¹ Unless otherwise specified, all exchange rate quotations are taken from A. Sapori, ed., I Libri di Commercio dei Peruzzi, Milan: Publicazioni … “Studi medievali,” Vol. I, cxxvii plus 571 pp. Other medieval account books edited by Sapori (e.g. the Alberti and Gianfigliazzi accounts) or by De Roover (e.g. the Guillaume Ruyelle accounts) do not have much information on exchange rates.

² The table omits several transactions (pp. 131, 134–36, 359, 365) in which the Florence office credited the London branch for purchases of Flemish currency with sterling. These allowed the London office to reduce slightly its debts on the Continent. As “capital transactions” they were not important, and as exchange rate quotations they were complicated.

³ An attempt was made to determine whether the market was “thin,” by investigating whether large transactions were conducted at rates differing from those on small transactions. There seems to have been no systematic difference between the two.

⁴ A History of Agriculture and Prices (7 vols.; Oxford, 1866–1902). This wine was presumably of relatively low quality, since (as James has shown) the King's butler paid substantially higher prices than this.

⁵ Presumably most of the insurance was self-insurance, but shippers would have to cover the risk somehow.

⁶ We used Tables for Testing Significance in a 2 x 2 Contingency Table, compiled by D. J. Finney, R. Latscha, B. M. Bennett, and P. Hsu (Cambridge [Engl.]: The University Press, 1963).

⁷ Correlations performed on randomly selected economic time series are, on the average, higher than those to be expected on the basis of ordinary significance tests. See Ames, and Reiter, , “Empirical Distributions of Correlation and Autocorrelation Coefficients in Economic Time Series,” Journ. Amer. Stat. Assn., Vol. LVI (1961).Google Scholar

⁸ The main reason for caution in the use of statistical significance tests involving time series data is that economic data change only gradually. Consequently the successive observations in a time series are not independent of each other in the manner assumed by significance tests.

⁹ The restriction to northern Europe made in the preceding paragraph seems to be important. At the time under study, the Florentines were the principal Mediterranean power to trade in northern Europe; but the economic affairs or northern Europe were largely independent of those of Mediterranean Europe, because of the difficulties in communication. Our conclusion will be that Florentine currency was a “key currency,” or roughly the equivalent of “gold and dollars”; and we assume that the market for this currency in northern Europe was largely independent of the market for this currency in the South.

¹⁰ SirCraig, John, The Mint: A History of the London Mint from AD 287 to 1948 (Cambridge [Engl.]: The University Press, 1953)Google Scholar. Shaw, W. A., The. History of Currency, 1252 to 1894 (London, n.d. [1895?]).Google Scholar

¹¹ Shaw lists the following French mint buying prices for one mark of silver: February 1336, 3/12/6; November 1338, 4/12/0; January 1339, 5/0/0; August 1340, 7/0/0; December 1340, 7/10/0; January 1341, 9/4/0; June 1342, 12/10/0. For gold, the buying price, per mark, was: January 1331, 39/0/0; February 1336, 50/0/0; November 1338, 58/0/0; May 1339, 61/10/0; August 1339, 69/0/0.

¹² The Calendar of Entries in the Papal Registers lists 54 different months over the period 1372–1489 in which one Florentine gold florin was worth 12 silver gros tournois. Thus not all exchange rates in the Calendar fluctuate, and not all are stable. This fact lends credence to the assertion that the quotations in the calendars represent rates which may actually have existed in the markets, rather than in bookkeeping conventions or aberrations.

¹³ Compare this range with the range of about 3 per cent suggested by the calculations of silver import and export points given above.

¹⁴ The long-run changes in sterling rates implied by Table 7 are roughly what one would expect from the changes in die metal content of sterling coin. In 1342 one Tower pound of siver made 243 d. of coin and in 1412–65 it made 360 d. In 1344 one Tower pound of gold made 15/0/0 in coin, and after 1412 it made 16/13/4. If the (apparently fixed) sterling rate of 6.0 prevailing after 1412 is computed backward we would therefore estimate the silver rate at 8.00 and the gold rate at 6.67 in 1344; these bracket the actual quotations for the 1340's.

¹⁵ Means are calculated for the entire period 1333–43 (or 1330–43).

¹⁶ It is known, of course, that the King had great difficulties in actually laying his hands on wool while the monopoly existed.

¹⁷ Calendar of Close Rolls, 12 Hen. III, pp. 31, 317; 43 Hen. III, p. 351. Thus Feaveryear is incorrect in stating that the Statute of Stepney in 1299 is the first prohibition on the export of coin (The Pound Sterling [Oxford: Clarendon Press, 1931], p. 3)Google Scholar. See also CCR, 7 Edw. I (1280), p. 519.

The following abbreviations are used in citations in the remainder of this paper: CCR—Calendar of Close Rolls; CCW—Calendar of Chancery Warrants, 1244–1326; CFR—Calendar of Fine Rolls; CLB—Calendar of Letter Books Preserved among the Archives of the Corporation of the City of London (the letter following “CLB” designates the book cited, e.g. CLB-C is Book C); CMI—Calendar of Inquisitions, Miscellaneous. (All these series are publications of the Historical Manuscripts Commission.) Rotuli—Rotuli Parliamentorum, ut et Petitiones et Placita in Parliamento, Anno 6 Edward I—Anno 19 Henry VII (London, 1832); Ruding—Ruding, Rogers, Annals of the Coinage of Great Britain and. Its Dependencies (3rd ed.; London, 1840)Google Scholar; Ruffhead — The Statutes at Large, from Magna Charta to the Twenty-Fifth Year of the Reign of King George the Third, Inclusive, Ruffhead, Owen, ed. (London, 1786)Google Scholar.

¹⁸ Permits only; prohibitions only; both; neither.

¹⁹ In 1304–7, the monasteries were required to make such payments in coin, in contrast to the papal nuncios, who were required to transfer money to Rome by letter of exchange (CPR, 35 Edw. I, p. 514; Rotuli, I, 222, and II, 217; Ruffhead, I, 161). Evidently some change was later made, for tie Peruzzi made payments which were evidently apportum (Sapori, p. 195, gives an example).

²⁰ One may compare the 334 persons who traveled abroad in 1337 on the King's service, and mostly “with the King” in Flanders, with the over 1,200 persons pardoned for crimes from murder on up as reward for military service overseas in 1338–39. Moreover, presumably not all the troops were felons.

²¹ Moreover, the large number of protections going to non-officials traveling in Gascony and Aquitaine in 1337–38 seems to have been associated with naval convoys of ships going in those years (and only in those years) to these destinations.

²² That is, the number of pennies made from a Tower pound of silver was increased from 243 to 270—about 10 per cent.

²³ The word “apparently” is used advisedly. In other years the qualification “except by license” appears in the calendars, and, as Table 13 shows, licenses to take coin abroad for travel expenses were certainly issued. Either the phrase escaped the notice of the editors of the Calendars, or else it was taken for granted by royal officials.

²⁴ The violations include illegal exports of precious metals and also failures to deliver “false” and clipped coin to the Exchange for minting. (False coin was non- English coin which resembled the penny but contained less silver. In the 1340's it was described as Lussheborne—Luxemburg—and at other dates it had other designations.

²⁵ The page references for given years are: 1333, p. 347; 1337, pp. 6, 56, 60, 61; 1338, pp. 81–82; 1339, pp. 113, 135; 1341, pp. 216, 252; 1342, p. 309; 1343, pp. 318–19; 1344, pp. 357, 365; 1345, pp. 415, 443, 448; 1346, pp. 467, 474, 485; 1347, pp. 17, 18, 29, 30; 1348, pp. 67, 68.

²⁶ “Two actions” means “action on two occasions,” for a single action often involved instructions to several local officials.

²⁷ There is no inherent absurdity in the proposition that the two offenses might occur simultaneously. In the twenty years since 1945, there have been many occasions when a traveler from the United States could go to Canada with one silver dollar, purchase eleven Canadian dimes with it, and return to the United States to purchase $1.10 in U. S. goods with them. The twentieth century traveler would have committed no crime, while the fourteenth century traveler would have placed his body and goods at the King's pleasure. The transaction described would have been profitable whether or not Canada was adding to its reserves of gold and dollars at the time.

²⁸ The converse is not true. It may be profitable to do something if one has a license but not profitable to take the risk of capture and punishment if the act is illegal.

²⁹ Miskimin, Harry A., Money, Prices and Foreign Exchange in Fourteenth Century France (New Haven: Yale University Press, 1963)Google Scholar, Appendix B.

³⁰ The only reference to the prohibition on the use of letters of exchange by the alien priories relates, as noted in an earlier footnote, to 1304–7; by the 1330's they seem to have been using letters.

³¹ Craig, pp. 66–69; Ruding, I, 217–19; Ramsay (cited in Table 9), I, 339; Hughes, , Crump, , and Johnson, , “The Debasement of the Coinage under Edward III,” Economic Journal, VII (1897), 197Google Scholar.

³² Rotuli, II, 138.

³³ CGR, 16 Edw. III, p. 415.

³⁴ Ruding, p. 213, states that in 1340 two marks of silver were to be imported per sack of wool exported. I have not found any other source for this statement. In 1343, silver imports were to be one third of the value of exports (Rotuli, II, 138). This provision, in some form, lasted until 1348, when the Flemish instituted a similar regulation, thereby forcing the English to abandon it.