Distribution of Unfair Labor Practice Situations Received, by Number of Employees in Establishment, Fiscal Year 1967

p. 265. Crop is cut off at the bottom: the body ends after the "110-119" row with the column rules still running, so the remaining size classes (120 and over, not reported, etc.) continue below/on the next page and are not in this image. No footnote texts are printed in the crop, although two markers appear: a raised 1 on the caption (caption excluded per instructions) and a raised 2 before 14,425 in the Total row, transcribed as 14,425^2. Decimal points: this is a 1-bit scan in which many periods in the percent columns failed to reproduce and print as a gap (e.g. the Total row CC and Other C cells read "100 0", and cells such as "26 0", "8 1", "6 3", "63 1", "0 4"). All such cells are one-decimal percentages and have been transcribed with the decimal point restored; no digits were changed. Doubtful cells: CB "Percent by size class" in row "10-19" is read as 8.1, but 167/2,076 = 8.04, so the printed glyph may be 8.0 — this is the one cell where my reading disagrees with the implied ratio. Other low-confidence (faint-point) readings I cross-checked against the ratios and kept as printed: CA percent 8.6 for "20-29" (ratio 8.69), CP percent 5.8 for "40-49" (5.87), CP percent 1.1 for "80-89" (1.17), CP percent 1.7 for "90-99" (1.76), Other C percent 6.7 for "100-109" (6.78) — the table appears to truncate rather than round in several places, so these are plausible as printed. Blank cells: the CE "Percent by size class" column is blank or a dash rule in every row whose CE number is 0 (10-19 shows an empty cell, the later zero rows show a dash rule); the Other C percent cell for "110-119" and the Total row's cumulative-percent cell are dash rules. All are transcribed as "".

p. 266. The crop is the bottom continuation of a table carried over from the preceding page: no column headings are printed on this image, so header_rows is empty, and no caption, total row or upper body rows appear here. The grid has 20 columns: the stub (size class), a count, a percent, a cumulative percent, and then eight count/percent pairs. That column count is confirmed arithmetically: in every row the eight pair counts sum exactly to the row's total count, and the cumulative-percent column rises by exactly the percent column's value from row to row. Column boundaries: the dash rules ("------") that stand in for a zero percent occupy the percent column of the pair whose count is 0, and every such pair has count 0, in all 22 rows. Scan quality: the type is worn and many decimal points did not print or did not survive the 1-bit scan, so cells appear as "70 0", "0 3", "2 1", "8 3", "1 6" etc. These are transcribed with the decimal point ("70.0", "0.3", "2.1", "8.3", "1.6"); each such reading is consistent with the arithmetic checks above (e.g. 2/24 = 8.3 percent in row 2,000-2,999). Doubtful cells: row "130-139", last pair (columns 19 and 20) - a stray mark resembling a "1" sits just left of the final count, which itself reads as "0" followed by a dash rule. I transcribed the pair as 0 and "" because that is what the printed row total requires (85+7+4+2+0+2+5 = 105) and because a dash percent accompanies a count of 0 everywhere else in the table; if the mark is instead a real "1", the components would sum to 106 against a printed total of 105. Row "400-499", column 9 (the 4th pair's count): a small stray mark precedes the digit, giving the appearance of "~9"; the value is 9 (213+47+9+11+0+0+11+6 = 297). Row "180-189": the percent cells of the 3rd through 6th pairs are faint and may be blank rather than dash rules; either way they are transcribed as "". Row label "3,000-3,999" is printed with a very faint hyphen. In footnote 2 the opening words of the last line are smudged in the scan; "are based on" is my reading of the partly visible characters. Cells holding only a dash rule or leader dots are "".