Quince science

Cooking quinces is culinary magic. You start out with fruits that are yellow, hard and astringent. Three or four (or five) hours of gentle poaching later, they are fragrant, grainily soft, sweet and – most impressively – deep pink. Unless, that is, they have been cooked by me in the last year or so. I used to have the knack: see these beautiful specimens I cooked when I lived in the UK (and I made many similar-looking in Ireland, too, and I think I used to in Australia). But the last few times I’ve cooked them, they have refused to change colour, and after three or four hours of poaching I’ve been left with quinces the colour of bandaids. Not very appetizing! The liquid always goes at least a little pink, but not the jewel-pink it does when the fruit themselves also change colour.

Each time I’ve been confronted with beige quinces, I’ve scoured the internet looking for information on what could be going wrong. I haven’t found anything conclusive, but did see brief suggestions on a couple of cooking forums that absence of colour change could be due to one of two faults with the poaching liquid: (a) low sugar concentration, or (b) not adding lemon juice. Now, although I did make one beige batch with a lighter sugar syrup than usual, I’ve also had non-colour changing quinces when using the same sugar concentration that I’ve always used. So I thought it probably wasn’t the sugar. The lemon suggestion struck a chord, though.

When I was first cooking quinces, and for many years thereafter, I was a good, dutiful woman and followed Stephanie Alexander’s poached quince recipe to the dot. This included dropping the peeled and cored quinces into acidulated water (water with a good squeeze of lemon juice), before adding them to a light sugar syrup (1 part sugar to 2 parts water). A couple of years ago, though, I got uppity and started to think there was no point in the acidulated water, the stated purpose of which was to stop the quinces discolouring. What did I care if they discoloured a little? Once they turned dark pink you wouldn’t be able to see any discolouration, anyway. So I stopped doing that step, and began just tossing the quince quarters directly into the sugar syrup once I’d finished trimming them. Although I can’t pin it down precisely, in retrospect I do believe that it was around the same time I started doing this that my quinces stopped turning colour so successfully.

Last time I was faced with beige quinces, I found the lemon suggestion online, and, three hours into the cooking time, squeezed the juice of half a lemon into the poaching liquid. When I looked again, an hour later, the quinces were a lovely, reasonably deep pink. Ah ha!! But, being a scientist and generally sceptical, I wondered whether perhaps it was just the extra hour of cooking that had wrought the change, and the lemon had had nothing to do with it. So I decided to do….. Quince Science.

First, I obtained some experimental subjects.

Next, I thought about my experimental design. I wanted to test two hypotheses: (a) that low sugar concentration in the poaching liquid reduces quince colour change, and (b) that lack of lemon juice in the poaching liquid reduces quince colour change. I was also curious about whether sugar and lemon interacted somehow – for example, sugar and lemon might each boost colour a little bit, but both of them together would be required for full colour change. So I decided to set up four treatments:
– low sugar/no lemon
– low sugar/with lemon
– high sugar/no lemon
– high sugar/with lemon
This would let me separate out the effects of each of the variables, while also checking for an interaction.

I also thought about other sources of potential variation that might affect my results. The main ones I could think of were (a) random variation between quinces and (b) consistent variation between quince varieties. I could deal with the first fairly easily. Instead of cooking one quince in high sugar/no lemon, a second quince in high sugar/with lemon, etc, I decided to cut each quince into quarters, and cook each quarter in a different treatment. That way if one of my quinces was a weird one that would never change colour under any circumstances, I wouldn’t be misled into thinking that it was the treatment that was at fault for causing no colour change. The second source of variation was a bit more tricky. I know that there are about a dozen different quince varieties grown in Australia alone, and more overseas. But they’re never labelled, and I’m not sure which ones I’ve used for any particular batch of quinces. I know the last batch I cooked had waxy, fuzz-free skin and were and pear-shaped, while the ones I bought at the farmers’ market for this experiment were fuzzy, and two were pear-shaped and one more spherical. I wasn’t actually prepared to spend days driving all over Brisbane to source a wider variety of quince types for this experiment, so decided to just press ahead with what I had.

So I set up my laboratory. I usually poach quinces in the Le Creuset in the oven, but have occasionally done it on the stovetop. I don’t own four Le Creusets (or equivalents), unfortunately, so had to do this on the stove.

Here’s another potential confounding factor: different pot volumes and burner strengths. I tried to deal with this by making sure that all the treatments had ample liquid, and tweaking the burners throughout the experiment to make sure that all the pots were simmering very, very gently. And after all, I am not going to be publishing this in Nature.

The setup:
1. Rear left: high sugar/no lemon: A 1:2 sugar syrup, 2 cups sugar: 4 cups water.
2. Front left: low sugar/no lemon: A 1:4 sugar syrup, 2 cups sugar: 8 cups water.
3. Front right: high sugar/lemon: A 1:2 sugar syrup, 3 cups sugar: 6 cups water, plus the juice of half a lemon.
4. Rear right: low sugar/lemon: A 1:4 sugar syrup, 2 cups sugar: 8 cups water, plus the juice of half a lemon.

In go the quinces, to simmer gently for 3 hours.

After three hours, the differences were clear:

Fortunately this is science rather than art, so ugly flash photography is A-OK.

The bowls are in the same arrangement as the saucepans were on the stove. The two treatments with lemon are on the right, and are clearly a lot pinker than the treatments without lemon. There’s no obvious difference between the high and low sugar treatments with lemon. There’s a colour difference between the two non-lemon treatments, but I think that’s because the low-sugar treatment was on the front, wok burner, and so simmered a little harder than the back saucepan despite my trying to even them out.

Conclusion: Lemon is required for quinces to turn properly pink! Sugar concentration has no noticeable effect at the levels tested here.

I then went on to try a couple of further experiments.

First, I wanted to see whether I could get the lemon quinces to turn even darker pink, and whether this could be achieved simply by longer cooking, or if it required (or could be facilitated by) more lemon juice. I combined the two lemon treatments then split them back into two saucepans, to even out the different sugar levels, since I no longer believed this to be an interesting variable. I then added the juice of another half a lemon to one saucepan, left the other as it was, and simmered them both for a further hour.

There was no visible difference between the quinces with extra lemon and the ones without. Both sets, however, were noticeably darker and glossier after the extra hour of cooking. The syrup was also much thicker and darker. (I think the apparent glossiness of the fruits is just due to being coated in the beautifully cooked-down syrup.) You can’t really see an enormous amount of difference in the photos above, but trust me, there was a very clear change! I should have kept back some of the 3-hour quinces to use as a comparison in the photo, but I was too keen and rushed on without thinking my experimental design through, tch tch.

Finally, I wanted to confirm that a pot of beige-y quinces could be rescued at the last minute by the addition of lemon, for the benefit of anyone reading this after searching for answers to their quince troubles. So I combined then split the non-lemon quinces into two pans, added the juice of half a lemon to one of them, and simmered for a further hour. After that hour, the quinces cooked without lemon were pale pink (so at least no longer beige). But the quinces with the late-added lemon were utterly indistinguishable from the ones that had been cooked with lemon from the start. Both sets were equally dark pink and glossy. There was a difference, however, in flavour. The quinces that had been cooked with lemon from the start seemed to have a stronger flavour, that was more consistent all the way through. The ones that had had lemon added at the end were entirely fine, but less flavourful in the middle of each piece of fruit. I think the lemon may heighten flavour, and cooking with lemon for 4 hours rather than just 1 gives it a chance to penetrate all the way in.

Final conclusions: Long-cooking (at least 4 hours) contributes strongly to quince pinkness, as we all knew. For maximum colour change, add lemon juice during cooking. For the best flavour and earliest colour change, add the lemon at the start of cooking, rather than towards the end.

Future research directions: While waiting for these quinces to poach, I browsed the internet for more quince information, and came across this incredibly useful site from NSW DPI: Quince Growing. Table 1 lists 15 varieties of quinces grown in NSW, and includes information on what colour they are when cooked. This ranges from ‘yellowish’ to ‘deep pink’. I wonder whether some varieties really can’t be turned pink with cooking, even with lemon, or whether other varieties turn pink without lemon. Clearly an area that needs further research! I will just have to scour farmers’ markets for further specimens from a wider variety of cultivars. Maybe I can get ARC funding for it.