University of Virginia Library

Copyright Records and the Bibliographer
by
G. Thomas Tanselle [*]

In a country which provides copyright protection by statute or which requires the deposit of works for the benefit of one or more of its libraries, the operation of the law will usually produce a body of documents relating to the literary and artistic output of that country. The exact nature of such material will vary according to the requirements of the statute, but one may expect it to contain information of bibliographical significance which would be difficult (and in some cases impossible) to locate elsewhere. For example, if the law defines the term of copyright for any work as dating from its publication, the records may include publication dates; if there is a manufacturing clause in the law, certain facts about the printing and binding of individual books may be given; if there is provision for the deposit of copies, the recorded deposit dates will furnish proof of the existence of particular books. Thus the bibliographer who knows what copyright records survive for the period and country with which he is concerned, and who possesses enough historical understanding of the copyright law to interpret them, has at his command a valuable tool for research.

In the case of England and the United States, not only are the extant copyright records extensive; in addition, because of the unusual situation in which two countries with important literatures speak the same language, the copyright laws of each country have had an effect on the original publication of the literature of the other, so that bibliographers of authors writing in English must be aware of the

copyright records of both countries. The English copyright law required registration of copyrighted works between 1710 and 1912, and there are also some records of registered books before and after this period. The United States, from the time of its first federal copyright law in 1790, has always included a provision for registration in its copyright legislation. For obvious reasons these records, even when they survive intact, do not constitute national bibliographies — some works may be copyrighted but never published, and many others may be published but never copyrighted. They do, however, provide a preliminary basis for a record of the two nations' literary output.

Whenever a bibliographer finds information of interest to him in copyright records, he is making a secondary use of those records; he is finding them helpful in ways other than those for which they were designed, just as he may profitably consult a publisher's files which are no longer useful to the publisher in carrying on his business. Quite understandably and properly, the officials in charge of such records have normally been more concerned with maintaining whatever records are required for current purposes than with servicing archives of older material. In the United States, such men as William Elliot and Thorvald Solberg did recognize some of the bibliographical uses of the copyright records, but their hopes for a national bibliography based on them were not realized until 1891, and then only in part — a story which has been admirably set forth by Joseph W. Rogers in U. S. National Bibliography and the Copyright Law (1960). Since that time such men as Verner W. Clapp, former Chief Assistant Librarian of Congress, Joseph W. Rogers of the Copyright Office, and Frederick R. Goff of the Library's Rare Book Division have been particularly concerned with the historical importance of the old records. But the general indifference to this material, on the part of officials and scholars alike, has not yet disappeared, and bibliographers have remained peculiarly unaware of its potentialities.

Although an enormous literature of copyright exists,[1] practically

nothing has been written about the secondary use of copyright material by literary scholars and bibliographers. Walter Pforzheimer, in his well-known essay on "Copyright and Scholarship" in the English Institute Annual 1940 (1941), outlined the essentials of English and American copyright history for the use of scholars (pp. 164-99) but said little about the ways in which the records could be employed. In connection with the American records, only two writers have gone very far in discussing the specifically bibliographical significance of such material. On 30 December 1937, Martin A. Roberts of the Library of Congress read a paper before a joint session of the American Historical Association and the Bibliographical Society of America describing the pre-1870 records and urging their publication as a "national heritage."[2] Seven years later Ruth Shaw Leonard intensively studied the Massachusetts records of the period 1800-09 for her Master's thesis at Columbia and summed up her findings in 1946 in an essay entitled "The Bibliographical Importance of Copyright Records."[3]

At that time the American copyright records had been used by only a few bibliographers. Charles Evans, in the 1910's, had taken advantage of some of the early district court records to add titles and deposit dates to his American Bibliography for the last decade of his coverage (1790-1800). And Miss Leonard, with Harry C. Bentley, had conscientiously read through the records in preparing the Bibliography of Works on Accounting by American Authors (2 vols., 1934-35) — which also contained an introductory chapter entitled "Copyright Laws and Administration — Their Significance to Bibliographers" (I, xi-xxi). In 1940 Louis C. Karpinski published his Bibliography of Mathematical Works Printed in America through 1850, which made effective use of the copyright records to establish the publication of many titles no longer extant, just as Lyle Wright's American Fiction, 1744-1850 (1939) included some otherwise unknown titles discovered in a search through the collection of deposited title pages, and Marcus McCorison's Vermont Imprints 1778-1820 (1963) contained some

unlocated titles taken from district court copyright records. Sidney Kramer, one of the pioneers in the use of copyright records for bibliographical purposes, included deposit dates in his A History of Stone & Kimball and Herbert S. Stone & Company (1939), and this practice of giving deposit dates in descriptive bibliographies has been used occasionally since then, notably in Jacob Blanck's Bibliography of American Literature (1955- ). For a more recent period, the published copyright catalogues were diligently searched by Wynot Irish for The Modern American Muse (1950), a list of American poetry between 1900 and 1925. Beyond these few works, American copyright records have rarely been drawn upon by bibliographers.

The English records have been used more extensively. The registers of the Stationers' Company between 1554 and 1708 have been regularly employed for decades by bibliographers of Elizabethan and Jacobean literature, and such literary and bibliographical scholars as A. W. Pollard, W. W. Greg, R. C. Bald, Giles Dawson, C. J. Sisson, and Leo Kirschbaum have turned their attention to the whole question of copyright in England in the sixteenth and seventeenth centuries. Pollard and Redgrave, of course, included references to the Stationers' Company registers in the Short Title Catalogue (1926) of pre-1640 books, and both Greg and W. A. Jackson edited parts of the Stationers' court records (1930, 1957). For the eighteenth, nineteenth, and twentieth centuries, the English copyright materials are perhaps lesser known; but the increasing recognition of the textual importance of overseas editions of both English and American authors is drawing more attention to the copyright records of both countries, especially to the way in which the copyright laws of each affected the literary properties of the residents of the other. I. R. Brussel, in his pioneer bibliographical work on trans-Atlantic literary publications, the two-volume Anglo-American First Editions (1935-36), understood the indispensability of copyright records, and Graham Pollard contributed to the first volume a thoughtful discussion of the bibliographical implications of the Anglo-American copyright situation. More recently Simon Nowell-Smith has gone into the question of nineteenth-century international copyright from a bibliographer's point of view.[4] And James J. Fuld's The Book of World-Famous Music (1966) illustrates some of the bibliographical uses of copyright data on a world-wide scale.

Copyright records have thus not been totally ignored by bibliographers, but they have not been drawn upon as frequently as one

might expect, considering the obviousness of the uses to which they can be put. They can help to establish publication dates; they are a source of information about anonymous and pseudonymous works; they can supply publication details for individual author-bibliographies and titles for comprehensive imprint lists. It is not too much to say that a check of the relevant copyright records should be a routine part of any bibliographer's investigations; and no bibliography can be considered definitive or thorough which does not utilize them.

Their precise bibliographical usefulness, however, depends on the provisions of the copyright laws at any particular time and place; in order to know how to interpret the records — or, indeed, what records to look for — the bibliographer must have some understanding of the copyright laws of the country and period with which he is concerned. It is with this in mind that I have drawn together a few suggestions for extracting from such records the bibliographical information they can yield. Since the American records are perhaps more complex and certainly less well-known than the English, the emphasis will be on them; and they can most conveniently be taken up in four groups: 1783-90, 1790-1870, 1870-1909, 1909-present. Following these four discussions are some briefer notes on the English records and some general considerations.

I. 1783-1790

Although the Massachusetts Bay Colony passed a law on 15 May 1672 prohibiting any printer from printing more copies of a book than its author had agreed to,[5] and although occasional ad hoc copyright acts were passed in the colonies upon the petition of particular individuals, there was no general copyright act in any of the states until Connecticut instituted one in January 1783. Massachusetts and Maryland followed in March and April, and on 2 May 1783 the Continental Congress passed a resolution recommending that each of the states work out a copyright law. Between that time and 1790, when the first federal law was enacted, all the other states except Delaware passed copyright laws, in several cases through the energetic sponsorship of Noah Webster.[6]

Of particular interest to bibliographers in these twelve laws are any provisions they contain for registration or deposit. Whenever copyrighted works were to be registered, some sort of ledger or record book for that purpose must have existed at one time and might be extant today; whenever copyrighted works were to be deposited, the deposit copies might still survive, or at least a record of the deposits to prove that certain books once existed. The list below sums up these features of the state laws:[7]

                         

	Date of Law	Term of Copyright (years)	Renewal Period (years)	Registration Required	Deposit Required
Connecticut	8 Jan. 1783	14	14	yes	no
Massachusetts	17 March 1783	21	--	no	yes
Maryland	21 April 1783	14	14	yes	no
New Jersey	27 May 1783	14	14	yes	no
New Hampshire	7 Nov. 1783	20	--	no	no
Rhode Island	Dec. 1783	21	--	no	no
Pennsylvania	15 March 1784	14	14	yes	no
South Carolina	26 March 1784	14	14	yes	no
Virginia	Oct. 1785	21	--	yes	no
North Carolina	19 Nov. 1785	14	--	yes	yes
Georgia	3 Feb. 1786	14	14	yes	no
New York	29 April 1786	14	14	yes	no

In addition, seven states (Conn., Mass., N. H., R. I., N. C., Ga., N. Y.) specifically included a provision extending copyright privileges to citizens of other states only if the states involved had similar laws. And in five states (Conn., S. C., N. C., Ga., N. Y.), an author who did not allow a sufficient quantity of his books to be printed or charged an excessive price for them could be required to have more printed or lower the price; if he failed to comply, the state could authorize the complaining party to print additional copies.

It can be seen from the list that nine states made the registration of titles a requirement for copyright. In six (Conn., N. J., S. C., N. C., Ga., N. Y.), registration was to be made with the Secretary of State; in Maryland it was to be handled by the "clerk of the general court," in

Pennsylvania by the "prothonotary's office in the city of Philadelphia," and in Virginia by the "clerk of the council." Two of the laws (Md., S. C.) specifically mentioned the "register" into which the titles were to be entered. Two of the states (Md., Pa.) with registration requirements, however, had the further clause that the law was not to go into effect until such time as "all and every" of the states had enacted similar laws; since Delaware never passed a copyright law, the state copyright laws did not go into effect in Maryland or Pennsylvania either, and the number of possible record books is therefore presumably reduced by two (although, as it turns out, copyright entries were made in both states). As for the deposit of copies, only two states had such a clause — Massachusetts and North Carolina. In Massachusetts, two copies of each work were to be presented to "the library of the University of Cambridge [Harvard], for the use of the said university"; in North Carolina (the only state to require both registration and deposit), one copy of each work was to go — before publication — to the Secretary of State "for the use of the executive of the State."

Of the nine possible sets of records and the two groups of deposit copies, the records for Connecticut, Maryland, and South Carolina have been located among state archives and the copies for Massachusetts have been traced in the Harvard library:

• Connecticut, 1783-89: Connecticut Archives (Colleges and Schools), 1st ser., II, 154-55, 159-61, 165-68. Published in The Public Records of the State of Connecticut, ed. Leonard W. Labaree, V (1943), 245-46, 459; VII (1948), 87. See also James Hammond Trumbull, List of Books Printed in Connecticut, 1709-1800 (1904), items 308, 954, 998. [5 entries]
• Maryland, 1786: Maryland Provincial Court Deeds, Liber T. B. H. No. 1, f. 532. Published in Irving Lowens, "Copyright and Andrew Law," PBSA, LIII (1959), 152-53. [8 entries]
• Massachusetts, 1783-90: Deposit copies in the Harvard Library. Published in Earle E. Coleman, "Copyright Deposit at Harvard," Harvard Library Bulletin, X (1956), 135-41. [11 entries]
• South Carolina, 1785-88: Georgia Grants Ledger, pp. 1-4, in the office of the Secretary of State. Published in "Copyrights and Patents Granted by South Carolina," South Carolina Historical and Genealogical Magazine, IX (1908), 56-58. [6 entries for books]

The located Connecticut records show only three entries (for Robert Ross, Joel Barlow, and William Blodget) under the 1783 act; but Trumbull, in his list of Connecticut imprints, supplies the registration dates of two more (for Andrew Law and John Lewis).[8] In addition, the Connecticut General Assembly granted Andrew Law a five-year copyright ("patent") for certain pieces of music in October 1781, fourteen months before the state passed its general law.[9] The eight Maryland entries of 7 January 1786 (for works by four men — Webster, Dwight, Barlow, and Law) were published in an article by Irving Lowens, whose thorough research has impressively demonstrated the bibliographical value of copyright records. The other most important investigation into pre-1790 copyright is Earle Coleman's essay, which records eleven Harvard deposit copies with the deposit dates for each (works by Barlow, Belknap, Billings, Bingham, French, Law, Webster, and Wood); undoubtedly other volumes were deposited which cannot now be located. The six South Carolina entries, published as long ago as 1908, are for books by David Ramsay, Henry Osborne, Noah Webster, Robert Squibb, Nicolas Pike, and John F. Grimke.

Another aspect of copyright legislation in which bibliographers are interested is whether or not printed notices of copyright were required in copyrighted books. Only one of the state laws — in Pennsylvania — had such a clause; but, since Pennsylvania's law was not to become effective until all the other states had copyright laws, one should not expect to find printed notices even in Pennsylvania books. Nevertheless, such notices do appear, and a list of books with these notices forms at least a partial record of Pennsylvania pre-1790 copyrights. Charles Evans, working on the sixth and seventh volumes of his American Bibliography (1910, 1912), located fourteen works with these printed notices and recorded the copyright dates given in twelve of them. These references constitute in effect another published copyright record:

Of the two books for which Evans did not give the date of copyright entry, one, Henry Wharton's Letter to the Roman Catholics (Evans 18883), contains only the statement (at the foot of the last page), "Entered according to Act of Assembly";[10] but the other, the second edition of the opening volume of the Transactions of the American Philosophical Society (Evans 21651), does carry the entry date of 29 April 1789. When Roger Bristol's list of additions to Evans is completed, it is possible that more books with Pennsylvania copyright notices can be traced.

Apparently none of the copyright records of the other five states with requirements for registration are in existence, for officials in those states have not been able to turn up any copyright material for this period.[11] It is occasionally possible, however, to learn something of the copyright registrations in these states through secondary documents — such as the 1842 letter from Stephen Dodd to George Hood, in the Boston Public Library, which gives the Pennsylvania (1784) and New York (1786) registration dates for Andrew Law's music books.[12] Judging from the located records, the number of registrations in each state was probably quite small, and the information could have been entered in ledgers mainly devoted to something else — just as the South Carolina entries are found on the opening pages of a volume labeled "Georgia Grants," recording plats of land granted by Georgia authorities. Such records are naturally difficult to locate, and it is not impossible that more of them will eventually come to light. In the meantime, bibliographers must be content with copyright information concerning fewer than forty American books before 1790.

II. 1790-1870

Since the United States Constitution (17 September 1787) gave Congress the power to grant copyrights, several authors in 1789 petitioned the federal government (rather than the states) for copyrights to their works — David Ramsay on 15 April, Jedidiah Morse on 12 May, Nicholas Pike on 8 June, and Hannah Adams on 22 July; and in granting copyrights to David Ramsay on 20 April 1789 for two of his books, Congress was making its first use of the copyright provision of the Constitution.[13] Noah Webster, at the same time, was drafting A Bill to Promote the Progress of Science and Useful Arts, printed in

New York on 23 June 1789, which led to the passage, on 31 May 1790, of the first federal copyright law in the United States (1 Stat. 124).

This law required three steps for the securing of a copyright. First, a printed copy of the title page of any work proposed for copyright was to be deposited before publication with the clerk of the district court for the district in which the author (or copyright proprietor) lived, and the clerk was to record this deposit in a ledger which he kept for that purpose. Second, the text of the clerk's registration was to be inserted, within two months, in at least one United States newspaper, where it was to run for a period of four weeks. Third, within six months of the publication of the work, a copy was to be deposited in the office of the Secretary of State. The term of copyright was fourteen years, renewable for another fourteen years if this same procedure were repeated within six months of the expiration of the first term.

Although later revisions of the law before 1870 modified the details somewhat, the basic procedures remained the same throughout this period, and each of them has important implications for the bibliographer who wishes to use the copyright records:

The amendments and revisions which followed are of interest to bibliographers in terms of the changes they brought about in the form of the records. The act of 29 April 1802 (2 Stat. 171, effective 1 January 1803) is significant from a bibliographical point of view only because it required all copyrighted works to carry a printed notice of the date of title entry, either on the title page itself or on one of the two pages immediately following (usually the verso of the title page),[19] a provision which has continued in effect to the present. On 3 February 1831 was passed the first general revision of the copyright law (4 Stat. 436), which as of that date superseded the 1790 act. It contained four important revisions: (1) the copyright term was henceforth to be twenty-eight years (instead of fourteen), renewable for fourteen; (2) the newspaper notice was required only for renewals; (3) copies of published works were to be deposited with the clerk of the district court (not the Secretary of State) within three (not six) months of publication; (4) the clerks of the district courts were to send these deposit copies and certified lists of all titles recorded, at least once each year, to the Secretary of State. In most other respects, the law remained as before.

One result of these changes was that clerks of some district courts began entering the deposit date for each work on the same page of the ledger which recorded that work's title entry. Another result was that the records were kept in duplicate, so that one set could be furnished to the Secretary of State. This system normally provides the bibliographer with more than one source for the dates of registration and deposit. Since the original record books and the duplicate set were kept in different places before 1870, it is unlikely that the same portions of each set would be lost or destroyed; thus the so-called "duplicates" can be used to fill in many gaps which now exist in the original records. Similarly, in cases where the deposit copies themselves are no longer available, the deposit dates can often be located in the record books (usually the original set only) — a fact not generally recognized by bibliographers, even those who utilize deposit dates. Although neither copy of Melville's White-Jacket in the Library of Congress, for example, has a deposit date written in it, the deposit date (26 March 1850) may be ascertained from the record books of the Southern District of New York; the Library of Congress copy of his Piazza Tales contains a note specifically pointing out that the present copy is a replacement of the deposit copy, but again the deposit date (20 May 1856) may be found in the district court ledger. Whenever both sources of information exist for a deposit date (both the copy itself and the record book) — or for a registration date (the two sets of record books) — some discrepancy is possible, particularly since these are handwritten records (another reason why the "duplicates" are not really duplicates). In such cases, for the registration date precedence should be given to the original record over the "duplicate," and for deposit dates to the actual copy of the work[20] over the record book. One should also remember that, after 1831, a publication date can be gauged with somewhat more accuracy than in the 1790-1831 period, since it can be no more than three months prior to the deposit date. But the most important point, worth repeating, is that there are few published and copyrighted American works, especially after 1831, for which the deposit dates cannot be ascertained, even if the deposit copies are no longer in existence.

The amendments to the 1831 revision have principally to do with the places where copies were to be deposited and the number of deposit copies required. The act of 10 August 1846 (9 Stat. 106, establishing

the Smithsonian Institution) provided that one copy of every copy-righted work be deposited — for use rather than record — in the Smithsonian and one in the Library of Congress (in addition, of course, to the essential copy of record deposited with the district court for transmittal to the Secretary of State). The copyright amendment of 5 February 1859 (11 Stat. 380), which repealed these deposit regulations in regard to the two libraries, transferred the copyright duties from the Secretary of State to the Department of the Interior: previous records and deposits were moved, and all future ones were to be sent there. Finally, on 3 March 1865, another amendment (13 Stat. 540) again required that a second deposit copy be sent to the Library of Congress — and within one month of publication. The various changes in deposit requirements[21] should perhaps be summarized:

• 1790-1831: One copy to Secretary of State within six months of publication.
• 1831-1846: One copy to district court (for transmittal to Secretary of State) within three months of publication.
• 1846-1859: Three copies: one to district court (for transmittal to Secretary of State) within three months of publication; one to Smithsonian Institution; one to Library of Congress.
• 1859-1865: One copy to district court (for transmittal to Department of Interior) within three months of publication.
• 1865-1870: Two copies: one to district court (for transmittal to Department of Interior) within three months of publication; one to Library of Congress within one month of publication.

One must keep these changes in mind when using the surviving deposit ledgers in order to interpret the dates properly. The district court deposit dates are usually preferable, when available, because in most cases the deposit copies had to travel a shorter distance to reach the clerk's office than to reach Washington and thus arrived sooner; on the other hand, for the five years after 1865, the law did not insist that copies be sent to the district courts as quickly as to the Library of Congress. A further complication is that many publishers did not comply with the deposit requirement, forcing Congress to make special provisions (in the acts of 3 March 1865 and 18 February 1867 [14 Stat. 395]) to penalize proprietors who failed to comply; so the recorded deposit dates for some works (particularly those of the smaller publishers) may not provide a reliable basis for estimating publication dates, since they may be late deposit dates resulting from a demand for compliance issued by the Librarian of Congress.

The records of the receipt of deposit copies by the various depository agencies in Washington are now available for reference in the

Rare Book Division of the Library of Congress and are complete except for a few gaps, as noted:

• State Department: 2 January 1796-29 May 1841 (11 vols.) [1841-59 not located]
• Department of Interior (Patent Office): 28 April 1859-5 July 1870 (3 vols.)
• Smithsonian Institution: 18 June 1852-1856 (unbound, in envelope 326); 29 February 1856-6 October 1858 (1 vol.) [1846-52 not located]
• Library of Congress: 1 December 1848-29 November 1852 (1 vol.); 31 March 1865-3 December 1886 (1 vol.) [1846-48 and 1853-58 not located]

None of these is indexed, and the titles in each are arranged according to the order in which they were received; usually only the year of publication of each book is given, along with the precise day of receipt (but the Smithsonian sheets from 1852 to 1856 list only the year of publication, and its ledger from 1856 to 1858 adds only the date the entry was made). Finding a particular title can be a time-consuming process, but if one already knows the district court deposit date, the search is facilitated. Melville's Mardi, which was deposited at the district court on 17 April 1849, was received by the Library of Congress on 19 April; his Battle Pieces, deposited at the court on 17 August 1866, was received by the Library on 20 August. These facts suggest that bibliographers should always check the district court record books for deposit dates (at least after 1846) even when deposit copies are still on the shelves of the Library of Congress, for those deposit copies may not contain the earliest deposit dates. The district court deposit copies, forwarded to the State Department and then to the Department of the Interior,[22] should have been turned over to the Library of Congress in 1870; but copies in the Library may also be those originally sent to the Library in the periods 1846-59 and 1865-70. Two copies presently in the Library may contain different dates, one of them the date of receipt by the clerk of the district court, the other the date of receipt by the Library. So the deposit ledgers of the various agencies are helpful in understanding these discrepancies and in locating the earliest deposit dates.

In addition, these registers of deposits received may be supplemented by certain published lists, for three attempts[23] were made

during this period to provide a published account of copyright deposits:

• William Elliot, A List of Patents Granted by the United States (1822, 1823, 1824, 1825, 1826). [Supplements in these volumes list all deposits received by the State Department between 1796 and May 1825. The supplements in the 1822 volume cover 1796-1821, and the succeeding volumes contain annual supplements through May 1825.]
• Charles Coffin Jewett, "Copy-right Publications," in the Fifth Annual Report of the Board of Regents of the Smithsonian Institution for the Year 1850 (1851), pp. 146-325. [Two supplements list all deposits received by the Smithsonian between 10 August 1846 and the end of 1850 (the first, pp. 146-236, covers 1846-49; the second, pp. 236-325, covers 1850).]
• Catalogue of the Library of Congress (1849, 1861); Supplement (1846-60). [The Supplements from 1846 through 1856 contain separate sections listing copyright deposits; the four Supplements for 1857 through 1860, as well as the two complete catalogues of 1849 and 1861, do not list deposits separately but mark them with a double dagger wherever they occur in the main alphabet. Together these Catalogues and Supplements entirely cover the Library's first period as a depository, from 10 August 1846 to 5 February 1859.]

These published listings fill some of the gaps in the manuscript ledgers presently located: the Jewett list, covering 1846-50, reduces the gap in the Smithsonian ledgers to a year and a half, since those now available begin with 18 June 1852; and the Library of Congress Catalogues and Supplements completely fill the gaps in the located Library ledgers. The Library lists and the Jewett list for 1846-49 are the least useful of the published lists, however, because they do not give the dates of receipt, as both the Elliot and the 1850 Jewett lists do. The places to look for deposit dates, therefore, other than in copies of the works themselves, may be summarized in this way:

• 1796-1831: in the registers of copyrights received by the State Department (or, through May 1825, in Elliot's published version of these registers).
• 1831-1841: in the district court record books for certain states; and in the registers of copyrights received by the State Department.
• 1841-1846: in the district court record books for certain states.
• 1846-1859: in the district court record books for certain states; in Jewett's list of Smithsonian deposits and the handwritten registers of copyrights received by the Smithsonian; in the registers of copyrights received by the Library of Congress, supplemented (without dates) by the Library's published Catalogues and Supplements.
• 1859-1865: in the district court record books for certain states; and in the registers of copyrights received by the Department of Interior (Patent Office).
• 1865-1870: in the district court record books for certain states; in the registers of copyrights received by the Department of Interior (Patent Office); and in the registers of copyrights received by the Library of Congress.

It is clear from this summary that the most important tool for research into deposit dates after 1831 is the long series of record books kept by the clerks of the district courts. Because they record the dates of original title entries (the "copyright" dates) and because they cover the entire period 1790-1870, they are obviously basic to any kind of copyright research for these years and constitute what is usually thought of as the "copyright records" before 1870. The copyright law of 1870 required that all the previous district court records be sent to the Library of Congress in Washington, and when those ledgers (dating from 1790) were added to the "duplicates" already there (dating from 1831), the result was a collection of over 600 large volumes and many additional bundles of loose papers (containing approximately 150,000 entries from 35 states). All this material (including 92 boxes of title pages) is now shelved in the Rare Book Division of the Library of Congress and constitutes a largely untapped bibliographical storehouse. Any bibliographer of nineteenth-century American literature should be intimately acquainted with the characteristics and contents of these records.

The difficulties in using these volumes are a result of the lack of centralization in the administration of the copyright law before 1870, since complete uniformity would be impossible in records kept at scattered offices throughout the country. Most of the books contain only one entry per page, consisting of the formal phraseology required by law, the date of entry, and the exact wording of the title page; they may be entirely handwritten (as the majority were in the earlier years), or they may be made up of printed forms, filled in as required. But they vary in several respects: some include the actual printed title pages, and others give only a handwritten title; some have single-volume or multi-volume indexes (by title or proprietor or both), and others are not indexed at all; some report the dates of deposit of published copies (after 1831), and others do not. In addition, not all the district courts sent their records to Washington, or some were lost en route, for the "duplicates" sometimes cover periods missing from the original volumes, and gaps still remain. (An inventory of these records is provided below in Appendix A.)[24]

One of the problems in using these records in their present form is that a bibliographer can never be sure that he has checked all the relevant material. This uncertainty is compounded by the inevitable feeling that the missing records may still exist in the states, not having been sent to Washington in 1870. A check of other repositories reveals that certain pre-1870 copyright records (for nine states and the District of Columbia) do indeed exist outside the Library of Congress — in the National Archives, the Seattle Federal Records Center, the South Carolina Library, the Texas State Library, the Virginia Historical Society, and the federal archives in Kentucky, Maryland, Ohio, and Rhode Island. (A list of these materials is given in Appendix B.)[25] Out of this entire group of pre-1870 records, only a few have been published or transcribed. Most of these publications consist of small groups of entries taken from the record books in the Rare Book Division of the Library of Congress; but four of them — for Kentucky, Rhode Island, Texas, and Virginia (Richmond, 1864-65) — draw on the material housed in other locations, and one (for Richmond, 1790-1844) preserves the content of a record apparently now lost.[26] (See Appendix C.)

The collection of copyright materials in the Rare Book Division of the Library of Congress also contains, in addition to the district court record books, the boxes of title pages, and the registers of deposits received by the four Washington deposit libraries, a small number of miscellaneous papers: (1) Letters — Library of Congress correspondence, both incoming and outgoing, on copyright matters (particularly acknowledgments of deposit copies) between 1852 and 1855, a letter

book of the Commissioner of Patents, 1859-70, and some miscellaneous letters in Box 308; (2) Ledgers — a volume marked "Maps Labels &c." (1849-58), with some music entries (1849-51) at the beginning; a volume listing the returns made by the clerks of the district courts for the Department of State, 1853-68; three unidentified indexes; and a volume containing (pp. 3-6) a list of the district court records received by the Library of Congress in 1870 (the list which formed the basis for the one Roberts published in 1939). This last ledger is of further interest because it contains (pp. 540 ff.) lists of books by twenty-six selected authors,[27] with the date of title entry for each work and often a reference to the volume and page of the original entry in the district court record books; it was evidently an attempt by someone in the Library of Congress, after the records were received in 1870, to furnish a convenient index for the most popular authors. Even though it omits a number of books, it can still serve today as a short-cut to locating the original title entries for these authors.

The fact that such a list, incomplete as it is, can be useful suggests the complex nature of the records for this period. Until all the pre-1870 entries are published, with a complete index, no bibliographer can feel certain that he has found everything relevant to his subject. If the desired information happens to be in one of the indexed bound volumes, it can be found easily; but if it involves a district for which there is no index or for which the only surviving records are scattered loose sheets, the search can be time-consuming. The lack of centralization in the administration of the copyright law before 1870 made complicated records inevitable, and a consolidated index is essential if they are to be an effective tool for research. But the bibliographer who understands the changing provisions of the law which produced those records can find in them, even as they stand, a large amount of information not elsewhere available about the publications of many of the major writers of nineteenth-century America.

III. 1870-1909

The copyright act of 8 July 1870 (16 Stat. 212), the first general revision of the copyright law after 1831, is a landmark in American copyright legislation. Its significance lies not in any altered conception

of the nature of copyright but rather in a change of administrative procedure, for all copyright business was henceforth to be centralized in the Library of Congress. Title pages and copies were to be deposited with the Librarian of Congress, not the clerks of the district courts; and all previous records, from the various district court offices and the Department of the Interior, were to be transferred to the Library of Congress. It is impossible to overemphasize the importance of this centralization: the year 1870 is a dividing line which marks the end of fragmentary and inconsistent copyright records and the beginning of records that are complete and uniform.

The basic requirements of the law were scarcely changed: the term of copyright was 28 years, renewable (within six months of expiration) for 14 more; the newspaper notice was for renewal only (to be printed for a period of four weeks, within two months of the date of renewal); a printed title page was to be deposited before publication; and two copies of the work were to be deposited after publication (mailed within ten days of publication). The bibliographer can use these dates to establish the approximate time of publication in exactly the same way as for the preceding period, since the publication date should fall, according to law, in the interval between the date of title entry and the date of copy deposit. This interval will generally be smaller (and its usefulness for determining a publication date presumably greater) than for the preceding period, since the deposit copies were supposed to be put in the mails no later than ten days after publication. From the bibliographer's point of view, therefore, the records for the period 1870-1909 are different from the earlier ones not in the kind of information they provide but in their completeness and greater ease of reference.

Between 1870 and 1909 the various amendments and revisions of the copyright law generally had no effect on these provisions for title entry and deposit of copies. The revision of 1 December 1873 (Rev. Stat. 957) retained them as they were set forth in 1870; and the only change in them effected by the act of 3 March 1891 (26 Stat. 1106) was that copies be deposited (or placed in the mails for deposit) no later than the day of publication, thus reducing still further the legal interval between the date of title entry (which could be as late as the day of publication) and the date of copy deposit. Books were not always deposited as promptly as the law stipulated, however, and the act of 3 March 1893 (27 Stat. 743) specifically granted copyright privileges to works, however belatedly deposited, that had been received before 1 March 1893. Of the other amendments, those of interest to bibliographers are the act of 18 June 1874 (18 Stat. 78),

which transferred one category of printed material — commercial prints and labels — to the jurisdiction of the Patent Office; the act of 19 February 1897 (29 Stat. 545), which established the position of Register of Copyrights; and the act of 7 January 1904 (33 Stat. 4), which described a special two-year copyright privilege for foreign works that were to be exhibited at the Louisiana Purchase Exposition.

The act of 1891, if it made only one small change in the deposit regulations, was nevertheless of the greatest importance in other ways, for it inaugurated certain features which have caused American copyright records to be of particular usefulness to bibliographers. For one thing, it provided for the granting of copyright protection to citizens of foreign countries, if those countries gave reciprocal protection to United States citizens — of bibliographical significance because of the publication data about foreign works which were thus inserted into the American record books. Along with this provision came the much-discussed requirement of American manufacture (the books deposited for copyright "shall be printed from type set within the limits of the United States, or from plates made therefrom"). Although the name and address of the printer were not entered into the record before 1909, it was this manufacturing clause which resulted, after 1909, in a large quantity of information of bibliographical interest. Finally, the 1891 act required the Librarian of Congress to report title entries of all works received to the Secretary of the Treasury, who in turn was "directed to prepare and print, at intervals of not more than a week, catalogues of such title-entries for distribution to the collectors of customs of the United States and to the postmasters of all post-offices receiving foreign mails." While the purpose of this provision was to facilitate the enforcement of the manufacturing clause, the more important result was the complete publication of the American copyright record (for deposited works) from 1891 forward.

The record materials which have accumulated since 1870 under the centralized administration of the copyright acts (with a few exceptions) are now housed in the Copyright Office of the Library of Congress and are available there for reference. After 1891 the published records make a great deal of the copyright data readily accessible without a trip to the Copyright Office, but even for that period there is no substitute for the indexes and other resources to be found in the Office itself. For the years between 1870 and 1891, since there are no published records except for drama, most copyright information can be obtained only in the Copyright Office records; bibliographers dealing with this period should search these records in person whenever possible, even though the Office provides, for a fee, an excellent search

service. To the uninitiated, the stacks and indexes at the Copyright Office may appear bewildering, but they actually are not difficult to use and are far less confusing than the pre-1870 records. For the period 1870-1909, the bibliographer should know in particular about three kinds of materials: the record books, the card indexes, and the published records.

(1) Record Books. For the entire period 1870-1909, the record books are uniform in the kind of information they contain: the date of entry of a printed title page on or before the day of publication, a transcription of the title page, the name of the claimant, and the date of deposit of two copies on or after the day of publication. As before 1870, some titles were entered for works never finally published, and copies were sometimes not deposited for works that did get published. In a few cases the copies were received, but the date was not posted onto the record (in some of these instances the Library still has one or both of the copies bearing the copyright deposit date or stamp); in other cases the copies probably meant for deposit were received by the Library but not by the proper office and were therefore not recorded. The only changes in the record books during this period are changes of form, not of content. Between 1870 and the end of 1897, when application for copyright was made by letter, the record books have four printed forms per page for the insertion of information regarding four applications; each entry has a serial number, and there is a separate sequence of numbers for each year, with an attached letter designating the year, from A through C2. From 1898 through 30 June 1909 application was made on printed forms; and on 2 January 1900 a new series of volume numbers began, since in that year copyrights were divided into separate classes (for books, periodicals, and so on), with separate series of volumes for each class (but the volumes for drama, Class D, did not begin until 1901). For book applications there was now only one printed form per page (500 per volume), and the new sequence of serial numbers was prefixed with "XXc" (for "twentieth century"), running continuously from year to year; by 30 June 1909, volume 488 and XXc 243564 had been reached. For Class D (drama), the number of record books by 1909 amounted to seventeen (with four applications per page). Record books before 1900 are at present housed in closed stacks and may be requested by searchers; those from 1900 onward are shelved in open stacks.

Assignments and renewals were handled in different ways. All assignments were entered in a separate series of record books, beginning on 25 July 1870 and continuing in an unbroken sequence

throughout this period. These records are an unparalleled source of information, especially for bibliographers interested in the history of publishing firms, because it is possible here to trace mass transfers of titles and establish the line of succession of various firms. Renewals, on the other hand, do not form a separate category of records. Since the law required that copyright proprietors comply with the same requirements for renewal as for original registration, renewal entries are like the entries for original copyright and occur in the same sequence of serial numbers.[28] The deposit requirement of two copies also applied to renewals, and some publishers, such as Houghton Mifflin, occasionally printed cancel title pages bearing the renewal date for insertion into the deposit copies; this production of bibliographical oddities was by no means widespread, however, and was certainly not required.[29]

The changes of form in the records are of little practical significance to most searchers, who will come to the books with a list of particular serial numbers they wish to look up. The important point is that these volumes constitute the basic copyright record for the period and that they are complete. The printed title pages and the applications for these years are stored in the Federal Records Center in Alexandria, Virginia; although they are available upon request, the completeness of the record books makes them of less significance than in the pre-1870 period, for there are no gaps in the records which they might fill and no information in them not included in the record books.[30] The deposit copies themselves no longer form a part of the

records of the Copyright Office, but many of them are on the shelves of the Library of Congress, and others may turn up almost anywhere. The 1909 act empowered the Librarian to make requisitions for the second deposit copies of any previously deposited works which he considered desirable for the Library to possess in more than one copy; in this way both deposit copies of certain works by a number of major writers have been preserved.

(2) Card Indexes. Because the entries were made in the record books in chronological order, some kind of index is essential for locating particular items with any efficiency. The years from 1870 through 1897 are covered in one card index, affectionately referred to by the Copyright Office staff as "Old High." This index consists of hand-written cards and slips of various sizes: some cards contain many entries, others only one; some were written at the time of registration, others much later in an attempt to improve the index; some are Library of Congress requisitions for second deposit copies, and others have printed strips mounted on them. Entries are under copyright claimants, authors, and titles. The claimant entries are supposed to be complete, but the entries for authors and titles were not consistently made throughout the period. As a result, the index is not entirely complete; in addition, cards may sometimes be misfiled, since they are not held in the drawers by rods and since there have been various attempts to improve the index by adding cards. Despite its deficiencies, the index generally provides the desired reference, especially if one knows the name of the copyright claimant and can check under it. These cards do not normally give the dates of registration and deposit but only the registration serial numbers, which can then be looked up in the proper record book. The card for Herman Melville, for example, lists beside the title Clarel simply the number G7090; one then locates this entry in the 1876 (G) record books and finds that the title for Clarel was entered on 8 June 1876 and the copies deposited the following day — information not available, incidentally, in the deposit copy in the Library of Congress.

The remainder of this pre-1909 period is included in a card index which covers the years 1898 to 1937. This index is divided according to the various classes of copyrightable material, but the index for books is usually the only one bibliographers will need to consult. The book index for 1898-1937 consists of two alphabets, one of claimants and

one of authors (and some titles). For the years 1898 to 1909 the claimant entries are on printed "proprietor cards" which record the dates of title entry and copy deposit, transcribed from the original applications, along with the registration serial number. It is not necessary, therefore, to consult the record books themselves during this period, except for the purpose of double-checking (and the information in the record books is also a transcription from the applications). The author index is complete for copyrighted books represented by Library of Congress cards, but its completeness in regard to the authors of works not so represented cannot be assumed. Claimant indexes are always the basic and most thorough ones for copyright purposes, a fact which bibliographers, generally accustomed to searching by authors, should keep in mind (and this very fact is what makes copyright records so useful for research into the history of publishing firms). To sum up, the entire 1870-1909 period is adequately (if not quite completely) indexed in two basic card catalogues: a claimant-author-title catalogue covering 1870-97, and separate claimant and author files covering 1898-1937. Books falling near the dividing line (that is, in 1897 or 1898) could conceivably turn up in either index and should be checked in both.[31]

(3) Published Records. The copyright records for this period have been partially published, in two publications, one covering all entries for drama and the other containing deposits of all classes beginning with 1891. Both publications should be basic tools for all bibliographers of American literature, but they are not as well known as they deserve to be. While they are not a substitute for the materials in the Copyright Office, they are in some respects easier to use, and they at least provide the basic data about many works from this period without the trouble of a trip to the Copyright Office or correspondence with a member of the search staff.

In 1916 and 1918 the basic tool for copyright research in the field of drama during this period was published:

The arrangement is by title, with an index of authors and claimants; for each title are given the dates of registration and deposit and the registration serial number. It is important to note that this work lists all titles of dramatic compositions registered, not just those that were eventually deposited, and consequently is a substitute for the record books as far as drama is concerned. The thoroughness of the work and its indexing also make it a more dependable guide than the sometimes erratic card catalogue for 1870-97. Anyone seeking copyright information about drama during this period should consult this work first and will probably have no reason to check further in the Copyright Office itself.[32]

The other work, authorized by the act of 1891, is surely one of the most important bibliographical tools in existence, at least in the form which it eventually assumed. The 1891 law provided for the publication of all title entries for which copyright had been completed by the deposit of two copies, and the Catalogue thus started has continued to the present, with one change of title:

During its early years the Catalogue was not an effective instrument for research, both because it did not contain sufficient information and because the information it did contain was not conveniently arranged. For the first 227 weekly issues, the titles were listed in one column and the names of proprietors in another, but in no particular order and with no index and no specific dates (although one could conclude that the works had been deposited during the week covered by the issue). These early issues do serve as a record of all works deposited, but they can be used for reference only by running through the entire list of titles. Gradual improvements occurred in the following stages:

• No. 228 (11-16 Nov. 1895): books arranged alphabetically by proprietors in each issue.
• No. 323 (6-11 Sept. 1897): specific date of receipt of deposit copies recorded.
• No. 327 (4- 9 Oct. 1897): books arranged alphabetically by authors, followed by index of proprietors.
• No. 340 (3- 8 Jan. 1898): registration number as well as deposit date recorded.
• No. 348 (28 Feb.-5 March 1898): specific date of title entry as well as registration number and deposit date recorded.

It was not until nearly seven years after the start of the Catalogue, with No. 348, that the entries contained the information one would expect — registration number and dates of registration and deposit — and even then the only indexes were weekly. With volume 26, for the first three months of 1901, appeared the first index for an entire volume (giving the names of proprietors and authors and the titles of plays and periodicals). When the new series began on 1 July 1906, the Catalogue was divided into four parts, for books, periodicals, musical compositions, and fine arts. The successive sections of Part I (Books) were cumulated into semi-annual volumes with excellent author-proprietor indexes, so that from the middle of 1906 there are only two indexes per year (for books).

The Catalogue is essentially a twentieth-century tool, therefore, because its indexing is inadequate before 1901 and the information it contains not sufficiently detailed before 1898. For the years before 1901 the card indexes in the Copyright Office are much simpler to use than the published Catalogue, although before 1898 they may be some-what less dependable; even after 1901 they may be more convenient for certain kinds of research, since they bring together entries from large spans of years and obviate the drudgery of going through numerous annual indexes. And for information about titles which were never deposited, the Copyright Office card indexes are essential, since this information is not included in the published catalogues (except the cumulated one for drama). The Catalogue does not supersede the Copyright Office files, but it does make possible extensive copyright research, at least from 1901, without recourse to those files.[33]

This period may be divided into three segments, according to the kind of indexing available (aside from the published index for drama):

• 1870-1891: card index, not entirely complete, containing references to record books.
• 1891-1897: card index, not entirely complete, containing references to record books; complete published record of deposits, lacking important information and index.
• 1898 and after: complete card indexes, duplicating information in record books; complete published record of deposits, with basic copyright data and indexes (cumulated volume-indexes beginning in 1901).

These changes are evidence of the fact that the period was a transitional one, in which the newly centralized Copyright Office was gradually developing the efficient techniques which would characterize it in the twentieth century.

IV. 1909 — present

Whereas the 1870 law represented a change in administrative procedure rather than in definition of the term of copyright, the act of 4 March 1909 (35 Stat. 1075, effective 1 July 1909) made fundamental changes in the procedures required for securing copyright, the first such changes since the original law of 1790. From 1790 to 1909, the term of copyright had dated from the deposit of a printed title page prior to publication, in order to protect a work during the course of publication; the 1909 statute eliminated entirely the necessity of registering title pages in advance of publication and designated the term of copyright to begin with the date of publication. (The term was still 28 years, but renewable for another 28, rather than 14 — without inserting a notice in a newspaper.) Henceforth the "copyright date," by legal definition, was to be synonymous with the publication date; and as a result bibliographers have at their fingertips, in the copyright records and the published Catalogue, the official publication dates of all copyrighted works for which registration has been made in the United States since the middle of 1909. To be sure, the actual publication date (or date of release to the general public) may in some cases be different; but deciding what constitutes an "actual publication date" is an arbitrary matter anyway, and the fact remains that the date in the copyright records is the one reported by the publisher (or author, or whoever was the copyright proprietor) as the publication date (and therefore the date on which he wished his copyright protection to begin). So, in the post-1909 records, there are no title registration dates and no collection of title pages, but only the dates of publication — and, as before, the dates of deposit of two copies, important for the bibliographer as proof that physical copies of the books existed on those dates.

The other most important feature of the 1909 law, from the bibliographer's point of view, has to do with the requirement of American manufacture, initiated in the 1891 act. Whatever its merits (or demerits) from other points of view, it became a great boon to the bibliographer in 1909, for the new statute required, as part of each application for copyright, an affidavit giving the names of the printer and binder and the date of the completion of printing. Since the printer was not obliged, as in England, to place his name on the printed matter itself,[34] this manufacturing information for American books is

usually not elsewhere available, unless the printers' or publishers' files exist and are accessible. Descriptive bibliographers generally record the names of printers when they are provided in a colophon or printer's imprint and sometimes go to elaborate lengths to consult publishers' records. While there is no doubt that publishers' records are valuable (in furnishing facts about sizes and dates of impressions, for example), the point here is that those records are not the most convenient source for the names of the printers and binders of all copyrighted American books after 1909. The copyright records, open to public inspection, contain them, and often in addition the officially sworn date of the completion of printing for each book (but the presence of this date depends on the form of affidavit in use at any given time). One can easily discover that Henry James's Gabrielle de Bergerac was printed and bound by J. J. Little and Ives (completed on 12 November 1918) and that Sherwood Anderson's Winesburg, Ohio was printed by the Van Rees Press (finished 28 April 1919) and bound by the H. Wolff Estate — information not provided in the standard bibliographies of these authors.

One other provision of the 1909 law will serve to suggest the bibliographical usefulness of twentieth-century American copyright records even for students of English and foreign literatures. Sections 21 and 22 of the act provided for "ad interim copyrights" of works in the English language first published abroad. Depositing one copy of any such work in the Copyright Office within thirty days after publication would secure to the author (or proprietor) an American copyright to last for a period of thirty days; if during that period the author (or proprietor) arranged to have an American edition of the work printed and deposited, he could secure a regular 28-year copyright. Although the lengths of these periods have varied in the years since 1909, the application for an ad interim copyright has always necessitated a statement of the original date of publication of the foreign edition; the ad interim record books (which begin with 16 July 1909) therefore contain the English publication dates of a great many works of twentieth-century English literature. The works entered naturally represent only an erratic sampling of all English books published (amounting to something more than 29,000 entries by the end of 1945); but for the works included, this source of information about publication dates is frequently a more convenient and accessible one than any which exists in England. From the ad interim record books one can learn that Norman Douglas's South Wind was published on 5 June 1917 and D. H. Lawrence's Pansies on 4 July 1929, facts which are not recorded in the standard Soho bibliographies of these authors. A similar statutory

arrangement in regard to works first published abroad in a language other than English had existed from 3 March 1905 (33 Stat. 1000) until 1 July 1909, when a provision went into effect, as a part of the general revision of that year, specifying that a full-term copyright could be secured for works of foreign origin in a language other than English whether or not such works or translations of them were manufactured in the United States; the resulting "Class A Foreign" record books contain the original publication dates for a large number of foreign works (over 3000 entries between 22 March 1905 and 1 July 1909, and over 85,000 more by 1945).

The copyright records for the post-1909 period are not strikingly different from those for the preceding period, except for the kinds of dates they include, but certain points about them are worth noting:

In the second series of CCE (1906-46), bibliographers will generally need to refer only to Part I (Books) and Part II (Periodicals). Beginning with 1909, however, pamphlet and dramatic material was designated as Group 2 of Part I, listed in separate volumes with separate annual indexes; since some works could be classified either way, it is generally wise to check the annual indexes for both Groups 1 and 2. A Group 3, for dramatic compositions, was added in 1928; again, published drama could appear in Groups 1, 2, or 3, and all are worth checking. The annual indexes, by claimant and author, are excellent through 1938, but from 1939 through 1946 they contain only the names of authors, not of claimants, making the CCE less useful during this period for work on publishing history.[39] The war also necessitated certain abbreviations in the scope of each entry: from 1938 through 1945 some of the information available on Library of Congress cards (such as the total pagination) was eliminated, and from 1940 through 1946 the publication date alone (and not the deposit date) was given. In the third series of CCE (1947- ), bibliographers may be interested in Parts 1A (Books), 1B (Pamphlets and Contributions to Periodicals), 2 (Periodicals), 3/4 (Drama and Works for Oral Delivery), and possibly 5 (Music) and 6 (Maps and

Atlases). The entries for Part 1 are now arranged alphabetically by authors, obviating an author index, but with two alphabets per year (since there are two semi-annual issues); not until 1951 and 1952 are there claimant indexes, and beginning with 1953 there are cross references, in the same alphabet, from the names of claimants. Only publication dates and registration numbers are given in the third series. Despite a few deficiencies between 1938 and 1951, the CCE is a remarkable work of reference that can serve many bibliographical uses for which it was not originally designed.[40] Bibliographers working in the period after 1909 will find that the American copyright records contain more information of the kind they are seeking than has been preserved in the official records of any other nation.

V. A Note on English Copyright Records

England was the first country to institute statutory copyright, in the famous Statute of Anne of 4 April 1710 (8 Anne, ch. 19). Before that time the closest counterpart to copyright was the control which the Stationers' Company exercised over the printing of books by its members, but this regulation did not resemble the modern concept of copyright, since it was designed as protection for the printer rather than the author. Nevertheless, the requirement of entering titles in an official register maintained by the Stationers' Company has resulted in records of great bibliographical value similar to those which are produced by a system of copyright registration. The registers covering the years 1554 through 1708 were transcribed and published many years ago[41] and are well known to bibliographers, as are the published records of the Court of the Stationers' Company between 1576 and 1640.[42] The bibliographical importance of these materials has long been recognized, and extracts from them were published by J. P. Collier as early as 1853;[43] since that time several important bibliographical

publications relating to the period before 1710, by such scholars as Rollins, Greg, and McKenzie, have been based on records at Stationers' Hall.[44] Anyone interested in pursuing the complications of sixteenth- and seventeenth-century English copyright should acquaint himself with the papers available in Stationers' Hall, as listed in the Library, 4th ser., VI (1925-26), 349-57,[45] as well as with certain basic work in this area: Harry Ransom has recounted, in historical terms, the operation of English copyright before the act of 1710;[46] Leo Kirschbaum has described "The Stationers' Company in Operation" and has helped to clarify the difficult relationship of registration to both license and copyright;[47] and W. W. Greg has taken up in detail the bibliographical implications of the records of the Stationers' Company.[48] Although problems remain to be solved, it is clear that the essential researches of these and a few other scholars[49] have provided a fuller treatment, from a bibliographical and literary point of view, of this period than of any other in Anglo-American copyright history.

The act of 1710 is the principal dividing line in the story of English copyright: it provided for a copyright term of fourteen years, renewable for 14 more (but not clearly established as author's, rather than printer's, copyright until 1769 in Millar v. Taylor); it made prepublication title entry in the register of the Stationers' Company a statutory requirement; and it stipulated that nine deposit copies be furnished upon demand (for the Royal Library [later the British Museum], the university libraries at Oxford, Cambridge, Edinburgh, Glasgow, and St. Andrews, the Library of the King's and Marischal Colleges Aberdeen, Sion College London, and the library of the Faculty of Advocates in Edinburgh). Although the force of the Stationers' Company declined after the lapsing of the Printing (or Licensing) Act in 1695,[50] the stationers continued, even after the 1710 act, to turn copyright to their own advantage, through trade boycotts and court injunctions. Their contention that common-law copyright still obtained for any work after the expiration of its term of statutory protection was finally invalidated in the case of Donaldson v. Becket (1774), and a stream of cheap reprints followed.[51] The upshot of this situation for bibliographers in search of records is that the "Entry Books of Copies" at Stationers' Hall for the eighteenth century[52] (with a printed index 1710-73, a manuscript index thereafter) are even less full than those for the earlier years: stationers before 1774 considered their copyrights perpetual by common law; after 1798 (Beckford v. Hood) they saw that failure to register prevented only suits for the statutory penalties, not the more important suits for damages; and, in addition, they felt that by not registering they were not obligated to provide copies for the depository libraries, even though the Universities Copyright Act of 1775 (15 Geo. III, ch. 53) emphasized the requirement of prepublication registration and deposit. The records at Stationers' Hall from 1814 through 1841 have a slightly different

significance for bibliographers: because the 1814 amendment to the copyright law (54 Geo. III, ch. 156) required that entry be made within one month after publication (rather than before publication), only works actually published are listed during this time.

The next major revision of the copyright law, the act of 1842 (5 & 6 Vict., ch. 45), made the copyright term 42 years (or the life of the author plus seven years, whichever was longer), and it reenforced the requirement of registration at Stationers' Hall. The copyright registers and related documents which accumulated in the administration of this act, covering the period 1842-1923, are now in the Public Record Office, not Stationers' Hall. The entries in them fall far short of constituting a national bibliography, for understandably many publishers still did not bother with registration; it was not a condition of copyright, and, in case of a suit for infringement, it need be performed only before the actual court proceedings. But the records are worth checking, for American as well as English books, if one keeps in mind the fact that the entry may occur long after the date of publication. Indexes exist for the entire period, in seven segments, giving the date of entry and the date of publication for each work registered. The first five volumes, covering 1842-1907, have been published as the Index of Entries (1896-1907) and are available in many reference libraries; the remaining years, 1907-23, are covered in six unpublished manuscript volumes in the Public Record Office. All eleven are listed here with the index numbers assigned them by the Public Record Office:

• 1842-1884 (literary): Ind 16908 (titles)
• 1842-1884 (commercial): Ind 16915 (titles)
• 1884-1897: Ind 16909 (titles, with author-publisher index)
• 1897-1902: Ind 16910 (titles, with author-publisher index)
• 1902-1907: Ind 16911 (titles, with author-publisher index)
• 1907-1911: Ind 16912 (titles), 16913 (authors), 16914 (publishers)
• 1911-1923: Ind 20293 (titles), 20294 (authors), 20295 (publishers)

These indexes furnish the same information that is available in the original registers; but if one wishes to check the original entries, the next step is to ask at the Assistant's desk in the Long Room for the typewritten Guide to the Records of the Copyright Office (classified as ff. 78), which indicates the dates covered by each of the registers (Ind 5791 through 5869).[53] It is also possible to send for the original copyright entry forms, stored at Ashridge (classified as Copyright 1/567 through 887, and designated by year in the Guide.) After 2 July 1883 copyright assignments are entered separately (Ind 5870-5872, indexed by Ind 6005-6007), as are certain foreign entries after 1854.[54] Other materials from this period in the Public Record Office, particularly papers relating to copyright suits, can furnish publication data of value to bibliographers. By checking the bills of complaint (in C.14) and the cause books (C.32) — as well as the depositions (C.24) and affidavits (C.31) at Ashridge — one can discover, for example, the publication dates of the Routledge piracies of Melville and Irving.

The act of 1911 (1 & 2 Geo. V, ch. 46), which made the term of copyright (for unpublished as well as published works) the lifetime of the author plus fifty years, abolished the statutory requirement for registration, although this requirement remained in force in some parts of the United Kingdom (e.g., Canada) through 1923. Voluntary registration was still possible, however, and entries made on this basis since 1 January 1924 are in registers at Stationers' Hall. But the new law retained a provision for deposit, emphasizing the split between registration and deposit which had prevailed throughout English copyright history. A sort of deposit system had existed as early as 1610 in an arrangement Thomas Bodley had made with the Stationers' Company, and the Printing Act of 1662 had required the deposit of three copies (for the Royal Library and the universities at Oxford and Cambridge). The list of nine copies established by the 1710 act was increased by two in 1801 (41 Geo. III, ch. 107), with the addition of Trinity College and the Society of the King's Inns, Dublin; then in 1836 (6 & 7 Will. IV, ch. 110) the four Scottish universities were removed from the list; and in 1842 (followed by the 1911 act) only six depository libraries were designated: the British Museum, Oxford, Cambridge, Trinity College Dublin, the Library of the Faculty of Advocates in Edinburgh (after 1925 the National Library of Scotland), and the National Library of Wales. The number of places to look for deposit copies thus varies with the period and, in addition, with the requirements of particular libraries, since copies were to be deposited upon request by the libraries. The necessity of depositing on demand, however, was not limited to registered books; after 1812 the interpretation of the law was that the deposit provision applied to

all books published in England, whether or not they were entered at Stationers' Hall.[55] Libraries often mark the date of receipt in books; the British Museum began this practice in 1833 (the exact day from 1837 forward), stamping copyright deposit copies in blue and colonial copyright deposits in black.[56] Since dated deposit copies may exist for books which were never registered, both sources — official registers and depository libraries — should always be checked. English copyright records, at least after 1710, are not particularly complex and for much of the period are conveniently indexed in printed form; at the same time they are not especially full and represent an unpredictable sampling of the national literary output.

VI. Some General Considerations

Although copyright records are national documents, the operation of a national copyright law can at times produce records of interest to students of another country. Even when the records themselves do not contain information about the publication of foreign works, the international copyright situation is always a principal factor in publishing history. Whenever two countries employ the same language, authors in one country sometimes deliberately publish their works first in the other, in an effort to forestall unauthorized (or "pirated") editions. In regard to English-speaking countries, this situation existed between England and Ireland in the eighteenth century (since the English copyright law did not extend to Ireland until 1801) and between England and the United States in the nineteenth (since the American law did not include English works until 1891). Many famous English books of the nineteenth century were first published in America because, even though the works could not be copyrighted in the United States, the authors could at least gain something from the

American market by selling advance sheets to an American publisher; many famous American books were first published in England because American authors could secure English copyrights during most of this period only if they published their works first in England or were residents of England.[57] A great deal of bibliographical and literary research has been directed toward this aspect of copyright history — from I. R. Brussel's Anglo-American First Editions (1935-36) to individual studies of the trans-Atlantic publication of Scott, Marryat, Cooper, Irving, Dickens, Tennyson, Longfellow, and Mark Twain,[58] among others. But copyright records are not involved, except when a country allows foreign works to be registered or deposited. American titles do appear in the nineteenth-century British records; but foreign titles are listed in the American records (under authority of the 1891 act) beginning at different times for different countries: copyright privileges were extended to some countries (e.g., France and England in 1891, Germany and Italy in 1892) by individual Presidential proclamations, to others (China, Hungary, Thailand) by bilateral treaties, and to still others through the Mexico City Convention (after 1908), the Buenos Aires Convention (after 1914), and the Universal Copyright Convention (after 1955).[59]

Despite the differences between the English and American copyright laws and records, the two countries have held the same basic traditions of copyright. The English concept of perpetual common-law copyright in unpublished work and limited statutory copyright in published work (implying that copyright is not a natural right of authorship), established in Donaldson v. Becket (1774), formed the tradition inherited by the United States and supported judicially in Wheaton v. Peters (1834). The philosophy of copyright which exists in any country naturally affects the kinds of records that accumulate, but not their thoroughness, for compulsory registration and deposit are not necessarily tied to statutory provisions for limited-term copyright. Further-more, official copyright information not found elsewhere may be embodied in court decisions; thus the judicial process by which a national attitude toward copyright is evolved serves at the same time to record specific facts about the individual books which occasioned the decisions. Changes in the law or its interpretation may not always result in changes, from a practical point of view, in the bibilographer's approach to the records; but since the records are a creation of the law, shifts in policy will usually alter the theoretical significance of particular inclusions or omissions.

The most striking conceptual change, since the Statute of Anne, was effected in the 1911 British act by removing unpublished works from the realm of common law; a similar provision has been proposed in the copyright revision bill in the United States (H.R. 2512, S. 597, 90th Congress, 1st session).[60] If this bill is passed as presently written,

all works, unpublished as well as published, come within the scope of statutory protection, and the term of copyright is the same for both — the life of the author plus fifty years. Bringing unpublished works under the statute and basing the term of protection on the author's lifetime will result in the recording of useful biographical information for literary historians. Both registration and deposit, however, are specifically stated not to be conditions of copyright protection. Registration is permissive rather than compulsory and may be performed at any time during the term of copyright (though it is a prerequisite for a suit for infringement); deposit of two copies within three months of publication is requested for the Library of Congress (enforceable upon the Register's demand). Under these regulations, most copyright holders will see the value of prompt registration and deposit, and the bibliographer will find no great change in the information he can derive from the records. At the same time, the view that copyright protection is not dependent upon any administrative technicalities of registration — with the consequent permissive nature of the registration procedure — means that the records which the bibliographer searches fall short, by legal definition, of constituting a list of all copyrighted works.

Bibliographers and literary historians do not often have the opportunity to document their research with sworn testimony, and evidence from copyright records therefore comes to the scholar with a kind of guarantee he rarely encounters. Information in the American Catalogue of Copyright Entries, for example, can be presented as prima facie evidence in a court of law; thus when a currently unlocated book is recorded as having been deposited on a particular day, the bibliographer, even though he may wish he personally could see the book, has the strongest assurance that it did actually exist at that time. No materials except official documents can carry such authority, and for this reason copyright records are a uniquely important source for the bibliographer. The English and American records, as described here, vary from period to period in their completeness, but it is evident that copyright registration has been carried out more punctiliously and in greater detail in the United States than in England and that the American records, especially after 1870, constitute an unparalleled storehouse of bibliographical facts. The Copyright Office of the Library of Congress should be familiar territory to every bibliographer; as a center for bibliographical research into American books (and many foreign ones after 1891), there is no other place equal to it.[61]

The convenience of using the English and American copyright records has been greatly increased by the publication of substantial portions of them. Even when published indexes do not carry the official sanction which the Catalogue of Copyright Entries does, they serve a useful purpose. It is to be hoped, therefore, that the remaining sections of these records can be published, either as official governmental projects or as separate volumes prepared by individual scholars. For English copyright, the Bibliographical Society has undertaken to publish part of the remaining records, beginning with David Foxon's edition of the Register for 1710-46; for the rest of the period before 1842 and after 1907, at least the kind of index which has been published to cover the years from 1842 through 1907 should be made generally available. As for the American records, the two unpublished segments present different problems. The record books from 1870 to 1891, which are complete and orderly, offer little difficulty; the essential facts in each entry could be transcribed and indexed (perhaps in a form similar to Dramatic Compositions), and the resulting volumes could be issued seriatim. But for the district court records, from 1790 to 1870, all the surviving documents (including the title pages) should be thoroughly indexed on cards before any volumes are published, since the nature and physical arrangement of these materials are such that information relevant to one period and state can turn up at a number of places. The published entries could then be arranged chronologically by states, with one master-index, or else the master-index itself could be constructed to contain the information from the entries: the most important requirement is that all titles registered in any given period, regardless of the state in which the registration took place, eventually be brought together in one alphabet. Such projects, though expensive and time-consuming, are justifiable and desirable. The publication of these entries would facilitate and stimulate bibliographical and historical research into the Anglo-American cultural heritage; it would also insure the preservation of the most detailed official record ever produced of a nation's literary and artistic product.

Appendix
UNITED STATES DISTRICT COURT COPYRIGHT RECORDS, 1790-1870

A. Records in the Library of Congress

The following inventory of the records now located in the Rare Book Division of the Library of Congress has been prepared as an aid to bibliographers who wish to check the copyright facts about a particular book but have no idea what sort of record exists for the period and district involved. Dates given by month or day below refer to the original entry books; those given only by years are the "duplicates" (often submitted in yearly batches), which may consist of full copies of the original entries, or of abbreviated lists, or of both. Records of copyright assignments, when kept in separate volumes from the title entries, are indicated by "A." The approximate size of each record is suggested by the number of volumes, but considerable quantities of material remain as unbound sheets (indicated by "u"). For the original volumes the presence (+) or absence (-) of three characteristics is recorded: whether they usually contain copy-deposit dates after 1831 (D) in addition to title-entry dates; whether they occasionally have printed title pages and labels pasted in (T); and whether they are indexed (I) by names of proprietors and/or authors (n) or by titles (t). The "duplicates" usually lack all three of these features, and exceptions may occasionally be found in the original volumes to the general attributes assigned them here:

• Alabama (Mobile): 1837, 1839, 1858-59 (u).
• Alabama (Montgomery): 1850-51, 1855, 1857-58 (u).
• Arkansas: 1837-70 (u).
• California: 23 July 1851-9 July 1870 (2 vols; -D, +T, -I); 1865-66(u). A 12 Sept. 1854-19 July 1870.
• Colorado: 14 Oct. 1864-6 May 1870 (1 vol; -D, +T, -I).
• Connecticut: 10 Sept. 1804-9 July 1870 (4 vols; -D, +T, +In); 1831-53 (3 vols + u). A 29 March 1837-2 Sept. 1868 (1 vol).
• Delaware: 1849-54, 1855-56, 1865 (u).
• District of Columbia: 16 April 1829 (u), 8 Aug. 1845-6 July 1870 (2 vols; +D, +T, +In); 1818-45, 1848-53, 1855, 1858-64, 1866-67, 1869-70 (u).
• Georgia (Atlanta): 23 Feb. 1849-20 June 1870 (1 vol; -D, +T, -I).
• Georgia (Savannah): 30 Dec. 1800-10 July 1870 (4 vols + u; -D, +T, -I); 1856-59 (u).
• Illinois (Chicago): 24 Aug. 1821-7 July 1870 (7 vols; +D, +T, +In); 1861-63, 1865-69 (u).
• Illinois (Springfield): 9 July 1855-12 July 1870 (1 vol; +D, -T, +In); 1864-69 (u).
• Indiana: Oct. 1822-24 June 1870 (3 vols; +D, -T, +In); 1853-60, 1866-67 (u).
• Iowa: 20 May 1868-6 July 1870 (1 vol; -D, +T, -I).
• Kansas: 1865-69 (u).
• Kentucky: 22 Nov. 1860-8 July 1870 (1 vol; -D, +T, +Int); 1807-45, 1849-55, 1857-70 (u).
• Louisiana: 14 Nov. 1833 (u), 25 March 1835, 17 Nov. 1851-23 Sept. 1856, 13 Aug. 1863-17 June 1870 (3 vols; +D, +T, -I); 1832, 1838-44 (u).
• Maine: 14 Aug. 1790-1 July 1870 (2 vols; +D, -T, +In); 1792, 1819-28, 1832-55, 1857-69 (2 vols + u).
• Maryland: 24 Feb. 1831-5 July 1870 (6 vols; +D, -T, +In); 1831-55, 1867-69 (2 vols + u).
• Massachusetts: 10 July 1790-18 June 1870 (45 vols; +D, -T, +In except Feb. 1804-July 1806, April 1814-1825, 1829-30); 1831-70 (13 vols + u). A 1834-70 (4 vols).
• Michigan (Detroit): 5 May 1824-9 July 1870 (5 vols; +D, +T, -I); 1833-70 (5 vols + u).
• Michigan (Grand Rapids): 23 Sept. 1863-18 June 1870 (2 vols; -D, +T, -I); 1863-66 (u).
• Minnesota: 20 July 1858-4 June 1870 (2 vols; +D, +T, +In after 24 Sept. 1866); 1858-62 (u).

---

123




• Mississippi: 12 Aug. 1850-27 April 1870 (1 vol; -D, +T, -I).
• Missouri (Jefferson City): 10 Oct. 1857-22 June 1870 (1 vol; -D, -T, -I); 1840-49, 1852-53, 1860, 1865-70 (u). A Dec. 1869-May 1870 (1 vol).
• Missouri (St. Louis): 24 March 1857-9 July 1870 (2 vols; +D, -T, +In); 1854-59, 1861-62 (u).
• Nebraska: 29 Feb. 1868-13 July 1870 (1 vol; -D, -T, +In).
• New Hampshire: 4 Sept. 1848-21 June 1870 (2 vols; +D, -T, +Int); 1848-61, 1867-70 (u).
• New Jersey: 8 Jan. 1846-9 July 1870 (2 vols; -D, +T, -I); 1863-69 (u).
• New York (Brooklyn): 24 March 1865-9 July 1870 (2 vols; +D, +T, +Int). A 29 Dec. 1866-21 Oct. 1869 (1 vol).
• New York (Buffalo): 21 Sept. 1813-8 July 1870 (16 vols; +D after June 1861, -T, +In); 1853-68 (3 vols + u).
• New York (New York City): 30 April 1791-11 July 1870 (112 vols; +D, +T before 1834, +In through 4 Feb. 1831 except for 4 May 1820-11 Sept. 1824, +Int after 14 Feb. 1831); 1831-62, 1869 (136 vols + u; + I 1831-53). A 4 Oct. 1834-7 Oct. 1870 (3 vols; +Int).
• North Carolina (Cape Fear): 30 Sept.-14 Nov. 1853, 4 Jan. 1859(u), 28 Nov. 1865-10 May 1870 (1 vol; +D, +T, -I).
• North Carolina (Pamlico): 1865-68 (u).
• Ohio (Cincinnati): 12 March 1806-9 Oct. 1828, 13 Sept. 1841-9 July 1870 (10 vols; +D from 1841, +T before 1828, +In from 1841, +Int from 1859); 1816-53, 1859-70 (2 vols + u). A 18 April 1843-13 April 1870 (2 vols).
• Ohio (Cleveland): 20 March 1855-20 June 1870 (3 vols; +D, -T, +In); 1859, 1861-69 (u).
• Pennsylvania (Erie): 1 April 1867-1 Aug. 1870 (1 vol; +D, -T, -I).
• Pennsylvania (Philadelphia): 9 June 1790-8 July 1870 (36 vols; +D, -T, +Int 1805-31, +In after 1831); 1831-38, 1842-46, 1850-59, 1861-69 (26 vols; +I 1831-53). A 12 Aug. 1834-10 May 1870 (2 vols).
• Pennsylvania (Pittsbugh): 6 Jan. 1819-5 July 1870 (5 vols; +D, +T, -I); 1839-41, 1856-60, 1863-68 (1 vol + u).
• Rhode Island: 23 April 1831-23 June 1870 (2 vols; +D, +T, +In); 1831-59, 1861-70 (1 vol + u).
• South Carolina: 1831-59, 1866-70 (u).
• Tennessee (Knoxville): 1831, 1834, 1855, 1857-60 (u).
• Tennessee (Memphis): 31 Oct. 1865-22 June 1870 (1 vol; -D, +T, +In).
• Tennessee (Nashville): 5 Jan. 1865-6 June 1870 (1 vol; -D, +T, +Int); 1831-34, 1839-55, 1857 (u).
• Texas (Austin): 1857 (u).
• Texas (Galveston): 5 June 1867-8 April 1870 (1 vol; -D, -T, -I).
• Texas (Tyler): 1858 (u).
• Vermont: 21 March 1831-28 May 1870 (1 vol; +D, +T, -I); 1831-59 (1 vol + u).
• Virginia (Norfolk): 15 Sept. 1837-20 April 1870 (1 vol; -D, -T, -I); 1831-32, 1844-59 (u).
• Virginia (Richmond): 9 May 1867-22 June 1870 (1 vol; -D, -T, -I); 1831-32, 1834-36, 1838-39, 1847-48, 1864, 1866-70 (1 vol + u).
• Wisconsin: 26 Dec. 1848-2 June 1870 (1 vol; +D, -T, +Int); 1848-55, 1857, 1860-69 (u).

B. Records Not in the Library of Congress

• District of Columbia: Title pages, 1854-63. In the National Archives — Record Group 21, Series NC-2, Entry 40.
• Kentucky: Copyright entries, 13 Sept. 1800-15 July 1864 (2 vols). In the United States Courthouse, Frankfort. See American Imprints Inventory, No. 38 (1942).
• Maryland: Index (by proprietor and title) to copyright record, 1831-56 (2 vols); copyright assignment record, 1834-70 (1 vol). See Inventory of Federal Archives in the States, Series 2 (Federal Courts), No. 19 (1938), p. 18, items 127-28.
• North Carolina: Copyright entries, 4 March 1796-30 July 1802, 2 Oct. 1811-10 Sept. 1831, 13 Aug 1857 (2 vols). In the National Archives — Record Group 21, Series NC-21, Entry 122.
• Ohio: Copyright entries, 20 Nov. 1829-29 Oct. 1842 (1 vol). See Inventory of Federal Archives in the States, Series 2, No. 34 (1938), p. 81, item 691.

---

124




• Oregon: Copyright entries, 28 Dec. 1865-18 April 1870 (1 vol). In the Federal Records Center, Seattle, Washington 98115. (Microfilm in Library of Congress)
• Rhode Island: Copyright entries, 1790-1867; card index by author and title; record of copyright assignments, 21 Jan. 1842-20 Jan. 1865 (1 vol). See Inventory of Federal Archives in the States, Series 2, No. 38 (1938), pp. 7, 29 (items 39, 195-6).
• South Carolina: Copyright entries, 8 July 1794-12 Nov. 1858 (1 Vol.). In the South Carolina Library, Columbia 29208. (Microfilm in Library of Congress.)
• Texas: Copyright entries, 1840. In the Texas State Library, Austin (Letter Book I, 247). See Journal of Southern History, XII (1946), 215-16.
• Virginia (Richmond): Copyright entries, 1864-65. In the Virginia Historical Society. See Virginia Magazine of History and Biography, XX (1912), 425-29.

C. Published Records

• Alabama, 1861-65: Raymond V. Robinson, "Confederate Copyright Entries," William & Mary Quarterly, 2nd ser., XVI (1936), 248-66. [122 entries, including entries from Georgia, Mississippi, and North Carolina]
• California, 1851-56: Edith Margaret Coulter, "California Copyrights, 1851-1856, with Notes on Certain Ghost Books," California Historical Society Quarterly, XXII (1943), 27-40. [118 entries]
• California, 1851-62: Robert Greenwood, S. J. Suzuki, Marjorie Pulliam, "California Copyrights, 1851-1862," in California Imprints, 1833-1862 (1961), pp. 480-504. [293 entries]
• Georgia, 1861-65: See Alabama above.
• Illinois, 1821-50: Douglas C. McMurtrie, "Early Illinois Copyright Entries, 1821-1850," Bulletin of the Chicago Historical Society, II (1937), 50-61, 92-101. [84 entries]
• Kentucky, 1800-54: John Wilson Townsend, Supplemental Check List of Kentucky Imprints 1788-1820 (American Imprints Inventory, No. 38, 1942), pp. 119-97. [448 entries]
• Massachusetts, 1800-09: Ruth Shaw Leonard, "Transcript of the Copyright Entries for the District Court of Massachusetts, 1800-1809, with Bibliographical Notes," in A Bibliographical Evaluation of the Copyright Records for the United States District Court of Massachusetts, 1800-1809 (Master's thesis, Columbia University, 1944), pp. 69-176. [342 entries]
• Michigan, 1824-29, 1837-70: William L. Jenks, "Michigan Copyrights," Michigan History Magazine, XI (1927), 110-43, 271-87, 445-58, 630-52; XII (1928), 108-24, 584-89, 740-43; XIII (1929), 121-26, 555-59; XIV (1930), 150-55, 311-13; XV (1931), 126-29. [1472 entries — but the actual total is somewhat smaller, since certain entry numbers are skipped]
• Mississippi, 1850-70: Charles F. Heartman, "Mississippi Copyright Entries, 1850-1870," Journal of Mississippi History, II (1940), 79-87. [71 entries] See also Alabama above.
• North Carolina, 1861-65: See Alabama above.
• Pennsylvania, 1790-94: Frederick R. Goff, "Pennsylvania Registrations for Copyright, 1790-1794," in "The First Decade of the Federal Act for Copyright, 1790-1800," in Essays Honoring Lawrence C. Wroth (1951), pp. 113-28. [80 entries]
• Rhode Island, 1790-1816: Howard W. Preston, "List of Rhode Island Books Entered for Copyright, 1790-1816," Rhode Island Historical Society Collections, XIII (1920), 69-72, 95-101. [40 entries]
• Texas, 1840-41: Andrew Forest Muir, "Patents and Copyrights in the Republic of Texas," Journal of Southern History, XII (1946), 204-22 (esp. pp. 215-16). [3 entries]
• Virginia (Norfolk), 1837-70: Barbara Harris and John Cook Wyllie, Norfolk Copyright Entries 1837, 1851-3, 1856-7, 1858-9, 1864, 1866-71 [i.e., 1870] (Charlottesville: Bibliographical Society of the University of Virginia, 1947), 48 pp. [82 entries]
• Virginia (Richmond), 1790-1844: James H. Whitty, A Record of Virginia Copyright Entries (1790-1844) (Richmond: Davis Bottom, 1911), 54 pp. Also printed as a supplement (52 pp.) to the Seventh Annual Report of the Virginia State Library, 1909-10 (1911). [250 entries, fully indexed]
• Virginia (Richmond), 1864-65: "Confederate Copyrights," Virginia Magazine of History and Biography, XX (1912), 425-29. [28 entries]

Notes