring].

The Construction.
Oosterwijck's simple box has no side windows, an early feature that Coster abandoned with D10, in 1659. The dial’s single winder gives us no clue to the presence of a new striking train, but two fabric backed sound holes at the lower left corner clearly denote its striking movement within.

Fig. 29 (click to enlarge)
e fabric backed sound holes at the lower left corner denote its striking movement.

Similar sound holes are also seen on the two extant Coster striking clocks, 'D8' and 'D10', although not on the alarum timepiece, 'D5', which has its bell mounted outside, above the case.

Fig. 30 (click to enlarge)
hick veneers protrude around
 its box to exclude dust.

The clock winder key opens a sturdy internal lock at the large round hole on the right (III ) face. The lock releases the frame's flat brass staple, from the slot in the right stile. A brass slider then unlatches the pinned dial, which swings open with the movement attached, just as Coster arranged. The left stile is chamfered to clear the pivoting dial. The case does not require the furniture key of Coster's earliest spring-clocks, also Van Ceulen's very finest commissions.

With the movement exposed, its split-barrel and strike train are self evident. The thick bell stands proud below the movement on the IX side of the dial-plate, dome facing out, set on an obelisk bell-stand.

The dial plate is located into the box by two swivel-pins, held in wire staples set below the veneers with ends just visible. The upper dial pin is fixed to a flat plate riveted onto the dial. The lower dial pin-plate has integral steady-pins, and is screwed to the dial after its pin has been set into the recessed lower staple. This arrangement allows the dial and movement to be demounted as a single unit.

Visible outer faces of the box case are veneered in thick sawn ebony; the forward edges are mitred; bottom veneers overlap the side veneers, which overlap the top veneers. Thick sides, (2.5 cm including inner and outer veneers), have no inner returns but a small rebate seats the dial. The dial latch has its own small recess.

Fig. 31 (click to enlarge)
Case chopped out for the pendulum.

Thick door stiles (2.5 cm x 1.65 cm) match the box, but are marginally larger because protruding veneer sills overlap the box on three sides to exclude dust. Coster’s cases share this feature; presumably by the same case maker, being yet more evidence of these two makers cooperating.

The lower right corner shows evidence of impact damage. The upper hinge is fatigued, I advised its immediate repair. Paul Shrouder’s chosen glass has nice irregularities; I took it as old. Thin brass hinges are set with straps invisible under the facing veneers.

Fig. 32 (click to enlarge)

Everything so far mentioned places this case in the earliest period, almost in a class of its own. An accident has left its mark, the backboard split, the hanging-eyes are lost, the door edge crushed, and its movement ‘ran-on’. Perhaps it fell, but these repairs are more recent than Huygens’ clocks that were crushed in book-piles during carriage to Paris in 1664, (Plomp, “Pendulums”, Op.Cit. p19).

Damage Control.
nfortunately, some damage and alterations detract from the case's structure and interior. The evidence supports an  accidental fall, that also damaged its movement. The back board was split, and is now filled and glued. Mismatched veneers front the dial-pins, where the heavy movement broke away from its box at this weakest point, despite an internal reinforcing staple. (Currently repaired with black stained pine). The hangers are gone, probably capstans with small rings, or one-piece eyelets, both their screw-holes are now capped off with plugs. The backboard now has two holes drilled to secure it to the wall.

Abrasions to the backboard testify to earlier repairs of the escapement. Sensitive conservation is being recommended.  

The box’s inner faces are now chopped-out for a wider pendulum arc. Whether for a King or a commoner, Severijn Oosterwijck would never have despatched his exotic Kingwood and Ebony case in this state. [Note, the inner sides of Coster D4's box are also scalloped out, but in neater ellipses].

First I assumed that some miscreant had failed to properly repair the once damaged escapement, so he simply rebated the box for his wider arc. But Paul Shrouder disabused me, saying ”the original amplitudes are never bettered by later repairs”. Berry Van Lieshout suggests that the real culprits for these chopped-out side were a stronger second spring (replaced), probably with shortened pallets and re-cut escape teeth. [My abstention from the bench has disadvantages].

Redundant wooden pegs are now set through the upper box. On reflection, these probably supported a pediment or crest, added in France just before its putative arrival in England in June 1660. Dr.Plomp identified several of Claude Pascal‘s earliest Hague clocks where his Dutch cases were similarly modified after their arrival in Paris.

[Van Lieshout suggests these peg-holes might even be evidence for a Cardan mount or gimbals, for use on board ship?  KP. I had not considered that possibility; there are no pressure marks to indicate clamping bolts, but a wide 'Knibb Phase 1' handle might, conceivably, have been added, although I still incline to a pediment or crest. One of Oosterwijck's sea-clock was damaged by a fall in December 1662, but they were timepieces; they showed Seconds'; and they had fusees. However, if Berry is right about the extant verge being an English repair, a different crutch might conceivably have been removed; But why the pinholes to the jaws of crutch? It would suggest a later also a very different clock, and one having quite another provenance. Here I do not anticipate any debate].

Beneath and right across the box, two black-stained wood strips are roughly nailed to steady the clock, reflecting its so called, “mantle clock” status.

Mortised Hinges.
Since I made my inspection, Philip Oliver was instructed to repair the fatigued top hinge. Unfortunately he took no photographs, but he reports that the hinges are not fixed square on the box or door. Each hinge-plate is deeply rebated, into 45 degree mortises; ebony veneers obscure that detail.

Fig. 32a
The hinge-plates are rebated
 into 45 degree mortises, veneers obscure that detail.

This aspect of original construction may assist researchers, and prompt investigations into other early case construction techniques. [Berry Van Lieshout informs me that he has already noted this form of hinge construction]

Oosterwijck’s abused box may now be attended to by one of today's most competent conservators, who can recover its original beauty, once befitting a King. (see Appendix Two).


Certainly, Oosterwijck’s Royal Hague clock proved more challenging than I had anticipated. On first reading Paul Shrouder's article, I recalled Huygens' price list of January 1659, to Boulliau in Paris, which omits any mention of 30-hour spring clocks with striking. That seemed to be a certain bar to earlier dating, but its Coster features suggested 1659-1660 as a safe date; which I later found to coincide with the family's extraordinary provenance.

Dr.Plomp suggests that Oosterwijck and Visbagh did not make ‘prototypes’ before Coster died in December 1659, (Plomp R, “Prototypes”, Op.Cit. p.203, pf.2). Whereas he puts Oosterwijck's ‘D9’, also bearing the legend "Met Privilege", before Coster's ‘D10’. He was, then, unaware of Oosterwijck's even earlier Royal clock. So, clearly, there remain uncertainties, both about historical events and chronologies, which the 'open-research' project will address.

At my own inspection, constructions not seen in another Hague clock sparked associations to research long in hand with Mr 'Berry' van Lieshout in Holland. We return to these subjects as new evidence presents itself, as in this instance the 'Royal Haagseklok'.

Oosterwijck's clock demonstrates the earliest use, and probably the only extant in any Hague clock, of the split-going-barrel with a hidden stop-work. Dr Plomp ascribes what he calls 'tandem-barrel' to Coster. But the ramifications of his claim led me into uncharted historical waters, and eventually to my concluding perspectives.

The greatest of horology's early innovators, Jost Burgi of Kassel (later Prague), first incorporated his original 'split-barrel' in a series of mechanical Globes, circa 1582, to drive two strike systems.

The subject clock's split-barrel with going and strike, surely, is an adaptation from Burgi, being in the parlance of that time 'a new invention’. But how did Oosterwijck and Coster come to it? With no English comparable now extant, evidentially, it would appear to be a Dutch adaptation as Dr Plomp has claimed. But was Coster or Oosterwijck its Dutch ‘originator’? Who actually made or licensed the craft ‘secreet’ that would be disclosed on Mayday 1658? And is the split-barrel part of that secret?

After long deliberations I now consider this clock to be a forerunner of its type, made no later than 1658; if the dating of its extant comparables can be relied on? Huygens' price list to Boulliau, dated 16th January 1659, omits 30-hour spring-clocks with striking, but shows 8-10 day duration weight clocks with striking. Would a King choose a 30-hour clock if an 8-10 day clock was already available? Therefore I should not be surprised if my provisional dating were moved nearer to the time of signing the Fromanteel-Coster Contract, on September 3rd 1657, because striking clocks were long established and popular demand for striking already existed at all social levels.

The famous Notarial Akte (Contract) is somewhat obtuse, therefore I did consider other scenarios. The split-going-barrel might be a reliable work-horse, but it is also a Chimera. Like its simpler going-barrel, it is not suitable for precise timekeeping - even were it to indicate Seconds', no scientist would use it. Huygens' experiments always used weight clocks, (see below, "Hiatus of Seconds"). So was the split-going-barrel a commercial gimmick? In his own Patent Application, on 9th August 1658, Simon Douw correctly identifies several defects of the Huygens-Coster pendulum, also their spring and weight systems; evidently Douw was neither a simple craftsman nor a sly plagiarist as Huygens' always alleged!

Might another clockmaker, possibly even its true originator, have understood or discovered the going and split-barrels' defects as a timekeeper, then abandoned it on a Hague negotiating table to extract a bigger prize? It is an aspect none have considered, and it certainly makes for a rather more intriguing three-cornered suit.

I became confident that Oosterwijck's Royal Hague clock was made during Coster's lifetime and, very probably, it is the first of its type now extant, antedating both Coster's D8 and D10. It also demonstrates more clearly than words, just how Fromanteel's influence extended deep into Coster's workshop, and perfectly explains the use of an English word, 'secreet', in a Dutch Contract (3 September 1657). That linguistic clue must assume a significant new role in resolving this 352-year old puzzle.

These opinions will not sit well with all authorities, and may even be regarded as controversial. I suggest we let open-research provide the evidential base for an informed debate on the evolution of components and chronology of Hague clocks. Only similarly thorough re-examinations of all the extant comparables will suffice to fairly answer my hypotheses.

For those who weigh conundrums, my concluding perspectives may confirm, entertain, challenge or even awaken interest in this period. Here, I leave behind the artefact, and embark on what I regard as its more significant historical perspectives of context, with further evidence and personal insights.

In the meantime, Mathew King has accepted  the challenge of carrying out the sensitive conservation work to the furniture of the unique case. At this stage of conservation, the distressed, later, inappropriate Indian-red velvet shall be left in place until a census of experts agrees on replacement.

The first fruits of Mathew King's conservation of the case furniture. The ebony veneer at lock-slot is repaired, well selected Kingwood veneers now conceal the chopped-out sides - but leaves the underlying evidence intact for future investigators. These Kingwood veneer covers will be thinned down before final fitting. (Image by courtesy of Mathew King).


PART II  "OSCILATORIUM Perspectives & Hypotheses

CONCLUDING PERSPECTIVES (click here to view)
  Coster’s Other Contracts?
     Makers of Coster Striking Clocks?
   Fromanteel Connections?
   Hidden ‘Secreet’ Constructions?
Personal Associations?
A Seconds’ Hiatus?
   Claims to Priority.

   Reflecting on the 1657 contract

        Oosterwijck's Box Case 

(click here)
   Significant early makers
        Comparable Wheel Trains
        Early Pendulum Clocks 'Matrices'


First I thank the private owners who kindly invited me to examine their rare clock; also, in the first instance, the British Horological Institute (BHI), for making contact with Paul Shrouder, also Jayne Hall who promoted and edited publication of this review. Paul has facilitated examination and answered my questions with helpful advice and sketches, he also advised the BHI in production of this review. Similarly, in Holland, Mr Fred Kats also made invaluable contributions, to bring this onto the Horological Foundation website.

I thank all those whom I have cited in my review; especially Drs Kees van Grimbergen, Carel Hofland and Pier van Leeuwen, respectively the director, manager and consultant of Museum van het Nederlandse Uurwerk at Zaandam, for providing records of wheel-trains also images of two Coster clocks (D3, D8), with consent to use; would that all museums held such records and were so co-operative. I am indebted too to Sothebys' Jonathan Hills for images and consents to show Coster timepiece (D3), and other clocks; also to Christie's Jamie Collingridge for images and consents to show Coster alarum (D5), and other clocks.

Dutch colleagues and friends, Hans van den Ende, Hans Kreft, Ruud Mestrom, Berry van Lieshout, Dr Reinier Plomp; also Nicole Brandt of the Nationaal Archief; each provided useful insights into other early Hague clocks, also the Netherlands' history. David Thompson, Head of Horology at the British Museum,  for images and train count of Bernard van Strijp's hybrid, Antwerp, Hague clock. Brian Walton, world pilot and solo-sailor, also old squadron colleague,  who gave insights into basic longitude finding and relevance of 4-minute dial. I must also acknowledge the late Ronald Lee and Peter Gwynn for their encouragement, also the private owners of Samuel Knibb's Royal kingwood longcase and of Van Ceulen clocks.

In anticipation, I thank Georgio Strano, Curator at the Museum of the History of Science, Florence, who has kindly undertaken to re-investigate Philip Treffler’s 1657 timepiece (DŘcopy); Alison Boyle and Rob Skitmore of the Science Museum, London, for access and for consent to publish Coster (D4); similarly to Simon Chesters-Thompson of the National Trust also Tom Boggis, the House and Collections Manager at Lyme Park, regarding the Ahasuerus Fromanteel timepiece dated 1658. Finally I commend Dr.Elspeth Knights D.Phil. for proof reading my drafts despite her unfamiliarity with horology.

Museum Boerhaave,Leiden, presently withholds train information or access to Coster timepiece (D1), on grounds that the museum is studying it themselves, with a view to publishing an article. It is to be hoped that every custodian of our heritage will now realise the advantage of 'open research' - also to themselves and their organisations too.

Image copyrights:

Views herein reproduced or accessible as digital links are the legal copyright of diverse persons or bodies, identified wherever possible; the copying of any View is not permitted.

Figure 6a is the legal copyright of Mr Paul Shrouder. Other Figures whether published in printed or digital media or any other form or distributed by any method, are claimed by the author, Keith Piggott, as his legal copyright.

However, in the interest of ‘open research’ (see Appendix Three), all Figures herein may be used and be reproduced freely without the express written consent of the copyright holder providing only that at each use of an image the user shall acknowledge the copyright holder. Figures may eventually be upgraded by the author, when the same free license shall also apply. However, any present or later upgraded Views may not be copied.

(before 1637-1690/94)

1. “Oeuvres Completes de Christiaan Huygens”, now available to all researchers at the Biblioteque Nationale Francaise website <http://gallica.bnf>.

2. Robertson, J Drummond, “The Evolution of Clockwork”, Chapt. IX, Huygens and the determination of Latitude, p.149, p.154. (Cassell 1931) .

3. Morpurgo E, “Nederlandse klokken- en horologemakers vanaf 1300” , p.94, (Scheltema & Holkema NV. 1970)

4. Plomp, R, “Spring-driven Dutch pendulum clocks 1657-1710”, pp.19, 28, 31-33, 37,38, 178-181 (Interbook International BV, Schiedam, 1979).

5. Cataloue “Horological Masterworks”, Museum of the History of Science, Oxford,
(Antiquarian Horological Society, 2003)

6. Catalogue “Huygens’ Legacy”, Paleis Het Loo, Apeldoorn (Fromanteel Ltd.2004).

Vehmeyer, HM, “Clocks Their Origin And Development 1320-1880, (Snoeck, Gent 2004).

8. Catalogue, "Octrooi op de Tijd", Museum Boerhaave, Leiden 1979.

Copyright: R.K.Piggott, February 20, 2009.

(about R. K. Piggott)                                      APPENDICES

This article is subject to ongoing revisions.