The tubular loom is suited to long, narrow fabrics. Peruvian weavers, when they wanted wider cloth, used to sew several strips together. Sometimes, one student has guessed, they even set several waist-loom weavers side by side, having them interlock the edges of their webs as they worked. The pueblos, somehow, developed a wide loom. No one knows how this came about but scraps in the ruins suggest that it happened between 1100 and 1300 A.D. Perhaps those old looms, whose poles and string have long since rotted, were like the blanket loom of today. This is the famous loom of the Southwest, thought of by many people as the only one. It is still used by Hopi and Navaho and once by many other pueblos.

The photograph (111-12) shows such a loom outside a Hopi house with a weaver of the 1890's sitting before it. This weaver is a man, as all Hopi weavers are except the weavers of a few women's belts. However, men seem to have used the blanket loom in mcny pueblos, in contrast to women, who used the waist loom. Of course the rule is not a fixed one. The Navaho, who learned weaving from the pueblos, leave it all to women, blanket loom and waist loom alike. From this we imagine that, even for the blanket loom, they may have had women teachers.

However, there is a difference of tradition connected with the narrow loom and the wide. The waist loom, throughout Peru and Middle America, is woman's work and perhaps the Southwest learned about it in that form. On the other hand, the wide frame for twining rabbitskin blankets was often used by men, the skinworkers. Old Paiute and Papago men, in Nevada and Arizona, remember the custom yet. Perhaps pueblo men also did such weaving, and when the blanket loom was somehow developed they considered it suitable for men's work.

The diagram (111-13) shows how this loom was rigged, a and b are the already familiar bars which hold the warp, a instead of being fastened to a weaver's waist is attached by loops stakes in the floor h and h. b, the upper bar, is fastened in a more complicated way, since it is from b that the warp is tightened or loosened. A heavy beam, d, is hung from the ceiling and, in many houses, this is a permanent fixture. From it, by rope loops, there hangs a lighter bar, c. These ropes can be pulled tighter so as to produce more tension in the fabric. From c hangs the upper loom bar. Looking at the photograph, you can see the same arrangement.

Hopi blanket loom

Plate 111-12. Hopi blanket loom.

So far, we might be speaking of the tubular loom. However, though this does not show in the diagram, the warp j is not in tubular form. The finished fabric, therefore, will not be twice the length of the loom but, with a blanket, this is not necessary. In fact, the new situation has brought about a very clever arrangement. The warps are crossed while being strung, so that when the loom is set up, one shed is already in place. The weaver who first saw this possibility ranks as a real inventor, though we do not know when and where he worked. Second, the warps are not passed directly over bars a and b but over strings attached to these bars. These "loom strings" are g in the diagram. The warp is j. At each side of it are two heavily twisted strings, k, which will form the selvage of the blanket.

Warp Stringing for the Blanket Loom. For particulars as to warp arrangement, we turn to the more detailed diagrams, (111-14). For stringing this warp, bars are set in place just as for stringing the waist loom and, infact, the same bars may be used. The number of holes in the wall allows for their being set any distance apart. Warp is wound over these bars, not in the tubular form, over A over B but over A under B, forming a figure 8, as in Figs. 1 and 2 (111-14). Now the weaver inserts the loom strings, which will take the place of A and B when the warp is held upright in the loom. Fig. 3. Why strings instead of rods? Because the finished product is not to be a sash with fringe but a blanket with strong edge. The cords will form its upper and lower edges but meantime they have to bear the pull of the stretched warp so they must be strong. Each cord is made of specially spun string, or perhaps of two strings twisted. And then it is doubled. Its two parts twine through the warp, one over and one under, in the manner of the old fur blanket. If the warp is a fine one, with strings close together, this weft is twisted once between each two strings. If the warp is coarse it is twisted two or three times. Now the warp threads are held firm and evenly spaced.

Warp stringing for blanket loom

Plate 111-13. Warp stringing for blanket loom.

Next, the loom strings must be stretched taut, so the warp will hang properly. The weaver takes the light rods, A and B in the diagram, and loops them to other rods, H and I. The cord loops over the rod, through the twining of the loom string, over the rod again, as in Figs. 4 and 5 (111-14).

Next the heddles are put on. If this were a tubular warp the weaver would have to open the first shed with his fingers and run a stick through it. As it is, the figure eight arrangement has already opened a shed. He runs a long stick through this, then picks up the alternate threads and attaches a stick to them. This is the heddle. If this is an elaborate weave, he may put in more heddles. He ties the ends of all heddles together so that they will not slip out, then takes the warp off the frame and perhaps puts it away. He may set it up for weaving in some other house or, if he is a Hopi, in the kiva.

Setting the Warp

Old houses and kivas too had permanent fixtures for warp attachment. We have noted the heavy beam hung from the ceiling in diagram 111-13 and outside the house in photograph 111-12. For the floor end of the warp, there might be stakes as in diagram 111-13 h or a beam attached to the floor, as in 111-11 or 12. Sometimes the lower bar was held by leather loops put through grooves in a stone floor or a log sunk in an earthen floor. Old housebuilders provided for weaving fixtures as modern builders provided for wiring.

The bars in place, the warp is tightened. Now comes the last operation, the placing of selvage cords at the two sides (K in diagram 111-13). Zuni people run three cords along beside the warp at each side, tying them to bars A and B instead of to the loom string. As each weft came to the edge of the warp, it passed over one of these and under the other two, always in alternate succession. Jemez people use one cord made up of two or three smaller ones, tightly twisted, and each weft passes between the cords of the twist at a different place. The twist becomes loosened in the operation and often has to be untied and twisted again. There must have been other forms of selvage strings but they all insured a strong edge at the side of the fabric, as the loom strings Insured it at top and bottom.

Details of warp stringing

Plate 111-14. Details of warp stringing for blanket loom.

In any case, stringing and setting up the warp was a day's work. Hopi men preparing to weave the square white blanket for bride set the date days ahead so that a group could gather. They worked all day in the kiva and expected a feast of mutton stew from the bride's relatives at the day's end.