Collecting Antique Clocks A Lantern Clock Showing 'Thirds' of an hour
The earliest household clocks in England were lantern clocks. The earliest of these might have been made as far back as the year 1600, but lantern clocks before 1640 are pretty ancient things and we usually call these First Period clocks. All English lantern clocks at this period were driven by balance wheel regulation and showed time by a single hand, which indicated the hour and such subdivisions of an hour as could conveniently be indicated by this single hand. Normally the subdivisions were shown by what we call a half-hour 'marker', commonly a star or asterisk, sometimes 'floating', but usually suspended on a stalk like a balloon on a string. Further subdivisions of the hour were normally quarter hour units, achieved by splitting the half hour units into two with a further marker, the quarter-hour marker.
Because of the tiny size of these dials, it was impractical to split the quarter hours further, which means the quarter hour unit is the smallest we see on single-handed lantern clocks. Later on with the much larger dials of single-handed thirty-hour longcase clocks, and even on larger horizontal sundials, we do sometimes see this quarter unit split down further still, usually into three, which in other words would give five-minute divisions. It is therefore possible to read the time indicated by a single-handed clock to within five minutes, even on very rare occasions to within smaller units still. If the presence of five-minute divisions on a single-handed longcase was very unusual, then smaller divisions still were rare but are known. Five-minute divisions were unknown on lantern clocks.
For our purpose here we can take quarter-hour units as the smallest divisions normally shown on single-handed lantern clock dials - virtually all were single-handers but there are a few two-handers, to which these remarks do not apply. It is however perfectly possible when reading time on a lantern clock dial to visually split the quarter-hour unit, by deciding when the hand is half-way through the quarter - which would indicate a half-quarter hour or a seven-and-a-half minute unit, if you wished to know such a thing. Not only was the inner chapter ring diameter too small conveniently to split these lantern clock quarter-hour units into even smaller engraved divisions, but there are further reasons why they didn't. One is that any slight inaccuracy in the layout of those divisions by the engraver (that is inaccuracy in the spacing of the divisions), would make trying to read time more precisely than in quarter-hour units a pointless exercise, the more so when they are spaced round such a tiny diameter as an inner chapter ring, where the length-of-hand time reading distance is as little as two inches. The shorter the hand the more noticeable would be the drift through inaccuracy of spacing. Furthermore the hand itself has a certain amount of what we call' slop', in that the slight gaps (the play) between the teeth of the driving wheels will cause the hand by virtue of its own weight to ride slightly ahead of its true position when going 'downhill' (from XII to VI) and lag slightly behind its true position when going 'uphill' (between VI and XII).
Add the slop distance to any slight inaccuracy of engraving of the quarter-hour positions, and time reading is less than accurate. Add to this the fact that the clock would run fast or slow according to whether it was a cold or a hot day. Add to all this the fact that the clock was set in the first place by the owner dashing outside to read the sundial. Add time for him to try to work out in his head the equation of time value between solar and mean time. Add time for him to dash back inside and correct the clock hand. Add to that the fact that only with great patience in adding and removing lead shot from the weight canister to increase or decrease the going rate of the clock can accuracy of timekeeping be expected within a quarter of an hour a day, and it is no wonder they did not bother with units of less than a quarter hour. Then if the clock ran down during its eight-to-twelve-hour running time per winding of a normal balance-wheel lantern clock, when your back was momentarily turned, you had to start this all over again......!
At this time (our First Period), when domestic clocks were in their infancy, there were clocks (other than lantern clocks), which had two hands showing respectively hours and minutes. They were rare and were made by very specialised clockmakers only for very wealthy or learned people, and even then they were not what we would think of as accurate, but the concept of minutes was known to scientists and the well-educated. There were even a handful of lantern clocks built originally with two hands, but these are exceptionally uncommon especially in the First Period. For the normal person the smallest unit of time to be bothered with (in the first half of the seventeenth century, just as now) was the quarter-hour, but if you really wanted to split hairs, you could speak of half a quarter of an hour.
The only example I can think of which shows that this unit of time was in common use and parlance appears in the advertisement placed in the year 1658 by Ahasuerus Fromanteel, the man who had just introduced the pendulum into clockwork in England thereby enabling the longcase clock to be born which was far more accurate, so accurate that there was a point in its showing hours and minutes by two hands. Most of his advert set out the benefits of his new pendulum-controlled clocks, which 'go exact and keep equaller time than any now made' and are less 'subject to differ by change of weather'. But at the end he cannot resist advertising his new fire extinguishers, which could also be used for garden watering, crop spraying and so on: 'Engins made in a new way of his own invention for quenching of fire ...'.
The main selling point of these would that 'they are not subject to choak with mire', in other words they don't clog up with dirt. But when they do, he adds, which rather lets down his argument, 'when they are clogged with dirt, (they) may be presently cleansed without charge, 'in half a quarter of an hour's time'. They could easily be cleaned in the smallest time you can imagine and the smallest unit of time the public could call to mind in 1658 was half a quarter of an hour, or what clock enthusiasts now term a 'half-quarter'.
Those familiar with lantern clocks already know that the dials are normally marked out in quarter-hour units, and that you can, if you so wish, imagine a half-way point through the quarter where the hand might indicate a half-quarter - just as you can with a single-handed longcase dial. But there were always clockmakers who were contrary, and some clocks indicated half hours without quarter hour divisions. These were unusual except during this First Period, when examples are seen, after which quarter-hour markers were normal.
One practice which was extremely uncommon but was done just occasionally, was to engrave hatched lines inside alternate quarter hour divisions. The result was that when the clock dial was waxed and silvered, each quarter hour unit would appear alternately black and white, the idea being that the contrast would make for greater clarity. The only lantern clocks of this First Period that I know of which show this feature are two pictured by George White in his book English Lantern Clocks, one made by John Hobson who is known to have been working between 1603 and 1630, and another one by John Cattle dated 1633. I cannot recall having seen any other such as early as this, though I do know two others made much later, about 1700 which happen to have this, by then very archaic, feature. These are the only two clocks known by Christopher Carter of Galby in Leicestershire, a village smith, who was born in 1676. That is until this present clock pictured here came to light.
The extra-ordinary thing about these inner chapter ring markings on this clock is that they are marked in thirds of an hour, twenty-minute units. Why, I cannot say. Why would someone want to split hours with a marker into three rather than into the normal fractions the human mind can easily picture of halves and quarters? It is possible to split each division further with the eye, which would make it readable in units of ten minutes. Is it just a case of an engraver who was bored with the regular way of doing things? Very few clockmakers did the same thing repetitively, but they often enjoyed variety just for its own sake. So too, one assumes, did their customers. Perhaps they too wanted something that was different from the one their neighbours owned. I have no other explanation as to why, but it makes an interesting variation on a theme.
This clock is unsigned, as many were in the First Period, probably dates from the late 1630s, and has the inner chapter ring divisions boldly cross-hatched for clarity. Where was it made? Probably London, which is where most were made at this period and it offers no clues to any other location. Most of its features are clearly from Period One. The integral pillars (being a single casting incorporating the pillar, the finial and the foot all in a one-piece casting), the finial style, the foot style, the chapter ring style, ... in fact almost everything visible on this clock clearly dates from Period One. Hidden away inside are two casting marks, made by the brassfounders.
One casting mark, cast twice on the back of the chapter ring, is rather indistinct but appears to be the known monogram I over S, appearing as IS. We do not know whose name this represents but his casting mark is known on lantern clocks (mainly on chapter rings and frets) made from as early as pre-1615 by the two Harvey brothers, Robert and Thomas, the first native-born craftsmen by whom domestic clocks are known to survive, up to as late as 1650 or even 1660. In other words this casting mark spans a period of about half a century. The other casting mark is the well-known 'matchstick' man, which appears principally on the dial wheels, ratches and countwheels of some lantern clocks from as early as the 1630s to the 1680s, again a period of about half a century.
It is not unusual for both casting marks to appear on the same clock, which presumably implies that each of these two unidentified brassfounding companies specialised in different types of castings. I use the word company because this length of manufacturing time would seem to be longer in each case than the working life of one man. The IS castings are ones which carry, or might carry, engraving work - frets are sometimes engraved on all three, which form the set used on each clock, but sometimes just the front one is engraved and the side two left plain. The matchstick man appears principally on parts which have no engraving, usually mechanical parts. The two casting marks are not always entirely divisible into these two types of product, but mostly are. Does this help us narrow down the two different types of suppliers?
There must have been a reason for a clockmaker to buy castings from two different sources simultaneously. Nobody in his right mind would buy from two sources if one would do. Did one brassfounder supply castings already engraved, for the clockmaker to fit for himself, and one supply those which did not need engraving, but perhaps just finishing work? We know that some matchstick man parts, such as countwheels, sometimes have the casting mark interrupted by later cutting, which leaves the casting mark incomplete. We can also guess that some, perhaps many, clockmakers could not engrave. Spellings vary drastically, even of the maker's name as engraved on a dial, and we would expect a man to be able to spell his own name consistently. But how else do we explain clocks made by a few clockmakers we know could not write, who signed documents with their marks? If the master of the IS mark supplied engraved, decorative castings, and the master of the matchstick mark down the road supplied unengraved, mechanical ones, that would be one possible explanation of the two different sources being used simultaneously. Both casting marks appear principally, though not exclusively, on lantern clocks made in the Lothbury district of London. The implication is that these brassfounders were located in, or close to, Lothbury.
One stylistic feature of this unsigned clock is at first sight puzzling. The fret is one version (in fact a very rare version) of what is known as a 'heraldic' fret, so called because it carries a central shield, on which one might expect might be engraved the owner's coat of arms. In actual fact I think I am right in saying that no example is known of this First Period heraldic fret with such a coat of arms (the Royal Arms are known much later on a quite different fret carrying a shield supported by a lion and unicorn) but the heraldic name has stuck. This pattern of heraldic fret (there are two or three slight variations of it, clearly resulting from different brassfounders making the same pattern) was common on First Period clocks and some Second Period ones too, this second period being defined as between 1640 and 1660. But this particular variation of that fret is exceptionally unusual, and I can find it on only one clock, being one made by the eccentric village blacksmith, George Newton of Seend in Wiltshire, described by John Aubrey the diarist, in 1666 as 'an ingeniose man, who from a blacksmith turned clock maker and fiddle maker'. George tended to spell his name as Newnton -even though his father wrote Nuton and everybody else wrote Newton. The one clock of Newnton's which has this fret carries the date 1660, which is probably genuinely the date that clock was made, though oddly this clock is signed 'Newton' not 'Newnton' as his others are.
Yet these frets of the identical pattern of Newton's 1660 clock are original to this unsigned First Period clock 'showing thirds'. So how could George Newton be using in 1660 a fret, the only other example of which dates from twenty or thirty years earlier? The answer seems to be as follows. Only four lantern clocks are known by George Newton, ranging in date between about 1645 and 1677. His earliest two are eccentrically old-fashioned for their time in virtually every stylistic and constructional respect. His 1660 clock is totally different (much more 'modern') in all respects but still using old-fashioned integral pillars and a fret thirty years out of date. George's father, Richard Newton, also a blacksmith, died in 1625, leaving a will in which twenty-year-old George was left his forge. But also amongst Richard's possessions was a clock - which could hardly have been anything other than a lantern clock. Could it be that when George came to make clocks of his own, and the evidence is that he did actually make them personally, including the castings, he was using this late father's clock as a model, copying the style? This would certainly explain George's old-fashioned styling, but for whatever reason, there is no doubt that his clocks were old-fashioned at the time they were made.
The dial sheet of this clock is most spectacularly engraved with a three-dimensional design based on tulips. Tulip-engraved centres tend to appear on clocks after 1640. Period One engraving is often two-dimensional and rather primitive. This more stylish engraving suggest we are approaching or just into Period Two and puts the clock into the late 1630s. This particular clock has no alarm, hence the engraver can use the entire dial surface, as opposed to being cramped into an engraved circle by a central alarm disc. On such (alarmless) clocks the engraver is more tempted to run riot, as here into an overall composition, where he can give wider rein to his shading and hatching talents.
This article is published in enlarged form in the Horological Journal.
Copyright © 2013 Brian Loomes
click for details of home page || clocks for sale: 3 - 4 - 6 - 7 - 8 - 9 - 10 - 11
books || contact || clock how-tos || articles on clock collecting || Finding out about your antique clock: identification/valuation/appraisal
finding a clock by a particular maker
Archive (sold clocks): 1 - 2 - 3 - 4 - 5 - 6 - 7 - 8 - 9 - 10 - 11 - 12 - 13 - 14 - 15 - 16 - 17 - 18 - 19 - 20 - 21 - 22 - 23 - 24 - 25 - 26 - 27 - 28 - 29 - 30 - 31- 32 - 33 - 34 - 35 - 36 - 37 - 38 - 39 - 40 - 41 - 42 - 43 - 44 - 45 - 46 - 47 - 48 - 49 - 50 - 51 - 52 - 53 - 54 - 55 - 56 - 57 - 58
Our valuation/identification/appraisal service costs £50 (currently $100 US)
Full price list of clocks sent on request