(warning: lots of technical knitting details ahead…so if you aren’t into that kind of detail, just skip this, or read the first half, which isn’t so technical.)
Yesterday I spent four hours in a knitting store painstakingly paring down a pattern in Barbara Walker’s Third Knitting Treasury. It was a seven-rayed leaf or feather pattern (I’ve forgotten the name) that looked like it might be interesting for the peacock “scale feather” pattern, but at 29 stitches across (15 per feather) it was going to come out too large, so I scaled it down to a five-rayed feather pattern.
Scaling down lace patterns isn’t that hard as long as you understand the basic rules, are willing to spend a couple hours fine-tuning to make sure the stitch counts work out right, and accept that your first few attempts will to look like a mutated grapefruit. (Particularly advanced knitters may wind up with a mutated squid. 🙂 ) Usually you have to trim the length as well as the width of the pattern, which means you have to pay close attention to what’s going on in each row–this is where a photo of the finished work comes in handy.
Then, of course, after you’ve designed the basic pattern, you have to work out the half-drop–how to nestle the individual motifs against each other so that, when knitted, you get a nice even grid. This is tricky at the best of times, and when you are working with a pattern whose number of stitches changes from row to row, it can become a nightmare. I’ve found that it makes life MUCH MUCH simpler if you just grid out two patterns side by side, running parallel to each other, with blank boxes in between, like this:
x–
–y
x–
–y (where x and y are different rows of motifs; the -‘s are just there because HTML doesn’t like blank spaces)
This makes it really easy to see where the half-drop occurs, and which stitches are part of which pattern. Then you muck around with the stitches in between until you get an effect you like.
That’s also a nice way of charting out lace patterns that include a half-drop (two iterations of the pattern offset from each other)–then you can clearly differentiate which stitches belong to which motif. Makes it a lot easier to see what’s happening. I’m a visual person, so I chart everything–I LOATHE the written-out patterns, you can’t tell *anything* about what’s going on. All you can do is slavishly reproduce them exactly as written. Chart it out, and you can SEE what’s happening.
At any rate, I spent several hours painstakingly reworking the pattern, doing a half-drop, making sure the stitch counts worked out, normalizing the number of stitches between patterns…then got home and started knitting the first few rows, and realized, to my horror….that the pattern was going to come out UPSIDE DOWN from the direction I needed.
Aargh.
So I have just spent the last hour or so reversing the pattern, which (as knitters know) can be pretty tricky, because knit stitches really don’t reverse all that well. In particular, there are a few stitches that *really* don’t reverse well.
So, a few tips on “reversing” difficult stitches:
First, a chart index:
/ = decrease (usually k2tog, but I’m not too fussy about the direction for this discussion)
– = purl stitch
1 = stockinette stitch
0 = yarn over
M = make one stitch (pick up from row below and knit)
An increase in one direction becomes a decrease in the opposite direction. Fortunately, since most patterns have balanced increases/decreases in each row, you can pretty much ignore this and flip the pattern top to bottom:
/11110
/11101
/11011
/11011
/11101
/11110
Technically, you are swapping the locations of the increases/decreases–but unless you want to think deeply about it, let’s just say that these patterns do flip easily and reasonably intuitively, though you may have to fiddle with the direction of the decreases.
However, if you don’t have balanced increases/decreases, life gets a little more interesting. For example:
–1–
–1–
–0–
—-
looks like a ladder of stockinette stitches rising up out of a little hole (the yo). Unfortunately, when you reverse it, it doesn’t work right:
—-
–/–
–1–
–1–
preserves the stitch count but hasn’t got the “hole”.
—-
-/0/-
–1–
–1–
preserves the stitch count *and* has the hole, but the hole is offset to one side–it doesn’t end the line of stockinette stitches, it’s half a stitch to the side. Looks icky.
So how to create a hole at the *top* of a line of stitches? I did a little bobble:
-/ – \–M3–
–1–
–1–
The M3 (p1,k1,p1 into same stitch) forces the top stitch open, creating a little hole; then the next row takes out the extra stitches. (Doing it as two p2togs plus 1 p2tog to the side, rather than one p3tog pulls the stitches of the bobble apart, holding the hole open.) It creates a dense bit at the top, but you get the nice line of stitches terminating in a hole.
Stitches like K5tog are very hard to reproduce; I’ve just been replacing them with M5s. If anyone knows of a better option, lemme know.
Better yet, if anyone knows of a lace book with detailed instructions on how to do these things, let me know. I wish they were more common, but from what I’ve seen most knitters are content just to work within a pattern–not much innovative thought there.
Anyway, I have to run off, I’m going to a farm today to check out some fleeces! Utopia Ranch, run by Jean Near, produces some of the most outstanding merino fleeces I’ve ever seen, and I’m looking for an extra-lustrous one to sample for the peacock shawl. I’ve never met a sheep whose fleece I’ve used, so it should be pretty cool; and I’m meeting an old college friend who also has a farm, so I think it’ll be lots of fun. 🙂
Tien
