GENETIC JOYCE STUDIES - Issue 3 (Spring 2003)



The Tulsey Town Treasure Trove: Preliminary Report on the Rediscovered Finnegans Wake Proofs


Robbert-Jan Henkes and Erik Bindervoet


- Hey, the 'lost last proofs' have been found! There are 1500 pages of unexplored Finnegans Wake-materials in Tulsa! Lying and waiting to be studied in what is called The Paul and Lucie Léon Collection in the McFarlin Library in the University of Tulsa, Oklahoma. For the sake of the papers themselves, for the insights in the compositional process and of course for fresh transmissional departures to incorporate into our 'corrected translation'.

- Tulsa? How did they end up in Tulsa?

- I don't know for sure but apparently the son of Lucie Léon, widow of Paul Léon, Joyce's closest amanuensis during the years of composing Finnegans Wake, sold the collection and Thomas Staley managed to acquire it for Tulsa in 1984.

- Tom Staley the founder of the James Joyce Quarterly in 1963?

- Himself. Up till now the collection was relatively unknown. There was an exhibition in 1985, with a description of the treasures by Bernard Benstock, but then genetic criticism was still in its infancy, so the information lay dormant, as an alien in its egg, a smoking gun in its holster.

- And the existence of the Wake proofs was unknown even to the editors of the allencompassing James Joyce Archive when they assembled all manuscripts, typescripts and proofs for their 1978 edition, and hence were not included and therefore remained even more unknown than they already were. And unknown makes unloved, as we say in Holland.

- Until Luca Crispi, the Columbus of Joyce studies, came along and rediscovered the treasures for the Joyceans, and especially Wakean world, and even more specifically, for us, the translators into Dutch of Finnegans Wake. And now the Tulsey Town Treasure Trove has yielded the very very last page proofs of Finnegans Wake, from late 1938 and early 1939, mere months before the novel hit the bookshops with a thunderclap (4 May 1939).

- All right! Right on!

- This makes Tulsa the fourth biggest Joyce-repository in the world and in the universe.

- What's in it then? Should we go there to see for ourselves?

- Of course! Fortunately a very helpful Lori Curtis, librarian of the Special Collections already provided us with photocopies, so we were able to study the material before we got to see the original documents live on the fifth floor in the McFarlin Library during the 1132nd International Bloomsday Conference, held in Tulsa and entitled Post-Colonial Joyce (16-20 June 2003).

- We navigated our way through approximately 1500 pages of proofs, sometimes annotated, more often not, again in different stages as the proofs were corrected and sent back to Joyce/Léon by the printer, Robert MacLehose in Glasgow, Scotland. First we had to bring some order into the mass of materials, if only for ourselves, lest we couldn't see the wood for the tree. The Tulsa librarian who originally sorted the Paul and Lucie Léon Collection devised an intricate and for us far too difficult system in which the quires or signatures were alphabetically numbered according to the datestamp of the printer.

- We know that Luca Crispi is reordering the material and assigning a better chronology to it, but until that time, here's what we came up with to describe the incredible mass of new material. It is just to get an impression of the chronology and the wealth: the real description will be from the hands and brains of Luca Crispi.

- So our own personal and preliminary chronological arrangement of the proofs runs as follows, being a reconstruction of the final hectic months of Work in Progress.


Tulsa 0 and Tulsa 1

Tulsa 0: the first set of proofs, not annotated, p. 1-355, dated by the printer from July to October 1938.

Tulsa 1: a duplicate set of Tulsa 0, also p.1-355 and dated by the printer July-October 1938, this time heavily annotated. Most of it is in Paul Léon's hand, but obviously more people worked on it, one of whom was Giorgio Joyce. Luca Crsipi undoubtedly can identify the identities in the writer complexus, being versed in Notebook graphology. The revisions of all stages of the Tulsa page proofs are in many colours: green ink, black ink, black pencil, blue ink, red pencil (printer's remarks), while for the final stages of II.2 a red-orange pencil was used as well as red ink. So probably Joyce, as was his wont, had many hands helping him.

From July-October 1938 MacLehose regularly sent proofs to Joyce/Léon, as soon as they had finished four signatures (64 pages), sometimes less, sometimes more. Tulsa 0 is in a virgin state, Tulsa 1 is probably the proof that was sent to MacLehose. There must have been a complete third duplicate set of these proofs, the one on which Joyce himself made revisions, before Paul Léon copied the revisons in his own, neater hand. We know this because a small portion of Joyce's proof has survived and was reprinted in the James Joyce Archive (JJA 51:449-489), the proofs for the Mime, chapter II.1, FW 219-259 (dated 23 September 1938 on JJA 51:487, p. 257).

This circumstance, the survival of a part of both annotated sets of proofs (only eight percent of the entire Wake though), offers us a unique opportunity to check if Léon copied faithfully. He usually did, but there are differences: Joyce changes on FW 219.06 (JJA 51:449) 'taken in taken' into 'token in taken', but Léon copies this as a desired change into the much blander 'taken in token'. And so it is a new transmissional departure we shall have to take in account in our translation.

Sometimes the Léon copy is more complete: FW 224.08 'Towhere byhangs ourtales.' is not on Joyce's copy (JJA 51:449), only on Léon's copy. Does that mean that Joyce was in the immediate vicinity when Léon copied Joyce's emendations?

In this chapter (II.1), besides the 'token in taken', Léon missed two more of Joyce's corrections, but Joyce catches these himself, in the Corrections after the first edition: FW 224.04 'thinker,' -> 'thinker's' and FW 254.04 'the, Ipse' -> 'the Ipse'. Furthermore Léon made two very minor mistakes in an added sentence on FW 256.12-14: Joyce wrote 'sherrygoldies' not 'sherrigoldies' and put a comma after it (JJA 51:486)

The revisions of Tulsa 1 mainly concern the correction of small typos and the addition of a fair number of commas, a practice that Joyce continued in the Corrections. He completes the remaining thunderwords and adds a word or sentence here and there. For instance, he now finishes the first sentence of the book by adding ", {from} ^by^ swerve of shore to bend of bay," Another motif Joyce expands at this stage is the Tom-Tim motif. He takes care to add a Tim whenever there is a Tom mentioned: at FW 27.01, 39.14, 210.18, 215.33, 331.11 and 342.03. And at a later stage, christened Tulsa 4 by us, he adds another two: FW 505.01 & 507.02 (on the second set).

One of the very few instances of Joyce explicitely accepting a typo occurs when he changes 'marrogbones' back into its original intention 'marrowbones' (FW 16.03), but on second thought allows it to stand (by writing 'stet' in the margin), probably because 'marrog' is a happy conflation of 'marrow' and the German 'Marg' and/or Dutch 'merg' that mean the same. In our seven years of translating the Wake and studying the whole of the Finnegans Wake-portion of the James Joyce Archives we came across no more than seven or eight instances of Joyce actually approving of a typo, by using it further on or by reversing his initial correction. On the contrary: we found that Joyce didn't like typos at all: he wanted to be a very meticulous proofreader, but unfortunately he wasn't always.

At this stage the mysterious full stop between "Coldstream" and "Guards" appears for the first time on FW 58.25. Is it an intentional sign? Judging from the actual marginal correction, we think not: Joyce/Léon changes the roman "g" of "guards" into a capital "G" and about one centimeter to the left of it, just under the insertion mark, there is this 'full stop'. Now, if it would have been an intentional full stop, it would have been written much closer to the "G". As it is, we conjecture Joyce/Léon accidentally tipped his green pen onto the page and allowed it to rest for a fraction of a second.

Another interesting transmissional departure occurs on FW 71.24: in "Hooshed the cat from the Bacon" Joyce/Léon crosses out the "c" of "cat" and inserts a "b" twice underlined in the margin, indicating that it should be a capital "B", so as to get: "Hooshed the Bat from the Bacon". Here we'll have to revise our translation.

Next, the proofs were sent back to Scotland, the corrections and additions were processed, and another proof was sent to Joyce:


Tulsa 2

Tulsa 2: the corrected proofs of Tulsa 1, not annotated (except the first page), p. 3-355, dated by the printer 29 November 1938, 6 and 8 December 1938.

The Tulsa 2-proofs were also sent in signatures as soon as some were finished and in three copies, although not everything is extant. There are two copies of page 186 (maybe an error as the copies were made for us?), pages 321-336 and 337-352. Three identical unmarked copies exist of pages 273-288, 289-304 and 353-355.

Tulsa 2 has only one annotation: a new paragraph should start after 'Environs' (FW 3.03).

For the greater part of Finnegans Wake - the printer apparently still doesn't know the title! The signatures are marked with J.W.P., Joyce's Work in Progress - this is the final proof, at least up to page 355, because the rest is missing. There are no more revisions for these pages by Joyce/Léon. But there remain two nasty typographical problems that have to be solved. The marginal notes in chapter II.2 Night Lessons (FW 260-308) and a question of capitalization on FW 353.02-03.

These are dealt with posthaste on what we call Tulsa 2a and Tulsa 3:


Tulsa 2a and Tulsa 3

Tulsa 2a: duplicate set of Tulsa 2 proofs, 3 pages (353-355), with a written request (in Joyce's hand?) in the left upper corner: 'Revise pp. 353-355 when you send the rest of that signature' and (in Léon's hand?) 'only the first letters of each word in capitals - the rest in ordinary type - the Capitals also smaller)-' and underneath this request (in a printer's hand?): 'try sm. caps.'

Tulsa 3: duplicate set of Tulsa 2 proofs, same dating, but only pp. 257-355 survive, annotated.

On Tulsa 1, Joyce had requested the words 'most marsiful' 'aweghost' and 'gracious one' (FW 353.02-03) to be set in capitals. The printer wonders: 'all?' in the margin, twice underlined, but executes the order nonetheless. Joyce doesn't like the resulting capitals and immediately sends back the three pages at some date very soon after 29 November 1938. At least, that is how we see it, because already on 7 January 1939 he receives a revised proof (Tulsa 4), that isn't any good either, as the first letters of the words are capitalized in small capitals. The re-revision of the small capitals into ordinary capitals must have been ordered telepathically, as no further page proofs remain and the first edition has the normal capitalization.

Tulsa 3, a duplicate set of proofs of Tulsa 2 intended for substantial marginal revision of Chapter II.2, is dated 8 December 1938 and headed (in Joyce's hand?): 'Revise pp. 257-320', twice underlined. The placement of the marginal notes was such a mess on Tulsa 1 (the printer had swapped the left and right margins on all left-hand pages), that a real checkup could only start on a new proof. Tulsa 3 is that proof. Joyce takes very special care to align the left marginal comments with the appropriate line of the body of the text, indicating by the way that he really wanted the upper line of the comments to be situated exactly opposite a line of the main text - and that is what the Faber and Faber reprint doesn't do: when Faber and Faber published Finnegans Wake, they scaled down the marginal texts, and what's more: replaced them not at all accurately. Sometimes they stuck two different sidenotes together and sometimes they put the sidenotes facing wrong lines. For any next edition of Finnegans Wake, Faber and Faber should reset the marginal notes using the Penguin reprint of the first edition as a guideline. Even the Compass Edition, sixth printing, 1967 (from the Viking eighth printing 1958), bought on the twentysecond of May 2003 for a mere 7 euros of borrowed money at Barry's Used English Bookshop on the Kloveniersburgwal in Amsterdam, is better in this respect.

Another interesting feature of Tulsa 3 is the request to have the Doodles-drawings remade (in Léon's hand?) 'these geometrical figures have been copied from my drawing in pen - hence their unevenness - if possible redraw them without keeping the uneven lines'. This was never done, and the request was crossed out by a waving line, but by whom? In our translation, and in a future "variorum'-edition, this request can be effectuated, as there is the electronic 'Wake'-font used in the complete A Wake Newslitter cd-rom by Split Pea Press in Dublin. And so we will!

By the way, someone put a "<" in front of practically all footnotes of the chapter. Maybe it was someone checking if all footnotes were there? Or were the signs meant to be taken as corrections. indicating that the footnotes shouldn't be indented? Anyway, indented they remained.

Judging from the speedy resetting of page 353, now with the initials in small capitals, the next proof Joyce/Léon receives is:


Tulsa 4

Tulsa 4: corrected proofs in which the revisions of Tulsa 1 and of Tulsa 2a have been incorporated, p. 353-496, not annotated, dated by the printer 7 January 1939 and (from p. 481) 16 January 1939. Pages 449-480 are missing. Only a single copy survives.

We have to believe that all corrections and revisons made on the non-extant pages 356-628, ideally following Tulsa 1 or 2, have been incorporated flawlessly. There are only two pages on which we can check this, the first two, 353 and 354 (page 355, although extant in Tulsa 1, had no revisons), and yes, the changes 'senate' -> 'senatas' 353.28 and 'fiannaship dook' -> 'fiannaship, dook' have been effectuated (as well as the small capital initials in 'Most Marsiful' etc on FW 353.02-03, see Tulsa 2a).

Also belonging to this proof, because dated and sent at the same time as the second batch of Tulsa 4 (that is, 16 January 1939) is:


Tulsa 5

Tulsa 5 belongs to the same batch of proofs as Tulsa 4 and maybe should be named Tulsa 4 as well, being the continuation of 353-496, also dated 16 January 1939, but these pages, 497-512 are a) in three sets and b) annotated, so we call them for the time being Tulsa 5/1, 5/2 and 5/3. Maybe these proofs are the first post-galley proofs of this part of Finnegans Wake, on a par with Tulsa 0 and Tulsa 1: maybe MacLehose reached this part of Finnegans Wake only now, which would explain that Tulsa 0 and Tulsa 1 seem incomplete sets of proofs. This part apparently hadn't been checked well by Joyce/Léon before.

Tulsa 5/1: 497-512 first set, with only one emendation, on 512.35, 'Tltass' -> 'Tiltass', also on the next two sets.

Tulsa 5/2: 497-512 second set, annotated (in different hands), mainly added commas and corrections of typos, plus three new sentences. Among them the first appearance of the last instance of the 'Title'-motif or 'Ballad of Tim Finnegan'-motif, which was hunted down by Finn Fordham in his exemplary study Mapping Echoland (in: Joyce Studies Annual 11, 2000, p.166-201): '- But there's leps of flam in Funnycoon's Wick. The keyn has passed. Loud lift the keying!' The word 'Loud' is overwritten in Tulsa 5/3 with 'Lung'.

Tulsa 5/3: third set, prepared for the printer, with the same annotations as Tulsa 5/2 (except the Loud -> Lung, see 5/2), marked with 'RUSH' probably by the printer, and 'revise', probably by Léon. The word 'SILENCE.' on FW 501.06 is requested to be put 'in the middle of the space and smaller capitals'. In the first edition the placement is okay and the capitals are a tiny fraction smaller (ca. 0.3 mm): real small capitals wouldn't work, as Joyce/Léon had already found out on Tulsa 2a and Tulsa 4.

Time was getting tighter and tighter and still the list of names in the left margins of FW 306-308 was all wrong.


Tulsa 6

Tulsa 6, two sets of the signature of page 305-320, is the corrected proof of Tulsa 3, dated 17 January 1939 (the unannoted set) and 18 January 1939 (the annotated set). It could be both 14 January 1939 too: Tulsa 4, 5 and 6 came to Joyce as an avalanche. In the left upper corner of the annotated set Joyce writes 'Revise pp. 305-320' and the printer puts a 'revise' stamp next to it and the written word 'RUSH' underlined.

By this time the printer must have been informed about the title, because the mark on the signatures, 'J.W.P.' is crossed out (or is it common practice to delete this part of the signature in the last stage?), but not replaced by for instance J.F.W. It was too late to delete the letters on all signatures: the 1st edition still bears a 'J.W.P.' on page 33.

The revisions only concern the list of names in the left margins of page 306-308. Page 308 was done well on Tulsa 3, but now page 306 was a mess, forcing Paul Léon to make a list indicating which name should face what words: 'Ovid should be in a line with = The Voice of Nature...' etcetera. And the main text of page 307 has an indentation where the marginal names take up too much space. Joyce/Léon comments on the sideline: 'I do not think that the indentation of the text is a good solution. Notes in a textbook can never make the text move Can you see any other solution?'

Apparently this is the very last work on Work in Progress before it became Finnegans Wake, some short time after 18 January 1939.

Or is it? We still found an unaccounted for difference between Tulsa 5 and the first edition: Whereas Joyce/Léon makes it very clear on Tulsa 5/2 and 5/3 that on page 499.08 he wants to change 'Wo Hallall!' into 'Woh Hallall!', we get in the first editions 'Woe Hallall!' And the previous exclamation is also changed, from 'Who Hillill!' into 'Woh Hillill!' And maybe there are even more last minute changes. The more textual mysteries you unravel, the more remain.

- And a good thing too we went to Tulsa to check the proofs for ourselves.

- Encased in a heavy-duty winered cassette.

- We cleared up the last things we couldn't decipher on the copies.

- And mixed our fingerprints with those of James Joyce, thereby absorbing some molecules from his genepool.

- And we saw a beautiful exhibition of Joyce-papers, publications and paraphernalia mounted by Lori Curtis of the McFarlin Library.

- And came home with an extensive and richly illustrated must-have catalog by Luca Crispi and Stacey Herbert, 'In Good Company'.

- It turns out, thanks to this exhibition, that Tulsa, through the indefatigable collector Thomas Staley...

- The founder of the James Joyce Quarterly?

- In 1963, yes, the same, can boast of one of the largest Joyce-collections: he managed to acquire the libraries of Edmund Wilson (10.000 volumes), of Cyril Connolly (8.000 volumes), the libraries of Rebecca West, the papers of Harriet Shaw Weaver, the Paul and Lucie Léon papers and, in 1987, the papers and library of the great biografiend himself, Richard Ellmann.

- It's an Aladdin's Cave.

- Stuffed to the rim with Pandora's Boxes.

- The Joycean Science enters into a new stage: biographing the biograph.

- Finally we can find out where Ellmann got his wisdom from.

- And where he got it wrong.

- Whom he quotes and misquotes.

- And there was a wealth of original Joyce-publications to be looked at behind glass.

- Like the first ever magazine Joyce published in, Dana. An original Gas from a Burner broadsheet. Unique Pomes Penyeach editions, with original burn stains from the fire in the garage of Miss Weaver.

- Too much. Entirely too much.

- But for us the final Wake-proofs were the most important. There was more, however. Major new discoveries of unknown Wake-material in the lines of transmission were in store for us.

- The Tulsa treasures are really the X-files of Genetic Wakean Studies. No serious scholarly human hand has ever set foot in these documents since Paul Léon closed them once and for ever.

- For instance we saw a completely unknown typescript with handwritten marginal notes for Tales Told of Shem and Shaun (we think it was). (Or was it Storiella As She Is Syung?)

- We couldn't get to them behind glass, so we'll have to study those another time.

- Not behind glass fortunately were some marked, that is annotated, transition-issues from Paul Léon's library.

- Annotated! Wow!

- Lookey here: The Mime of Mick Nick and the Maggies in transition 22 (February 1933, published by Servire, Rietzangerlaan 15 The Hague, price fl 1,90), has colours in the margin, by Léon or even Joyce: thick orange, green and blue pencil lines. Sometimes a paragraph long, sometimes several paragraphs and occasionally breaking off before the end of a paragraph. As the game which is played in II.1 is called Colours, and Glugg has to guess the colour heliotrope, we think that these coloured lines have something to do with the game. Or with whose point of view is on - we'll have to do some more research on this.

- And there are four handwritten marginal corrections: they all made it into the final text, probably via another way.

- Exciting! Look what I've got here. Another transition, nr. 23, also published by Servire (July 1935, price fl 2,50). The subtitle here is 'An International Workshop for Vertigralist Transmutation.' What did nr. 22 have?

- 'An International Workshop for Orphic Creation.'

- It's like our Platforum! Different each time. Nr. 23 has the closing Nightletter on a separate page. And the second line of the first paragraph is already missing. It should have read: 'As we there are where are we are we there haltagain. By recourse, of course, recoursing from tomtittot to teetootomtotalitarian.' The middle words, comprising exactly one line, were overlooked either by the Storiella or the transition printer.

- We spotted that transmissional departure already and put it into our appendix to the translation.

- No handwritten corrections here though. But the next issue has corrections. Transition 27, the Tenth Anniversary number from 1938.

- The Butt and Taff-episode. Paul Léon corrects typing errors in his copy of transition. But Joyce doesn't reproduce them in his copy.

- Seven of the corrections Léon made were not spotted independently by Joyce. That means seven unintentional transmissional departures (from the typescript to transition) went into the final text.

- Shocking.

- Yes, but (except for two comma's) we already noticed the typo's and incorporated the right readings in our translation. For instance, 'I cantenue' on FW 347.22 was misset as one word, and 'fiend' was misset as 'fined' in 'the lelias of the fined' on FW 340.22. But we translated the old, correct version as 'de lelia's des viands'. Phew.

- But what are the consequences of all this for our 'do-it-yourself' corrected text?

- We'll explain right away.


Transmissional departures

In our very recent bilingual (the English text entirely reset) translation of Finnegans Wake (Athenaeum-Polak&VanGennep, April 2002, to be ordered at or for 75 euro plus postage/shipping), we appended a list of 1293 transmissional departures: unauthorized and unaccounted for changes in the transmission from one stage to another in the textual history of Finnegans Wake, as located in the James Joyce Archives (and sometimes Joyce's Letters). Not so many as Jack Dalton in 1963 estimated that a 'revised' edition of the Wake would contain (6.000), but still a respectable number of possible and probable errors and oversights.

We cannot call them mistakes, as Joyce gave his imprimatur and nihil obstat to the published book, so it has to be assumed that he saw every little change. Nonetheless, it is our contention that most of these transmissional departures are indeed oversights and mistakes.

With our edition the reader can make his own 'corrected text'. That is what we would do if we hadn't already done so: buy a new copy of Finnegans Wake and copy the 1293 items of the list into the main text. You will be amazed by the results: the Wake was never cleaner. Just to see what a word or a phrase was in a previous stage is in many cases very enlightening for the understanding of the passage in question.

Behind each and every transmissional departure there is a whole story to tell; and each and every item is ultimately a judgement call; but a discussion about what might be a mistake and what might not be can only begin if there is something to discuss. With this list we hope to make a beginning with a real 'variorum' edition of Finnegans Wake.

In the light if the new findings in the transmissional stages in Tulsa our list of transmissional departures has to be amended. Sometimes we found what we missed, sometimes we had to change the variant and sometimes we found a new transmissional departure.

These variants should be deleted from the list, as we found the spot in the Tulsa proofs where Joyce/Léon changed the words as they appear in the final text:

07.14: "bodey, behold" (jja 44:109). 07.20: "brontoichthyán" (jja 49:7, 50:9). 15.34: "dragonman..." (jja 44:120, 44:160, 49:15). 16.03: "marrowbones" (jja 44:243, 44:268). 16.26: "Has? Has af? Hasafency?" (jja 49:17, 49:16, Corr.). 25.04: "The Madison Men" (jja 50:277). 25.12: "pigscheeks" (jja 50:281). 44.21: "...koupkos!" (jja 45:71, 45:81). 60.10: "blame not the" (jja 49:375, 49:374). 104.05: "Augustissimost" (jja 49:140, 49:138). 105.12: "Him My Juckey and" (jja 49:433, 50:140). 105.31: "Forward Two" (jja 49:141, 49:138). 146.13: "ballshossers" (jja 49:481, 49:480). 187.06: "him urged" (jja 49:531, 49:530, 1st ed.). 315.07: "shifk" (jja 54:15, 54:105, 54:263). 317.05: "houming" (jja 54:201, 54:265). 303.f1: "eelyotripes, Mr Tellibly" (jja 53:224, 53:263, 53:331, 53:391). 321.05: "corollos" (jja 54:270, 54:271). 329.19: "homeymoon" (jja 54:211, 54:247, 51:281). 499.01: "fending" (jja 62:126-127, 62:376-377). 511.09: "red Fox-Goodman" (jja 58:226, 58:364, 59:25, 62:143). 550.27: "get" (jja 58:116, 58:447, 62:129).

These variants remain, but in a changed form:

06.32: "platterplate $E." (jja 44:11, 44:51, 44:108, 44:149, 44:207, 44:234, Tulsa 1). 09.02: "Cromwelly, Looted." (jja 44:258, 49:8, Tulsa 1). 58.25: "Coldstream Guards" (Tulsa 1). 61.19: "coaffianced" (jja 49:83, 49:82, Tulsa 1, cf. FW 61.21). 101.24: "vinom" (jja 46:110, 46:177, 46:225, Tulsa 1). 121.07: "a" (jja 46:426, 46:435, Tulsa 1). 136.05: "rain wets" (jja 49:469, 49:468, Tulsa 1). 190.32: "paragargle" (jja 49:535, 49:534, Tulsa 1). 241.34: "jompetter" (jja 51:423, 51:471, Tulsa 1). 295.33: "ocher. By Betcha Bob! Lucihere.!" (jja 53:5, 53:30). 317.29: "unium! ¶ - Place" (jja 54:19, 54:51, 54:228, 54:265, Tulsa 1). 344.27: "Graste" (jja 55:53, 55:157, 55:213, transition 27 Tulsa p.67, jja 55:242). 500.04: "rescue!" (jja 58:221, 58:288, 58:349).

These variants are brand new:

64.20: "nigh" (Tulsa 1, 2). 71.24: "Bat" (Tulsa 1, 2). 102.21: "wayfarer" (jja 46:111, 46:227, 49:137, Tulsa 1, 2). 219.06: "token in taken." (jja 51:449, Tulsa 1). 262.f1: "Jussive" (Tulsa 1, 2, 3, 1st ed., Corr.). 263.06: "incenstrobèd" (Tulsa 1, 2, 3). 276.L1: facing line 2 (jja 52:357, Tulsa 3, 1st ed.). 277.L: L2 facing line 6, L3 facing line 8, L4 facing line 11, L5 facing line 12 (jja 52:359, 1st ed., 3d ed. 1971). 281.L3: facing line 10 (Tulsa 3, 1ste ed.). 285.L1: facing line 7 (jja 52:369, Tulsa 1). 294.L2: facing line 18 (jja 52:215, 52:250). 299.F4: "$E, $A, $T, $X, $[], $/\, $[." (Tulsa 3, 1st ed.). 301.L3: facing line 25 (jja 52:22, 52:262, 52:391). 303.F3: "Browne-Browne" (jja 53:224, 53:264, Tulsa 1,2). 304.L1: facing line 29 (jja 52:397, Tulsa 1). 342.20: "), with" (jja 55:148, 55:210, transition 27 Tulsa p.64). 348.11: "fuselaiding and" (jja 55:170-171, 55:217, transition 27 Tulsa p.71).