I had a goal: To create classic English marmalade. I had everything lined up: I bought a special jam-making pan; I researched recipes in cookbooks and on the internet.
Bitter oranges, I told my father, Procurer of the Finest Fruit: Seville.
A case of the finest Satsuma mandarins arrived.
Satsumas are not the foundation for classic English marmalade, but they are very tasty, and since you can make marmalade out of pretty much any citrus, I did. I used Alton Brown’s recipe, which came up on Google when I searched “Satsuma Marmalade” but oddly includes no mention of Satsumas on the page itself.
The recipe was easy enough to make, and included unnervingly specific temperature details that I don’t usually see on jam recipes, and since I am nothing if not a rule-follower when it comes to recipes, I got out my candy thermometer and did what Alton told me to.
The marmalade not only declined to set at the temperature specified, it steadfastly continued to refuse for some time, even when I reluctantly allowed the temperature to climb higher. I tested it repeatedly on a chilled plate, nervous that I might ruin the recipe by not following it precisely, even though I’ve never used a thermometer in the past when making jam – only the chilled plate test.
When the marmalade seemed to have sort of set – meaning it didn’t run in a thin stream off the chilled plate when tested – I put it into sterilized jam jars.
The next morning, I tipped one of the jars on the counter, to see if it had set. The marmalade sloshed around soupily.
So did the chilled marmalade in my refrigerator.
So much for marmalade; so much for science.
I left the jars – nine of them – on the counter of my kitchen that I put things I don’t really feel like dealing with, and debated whether I should just dump the marmalade and use the jars for something else, or give the marmalade to friends who would appreciate a tasty, if slightly runny, preserve.
After I made the Satsuma Marmalade, I did some research online and found a source for Seville oranges. The window of opportunity to obtain Sevilles is relatively short – basically, you can get them for a few short months in the winter, and that’s it. I ordered a box from The Florida Orange Shop. I debated ordering multiple boxes – how many Seville oranges do you need to make marmalade? I didn’t know, but I do now: not that many. The box contained enough for two full batches of marmalade – about 18 eight-ounce jars – which turned out to be just the right amount, because what seemed like a simple exercise in jam-making turned into a series of somewhat humbling learning moments.
The Orange Shop ships once a week, on Fridays, and although I thought I had placed my order in time to receive the oranges sometime the following week, I didn’t, which meant my oranges were delayed a week, which, in turn, meant the oranges arrived just as we were getting ready to leave for a week’s vacation.
Fortunately, I tend to hyper-organize right before I leave on a trip, so that I usually find myself with nothing in particular to do the night before we actually fly somewhere.
Making marmalade struck me as a perfectly reasonable way to spend such an evening; making Seville orange marmalade would be a perfectly reasonable choice if I planned my jam-making the way I plan trips. There is, unfortunately, very little I plan in such detail as a trip, certainly not jam-making, something at which I fancy myself fairly adept.
I used the recipe from the generally reliable Wednesday Chef, which involves removing the orange peels, slicing them fine, squeezing the juice from the flesh, then de-seeding and chopping the flesh, adding it all to the sliced peels with water, and letting it soak for 24 hours. Then a long simmer, and finally, sugar and jam. I was a bit perplexed by some of the directions: It seemed odd to soak the oranges in water that is then used to cook the oranges – even if the soak removes some bitter elements, wouldn’t the water retain them? Of course, I had not allowed myself time for a 24-hour soak, but since it seemed a bit unnecessary in any case, I thought it safe to skip this part of the recipe.
Even taking this shortcut, I found myself pressed for time. Slicing the peels takes a certain amount of time, but de-seeding the fruit is a real exercise in patience. You cannot comprehend the true greatness of a seedless orange until you have attempted to de-seed a Seville, whose seeds are as plentiful as the sand on the beach to which I was headed, and also as difficult to completely remove.
I cooked the peel for the exact time specified in her recipe – 45 minutes. At that point, the peel was mostly soft and mostly translucent, so I added the sugar, cooked it all until it reached the set point – as measured by a chilled plate – then poured it all into sterilized jars and sealed them. It didn’t taste like much; the marmalade had a very intense bitterness that made me think that perhaps this was one of those hip foodie things like kale that everyone pretends to like, even though nobody actually does.
I left the jars to cool, put the remaining half box of oranges in the fridge, and boarded a plane to Mexico.
When I returned a week later, the oranges in the refrigerator were still good, and since bitter oranges aren’t good for eating, I decided to experiment with one final batch of marmalade. This time, I went to the source of Wednesday Chef’s recipe, which is a recipe from Nigel Slater, and although the two recipes are somewhat similar, there were some important differences.
Although Slater also starts by removing the peel from the oranges and keeping the fruit whole, he mercifully makes no effort to de-seed the orange flesh after squeezing the juice in. Instead, the orange seeds and pulp are placed into a muslin bag, cooked with the peel, then removed and discarded. This was a much easier process, for which I unimaginably grateful. I juiced the oranges over a strainer, to catch all the seeds, then bundled up all the orange and seeds in pieces of cheesecloth, which worked quite nicely.
Slater also recommends a 24 hour soak; this time, I complied, and since I had plenty of time, I also didn’t rush the cooking. Instead, I simmered the peel until it was, as instructed, completely soft, and there was a point at which I bit into a peel and there was a marked, unmistakeable change in the degree of bitterness. It took about an hour and a half. At this point, I added the sugar, and it took another half hour or so until the marmalade jelled. The marmalade was more translucent than the first batch, but more to the point, it tasted like classic English marmalade – mostly sweet but with a nice bite to it.
Where I deviated from Slater’s recipe was in the ingredients: I didn’t have a lemon, so I omitted it. He calls for 12 oranges, but even if I’d had that number, my oranges ranged quite a bit in size, so I had no idea how much orange was really being called for. Since I had exactly half of the oranges remaining, I used the volumes called for in the Wednesday Chef recipe.
A friend of mine stopped by while the second batch of Sevilles was simmering, and so we chatted for a bit about jams, and I asked for her opinion on salvaging the satsuma marmalade, holding a jar upside down to show her the problem.
She said, it looks set to me.
We turned it upside down again: still set.
I gave her a jar of mysteriously set marmalade, followed by two more jars, one from each batch of Seville marmalade, and asked her to taste-test. After her initial feedback, and considering the fact that I now had more than thirty jars of marmalade on my counter, I posted a request for taste-testers on Facebook, and sent jars to foodie friends near and far. I received detailed feedback from three of them: Lori, a neighbor and fellow jam-maker; Bill, a high school friend and foodie; and Dorina, a former investment bank colleague turned professional pastry chef.
The results, inevitably, were not what I expected.
Lori adored the Satsuma marmalade, describing the taste of it as being like walking through a citrus grove on a sunny day. Dorina and Bill felt otherwise, finding it too loose and too sweet. Dorina made the helpful suggestion that adding a sharper citrus would brighten up the flavors and cut through some of the sweetness, a suggestion I thought would help a lot.
Similarly, Lori didn’t care for the first batch of Seville marmalade at all. The initial bitterness is nice, she said, but there’s a bitter aftertaste that is very unpleasant. Dorina, on the other hand, liked its flavor the best, but found the texture “too hard” – too much peel, not enough jelly. She also noted (correctly, based on Lori’s comments) that the general public might find it a bit too intense. This was Dorina’s favorite in terms of taste. Bill liked this batch too, and praised the level of bitterness, calling it an adult take on marmalade.
Lori liked the second batch of Seville, which was my preferred batch, and pronounced it Just Right – just the right amount of sweetness and bitterness, and a much more pleasant texture, presumably due to the softer peel and the removal of the orange flesh. Dorina praised the texture as well, but found it overly sweet, a problem that was likely caused by omitting the lemon called for in Slater’s recipe. Bill pronounced this one also very good overall, but commented on the fact that the peels had floated to the top, making it a bit fiddly to scoop out a good jam to peel ratio. Dorina remarked on this issue in all the marmalades, but it was most pronounced in the second batch of Seville.
Dorina also amusingly assumed I was asking for input because I was thinking of marketing small-batch marmalade, which isn’t a bad idea if I can perfect my recipes.
I learned quite a bit, although much of it was, not surprisingly, things I already know or that should have been obvious. First, take care when tinkering with recipes: a recipe that calls for a long, slow simmer probably does so for good reason, so that’s not a good place to take a shortcut. Similarly, omitting the lemon from the second batch of Seville was not a good idea, as it would have lent exactly the tartness one of the tasters said was missing.
Second, a candy thermometer and a kitchen timer are useful tools, but they aren’t a substitute for your senses. I could see perfectly well the Satsuma wasn’t quite set, regardless of what the thermometer said; similarly, I knew the peel in the first Seville attempt wasn’t quite ready yet, regardless of the amount of time the recipe should officially have taken.
Third, when something tastes off, it probably isn’t me just being self-critical, it is off – or at least, it is to me. My friends’ responses reminded me how very personal one’s food preferences are. Maybe there’s a place for kale after all, even if it isn’t on my own dinner plate.
The most useful reminder I got was this: Many food mistakes can be salvaged with a little creativity. Lori didn’t care for the first batch of Seville on her morning toast, but emailed me a few days later to say that with the addition of a bit of mustard, vinegar, and oil, it makes a delightful dressing for spinach salad. Bill made a similar observation about the not-quite-set Satsuma marmalade. Both of them got me thinking about other ways I could use the marmalade, or even the possibility of gifting something I considered a failure, with a recipe card included.