Apologies because this is an old post, first published in 2007. It’s about one of the first kilnformed glass techniques I ever experimented with, around 1998, but I’ve gotten a couple of follow-up questions, so thought I would highlight.

I was trying to slump a star-shaped piece through a drop-ring that was a bit too wide and it fell through. I peeked into the kiln and saw the half-slumped piece curling up; I liked the shape so much I started trying to do it in earnest. Enjoy!


Hermissenda crassicornis, a type of nudibranch I used to see when diving in northern California. Depending on the time of year, its body can look just like glass.

I really like nudibranchs, the wildly colored, shell-less critters that wend their way through kelp beds, coral formations, and sea floors around the world.

In fact, I’ve been making glass in honor of nudibranchs for maybe 20 years.

If you’re interested, you can learn more about these critters on  Sea Slug Forum. It’s a very nicely put-together website, with clean navigation, lots of photos, and an unswerving purpose: Get folk hooked on the wonders of sea slugs.

It’s worth a visit.


OK, so it doesn’t really look a LOT like a nudibranch, but that was the intent… 😉

My glass sea slugs are fun to make, and a good way to use up scrap glass. I’ve done them in float glass, but float’s extreme tendency to devitrify (i.e., start to crystallize and become not-glass), can make them kinda crusty-white looking after the three to four firings required.

I’m using a pretty rough firing schedule, almost guaranteed to bring out scum, if not full-blown devit. If you want sinuous, graceful nudibranchs, use fusible glass.

Usually I’ll wait until I have a fair amount of clear glass sheet glass saved up, then add spare frit, stringers, dichroic bits, and/or powder for color.

Devit, looking a bit like scum or rough spots, can start to appear on the tiny terminals (lower center) by the second or third flat fusing. When you’re using scrap and frit together, you’ve got a lot of surface area with devit potential.

Even with regular kilnforming glass such as Bullseye or Uroboros, you may find devit at terminations by the second or third flat fuse.

You can avoid it to some extent by making sure all scrap is well-cleaned. Still, with the addition of little particles from frit and powder in random spots, there’s simply a LOT of surface area here. Devit is tough to avoid.

I build mine from leftover corners of circle-cutting–those long, thin pieces with the internal curve. I stack them up, corners at the center, so that they resemble stars, with long-thin arms radiating out to different lengths from the center.

I build directly in the kiln, to avoid having to transport a bunch of loose little glass pieces on a flat kilnshelf.

As I stack, I’ll toss streaks of color here and there with frit powder, following the shapes of the clear glass. Sometimes I’ll paint a little CMC or gum arabic on the edges of some pieces and dip those directly in fine or powdered frit, to make a very thin colored line.

When fired, those lines will begin to mimic some of the features of aeolids, my favorite nudibranchs.

The aeolids can be nearly perfectly clear, with only white, shimmering outlines. Or they can be gaudy as all get-out, with big iridescent slashes of color in their cerata (those tentacle-like protrusions on their backs).

In fact, they’re where I got the idea to make these as nudibranchs. I base my designs in clear glass to get them as close as possible to their namesakes. I’ll add frit balls or scrap to the very tips of my stars, often in colors or dichroic glass, to further enhance the balling-up effect on the tips.

Assembly and fuse

Nudibranchs are what I call a hot-fuse project, i.e., the opposite of my normal low-and-slow firing schedule to keep the glass in place and control it gently into the mold. Usually, I want only just enough heat to achieve my adhere/flatten/whatever goal, then I get the heck out.

Disclaimer: I’m not claiming to have invented the technique, nor do I own it, or want to. It’s an obvious discovery that I’m sure dozens of people have “invented” at some time or other. More than likely, the Romans or Phoenicians or Mughals or Chinese were doing it 2,000 years ago anyway.

I DO copyright my words, and (strongly) request that you ask permission and give me credit  for them, before reproducing). Thanks.

This is almost the opposite: I want to shock the glass, potentially shatter it further into interesting patterns (which doesn’t happen all that often), and toss in so much heat that it starts to flow. I’ll hold at processing temperatures at least twice as long as a normal schedule.

I want the blank to lie as flat and even as possible without actually separating, and to smooth out every sharp angle as much as possible.

In the end, the glass will “dogbone,” i.e., start drawing in on itself to get as close to an even 6mm throughout as possible. Those thin tips at the edges will contract and pull in. The thicker centers will coalesce and push out, into the tips.

The end result will be soft, sinuous shapes that, when slumped, can look more blown than fused.

I generally give these two full-fuses on the first side. After the first fuse, I review the result, figure out where needs more color or arms or whatever, and apply. I want to make sure everything on the surface is perfect before the flip-over.

For the third firing, I’ll flip the glass over so that the shelf-side is now facing up. I’ll trim off any tiny protruberences that might encourage devit, prune stuff that doesn’t look quite right, add more color/sparkle, and then fire particles didn’t fully fuse on the underside.

A kilnshelf acts as an insulator; without a flip-over, you’ll have one side of the glass feeling grainy/sandpapery. This time, though, you must be careful NOT to trap frit between glass and shelf. I usually stick with powder, and larger chunks that I can precisely control.

After the third firing, I’ve got a smooth blank that sorta looks like an octopus on LSD.

Slumping the blank

If the piece is small enough, like the nudibranch at the top, I’ll simply mound up used fiber paper and set the glass on top. When it softens, I’ll manipulate it into a near-final shape in the kiln when it hits process temps.

I can also do what I did originally, i.e., put the blank on a drop-ring that’s a wee bit too big. When the glass begins softening, those thin arms will soften first, and the heavy center will drive the glass down, out of the mold.

It can either land on the kiln shelf, or (safer) wind up in a bowl loosely filled with sand, with irregular used ceramic fiber covering the edges. (You NEVER want a bare, hard edge from firebrick or steel to contact the glass; it shows up as an artificial crease in the result.)

Your schedule should be slow enough on the upramp to bend the arms without going too fast. Too fast, and the weight will drive down and crack the glass. You’ll need to do some testing to figure out how your kiln responds on this.

Once it lands in the bowl, I’ll check the slump, and do whatever manipulation is required to pull those octopus arms up slightly past vertical (i.e., perpendicular to the floor plus about 3 degrees). I don’t want them near any hard edge until they’re completely incapable of movement.

Then I close the lid and hit the cool-down button. There’s usually enough residual heat to keep softening the glass, so gravity pretty much removes the artificial shaped look of my kiln manipulations. I go fast down to the anneal temperature, and I’m done.

(Note: If you use a fiber kiln, heat recovery might go quicker, so again: Test)

Three nudibranch layers, roughly 16 inches tall, 10 inches wide.

Big, layered pieces

The drop-manipulation process I use for small pieces doesn’t work well for big pieces like the green one shown, which is about 16 inches tall.

Those I slump over a stainless steel form, well-padded with refractory fiber in random mounds and bumps.

Finding stainless steel forms

The best I’ve found, so far, are KitchenAid K45 stainless steel mixing bowls. They fit the smaller KitchenAid mixer (I know this, because I sacrificed my mixing bowl to the kiln gods), and have a glorious upside-down tulip shape that’s one of the prettiest slumpover forms I’ve seen for vessels.

They’re pricey (although not as expensive as a good stainless former), but you can frequently find them for less than $15 in thrift shops. They’re usually on eBay for around $20.

The old ones, without the handle, are best. If you do find one with a handle you’ll want to saw it off with a hacksaw and file it down.

Pack the screw threads at the bottom of the bowl (which becomes the top of the slumping form) with a little used fiber paper. Then drape the bowl with fiber paper or blanket. In this case used actually works better, because you can shape it.

Just be VERY careful to damp used fiber down first–mist with a sprayer, for example–and wear a respirator whenever it’s out in the open and not sealed in an airtight bag. Silicosis is no joke.

I try to keep the shape as organic as possible, so I’ll prop my steel form up on firebrick so that I don’t get a hard delineation where the form meets the kiln shelf (which gives you a very artificial-looking right angle), and carry the fiber blanket all the way down to the shelf like a skirt. Don’t worry about symmetry–you want the shape as irregular and organic as possible.

Slumping schedules

Unlike most of my very controlled slumping schedules, this one’s relatively fast and furious. The glass is reasonably symmetrical and mostly in a star shape, a fairly stable form. It has been fused flat and allowed to reach a semi-uniform “glass wants to be 6mm” thickness, with absolutely no right angles contributing new stress points.

You’re also slumping over a form covered with fluffy refractory fiber, with nowhere to trap the glass. The chance of “hotspots,” i.e., temperature differentials between the glass hanging out in space and the insulated is minimal.

If you’re worried that very long arms will crack during the slump,you can prop up the very tips of your blank with kiln furniture, again with some fiber underneath. When the glass starts to slump, past the worst thermal shock danger point, it will fall off the kiln furniture naturally.


Once you’ve started making these things, it’s fun to layer them. This one is roughly 2 feet tall.

In other words, it’s about as safe as it’s ever going to get. As long as you just didn’t go nuts with the annealing stage, you can use a standard normal-to-fast slumping schedule. If there’s some danger of devit (tossing random scraps and frit together increases the chance that something wasn’t properly cleaned, and there’s a lot of surface area open to devit), a faster schedule can potentially bypass the zone.

Update, 2017: Got a question about firing “coral” designs, which are a bit like this, but with thinner, longer branches that retain some angular shape. As a general rule of thumb, the sharper and more distinct the connections between individual pieces, the more careful you must be with your schedule. (IOW, go lower and slower on your slump firing)

Second rule of thumb: The more variation in thickness within the piece, the more careful the schedule. Nudibranch designs are overfired a bit, to ensure the thinnish areas draw up/round over completely, and that makes the piece about as flat and even as possible. So…adjust schedule accordingly.

Whichever method you choose, this is a process that obviously requires you to watch what’s going on in your kiln.

If you’ve got a nice big kiln window, watch the progress from the comfort of your cool studio. You’ll slap the kiln controller into cooldown and open the lid to manipulate when the slump is maybe 5 degrees away from perfect.

The residual heat will continue to carry the glass down the mold for awhile, but every kiln is different. With practice you’ll learn to see exactly when to go in.

If you’ve no window, it gets a bit trickier; opening the kiln cools things down dramatically, and the reheat will make it a little more difficult to time the slump.

I usually let the glass hit its process temp and wait five minutes, then check. Then I recheck every 2-3 minutes until it’s where I want it for a fiber kiln. A brick kiln might take longer.

Caution with slump-overs: You must be careful to let the fiber blanket form a skirt all the way around the steel form, and not overslump the glass. If you don’t, I can practically guarantee that at least one of the nudibranch arms will curve back under the form and trap the mold.

If you DO trap the mold, you can repeat the slump cycle, open the kiln at the hottest part of the firing and pull out the trapping glass arms, then prop them back with kiln furniture. It’s a royal pain to do that, though, and increases the potential for devit.

I use a similar method to make bases for these things, only this time around I let the glass touch the shelf to make stable feet. Then I clean the pieces up, play around with stacking them in different directions, grind the bottoms slightly to make a flat surface for gluing, and “cold-fuse” them together.

They’re not great art, they’re not functional in the slightest, but they’re a fun test project.