Acknowledgments

I GRATEFULLY ACKNOWLEDGE the assistance of the Amherst College Library, Willis E. Bridegam, Librarian, John Lancaster, Special Collections Librarian, and of the Houghton Library of Harvard University,W. H. Bond, Librarian, Rodney G. Dennis, Curator of Manuscripts.  Without the fine cooperation of these libraries, this edition of Emily Dickinson's manuscript books would not have been possible.  They allowed close study of the holographs, including a period of study in which manuscripts from both collections were together; like the Library of Congress, to whom I also express thanks, they gave permission for manuscripts they hold to be photographed by the Meriden Gravure Company.       I have relied upon generous assistance from many colleagues and friends. I am particularly indebted to LaVerne Goman and Dirk Stratton, who helped to prepare copy and to read proof, to G. Thomas Tanselle, who responded to requests for bibliographical advice, to Mary L. Hampson, who made available papers of Susan Dickinson and Martha Dickinson Bianchi, and to Richard B. Sewall and Jay Leyda, distinguished Dickinson scholars, who suggested a facsimile of the fascicles and gave steady encouragement.       Part of my research was supported by a summer stipend from the National Endowment for the Humanities and by a leave from Whitworth College.  Little, Brown and Company and Houghton Muffin Company kindly granted permission for the reproduction of poems originally published by them.  The University of Wisconsin Press allowed the inclusion of a chart, in revised and extended form (Appendix 2), which appeared in my book The Editing of Emily Dickinson: A Reconsideration. A few poems typeset in lieu of facsimile reproduction are reprinted by permission of the publishers and the Trustees of Amherst College from Thomas H. Johnson, editor, The Poems of Emily Dickinson, Cambridge, Mass.: The Belknap Press of Harvard University Press, Copyright © 1951, 1955, 1979, by the President and Fellows of Harvard College. Like all other Dickinson studies since 1955, my work is indebted to the fundamental contributions of Mr. Johnson and his colleague Theodora Ward. R.W.F.

INTRODUCTION

EMILY DICKINSON, although she did not publish, wrote nearly eighteen hundred poems and organized the largest portion of them with her own form of bookmaking: selected poems copied onto sheets of letter paper that she bound with string. In her isolation and poetic silence, these manuscript books, known as fascicles, may have served privately as publication, a personal enactment of the public act that, for reasons unexplained, she denied herself.  In time the poems became an extended letter to the world, gradually published after her sister, Lavinia, upon finding the manuscripts, set about with determination to get them printed.  Yet no edition of the Dickinson poetry has followed the fascicle order; indeed for much of the complex manuscript history the fascicles have been in disarray, divided between families and, finally, between libraries. This edition makes the manuscript books of the poet available for the first time, restored as closely as possible to their original order and, through facsimile reproduction, presented much as she left them for Lavinia and the world.       A facsimile edition is of particular importance to Dickinson studies, for the manuscripts of this poet resist translation into the conventions of print.  Formal features like her unusual punctuation and capitalization, line and stanza divisions, and display of alternate readings are a source of continuing critical concern.  Because she saw no poem through the press and left her manuscripts unprepared for print, judgments must he informed by the manuscript conventions themselves.  Perhaps no less important, interest has developed in the fascicles as artistic gatherings — as gatherings intrarelated by theme, imagery, emotional movement. In general, we need to understand why she assembled the fascicles — by what principles and for what purposes — and to have them available in the way she viewed them. The variorum edition, The Poems of Emily Dickinson, edited by Thomas H. Johnson, translated the mechanics of the poems into conventional type and, in presenting them chronologically, obscured the fascicle structure.  Such an edition, though essential, does not serve the same purposes as a facsimile of the fascicles.       One motive Emily Dickinson had in constructing these books was to reduce disorder in her manuscripts. As she copied poem after

poem onto uniform sheets of stationery, she destroyed earlier versions, so that today few worksheets survive for poems included in the fascicles.  Each bound unit replaced multiplicity and confusion.  A sense of the disorder thus forestalled may he gained by examining the manuscripts left from the last years of her life, after she had stopped copying onto fascicle sheets. Mabel Loomis Todd, the first Dickinson editor, called them "scraps" — a profusion of shapes, sizes, and materials, from brown paper sacks and used envelopes to notebook pages and the backs of recipes.  The growing number of fascicles must have made tangible to Emily Dickinson her poetic achievement, but as they multiplied, the simple order they provided must have become an increasingly difficult virtue.  Because she did not number or otherwise label the fascicles, did not index them or apparently maintain them in a particular order, one may wonder how she found her way among them. Except for browsing, ease of access could only have come from living among them as her own society.       However difficult, the fascicles did serve as a record from which Dickinson made copies to send to friends, the only external form of publication she allowed herself during her lifetime.  The earliest fascicles have no alternate readings, the first such unresolved choice appearing in Fascicle 5, with only five others through Fascicle 10.  Since almost all the poems known from these years were entered in the manuscript books, her earliest goal appears to have been a systematic and comprehensive record of completed poems.  But variants developed as she made copies for friends, and about 1861, some three years after she began assembling manuscript books, alternate readings became abundant in the fascicles.  In many instances, as may be seen in the facsimiles, she returned to this record, sometimes years later, to make further revisions, usually in pencil, whereas the initial writing had been in ink.  What started out as a comprehensive record of completed poems, serving as a source for additional copies, broadened to include intermediate stages and became in a sense a continuing workshop where, in producing a new copy for friends or in reading among the poems, she would enter the specific poetic process again.       The manuscripts Lavinia Dickinson found included forty bound fascicles and enough unbound fascicle sheets for several others — plus the worksheets, intermediate drafts, and miscellaneous fair copies Mabel Todd called "scraps." The Todd sequence of numbers for these manuscripts was 1-38, 40, 80-112. Todd worked on the manu-   x

scripts over several years, and the sequence evolved slowly. Although numbers were not placed directly onto manuscripts before 1891, there had been an earlier sequence, probably recorded on envelopes: 1-10 for the bound fascicles.  From this, 39 became 80, leaving a permanent gap at 39.  Before that, the fascicles considered to be 14, 33, 35, 36, and 38 had been transferred to 81-85, but Mabel Todd had filled these gaps with a group of displaced sheets (Todd 14) and four groups from the unbound manuscripts in Lavinia's possession (Todd 33, 35, 36, 38).  The editor took possession of 80-85 and, from the unbound manuscripts, assembled additional packets, numbering them from 86 on.  The break between 40 and 80 does not represent lost manuscripts, as has been feared, but rather an attempt to create a separate collection whose numbering, by doubling the tens digit, was also separate.  By 1891, when Todd indexed the poems, she had reached 98, though 96, presumably because of manuscript shifts, was vacant.  She placed the remaining manuscripts in envelopes which her daughter, Millicent Todd Bingham, during the 1930s numbered 96, 99-112.       This facsimile edition relinquishes the Todd sequence and restores the original forty fascicles, arranged chronologically and renumbered.  The unbound fascicle sheets, grouped here by similarity of paper and date, are called sets to distinguish them from the poet's completed books.  The other manuscripts Lavinia found, as well as fair copies sent to friends, are not included, for the intention has been to present only the manuscripts Emily Dickinson included, or copied as if for inclusion, in her manuscript books.  The bound units are present in facsimile as Fascicles 1-40, the unbound sheets as Sets 1-15.       For both the fascicles and the sets Emily Dickinson wrote in ink on sheets of letter paper already folded by the manufacturer to produce two leaves.  (Two sets consist of an unfolded paper, tablet style.)  She wrote on each of the four pages, rarely leaving one blank.  A sheet was ordinarily self-contained: in two instances a poem over-flows onto the next sheet; in seven instances onto half-sheets inserted expressly for overflow; and in six instances onto a small slip of paper pinned or bound into place (Appendix 6).       To assemble a fascicle she stacked several copied sheets on top of one another.  She started out in 1858-59 consistently using four sheets per book and ended in 1863-64, when she ceased binding, with a norm of six (Appendix 4). The sheets, copied separately, were not   xi

inserted inside each other but were stacked and then stab-bound; she punched two holes through their sides, from front to back, and threaded them with string, tied on the front. That completed the construction of a fascicle, as there was no title page or pagination, and no contents list or other apparatus to lead to specific poems.       When she stopped binding fascicle sheets about 1864, it was a conscious change rather than an action deferred and then forgotten.  It had been anticipated as early as 1862, when she began leaving a few miscellaneous or leftover fascicle sheets unbound (Sets 1-4), whereas earlier such sheets had been bound, even if into a grouping of mixed paper or date (Appendix 3).  With Set 5 the change was complete.  She never bound sheets again, and she altered her technique accordingly.  During construction of the fascicles she had come to prefer the use of half-sheets to accommodate overflow when she ran out of space.  As of Set 5 she abandoned this technique, which requires binding to maintain the association, and reverted to a much earlier method of pinning a small slip of paper onto the individual sheet to provide, in clear and constant association, the space needed to complete a poem.  Her unit, which in one sense had always been the sheet, became more so.       Dickinson's patterns of copying were otherwise much the same, especially through the 1860s, when she used batches of stationery sufficient to make up several bound fascicles had she wanted to.  In Fascicles 1-40 there is an impulse toward uniformity of paper, ink, and handwriting within each unit.  Even during the difficult year of 1861, when she faltered, mixing papers and dates in several fascicles, the variation occurs within a similarity of embossed design or of size and character.  With the annus mirabilis of 1862 her practice became regular, and uniformity continued in the unbound sheets (Appendix 3).  Correlation of paper and date, observed in the bound fascicles, has thus been used to gather the unbound sheets into Sets 1-15.   Emily Dickinson may have stopped binding because, once she had survived the crisis and drive of 186 1-1863, her need for self-publication declined, and with it the desire to leave an organized legacy for the world. That she continued to copy fascicle sheets without binding them suggests that she found the bound books difficult to use. In growing number (Fascicles 1-40 contain over eight hundred poems), they have a disorder of their own. By 1864 unbound sheets may have been easier for her to to use — connected perhaps to the eye trouble of 1864 and 1865 that forced her to spend several months in   xii

Cambridge under the care of a physician.  In any event, the unbound sheets still prevented the disorder of the "scraps" and provided a record from which subsequent copies could be made and onto which she could make revisions.  In the later 1860s, her poetic drive somewhat spent, Emily Dickinson stopped copying fascicle sheets.  The poems of 1867-1870 are few and are unorganized. For a time in the 1870s she revived such copying (Sets 8-15) but finally gave herself up to the proliferation of shapes and sizes of her worksheets and miscellaneous manuscripts.  It has been suggested that in the 1880s she may have returned to making fascicle sheets once more, this time in pencil, but a study of the paper and of the copying and folding patterns of these manuscripts shows them to be more like worksheets and miscellaneous manuscripts than fascicle sheets. The attempts at bookmaking appear to have ended with the 1870s. AFTER THE POET'S DEATH IN 1886 the Dickinson manuscripts suffered disarray and some mutilation.  Thomas H. Johnson's variorum edition was the first attempt to identify the original fascicle arrangement. My book, The Editing of Emily Dickinson: A Reconsideration, revised the variorum order, including the addition of a number of missing poems.  This revision was based on the evidence of the holographs and of a variety of secondary materials — nineteenth-century transcripts of the poems, the typewriters and papers used, and the diaries, journals, and correspondence of the early editors.  The revision proposed has been supplemented by a series of articles in the Papers of the Bibliographical Society of America and the Harvard Library Bulletin in which distribution problems unresolved or unperceived at the time my book was published have been treated.  The changes affect Todd numbers 2, 3, 6, 10, 14, 20, 26, 27, 29, and 80.       The Editing of Emily Dickinson, although an attempt to identify the proper groups of sheets for the fascicles, did not address the sequence of sheets, and thus of poems, within each unit.  Sufficient evidence is available to do so, and the internal sequence for each of the forty bound fascicles has been established in this facsimile edition.  The 1891 notebook in which Mabel Todd recorded the fascicle location of poems, although of uncommon value in determining the proper sheets, is arranged alphabetically and does not reveal sequence.  But other editorial lists from this early period do.  The most important of these is an eight-page list, now in the Amherst College   xiii

Library, that Todd prepared in 1889.  When understood, it indicates the order of poems within fascicles so represented.  On occasion the transcripts Todd and her assistant made have been useful in establishing internal sequence.       The primary evidence is from the manuscripts themselves.  Soiling on first and last pages usually identifies the first and last sheets of a group, and the various links afforded by stain offsets, matching smudge patterns, pin impressions, and manufacturing defects like paper wrinkles place one sheet ahead of or behind another.  Puncture patterns, where the needle pierced the paper for binding, and stress effects, caused by the pressure of opening a fascicle against the tension of a stabbed binding, vary within fascicles, with initial sheets differing from subsequent ones in amount of curvature along the fold edge and in the direction and extent of damage to the binding holes.       Usually the available evidence is substantial and mutually corroborative. Fascicle 10 (Todd 37), for example, has five sheets identified through the 1891 notebook, and the eight-page list from 1889 indicates their order to have been H 200 through H 204 (to use the Houghton Library numbering).  This sequence is confirmed by front-page and last-page soiling (H 200 and H 204) and by stain links be-tween H 200 and H 201 and between H 201 and H 202.  The puncture patterns and stress effects are progressive when the sheets are in this order, which, it may he noted, mixes Dickinson sheets dated by the variorum as 1860 and 1861 and does so thus: 1861, 1860, 1860, 1861, 1861. The sequence of dates notwithstanding, the sequence of sheets is clear.       No internal sequence belonging to the poet has been established for the unbound sheets in Sets 1-15.  Transcript patterns indicate that Mabel Todd assembled her editorial groups after transcribing the sheets and that the manuscripts were disordered when she copied them.  Since the sheets were not bound, puncture patterns and stress effects do not exist to provide evidence. Indeed the lack of binding itself suggests that Emily Dickinson did not intend a specific arrangement.  Yet, as with the fascicles, there is a uniformity of paper through the 1860s that shows her working with large batches of stationery (Appendixes 3 and 4), and a few stains and pin impressions link sheets when they are gathered into sets by paper and date.  The sets that contain more than one sheet have generally been arranged by variorum date and within that, where stains and pin impressions provide links, certain sheets have been brought together. In Set 8b   xiv

where a distinctive stain moves progressively through ten sheets, they have been arranged in the order of the stain.       The dating attempted by Theodora Ward and Thomas H. Johnson has been used with modification.  In certain instances manuscripts from the same bound fascicle, of identical characteristics but arbitrarily separated from each other, were assigned to different years, and other bound fascicles, although not separated, have problematical dating.  After 1864 the variorum chronology is weaker, apparently influenced by the artificial Todd groups — broadly based on decade with some attempt at grouping by paper — and hampered perhaps by restricted editorial access to these particular manuscripts. Variant dates between Fascicle 38 and Set 2 and between Fascicle 39 and Set 3, when there are similarities of paper, ink, and handwriting, are questionable.  So are the ranges of dates, beginning with Set 5, assigned to similar sheets, sometimes linked by stain or pin impressions, that Todd had put into different groups.  By the 1870s (Sets 8-15) the variorum dating is erratic within fascicle sheets, as in Set 14 where one sheet onto which Dickinson copied three poems, one following upon the other, was given three different dates — in reverse chronology to their sequence on the sheet: about 1877, early 1876, and about 1873.       Although variorum dates are included in this edition, additional evidence has been used to arrange the fascicles and sets, sometimes in disregard of those dates (Appendix 3).  Besides characteristics of the handwriting, considerable attention has been given to the distribution of paper and to other formal patterns.  The poet's habits changed over the years, passing through various styles of displaying alternate readings, of using underlining and quotation marks, of employing disjunct leaves, and of accommodating overflow.  Such features have been useful in arranging the fascicles and sets. The resulting order, although open to refinement, should be a clarifying supplement to the variorum.  It may be emphasized that although variorum dates have been applied to the developed sequence, there is necessarily some imprecision about the beginnings and endings of years.       The year of composition for a poem — that is, the date of the worksheets now destroyed — may not have been the year it was transcribed onto a fascicle sheet.  Some lag is to be expected, and there are examples to suggest it — notably poem 174 in Fascicles 8 and 21 and poem 259 in Fascicles 13 and 36. Presumably because she failed   xv

to destroy a worksheet, two years later Emily Dickinson copied each poem a second time, with little or no change.  The pool of worksheets from which she was selecting poems for the fascicles may, thus, have had a range of dates greater than the year in which she was copying poems.  On the other hand, the infrequency of such repetition (Appendix 9) and the lack of worksheets for poems during the years continuously covered by the fascicle sheets, through the mid-1860s, suggest that her copying was systematic and may have kept up fairly well with her poetic production. A NUMBER OF PEOPLE besides Emily Dickinson have written on the manuscripts.  Almost always in pencil, much of this writing falls into recurrent patterns identified in this Introduction; instances which are not part of a pattern, or which have been erased, are identified in the Notes. The earliest writing except for the poet's was probably that of her sister-in-law, Susan Dickinson, to whom Lavinia turned after finding the manuscripts. Her letters D, F, L, N, P. S, and W occur at the head of poems in several fascicles and a few sets (see Fascicles 14 and 29). The letter H, written with a different pencil and at a different time, probably was also hers (see pp. 227, 381). The meanings these had for Sue are not fully clear, though some of them — W for a poem about the wind, D for death, and N for nature — seem to have been topic indicators.  A few may refer to Sue herself or to other persons (see pp. 67 1-672).  Her numbers 1, 2, and 3 also appear on the manuscripts (see Fascicle 11).  These were probably a form of ranking or other indication of interest and may have been a version of the crosses (x, xx, and xxx) found throughout the manuscripts. Tradition has ascribed the crosses as Lavinia's marks of choice, but they are similar to marks on documents in the Houghton Library edited in Sue's hand as well as to marks on manuscripts Emily Dickinson sent her.  They often appear in the fascicles and sets close to Sue's letters and in conjunction with her numbers, as in x2 and xx3.  Like the numbers, the crosses were made at more than one time, with one or two crosses increasing to two or three in a pattern similar to the increase suggested by the numbers.       After about two years without publication, Lavinia retrieved a box of manuscripts from Sue and asked Mabel Todd for help.  About 1890 she turned another box of them over to her. Whereas Sue's approach had been to read through the manuscripts marking poems for possible publication, Todd decided to copy everything in the first   xvi

box and to make a selection afterward.  In organizing and indexing the manuscripts in 1891, she numbered them in blue pencil (1-38, 40, 80-85) and in lead pencil (86-95) and made two marks in blue pencil - an x on the last verso of the bound fascicles in her possession (80-85) and an occasional mark in the upper right corner of the first recto of others.  But she chiefly wrote on the manuscripts in preparation for her assistant, Harriet Graves, the principal copyist for Fascicles 11, 19 (part), 21, 22 (part), 23, 24, 27, 30, 35, and 36.  Before Graves took a fascicle, Todd would determine which of the unresolved readings were to be copied. She crossed out words and wrote on the manuscripts alternates Graves was to include.  At times notes instructed Graves to forgo copying a particular poem ("Not Copy").  Before she finished, Todd herself copied part of the second box ("scraps") and part of some additional groups of manuscripts brought to her.       After Mabel Todd had published three series of Emily Dickinson's poems (1890, 1891, 1896), a quarrel with Lavinia over a piece of land, aggravated by other fears and hostilities, ended her work on the poems.  Manuscripts numbered 80 and above remained in her possession, the others in Lavinia's. Undaunted, Lavinia turned to another friend, Mary Lee Hall, to copy poems.  To know what to copy, Hall had to determine what had been published already. Two marks, indicating prior publication or the lack of it, appear to have been hers.  Working from the first series, she wrote 1- to the left of poems in manuscript (see pp. 129, 138).  But it was easier to compare the manuscripts to the published editions, instead of vice versa, so she shifted her approach, using a zero for poems not published in the first three series (see p. 44).  Hall kept the manuscripts from late 1898 until, concerned at Lavinia's poor health, she returned them about June 1899, only a few mouths before Lavinia's death in August.       The division of the manuscripts between the Dickinson and Todd families continued into the twentieth century.  Each family published selections from them before the manuscripts came to rest, still divided, in the Houghton Library of Harvard University and in the Amherst College Library.  Martha Dickinson Bianchi, the poet's niece, issued The Single Hound (1914) from manuscripts sent to Sue; Further Poems (1929); Unpublished Poems (1935); collected editions (1924, 1930, 1937); and biographical treatments that included poems. From the Todd portion, Millicent Todd Bingham, daughter of the first editor, published Bolts of Melody (1945) and some articles. Like   xvii

Mary Lee Hall, both descendants needed to know which poems remained unpublished.  To indicate which ones, Bianchi wrote No on many manuscripts in her possession; she also conducted several counts and recorded her tallies (see p. 285).  For Bingham the letter u indicated that a poem remained unpublished, To show publication she used I, II, and III for the three nineteenth-century volumes and abbreviations or short titles, such as SH, EP, UP, and Letters, for the others (see Fascicle 1).  Both libraries numbered the manuscripts they received. INTRODUCTIONS TO THE FASCICLES AND SETS list the poems contained in them and describe their date, paper, and makeup.  The Todd numbers, as applied in 1891, are recorded for each (compare Appendix 2).  The description indicates whether the paper is laid or wove, ruled or unruled, and includes the nature of the design, if embossed.  Common color names have been used: white, cream, bluish white, greenish white, grayish white, and beige.  These may be compared, respectively, to the ISCC-NBS Color-Name Charts (Supplement to National Bureau of Standards Circular 553) for white (Centroid 263), yellowish white (92), bluish white (189), greenish white (153), light gray (264), and yellowish gray (93). In most instances the actual color is much lighter than the standard.  Because of minute but frequent variation in size from sheet to sheet of the same paper type, overall dimensions, and those for the embossed design, are summarized in Appendix 5.       Although the fascicles and sets were not printed and folded like a book, the conventions of standard bibliographical description can be used to express the patterns of sheets, half-sheets, and attachments.  The descriptive formulas begin with a statement of format, almost always folio (2o) since most of the sheets had been folded by the manufacturer to produce two conjugate leaves.  In two sets Dickinson used a tablet style paper without a fold (1o).  Because she did not mark sheets, their numbers in the formulas have been supplied editorially and thus recorded in italic type.  A superscript figure indicates the number of leaves in each sheet. 2o: 1-42, 8 leaves. This description of Fascicle 1, typical of many, states the format to be folio (2o) and records four sheets (1-42), unmarked, for a total of eight leaves.   xviii

When disjunct leaves, half-sheets, are present, they are included in the sequential numbering; a subscript figure indicates their singleness. 2o: 1-42 51 62 71, 12 leaves.  Slip pinned to 42v (H 1462v). This description shows that Fascicle 19 begins with four folio sheets (1-42) followed by a single leaf (51) then a full sheet ( 62), with another single leaf (71) at the end.  Reference is made by italic number, with specific leaves identified by a subscript figure and recto and verso by r and v. In Fascicle 19 Emily Dickinson pinned a slip to the verso of the second leaf of the fourth sheet (42v), whose Houghton Library number is 146.       Although a few manuscripts are missing from the fascicles and sets (Appendix 10), we know most of their contents and where they belong.  Typeset texts have been included for these among the facsimiles; such poems are listed in the contents preceding the fascicle or set.  Because the manuscript books are unique, the formulas cannot describe an "ideal copy" developed by comparative collation, but they do describe an ideal assembly that includes the manuscripts now missing.  Absence is noted following the statement of collation, as in Fascicle 25 where the last sheet lacks its first leaf: 2o: 1-62, 12 leaves (61 missing). Citations to literature discussing the identification and placement of missing manuscripts have been included in the Notes.       Seven fascicle sheets were folded inside out when inventoried for the Houghton Library in 1950 or in the 1930s by Millicent Todd Bingham.  One of these, H 19 in Fascicle 24, has been reversed in this edition, for a study of the Graves transcripts, as reported in the Harvard Library Bulletin (1980), showed that as late as 1889 the sheet had been right side out.  The others have been left as inventoried and so reported following the statement of collation in Fascicles 2, 18, and 31, and Sets 7, 9, and 10.  Emily Dickinson may have folded all these sheets backward after copying poems onto them.  Some of them she certainly did: the "improper" second verso of one sheet is soiled from having served as the final page of Fascicle 31; another shows pin impressions on its "improper" first recto from the sheet bound ahead of it in Fascicle 18.  The six inversions may be left without dividing any poem, though the order of poems is affected.   xix

The contents lists indicate the manuscript number assigned by the repository (prefixed by A or H), the poem numbers in the variorum, and the first lines of the poems. H 23      495.   It's thoughts - and just One Heart -               337.   I know a place where Summer strives               496.  As far from pity, as complaint - Because the Houghton Library sequentially numbered pieces, usually a sheet with two leaves, and the Amherst College Library numbered leaves, while preserving the Todd numbering, their treatment of similar items was often dissimilar (H 23 as compared to A 80-1 and 80-2).  The Amherst numbers sometimes differed from the corresponding Bingham inventory numbers recorded in the variorum.  One manuscript reproduced in this edition was given to the Library of Congress by Millicent Todd Bingham.  The first lines of poems, with a few corrections, are quoted from the variorum.       Manuscript numbers appear at the top of facsimile pages the first time applicable; they are not repeated.  The variorum poem numbers are displayed in the outer margin at the beginning of poems. Symbols in the inner margin — occasionally in others — indicate the structure of the manuscript reproduced. For the usual sheet of two leaves, four pages, the symbols are: (See pp. 3-6 and pp. 25-28, respectively.) The broken line also appears alone — for a half-sheet used independently (see pp. 155-156) and, in general, for edges torn or cut.  A solid vertical line in the inner margin indicates a manufacturer's trim edge (see Set 8); other margins, except as noted, may he assumed to represent the original trim.   xx