In the last article, we explored how a gluten-free cake recipe might get a bit more leavening (Read FC’s coverage on gluten’s role in baking to learn more). Eventually, GFBaker (or “Rebecca”), who asked the original question, got in touch with me and we had a few days of brainstorming with the full recipe. Unfortunately, we didn’t find the magic changes to make to the gluten-free cake to give it enough rise, so she made some changes to the cake assembly process to make it work well enough for her purposes.

What I did discover was that the cake was assembled with the creaming method, which is basically what I had presumed in the previous post. You sift and combine the dry ingredients. You beat (or cream) the sugar into the butter, and you add the liquids and dry ingredients in a few alternating batches.

She managed to make a couple of variations of the cake, but the problem was the amount of time that we had to make the variations. She had a chance to do maybe three different trial cakes, the first being to try the suggestion from the original article which was to whip the egg whites and fold them in. This had a bit of an effect, but not enough of one.

The next change was to try to stabilize the egg foam in order to give it more support, and also to reduce the amount of chemical leavening, so she added some cream of tartar and some sugar to the egg foam, but that didn’t really provide enough rise, and dried out the cake a bit.

Now, it may seem strange to reduce the amount of leavening when trying to make the cake rise more, but this recipe called for a large amount of baking soda, because the original recipe that the gluten-free version was based on called for that large amount of baking soda, and I was concerned that the recipe was overleavened. In a traditional, wheat-flour cake, the batter can hold in bubbles from roughly 1/4 teaspoon of baking soda for every cup of flour, and this recipe called for enough baking soda for 12 cups of flour. So the theory was that most of the bubbles were escaping the batter when baking, and using an egg foam for structure and leavening might help that.

Those familiar with test kitchens, science, and computer debugging will notice one of the big problems with our methodology, which was that we changed two things at the same time. In the last version of the cake, we reduced the amount of baking soda at the same time we increased the stability of the egg foam. There were even other changes that happened at the same time with respect to cake pan sizes and similar that may have had an impact. And that causes trouble.

When you’re not entirely sure what’s happening with a recipe, what you want to do is to change only one thing at a time. The better you do at ensuring that everything is the same except for the one thing you change, the better chance you have at knowing what your change did. If you change two or more things at a time, you can guess at which change did what, but you’ll never be sure.

When you’re studying something scientifically, this method of only changing one thing is vital to a certain class of experiments, and the better it is done, the more we trust the explanation of what’s happening. In the kitchen, though, we are generally less concerned about why something happened as we are with the final results. So can we get away with multiple changes at a time?

As a general rule, yes. We have thousands of years of cooking experience, depending on the type of food, and hundreds of years with most of the more modern techniques. However, when we start replacing something that is integral to the structure of the food, such as gluten, then there’s a lot of things that we could take for granted that just don’t work any more.  Ideally, if you would find a replacement that works exactly like gluten in all cases and doesn’t make people sick, but in reality, we can only approximate the effects of the gluten in cooking.

So, with more time and energy, we would have tried the following variations:

1. Simply reduce the amount of leavening and see if the cake was taller, shorter, or the same height at the end of cooking.
2. Change the application of the leavening, so that we use the soda only to reduce the acid in the cake with baking soda (which is probably why the original recipe called for so much), let that settle so that any excess bubbles are released, then add a slow-acting baking powder to the cake to get additional rise during the baking cycle. See if that is taller, shorter, or the same.
3. Take the best variation from 1 and 2 and try adding in the egg foam.
4. Take the best variation from 1 and 2 and try adding in a stabilized egg foam.