Woven Structures Update

Marla Mallett

I suppose it was inevitable that as soon as my book, WOVEN STRUCTURES was sent off to the printer, intriguing new technical features would pop up everywhere. A mysterious rug and yastik appeared... Then some extraordinarily complex selvages... Then Kyrgyz and Yomut weavings with strange features... And more!  My second edition has incorporated a few of these, but new features continue to appear.  

To share new findings, I have posted occasional research notes and photos on these pages.  Anyone looking for a quick introduction to the basic weaves used in Middle Eastern textiles should go to our page on The Basic Tribal  Weaves.


Part 1 - Below

  • A Mystery Rug: Needle-Knotting
  • An Odd South Persian Saddlebag Detail
  • A Distinctive Talish Selvage?
  • A Curious Yastik: Tufting
  • An Unusual Bag Face...and a Few Loom Questions
  • Some Yomut Knotted Aberrations
  • An Unusual Beshir Weave
  • Pet Peeves: Rug Book Weirdities

Part 2  

  • A Kyrghz Rug?...or Band?...or Bands?
  • The Most Complex Selvage Yet: A Kirman Example
  • Chinese Art Deco Rug Problems
  • Anne Rowe Review of Woven Structures
  • The Emery Question

Part 3

  • A Sticky Terminology Question: Yarn Singles
  • Some Odd Knots: Stacked, Halved and Overlapped
  • What's Up? In Kilims, That Is...
  • A Specialized Kurdish Selvage
  • A Baluch Bag and Its Lessons
  • A Distinctive Senneh Weave
  • Two Konya Selvages

Part 4:  End Finishes Project

  • Rugs and bags with distinctive warp-end finishes and
    decorative borders. Thirteen pages with over seventy 
    examples now illustrate a wide variety of structures. 

Part 5:  A Simple Frame Loom

  • How to set up a picture-frame loom for experimenting
    with the basic tribal weaves.

Part 6

  • Kurdish woven covers from the Djezire area of southeastern 
    Anatolia. These weavings display an unusual single-interlocked 
    twill-tapestry structure.  

Part 7

  • Offset knotting as used by Anatolian, Persian and Turkmen
    rug weavers. Classic early examples are also shown.

Part 8

  • Turkmen tent band esoterica.  A beginning investigation of the 
    offset knotting structures in mixed flatweave/knotted-pile bands.


A Mystery Rug...Still

Steve Hofmann brought me an unusual rug that had puzzled almost everyone.  Roughly square--about 5'x 4½'--it has soft wool, pale colors (probably synthetic), and a simplified design with a Lesghi-type star in the center.  With a close look, I found that that the rug had been needle-knotted.  On a pre-woven fabric, off the loom, the pile yarns were inserted with a heavy needle, much as carpet restorers do. A right-handed person would simply make two right-to-left stitches, then leave a loop (to be cut later) before the next pair of stitches. 

With this kind of stitchery, few clues to the technique may be immediately apparent.  Under magnification, we may find places where the needle pierced the warp or weft yarns.  Selvages rarely duplicate those on any woven carpet, as extra interlacing or substantial reinforcement at the edges is unlikely.  The handle of any needle-knotted piece is soft and flexible.  Steve's rug displays all of these characteristics, but also some additional peculiarities.
Normally, we should expect needle knotting to duplicate typical symmetrical (Ghiordes) knots, with each knot neatly encompassing two warps or two warp pairs. Throughout most of Steve's rug, however, pile yarns encircle both warps and wefts.  It is impossible to produce such a structure on a loom.  The stitching on Steve's rug is irregular throughout, with several different knotted configurations. I have shown three variations in my drawing. On the rug's back, irregularities are clearly visible, as you can see in the photo.

Needle knotting has a long history. It was a popular domestic craft in mid-17th century England, where the products appeared most often as upholstery on so-called "farthingale" chairs.  With short clipped pile, the fabrics were labeled "Turkey work," since they resembled fashionable Anatolian carpet weaves.  The upholstery designs, however, were very English in style, with sprawling floral motifs. These designs were much more naturally produced with needlework than with sequential horizontal rows of knots on a loom.  Designs could be drawn directly on the ground cloth, just as for needlepoint or crewel embroidery

I remember that in the 1960's needle-knotting was a popular American craft pastime.  Embroiderers made small rugs with long pile that imitated lush Scandinavian rya rugs.  When my son David was six or seven years old, and hanging around my weaving studio, I showed him how to do needle knotting on loose cotton canvas.  With a big blunt tapestry needle and heavy yarns, the child made a tiny pillow cover--using every brilliant color from my yarn shelves.  Although the process is slow, it is simple.

Needle knotted pieces should not be confused with rugs that are tufted with hooks or punches, nor with those having simple inlaid looped pile. A couple of the discussions below will touch on these structures.

So what is the origin of Steve's rug?  Speculations thus far have ranged from American, to English, to Balkan.  What do you think?  Please let me know if you have encountered anything similar, and can help to solve the mystery.

An Odd South Persian Saddlebag Detail

Daniel Deschuyteneer has sent photos from Belgium of a curious detail he found on a Basseri Kamseh saddlebag. He had figured out the construction and even made a small mock-up of it. The question was simply what to call it.

This detail does not represent a new and unique weave, but merely a reversed and expanded version of the rather standard two-color wrapped and bound selvage construction that appears on the Afshar weft-substitution rug I've posted below. I have also included diagrams below. The ordinary ground weave continues underneath the heavier reinforcing yarns and is not shown in the drawings. Any number of warp yarns may be enclosed by the wrapping.

These common decorative selvages sometimes edge the slits on the closure strips of South Persian saddlebags. They provide sturdy reinforcement that is easy to do. You can see examples on pages 181 and 197 of Jim Opie's Tribal Rugs of Southern Persia. But the scan which Daniel sent shows multiple columns of this reinforcement at one end of his saddlebag's bridge between the two pouches. This part of such a bag is extremely vulnerable, and I have repaired tears in such spots on more saddlebags than I care to remember! So the special attention given to this bag was for good reason. This weaver also extended the heavy groups of twined black and white yarns that made her closure loops, braiding them to form "handles" on each side of the bag-- very convenient for carrying the object.

When this knotted-pile saddlebag was on the loom, the pouches' back sides and the back of the bridge faced the weaver. She produced her wrapped and bound selvage in the short bridge sections exactly as on a rug, but expanded the idea, producing four decorative adjacent columns with the same construction. The ground weave is continuous under them. Once the bag was off the loom and assembled, the reverse side of the bridge became the front.

Front of the saddlebag bridge

Back of the saddlebag bridge
(Side facing the weaver)

Wrapped and bound selvage in its simplest form

Two-color wrapped and bound selvage as on the saddlebag above and the Afshar rug below
So what should we call Daniel's saddlebag detail?  I would probably just write, "Two-color wrapped and bound selvage reinforcement, in four adjacent columns." Or something like that. One could also indicate that the binding occurred on the back side, if desired.

Daniel also came across another unusual use of the wrapped and bound construction. His photos below show the detail used both for multiple borders and as a decorative feature within the field of a small soumak weaving said to be Baluch. Astonishing! The actual selvage is simply overcast. Each of the columns is edged with vertical wrapping; to continue the distinctive two-color wrapped and bound design feature horizontally, the weaver used twining.  I am wondering if anyone has found similarly ingenious uses of the wrapped and bound structure elsewhere?  It is delightful to see such creative structural designing in a fairly recent piece.

A Curious Yastik

Charlie Whitfield recently showed me an interesting old "yastik." The handsome piece has lovely colors, lustrous wools,  full pile, a conventional Turkish design, and evidence of substantial age. But on examination, it proved to be not knotted, not woven, and not even Turkish!

Instead, this curious and remarkable little piece is tufted.  Loops of yarn were pushed through a pre-woven cotton twill fabric with a shuttle hook or punch, as in the drawing below.  It was made in the same way as the tufted or hooked rugs that are indigenous to North America. Finally, the loops were cut and the surface sheared. This small piece was worked on a much finer scale, however, than we most often see in domestic tufted rugs.

On a piece like this, when we separate the pile on the front, there are no knot "collars" to be seen. Distinctive signs of the technique are visible on the textile's back. As you can see in the photo above, the rows of tufting are oriented in all directions. The work did not proceed just horizontally, but diagonally and vertically also. Often the tufting followed the shapes and contours of the design.  The motifs in this piece are free-flowing in character, unconstrained by a woven grid. 
The earliest pieces were made with simple hooks. With this rather tedious method, the artisan works from the front, and pulls yarns upward through a loosely-woven ground fabric. She may just hold this fabric on her lap, but it's more satisfactory to stretch it on a frame. It is fastest of all to work from the back of the fabric, pushing loops of yarn or cut cloth strips through to the front. This is also the most practical method with fine materials. The photo below shows four kinds of commonly used implements: a hook (with red handle), a punch, and two shuttle hooks. Technically, rugs made with shuttle hooks or punches from the back should be called tufted, although the term "hooked" is often applied.

The domestic rug-hooking craze hit its peak in the eastern US and Canada in the 1920s and '30s. In some places the craft was developed as a cottage industry. But Oriental rug patterns were first popularized among rug hookers 40 or 50 years earlier--at a time when Oriental rug merchants were furnishing fashionable Victorian parlors. Housewives who could not afford exotic imported rugs found that they could make their own substitutes.

In the 1860s and 70s, Edward Sands Frost, a New England tin peddler, designed and sold hooked rug patterns with a wide variety of subjects, including freely interpreted Near Eastern motifs. From metal stencils, he produced colored patterns on burlap. Some were imitated by Ebenezer Ross of Toledo, Ohio, and others.  The illustrations at the right are from the E.S. Frost Company's 1884 catalog. By this time Frost had sold his business and it was operated by James Strout in Biddeford, Maine. The circular promoted not only the company's "Turkish Rug Patterns," but also an innovative shuttle hook, patented in 1981 and billed as an "Embroidering and Ornamenting Machine." In 1886, Ross developed a shuttle hook nearly identical to the wooden one in the photo above. Apparently designs with a Turkish flavor were first devised for small tufted pieces, but they were eventually produced by Frost in 36" by 72" sizes.  By the late 1880s even the Montgomery Ward catalog was offering burlap rug patterns and listed "Turkish Designs" among the subjects.*

Rug hooking and tufting enjoyed a revival in the 1960's, and during that period, even I designed and produced quite a number of tufted pieces. It's a pleasurable process, and the technique gives an artisan considerable freedom. Electric tufting guns had been refined just a bit earlier, and led to a flourishing special-order carpet business in North Georgia.  Fiber artists elsewhere also bought and used these guns, and their products should not in any sense be considered machine-made;" the electric tool merely speeded the work. The adaptation of tufting for mass production was the next step. Rows of tufting needles were mounted and the tufting process became the primary method by which wall-to-wall carpeting was -- and still is -- produced.

Does anyone have more information on the earliest North American domestic craft products that were Oriental rug design spin-offs?  Or of the British examples that began appearing in the 1890s?  We have been aware for some time of the influences of Near Eastern designs on Navajo slit-tapestry weavings, but tufted folk art products have been largely ignored in Oriental rug literature.
I've been asked how we can tell the difference between a tufted piece and a weaving made with inlaid looped pile. The drawing at the right shows how inlaid pile is made by pulling up loops from a yarn inserted within an ordinary shed. Color usage with inlaid pile is limited, single loops (or single cut loops) are impractical, and all patterning proceeds neatly, row by row in a rigid, orderly fashion -- unlike the tufted example shown above.  Inlaid pile yarns are crossed by warp yarns on the fabric's back. Sometimes multiple colors are inlaid together, so the pile colors can be selected and pulled up wherever the artisan pleases.  PLEASE NOTE: This drawing is shown without the ground wefts that are normally placed within the shed along with the pile wefts.

For more information see J. and K. Kopp, American Hooked and Sewn Rugs, New York, 1975, and Sarah Sherrill, Carpets and Rugs of Europe and America, New York, 1996.

A Distinctive Talish Selvage ?

While working on the book, I was asked about the significance of Talish selvages with their blue incursions into the field. How should they be catalogued? At that time I could not find any Talish rugs to examine, but I have seen a few since, and the answer is simple. On the rugs I have examined, interlaced selvages with two groups of four warps are reinforced with blue single-ply wool. Periodically, the blue selvage yarn extends erratically into the field -- 4 to 16 warps into the knotted area -- but it interlaces these warps in pairs, rather than singly, as do the ground wefts underneath them. You can only see this feature clearly on a rug; it is difficult in my photo.

It is not unusual for a weaver to extend selvage reinforcing to make a firm edge; the amount of reinforcement often varies within a single rug, as the craftsman makes adjustments to account for the yarn's bulk and the way it compacts. By interlacing pairs of warps in the field, however, instead of using the loom's plain-weave sheds, the Talish weaver could use a slightly heavier selvage yarn. This construction may help us to identify this group of South Caucasian rugs, and include pieces without the standard Talish design layout. The
construction, not the color, of the selvage is the critical factor. 

We need to look further for similar structures elsewhere. Daniel Deschuyteneer has reported finding similar selvages on some northwest Persian rugs, (pieces usually labeled Sarab and Karadja) and it should be interesting to discover how widespread and consistent this approach to construction has been.

Daniel posted a photo of a rug on the Turkotek discussion board a short while back that most viewers immediately assumed to be Talish. It is shown below. But the assumption proved to be erroneous.
Not only is this rug's weave a coarse 36 knots per square inch, quite unlike typical Talish constructions, but the selvage structure is different. This rug has an attached interlaced selvage, in which a free-floating outer warp unit is attached to the rug only by separate red wool selvage yarns. This is a distinctive weaving practice, with a construction that differs significantly from a simple reinforced selvage. A different attribution is surely required. In Daniel's opinion, a Gendje label may be most appropriate for this Talish look-alike.

An Unusual Bag Face...and Some Loom Questions

Philip Lichtman, rug restorer and author of a series of excellent articles on restoration techniques in RUG NEWS (also on line at http://www.rugnews.com), sent photos of a piece with unusual construction.  He describes the weaving as a "fairly recent Persian bag face."  He says the wools are very glossy and stiff, with bristly pile, and yet the piece is supple.

The curious feature of this little bag face is its expanded plain-weave ground cloth, sometimes called a half-basket weave.  Each weft interlaces pairs of warps (over two, under two) instead of interlacing them individually as in normal rug constructions.  The wefts are two-ply wool yarns, and there are two shots between rows of symmetrical knotting.  Although knots and weft spans could have been offset, they were not.  Unfortunately, the ply in the weft yarns makes it difficult to see the structure clearly in the photos. You can see though, that the knots are not perfectly lined up--they jiggle up and down slightly, in accord with the weft alignment.

So what could be the reason for using a weave that is inherently less secure?  It has been suggested that the weaver may have wanted a soft, flexible textile, but this seems an unlikely consideration for a bag.  It is more logical to suppose that the piece was made NOT on a rug loom, but on a loom used predominantly for other kinds of weaving.  It was probably a four-harness counter-balanced treadle loom. The weaver very likely found that the yarns she had available for wefts were just a little too heavy for ordinary single interlacing, and so she expanded the structure to accommodate them.

We tend to think that nomad and village looms are of just two major types -- either very sturdy fixed-heddle upright rug looms, or horizontal fixed-heddle versions staked to the ground. We forget that looms with harnesses and treadles existed in the past among settled people nearly everywhere. In my jaunts around Anatolia, for example, I've come across some truly strange home-made contraptions used for everything from blankets and curtains to occasional brocaded or tapestry-woven rugs; not very long ago country people used such looms for clothing fabrics.  I've also seen very simple slit-tapestry kilims made on these looms (with widely spaced warps) and woven row-by-row, instead of in the more typical separate design sections.  On most conventional treadle looms, a swinging beater the width of the loom holds a metal reed that separates the warps and packs the wefts, greatly speeding up the production of simple weavings.

So here are the applicable points: It is impossible to pack wefts as tightly with a wide mechanical beater as when they are pounded down one narrow area at a time with a heavy hand beater on a rug loom.  Thus the expanded weave on the bag in Philip's photo allowed the most dense and sturdy construction possible under the likely circumstances. If warps were threaded consecutively through four different harnesses, the weaver could easily switch from raising alternate warps, to raising the warps in pairs.  I have frequently made such accommodations in my own weaving, although not with knotted pile.
Has anyone else come across a knotted bag face or rug woven on a paired-warp ground? 
Similar loom technology and procedures almost certainly account for the construction of simple tapestry shäddä, or covers, from the southern Caucasus. These are either a balanced plain weave or a weft-predominant weave; they are not weft faced like conventional tapestry. 

The detail at the right is from a handsome piece in the collection of Mike Tschebull. For such a piece, the loom was warped with wide solid stripes of varying colors that matched a majority of the intended wefts; thus the weaver avoided the speckled look of exposed light warp yarns. In our photo detail, warps on the left are blue, warps on the right are red. In transitional areas the warp and weft colors mix, and the blue areas are speckled with red.  On a colored warp, with a full-width reed and beater, one could easily produce this kind of weft-predominant cloth, although such a fabric would be most difficult to produce with any consistency on a rug loom. Only an inexperienced weaver would attempt to do so. Mike is convinced that these shäddä were village products, perhaps produced for trade with affluent nomads; I think that is a reasonable

Likewise, the double-interlocked twill-tapestry Persian covers sometimes called moj are not weft-faced, and their twill constructions required four loom harnesses or heddle bars. The photo shows the reverse of a piece belonging to Allan Arthur. With this particular tapestry technique the weaver had no choice: the structure required row-by-row execution, and so for a treadle loom with a reed and beater, this kind of cover was a natural product.

After thinking through these matters, I've begun to wonder if the unusual Anatolian Malatya ala çuval that feature slit-tapestry (on a minuscule scale) on offset paired warps weren't also woven on four-harness looms. It's logical. One can take advantage of a loom's treadles, multiple harnesses and reed, and still weave separate sections without using the wide beater to pack the wefts. This is the way the miniature cotton Egyptian tapestry shown on the loom on our Egyptian Tapestry page is being woven. This specific tapestry structure is discussed in WOVEN STRUCTURES on page 78. For a larger photo click here.  I have not seen recent examples of this rare weave, and so do not know if it is still being produced in the Malatya area, although reciprocal brocading with a superficially similar look is found frequently.  

I thought you might enjoy seeing one of the more peculiar upright treadle looms I've come across in Turkey. I found the roughly framed construction in the photos below in a small winter settlement of Tarus Mountain nomads. This loom had only two harnesses, and the warp had been stretched right out the door of the weaving shed to wrap around a post in the muddy animal yard outside. On this loom you can see harnesses hanging from pulleys, and a wide mechanical beater that swings from above. Two treadles dangle below. A very soft, blanket-like kilim was in progress. Ever wonder why so many villagers prefer bright colors?  Just try weaving (or living) in this kind of dark, dreary, windowless dungeon! 

Some Yomut Knotted Aberrations

As my friend Allan Arthur and I have looked for knotting irregularities in Turkmen rugs, we have found plenty of examples in which knots were offset to better articulate designs. On a large percentage of the asymmetrically knotted Turkmen weavings that we have examined, symmetrical knots have been used to fortify the edges. On old Saryk and Yomut pieces, we have found short rows of symmetrical knots frequently overlapped (as in the drawing) to add bulk and thus straighten crooked weaving. These practices are all discussed and illustrated in WOVEN STRUCTURES.

But we discovered another curious feature on a Yomut ensi -- a rug now in the collection of Sophia Gates. The weaver of this rug overlapped from 6 to 9 knots at both ends of each knotted row next to the selvages (also as in the drawing).  She did this along the entire length of each side.  Does this detail indicate that Sophia's rug was almost certainly made for a yurt door hanging? Because the edges were expected to be repeatedly grasped and pulled at and so were strengthened with this extra bulk?  Might such a feature help to distinguish a truly functional, ethnographic ensi from later, purely commercial products?  Seems logical to me. 
Two more ensis have now appeared with this same detail: rows of several overlapping symmetrical knots along both selvages. One is again a Yomut piece (shown below) in Sophia Gates' collection, while another is a very early Saryk ensi.  Thus this proves to not be a distinctively Yomud practice. Has anyone else checked their ensis for this feature? 

On occasional small Yomut knotted-pile weavings, we see pairs of symmetrical knots overlapped along the side edges -- sometimes regularly and sometimes erratically. As on the ensi, the extra bulk from this overlapping strengthened the edges a little.

It is easy to confuse pairs of overlapping knots with Yomut edge knots, because in both cases the knots are large on the rug's back side and an uneven number of nodules appears in the border area. (See page 37 in WOVEN STRUCTURES.) With the distinctive edge knots, however, the large nodules form the outer column. In contrast, when a pair of ordinary symmetrical knots is overlapped, large nodules form the second column on the back. This is shown in the asmalyk photo at the right, although it is hard to see the outermost half knots next to the very heavy modern overcasting. 

An Unusual Beshir Weave

Nick Wright and Seref Ozen have sent scans from Istanbul of a most interesting Beshir rug. They asked that I please provide some "enlightenment"! 
They described the piece as having a mix of cotton, wool and cotton, and wool wefts, with two weft shots between rows of knotting. They went on to say, "The all-cotton weft appears in white triangulated sections, sometimes separated by wool wefts that span the width of the rug. In other places the cotton and wool together, in the same shot, also span the width of the piece. The triangulated all-cotton sections are joined to the wool wefts around the same warps, or crudely tied in a few places." The question was whether any of this might have been done for structural reasons or if it represented pure fancy.  

It is risky to diagnose structural details or problems from photos, but it appears to me that this rug was simply made by a couple of frugal weavers intent on using up odds and ends of left-over yarns, and doing so quite randomly. Since they were working from the front of the rug, they were not so aware of its peculiar reverse-side appearance. The apparent use of varying left-over materials appears also from time to time in other work, particularly in South Persian and Kurdish rugs.
The strongly contrasting white cotton in this piece accentuates the fact that in some places the weavers have used discontinuous wefts and have offset their ending points diagonally. Such diagonally aligned breaks -- usually referred to as "lazy lines"-- most often occur when there is a significant disparity between the weaving skills of two or more people working together. They are most likely when the speed of one person changes repeatedly as she confronts unfamiliar patterning. But they also occur when a solitary weaver works first in one area, then moves sideways to weave another section. That wefts in some parts of this rug are continuous merely tells us that the situation changed during the rug's manufacture. A couple of people (or maybe more) working together were able to synchronize their work part of the time, but were unable to do so efficiently when a less skilled person joined in. If the wefts were all of a single color, the diagonal lazy lines might still be present, but simply less noticeable.
It is difficult to see clearly in the photo, but it seems that the white discontinuous wefts have been carried from one weft row upward to the next over a row of knots along the slanting edges of the wedge-shaped sections. This is another sign of an inexperienced weaver, as the upward movement normally occurs at the selvage, making these diagonal floats unnecessary.  A most interesting Central Asian weaving!

Pet Peeves:  Rug Book Weirdities

Rug literature has often been the butt of jokes in the wider textile community because of its technical inaccuracies and quaint language. I would like to spout off on a few of my pet peeves. 

PET PEEVE NUMBER 1.  "Wefting." Believe me: No such word exists in the wider textile world -- among weavers, textile engineers or textile scholars!  "Weft," "wefts," or "weft yarns" are the appropriate words.

PET PEEVE NUMBER 2.   "Shoot." As in "weft shoot." The proper word is "shot." Even this best applies to a weaving process in which a bobbin enclosed in a sleek boat shuttle is thrown or shot through a wide open shed -- something that rarely happens on a rug loom. Most properly, "shot" applies to the process, while "pick" is the textile term that describes an interlaced weft in a finished fabric. I'm under no illusions about instigating Radical Changes in rug literature, but we can at least avoid "shoot." That inevitably conjures up visions of
bean sprouts, not yarns, for some of us.

PET PEEVE NUMBER 3.   "Cord" is among the most widely misused terms in rug books. With current usage, if a rug's wide four-rib selvage is called a "four-cord selvage," we do not know if the structure incorporates actual cords or something else. A cord, after all, is two or more plied yarns that have been re-plied; the term cannot describe a group of two or three ordinary warp yarns. The confusion is compounded when actual cords are called "cables." A cable, in fact, is two or more cords that have been re-plied. When we routinely misuse these terms we forfeit opportunities to convey significant diagnostic information. 

PET PEEVE NUMBER 4.  "Weft-float brocading."  Lots of people, when confronted with unfathomable flatweave structures, seem compelled to slap on this label. As a favorite catch-all, it has been applied erroneously to at least half a dozen different weaves. It is surely preferable to see mysterious structures ignored, rather than identified erroneously. The only worse offense is calling flat woven details "embroidery."

PET PEEVE NUMBER 5.  "Wrapped." Another catch-all that has  often been used to describe selvages improperly. Even a cursory look at chapter 15 in WOVEN STRUCTURES should disclose the simple elemental differences between selvages that are interlaced (woven: over, under, over, under) and those that are wrapped (overcast: around and around and around).  Without understanding this elemental difference, it is impossible to identify more refined and distinctive selvages.

I could go on ad nauseaum. By merely correcting the simple abuses of language noted above, however, any ruggie who likes to talk about his favorite pieces is guaranteed to sound more textile savvy.

Go to WOVEN STRUCTURES UPDATE - Part 2, 3, 4, 5, 6, 7, 8

1690 Johnson Road NE
Atlanta, GA  30306   USA

Phone:  404-872-3356 
E-mail: marlam@mindspring.com
© Marla Mallett

  Update - 2  Update - 3  End Finishes  Frame Loom  Update - 6  Update - 7  Update - 8