In Praise of Elizabeth Zimmermann's Invisible Increase
- Susanna
- 5 hours ago
- 7 min read
Everybody knows about M1L and M1R increases. If you're a bit more adventurous, you might have even tried lifted increases. In this post, I want to take a closer look at a relatively unknown increase type that doesn't get nearly the credit it deserves: Elizabeth Zimmermann's invisible increase. You'll learn how it works, how it compares to more widely-known increase methods, and why it merits a permanent place in your knitting toolkit.
The Quest for Invisibility
Knitters are always looking for the best, most indistinguishable ways to do things in knitting. We want knitted fabric to follow the contours of the body without it being evident where and how it was shaped.
This quest for invisibility is the driving force behind the invention of many a knitting technique. For eons, the SKP (slip 1, knit 1, pass slipped stitch over) was the usual counterpart to K2tog decreases… until Barbara Walker's SSK came along. Nowadays there are myriads of ways to do left-leaning decreases. I've written about them here: 7 Ways to Knit a Left-leaning Decrease.
And what about binding off? The standard bind-off leaves a chain of stitches running along the cast-on edge which can disrupt the flow. Want a bind-off technique that just disappears into nothingness? Tubular bind-offs to the rescue.
Same thing with increases. Even the neatest shaping can leave behind a telltale mark: a little bar, a slight pucker, a line that draws the eye whether you want it to or not. No increase method is completely unnoticeable but the ones widely used today can be improved upon. Elizabeth Zimmermann's Invisible Increase is the most indistinguishable one I've encountered. It's even in the name!
Later on in this post I'll show you how to work this remarkable yet unremarked increase. But first, to understand the building blocks of the EZ invisible increase we first need to take a look at the two increase techniques it's closely related to: M1L and RLI.
M1L: A Left-leaning Strand Increase
M1L (read: make one left) is a left-leaning strand increase, meaning that it's worked into the running strand or thread between stitches. In The Principles of Knitting, June Hemmons Hiatt calls increase methods belonging to this family twisted running thread increases. Self-explanatory albeit a bit clunky.
For many knitters, M1L is the default increase. It's widely used, easy to knit, and often treated as the baseline against which all other increases are measured. Having said that, strand increases aren't entirely invisible. Especially when worked on consecutive rounds or next to twisted stitches, they can leaves quite large gaps that stand out.
To work a M1L:
Use the right-hand needle tip to pick up the running strand between stitches from back to front,
knit through the back loop of the strand,
twisting it to the left.
Want to learn more about strand increases? Read these posts:
RLI: A Right-leaning Lifted Increase
Lifted increases are done by lifting or raising the stitch from the row below, then knitting into that stitch. They are often touted as being more invisible than strand increases. In fact, the Cat Bordhi video that popularized the technique calls them "the most invisible, smoothest paired increases for your socks".
But lifted increases are not entirely indistinguishable, either. The very act of lifting the stitch onto the needle can stretch it out, resulting in elongated stitches and distorting the fabric.
The name lifted increases has become the norm today but in The Principles of Knitting, June Hemmons Hiatt calls increase methods belonging to this family raised increases. In the book she also raises a concern about stacking these increases on consecutive rounds (second edition, page 209):
"Because the technique requires working into a stitch in the row below, it cannot be used in every row of the same stitch column; when necessary to work on the next row, shift the increase to an adjacent stitch column."
To work a RLI:
Use the right-hand needle tip to lift the right leg of the stitch below the next one,
place the stitch on the left-hand (holding) needle,
and knit it through the front loop.
Want to learn more about lifted increases? Read these posts:
How to Knit Elizabeth Zimmermann's (Practically) Invisible Increase
In Knitting without Tears (1971), Zimmermann includes a laundry list of many useful tips and techniques that are part of the advanced knitting vocabulary today, such as the aforementioned Walker SSK, provisional cast-on, sloped bind-off for shoulders, and her version of a stretchy bind-off.
She also introduces this increase method which she modestly titles as a "practically invisible increase". Putting RLI and M1L together, here's how to work an EZ invisible increase:
Use the right-hand needle tip to lift the right leg the stitch below the next one (like in RLI),
place the stitch on the left-hand needle and knit it through the back loop (like in M1L),
twisting it to the left.
The resulting increase get sucked between two existing stitches. A new stitch columns appear out of nowhere! Like I wrote earlier, it's the most indistinguishable increase I've ever encountered.
The drawback of using the stitch from the row below as the base for the new stitch is that EZ invisible increases can't be stacked on every row or round. If you're working on something that requires every-row increases — such as a compound raglan or a contiguous sleeve — it needs to be paired with another increase type.
Is There a Right-leaning Counterpart?
You have to look very, very closely but the EZ invisible increase has a left lean like a M1L increase. But can it be done as right leaning? As far as I know, Elizabeth Zimmerman only used the left-leaning version, the reason for which will become evident soon.
To find a right-leaning version of this increase, I experimented with two things.
Experiment 1: RLI + M1R
First, pick up the right leg of the stitch below the next one but place it on the left-hand needle reverse mounted. Then comes the awkward bit: knit it through the front loop. This is akin to combining a RLI with a M1R. Although the increase now has a distinct right lean, it leaves a very visible bar.
Experiment 2: LLI + M1R
The second experiment was to combine a LLI with a M1R. Pick up the left leg of the stitch two rows below the last stitch (like in a LLI), then knit into the front loop (like in a M1R). The result is slightly better: it's a right-leaning increase alright but still a very visible one.
The results of two experiments are pretty similar but I wouldn't call either one invisible. While a right-leaning counterpart to the EZ invisible increase can technically be done, it's not worth the hassle.
The original increase blends in so well into the surrounding fabric that a mirrored counterpart is not even needed. After all, it's the position of the increase that determines which way the fabric grows, not the lean of the increase itself. If you were to use this increase type in raglan sweaters, for example, you can use all left-leaning ones and none the wiser.
What about the wrong side? Surely the increase leaves some kind of mark? Nope. The wrong side is as smooth as the right, with barely an evidence of where the increases were made.
Why Isn't It More Widely Known?
The EZ invisible increase is an increase method that's easy to knit, made up of familiar components of increases you already know, and — most importantly — is truly invisible. The result is a stitch that blends quite seamlessly into the surrounding fabric both on the right and the wrong sides. Especially in stockinette you have to actively look for it to spot where the increase happened.
Since you don't need a mirrored counterpart, there's also no need to remember which way to twist the stitch. The only drawback I can think of is that they can't be stacked on consecutive rows or rounds, at least not in the same position.
Outside of Elizabeth Zimmermann's own writings, the invisible increase rarely (if ever) gets a mention, not even in the almighty Principles of Knitting which practically has every obscure knitting technique under the sun. One can only speculate about the causes for this oversight.
Part of it is likely simple inertia. M1L and M1R have become the standardized language of increases in modern pattern writing. And what gets used in patterns is also widely taught in classes, books, and online tutorials. Once a convention reaches that level of ubiquity, alternatives have a hard time gaining a foothold, no matter how elegant they may be. Just think of VHS vs. Betamax. (Is that still a relevant example?)
Then there's the question of naming... or rather, lack thereof. Zimmermann wasn't concerned with formalizing terminology but actively encouraged knitters to "unvent" things and forge their own path. Her "practically invisible increase" is more of a vague description rather than a standardized term. It doesn't exactly roll off the tongue and without a succinct abbreviation or acronym, it's harder for the technique to travel. (I hereby motion that we abbreviate it EZinc in knitting patterns.)
Whatever the reasons might be, this hidden gem deserves much wider recognition than it has today. Next time a pattern asks for a particular type of increase, swap it out with an EZinc one and see what changes. As Zimmermann writes in Knitting without Tears (page 27): you will note quite a difference.
Pin this post!
