Number of representation cases and number of workers involved in representation cases received, closed, and pending 1940-41, by affiliation of petitioner

p. 41. The crop is cut off at the bottom: the last printed row visible is "By recognition of representatives" (4 / 1,400 ...), which is clearly mid-table (the "Cases settled after hearing" breakdown continues, and the "Cases closed after hearing" / dismissed / withdrawn lines and the table's footnote block are not on this image). No footnote texts are printed on the crop, so `footnotes` is empty even though markers 1, 2 and 3 are used in the body. Footnote markers are printed as small raised digits to the LEFT of the figure; I have moved them behind a caret after the value as instructed. Where two markers sit together I wrote them after a single caret separated by a space, e.g. "5,081^1 2" on the "Cases on docket" row. Marker legibility: the raised digits on "Cases closed July 1940–June1941", "Cases closed before formal action", "Cases settled" and "By consent election" are small and partly broken in the scan; the second marker could be read as either 3 or 8. I transcribed them all as 3, since 1/2/3 is the self-consistent set used elsewhere in the row block, but a reader should verify against the (not shown) footnote list. Suspected misprint: "Cases closed after formal action", column "Cases filed by unaffiliated unions / Number of workers involved" is printed as 27.071 with a period rather than the digit-grouping comma used everywhere else; transcribed as printed (intended value presumably 27,071). Printing oddity in a label: the "Cases closed July 1940–June1941" row has no space between "June" and "1941" in the original; the parallel rows above print "July 1940–June 1941" with a space. Transcribed as printed. In the same row group, the "Number" cell for employers on "Cases closed after formal action" prints with a serif-heavy first glyph that reads like "I4"; it is 14. "Cases closed before hearing" is set with a wide word space ("Cases closed before hearing"); normalized to a single space. Cells shown as a dash rule (a row of hyphens) under the unaffiliated-union and employer panels are transcribed as "".

p. 42. Continuation page: no column headings are printed on this crop (the table's caption, headings and opening rows are on the preceding page), so header_rows is empty. The grid has 11 columns — the stub plus five apparent pairs of number-of-cases / number-of-employees columns. Footnote markers 2 and 3 are not attached to anything on this crop; they must occur on the preceding page of the table. "Cases dismissed" (first occurrence, 3rd body row): a small raised "2"-like mark sits inside the dash rule of column 7 (the 4th cases column). It is not a value — the printed row total of 4 equals 3 + 1 from columns 3 and 5 — so that cell is transcribed as "". It is most likely a stray/offset mark or a displaced marker; I did not record it as a footnote marker because nothing in that cell is printed to carry it. Glyph calls resolved by arithmetic in this low-quality 1-bit scan, where 9 frequently prints like 0: "Cases closed after stipulated election" (first of the certification group), column 1, read as 69 — 69 + 432 + 17 = 518, the printed certification total; 60 would not reconcile. Likewise 1,929 in "By consent election" column 6 (1,219 + 1,929 + 26 = 3,174), 12,657 in "Cases closed after ordered election" column 6, and 69 in "Cases closed otherwise after board decision" column 6 (13 + 69 = 82). All other rows and columns cross-foot exactly as printed. The one apparent discrepancy is benign and self-documented: in "Cases pending, July 1, 1941", the employee columns sum to 533,791 against a printed total of 533,861 (and the case columns sum to 1,382 against 1,383); footnote 1 accounts for the extra joint-petitioner case, counted in the totals but in no affiliation column.