¹ In 1792 Irving was the Inspector General of Imports and Exports of Great Britain, and his statements were contained in a reply to an inquiry into “the Number of American-built vessels” produced each year from the Commissioners of the Land Revenues who had been appointed to look into the state of Great Britain's timber supplies. His letter, dated 7 January 1792, was printed as an appendix to their Eleventh Report, and the information he gave them was incorporated into it, including his warning about the tonnage. See The Journal of the House of Commons, XLVII, 272–73, 356–57. Irving says that this information “is taken from the Books of my Office, when I was Inspector General of the Imports and Exports of North America.” What must have been his original table—it is dated Boston, 11 May 1771 —was published in Holroyd, John Baker, First Earl of Sheffield, Observations on the Commerce of the American States (6th rev. ed.; London: J. Debrett, 1784), p. 96. Appended to the table before Irving's signature is essentially the same assertion that he made in 1792: The figures are in error by one third. North America, in Irving's administrative context, included all the British colonies on the mainland from Canada to East and West Florida, the Island of St. John's (Prince Edward Island), Newfoundland, Bermuda, and the Bahamas. For more about Irving, see Papers Printed by Order of the House of Commons from the Year 1731 to 1801, XLII, Nos. 267, 747Google Scholar; McCusker, John J., “The American Invasion of Nassau in the Bahamas,” The American Neptune, XXV (06 1965), 214Google Scholar; and SirClark, George N., Guide to English Commercial Statistics, 1696–1782, Royal Historical Society Guides and Handbooks, No. 1 (London: Offices of the Royal Historical Society, 1938), pp. 32, 38.Google Scholar

² The Earl of Sheffield, on literally the last page of his Observations on the Commerce of the American States (both the “New Edition, Much Enlarged” [3rd ed.] and the 6th ed. of 1784), announced categorically that “the Tonnage given in the. Register, is, upon an average, about a third less than the real measurement.” This follows immediately the tables in his appendix and presumably is meant to be applied to the tonnage figures in them including, therefore, Table VII, which has reference to all British and colonial ships entering and clearing North American ports from 5 January 1770 to 5 January 1771Google Scholar. David Macpherson, concerning a table giving “an account of vessels belonging to the ports of Great Britain” in 1760, remarks: “The real tunnage may in general be reckoned full fifty percent above the reputed;” Annals of Commerce, Manufacture, Fisheries and Navigation … of the British Empire and Other Countries, From the Earliest Accounts to … 1801 (4 vols.; London: Nichols & Son, et al., 1805), III, 340Google Scholar. He later repeats this comment when he reprints Irving's table ibid., p. 571.

³ See, for instance, Series Z 56–75, “Number and Tonnage Capacity of Ships Outward and Inward Bound, by Destination and Origin: 1714 to 1772,” prepared by Harper, Lawrence A. for U.S. Bureau of the Census, Historical Statistics of the United States, Colonial Times to 1957 (Washington: Government Printing Office, 1960), pp. 759–60, and Harper's discussion of these series (p. 745)Google Scholar. See also Jensen, Arthur L[ouis], The Maritime Commerce of Colonial Philadelphia (Madison, Wis.: State Historical Society of Wisconsin, 1963Google Scholar), passim. Compare Heers, Jacques, Gécle: ActivitS dconomique et problimes sociaux ([Paris]: S.E.V.P.E.N., 1961), pp. 453–54, andGoogle ScholarJeannin, Pierre, “Le Tonnage des Navires Utilises dans la Baltique de 1550 à 1640 d'apres les Sources Prussiennes,” in Le Navire et I'Economie Maritime du Nord de YEurope de Moyen-Age au XV1W siècle, Mollat, Michel, ed., Travaux du Troisième Colloque International d'histoire maritime … 1958 ([Paris]:] S.E.V.P.E;N., 1960), pp. 45–71Google Scholar. Several recent studies offer additional testimony to the growing awareness of the necessity of establishing precisely what tonnage measure involved: Davis, Ralph, “Tonnage Measurement and Its Meaning,” Appendix C of “The Organization and Finance of the English Shipping Industry in the Late Seventeenth Century” (unpublished doctoral diesis, University of London, 1955), pp. 471–81—his conclusions, but not the appendix, appear in his bookGoogle ScholarThe Rise of the English Shipping Industry: In The Seventeenth and Eighteenth Century (London: Macmillan, 1962) and at least one reviewer took strong issue with him (W[illiam] Salisbury inGoogle ScholarThe Mariner's Mirror, XLIX [08. 1963], 235–36)Google Scholar; Gille, Paul, “Jouge et Tonnage des Navires,” in Le Navire et I'Economie Maritime du XVe au XVWe siècles, Mollat, Michel, ed., Travaux du [Premier] Colloque d'histoire Maritime … 1956 ([Paris]: S.E.V.P.E.N., 1957), pp. 85–100Google Scholar; Lane, Frederic C., “Tonnages, Medieval and Modern,” The Economic History Review, 2d. ser., XVII (12. 1964), 213–33Google Scholar; Salisbury, William, “Early Tonnage Measurement in England,” “Rules for Ships Built for, and Hired by, the Navy,” and “Early Tonnage Measurement in England,” Part III,“H. M. Customs and Statutory Rules,” The Mariner's Mirror LII (02., May, Nov. 1966), 41–51, 173–80, 329–40Google Scholar.

⁴ “An Act for Preventing Frauds and Regulating Abuses in the Plantation Trade;” 7 & 8 Wm. IlI c. 22. Section xvi included the form of the oath to be sworn and this became the form of the actual certificate. All references to English laws enacted before 1714 are taken from The Statutes of the Realm, [ed. Ludens, A., et al.] (11 vols. in 12; London: Record Commission, 1810-1828; reprinted London: Dawson, 1963)Google Scholar.

⁵ “Tonnage” in the eighteenth century was a stylized measure of the cargo capacity of a ship. It was based in theory on the number of wine casks which a ship could carry in her hold (the standard medieval wine barrel was the French tun). It thus bore no relation to weight and was irrelevant, directly at least, to all but a trade in wines. The expanding commercial community of post-Elizabethan England struggled with the problem until the nineteenth century, when the introduction of the standard “ton” of 100 cubic feet and the concepts of gross and net registered tonnage brought a new order to sea-borne commerce. See Driel, A. Van, Tonnage Measurement: Historical and Critical Essay (The Hague: Government Printing Office, 1925), pp. 6 ffGoogle Scholar.

⁶ The 1786 law Irving seems to have had in mind was the “Act for the further Increase and Encouragement of Shipping and Navigation;” 26 Geo. Ill c. 60. See section fourteen for the passage quoted. All references to British laws of 1714 and after are taken from The Statutes at Large of England and of Great Britain, [ed. Tomlins, Thomas Edlyne and Raithby, John] (10 vols.; London: George Eyre and Andrew Strahan, 1811)Google Scholar. The rule of 1773 was set out in “An Act for the better ascertaining the Tonnage and Burthen of Ships and Vessels;” 13 Geo. Ill c. 74. Algebraically expressed this formula reads

where “L” is the length of the keel for tonnage measured “on a straight Line along the Rabbet of the Keel of the Ship from the Back of the Main Stem Post to a perpendicular Line from the Forepart of the Main Stem under the Bowsprit” and “B” is the breadth to “be taken from the Outside of the outside Plank, in the broadest Place of the Ship.” It was repeated in 14 Geo. Ill c. 56 xlvi. For a discussion of this formula see Lymon, John, “Register Tonnage and Its Measurement,” The American Neptune, V (06 1945), 226–30; andGoogle ScholarSalisbury, , “Rules for Ships Built for, and Hired by, the Navy,” pp. 173–77. The most reasonable explanation yet encountered by this writer of the use of “94” as the divisor is that offered byGoogle ScholarSalisbury, in The Mariner's Mirror, XLV (02. 1959), 83–84Google Scholar.

⁷ “An Act for granting to theire Majesties severall Rates and Duties upon Tun-nage of Shipps and Vessels,” 5 & 6 W. & M. c. 20. viii; “An Act … for ascertaining the Admeasurement of the Tunnage of Ships,” 6 & 7 W. & M. c. 12. ix (to become effective 1 June 1695); “An Act for making a.convenient Dock or Bason at Leverpoole,” 8 Annec. 8 [12]. iv; “An Act for Preventing Frauds and Abuses,” 6 Geo. I c. 21. xxxiii. The act of 8 Anne was repeated by a private act of 3 Geo. I, and by 11 Geo. II c. 32, and 17 Geo. II c. 20. See also 17 Geo. Ill c. 7. Expressed algebraically the formula of 1695 appears thus:

“L” equals the length of the keel for tonnage measured “so much as she treads on the ground” and “B” equals the ship's breadth “to be taken within board by the midship beam from plank to plank.” The 1694 formula had required the actual measurement of the depth. The result was called measured tonnage or shipwright's tonnage; see Bushnell, Edward, The Compleat Ship-Wright (4th ed.; London: William Fisher, 1678), p. 41Google Scholar.

This formula was not entirely satisfactory for several reasons—as the 1773 change suggests—but it was the basic form of the formula until the change. To demonstrate this it is sufficient simply to compare what contemporaries called the tonnage of a vessel with the results obtained by calculating the ship's tonnage oneself, using this formula. In our immediate case Table I establishes this adequately. Even one of its most articulate critics, Sutherland, William, referred to it as “the customary Method” of ascertaining tonnage; The Ship-builders Assistant: Or, Some Essays Towards Compleating the Art of Marine Architecture (London: R. Mount, A. Bell and R. Smith, 1711), p. 94. CompareGoogle ScholarSirSteele, Richard and Gillmore, Joseph, An Account of the Fish-pool (London: H. Meere, 1718), p. 16. The builders of at least one of the ships used in this study (the Black Prince), Wharton and Humphreys of Philadelphia, set down this formula in their “Notebook of Ships” in the early 1720's to be employed when measuring merchantmen; The Joshua Humphreys Papers, MS Collection No. 306, in The Historical Society of Pennsylvania, Philadelphia [hereafter, HSP]. We have furthermore been told by the eminent historian of the Royal Navy, Sir William Laird Clowes, that for eighteenth-century naval vessels this was “the more commonly-used method,” the explicit instructions of the 1719 Establishment notwithstandingGoogle Scholar; The Royal Navy: A History from the Earliest Times to the Present (6 vols.; London: S. Low, Marston and Company, Ltd. 1897-1903), III, 9Google Scholar, n. 1. Compare my position in “The Tonnage of the Continental Ship Alfred,” The Pennsylvania Magazine of History and Biography [hereafter, P.M.H.B.], XC (04. 1966), 230, nn. 12, and 232, andGoogle Scholar“The Continental Ship Alfred,” Nautical Research Journal, XIII (Autumn 1965), 60, n. 19Google Scholar.

⁸ Irving said that the reason for this reduction in tonnage was “in order to evade the Payment of Light House Money, and various Port Charges collected upon Tonnage.” Sheffield echoed this explanation and there appears to be sufficient prima facie evidence to support it; see Harper, Lawrence A., The English Navigation Laws: A Seventeenth Century Experiment in Social Engineering (New York: Columbia University Press, 1939), p. 423. CompareGoogle ScholarChaplin, W[illiam] R., “The History of Flat Holm Lighthouse,” The American Neptune, XX (01. 1960), 12–24. The abundance of colonial port duties which were prorated on a ship's tonnage and the spectacular growth in the number of lighthouses in North America in the eighteenth century both provide convincing though circumstantial evidence to support such a contention. Port duties very often took the form of that of Barbados quoted below in note 17Google Scholar. See Johnson, Emory R., “American Commerce to 1789,” Part I of Johnson, Emory R., et al., History of Domestic and Foreign Commerce of the United States (2 vols.; Washington, D.C.: Carnegie Institution of Washington, 1915), I, 63–64. The first lighthouse in North America was erected at Boston in 1716, and the town of Boston was anxious to keep for itself the “Profits and Incomes thereof” which amounted to “one penny per ton inwards and the same rate outwards;”Google ScholarStevenson, D[avid] Alan, The World's Lighthouses Before 1820 (London: Oxford University Press, 1959), p. 173Google Scholar. They had multiplied ten-fold in North America by 1775; ibid., pp. 171–83. Even the 1709 law of Queen Anne mentioned in note 7 (above) had as its purpose the collection of a duty, based on a ship's tonnage, to pay for the improvements of the harbor of Liverpool.

⁹ See the references cited in the notes to Table I.

¹⁰ See Clark, , Guide to English Commercial Statistics, p. 50Google Scholar; Jarvis, Rupert C., “Liverpool Statutory Registers of British Merchant Ships,” Transactions of the Historic Society of Lancashire and Cheshire, CV (1953), pp. 109, 111Google Scholar.

¹¹ The Boston “Register of all such Ships and Vessels Concerning the Owners and Property whereof Proofs hath been made upon Oath … ” is in the Massachusetts Archives, Vol. VII, pp. 85–523, Boston, Massachusetts. It was thoroughly analyzed in Bernard, and Bailyn, Lotte, Massachusetts Shipping, 1697–1714: A Statistical Study (Cambridge, Mass.: Belknap Press of Harvard University Press, 1959). The “Ship Register Books of the Province of Pennsylvania, 1722–1776,” are MS Collection No. 594, HSP. They have been published in a digested but essentially complete form inGoogle Scholar“Ship Registers for the Port of Philadelphia, 1726–1775,” P.M.H.B., XXIII (1899), 254–64, 370–85, 498–515; XXIV (1900), 108–115, 212–23, 348–66, 500–19; XXV (1901), 118–31, 266–81, 400–16, 560–74; XXVI (1902), 126–43, 280–89, 390–400, 470–75; XXVII (1903), 94–107, 238–47, 346–70, 482–98; and XXVIII (1904), 84–100, 218–35, 346–74, 470–507. The Liverpool and Campbeltown registers were still kept at these ports as of 1961. For Liverpool, registry information is available from 1739 to the end of the colonial period and beyond; for Campbeltown, from 1763Google Scholar. See Crick, B[ernard] R. and Alman, Miriam, A Guide to Manuscripts Relating to America in Great Britain and Ireland ([London]: Oxford University Press, 1961), pp. 201, 203; andGoogle ScholarJarvis, , “Liverpool Statutory Registers,” pp. 107–22Google Scholar.

¹² The first naval officers were appointed in the last quarter of the seventeenth century and the earliest accounts date from this period; Beer, George Louis, The Old Colonial System, 1660–1754 (2 vols.; New York: The Macmillan Company, 1913), I, 267–72. Still extant are some returns from almost every colonial port but serious gaps exist in a number of series. Those which have survived are preserved in the Colonial Officer Papers in the Public Record Office, LondonGoogle Scholar; see Andrews, Charles M., Guide to the Materials for American History, to 1783, in the Public Record Office of Great Britain (2 vols.; Washington, D.C.: Carnegie Institution of Washington, 1912), I, 157, 160, 165, 167, 171, 174, 175, 178, 185, 186, 187, 189, 191, 194, 200, 201, 215. CompareGoogle ScholarCox, John H., “Compilations of Colonial Imports and Exports on Film,” The Journal of Documentary Reproduction, II (12. 1939), 198–201. They can sometimes be supplemented by extracts from the returns published by contemporary newspapers but caution is required in using the extracts, as they were not always carefully compiledGoogle Scholar; see Morison, Samuel Eliot, “The Commerce of Boston on the Eve of the Revolution,” Proceedings of the American Antiquarian Society, new ser., XXXII (1922), 43. Some of these returns have been published in one form or anotherGoogle Scholar; see [Topley, Harriet Silvester, ed.], “Early Coastwise and Foreign Shipping of Salem: A Record of the Entries and Clearances of the Port of Salem, 1750–1769,” The Essex Institute Historical Collections, LXII (1926), 193–200, 305–20; LXIII (1927), 49–64, 145–60, 349–64; LXVII (1931), 281–88, 409–24; LXVIII (1932), 49–64, 241–56, 337–52; and LXIX (1933), 49–64, 155–98. See also Vaughn W. Brown, Shipping in the Port of Annapolis, 1748–1775, Sea Power Monograph No. 1 (Annapolis, Md. [United States Naval Institute]: 1965). Beginning with the latest volume, a yearly summary of these reports will be published in theGoogle ScholarCalendar of State Papers, Colonial Series: America and West Indies, ed. Sainsbury, W. Noel, et al. (43 vols. [to date]; London: H. M. Stationery Office, 1880–1963), XLIII [1737], 322–25. The obvious potential for a very profitable employment of computerized data processing techniques to sort and process this mass of material has yet to be acted uponGoogle Scholar.

¹³ In a strictly statistical sense the universe is the whole of the Philadelphia register, and the conclusions might be seen as applicable to all of the ships included therein.

¹⁴ The mean of column (6) is 0.316, with a standard deviation of 0.042. The standard error of the mean is 0.0211.

¹⁵ Compare the Bailyns', Massachusetts Shipping, pp. 8–9Google Scholar.

¹⁶ On the basis of this sample of five we can expect to find the mean correction factor for 95 per cent of the samples of the merchant ships registered at Philadelphia during these thirty-five years to fall within 0.375 and 0.275.

¹⁷ I will, however, without further comment quote from a letter written by one Colonel Robert Rich of Barbados, dated 31 May 1670, to Ogilby, John and printed by him in his America: Being the Latest and Most Accurate Description of the New World (London: Printed by the author, 1671), p. 380. It reads, in part, as follows:CrossRefGoogle Scholar The number of Vessels which come hither [to Barbados] to Trade in one year is found upon search to be about two hundred of all sorts, (some years more, some less) as Ketches, Sloops, Barques, &c. containing in Burthen fifteen thousand five hundred and five Tun, according as they were nere Entered, which is at least a third part less than their true Burthen, by reason every Ship pays one pound of [gun]Powder per Tun, the means ordain'd by this country for storing the Magazine …

¹⁸ Irving suggested this procedure in the note to his original table (as quoted by Sheffield, , Observations on the Commerce of the American States, p. 961Google Scholar) and adopted it himself in the table which was presented to the Commons in 1792. To accomplish such a correction one need only increase the registered tonnage figure by 50 per cent of itself. See, for instance, Bishop, J[ohn] Leander, A History of American Manufactures From 1608 to 1860 (2 vols.; Philadelphia: E. Young & Co., 1864), I, 71Google Scholar; Johnson, , “American Commerce to 1789,” I, 91; andGoogle ScholarHutchins, John G. B., The American Maritime Industries and Public Policy, 1789–1914: An Economic History, Harvard Economic Studies LXXI (Cambridge, Mass.: Harvard University Press, 1941), p. 150, n. 71 and p. 152, n. 91. Hutchins' confusing (and baseless) indecision as to which tonnages to increase and which to leave unaltered is alone sufficient justification for this present studyGoogle Scholar.

¹⁹ With the enactment of the British Registry Law of 1786 (26 Geo. Ill c. 60), the practice of discounting tonnage ended, as Irving implicitly maintained. See Chaplin, “The History of the Flat Holm Lighthouse,” pp. 20 and 24. In the United States, a new formula and federal enforcement inaugurated a new era after 1789: 1 Stat. 11 (1 Sept. 1789). See Peters, Richard et al, eds., The Public Statutes at Large of America From the Organization of the Government (Boston: Little, Brown, 1845-1873), I, 55–56Google Scholar.