Skip to main content

Last week, Sarah was away visiting her sister and some friends on the east coast, leaving me with a bit of “Bachelor Time”. Excitingly, “Bachelor Time” very closely resembles “Relationship Time”, except with more meat. I ate hot dogs for most of the week (before I completely lose your respect, I’ll say I was working heavy overtime at work trying to finish some deadlines that were looming). I also got to make some extra-special messes in the kitchen THAT I DID NOT IMMEDIATELY CLEAN!

Sarah being away mostly meant I had to pick something to make from the cookbook that she wasn’t likely to be sad about missing. She’s not the biggest fan of raisins, and the title of this one hearkened images of some sort of raisin loaf in both of our heads, so I pressed forward with it over a few nights while simultaneously texting a drunk Sarah in DC. Not wanting to be left completely out in the cold on that front, I cracked open a few beers myself, and before I knew it this dish had earned the dubious honor of being completed largely under the influence. Milestone?

One of the ingredients needed for this dish is “hazelnut praline”. Some intertubez searching led me to discover this is pretty much just hazelnuts tossed in caramelized sugar. I couldn’t find any prepackaged hazelnut praline at Berkeley Bowl, so I just decided to make some myself. I started with a bag full of raw hazelnuts (or, as some people call them, “filberts”). Problem #1 with these is getting the skins off. Don’t even try to do this while they’re raw, you’ll just end up ever so sad. I…(sigh)…spread my nuts on a sheet tray and toasted them in an oven at 350F for about 20 minutes or so; just until they smelled fragrant. I am so sorry for that sentence.

After letting the toasted hazelnuts cool completely, I could grab a fistful of them and shake them around in my hands to cause the skins to flake off. This is only slightly tedious, and I was left with a nice fragrant bowl of skinless filberts to work with.

I measured out 70g of nuts and melted an equal part sugar in a small saucepan, then poured in the nuts and let the mixture cool. I pulsed the result in a food processor into a fine crumby consistency.

The praline is used in a hazelnut cake, which is made with some eggs, whole wheat and white flours, and brown sugar. This is poured into a sheet tray and baked to a spongy consistency. Then it’s cooled and ripped voraciously into little bits (no seriously).

After laboring on this back-breaking cake, I decided to get my refreshment on for the next step, which involved making a raisin gel. What better way to do this than with a beer made from raisins?! I’m always a sucker for any New Zealand beer I can get my hands on. One of my favorite beers of all time is an IPA by Epic Brewing; it’s floral and grapefruity and awesome.

BUT NOT TODAY! No, today I’d forgo the Epic for another NZ beer I found at Berkeley Bowl, by one of my other favorite breweries down there. Kiwi for raisins is “sultana” (when I first moved there, I was really thrown by the familiarly-packaged “Sultana Bran” I found in the cereal aisle), and this beer seemed a poetic choice while working with this dish.

Of course, I also needed some raisins. The Bowl had packaged bags of just the right amount of something called “Jumbo Prima Raisins”; they’re way bigger than the little Sun-Dried-style raisins I’m used to seeing. They also taste way more genuine and punchy.


I let these steep in hot water overnight to reconstitute them a bit, then blended them with a few magic white powders and poured the resulting liquid into a sheet tray I’d lubricated with some cooking oil. Once they’d set into a gel, I cut them into small flat squares and set them aside.


The ‘magic white powders’ in this case were two types of Gellan gum. Gellan is available in many ‘flavors’; at one end of the spectrum, “Low Acyl” Gellan forms thermoirreversible (unmeltable) gels that are hard and brittle in nature (in the world of gels, this means something that tears easily rather than being very elastic). At the other end of the spectrum, “High Acyl” Gellan forms thermoreversible gel that are soft and elastic. There are lots of blends of these two extremes that try to mix-and-match the best of their characteristics; Kelcogel JJ Gellan is such an example. An advantage of Gellan is that it doesn’t need to reach boiling temperatures to dissolve into a mixture, and in the case of High Acyl Gellan, the melting point is quite a bit higher than that of Agar. The gel I ended up with was soft and slightly toothy, sort of like a very delicate gummi bear.

Despite being in a bit of a raisiny-alcoholic fog (the raisin beer was pretty potent), I pressed on with making some carrot puree. I first juiced several carrots, and sliced several more into thin discs on a mandoline. I packed the discs in a vacuum bag with some honey and the carrot juice, and cooked the lot in a warm water bath until they were very tender. The cooked carrot slices were then blended with some Xanthan gum to thicken the resulting puree into a thick puddingy consistency and put in a little squeeze bottle.


Next I made what the book calls “Hazelnut Nougatine”. When I picture nougat, I think of stuff like Three Musketeers bars; Sarah recently made her own nougat for use in a Candy Bar Pie from the Momofuku Milk cookbook, and it involved cooking sugar to a caramel and folding it into fluffed egg whites (doing this requires a dazzling bit of kitchen timing acrobatics). What I made is what Wikipedia calls “brown nougat (referred to as “mandorlato” in Italy and nougatine in French)” and is basically just more hazelnut mixed with cooked sugar. The result is very crunchy, almost identical to my hazelnut praline, only with a slightly different sugar-to-nut ratio. This was again chopped up into a coarser crumby texture and set aside.

Finally, the big neat surprise: the cool yellow orb, which is spherified melted butter. I’ve done this trick several times; enough to feel like I understand it pretty well and am slightly less-dazzled buy it now. But right when I get comfy, BAM. Something neat and new to learn.

Spherifying butter presents a fundamental problem: butter is fat, fat is oily, oil and water don’t mix. So, how does one get the magic white powder vital to spherification (calcium lactate) dissolved into the butter? A while back I made Beurre Monte for a different dish, which is an amazing sort of mayonaise made just by whipping butter gradually into a tiny amount of water, effectively emulsifying the water into the butter. We use the same strategy here, but we have to do it extremely carefully I learned!

I started by picking a butter. If I’ve learned one thing from my baking experience so far, it’s that not all butters are created alike. “European-style” butters are cultured during the creation process, which results in generation of lactic acid. It gives the butters a nice slight sour note and intensifies the ‘buttery’ flavor. While cooking with cultured butter is a little hit-or-miss in terms of how much flavor you get from it, eating it spread on toast or in otherwise ‘purer’ form highlights the flavor differences nicely. For this, then, I went straight for one of the nicer ones I’m familiar with:

I brought a tiny amount of water mixed with a few grams of calcium lactate to a boil, then whisked in cubes of butter one at a time to form an emulsion. This took me several tries to get right, and it’s important to note (as I learned) that the calcium lactate–the magic ingredient that’s gonna make us up some real nice butter balls–does not dissolve in the butter. We want to basically keep the calcium lactate/water in suspension long enough for the butter to solidify, at which point we’re hoping for as uniform of a dispersal in the butter as possible. If the emulsion breaks even in the slightest, the oil in the butter rises to the top of the molds as the butter sets, which means there’s no water (and therefore no calcium lactate) on that upper surface, which means the sodium alginate bath has nothing to ‘latch onto’ to form a sphere membrane in that area. Tricky.

Once the butter has set, I carefully lowered the half-spheres into a (cold) bath of sodium alginate. I let the butter sit for 20 minutes to allow the calcium lactate and sodium alginate to form a membrane, then I flipped all the half-spheres and waited another 20 minutes. At this point, you can tell if you’ve messed up; you’ll see obvious holes in the membrane if too much oil has settled to one spot.

When I transferred the butter to a water bath to rinse them, I could see a nice membrane had formed over most of them.

Then, the magic: lower the half-spheres into a warm water bath held at 94F–warm enough to melt the butter. Surface tension causes the shape to change to that of a sphere, once the sphere is fully liquefied, I can pull them out and serve them. They look like tiny bright egg yolks; one might mistake them for quail yolks even.


Finally I was ready to plate. I bought a “Dunes” plate from Crucial Detail just before Christmas specifically because of how it works with this dish. There’s a neat sort of ‘lagoon-like’ area of the plate where the butter sphere can do. When a diner breaks the sphere, butter runs like a little river down the plate into a larger lagoon area, where it mixes with the rest of the components for easy swishing and dipping fun.

The sphere, some shards of the hazelnut cake, and some squares of raisin gel are heated under a broiler until the cake is slightly-toasted and the whole plate becomes aromatic. It’s then topped with a sprinkle of the crushed nougatine, dots of the carrot puree, and some ground cinnamon. The dish tastes more or less like (a really great version of) what I expected; the flavors are warming and autumnal, and the plate is literally warm and aromatic and yummy. I thought I would find the raisin/carrot-cake-vibe boring but it’s really very delicious. The carrot puree tastes sweet and honeylike, and I like the vibrant color it has to offer. The cake is molassesy and toasty and nutty, which is great in and of itself. But obviously drizzling everything through a river of warm melted butter only makes everything more delicious; you might as well throw a few slabs of bacon on the plate, as predictable as that is.

As I was shooting photos of this, I found the lower I got to the plate, the more it started resembling a weird martian landscape. I futzed with my lights a bit to try to allow enough room to close down my aperture and get a nice deep depth of field. I was so close this didn’t work very well, but shooting this dish and the last Bison one both gave me an itch to try to get ‘closer’ to the food, to shoot it as if I were a little guy hanging out in this giant world of food. I’d like to see if I can find ways to pursue that idea, it seems like it could be neat.

Join the discussion 4 Comments

  • Norma Vogel says:

    Hi Allen,

    I hope you are doing great! Very interesting stuff you did with the pedestal for the glass tube. By the way, I know that part of the project was to do it yourself, but if you ever come up with other serving pieces that you would like made, here in Nicargua, where I live, you have have ANY shape made for next to nothing. For example, I took a small road trip like an 30 min from my work to talk to these ppl who work with steel and aluminum and other material. I printed a picture of what you had made just out of curiosity, and they where willing to do as many pieces as I wanted for under 2 dollars a piece. I then proceeded to ask if they could do anything I wanted… they just said to bring the design. Anyhow, if you are ever interested in having some design of your made, I am more than happy to help you out and ship it to you. Since I travel to the states for work almost monthly, that would be no problem. Anyway..just thought I’d offer.

    Now I have a question for you… where did you get spray dried licorice powder?

    Take care

  • Norma vogel says:

    Hi Allen,

    Now I have a question for you… where did you get spray dried licorice powder?

  • Norma Vogel says:

    Thank you. Sorry bout that.