I learned to knit when I was 4 years old, and grew up working from vintage/ retro patterns, mostly from the 1970’s. One of my favorite pattern magazines to this day is Women’s Day’s “101 Sweaters you can Knit and Crochet,” published in 1976 (and sold for 95 cents). The patterns are terse, and directions for an entire sweater often fit in 2 columns of a 3-column formatted page. Implicit in these patterns is the assumption that readers will be able to increase/ decrease a given number of times evenly over the total number of stitches, an assumption that I think is no longer valid. In order to assuage the impending sense of doom that comes with directions to “increase evenly…,” I’m going to work through distributing 6 Events (which can be increases/ decreases/ special stitches/ beads/ whatever) evenly over 87 stitches. I recently followed this procedure when distributing button-holes in my “Moorish Lattice Cardigan”, although the details of the numerical values were a bit different.
The most important aspect of this task is to consider not the number of Events (which I’ll call “E”), but the number of spaces between Events, which I’ll call intervals (or “I”). There are a few different ways in which the Events can be distributed:
A) The Events are distributed such that there is one at each end of the knitting and 4 more in the middle. This distribution is useful for, say, placing beads along the short edge of a scarf. In this case, there are 5 intervals, or one less than the number of Events. I= E-1.
This set-up is closest to the one I followed for my Moorish Lattice Cardigan. To avoid placing button-holes in the first and last rows of the button band, I allowed 2 rows on either side of the first and last button-hole, and distributed the holes over the remaining number of rows.
B) The Events are distributed in the middle of the work, such that there is a plain interval at each end. This distribution is useful when the edge might be a transition to a separate section (rather than an endpoint), as in a shawl. Here there are 7 intervals, or one more than the number of Events. I= E+1.
C) This set-up is similar to B, but instead of a full interval at each end, there is half an interval. This set-up is useful if the two edges of the work will later be seamed together, as in a sleeve. The total number of intervals is equal to the number of Events. I=E.
D) Here the row starts with an Event, and ends with a plain interval. This distribution is often used for pieces that will be seamed to look as if they were worked in the round, e.g. for a sweater back and front. There are 6 intervals, which is the same as the number of Events. I=E.
E) This set-up is similar to D, except worked in the round. The number of intervals equals the number of events. I=E.
To start, figure out how many stitches will fit in each interval. For each situation:
A) I= E-1= 6-1= 5. The length of each interval is 87/5= 17 Remainder 2.
B) I= E+1= 6+1= 7. The length of each interval is 87/7= 12 Remainder 3.
C-E) I= E= 6. The length of each interval is 87/6= 14 Remainder 3.
The quotients 17, 12, 14, respectively, tell us how many stitches will be in each interval. If you worked exactly that many stitches in between Events, there would be 2, 3, 3 stitches, respectively, left over at the end of your row/ round in each case.
What to do with those extra stitches? The easiest solution is to put an extra stitch into some of your intervals. The value of the remainder tells us how many intervals will get one extra stitch. In case (A), there will be two 18-stitch intervals amongst a total of five intervals (the remaining three intervals will have 17 stitches each, as initially calculated); you can incorporate the extra stitch into the 2nd and 4th intervals. In case (B), three of the intervals will contain an extra stitch (13 stitches), while the remaining four intervals will contain the calculated 12 stitches. Here we have to distribute 3 larger (13-stitch) intervals in a total of 7 intervals. Intuitively, the larger intervals can be alternated with the normal-sized ones (i.e. intervals #2, 4, 6 can contain 13 stitches each).
What if intuition fails you in distributing the larger intervals? Say you had 7 large intervals to distribute amongst a total of 20…. Then simply repeat the process again! Now the number of Events is the number of large intervals (7), and the value over which to distribute these is the total number of intervals (20).
20/7= 2 Remainder 6.
This means that you could work every 2nd interval as a large one (i.e. in pairs of normal and large-sized intervals), in which case you’d have 6 normal ones left at the end of the row/ round. Instead of stacking them all together, you could distribute them such that 6 pairs of (normal, large) intervals have an extra normal-sized interval immediately following, while the 7th does not.
Finally, how do you go from the quotients and remainders to the final pattern? The key here is deciding whether your Event consumes a stitch. If your Event is an M1 increase, no stitches are consumed. In that case, the patterns in each scenario would be as follows:
A) M1, (work 17 sts, M1, work 18 sts, M1) 2 times across, work 17 sts., M1.
B) (Work 12 sts, M1, work 13 sts, M1) 3 times across, work 12 sts.
C-E) I’ll leave you this one to work out!
In case (A), doing an M1 at the beginning or end of a row is tricky (if not impossible), so if you’re planning on using an M1, it might be easier to use one of the other distribution schemes (B-E), a different type of increase, or move the M1 one stitch in from each end.
If your increase consumes a stitch (e.g. kfb), then your pattern will be as follows:
A) (Kfb, work 16 sts, kfb, work 17 sts) 2 times across, kfb, work 15 sts, kfb.
B) (Work 11 sts, kfb, work 12 sts, kfb) 3 times across, work 12 sts.
C-E) I’ll leave this one to you!
This workflow is universally applicable to distributing any number or types of events over any number of stitches. The devil lies in the details of choosing a distribution scheme, and deciding how many stitches will be consumed by your Event.
Or… you could take my mother’s approach and just eyeball the interval size!