A solid grey tweed can hide almost anything. The same mill, the same loom and the same yarn woven into a plaid will expose every distortion in the weft, and the fault shows up not at inspection but on the cutting table, where the checks refuse to line up along the seam. The two distortions responsible have names: skew and bow.
This guide explains what each one is, how they are measured under the standard method the trade uses, where they come from in weaving and finishing, and the acceptance line that belongs in a fabric specification before the order is placed.
What Skew and Bow Actually Are
Bow is a weft that curves. Instead of running straight and perpendicular to the selvedge, the weft thread bends through the middle of the piece, so the fabric looks like it was pulled into a shallow U shape. Skew is a weft that runs straight but at an angle, so the pattern leans instead of sitting square.
The two faults behave differently downstream. A bowed weft pushes a stripe out of line in the middle of a panel; a skewed weft makes the entire panel lean, which is more visible in a short garment piece with a strong vertical stripe. Plaids combine both, since the check is built from weft and warp stripes crossing each other.
Both are measured as a deviation, in centimetres or inches, from where the weft should be. That single number is what a buyer and a mill can argue about productively, because it is repeatable and it is defined by a published method rather than by opinion.
How the Measurement Works
The trade method is set out in the standard test method for bow and skew in woven fabrics, ASTM D3882, and it is catalogued in the ASTM standards library that mills and inspection houses work from. In outline, the fabric is laid flat without tension, a reference line is drawn perpendicular to the selvedge, and the position of the weft is measured against that reference at defined distances across the width.
The largest deviation found, expressed against the measurement length, is the result. Two details decide whether two measurements are comparable: the length over which the deviation is measured, and whether the fabric was relaxed before measuring. A bow measured on a roll straight off the loom and one measured on a relaxed bale are different numbers for the same cloth.
This is why the standard defines the method but does not set a universal pass mark. Acceptance values are commercial agreements, and they belong in the purchase order next to the measurement length, so both sides are measuring the same thing the same way.
Where the Distortion Comes From
The root of most bow and skew is tension. If the warp threads on one side of the loom run tighter than the other, or if the take-up mechanism is uneven, the weft is inserted into a warp sheet that is not uniformly stretched, and it settles into a curve. Loom width matters too: the wider the loom, the more a small tension difference shows across the piece.
Finishing can add distortion as well as remove it. Stentering, the step where fabric is held by its edges and dried under width control, can pull skew into a piece if the clip chains are not aligned. Conversely, a well-run stenter is where a mill corrects a small bow by overfeeding or skewing the fabric deliberately.
Relaxation finishes the story. Wool moves after wet finishing and again after steam, so a piece that measured within tolerance at the loom can shift in the bale. Buyers who cut immediately on arrival, without relaxing the fabric, inherit whatever movement happened in transit. The measurement that matters is the one taken after relaxation, in the state the cutting room will actually use.
Why Some Fabrics Expose It and Others Hide It
Pattern scale decides visibility. A fine herringbone hides a deviation that a large windowpane check displays immediately. Yarn character matters as well: a hairy, melange wool softens the edges of a pattern and hides small deviations, while a clean, tightly spun yarn draws a crisp line that shows every degree of lean.
Nap adds a third variable. A raised or brushed surface scatters light, so a leaning weft is less obvious to the eye even though it is physically there, which is a reminder that visual inspection alone is a poor substitute for the measurement. A fabric can pass a look test and still fail a pattern-matching cut.
The practical result is a two-tier acceptance approach. Buyers cutting solids can accept a slightly wider band, because the fabric has no reference line for the eye. Buyers cutting checks, stripes or anything that must match at a seam should specify the tighter band and measure before cutting, not after the panels are already out. The same discipline that governs width and selvedge planning applies here: measure the property that your cutting plan depends on, and measure it before you commit the cloth.
Writing It Into the Order
A complete clause runs to five lines: the test method, the measurement length, the state of the fabric at measurement, the acceptance value, and what happens if a bale fails. The last line is the one that saves the most money, because a failed bale is a decision point, and a decision made in advance is a negotiation that has already been won.
The workable options for a failed bale are rejection with replacement, acceptance with a price allowance for the affected metres, or acceptance with a re-cut plan that uses the fabric in panels where pattern matching is not required. Mills accept all three; what they resist is a claim raised after the fabric has been cut, when no one can prove what the roll measured on arrival.
Finally, keep the records. A photograph of the fabric on the table with the reference line drawn, a note of the measurement length, and the date, all stored with the bale number. This is the same evidence discipline used for selvedge and marking checks, and it turns a quality argument into a documented comparison. Where skew interacts with the way the cloth hangs in the finished garment, the selection notes in how weight turns into drape are worth reading alongside this one, because a fabric can pass both measurements and still be the wrong choice for the silhouette. The programme side of the same question, matching the cloth to a line and a price, runs through our wool fabric manufacturing service.
Ofte stilte spørsmål
What is the difference between bow and skew?
Bow is a weft that curves through the middle of the piece; skew is a weft that runs straight but at an angle to the selvedge.
Which standard covers the measurement?
ASTM D3882 defines the measurement of bow and skew in woven fabrics. The acceptance value is a separate commercial agreement.
Why did my plaid fail if the mill passed it?
The fabric may have shifted during relaxation or transit. Measurement should be taken in the relaxed state the cutting room will use.
Can skew be corrected after weaving?
A small amount can be corrected during stentering. Larger distortions are a loom tension problem and should be fixed at the source.
What should the order specify?
The test method, measurement length, fabric state, acceptance value and the remedy if a bale fails.
Video: Hidden Craftsmanship in Tweed Fabric: What You Need to Know
If your next programme uses a check or a stripe, agree the skew and bow clause before the loom is set. The team behind our wool fabric manufacturing service measures every finished bale and can quote against your acceptance band from the first sample onwards.


