Last fall my curiosity turned toward the variety of visual things that can happen when a vertical line and a horizontal line meet and intersect, and while I was thinking of squares, I bought some very beautiful yarn made by Flying Goat Farm, a local Maryland goat farmer and dyer. The yarn was their Zephyrette, a blend of mostly alpaca with some silk and cashmere mixed in to create a fabric that no one can resist petting. I bought six skeins each in a different soft, natural color, to create a rough pink-orange three-color gradient and another in a green three-color gradient. The yarn immediately told me that it wanted to be used in gradients stranded in a floral pattern, and while I stood there looking at the yarn, the design came to mind. When I think of a design, my thought process isn’t verbal, it’s a series of images: squares; checkerboards; positive-negative color alternation; natural plant-like colors; gradients; stranded knitting; flower shape in gradients; positive-negative alternation of the flower shape in a checkerboard pattern; simple checkerboards at the shoulder-shaping.
Zephyrette is a luxury yarn, and a sweater-quantity is a chunk of change, so the price tag influenced my design. I knew I wanted a boxy shape with at least 10 inches of positive ease at the bust. It didn’t have to be terribly long, 16 to 18 inches from armpit to hem would suffice, and an opening of about 7 inches per side at the dropped shoulder. I thought I would dispense with sleeves and use the garment as a layering piece over shirts and light sweaters, and we did the necessary calculations to determine how much yarn I would need for the dimensions I was planning. Zephyrette comes in 50-gram skeins, 210 yards each. The yardage for a boxy shape of 50″ around and 23″ long, without sleeves, would be around 1100-1200 yards. Six skeins at 210 yards each would give me 1260 yards. Six skeins, sold.
I have certain design gifts, but creating my own charts and stitch patterns isn’t one of them. My imagination is inspired when I look at an existing design element and invert it, turn it sideways or upside down, elongate it, shorten it, widen it, narrow it, fill in its spaces, color it in ways that alter its character, and I can improvise variations on shapes that I understand. But staring at an empty grid and filling it in is generally not a happening thing for me. Sometimes I can do it, but usually creating new charts and stitch patterns is something I leave to others who do it better than I can. (Delegation is one of my managerial talents.) I have several hardcopy books in my personal library that I always go to when I want a chart for an elaborate flower, and the first book I went to eliminated any need go to the others, because I found what I wanted in about 30 seconds. That book is the elusive Poetry in Stitches by Solveig Hisdal, which happened to be in print for five minutes in 2010, when I got my very precious copy. The chart I used is in the garment featured on the cover of the book.
I acquired the yarn and formulated the main features of the design in October, when I was knitting the items that were the subject of my December 2018 blog post When Vertical Lines Meet Horizontal Lines. My original thought was that this design, with the elaborate flower placed in a grid in a positive/negative alternation, would be the culmination of my thoughts about squares when I wrote that blog post back in December. But everything I was working on took longer than my rosy scenarios allowed for, and other things suddenly needed to be knitted first, so I couldn’t even get to the swatch until the January 2019 Swatchathon. As I said in the 2019 Swatchathon post, knitting the swatch made me alter my plan for the yarn, because the feel of the fabric was too nice to waste on something that was designed never to touch the skin directly. I thought I would make the sweater narrower and use the yarn that didn’t get used on girth to make sleeves whose length would be dictated by how far the yarn went before I used every inch. That would have worked. That’s not what I did.
Work on the sweater kept being preempted by one thing after another until February, and even then it was a secondary priority because I was working on a lengthy to-do list of other knitting that all had February and March deadlines. But I did the math based on my swatch, which gave me numbers for a 50″ circumference at 310 stitches and 10 pattern repeats, after I added three stitches to the edges of the pattern chart, from 28 stitches to 31. At this point I had given up the idea of making a slimmer-fitting garment and using the conserved yarn for sleeves. Oversized clothing is my comfort zone, and I had absorbed the cost of the first set of yarn. So buying more, if I needed to, was now a viable solution if/when I ran short, especially since Melissa, the LYS owner, had invited me to join her on a field trip out to the goat farm, an hour away in Frederick. So I was back to the original plan for the size and shape of the garment. Now that I’m wearing the finished sweater, I can see that it would have looked just fine with eight pattern repeats instead of 10, that is 248 stitches instead of 310, because I knitted loosely on U.S. size 5 needles, which produced a drapey fabric that ended up a larger than gauge. Well, next time. And there probably will be a next time, because I really like this yarn. You have to feel it to believe it.
My first cast-on was a bust, because it was a provisional cast-on that theoretically would have enabled me to add an edging after finishing the body using whatever yarn was left over. But it was really hard to get an accurate number of stitches, and then I dropped stitches and couldn’t reconstruct the provisional cast-on, so I ripped it all out and just cast on the normal way. My swatch stopped curling at the bottom edge after I washed it and laid it flat, which was a sign that the finished garment wouldn’t need to an edging to keep the bottom hem from rolling up. Establishing the pattern in the positive/negative alternation was a spacial challenge, and the first few rows were a very slow go. Actually the first couple of pattern repeats were a surprisingly slow go because the structure of the flower took longer to establish itself in my muscle memory than I was expecting, and carrying out the color reversal required a lot more conscious effort than I thought I ought to have to exert. I think I now understand why: I was ill at the point when the pattern should have entered my muscle memory, and the illness was blunting all aspects of my cognition and mood.
It also took several pattern iterations before the flowers and the checkerboard alternation started to read enough for people to stop saying to me, “Pretty colors. What is it?” But by the time I got to the front-back divide on the fourth sequence of the repeat, requiring the spacial challenge of wrong-side stranding, I had done the pattern 30 times, that is 15 color-reversal pairs, and finally I had a fairly solid concept of how to build the pattern, although the first few rows of the final color change never got automatic and presented an opportunity for the pattern variation that exists in nature. At least that’s how I’m rationalizing “mistakes.”
I have made this boxy dropped-shoulder garment shape for my body often enough to have developed a reliable formula to get an easy and flattering fit without bunching under the armpits, unlike the oversized boxy shapes of the 1980’s and ’90’s, which were tragically unflattering on mere mortals. The garment’s circumference can be very significantly larger than the size of my actual body without looking as if it has devoured me whole, provided that the opening for the sleeve is the minimum size for a close-fitting but comfortable sleeve, and the shoulder is sloped with short rows every inch and a half or so. For me, the magic measurement for the front-back divide is 7 inches, to give a sleeve circumference of 14 inches, which is roomy enough to wear a shirt underneath but not voluminous. Based on the length of the already-knitted pattern repeats, I figured that the divide needed to start at the beginning of the fourth flower sequence, and then I would need to cover a couple more inches, not enough for another flower sequence, but plenty for several iterations of simple checkerboard alternations, into which I would work the short-row shoulder shaping. As I worked my way up the garment, I discovered that my calculations for the amount of yarn I would need for my original idea were completely accurate. But now I needed sleeves.
Once I finished the fourth and final iteration of the flower pattern, knitted back and forth to make the sleeve opening, I rearranged my color pairings for the checkerboards. I wanted to conserve the dark green to make sure I had enough for the edgings at the neck and cuffs, and also it was time to mix up the colors to give the eye something a little different to chew on, to mix a metaphor. It was also time for a trip to the goat farm.
Flying Goat Farm’s yarn wall:
And Flying Goat Farm’s goats. They stop flying and land on the ground when they meet guests.
I already was pretty sure I knew what colors I wanted to get. I no longer had enough of the original colors to get very far with them in the sleeves, and getting new skeins of those colors would have been expensive and much more yarn than I needed. And I felt that those soft, muted colors needed something to electrify them and shake them up. I remembered from when I originally chose the colors, six months earlier, that there was an aqua and a lavender, and those two colors kept coming back to my mental eye. When I saw them in real life, I knew I had remembered accurately. Generally I don’t like aqua and lavender together because the combination has been overused in juvenile contexts, but here a checkerboard arrangement of the two was going to work. The lavender blends nicely with the muted colors, and the muted colors make the aqua more vivid than it is in isolation. The aqua sleeves are the focal point of the composition, where the eye is first attracted, and from there it travels to the other simple checkerboards in different colors and spends some time working through the similarities and differences between the little plain squares in alternating colors and the big squares with the complicated flowers inside, also in alternating colors.
I planned the shoulder shaping by deciding how wide I wanted the neck opening to be, which was eight inches, that is, the center 40 stitches, and then I placed markers outside those stitches and from there I placed more markers every 8 stitches to delineate where the short rows would be, for a total of seven short row turns per side. I shaped the back first to delay dealing with the neck shaping. One thing that I always have to puzzle out about shaping shoulders on the back is the issue of symmetry, since you start at the outside edge on one side and knit to a certain point inside the outside edge of the other side for the short-row turn, which makes the side you start on one row longer until you get to the end of the process. The result isn’t perfectly symmetrical, but the two sides do end up with the same number of rows. Drawing the back and forth path of the knitting on the back of an envelope helps with visualizing it.
The German short row turns, which I executed without any trouble at all when I made my Swing Knitting swatch during the 2019 Swatchathon, kept flummoxing me when I did them this time. I just couldn’t remember the sequence for knitting the turn stitch, turning the work, slipping the turn stitch to the right needle, and pulling the working yarn up and over the turn stitch to put the two legs of the stitch onto the needle. Simple enough, right? My flu-addled brain couldn’t retain the procedure for any of the 28 short row turns I executed for the shoulder shaping of this garment. But I did them, and not badly either. There is a slight structural asymmetry between right and left and front and back of the shoulder construction that I don’t know how to avoid, but I am pleased not to be so burdened by perfectionism that this bothers me. I could have done the fronts perfectly symmetrical with one another because the divide at the neck makes it possible to start the knitting at either the neck edge or the shoulder edge and mirror it on the other side, but I decided to treat the front the way I had treated the back in order to balance that minor asymmetry so that where the back has fewer rows, the front has more, and where the back has more rows, the front has fewer.
The knitting I had already done in the body of the garment gave me a stitch gauge of 5 stitches to the inch (fully a stitch more per inch than my original swatch!), so the number of stitches I picked up at the opening of the front-back divide came to 70 for a 14″ sleeve circumference. However, the actual garment, as it grew and especially after I steamed it after joining the shoulders, became looser than it was at earlier stages, and the opening I had made for the sleeves was more like 8″ per side than 7″, so I stitched together the excess and cast on the sleeves to work them top-down. The main color arrangement for the sleeves was the aqua and lavender checkerboard, but I threw some other colors into the arrangement because everything about this sweater had been regular and too much regularity gets boring.
I had a really good hair day the day I got my modeled project pictures. I also figured out the self-timer on my phone camera so that I can get my best angles and not have to tell my husband that every single one of the 100 photos he took of me makes me want to kill myself.
I am very pleased with this sweater. I’m so glad that I decided to make it something that touches the skin instead of the layering piece of my original idea. The yarn does pill and shed a bit, and I’m brushing the sweater more than I’m in the habit of doing, but the feel of the fabric compensates for that characteristic. It makes me happy to wear it, because the shape and fit suit me and I’m not yanking on bits that trouble me, since there aren’t any. It also does some aesthetic things that motivate my knitting: as a meditation on the simple, ubiquitous shape of boxes, the eye takes a journey vertically, horizontally, and diagonally to absorb the likenesses and contrasts of the squares and the shapes inside the squares. Maybe other people don’t have this experience, but this visual exercise gives me some deeper, nonverbal, intangible, unmeasurable, un-montetizable, totally impractical understanding of the geometry of daily life.
I’m not done with squares. Watch this space.
4 thoughts on “Back In the Box”
I love how this looks on you! And, as always, I’m in awe of your color sense and designing. And I love the goats!
Thanks so much, Laura!
This is such a gorgeous sweater, and it looks perfect on you. Thank you for sharing your process as well as the beautiful result.
Thank you! I’m glad you like it!