West Sea Company


By Rod Cardoza

Edited by A. N. Stimson, Head of Navigation Section, Department of Astronomy and Navigation, National Maritime Museum, Greenwich, England.

All photographs courtesy West Sea Company unless otherwise noted
©2000 West Sea Company. All rights reserved. Reproducing any part of
this article without the expressed written consent of the author is
forbidden by law. Violators will be prosecuted.



In modern times the navigator's sextant has become widely recognized as a universal nautical symbol. Indeed the sextant and the magnetic compass were the two basic tools of navigation on the high seas for more than two centuries. Often the mariner's most prized possession was his sextant. Witness the drama evoked by the handwritten account found with this 19th century English sextant:

"This sextant was salvaged from the pilot house of the Norwegian steamship VICTORY by her master, after being sunk by gunfire from a German submarine 35 miles north of Ushant at 1:00 PM on July 6, 1917. The master, 2nd mate, nine crewmen, and a stowaway were rescued from an open boat by the USS O'Brien at 5:30 A.M. on July 7th, 20 miles west of Ushant and were later landed at St. Nazaire, France. As a token of his gratitude for the rescue the master gave this sextant to the Captain of the O'Brien."

Sextant from the SS VICTORY
(Private Collection. Ex. West Sea Co.)

It is only in the last 30 years, with the advent of inertial guidance and satellite navigation, that the demise of the sextant has been heralded. Yet, despite its obsolescence in the computer age, the simplicity, accuracy, and relatively low cost of the sextant will ensure its survival as a backup navigational tool for years to come.

In tracing the evolution of the sextant and searching for remaining examples, the first question a collector is likely to ask is "How old is it?" In attempting to date an early instrument one must consider the state of communications in the 12th through the 18th centuries. The word of new inventions and discoveries was slow to travel. For this reason it was not uncommon for an innovation to actually be "reinvented" many times over! Moreover, such was the scientific understanding of the average Renaissance man that new inventions were often looked upon as unnecessary complications of methods tried and true. Inertia to change was great. As recently as 1925 for example, the German Maritime Ministry, Deutsche Seewarte, was still certifying octants for use at sea!

Unless an instrument is specifically dated, the margin of error in dating can be as much as 20 years - more in older instruments. Bear this in mind as we explore the history of these fascinating instruments.

The earliest attempt at navigation was undoubtedly simple coastal piloting. Mariners would venture no further than the sight of land. The limitations of such navigation held trade and exploration to a minimum for thousands of years, while open water sailing was reserved for the incredibly brave or foolhardy.

The early maritime cultures of the Chinese, Phoenicians, Polynesians, and Vikings certainly made open ocean travel a reality. Yet we have no tangible proof of their having used navigational instruments.

The knowledge required of a mariner in those instrumentless times was set forth in the Sanskrit Mu'allim of 434 A.D. "He knows the course of the stars, both regular, accidental, and abnormal, of good and bad weather: he distinguishes regions of the ocean by the fish, the colour of the sea, the nature of the bottom, the birds of the mountains, and other indications. And the only aids he possesseth are his memory, helped by a pilot book, and a sounding lead or staff."

Quite apart from one another the Chinese, Egyptians, Babylonians, Greeks, and later the Arabs had discovered that they could relate their position on the earth relative to the stars. The observations by their astronomers would give birth to celestial navigation as it progressed from the 15th century onward. Simply stated, one's position on earth could be ascertained relative to a star (fixed point) by measuring the angle of elevation (altitude) of the star from the observer (apex) and the earth (horizon).

The earliest practical form of celestial navigation was probably what was known as "Running Down The Line." When a ship departed homeport the navigator knew its latitude. At sea, the navigator could also ascertain his vessel's latitude by observing the height of Polaris the north star. When it came time to return to port, the vessel was steered on a northerly or southerly track until the altitude of Polaris matched that of the homeport. Then course was altered east or west to "sail down the line" of latitude. In perfecting this method, the Arabs developed a very simple device called a Kamal which, by means of a knotted cord, indicated the height of the pole star at various latitudes. However, the obvious implications of having to take an indirect course home stirred navigators and astronomers to find a better way.

During the 15th century the Portuguese began to explore the west coast of Africa using coastal piloting. As word of Marco Polo's adventures in China and the treasures of the Orient spread, ocean travel to India and China demanded improved navigation. Prince Henry the Navigator founded a navigational school for his officers where he recruited astronomers, cartographers, mathematicians, and craftsmen to expand the science of navigation, construct navigational instruments, and draw up accurate charts.

Perhaps the earliest instrument, of which a rare few still remain, is the astrolabe or "astrolage". The first astrolabes were non-marine and were constructed in the Islamic countries of the Middle East which had absorbed and applied the remnants of Greek science and technology beginning in the 9th century A.D. The earliest surviving example dates from the 10th century. The astronomer's astrolabe was a complex and costly affair. In essence it was a mechanical computer coupled with an alidade mounted on a two-dimensional planisphere which could be rotated over a plate on which a network of azimuths and altitudes dividing the heavens were engraved. Around the rim were scribed the hours of the day, the days of the year, and the signs of the zodiac. Engraved on the backplate was even more data.


Copy of an Islamic astronomical astrolabe showing the various interchangeable plates for use with differing areas of the celestial sphere. (Offered by West Sea Co.)

A view of the reverse side, clearly showing the signs of the Zodiac. (Offered by West Sea Co.)

The sea astrolabe was an adaptation of the astronomical type. It was much simplified, as the alidade and degree scale were really all that was required for the mariner's use. It was made much heavier to keep it vertical on the rolling platform of a ship, and cut-out to reduce disturbance in the wind. The sea astrolabe was introduced about 1460, but did not see general use until the beginning of the 16th century. Its use persisted until after 1670, particularly in the fleets of the Spanish and Portuguese, where it was in evidence early into the 18th century.

For the serious nautical collector the astrolabe is perhaps the ultimate. Fewer than 100 examples are known to have survived, and of those most are in poor condition, having been retrieved from shipwrecks. Obviously one must be extremely cautious in attempting to acquire an astrolabe. A good many reproductions, primarily of the astronomical type of these instruments, were made in the Islamic countries in the late 18th and 19th centuries. More modern reproductions of the sea astrolabe have also been produced and faked.

A contemporary of the astrolabe, which may actually have been a predecessor of the sea astrolabe, was the simple quadrant. Like the sea astrolabe, the mariner's quadrant was adapted from its earlier and more complex astronomical counterpart. In design and function it was remarkably simple. It consisted of nothing more than a triangular plate, the apex of which was fitted with a plumb bob, with a pair of sighting pinholes on one edge. On the lower limb a degree scale was scribed, over which the plumb bob swung. In use, the observer merely lined up the celestial body viewed through the pinholes then "pinched" the line of the plumb bob on the scale to ascertain the reading.

Apparently the quadrant was in use well before 1450, although that was the first recorded mention of the instrument. Early types were often embellished with the appropriate landmarks at the point on the scale where their corresponding readings would fall (e.g., 39 - Lisbon).

The sea quadrant never seemed to gain much favor with the English sailors, although its use by the Dutch persisted through the 18th century. Both brass and paper covered wooden examples survive.


A 17th C. version of a simple quadrant missing its plumb bob, but with the two sighting pinholes clearly in evidence at the apex and to the right. This example represents the late evolutionary period of such instruments. It is engraved brass whereas earlier versions were of wood and it performs several functions which the earliest quadrants did not. The upper portion is engraved with a "shadow square" for use as a sundial. At the bottom it is divided to single degrees to allow for sighting the altitude of the sun or stars. Across the face of the instrument is a depiction of the celestial hemisphere showing major stars, the Equator, and Tropic of Cancer. Below these are engraved the months of the year, accounting for seasonal changes in the sun's declination and providing a Zodiacal calendar. This form of quadrant was also known as a "Gunter quadrant" after Edmund Gunter who described it in 1623. (Private collection)

The first real ancestor of the modern-day sextant as a multipurpose nautical instrument was the cross staff or Jacob's staff. It was first described in 1342 by a Jewish scholar named Levi ben Gerson. The instrument, as its predecessors, was an adaptation from an earlier astronomical surveying device and performed the same function, albeit with a higher degree of accuracy, as the Arabian Kamal. In its earliest form the cross staff consisted of the frame or "staff" and a perpendicular sliding piece called the "transom" or "cross" (hence the name). By lining up the horizon with one end of a cross and the celestial object with the other, the observer had a simple trigonometric computer. Later it evolved into a more complex instrument consisting of a frame over 30 inches long with scales engraved on all four sides and two or more transoms.

The cross staff represented a great leap forward in the art and science of navigation, since it embodied all of the functions for recording the altitudes of the sun, stars, moon, and planets, as well as terrestrial sights - a function lacking in the astrolabe and simple quadrant.

Most frequently the observer's latitude was found by "shooting" the sun, an expression popularized because of the resemblance of the instrument to the cross bows of the period. To this day the navigator's celestial observations are still referred to as "shots".


The simplest sight to be taken with a cross staff was for the observer A, to place the lower limb of the cross on the horizon B, and the upper limb on the celestial object, C. The resultant "altitude angle" BAC, was read off of the divided staff at the intersection of the cross. (From a 16th C. woodcut)

Cross staves were generally constructed of hardwood (to prevent warping), although an ivory example has been preserved. Very few have survived even though their use was in evidence early into the 19th century. Because of their simplistic design it is likely that most were discarded as useless once they were brought ashore. When the mariner took his sun shots with cross staff, the blinding glare of the sun often caused him to turn his back. By using two crosses and adjusting their angles of incidence, the observer could sometimes read the shadow cast by the sun on his instrument. But this procedure was fraught with error. The next logical step in the evolution of navigation instruments was the development of the backstaff or Davis quadrant


A rare Colonial American backstaff with an ivory maker's nameplate signed "Made by . C . Eliot . in . New . London . For . Mr. Prentice Peabody . 1768"   The arcs and the horizon vane are of indigenous American fruitwood (apple or pear) however the limbs are of ebony indicating that African trade with the colonies had been established, perhaps via Britain. (Offered by West Sea Co.)

A close-up of the Eliot backstaff nameplate. Note the unevenness of the signature indicating that Eliot was using individual letter and number stamps. This confirms the fact that this was a hand made instrument and that the scale was laboriously hand-divided. (Offered by West Sea Co.)

This ingenious device was first proposed by English captain John Davis in 1594. The name "quadrant" came from the fact that 90° (or one quarter of a circle) could be measured even though there was no full 90° arc on the instrument. The backstaff consisted of two triangular arcs, the larger of which was calibrated to 30° and the one at the apex of the instrument which was calibrated to 60°. In practice the mariner would stand with his back to the sun. While peering through the sight vane he aligned the slit in the horizon vane with the horizon and at the same time adjusted the shadow vane so as to have its shadow fall on the slit simultaneously. With this ambidextrously accomplished, the sum of the readings on both arcs indicated the sun's altitude in degrees.

Most backstaves were of English manufacture, but American, Dutch and Irish examples do exist. The backstaff gained rapid popularity after its introduction, particularly with the English and Dutch sailors. Much care was lavished on its construction in order to avoid warping and to ensure the accuracy of its scales. Early instruments were constructed of walnut, lignin vitae, and fruitwood and were characterized by in-line scales. Later models included ebony and even mahogany in their construction and featured diagonal, interpolative scales.

But backstaves were only usable when the sun was visible, limiting that form of navigation as dependent on daytime sightings. Accordingly, until the very early years of the 18th century a mariner's navigation consisted of sun shots to determine his daily latitude and dead reckoning, coupled with piloting, to estimate the longitude. Latitude, the distance north or south of the equator, is the horizontal component of the imaginary grid system encircling the earth, unaffected by the earth's rotation relative to the stars. Longitude, the distance east or west on the earth's surface, is the vertical component of these lines of position. It changes constantly, with respect to the heavens, as the earth rotates. Thus a key element in most methods of determining longitude is precise time keeping.

The onset of the 18th century saw new methods and instruments innovated for finding the elusive longitude. Among these was the lunar distance method, by which the angular "distances" of stars from the earth's moon were measured and compared with complex tables. Although never an unqualified success, the method found favor with the English, culminating in the perfection of the reflecting circle by Mayer, Borda, and Troughton toward the end of the century. Another method, longitude by change in compass variation, promised an easy solution in theory, but was not precise enough to be of any value in practice.


Rare French reflecting circle of brass with inlaid silver scale divided from 0 to 180 degrees on either side of the zero point and signed "LORIEUX, LEPETIT suc Paris No 313" on the arc, with the additional mark "SH 140" on the index arm. The original dove-tailed fitted mahogany case with brass hardware also bears the brass tag reading "Service Hydrographique Cercel Hydrographique Lorieux Lepetit suc Paris No. 140." This instrument, with folding turned hardwood handle, was obviously used in the French Hydrographic Service for charting and mapping. (Private Collection. Ex. West Sea Co.)

The search for the longitude generated some bizarre proposals. In one case Sir Kenelm Digby claimed that he had caused one of his medical patients to jump with a start, even though the two were separated by a great distance. This was accomplished by placing some specially invented "powder of sympathy" into a bucket of water and then adding a bandage taken from the patient's wound. This "fact" led to the suggestion that every ship should be equipped with a wounded dog. On shore, a diligent individual equipped with a standard pendulum clock and a powdered bandage from the dog's wound, would dip the bandage into water at the stroke of each hour causing the dog aboard the ship to yelp at the appropriate instant!

The impractical application of all these systems was becoming tragically obvious. Several instances of entire squadrons of British ships being lost due to imprecise navigation occurred in 1691, 1707, and again in 1711. These losses provided a final impetus to the British Admiralty to pass a bill "for providing a publick reward for such person or persons as shall discover the Longitude," in 1714. The amount of the reward was £20,000 - a phenomenal sum at the time - indicative of the importance placed upon perfecting an accurate means of navigating.

As a result several ingenious methods were devised. But the one that ultimately proved its worth was the 4-decade work of Yorkshire carpenter John Harrison. A number of contemporary books, articles and a television series have recently documented this fascinating story. In 1735, Harrison successfully constructed the first marine chronometer having some components of wood and weighing 125 pounds! Because of its precise timekeeping ability, the chronometer, in perfected form, was later to become an indispensable addition to nearly every ocean-going vessel afloat. After years of tough testing and sometimes undue demands on him by the Board of Longitude, Harrison finally received the reward due him nearly 40 years later. In the interim, the modern era in navigation had begun.

The increased activity in "the search for the longitude" also spurred innovative interest in other areas of navigation. In 1731 John Hadley, an English mathematician presented a paper to fellow members of the Royal Society in London describing the use of a double reflecting "octant" or quadrant. His quadrant was based on the principle of light reflection and angles of incidence described by Robert Hooke, Isaac Newton, and Edmund Halley nearly a century earlier. The principle is that when an object is seen through a double reflection its angle from the eye is twice the angle between the two reflecting surfaces. Thus Hadley's quadrant, reading to 90°, actually required an arc of only 45°, one eighth of a circle, making it an "octant." Basically the instrument consisted of a triangular wooden frame with a swinging index arm pivoted at the apex. A mirror was fixed at that point which would move with the arm. A second mirror, half of which was transparent so that the user could view the horizon, was fixed to one limb with the sight attached to the opposite limb. A precise scale, calibrated in degrees, was scribed on the arc of the bottom limb of the triangle, across which the index arm moved.


A classic mid-18th C. "Hadley Octant", signed on the inlaid ivory nameplate "Gregory, London." Known in common usage as a reflecting quadrant, or simply "quadrant," this instrument has a single piece wooden "crab claw" index arm and a diagonally engraved boxwood scale. The ivory "line of faith" from which the reading was made is clearly visible across the scale. (Private Collection. Ex. West Sea Co.)

The back of the Gregory quadrant. At this early stage, brass hardware was minimal. Even the supporting "feet" on the back of the instrument are made of wood. (Private Collection. Ex. West Sea Co.)

Quite independently of Hadley, Thomas Godfrey, a Philadelphia glazier, had also devised an improved altitude measuring device based on the same principle. The Royal Society recognized the equal contributions of both men and awarded them a prize of £200 each. Godfrey also received a prize from the Board of Longitude (of chronometer fame) for his work. However it was Hadley who generally received credit for the invention.

The improvements of the Hadley octant, or "quadrant" as it came to be known, over previous instruments was immense. Not only was it more accurate, it provided simplicity of operation, and the ability to "capture" the object being sighted for rapid, multiple sightings. The merits of the quadrant were immediately noticed by the British Admiralty and it was quickly put into commercial production. Even so, the instrument did not find popular acceptance and general use amongst traditionally minded mariners until about 1750.

The earliest Hadley quadrants, like backstaves, were constructed of walnut or other indigenous woods, with the scales being engraved on boxwood, although examples on brass do exist. With the discovery and growing importation of exotic woods such as ebony and African mahogany around 1750, the use of mahogany was quickly implemented, gradually giving way to the exclusive use of ebony.


A really exceptional mid 18th C. Hadley quadrant, with ivory maker's plate signed "Charles Woods of St. Eustacious, 1761." This monster measures 21 inches high with limbs of ebony -- one of the earliest known examples to incorporate the exotic black wood in frame construction. Three very important revolutionary and transitional features are: A flat brass index arm instead of wood, an early form of vernier scale divided on the vertical edge of the crab claw marked "Minutes," and a solid brass diagonally divided scale with a second vertical scale below, all riveted over a boxwood arc! Note that the ends of the arc still retain their characteristic ogee embellishments -- a carryover from traditional backstaff construction. (Private Collection. Photo West Sea Co.)

Around 1760 ivory was introduced as the material for scales and nameplates because of its durability, ease of engraving, and light color which provided for easier reading in dimly lit situations. About the same time brass began to replace wood as the preferred material for the index arm, a trend that would eventually culminate in the elimination of wooden components altogether


A large, 3rd quarter 18th century mahogany quadrant. This example measures 20 inches tall and is signed "*Made by JOHN GILBERT on Tower Hill LONDON* for IOHN (sic) LAW * April 11th 1772." Scale division techniques had still not been perfected to allow makers (and users) the benefit of smaller instruments up to that time. Yet numerous "state-of-the-art" advances are already in evidence. The early form vernier scale is calibrated "10 - 0 - 10." The scale on the large arc is now of 2 joined pieces of ivory instead of boxwood. The evolution from the "crab claw" to the brass-tipped wooden index arm is obvious. But note that the very early form 2-shade interchangeable filter is still being used! Dated instruments such as this are invaluable in tracking the evolutionary progress of the early makers. (Private Collection. Photo West Sea Co.)

The early quadrants were characterized by large frames, 18 to 20 inches in length, sometimes 24 inches and larger! The size of the instrument was dictated by the fact that the scales had to be calibrated by hand - the larger the instrument the easier the division of the scale. The work by Jonathan and Jeremiah Sissons and that of John Bird perfected the "art" of hand division. But it was the advent of Ramsden's dividing engine, and the incorporation of a secondary "vernier" scale on the index arm, that significantly increased reading accuracy to about one minute of arc and led to smaller, more easily handled instruments.


Large 18th C. mariner's quadrant with limbs of ebony, scales of ivory, and brass furniture. The arc is divided to 95 degrees and is signed in script "SBR" denoting the prolific makers, Spencer, Browning & Rust, of London. This characteristically large instrument has a flat brass index arm measuring over 18 inches long. Note the early form thumb screw stop, ivory vernier scale marked from 0 to 20 arc minutes, backsight, and triple sun shades indicative of manufactory soon after 1780. (Private Collection. Ex. West Sea Co.)

Early on, two colored glass shades were added to aid the mariner when taking his sights in the sun's glare or in hazy conditions. By 1780, two filters had given way to three which were interchangeable for use with the foresight or the back horizon sight known as the "backsight." This backsight was another original feature which was fitted below the horizon mirror. It enabled the observer to sight on a celestial object using the opposite horizon (over 90) in cases where the fore horizon was indistinct. Soon after 1780 the introduction of the tangential screw fine adjustment represented the last major change in the basic operation of the octants and sextants for the next 150 years!

10th. C. SEXTANT
18th. C. SEXTANT

A massive 18th C. sextant by the esteemed London maker and noted optician, Peter Dollond. This sextant measures 18 inches high and an incredible 19 inches wide. It has limbs of mahogany and a strong observer would have held it by means of the octagonally faceted ebony handle on the reverse. It has all of the characteristics of a sextant including its handle, an arc divided to 125 degrees, tangential fine adjust screw, detachable telescope, and built-in horizon shades. Yet the addition of a vernier magnifier was not yet necessary due to it huge size. Circa 1780. Note the engraved brass scale, popular with the earliest sextant makers. (Private Collection. Photo West Sea Co.)

While the "Hadley quadrant" was technically a "Hadley octant" because its arc described one eighth of a circle, the term "quadrant" is still applied to those large instruments produced prior to 1780. The difference is in name only and is used primarily to differentiate the early instruments from the more refined and generally smaller later types. Fine examples of these early quadrants are still to be found by the collector, but at premium prices. Later, and smaller, octants are much more common and are priced accordingly.

Along with the development of the octant came the more familiar sextant. A common misconception is that the sextant is a relatively recent innovation in comparison to the octant. In actuality the sextant is very nearly contemporary with the octant. Another generalized assumption is that octants were constructed of wood and sextants of brass. Some very lovely examples of ebony and ivory sextants have been preserved from the late 18th century. Likewise, in the early to mid-19th century, numerous examples of brass octants were produced.

When Nevil Maskelyne, England's future Astronomer Royal, published his lunar distance method in 1764 the immediate need and popular demand for an instrument with the capability to record celestial angles up to 120 of arc was established. The result was the limited production of the sextant, encompassing 60 degrees of actual arc (1/6 of a circle) using the Hadley principle. The reflecting circle, already mentioned, was also refined during this period. The earliest known sextant is dated 1757 and was the work of that great master of hand division, John Bird.

Perhaps the greatest boon to the production of the sextant came in the period from 1768 to 1774 during which time Jesse Ramsden was given a patent for his perfected dividing engine. With the advent of this tool, scales on instruments could be engraved swiftly and accurately at minimum expense. The size of an instrument was no longer a factor in its accuracy, allowing makers to produce smaller, easier handled, and more economical models. As such, demand increased and business flourished, particularly during the period of the Napoleonic Wars near the turn of the 19th century.

One of the greatest concerns of the nautical instrument makers throughout history has been accuracy. Because of the severe conditions and weather extremes encountered at sea, a poorly constructed instrument was apt to shrink, expand, warp, or crack rendering a false, and potentially fatal reading. Numerous materials and innovations were tried in an attempt to ensure rigidity and stability of the octant and sextant.


Double Frame Sextant by the inventor "Troughton, London" circa 1810. On the vertical limb above the scale is the engraved serial number "1086" with the word "Platina" (platinum) engraved just below. The large arc is calibrated up to 150 degrees and the unique vernier magnifier (seen as a brass disc) is backed by white porcelain to aid in observing readings. Troughton invented the double or "pillar frame" sextant in 1788, so named because it consisted of 2 thin sheet frames of brass braced together with brass pillars to form a lightweight and rigid platform on which to construct a hand-held instrument for taking celestial observations. While the double frame proved to be an effective 18th C. concept -- one which gave Troughton great notoriety as a respected instrument maker -- the complexity and cost of producing such sextants gave way to more simplified designs by the mid 1800's. (Private Collection. Ex. West Sea Co.)

To address the problem, perhaps the most famous of these innovations was the pillar frame sextant patented by Edward Troughton in 1788. The frame was constructed of two parallel strips of sheet brass joined together by machined pillars secured with screws, much like the trusses incorporated in "modern" iron bridges of the time. This "double frame" sextant was made by numerous makers using slight modifications for more than one hundred years thereafter. A variation of the double frame was the "bridge frame" sextant made by Ramsden and a few other makers late in the 18th century. Examples of both forms of these early sextants are quite scarce and highly sought after by collectors.


A rare hydrographic surveying sextant also known as a "Quintant" because of the increased size of its scale, reading to 165 degrees. In this lovely example, signed "Cary, London," the scales are engraved on gold and platinum and it bears British Admiralty Hydrographic Survey markings. Such instruments were used to take triangulation sightings when making soundings for Admiralty charts. Note the addition of a bubble level on the index arm just above the vernier magnifier. Not visible in this photo is the unique system of sextant frame stiffening that the Cary firm incorporated into this design. Circa 1850. (Collection West Sea Co.)


Although numerous frame styles were tried including the "bell" form and variations thereof, lattice, straight, arched, and braced, the final solution to the problem of rigidity and stability was found by using bell bronze alloy in casting the frame. The most prevalent frame style going into the 20th century was three adjacent circles.


By the mid-19th C. the standard brass frame English sextant had evolved into this form -- one which would endure for a century. (Private Collection. Ex. West Sea Co.)


A scarce mid-19th C. miniature sextant by Troughton & Simms, London. This genuine working instrument measures only 5 inches long on the index arm -- about half the size of a conventional sextant. These "mini-sextants" were sometimes presented to aspiring officers for their accomplishments, as in this case. However their diminutive size actually made them "too small" to be handled effectively. As a result, most examples that have survived are in excellent condition, showing little use. (Private Collection. West Sea Co. photo).

Engraved silver presentation on the box lid of the miniature sextant. Such a sextant must have made for a very meaningful gift to its proud young recipient. (Private collection. Photo West Sea Co.)

Among the earliest accessorial improvements was optical enhancement of the image by means of a telescopic attachment. Evidently this innovation came very near the advent of the sextant itself, even though telescopes generally were not incorporated for use with octants until about 1830. Strangely, a similar such disparity was the use of a handle. Even the earliest sextants had handles, whereas the inclusion of handles in the construction of octants came at about the same time that optics were added. Another feature unique to the sextant and lacking in the octant was the scale/vernier magnifier. Because of its smaller size and finer scale the sextant was read by means of a small magnifier affixed to the index arm. The octant required none


This lovely example signed "Parkinson & Frodsham, London" also dates circa 1850. Note the addition of a handle and horizon shades on the instrument itself. As for accessories, a peep sight and telescopic sight have been added. At this point, the only thing that differentiates the octant from a sextant is the arc of its scale and lack of a vernier magnifier. The existence of the "keystone" box helps in pointing at its mid 19th C. or earlier origin. The trade labels in the lid are those of Thaxter & Son and Aaron Breed, both of Boston. Thaxter's were the American agents for Parkinson & Frodsham during the first half of the 19th C. Original trade labels are an excellent tool in helping to date the period of an instrument. (Collection West Sea Co.)

By 1850 the demise of the octant was imminent, even though its use persisted into the 20th century. The superiority of the brass sextant in terms of accuracy, compactness, and durability was indisputable.

The last half of the 19th century saw little change in navigational instruments in general and the sextant in particular. A good sextant would provide 50 years of service. As makers fulfilled the need of mariners and demand declined there was little incentive for improvement in manufacture or design.

With the onset of World War I and the loss of hundreds of vessels and the construction of many more, nautical instrument manufacture was given a second wind. The advent of the drum micrometer sextant by the end of the War was the greatest single improvement in the sextant during the 20th century. By employing a continuous tangential screw attached to a calibrated knob, an instantaneous reading visible in the low light situations of the navigating bridge was readily obtainable. A further improvement was the incorporation of a small battery powered light for easier reading.

Actually this "new invention" was not new at all. There is an example of the drum micrometer incorporated into a sextant made by Ramsden that dates circa 1780. Before that it was incorporated on some astronomical instruments dating back to the 17th century!


A post World War II Japanese sextant by "Tamaya, Ginza, Tokyo." Modern features include a lightweight aluminum frame, endless tangent screw drum micrometer read-out, a built-in reading light, a large adjustable light gathering telescope and over sized mirrors and filters. The triangular device at the left is an artificial horizon, useful in low light or foggy conditions. (Private Collection. Ex. West Sea Co.)

Just prior to World War II the long evolution of the sextant culminated in the invention of the ball recording sextant. It is ironic, and perhaps fitting, that the final form of the sextant was not a sextant at all but a much earlier ancestor, the true quadrant. Developed for use at night when no horizon was visible, the recording "sextant" used no reflecting mirror. Rather, the celestial object was viewed directly as with the true quadrant. Instead of using a plumb bob, a liquid-damped steel ball recorded the altitude of the object on a screen. A drum micrometer was used to determine the precise altitude reading. Because of its size, easy reading, and nighttime capability, the recording sextant found favor with airplane navigators, leading to "aircraft sextants" built on the same principle. Numerous examples still exist for the collector's choosing at relatively inexpensive prices.

This has been a capsule look at the history of the sextant, and in no way constitutes a thorough treatise on the subject. Numerous topics relating to early navigation and its instruments, including the nocturnal, traverse board, ring dial, mariner's bow, circumferentor, hydrographic surveyor's sextant, pocket or box sextant, and others have not been discussed. A selected bibliography is provided to aid the reader in further study.

Good hunting!


Andrewes, William, "The Quest for Longitude - Proceedings of the Longitude Symposium Harvard University November 1993," Harvard University, Cambridge, 1996.

Bedini, Silvio, "Early American Scientific Instruments and Their Makers," Smithsonian Institution, Washington, DC, 1964.

Bennett, J.A., "The Divided Circle," Phaidon - Christie's, Oxford, 1987.

Brewington, M. V., "The Peabody Museum Collection of Navigating Instruments," Peabody Museum, Salem, 1963.

Brophy, Patrick, "Sailing Ships," Galahad Books, New York, 1974.

Clifton, Gloria, "Directory of British Scientific Instrument Makers 1550-1851," The National Maritime Museum, Greenwich, England, 1995.

Cotter, Charles H., "A History of the Navigator's Sextant," Brown Son & Ferguson Ltd., Glasgow, 1983.

Gould, Rupert, "The Marine Chronometer," The Holland Press, London, 1960.

Ifland, Peter, "Taking The Stars," The Mariners' Museum, Newport News, 1998.

L'e Turner, Gerard, "Antique Scientific Instruments," Blanford Press, London, 1980.

L'e Turner, Gerard, "Nineteenth Century Scientific Instruments," Sotheby's Publications, London, 1983

Pearsall, Ronald, "Collecting and Restoring Scientific Instruments," Arco Publishing Co., Inc., New York, 1974.

Randier, Jean, "Marine Navigation Instruments," John Murray. London, 1980.

Stimson, Alan and Daniel, Christopher, "The Cross Staff - Historical Development and Modern Use," Harriet Wynter, London, 1977.

Taylor, E. G. R., and Richey, M. W., "The Geometrical Seaman - A Book of Early Nautical Instruments," Hollis and Carter, London, 1962.

Wheatland, David, "The Apparatus of Science at Harvard 1765- 1800," Harvard University Press, Cambridge, 1968.

Wynter, Harriet and Turner, Anthony, "Scientific Instruments," Charles Scribners Sons, New York, 1975

Significant events in the evolution of angle measuring marine navigational instruments

Circa 9th century A.D.: First astronomical quadrant

9th century A.D.: First astronomical astrolabe. Brass and/or wood

10th century A.D.: Earliest surviving example of astronomical astrolabe

11th century A.D. Directional properties of lodestone employed

1187: Earliest known description of the compass

1342: Cross staff described by Levi ben Gerson

1415: Prince Henry the Navigator establishes his navigation institute

Circa 1450: First recorded mention of the mariner's simple quadrant, although the instrument predates this period

Circa 1460: First mariner's astrolabe. Sheet brass construction

Circa 1500: Introduction of cast brass mariner's astrolabe

1555: Earliest surviving example of a dated mariner's astrolabe

1594: Captain John Davis introduces backstaff

1631: Pierre Vernier describes his method for making an improved scale reading device

17th century: Backstaff constructed of walnut, lignin vitae, and fruitwood. Non-interpolative linear scale.

18th century: Backstaff constructed of ebony, rosewood, and mahogany in addition to earlier materials. Diagonal (interpolative) scale.

1731: John Hadley introduces reflecting quadrant

1735: John Harrison's first chronometer

1756: Mayer's reflecting circle

1757: First "Hadley sextant" by John Bird.

1760: Brass, ebony and ivory introduced as materials for constructing instrument frames. Brass begins replacing wood as the preferred material, starting with the base of the index arm in quadrants. Existing vernier scales calibrated "10 - 0 - 10"

1768: Scale dividing engine by Jesse Ramsden

Circa 1770: Finite scale on vernier calibrated "20 - 0 - 20".

Circa 1780: Instruments decrease in size with widespread use of dividing engine in scale manufacture

Circa 1780: Introduction of vernier scale with right reading "0."

Circa 1785: Inclusion of fine adjust tangent screw feature in the production of most instruments. Index arm of brass construction.

1788: Edward Troughton patents the lightweight "pillar frame" sextant.

Circa 1790: Flat brass index arm gives way to braced type.

Circa 1820: Elimination of backsight feature in the construction of most octants

Circa 1840: Octants and sextants more similar in appearance with the inclusion of handles, optics, and all brass construction in octants

Circa 1880: Cessation of the production of wooden instruments of the navigational type

Circa 1918: Inclusion of the drum micrometer feature in modern sextants

Circa 1940: Ball recording sextant, aircraft sextant

Characteristics of an octant or sextant (as applicable) useful in dating

Dated instrument

Maker's name on instrument or box label

Owner's name on instrument or box

Presence of backsight

Wooden index arm vs. brass combination wood and brass index arm

Flat brass index arm vs. braced

Locking screw vs. locking screw and tangent screw combination

Presence of vernier scale

Type of vernier scale division ("10 - 0 - 10" vs. "20 - 0 - 20" vs. "20 - 15 - 10 - 5 - 0")

Presence of vernier magnifier affixed to index arm

Type of wood used in frame construction vs. brass

Size of instrument

Frame configuration

Existence of index arm "stop" on right limb of instrument

Use of ivory in some components

Presence of wooden "feet" vs. turned brass

Presence of optics vs. peep sights

Existence of pivoting peep sight "shade"

Presence of an ivory "pencil" and/or "notepad"

Presence of handle

Type of silvering used on mirrors

Configuration of box (angular shaped vs. stepped keystone vs. keystone vs. rectangular wood vs. bakelite)

Construction of box (type of dovetailing if any)

Configuration of filter shades (double vs. triple) and interchangeability of same

Existence of separate fixed horizon filters in addition to index filters

Existence of variable height sighting tube holder vs. earlier fixed holder

Information found on paper gasket inside pivot point

© 1998-2013 West Sea Company. All rights reserved. Reproducing any part of this article without the expressed written consent of the author is forbidden by law. Violators will be prosecuted.