We all learned at a very young age that even numbers split nicely and odd numbers create those troublesome leftovers, remainders. For some, that thought alone is the only seed needed to sprout an appreciation for optimal knitting design layout. For others, while the math might be easy, visualizing the concept’s application to knitting design can be difficult. Read through this article, ponder the diagrams and, if the layout of Fair Isle or Norwegian knitting motifs has confounded you in the past, hopefully, this will help.

As one gets into any type of stranded knitting, one learns that motifs can be lined up, shifted around, staggered, split, reflected and, indeed, transmogrified in countless ways to create an endless variety of enchanting designs. So I guess I shouldn’t be too surprised by the many e-mails I’ve received from readers in response to my post about sticking with an even number of repeats (and thereby limiting the sizes available) for my Rosy Outlook Vest design. Folks have written to tell me, some quite adamantly, that I’m wrong, that one need not be limited by an even number of motif repeats, that as long as the motifs are shifted over a bit, they’ll still produce a perfectly symmetrical Fair Isle design, whether the number of motifs is even or odd.

Um, no!

Then again, although I’m not inclined to use it (for reasons you’ll soon discover) there might be a wee bit of leeway for allowing odd-numbered repeats, in one special case. Whether or not you’ll choose to use it will depend upon your own take on “perfection” and “symmetry”.

There’s an interesting article on symmetry on Wikipedia. It discusses both the aesthetic and mathematical meanings of symmetry, both of which are important in knitting design. Aesthetic symmetry focuses on the visceral sense of balance in a design, allowing for artistic preference, while the more clear-cut mathematical sense of symmetry demands mirror images and leaves little room for argument.

For my all-time favorite exemplar of perfection, I give you the great Mary Poppins who, as we all know, was “practically perfect in every way”.

Let’s think about laying out our knitting motifs so that they’re “practically perfect in every way.”

My Rosy Outlook vest is a variation upon the ubiquitous “OXO” Fair Isle theme. An “OXO” theme is simply a display of alternating X’s and O’s which can be either lined up above one another like this:

OXOXOXO
OXOXOXO
OXOXOXO

or

XOXOXOX
XOXOXOX
XOXOXOX

or staggered, so that the motifs shift back and forth, like this:

OXOXOXO
XOXOXOX
OXOXOXO

or

XOXOXOX
OXOXOXO
XOXOXOX

You can start the design with either a whole X or a whole O, or you can shift it a bit and start with half of an X or half of an O.

My Rosy Outlook design is a variant of the 2^nd, staggered layout with a shifted starting point so that we begin and end with half O’s and half X’s. Despite the “OXO” name, the repeat actually consists of either “OX” (or “XO”) and it can be measured from the beginning of an O (or X) to the start of the next instance of O (or X), or from the middle of an O (or X) to the middle of the next instance of O (or X). In either layout, the horizontal sequence is the same: …one O, one X, one O…or …one X, one O, one X; who starts first is irrelevant. In fact, the Xs and Os are irrelevant, too. You could have an *#* design with an *# repeat, or a tree-wombat-tree design with a wombat-tree repeat. (Say, there’s an idea!) You get the picture, yes?

“Hey, nice vest!”

The confusing point for many seems to be: Just how many – oh, let’s just call them “XO repeats” - can you use and still produce a symmetrical design? This question often arises when the size a knitter desires lies between or beyond the sizes offered in a pattern. If a pattern has 2 sizes that differ by two repeats, why can’t we just shift things over a bit, fit in one more repeat and end up with just the right middle size in the very same design? And with all the beautiful knitters to be found in every conceivable size, why on Earth would anyone offer a design in only one size?

If you’re only concerned with achieving an aesthetically symmetrical design AND your take on “practically perfect” emphasizes “practically” over “perfect”, as I mentioned, there is a bit of room, and I mean just one teeny, tiny bit of room, in a very limited number of applications, for an odd number of repeats. For instance, if you’re knitting motifs go around the bottom of your sweater or vest, but end before they get anywhere near the top, you might not really need to bother with striving for the ultimate in symmetry. In such a case, as long as you center either an X or an O at the center back and the center front, you might be perfectly happy.

XOXOX layout on front with an odd number of repeats
OXOXO layout on back with an odd number of repeats

So, in this simplified layout, above, we have a total of 5 repeats going around our sweater, with the X centered on the front and the O centered on the back. What’s wrong with that? Granted, it’s not mathematically symmetrical, for the front and the back obviously differ. But, certainly, there is a fine sense of balance and aesthetic symmetry here, for we do have mirror images around the center front and the center back. If it’s just going around the bottom of your sweater and it ends before the armholes, there’s really nothing wrong with this layout at all. However, once you get into the armholes with an odd number of repeats, you’re headed for trouble.

Picture the sides, where the X’s from the front meet up with the O’s from the back. Well, that’s rather nice, isn’t it? The X’s and O’s are whole, they continue in sequence and everyone’s playing nicely, right? So where’s the problem? Let your imagination wander up those sides to the armholes. Naturally, you’re using steeks to manage those armholes with perfection Poppins would be proud of, right? If so, I know you’d much rather avoid the X on one side of the steek, the O on the other. If you had either an X or an O centered on the side line, in addition to those already centered on the front and back, the sides of the armhole steeks would be mirror images of each other. Oooh, wouldn’t that be nice?! That extra bit of symmetry certainly makes for a far more pleasant knitting experience. Knowing that the far side of your steek will be a mirror image of the near side you just knit goes a long way toward simplifying the chart reading you must decipher and remember as you knit along. It also produces a project which is one important step closer to being “practically perfect in every way.” (I know, just the thought of it makes you want to go sliding up a banister, doesn’t it?!)

The thing is, to have an OXO design that is centered at the front, back and on the sides, the only way to do it is with an even number of repeats.

And now, for my diagrams, proving once and for all that there is at least once case in life where I don’t instinctively leap to blind allegiance for all things “odd”:

Fig. 1:

Think of my Rosy Outlook Vest, or any other OXO garment, opened in the front and laid flat. Now check out the diagram, above. Each of these cases shows an OXO example that has been centered over both the front and the back. The solid red lines at either end of each case represent the center front (CF) closures. The dotted red line down the middle of each case represents the center back (CB). The dotted green lines at the midpoint between the center back and center front represent the armhole / side lines (S) of the garment. These vertical lines mark off the four sections which represent the left front, left back, right back and right front of a garment. The variable R represents the # of XO repeats across one of those 4 sections. Unless you’re knitting for Quasimoto, you definitely want those 4 sections, and their corresponding R values, to be equal.

The uppermost example shows you a simplified 4-repeat OXO layout which is symmetrical around CF, CB and S. As you can see, between each of its equidistant vertical lines, we have one full repeat (2 half X’s = 1 X, and there’s 1 O), and around every vertical line drawn, the motifs form those beloved mirror images.

In the middle case, I’ve added one more repeat, or more precisely, half of a repeat to each end, to get an odd-numbered repeat OXO design, as so many of my e-mailing friends have been suggesting I do. What happens now? We can still center the motifs at the center back and center front. We can still draw the vertical lines so that every section between them has the same number of repeats (R = 1 ¼, in this case) so we know our design is evenly distributed. So what’s wrong? Look at the S lines. You won’t have that nice mirror image around your armholes. Granted, I’m sure some of you are saying not only is that not the worst thing in the world, it may be the last thing in the world you’re apt to care about. I have to admit that, knitting efficacy aside, it’s probably better not to get too fixated upon other’s armholes. But let your imagination travel a bit further up the armholes to the shoulder joins. Here is a great example of what happens when one trades mathematical symmetry in knitting design in favor of squeezing in more sizes. That’s the great Alice Starmore’s wonderful Oregon Autumn design which was reprised in six, count ‘em, six sizes in the Holiday 2007 edition of Vogue Knitting. Each consecutive size has one more repeat than the last, so we know that 3 of them have an even number of repeats, and 3 of them have an odd number of repeats. Uh oh!

Fig. 2:

In my first diagram, Fig. 1, I showed you what happens when you add an extra repeat to an even numbered OXO design by keeping things centered on the same motif elements, but you add one more repeat, or more specifically, half of a repeat, to each end. The resulting odd number of repeats gives us mismatched shoulders. Fig. 2, above, shows you the problem with the odd-numbered repeat sizes (every other size) in the Vogue version of Alice Starmore’s Oregon Autumn. Here, rather than keeping the centers on the same spot by splitting the extra repeat at the ends, one whole repeat is just added at one end. In either case, the odd number of repeats results in the same problem: whether you shift or split your added repeat, if the rest of the design is centered on the front and back, using an odd total number of repeats will always lead to mismatched shoulders.

I can’t imagine there’s anyone in the world better schooled in Fair Isle design than AS. I also can’t image that, with such a popular design, any magazine editor would want the flood of disgruntled readers that would ensue if they’d didn’t offer such a masterpiece in their entire readership’s range of sizes. Hence the dilemma. Some people might not mind mismatched shoulders. Perhaps they think it’s a small price to pay to have the pattern in their size. For some of us, the design imperatives that limit the perfect result outweigh the marketing goals of offering everyone’s size. Personally, I’d be mad as a hatter if I put all of that time into knitting such a complicated design, only to have mismatched shoulders. I’m not comfortable with expecting others to do it, either. None of us is perfect, but why not shoot for perfection in our work, and come as close as we can? The magazines understandably prefer to offer the full readership’s range of sizes, even if not all of those sizes are created equally. For me, it pays to be on the internet and independent, so I can happily offer my even-numbered repeat designs, despite their limited range of sizes, without compromise. For you, before you delve into any OXO pattern, I think it will pay to grab some paper and pencil, scribble out your Xs and Os, chop up your four sections and find out just how symmetrical that design you’re thinking of spending so much time and money on really is.

So what can be done for those who would dearly love to knit Oregon Autumn, or Rosy Outlook, or any other Fair Isle design, but the even-numbered repeat patterns don’t come in their size and they rightly insist upon matching shoulders? Try swatching to see if you can knit the motifs a bit more tightly to get a smaller size, or a bit more loosely to get a larger size. Or try using a different gauge of yarn. You might have to lengthen or shorten the design by a motif or so, but you can do that and still keep the other design imperatives intact.

Does this mean that odd-numbered repeats are to be entirely avoided? Actually, no, there are some cases, in some different types of Fair Isle designs, where they’re worth considering. Look at these examples:

The top case shows a staggered OXO design using 4 repeats. Although it uses an even number of repeats, there’s a problem. The problem is, it has not been centered, so we have mismatches all over the place. Some folks might put this around just the bottom of a sweater and it would certainly be passable. But I’d rather see even that centered, as it was in the first diagram. Just shifting it ¼ of a repeat would do the trick, as we see in the middle case here (which is the first case from the first diagram). The bottom case shows the one rare case in which I’d use an odd number of repeats, with some important caveats. As you can see, the repeats are vertically staggered. I wouldn’t do this without vertically staggering the motifs, for, as you can see, since it’s not centered, we have O’s and X’s opposing each other all the way up the center front. As long as you vertically stagger motifs, and as long as there are plenty of rows in which the staggered effect goes back and forth, to visually distribute the motifs, I think the overall effect, while not mathematically symmetrical, is arguably aesthetically symmetrical, and it could work. As I prove daily in so many ways, there is joy to be found in being not-quite-so-practically perfect!

So why didn’t I do that in my Rosy Outlook design? I had my reasons – two, in fact. First of all, notice that the “O”s in my vest, which are really roses, are much weightier design elements than the “X”s. Furthermore, there are only 3 rows of them that line up in the buttoned section. So, if you shifted things over to fit in (or remove) a repeat, you’d end up with one side of the button placket having two heavy roses and one light X; the other side would have two light Xs and one heavy rose. If you stared at it long enough, you’d start listing to one side, and you might even topple over. We could be causing accidents all over the place! And, oh yeah, it wouldn’t be symmetrical. By now, I suspect you all know how I feel about symmetry, right?

So here’s a recap of my Fair Isle layout guidelines:

1) For a perfectly symmetrical OXO design, whether the motifs are staggered or stacked, use an EVEN number of repeats and CENTER them, so that either an X or an O is split at the front, at both sides and in the back.

2) For a potentially aesthetically symmetrical OXO design, you might get away with using an ODD number of repeats and staggering them, splitting either an X or an O at the sides and having whole Xs meet whole Os at the front, but don’t try this unless you have enough vertical repeats to create a well-distributed, staggered effect.

3) Remember, these are guidelines, not rules. There are no rules in knitting! While beloved traditions and strong opinions abound, the best advances come from breaking the “rules” set by those traditions and opinions.

Psst - did I mention that the swirls in the center of roses are not symmetrical?

For more of my designs, symmetrical and otherwise, click here.

Happy knitting from Mary Ann Stephens

If you’re wondering “Where in the world did she find such a handsome wombat!?”, look here.

Devastatingly handsome wombat appears with permission of Tom Dempsey / photoseek.com

Like ‘em? I’ve just finished my Chrysanthemum Mittens, above. I designed these to be knit in Dale of Norway Hauk yarn, which is essentially good old Heilo 100% Norwegian sport weight wool that’s gone high tech with a Teflon infusion to make the yarn especially water and stain resistant. The Teflon even makes the yarn slightly softer. And it knits at exactly the same gauge as Heilo. So Hauk is a perfect alternative for stain-prone comfort seekers like me. It’s great stuff - I love it!

By using needles a couple of sizes smaller than I’d normally use for sport weight yarn, I ended up with especially dense, warm fabric that I’m really happy with for mittens. But just how much of this yarn we need for these mittens is another question.

I used 1 ball of each color (black, barn red and natural) and ended up with a generous bit of the black left over, tons of barn red left over and a downright stingy four feet of the natural left over. It was a close call, too close for my taste. It’s all too easy to imagine knitting just a wee bit looser and running out of the 1 ball of natural before finishing up. I’m figuring kits to make 1.999 mittens might not be terribly popular . So, it’s time to look further into a question that’s been bothering me for a while now: Do Hauk and Heilo really have the same yardage???

According to the ball bands, they do. Like all Dale yarns, both Heilo and Hauk are put up as 50g balls. Both say they provide 100 meters. But I wonder if that’s precisely true, or merely an approximation, or…perhaps, not even so approximate an approximation.

Given: Heilo + Teflon = Hauk. Assume the 50g does truly stay constant. I’m no chemist, but last I heard, Teflon is not weightless. So, as Teflon is infused into the yarn in the dye bath, it increases the wool’s weight, right? So I’m betting there’s a proportionate decrease in the yardage per 50g ball. Is it so slight a change that there’s no sense in changing the yardage numbers? Or is there more to it?

Having made the first mitten in my second pair ofChrysanthemum Mittens, this time in Heilo, I’m taking a look at what I have left so far in the natural and it seems that I’ve got considerably more natural left with the Heilo than I did with the Hauk. Granted, I’m just going on memory at this halfway point. I’ll report back to you when the second one is done. It will be interesting to see just how much more, if any, of the Heilo natural there is to compare with my 4 feet of Hauk. Any guesses? Stay tuned!

A few knitters have asked me lately how I finish a knitted hem. I just came across the above old photo and thought it might be useful in answering that question.

This photo shows how I slipstitch the hem inside my hats. After I cast on and join my work circularly, I knit a couple of inches for the inside hem. Then I purl 1 round to create a foldline. (You’ll see a brown purled foldline to the right of the mostly solid grey hem.) Right after I purl the foldline, I change to a larger needle size. By using the smaller size on the inside, I know I’ll create a smaller concentric circle for my hem and that it will sit nicely inside the hat without pushing on the outside and making the bottom of the hat flare unnecessarily.

Rather than changing needle sizes, some folks might choose to stick to one needle size and increase several stitches immediately before or after the purl foldline to get the same nested circle effect. That can certainly work, but I prefer avoiding potentially unsightly increases which can rumple my foldline. I let the needle change do all the work for me on the sly. Just how different my needle sizes would be depends upon just how different the nested circles must be. For example, with heavier gauge yarn, we need more of a difference, so the needles might differ by two, even three sizes (for the really bulky stuff); for fingering weight yarn, one size difference will do.

Once I’ve worked the entire outside of my hat, I’m ready to slipstitch the hem in place. I started by tucking the hem inside, right on the purl stitch foldline. Once I’ve figured out just which round of knitting on the outside is the same distance from the foldline as the cast-on edge, one loop from each stitch on the cast-on edge (solid grey) gets joined with the purl bump from the same stitch column on the inside of the hat. It’s well worth double checking that the purl bump you’re scooping up is indeed from the same stitch column as your cast-on stitch - otherwise, your hem will be skewed. (Yeah, ask how I know! ) If you carefully match the first stitch you hem, you’ll be set for the whole round. Loosely join a few consecutive stitches, then snug up the tension, checking the right side of the work to make sure that you’re not pulling so tightly that you create a demarkation line.

I don’t really enjoy hemming on any sort of fiber. It’s not creative; it’s drudgery at best. Back in my days as a corporate slave, the scotch tape dispenser on my desk was probably accessed as often for hemming as for any legitimate office purpose. Same with the stapler. Of course, anything I knit is far more important to me than any old corporate get-up, so I don’t skimp when it comes to knitted hems. It’s well worth taking care to join every single stitch in the knitted hem. Short cuts on hems show up badly. (Okay, maybe not as badly as staples and scotch tape, but forget the shortcuts and you’ll be happier in the long run.)

I know some people like to knit their hems in place as they go by knitting stitches from the cast-on edge together with those from the equidistant round on the outside as soon as that round is encountered. No doubt, there are some folks who have perfected that method. But from what I’ve seen and what I’ve tried, that method is more apt to lead to a tight ridge that can show up unattractively. It’s easier to adjust the tension a few stitches at a time with this hand sewn method you see above. Especially on hats, a tight ridge squeezing around our heads is probably the last thing we need. Ouch! Give this slipstitched hem a try and tell us how it goes!

I’ve just added the free knitting pattern for these Norwegian Teddy Bear sweaters to my Kidsknits.com site. You’ll find links to this and a few other free knitting patterns I’ve designed on the left-hand side of my homepage, Kidsknits. You can also get to the pattern for the above sweaters by clicking here: Monkster Gets Stranded.

I’ve had a free pattern for a similar, but plain, Teddy Bear sweater (aka Monkster’s Sweater ) out there for a while:

Both of these wee sweater patterns use essentially the same construction method, with a few minor differences. The old one was made with worsted weight yarn; the new pattern uses sport weight yarn. The old sweater was solid-colored, (but you can add any motif your heart desires); the new sweaters, obviously, use “stranded” knitting technique, (but you can create your own Nordic motif, opt for stripes, or keep it simple and solid-colored, too.) Both patterns are worked circularly, from the bottom up, with raglan shoulders, in this fashion:

The fully-fashioned decreases in the old version were chosen to highlight the points at which the raglan shoulders were mitered. That method is especially nice for solid-colored knitting - the decreases form an interesting feature on an otherwise blank canvas. The decreases in the new version were chosen to minimize the visual disturbance of the raglan miters. I typically opt for subtle decrease methods for multi-colored, Nordic knitting so that my shaping and my motifs don’t visually compete with each other.

The new sweaters, at top, both use the same technique you’d find in most any life-sized raglan sweater: a few stitches at the base of the armholes are put aside on stitch holders while the yoke is worked as one circle. Once you’re all but done, those few held stitches are grafted together at each armhole base. This helps in two ways: it provides more flexibility at the underarm for the wearer (although I’m not sure guys like Monkster really appreciate that fact) and it makes the joining round, wherein the sleeves and body become one, a good bit more manageable. But, it also leaves a little bit of grafting work to take care of at the end. (See my “Grafting 911″ post, under “Technique”.)

In the original, plain sweater, I did not cast off any stitches at the base of the armhole. So, the joining round for that version has to be worked with the “circular” knitting flattened for a round or two, and it’s a bit trickier to wiggle your yarn and needles in and out of the flattened work. The most saintly among you may find yourselves cursing like sailors at that point, but it does save you a good bit of finishing work. It’s a nice shortcut for a quick toy sweater, but it’s not something I’d recommend for the opinionated, comfort-seeking, self-propelled types. Personally, I prefer the method in the new, stranded design.

At the bottom of the new pattern, in addition to the charts for the above motifs, you’ll also find an empty chart awaiting your own design. I’d love to see what you come up with, and I’m sure others would, too. Send a photo of your finished sweater and I’ll share it here. Have fun!

And now, for something completely different!

Ordinarily, if I want to produce wonderful effects with color, 99% of the time, I prefer stranded knitting over other methods. It’s a very rare day that I’m working on a monochrome project, such as you’d often find in lace knitting. Yawn! For color interest in lace knitting, I know you can use variegated yarns, but while many of the variegated yarns do provide plenty of color interest, they often do so while diminishing the impact of the lace’s texture. The Svale tank top on the cover of the new Dale of Norway Book 172 shows magnificent blending of colors and the color gradient produced actually highlights the lace’s texture. It’s all done in pretty common lace stitches using only 1 color per row/round - it’s not stranded at all.

I was lucky enough to snare a model of that Svale tank, so I thought I’d share the wealth, above and below. FYI, the tank instructions are in ladies’ XS - XL. A large size takes 6 balls of the main color (buttercup 2005) and 2 balls of each of the following: powder blue 5812, jade 7432 and lime 9425. BTW, Book 172 also has instructions for several other summer tops for all girls from 2 to 102! As always, feel free to write or call with any questions. Happy knitting!

I’ve just finished my new Rosy Outlook Vest. It’s made in 7 colors of Dale of Norway “Daletta” 100% washable wool. It’s knit at a gauge of 27×27. That’s slightly looser than the ball band gauge of 28×38 - the gauge Dale uses on most of their Daletta jackets (like Ingeborg, for you Dale aficionados out there.)

I wanted something a bit lighter, somewhat more fluid, more Fair Isle than Norwegian. It worked out that, by going up one extra needle size, I hit the 27×27 gauge. I wouldn’t recommend Daletta at 27×27 for a jacket, but it’s perfect for a Fair Isle-type vest. And symmetrical, too! (I’m forever delighted by symmetry.)

There’s very little weaving-in to be done with this vest. The color changes are all done mid-front so they can be secured with machine-sewn covered steeks, using the same technique I developed for my Tiger Lily Jacket.

Halfway through, the “Rosy Outlook Vest” was almost headed for the Hefty bag. The folks at Dale of Norway had just decided upon a major shift in the Daletta color range. The colors in my vest were headed for Discontinuation Station. Sure, it’s fun to conjure up new colorways, but, I loved those colors together with the rose motif! I decided to forge ahead and complete it. I can’t promise these colors will be available for long, but there will soon be other colorways, too. Here you have the original:

Whadyathink? Alas, it’s not for everyone - there’s only one size. Boo, hiss, I know, but I had my reasons.

The finished chest circumference is about 45″; the finished length is about 26″. The rose motif has a 30-stitch repeat. To be symmetrical (as we must ), we’d have to add or subtract 2×30=60 sts, or almost 9 inches, to get to the next size. So a smaller size, finished, would be about 36″ around; a larger size, about 54″. If you figure 4 to 6″ of ease, we’re talking about bodies that have chests as small as 30″ or as large as 50″. I know such folks are out there, and I apologize for excluding them. However, figuring yardage requirements for 7 colors is not an easy task with multi-colored, stranded projects unless each size is actually knit. That’s not a task my tired hands are up to. So, I’m sticking with this one size, which would be a ladies’ large in some labels, ladies’ extra large in others.

If you’re interested, here’s a link to one of my sites for buying the Rosy Outlook Vest. As with all knitting projects, before you dive in, I recommend comparing the finished dimensions to those on a comparable piece you already have. That’s the best way to preview fit. And as always, don’t hesitate to write or call with any questions. Happy knitting!

Knitting patterns start with the presumption that we will all follow every direction to a “T”. Experience teaches us that none of us is perfect, and, no matter the skill level, we will all eventually make some mistakes. Furthermore, it seems there is some perverse law of Nature that dictates that a fair amount of those mistakes will go entirely unnoticed in the making, yet they will become glaring blunders right about the time we think we’re done.

Certainly, none of us relishes the idea of unraveling most of our beautiful work just to get back to our old missteps. But as we bind off the last stitches, hold up our masterpieces for inspection and feel our hearts sink, is it really time for the Hefty bag? Not so fast! Grafting to the rescue!

Grafting (aka kitchener stitch) is often used for joining shoulders invisibly. It’s also the preferred method for joining the bottom of raglan sleeves to sweater bodies. Most knitters familiar with grafting learned the technique by following patterns with one or both of those applications. However, the most valuable use of grafting may be the one no pattern will ever recommend: rescue from the dreaded Hefty bag.

Having heard far too many stories of 99% gorgeous creations being banished to the purgatory of a Hefty bag, I’ve decided to share a lesson on grafting that will hopefully inspire some of you to rescue, perfect and proudly wear those hidden near-masterpieces. The steps below show you how to use grafting to repair isolated areas within one row of knitting without having to rip out any extra rows of stitches. Many of those previously-invisible-but-now-glaring mistakes can be essentially erased this way. You can also use the same technique to insert or delete rows to repair, lengthen or shorten your work and then rejoin it, effectively splicing your work and editing it to your taste, without undoing any of the good parts. Phew!!!

Note: In hope of showing a visually- friendly example, I searched my stash for the closest thing I had to solid-colored (yawn) knitting. I used red, contrasting yarn so that you’d easily see what was done, but of course, I suggest you use matching yarn. The “mistake” in these steps is imaginary; the stripes are irrelevant. But don’t worry, I have plenty of bona fide blunders to share with you – see bottom!

Find a well-lit table and spread your work out flat. For mistakes isolated within one row, pick up every stitch in the rows immediately above and below the row you intend to repair. The problematic row lies untouched, in the middle. Pick up the stitches by inserting the tip of your needle under just the right side of each stitch’s “V”.

In the steps below, notice that the work is done between the 2 knitting needles - not below the lower one, nor above the upper one.

Carefully snip a stitch in the middle of your problematic row. Unravel as many stitches within that row as needed to “erase” your mistake, leaving enough yarn length on each side for weaving in later.

Grafting Step #1: Thread a blunt yarn sewing needle (SN) with your yarn. Insert SN right-to-left through the first stitch on the lower knitting needle. Let that stitch slide off your knitting needle, then pull SN and the grafting yarn through that stitch, leaving a couple of inches of tail to be woven in later.

Grafting Step #2: Bring SN up and insert it right-to-left through the first stitch on the upper knitting needle. Let that stitch slide off the upper knitting needle and onto SN. Pull yarn through loosely. (If the next stitch on the upper knitting needle is a different color than the one you just pulled through, see notes on stranded grafting, at the very bottom. Otherwise, go to Grafting Step #3.)

Grafting Step #3: Insert SN left-to-right through the next stitch on the upper knitting needle. Pull yarn through loosely. Let that stitch stay where it is on the upper knitting needle.

Insert SN front-to-back through exactly the same spot the grafting yarn came out of in Step #1.

Okay, let’s see that again! Here’s Step #1 again. Remember, once we scoop up that first stitch, we take it off the lower knitting needle.

Here’s Step #2 again. We’ll take that stitch off the needle, too.

In Step #3, it’s important that we remember to insert SN from the left to the right and pull our yarn through but still leave it on the upper knitting needle.

Now here’s Step #4. See, SN is going from the front to the back, right into the middle of that stitch we had dropped off the lower knitting needle in Step One.

1,2,3,4,1,2,3,4,1,2,3,4…ta-da!

As you can see below, I really am woefully familiar with that heart-sinking feeling at final inspection. “Oops” was easily remedied with 1 row of duplicate stitch - a small price to pay for knitting in poor lighting and selecting the wrong color for 1 row. But, “yikes” shows a real head-banging blunder: three rounds entirely forgotten. Doh!

These sleeves were a fair amount of work and I really did not want to frog the second one. I decided to use long circular needles to pick up each entire round immediately above and below the point of the missing rounds, leaving one round in between for cutting. I snipped the sleeve open all the way around. On the bottom half, I reknit the cut row. Then, I knit the 3 missing rows. Finally, I grafted the new & improved bottom back on using the above technique, keeping in mind the special consideration for stranded knitting mentioned below. Fortunately, the 3 missing rows were not affected by increases. (Regrettably, splicing and shaping don’t mix.) As you might imagine, you can use this technique to shorten or lengthen all sorts of work, as long as there is no shaping within the spliced section.

In a case like this, where we want to pick up the entire circumference, it pays to use extra long circular needles that can be looped back on themselves. If you do this with double-pointed needles, you might want to secure the ends with rubber bands or some other sort of stopper.

Special consideration for any stranded grafting: Look closely at the photos on the simple striped swatch, at top. Notice that the work on the bottom needle joins the adjacent sides of two different stitches and the work on the top needle joins the two sides of one single stitch. Since you’re dealing with two different stitches on the bottom (in Step #4, then the following Step #1) you may or may not encounter different colors, depending upon your Fair Isle patterning. In any case, it really won’t matter for the bottom needle – Steps #4 and #1 for Fair Isle grafting are worked in exactly the same manner as in the above example. Since both elements joined at the top (i.e., Step #2, then Step #3) form one single stitch, they should always be the same color. With Fair Isle grafting, after you’ve done Step #2, the next stitch on the upper knitting needle may be a different color. Well, that just won’t do! If the loop for Step #3 is a different color than that for Step #2, once you’ve slid off your Step #2 stitch, notice that its other half has become a strand/float going to the left behind that differently-colored Step #3 stitch. Forget about that different-colored stitch we’d normally use and pick up the same-colored left side that becomes a strand/float. Now you can proceed with Step #4 as above.

kidsknits.com

skiknits.com