Hostname: page-component-586b7cd67f-g8jcs Total loading time: 0 Render date: 2024-11-28T13:36:03.614Z Has data issue: false hasContentIssue false
  • English
  • Français

Benjamin Martin and the Development of the Orrery

Published online by Cambridge University Press:  05 January 2009

John R. Millburn
John R. Millburn
Affiliation:
58 Bierton Road, Aylesbury, Bucks HP20 1EJ.
Get access Rights & Permissions [Opens in a new window]

Extract

The catalogues of mid-eighteenth-century mathematical instrument makers included not only instruments in the sense ‘tools of science’ but also devices intended primarily for educational use. Among the latter would be found mechanical models of the solar system, in varying degrees of complexity and under various names; for convenience such models may be, and often were, loosely called orreries.

Type
Research Article
Information
The British Journal for the History of Science , Volume 6 , Issue 4 , December 1973 , pp. 378 - 399
Copyright
Copyright © British Society for the History of Science 1973

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 See, for example, Clay, R. S. and Court, T. H., The history of the microscope (London, 1932)Google Scholar, chapter IX.

2 The young gentleman and lady's philosophy, i (1759), 200.Google Scholar This work, which was originally published in the monthly numbers of the General magazine of arts and sciences, is discussed further in note 52; the paragraph quoted appeared in December 1756.

3 The description and use of both the globes, armillary sphere and orrery (London, [1762]), p. 178.Google Scholar

4 Gentleman's magazine, lii (1782), 95.Google Scholar

5 ‘Brief memoirs of the late ingenious Mr. Benjamin Martin, accompanied with a portrait, elegantly engraved from an original painting’, Gentleman's magazine, lv, part II (1785), 583Google Scholar, and plate facing. The memoirs amount to only one paragraph, plus a list of 31 publications (not entirely accurate). In a footnote the editor states that he will give the painting to the curator of any public institution who thinks it worth preserving; a further footnote on p. 943 states that it had been given to Greene's Museum. It is listed in Greene, Richard's Particular and descriptive catalogue of the curiosities … in the Lichfield Museum (3rd edn., Lichfield, 1786), p. 84Google Scholar, but its subsequent history is unknown.

6 The philosophical grammar, by Benjamin Martin, фιλоτєχνος, was printed for John Noon, London, and is listed in the June 1735 Register of Books in Gentleman's magazine, v (1735), 335.Google Scholar Its preface is dated ‘from Chichester, December 14 1734’. An anonymous review appeared in The present state of the republic of letters, xvi (1735), 165–87Google Scholar (September). After the introductory biographical passage (which is qualified by ‘if I am rightly informed’), this consists largely of extracts from Martin's preface and text.

7 Manning, O. and Bray, W., The history and antiquities of the county of Surrey, iii (London, 1814), p. 89.Google Scholar Some reference books state (almost certainly in error) that Martin was born at Chichester, presumably because that is the first address mentioned in his publications.

8 In 1862 the Microscopical Society had just acquired a Martin microscope, and Williams attempted to obtain some biographical information to accompany a paper on this. See J. Williams, ‘Some account of the Martin microscope …’ and ‘A few words more on Benjamin Martin’, read to the Society on 8 January 1862 and 8 October 1862 respectively, in Transactions of the Microscopical Society of London, new ser. x (1862)Google Scholar, [31]–41, and xi (1863), [1]–4.

9 An entry in the first volume of the Worplesdon parish registers reads: ‘Benjamin son of John Martin Gent., bapt. March 1 1704’ [O.S.]. I am indebted to the staff of the Guildford Muniment Room, where this volume is now held, for this information. The date and place agree with the story given by Manning & Bray, but it has not been positively established that this entry refers to the Benjamin Martin who is the subject of this paper. Research on Martin's early life is still in progress.

10 A list of about eighty primary titles and a similar number of secondary and foreign-language editions has been compiled by Mr P. J. Wallis of Newcastle University (whose assistance with bibliographical aspects of this paper is hereby acknowledged) and will be published in his Bibliography of British Mathematics to 1850, now in preparation. No library has a full set of Martin's works. Some of the larger collections are at: (a) The British Museum, London (now incorporating the Patent Office collection); (b) The Museum of the History of Science, Oxford; (c) University College, London (Graves Collection); and (d) The Library Company of Philadelphia, Pennsylvania, U.S.A.

11 A new compleat and universal system or body of decimal arithmetic (London, 1735).Google Scholar

12 The young trigonometer's compleat guide (2 vols., London, 1736).Google Scholar The preface is signed ‘… from my school in Chichester, April 8 1734’. Probably this work, Arithmetic, and Philosophical grammar were all in manuscript together, and it was largely a matter of chance that Philosophical grammar appeared in print first.

13 Martin probably settled in Chichester about 1730, for on 29 October 1729 he obtained a licence to marry Mary Lover, spinster, of Chichester; in this document he is described as ‘merchant of Guildford’; see Calendar of Sussex marriage licences recorded in the peculiar courts of the Dean of Chichester and the Archbishop of Canterbury, Sussex Record Society publications, xii (1911), 172. The original document is in the County Record Office, Chichester, Ep. III/6/1729. I am indebted to the staff of the County Record Office for drawing my attention to this, and also for locating a lease showing that Martin occupied premises in South Street in 1737. Martin makes no reference to his school after 1736, so it was probably not a successful venture; by the end of the 1730s he was describing himself simply as an author.

14 The description and use of a new invented pocket microscope with a micrometer (Chichester, printed by W. Lee, n.d.). This tract is illustrated by two plates, one of which is dated July 1738; it was first advertised in August 1738. Copies with an Appendix describing a later form of eyepiece may have been issued in the following year. Advertisements in this and contemporary works state that these microscopes were made and sold by the author in Chichester. Martin's interest in the use of the microscope at this time is revealed by a letter to Sir Hans Sloane, in which he wrote that he could ‘now and then provide ye Society with some curious & uncommon microscopic objects’. See British Museum, Sloane MSS. 4056, ff. 26–7 (14 Jan. 1738/9).

15 A course of lectures in natural and experimental philosophy, geography, and astronomy … confirmed by experiments (Reading, 1743).Google Scholar The list of booksellers for whom this was published includes names in London, Oxford, Cambridge, York and Scarborough, Salisbury, Bath, Ipswich and Newbury. This work shows that by 1743 Martin had acquired some knowledge of experimental philosophy, as distinct from knowledge of the literature. It has been suggested that he acquired this by working for J. T. Desaguliers, but evidence for this is flimsy and seems to rest entirely on a satirical tract by (or attributed to) Horne, George, The theology and philosophy in Cicero's Somnium Scipionis explained (London, 1751).Google Scholar This tract mentions Martin's Course of lectures textbook and comments that he ‘having attended Dr. Desaguliers’ fine, raree, gallanty shew for some years in the capacity of a turn-spit, has, it seems, taken it into his head to set up for a philosopher …' (p. 23).

16 An essay on electricity, being an enquiry into the nature, cause, and properties thereof (Bath, 1746).Google Scholar On p. 18 Martin refers to experiments ‘as I daily shew in my course of philosophy’.

17 Lingua Britannica reformata, or new English dictionary (London, 1749).Google Scholar A second (enlarged) edition appeared in 1754. Martin also wrote various essays on the English language; for details see Alston, R. C., A bibliography of the English language (Leeds, 1965—in progress), volumes i and v.Google Scholar

18 Martin's shop on the north side of Fleet Street, between Crane Court and Red Lion Court, is variously identified in his publications as ‘The visual glasses and globe’ (1757)Google Scholar, ‘The globe’ (1758)Google Scholar, and ‘Hadley's quadrant and visual glasses’ (1762).Google Scholar It would appear that Martin altered his sign to suit the nature of his trade at the relevant period. For remarks on the premises themselves see note 46.

19 For details of the various parts, states, volumes, etc., of the magazine see Millburn, J. R., ‘Martin's magazine—the general magazine of arts and sciences 1755–65’, The library, 5th ser. xxviii (1973), 221–39.CrossRefGoogle Scholar Comments on volumes relevant to the present paper are given in notes 52 and 57 below.

20 Martin's wife Mary was almost certainly the daughter of Joshua Lover, maltster, of Chichester who died in mid-1729 (his Will was proved in June 1729; Public Record Office, London, PROB/11/630), after whom the son was presumably named. Joshua Lover Martin has left little record of his activities except four patents taken out between 1774 and 1782, in which he is described as ‘optician, of Fleet Street’. In the London directories ‘B. Martin, optician’ becomes ‘B. Martin & Son, opticians’ about 1778. At least one instrument signed ‘B. Martin & Son’ is known, namely a microscope at the Science Museum, London (Inv. 1921–755). Several tracts were issued in the name of the firm, mostly on new forms of microscope, but the only positively dated example is A curious collection of electrical experiments (London, 1779).Google Scholar It was probably not a formal partnership in the modern sense. After his father's death Joshua Lover Martin seems to have gone abroad (see note 23 below); he apparently never traded as an optician in his own right.

21 Martin's name is included in the list of bankrupts in Gentleman's magazine, lii (1782), 48.Google Scholar The date of sealing the bankruptcy order is given as 7 January 1782 in Docket Book Index Volume IND 22655 at the Public Record Office, but the detailed records of this case appear not to have survived. The circumstances of Martin's bankruptcy are obscure, for according to the memoir of 1785 (note 5) he had ‘a capital more than sufficient to pay all his debts’.

22 See the 1785 memoir cited in note 5; ‘the wound, though not immediately fatal, hastened his death’.

23 Martin's burial entry in the registers of St. Dunstan's in the West is quoted in Noble, T. C.'s Memorials of Temple Bar (London, 1870), p. 84Google Scholar: ‘1782 Feb. 14. Benjamin Martin from Fleet Street, East Vault’. Letters of Administration were granted to his daughter Maria, wife of Humphrey Jackson Esq., on 17 April 1782 (Public Record Office, PROB/6/158). A further grant of goods, etc., unadministered occurred in November 1787, presumably on Maria's death; on the occasion of the second grant Gabriel Wright (probably the instrument maker) acted on behalf of Joshua Lover Martin who was then residing in Naples.

24 The philosophical grammar (2nd edn., London, 1738), p. 27.Google Scholar For this edition Martin added numerous footnotes, increasing the number of pages by 40 and inserting ten more plates. Subsequent editions were almost identical with the second; the work was last issued in English in 1778 as the ‘18th’ (actually 8th) edition. It was translated into French, Dutch, Italian, and Greek.

25 Heath's orrery print of c. 1738 has not been traced, but an illustration of an orrery-sphere by Thomas Heath in the Conservatoire National des Arts et Métiers, Paris, is given in Rice, Howard C. Jnr's The Rittenhouse orrery (Princeton, N.J., 1954)Google Scholar, plate IV. This model agrees well with Martin's description. One of Heath's trade cards which includes a sketch of an orrery-sphere is reproduced in Goodison, N., English barometers (London, 1968), p. 53.Google Scholar

26 [Sir Richard Steele], The Englishman, No. 11 (29 10 1713).Google Scholar The origin of the term ‘orrery’ is often incorrectly attributed to Steele himself, but his actual words were: ‘The honest man [Rowley] calls his machine the Orrery, in gratitude to the nobleman of that title; for whose use and by whose generosity and encouragement he began and accomplished the undertaking’.

27 Gould, R. T., ‘The original orrery restored’, The illustrated London news, cxci (18 12 1937), 1102–3, 1106.Google Scholar This model was exhibited in various institutions after its restoration and can now he seen in the Astronomical Gallery of the Science Museum, London.

28 For a description, illustrations, and outline history of this model, which is now in the Museum of the History of Science, Oxford (Inv. 47/48–27), see Gabb, G. H. and Taylor, F. S., ‘An early orrery by Thomas Tompion and George Graham’, The connoisseur, cxii (09, 1948), 24–6, 55.Google Scholar It is also described and illustrated in Symond, R. W.'s Thomas Tompion, his life and work (London, 1951), pp. 294–5Google Scholar, and figure 261.

29 It is generally taken for granted that Rowley's model was derived from the one by Tompion and Graham. However, there is another similar but little-known Sun/Earth/Moon model signed by Graham alone at the Adler Planetarium, Chicago. Apart from a brief mention in the English edition of Michel, P. H., Scientific instruments in art and history, trans. , R. E. W. and Maddison, F. R. (London, 1967), p. 100Google Scholar, this model has received little attention. J. T. Desaguliers, who first related the story of Rowley's copying an earlier model while it was temporarily in his possession, stated that it was Graham who was ‘the first person in England, who made a movement to shew the motion of the Moon round the Earth, and of the Earth and Moon round the Sun, about 25 or 30 years ago’; see Desaguliers, J. T., A course of experimental philosophy, i (London, 1734), 430.Google Scholar Unfortunately for historians he added: ‘if I am rightly informed’.

30 This instrument, made in 1682 by Van Ceulen to Huygens's design, remained in the possession of the Huygens family until 1754. In 1809 it was bequeathed by the then owner to Leyden University, and it can now be seen in the Rijksmuseum voor de Geschiedenis der Natuurwetenschappen, Leyden. A brief description and two illustrations are given in Crommelin, C. A., Descriptive catalogue of the Huygens collection (Leyden, 1949), pp. 18–19Google Scholar and plate III. Its design and construction were fully described by Huygens in a Latin manuscript printed in Huygens, 's Opuscula postuma (Leyden, 1703), pp. 429–60Google Scholar, with four plates showing its external appearance and mechanism. A French translation with numerous additional notes and comments is included in volume xxi of Oeuvres complètes de Christiaan Huygens (The Hague, 1944)Google Scholar. This device not only reproduces the relative mean periods of the six planets but also shows their orbits to scale with the correct eccentricities and perihelion/aphelion velocity ratios. It was designed as a perpetual ephemeris machine rather than a demonstration model.

31 The history of Roemer's planetariums and their subsequent fate is confused. The usual story that they were all lost in the fire at Copenhagen in 1728 is unreliable, as it is not known precisely how many were made nor how they were distributed. A ‘table’ model contemporary with Huygens's planetarium is described and illustrated in Gallon, 's Machines et inventions approuvées par l'Académie Royale des Sciences, i (1735), 81–3Google Scholar; a ‘ceiling’ version (probably later) is described and illustrated in Horrebow, P., Basis astronomiae sive astronomiae pars mechanica (Copenhagen, 1735), pp. 131–8Google Scholar. Both the above were Copernican.

32 This model is described and illustrated in Harris, John, Astronomical dialogues (London, 1719), pp. 159–80Google Scholar. According to Taylor, E. W., Wilson, J. S., and Scott-Maxwell, P. D. in their history of Messrs. Cooke Troughton & Simms Ltd., At the sign of the orrery (York, n.d.)Google Scholar, it was commissioned by the East India Company at a price of £500. Their statement that it eventually went to China with Lord Macaulay's (sic) mission must be mistaken, however, as Rees (i.e. W. Pearson) gives an account of its wheelwork based on personal inspection; see Rees, Abraham's Cyclopaedia (London, 1819)Google Scholar, Art. ‘Orrery’. According to Reid, Thomas, Treatise on clock and watchmaking (Edinburgh, 1826), p. 50Google Scholar, the device that went to China in 1792 with Lord Macartney's mission was the much more elaborate model by P. M. Hahn and A. de Mylius; for an account of this see Description of a planetarium or astronomical machine, which exhibits the most remarkable phaenomena … completed by Mr. Albert de Mylius (London, 1791)Google Scholar. The ultimate fate of Rowley's second model is unknown. Until recently no further orreries directly attributable to Rowley were known, but another is described in Chenakal, V. L., ‘The astronomical instruments of John Rowley in 18th-century Russia’, Journal of the history of astronomy, iii (1972). 119–35CrossRefGoogle Scholar. Photographs of this, taken before it was destroyed in the Second World War, show a typical grand orrery with superior planets and an armillary hemisphere, but documents quoted in the paper suggest that it originally was similar to Rowley's second model. It appears to have undergone substantial repairs in the 1770s. and the mechanism may have been reconstructed at that time.

33 See A system of experimental philosophy prov'd by mechanics … to which is added … Rowley's Horary, a machine representing the motion of the Moon about the Earth, Mercury and Venus about the Sun, according to the Copernican system (London, 1719), pp. 194–201Google Scholar. This work was issued by Paul Dawson in the name of J. T. Desaguliers without his authority, but another edition, corrected by Desaguliers, appeared in the same year. The latter version is entitled Lectures of experimental philosophy … to which is added a description of Mr. Rowley's machine called the Orrery; it includes a long list of errata, but the spelling of ‘orrery’ was not amended (except in the title).

34 For a documented account of Rowley's and Wright's activities, see Taylor, , Wilson, , and Scott-Maxwell, , op. cit. (32), pp. 7–17Google Scholar. The information given there on Wright's addresses is not entirely accurate: the St Bride's rate books show that from 1720 to 1731 he shared [135] Fleet Street with George Graham, and did not move to [136] until 1732.

35 An advertisement in the fifth edition of Harris, Joseph, Description and use of the globes and orrery (London: printed for T. Wright et al., 1740Google Scholar) states that by then Wright had made grand orreries for Watt's academy, for the King at Kensington, and for the new Royal Academy at Portsmouth, as well as ‘several other large ones for noblemen and gentlemen’. This book, which was first published in 1731 and went through twelve editions up to 1783, includes a large folding plate showing a grand orrery with armillary hemisphere. This illustration was frequently copied (in essence) by later publishers who wanted to show a typical grand orrery; see, for example, plate XIX of Hutton, C., Mathematical and philosophical dictionary, ii (London, 1796)Google Scholar. Martin used an almost identical plate in The young gentleman and lady's philosophy, i (1759)Google Scholar, plate XIX.

36 Wright grand orreries known to be extant include: (a) Science Museum, London (Inv. 1927–1659). Made for George II in 1733, this model is described in Chaldecott, J. A., Handbook of the King George III collection of scientific instruments (London, 1951)Google Scholar, item 140. Colour illustrations are given in the Science Museum booklets Physics for princes (London, 1968)Google Scholar, plate 16, and Astronomy (London, 1967)Google Scholar, plate 13; (b) Museum of the History of Science, Oxford: orrery dated 1731. This model incorporates the inner planets but not Mars, Jupiter, or Saturn (so strictly speaking it is not a ‘grand orrery’) in a twelve-sided case without an armillary hemisphere. It is illustrated in Gunther, R. T., Early science in Oxford, ii (Oxford, 1923), facing p. 258Google Scholar; (c) Musées Royaux d'Art et d'Histoire, Brussels (Inv. D.243). This is a typical grand orrery with armillary hemisphere, but the case is round instead of twelve-sided.

37 Bailey, N.'s Dictionarium Britannicum (2nd edn., London, 1736)Google Scholar, Art. ‘Orrery’ states that ‘Mr. Glyn, mathematical instrument maker in London, was the first that applied the orrery to the armillary sphere’, and implies that Glyn had made one for the Earl of Pembroke. An orrery by Glynne with wheelwork for the Earth, Moon, and inner planets, with manually adjustable superior planets, mounted within a large brass armillary sphere, is in the Museum of the History of Science, Oxford (Inv. 69–190). Another, probably earlier in date, with wheel-work for the Earth's parallelism only, is in the National Maritime Museum. Greenwich, (Inv. p. 13).Google Scholar

38 For descriptions and illustrations of this model in two different configurations, see Desaguliers, . op. cit. (29), i. 430–8, and plates 31, 32.Google Scholar

39 Philosophia Britannica: or, a new and comprehensive system of the Newtonian philosophy … in a course of twelve lectures (2 vols., Reading, 1747).Google Scholar

40 Ibid., ii. 369. The figures may have been exaggerated. John Rowley died in January 1727/8.

41 Ibid., i. 142, and plate XII.

42 Neale's orrery is illustrated and described in the printed version of his Patent no. 605, granted in 1744, entitled A quadrantal planetarian machine, by which the altitude of any of the planets or fixed stars may be taken, as also all the uses of the orrey answered. Unless there were major errors in transcription, however, his description was largely nonsense; the wheelwork described would not give anything like the correct period ratios. Although basically coaxial, Neale's mechanism did not employ a cone of driving gears such as Martin described. Martin omitted this reference in the third and fourth editions of Philosophia Britannica (1771 and 1788)Google Scholar, though it is still in the second edition (1759).

43 This terminology is introduced here as a convenient way of specifying a system of wheel-work such as that shown in Figure 1. It should not be confused with Roomer's ‘conical gears’, which were skew gears with unequal teeth, formed by cutting an oblique section through a cone. For the latter see Gallon, , op. cit. (31), 89–91.Google Scholar

44 The wheelwork of Roemer's jovilabe (made in 1678) had been described in English in Derham, W., The artificial clockmaker (London, 1696), pp. 119–20Google Scholar; also, with a diagram, in Harris, John, Lexicon technician (London, 1704)Google Scholar, Art. ‘Satellite instrument’. Horrebow, op. cit. (31), gives a description in Latin, on pp. 121–31.Google Scholar

45 See note 31. As these books were published in France and Denmark respectively, it is possible that Martin had not seen them.

46 The quarterly rate books for the relevant period are missing, but the City Land Tax Books (Guildhall, London, MS. 11, 316) show that Martin first occupied premises two doors east of Crane Court sometime between September 1755 and May 1756. His next-door neighbour was the optician John Cuff, who left the area shortly after Martin arrived. Between mid-1759 and mid-1760 Martin moved to larger premises two doors further east; this latter address became No. 171 when the street was numbered about 1768.

47 The earliest short list of products sold at the Fleet Street shop occurs in Martin, 's Essay on visual glasses (London, 1756), p. 28Google Scholar, but the first detailed and priced catalogue appeared in the following year with his Theory of comets (London, 1757)Google Scholar. This covers 136 items, of which 33 are in the category ‘invented or improved by B. Martin’. It does not necessarily follow that all the items listed were actually made, but nevertheless this catalogue indicates that Martin was the proprietor of a retail (and wholesale) shop rather than an individual craftsman. Similar catalogues were appended to many of his publications during the next twenty years. The latest that has come to light is in a disbound collection of tracts at the Museum of the History of Science, Oxford; it is undated, but is headed ‘B. Martin & Son’ and from internal evidence is after 1775.

48 In the second edition of his Dictionary (1754Google Scholar, see note 17) Martin defined orrery as ‘An astronomical machine explaining the solar system by means of several clockwork movements’. There is no entry for planetarium.

49 The following double-cone planetariums by Martin are known to be extant: (a) Science Museum, London (Inv. 1919–463). All-brass model on tripod stand, with planets to Saturn; manual satellites; (b) Adler Planetarium, Chicago, U.S.A. (Inv. AM 391). As item (a). From the Mensing Collection; (c) Science Museum, London (Inv. 1912–222). Smaller version on wooden pillar stand, no satellites. This model has an arm and mechanism for Uranus, but I am informed by Mr A. G. Thoday, Deputy Keeper, that the Uranus work is probably a later addition. Other models by Martin, more complex than the above, are mentioned in notes 62, 64, and 70.

50 The publishing history of this work is obscure. What appears to be the first edition was published in London in 1751; another version (not called ‘second edition’) is dated 1754. The relatively common ‘fifth edition’, which has experimental inserted in the title before philosophy, is dated 1765. A sixth edition is mentioned in a later catalogue, but this may be a misprint. At the date of writing this paper only the 1751, 1754, and fifth (1765) editions had been located.

51 In the short list of products published in 1756 (see note 47) planetariums were placed first, but later catalogues were dominated by optical instruments. In the London directories Martin was always entered as ‘optician’, not ‘mathematical instrument maker’.

52 The young gentleman and lady's philosophy was first published as one of the four principal parts of the General magazine of arts and sciences, commencing in January 1755. Each monthly number of the magazine comprised eight pages from each of four different works, plus sixteen or twenty pages of ‘miscellanies’ such as correspondence, book reviews, etc. When sufficient material had been issued to make up a ‘subject’ volume the sheets were re-bound into the latter form, and the magazine ceased to exist. Volume i of The young gentleman and lady's philosophy, with its individual title and index, is dated 1759, but the initial chapters covering double-cone planetariums had been issued in 1755.

53 The first edition of this work is undated, but from internal evidence it can be firmly dated 1762. Several bibliographies have wrongly assigned a date of 1736 to this book, owing to an ambiguous entry in the 1785 memoir (cited in note 5). The date is important because the book contains an integral catalogue which actually belongs to Martin's ‘London’ period, not to his much earlier ‘Chichester’ period.

54 See note 49, items (a) and (b).

55 This demonstration was grossly unfair to Ptolemy. What it really showed was a simple geocentric system in which all the planets move with uniform velocity in geocentric circles. Without epicycles, many of the arguments used to refute the Ptolemaic system were invalid, and were really directed at a hypothetical system which quite clearly did not fit the observed movements of the planets. For example, the demonstration allowed the inferior planets to be in opposition to the Sun, a situation that cannot occur if the appropriate epicycles are incorporated.

56 Pearson's, W.mean-motion planetarium, now in the Science Museum, London (Inv. 1950–1955)Google Scholar, incorporates no less than fourteen coaxial members, of which the third from the centre is held stationary to provide a fixed reference for the parallel motion drive. See Rees, op. cit. (32), plate VII of ‘Planetary machines’.

57 A new and comprehensive system of mathematical institutions was the fourth principal part of the General magazine (see note 52). Volume ii in bound form is dated 1764, but the chapters covering tellurians and lunariums were probably issued in the previous year. ‘Institutions horolegicae’ occupies pp. 361–535Google Scholar; ‘Tellurian’ pp. 484–91Google Scholar; and ‘Lunarium’ pp. 492–9.Google Scholar

58 The plates accompanying the horological chapters are not numbered. ‘A tellurian with three globes’ and ‘Mechanism for a lunarium’ (on one plate) are normally found at p. 484.

59 Apparently as a result of misreading an ephemeris, Martin derived a nodal rotation period of 7082½ days—a surprising error, as it was well known that the true period was about 6798 days.

60 The requirements for parallel and nodal (and apsidal) trains were well illustrated by James Ferguson's ‘Mechanical paradox’. In this device the parallelism train consisted of three equal gears, of 39 teeth each; the middle gear drove also a gear of 37 teeth, which therefore exhibited a slow retrogression with respect to the ‘parallel’ gear. See Ferguson, J., The description of a new machine called the mechanical paradox (London, 1764)Google Scholar; this was reprinted in his Select mechanical exercises (London, 1773), pp. 44–71.Google Scholar

61 Wheatland, D. P., The apparatus of science at Harvard, 1765–1800 (Cambridge, Mass., 1968).Google Scholar

62 Harvard Collection (Inv. 52). See Wheatland, , op. cit. (61), pp. 48–51Google Scholar, for a description and illustrations of the planetarium and its attachments, and of the ‘cone of gears’ (shown removed from the planetarium).

63 Harvard MSS., quoted in Wheatland, , op. cit. (61), pp. 49–50.Google Scholar

64 For the story of Bowdoin's gift and Martin's bill, see Wheatland, , op. cit. (61), pp. 52–4Google Scholar. The orrery itself is also described and illustrated by Cohen, I. B., Some early tools of American science (Cambridge, Mass., 1950), pp. 156–7CrossRefGoogle Scholar. See also Rice, op. cit. (25), plate VI.

65 See, for example, catalogues in the tenth (1768) and later editions of Harris, op. cit. (35). Although the first priced catalogue of Benjamin Cole & Son seems to be dated after the Harvard order had been placed, earlier editions of Harris contain advertisements stressing B. Cole senior's ability to make orreries, and the firm's prices must have been available. A grand orrery by Benj. Cole & Son dated 1763 was presented to The Queen's College, Oxford, in that year, and is still preserved in the College Library; see Gunther, , op. cit. (36), ii. 271Google Scholar, and plate facing.

66 Catalogues in Adams, G., Treatise describing and explaining … new celestial and terrestrial globes (London, 1766Google Scholar; 2nd edn., London, 1769). As in the case of the Coles, these catalogues are slightly later than the date of the Harvard order, but Adams had been advertising orreries for at least twenty years and his prices must have been available. When the third edition of this book was published in 1772 the price range for large orreries was increased (possibly by George the younger) to ‘£130–1,500 gns’.

67 Comparison of prices quoted by Wheatland with those in Martin's catalogues reveals some surprising reductions. For example, a cometarium (Harvard Collection, Inv. 6) was supplied at 3½ guineas, although it had been listed at 5 guineas since 1757. The price charged for the planetarium/tellurian/lunarium (15 guineas total) was well below the list price of the best manual orrery (21 guineas), despite the extra work involved. On the other hand, a large item such as a pair of 28-inch Senex globes was charged at the full catalogue price of £35.

68 Description and use of an orrery of new construction (London, 1771).Google Scholar

69 According to Henderson, E., Life of James Ferguson (Edinburgh, 1867), p. 81Google Scholar, the use of continued fractions for orrery wheelwork calculation was unknown in England before publication of Martin's tract, but this statement may have been influenced by Martin's own comment that the only previous account he had seen was by Huygens.

70 Until recently the only extant planetarium/tellurian/lunarium by Martin was thought to be the Harvard model, but two further examples came to light while this paper was in preparation. Both have a toothed edge to the top plate, not a separate fixed gear, for actuating the attachments, so they are not representative of the 1771 pattern; but they illustrate a later stage in the development of the tellurian/lunarium attachments in that their mechanism incorporates a radial shaft driven by a contrate wheel on the central coaxial assembly. This type of driving mechanism is found on all the planetarium/tellurian/lunarium models made by Adams and by W. and S. Jones after Martin's death. Details of these two Martin models are: (a) Planetarium signed ‘A new planetarium B. Martin London’, brass drum-on-tripod type with planets to Saturn; wheelwork drum approx. 300 mm. diameter with toothed edge; tellurian attachment with 75 mm (3 inch) globe by James Ferguson, with auxiliary 24-hour dial on tellurian arm. Sold at Messrs. Phillips Son & Neale's auction sale no. 18944, 29 February 1972, lot 140; now in a private collection in England. The internal double-cone wheelwork corresponds with that shown in figure I except that the Earth gears have 48 teeth; the toothed top plate has 274 teeth (as does Harvard model, Inv. 52). I am indebted to the present owner for permission to examine the interior of this model. (b) Tellurian/lunarium signed ‘B. Martin invenit & fecit Londini’, brass drum-on-tripod type, wheelwork drum approx. 160 mm. diameter with toothed edge; tellurian attachment with 75 mm. globe (no auxiliary dial on arm); lunarium attachment. Apparently no planetarium. I am indebted to Dr V. Himmelein of the Wurttembergisches Landesmuseum, Stuttgart, for information on this model, which at the time of writing was in a private collection in West Germany.

71 Wright, T., Description and use of an astronomical instrument, being the orrery reduc'd (London, n.d.)Google Scholar. This tract must have been published before 1728 (and therefore in Rowley's lifetime) as the author describes himself as ‘Mathematical instrument maker to H. R. H. George, Prince of Wales’.

72 An orrery having some resemblance to Wright's ‘orrery reduc'd’ is in the King George III Collection at the Science Museum, London (Inv. 1927–1414).

73 See note 35. The first edition (1731) has no advertisements, but the second (1732) has an advertisement on verso of title commencing ‘Mr. Wright has lately made several very large ORRERYS, with all the motions of the primary and secondary planets, which makes the price very great: but being willing that academies, schools, &c, should have the benefit of this useful machine, he has contrived an orrery, which exhibits what relates to the motions of the Sun; Earth, and Moon; to be sold at a very moderate price’. Subsequent editions include a short list of purchasers of large orreries, plus a statement that Wright also made ‘small ones for schools’.

74 Inv. 1919–1 (Astronomy Collection).

75 An engraving of the orrery was published in 1746 with Ferguson, James's tract Use of a new orrery (London, 1746)Google Scholar, and was also used as the frontispiece of his Astronomy explained upon Sir Isaac Newton's principles (London, 1756)Google Scholar. The wheelwork is described and illustrated in his Select mechanical exercises (London, 1773), pp. 72–87Google Scholar, and plates VI, VII. An orrery Corresponding closely with this description is in the Museo di Storia della Scienza, Florence, but it is possible that several were made to the same pattern.

76 See, for example, the ‘Calculator’ described and illustrated in Ferguson's Astronomy explained, pp. 249–54Google Scholar, and plate VII. This was a development of the ‘4-wheeled’ orrery described and illustrated in his Dissertation on the … harvest moon, also, the description of a new four-wheeled orrery (London, 1747), pp. 19–48Google Scholar, and plate I. Another type of pulley-and-cord orrery by Ferguson is in the Science Museum, London (Inv. 1934–134).

77 Ferguson, , Tables and tracts relative to several arts and sciences (London, 1767), pp. 129–34Google Scholar. Ferguson's tooth numbers were a mixture of Martin's and Roemer's.

78 George Adams the elder's year of death is usually given as 1773, on the strength of a statement by his younger son Dudley nearly forty years later, but in fact he died on 17 October 1772; see Gentleman's magazine, xlii (1772), 496Google Scholar. His Will was proved on 24 October 1772 (Public Record Office, PROB/11/981).

79 One is in the Whipple Museum, Cambridge; see Gunther, R. T., Early science in Cambridge (Oxford, 1937)Google Scholar, item 201. A ‘pair’ of grand orreries of rather later date is in the Musées Royaux d'Art et d'Histoire, Brussels (Inv. D.241 and D.242); these models are in very similar twelve-sided cases with armillary hemispheres, but whereas one is a conventional grand orrery, the other has an enlarged Earth/Moon system and no superior planets.

80 ‘The description and use of a terrestrial planetarium … commonly called an orrery’, a manuscript of 33 leaves signed ‘G. Adams scripsit, Nov. 20th 1741’; Astronomy Department, Science Museum, London. I am indebted to the Director of the Science Museum for permission to examine this. Most of the text is an account of the Copernican system and the uses of the orrery, with only a brief description of its construction. It incorporated the motions of Mercury, Venus, Earth, and Moon, and the rotation of the Sun on its axis.

81 Adams, G., Micrographia illustrata (2nd edn., London, 1747), pp. 242–63Google Scholar. This catalogue was first issued with the first edition of the same title in the previous year. The simplest ‘planetarium’ listed was one foot in diameter and showed only the motions of the Earth and Moon round the Sun; a 2-ft. model incorporated Mercury and Venus as well; a 2½-ft. model had the superior planets; and a 3½-ft. model showed the motions of all the planets' satellites. The last two items in the catalogue were ‘A particularly new and curious machine, containing a movement which plays either an organ or harpsicord, or both … and sets going a transparent firmament… not unworthy the grandest apartment in any gentleman's or nobleman's house’; and a ‘COSMOTHEORION, or machine, of a new invention, which at present is without parallel’. The latter was ‘about four feet and a half diameter, standing on a pedestal of curious workmanship’, and reproduced the motions of all the bodies in the solar system, including their variations in velocity according to Kepler's Laws. A shorter catalogue printed with Description and use of the universal trigonometrical octant (London, 1753)Google Scholar includes ‘Orreries and planetarums, or both, greatly inproved by himself in a new machine called the Cosmotheoron’.

82 It is not clear whether the Cosmotheoron was ever made. In the 1746–7 catalogues Adams appears to be describing an extant device, but nothing further is heard of it after the brief mention in 1753.

83 See note 66. This work passed through five editions and was then completely revised by George the younger and replaced by a new title: Astronomical and geographical essays (London, 1789)Google Scholar. The original title was resurrected by Dudley Adams in 1810, who republished the book from his father's 1766 MS. and called it the ‘30th’ edition, thereby creating the myth (which still persists) that Treatise on the globes ‘went through 30 editions’ and so must have been very popular.

84 Adams's globes are highly prized today, but the fact remains that the form of construction which he advocated made them not only more expensive but also less convenient to use for solution of spherical triangles, an application that Martin considered very important.

85 Martin's attack on Adams and his products occupies the first 31 pages of his Appendix to the description and use of the globes (London, 1766)Google Scholar. Although this work is usually regarded as an appendix to Martin's own Description and use of both the globes, armillary sphere and orrery, and is sometimes bound with the first edition of this (see note 53), its appearance was prompted by Adams's book. Martin at that time was occupied with the Harvard order, and on pp. 33–9 he gives some figures for the performance of ‘a new construction of orreries’ which presumably refer to the Harvard grand orrery.

86 Martin's wide-ranging attack included not only Adams, his products, and his business methods, but also his style of writing. On one point at least Adams took note of his adversary's comments—the word tellarium, which Martin deplored as having no etymological justification, was changed to tellurian in the second and subsequent editions.

87 A catalogue of the very large, capital, and genuine stock in trade of Mr. Benjamin Martin … a bankrupt, deceased, which will be sold by auction by Mr. Herring, on Friday, March 8, 1782 and the four following days (Sunday excepted), Lots 51, 116, 119 (first day) and 62, 122 (fifth day). A copy is in the Royal Microscopical Society's collection, now at the Museum of the History of Science, Oxford. For the background to the collection of catalogues of which this is one item, see Turner, G. L'E., ‘The auction sales of the Earl of Bute's instruments 1793’, Annals of science, xxiii (1967), 213–42.CrossRefGoogle Scholar

88 Precisely when the ‘hybrid’ form of construction was first introduced is not clear, but it remained popular from the late eighteenth century right through to the early twentieth century. Catalogues of Messrs. George Philip & Sons Ltd. of the 1920s include a ‘complete wheelwork orrery’ which was almost identical with those sold by W. and S. Jones (and other firms) in the early nineteenth century. I am indebted to Mr. M. V. Fisher for locating an illustration of this model in the firm's archives. For a typical nineteenth-century ‘hybrid’ model see Inv. 1869–48 (orrery by Newton & Co.) at the Science Museum, London.

89 SirHerschel, John F. W., A treatise on astronomy (London, 1833), p. 287Google Scholar. Herschel objected to orreries because they failed to show the true proportions of the planets and their orbits; but that was never their function. Orreries were essentially mechanical models for demonstrating motions; proportional sizes can be adequately illustrated by static diagrams.

90 Numerous models of the planetarium/tellurian/lunarium type by Adams and by W. and S. Jones can be seen in the principal science museums. Messrs. W. and S. Jones bought the stocks and copyright of Adams, 's Astronomical and geographical essaysGoogle Scholar after his death in 1795.

91 Jones, W., Description and use of a new portable orrery (4th edn., London, 1794; 5th edn., London, 1799)Google Scholar; both editions incorporate price lists of orreries made and sold by W. and S. Jones. The attribution to Martin was dropped in later catalogues, by which time his name presumably had no commercial value.