Texas caviar, for the patience-challenged

This is a simplification of my earlier Texas caviar recipe to eliminate cooking by using canned products. I've also attempted to streamline the instructions and improve their readability by trimming away details that should be self-evident. So with brevity in mind, let's get on with it.

RECIPE

Makes 6 servings

INGREDIENTS

For the base

Qty	Unit	Item
1	14–16 oz can	Black-eyed peas
1	14–16 oz can	Black beans
1	14–16 oz can	Yellow corn kernels (not cream style)
1	small	Red bell pepper (about 2/3 cup diced)
3	stalks	Celery (about 2/3 cup diced)
1	small	Onion (about 1/2 cup diced)

For the dressing

2	medium	Garlic cloves
1	medium	Jalapeno chile, fresh
1/4	tsp	Ground cumin
1/2	tsp	Table salt
1	tsp	Ground black pepper
1/4	cup	Apple cider vinegar
1/2	cup	Neutral-tasting vegetable oil

EQUIPMENT

• Can opener
  
• Colander
  
• Large (6 qt) and small (2 qt) mixing bowls
  
• Knife and cutting board
  
• Garlic press (optional)
  
• Gloves (recommended for handling jalapeno)
  
• Measuring cup (1- or 2-cup size)
  
• Measuring spoons
  
• Whisk
  
• Wooden spoon or spatula

INSTRUCTIONS

Steps 1–3 are for the base, and Steps 4–6 are for the dressing. You can treat these as two separate tasks and do them at different times if you'd like.

1. Empty the cans of black-eyed peas, black beans, and corn into a colander and rinse well. After rinsing, let the colander drain for a few minutes.
   
2. In the meantime, dice the bell pepper, celery, and onion into 1/4" squares and place into the large bowl.
   
3. Add the drained ingredients from the colander into the large bowl.
   
4. Mince the garlic and place it into the small bowl.
   
5. Dice the jalapeno into 1/8" squares and place it into the small bowl.
   
6. Add the remaining dressing ingredients (cumin, salt, pepper, vinegar, and oil) to the small bowl. Whisk until the vinegar and oil are blended.
   
7. Pour the dressing over the base ingredients in the large bowl, then stir to mix well.

TECHNIQUE

You know what's better than me blathering on and on about technique? Having an expert show you. I've included a few links above, but you know what's even better than text links? Videos. Check'em out.

Finally, although I already have a link to mincing garlic, ya gotta watch this video. Too bad he doesn't allow embedding, but it's really impressive how he rocks the knife (literally, starting at about 1:50). A master mincer, right there.

Read more inside...

Raw sweet corn and cashew chowder (sort of)

[Direct link to recipe]

First of all, let me just say that since I'm not a dedicated follower of the raw foods diet, I've never heard of Ani Phyo until seeing her mentioned in The Kitchn a few weeks ago. But if you had been paying attention (unlike me), you'd already know about her from all the coverage. She got my attention this time, though. Corn and cashew make an unusual pairing for ingredients—does she develop her recipes by alliteration? Anyway, the instructions can hardly be simpler: the steps basically boil (can I use that word here?) down to "toss stuff in a blender and go." Problem is, it's nowhere near the season for fresh corn. What to do?

Well, not being a raw foodist came in handy at this point to save me from having to violate my own dietary rules. The only corn I had at hand were frozen kernels, meaning that they were blanched before being packaged. In other words: not raw. Being the experimental sort, I also wanted to find out if roasting the cashews would make any difference, so I halved the recipe and made two batches.

As you can see from the comparison image below, the bowl made with plain (that is, not roasted) cashews is just a bit less intense in color than the one with roasted cashews. If I had used raw corn, then I'd imagine the color would've been even paler. Flavor-wise, the roasted cashews contributed a bit of extra depth and, well, nuttiness, to the second batch. As the recipe contained only a small amount of cashews relative to the corn, though, the differences were subtle and probably not discernible unless the two versions were compared side by side. It's probably not worth the extra effort to roast the cashews.

Another factor that made the taste-testing somewhat inconclusive was that I thought the first batch was too oily and salty and so used less of each for the second batch. While kosher salt and extra virgin olive oil make for a great combination, the quantity used in the original recipe overwhelmed the corn flavor. Now, this isn't a knock against Ani Phyo specifically, since I find that just in general many recipes specify too much oil and salt, and I've come to expect having to reduce their quantities as a standard modification.

As for texture, I did not find it as heavy as the guys over at The Bitten Word, although I did have similar pureeing issues. Commenters both there and on The Kitchn advised soaking the cashews overnight to ensure smooth pureeing. They are right. With un-soaked cashews, chunks up to half-piece in size remained after my blender ran for 30 seconds on its highest speed; this did not happen with soaked cashews. Also, in both batches the corn did not puree completely, and bits of the hull were dispersed throughout the soup. Extending blending time by 30 more seconds (for a total time of one minute) made no appreciable difference, and I think the puree needs to be passed through a strainer to achieve smoothness. Technically, the rough texture may not be an issue because this is a chowder instead of a bisque, although I have a feeling that this soup was really meant to be liquefied with one of those King Kong blenders favored by the raw foods community instead of the weaklings the rest of us own.

RECIPE

Makes 4 servings. Adapted from Ani Phyo's original, with less oil and salt.

INGREDIENTS

Note: This recipe uses 450 g (1 lb, or 3-1/4 cups) of yellow corn kernels in total, either fresh or frozen. If using fresh kernels, this is equivalent to 4 large ears. If you're a raw foodist, please keep in mind that frozen corn is blanched, not raw.

For the chowder

Quantity	Alternate Measures	Item
68 g	2-3/8 oz  (1/2 cup)	Cashews, preferably raw, plus water for soaking (soaking is optional)
312 g	11 oz  (2-1/4 cups)	Yellow corn kernels
473 g	16-3/4 oz  (2 cups)	Water
60 ml	4 tbsp	Extra virgin olive oil
1 small	1 small	Garlic clove
5 ml	1 tsp	Kosher salt

For garnish

138 g	5 oz  (1 cup)	Yellow corn kernels
To taste	To taste	Cilantro leaves
To taste	To taste	Freshly-ground black pepper

EQUIPMENT

• Weighing scale and/or measuring cups
• Measuring spoons
• Small bowl, for soaking cashews
• Knife and cutting board
• Blender
• Pepper mill
• Strainer (optional)

INSTRUCTIONS

1. If you don't own a high-powered blender, it's recommended that you place the cashews in a small bowl, cover them with water, and let them soak overnight.
2. If you're using fresh corn, take off their husks, remove the silk, then use a knife to cut the kernels off the cob. If you're using frozen corn kernels, let them thaw before putting them in the blender.
3. To prepare the cilantro for garnish, pull or slice the leaves off the stems, then chop roughly with a knife (video).
4. If you've soaked the cashews, take them out of the bowl and put them into the blender.
5. Peel the skin off the garlic clove, then place it and the remaining ingredients from under the chowder heading (corn, water, oil, and salt) into the blender.
6. Turn on the blender and puree this mixture until smooth. For an even smoother texture, you may want to pour the puree through a strainer.
7. Divide the puree into four bowls.
8. Divide each of the garnish ingredients into four portions.
9. Into each bowl, place one portion of the garnish corn and then sprinkle with cilantro leaves and ground peppers.

VARIATIONS

• Parsley, either the curly- or flat-leafed variety, can be used instead of cilantro.
• If you're not a strict raw foodist, this recipe will likely work well with grilled corn.

ARE THE CASHEWS REALLY RAW?

Aside from the frozen corn, the reason that I attached the "sort of" qualifier to my post title is because my cashews came from a regular supermarket. Why is this a problem? Well, while the general public classifies unroasted nuts as being raw, cashews from non-specialist suppliers are likely to have been heat processed before reaching market and are thus unacceptable or at least suspect to raw-foods purists.

So why are cashews heat processed? This is because freshly-harvested cashews are encased in an urushiol-containing double shell. Urushiol is the active irritant in poison ivy, so it's critical that the nuts are extracted from their shells without contamination. As this irritant is destroyed by high temperatures (PDF, p. 2), traditional extraction methods all use heat—either drum roasting, oil roasting, or steam roasting—to decontaminate the nuts as well as to separate them from their shells. For the raw foods market, special tools must be used for safe low-temperature extraction.

What are the risks that raw (or semi-raw) cashews are dangerous to eat, then? For that, I have no answer. I've had no problems from eating semi-raw supermarket cashews, but as shown by the current peanut problems or last year's Chinese milk scandal (you're being way too optimistic if you think something similar can't happen in the US), contamination of our food—whether from bacteria and viruses, artificial impurities, or a plant or animal's own natural toxins—is an always-present possibility.

Read more inside...

Creamy split pea and petite pea soup

Happy St. Patrick's Day, everyone. Even if you're not going to make anything specifically Irish, a green soup still seems appropriate, no? This is based on a recipe originally found in the December 2006 issue of Bon Appetit, which spikes an ordinary split pea soup with green peas to boost its color and dill to brighten its flavor. The result was delicious, but its consistency was a lot thinner than what I had expected from a split pea soup. So I set to work, and after some tinkering, this is my take on it.

RECIPE

Makes 4 servings. Modified from Bon Appetit, December 2006.

INGREDIENTS

I tested this recipe with both the basic and preferred/optional ingredients. The basic ingredients will work, but the preferred ones will give you much more flavor.

Quantity	Alternate Measures	Item
140 g	5 oz (1 cup)	Frozen petite peas (also sold as baby sweet peas)
200 g	7 oz (1 cup)	Green split peas
115 g	4 oz (about 1½ cup)	Leeks (from one large or two small leeks, white and light green parts only)
1 ea.	1 ea.	Bay leaf
30 g	1 oz (5 tbsp chopped)	Fresh dill leaves (from about 8–10 stalks), or 1 package baby dill
30 ml	2 tbsp	Vegetable oil, or extra virgin olive oil (preferred)
950 g	2 lb 1½ oz (4 cups)	Water, or vegetable stock (preferred)
8 g	1 piece, about 4"–6" long	Dried kombu (optional)
5 ml	1 tsp	Table salt (if using water; use less with stock)

EQUIPMENT

• Weighing scale and/or measuring cups
• Measuring spoons
• Various bowls for mise en place
• Knife and cutting board
• 4-quart cookpot
• Spatula, for sauteeing
• Blender, or immersion blender
• Ladle

INSTRUCTIONS

1. Measure out the frozen peas and set aside to thaw.
2. Pick over the dried split peas and check for pebbles or other debris, then rinse and drain the peas.
3. Cut off the dark green parts of the leek and wash thoroughly to remove any dirt. Once the leek is clean, slice it crosswise into ¼" rings.
4. Separate the midrib of the bay leaf by either folding it lengthwise or running the tip of a knife along both of its sides. Discard the midrib.
5. Pull the dill leaves off the stems and chop the leaves. Save a few fronds for garnish.
6. Heat oil in the cookpot over medium-high heat until it shimmers, then dd the leeks and bay leaf to the pot and saute until the leeks wilt, about 3 minutes.
7. Add the split peas and stir to coat, then add the water/stock and kombu.
8. Bring the liquid to a boil, then turn the heat down to between low and medium-low. Cover the pot with a lid and simmer for 45 minutes, or until the split peas are soft.
9. Turn off the heat. Add the thawed petite peas and chopped dill to the pot.
10. Puree the soup until smooth. An immersion blender is most convenient, but if you don't have one, carefully fill the pitcher of a blender only halfway full (or less), then blend in batches.
11. Add salt to taste.
12. Ladle the pureed soup into bowls and garnish with dill fronds.

NOTES

• Depending on the supermarket, fresh dill may be located either with the produce or with the herbs. Baby dill is usually sold as a packaged herb.
• I did not test this out, but you should be able to use mature green peas if you can't find petite peas.
• The bay leaf and kombu stay in the soup and get pureed.

Read more inside...

Raw almond butter

Update: Please also see my other almond butter posts:

• Homemade almond butter (using roasted almonds)
• Can you freeze almond butter?

Ack. This is going to look like I just can't let go, but the fine folks from Hangry Pants (love that name, BTW) asked about processing raw almonds instead of roasted ones, so I figured it'd be much better to find out for sure if there are really any differences than to spout off some half-cocked nonsense.

Guess what. They're different.

In short, without adding extra oil to the ground almonds, it takes about twice as long for a food processor to turn raw almonds into a creamy paste/butter as it does for roasted almonds. In my specific case, the right consistency was reached at 13 minutes.

Why is this? It's possible that the roasting heat weakens the almonds' cell walls so that oil is released faster, but I'm only speculating and don't really know the answer. That being the case, I won't bore you with another 286,973-word dissertation and just let the pictures do most of the talking here. The total elapsed processing time, in 30-second intervals, is superimposed on the food processor's spindle:

A couple of other notes and observations:

• You can see my original almond butter recipe, using roasted almonds, here.
• The almonds did not turn into random shades of brown during processing. As the 00:30 and 01:00 photos show, most of the lighting for this setup came from the side (bottom left in the images) instead of the top. The specific shade in each photo only indicates how thickly the almond particles coated the sides of the workbowl (thus blocking the light) and does not reflect color changes in the ground almonds themselves.
• Overall, though, roasted almonds do make for a darker finished product than raw almonds, which should be expected. This is a color sampling from my two batches:
  
• Unlike making almond butter with roasted almonds, I had to scrape down the sides of the workbowl this time. To preserve the grinding and clumping characteristics, this was only done after each photo was taken.
• A side benefit of the extended processing time is that it really isn't crucial to stop the machine at some exact specified moment. Although I said the right consistency was reached at 13 minutes, the changes in texture from about 11 minutes onward were really pretty minor.
• I have no idea if the specific equipment matters or not, but I used a full-sized KitchenAid food processor to make both this and the roasted almond butter.
• I split this batch and froze part of it. The result after thawing? Skin, stir, and a side-by-side tasting: no difference. So, homemade almond butter can be frozen with no problems.
• Update 2009-09-11: Based on a question in the comments below—homemade almond butter should be refrigerated, and you should keep it there for no more than four months. This is to prevent (or delay) the oil from going rancid. If you need longer storage, freeze it.

A NOTE FOR RAW FOODISTS

The almonds started at room temperature and became warm as a result of the long processing time. I did not take any measurements, but the temperature increase may be a concern for strict raw foodists. I don't know what effects starting with frozen almonds would have.

Update 2009-09-06: By law, since September 2007 all almonds grown and sold in the United States must be pasteurized, either through fumigation or heat treatment (PDF). This means that domestically-grown almonds are no longer acceptable for people on a strict raw-foods diet, even though marketers can still label these almonds as "raw" for sale to consumers. This law does not apply to foreign-grown almonds imported into the US, however, so if you're on a raw regimen, it would be advisable to assume that regular supermarket almonds are not raw (no matter what the label says) and buy truly raw almonds only from a specialist supplier.

Read more inside...

Homemade almond butter

Update: Please also see my other almond butter posts:

• Raw almond butter
• Can you freeze almond butter?

[Direct link to recipe]

Man, that clementine cake post just won't leave me alone. One of the questions I had while making it was how long I could grind the almonds in a food processor before they turned into almond butter. Well, I went and found out.

Let's get straight to it, shall we? This is a photo sequence of the almonds being processed, taken at intervals of 30 seconds each, and stopping after 6 minutes. You can click on each thumbnail for a larger image:

0:00

0:30
The almonds are nicely ground at this point. Knowing what I know now, this (or earlier, even) is where I would stop if I were making almond meal.

1:00
I don't think the granule size here is really any smaller than at 0:30, and clumping has started as the almonds' cells begin to rupture and release oil.

1:30
Between 1 and 3 minutes, the clumps continue to get bigger as more oil is released.

2:00

2:30

3:00
The clumps start to come together.

3:30
Between 3:15 and 3:30, the clumps are swept into one mass. At this point, the consistency is still thick, dry, and gritty. It is not spreadable yet.

4:00
Soon afterward the ball of almond butter is broken up. With more processing, more oil is released and the texture gets creamier.

4:30

5:00
I didn't notice it at the time, but from these photos it appears that the almond butter got significantly smoother between 4:30 and 5:00.

5:30

6:00
A good visual consistency is reached between 5:30 and 6:00, which is later verified by taste. Further processing may thin out the almond butter a bit as more air is mixed into it and perhaps still more oil get released, but airiness or fluidity aren't qualities we usually look for in nut butters.

If you'd like, you can also see these photos together as an animated sequence (671 KB GIF file).

RECIPE

Makes a little over one cup almond butter (estimated 9–10 fl. oz., or 270–300 ml).

INGREDIENTS

Quantity	Alternate Measures	Item
286 g	10 oz (2 cups)	Raw almonds with skins

EQUIPMENT

• Weighing scale or a dry-ingredient measuring cup
• Rimmed baking sheet (for roasting almonds)
• Food processor, fitted with steel blade
• Rubber spatula
• Canning funnel
• Glass or plastic jar, 10 fl. oz. (300 ml) or larger in capacity

INSTRUCTIONS

Steps 1–4 are optional and only applicable if you want to roast the almonds. If you want to make raw almond butter or use store-bought roasted almonds, you can start at Step 5. Update 2009-03-10: Well, maybe roasting almonds is a bad idea. Yikes.

1. Make sure the oven rack is at the center position. Turn on the oven and set the temperature to 375 °F.
2. Spread the almonds out evenly on a rimmed baking sheet.
3. Once the oven has reached its temperature, place the baking sheet into the oven and toast until fragrant, about 12 to 15 minutes.
   NOTE: This timing is only a guide. Every oven is different, and baking sheets come in different thicknesses. The best way to tell if the almonds are done is by aroma. USE YOUR NOSE. If you adhere strictly to the time and get burnt almonds, I'm not responsible.
4. Remove baking sheet from oven and let the almonds cool.
5. Once the almonds are at room temperature, place them into the workbowl of the food processor and process for 6 minutes or until the consistency is smooth. Stop the machine a few times to scrape down the sides of the bowl if needed. Update 2009-03-12: Raw almonds will need between 12 to 15 minutes of processing time.
6. With a rubber spatula, scrape the almond butter from the workbowl into a jar. A canning funnel fitted over the jar will minimize messy drips and spills.
7. Update 2009-09-11: This had completely escaped me, but a question in my raw almond butter post was a reminder: homemade almond butter should be refrigerated, and you should keep it there for no more than four months. This is to prevent (or delay) the oil from going rancid. If you need longer storage, freeze it.

EXTENDING THE RECIPE

Look, I know the way I break down recipes into infinitesimal detail makes them look scary, but all that verbiage can be reduced to one essential procedure: Put almonds in a food processor and run it until the consistency is right. With this knowledge as a foundation, it's a small conceptual step to extend this recipe to other nuts or even seeds, such as sesame seeds for tahini—there is certainly no shortage of recipes on the Internet. Although the specific processing times for each nut or seed may be different, the general idea is the same. You can also add salt or honey to fit your own tastes if you'd like.

YIELDS

I used a one-pint jar for this post's header photo, and as you can see, this batch filled it to just a bit over the halfway mark. So, two cups of raw almonds makes a bit over one cup of almond butter. As a rough estimate, I'd guess around 9 or 10 fluid ounces (270 or 300 ml).

Another unresolved issue from the clementine cake post was whether roasting the almonds will result in their weight loss. I was disciplined this time and did not snack on any of the in-process almonds. For this batch, 286 grams of raw almonds yielded 280 grams after roasting, so it's apparent that some loss of moisture or perhaps also volatile oils (hence the "done" aroma) does take place in the oven. I don't know if the almond's skin and flesh evaporate in similar fashion, though, so we will need further confirmation with blanched almonds.

Update 2009-08-18: Just found some photos of roasted nuts under an electron microscope (scroll about halfway down the page). You can see that their cell contents have shrunken after roasting, so a weight reduction should be expected.

Read more inside...

Clementine cake

With inspiration from smitten kitchen and knowledge that oranges are (one of the) symbols representing fortune for the Chinese New Year, it seemed only appropriate to mark the occasion by baking this cake. The ingredients list is short enough: oranges, eggs, sugar, almonds, and baking powder. Among the distinguishing features of this cake is the exclusion of both flour and butter from its recipe; however, I think what people will find most unusual here is the usage of whole oranges—both flesh and peel—as its main ingredient.

Before I can make this cake, though, there are hurdles to overcome...like, not owning a springform pan. Okay, no problem, I'll just call my friends to borrow one. Now, with a springform pan in hand, are we all set to bake? Uh, not so fast. It's the wrong size: the recipe is written for an 8-inch cake, but what I have is a 10-inch pan. This means that, to avoid ending up with an orange-flavored chapati, I need to make changes to an untried recipe before even taking the first step. Yikes.

Volumetrically, a 10" diameter cake is 56% larger than an 8" cake of the same height, and the most straightforward way to modify the recipe is simply to increase all ingredients by this amount. However, orange and egg quantities aren't amenable to being altered with such precision. The only practicable solution, then, is to change the recipe in whole-orange or whole-egg increments, change the other ingredient quantities accordingly, and make a cake that will be somewhat different in height from the original recipe and monitor the baking time to accommodate these changes. As it happens, the original recipe makes it easy to increase the egg quantity by 50%, and I was able to parallel this increase fairly well with the oranges I had on hand.

Oh, Just so you know, I did bake this cake in time for Chinese New Year, but it's taken me until now to finish this post. On to the recipe!

INGREDIENTS

Makes one 10-inch cake.

Quantity	Alternate Measures	Item
560 g	1 lb 4 oz  (6 to 8 ea.)	Clementine oranges
375 g	13¼ oz  (3½ cups)	Ground almonds
338 g	12 oz  (1½ cups + 3 tbsp)	Granulated sugar
10 ml	2 tsp	Baking powder
9 large	9 large	Eggs
As needed	As needed	Butter or shortening, for greasing cake pan

EQUIPMENT

• 3- to 4-qt. saucepan, for boiling oranges.
  
• Electric kettle or small saucepan, for boiling water to blanch almonds.
  
• Medium-sized bowl (about 4 qt.), for blanching almonds. You can use this same bowl later to whisk the eggs.
  
• Large bowl (about 6 to 8 qt.), for final mixing of cake ingredients.
  
• Rubber spatula or wooden spoon.
  
• Whisk.
  
• Colander.
  
• Knife and cutting board.
  
• Wide and shallow baking pan (such as a rimmed baking sheet, aka jelly roll pan), for toasting almonds.
  
• Food processor, fitted with steel blade. This will be used for both grinding the almonds and pureeing the oranges. There is no need to clean the workbowl between the two operations.
  
• 10" springform pan.
  
• Parchment paper and kitchen shears/scissors.
  
• Toothpick, for testing cake for doneness.
  
• Kitchen timer, or other time-keeping device.
  
• Kitchen scale and/or dry-ingredient measuring cups and spoons.

INSTRUCTIONS

You can also click on each of the images below to zoom in for more details.

PREPARE ORANGES AND ALMONDS

Oranges

1. Wash the oranges thoroughly. Remember, you'll be eating the skin.
   
2. Remove and discard any stem and calyx still attached to the fruits. I've found that these hard remnants tend not to get pureed in the food processor (see Step 8 photo), so it's best to remove them beforehand.
   
3. Place oranges in a 3- to 4-qt. saucepan and fill with enough water to float the oranges up by one-half to one inch.
   
4. Place saucepan on the stove. Bring water to boil over high heat, then place a lid over the pan, reduce heat to between Low and Med-Low, and simmer for two hours.
   
   
5. While the oranges are being boiled, check occasionally and refill the water if necessary.
   
6. Remove oranges from the cookpot (or drain into a colander) and let the oranges cool.
   
   
7. Once cooled, cut the oranges in half crosswise to check for seeds (pips). Remove and discard any seeds that you find. As you slice open the fruit, juice will run out. Be very careful to save as much juice as possible.
   
   
8. Now chop the oranges into pulp. This can be done with a knife, but is easier with either a food processor or a blender.
   

Almonds

Commercially-ground almonds, sold as either almond flour or almond meal (different terms for the same product), can be purchased online or in natural- and health-food stores. I've seen Bob's Red Mill almond meal in the natural foods section at a Kroger store near me, so I'm guessing it's the most widely available brand for local purchase. If you have or plan to buy commercially-ground almonds, then you can skip this entire almond-prepping section.

Before grinding the almonds, I removed their skins by blanching to avoid a mottled appearance in the cake and possibly avoid any additional bitterness from the skins. This is probably an optional step, though, depending on your preference. You can also buy blanched almonds in the store, in which case you can grind them up immediately.

The advantage of commercially-ground almonds, beyond the obvious labor savings, is a finer and more consistent grain size than almonds ground with a food processor. However, toasting nuts before adding them to recipes enhances their flavor, and, as far as I can tell, almond flour sold in stores is not toasted before grinding. Can you toast store-bought almond flour? Maybe, but I'd imagine the margin of error between "toasted" and "burnt" is a lot smaller with the flour than with whole almonds. Commercially-ground almonds also retail for about $12 per pound, whereas raw (skin-on) almonds cost around $6/lb. Update 2009-03-10: Well, maybe you shouldn't roast the almonds. Yikes.

You should prepare the almonds while the oranges are being cooked.

1. Place raw (skin-on) almonds into a bowl. As I mentioned above, if you have already-blanched almonds on hand, steps 2–5 are unnecessary.
   
2. In a separate pot (not the one with the oranges in it), boil some water. The quantity is not critical—you just need enough to cover the almonds.
   
3. Pour the boiling water over the almonds to cover, and soak for 1 minute.
   
   
4. Drain the almonds into a colander and rinse with cold water. Pinch the almonds between your thumb and forefinger, and the skin should slip off. Discard the skins.
   
   
5. Place blanched almonds on a towel and pat to dry.
   
6. Make sure the oven rack is at the center position. Turn on the oven and set the temperature to 375 °F.
   
7. Spread the almonds out on a wide baking pan. Place the pan into the preheated oven and toast until fragrant, about 10 to 15 minutes.
   
8. Remove almonds from oven and let cool.
   
9. Place the almonds into a food processor and grind into flour. I did not record how much time this took and simply stopped processing once the almond grains appeared "fine enough." The top image at left shows the ground almond's texture in the food processor, and the bottom image is a closer-in photo of the almonds in a mixing bowl. I don't know if it's possible to get the grain size to be much smaller and more consistent than this without using millstones, and I was afraid that I'd end up with almond butter if I had let the food processor run for much longer. Update 2009-03-03: Grind for no more than 30 seconds. See my homemade almond butter post.
   
   
10. Pour the ground almonds into a large mixing bowl so that the food processor can be used for pureeing the oranges.

PREPARE BAKING PAN

• If you haven't already done so for toasting the almonds, make sure the cake rack is at the center position, turn on the oven, and set the temperature to 375 °F.
  
• The original recipe (see Nigella Lawson, below) instructs to prep. the pan with both butter and parchment paper but doesn't explain why. The 1997 Joy of Cooking mentions that butter can applied between the pan and the parchment to keep the paper in place, yet How To Cook Everything has a recipe that instructs adding butter to the top (exposed) surface of the paper, which seems redundant to me. (I mean, isn't parchment non-stick already? Why add butter on top of it? Isn't that just gilding the lily?) Now that I've baked this cake, though, and know how sticky it is, this is what I think the procedure should be:
  1. Cut a 10" diameter disc of parchment paper and place it onto the bottom of the cake pan. Apply butter to the top side of the parchment. With this cake, this is not redundant, as it needs all the help it can get to release from the pan.
     
  2. Butter the sides of the springform pan, then line with a parchment. A total length of 31½" is needed to fit a 10" pan, although it is not necessary to have this in one continuous strip. Add butter the exposed surface of the parchment.

COMBINE INGREDIENTS AND BAKE

1. Add the sugar and baking powder to the ground almonds in the large mixing bowl. Mix thoroughly with a rubber spatula or wooden spoon.
   
2. In a medium mixing bowl, whisk the eggs.
   
3. Pour the whisked eggs into the large bowl and mix with the dry ingredients already in the bowl.
   
4. Pour the orange puree into the large bowl. Mix thoroughly with the egg/almond batter.
   
5. Pour cake batter into the prepared springform pan.
   
6. Place cake pan into the preheated oven and bake at 375 °F for approximately 1 hour. The cake is done when a toothpick inserted into its center comes out clean. When I checked my cake at 45 minutes, it was still jiggly, but had firmed up at 1 hour. My cake was done at 1 hr 10 min., which is consistent with other recipes that call for 1-hour bake times.
   
7. Once done, remove cake from oven and let cool in the pan.

Note: The original recipe suggests that you may need to cover the cake with aluminum foil after 40 minutes to avoid burning its top. I did not find this extra step to be necessary.

RECIPE COMMENTARY

• First of all, this cake is easier than my recipe makes it look—it's actually easy, period. I just didn't want to leave out any details or render any steps ambiguous. If you've read this far, congratulations, and go make the cake. If you know anyone who wants to make a simple cake but got scared by the length of my post, tell them they can follow Nigella Lawson's recipe for it.
  
• This is apparently a very popular recipe, and copies of it are all over the Internet, most of which refer back to Nigella Lawson. However, from what I can tell this cake is probably Middle Eastern in origin, and Claudia Roden has a variant that uses large oranges (such as Valencias or navels) instead of clementines; you can find her recipe at the bottom of this page. It was also pointed out on Chowhound that James Beard had included a large-orange cake recipe in one of his cookbooks from the early 1980s.
  
• I think the original recipe is a bit vague in its instruction to boil the oranges. Should I cover the pot with a lid? How hard should I boil them? With the amount of water originally specified (just enough to cover the oranges), I'd be afraid of cooking the pot dry if it's left on high heat without a lid for two hours. It's much safer, I believe, to cover the pot and leave it at a gentle simmer for this amount of time.
  
• Speaking of vagueness, I have a feeling this recipe can tolerate pretty broad variations in the ingredient quantities, so measurement accuracy isn't a crucial factor in its outcome. As an example, the original recipe called for one heaping teaspoon of baking powder for an 8-inch cake, which is an imprecise quantity that can vary up to about 2x depending on how tall the heap is. For the sake removing ambiguity, though, I've specified exact quantities for all ingredients.

CAKE COMMENTARY

• The cake domed up during baking, but it fell as it cooled so that the top of the cake flattened nicely.
  
• The texture of this cake is denser than those made with wheat flour; however, it is not heavy.
  
• The cake is also very moist—so moist, in fact, that it almost feels underdone. However, as the toothpick came out clean, my cake was properly baked by that test. I've seen descriptions that range from a soaked sponge to being almost pudding-like.
  
• There is a bit of graininess from the almonds, though it is not unpleasant. I'd imagine that commercially-ground almonds will result in a smoother texture. The occasional bites of orange zest provide zing to the cake.
  
• I like this cake, and, judging by the comments from the Food Network page, so do most people; the few negative reviews, however, are rather strident. Aside from just a matter of individual taste, I think it's possible that the negative reviewers may have expected a cake similar to those made with wheat flour instead of something so dense and moist. Another issue is that clementine oranges are by no means uniformly sweet, and it may be difficult to predict their flavor until they're baked into the cake; it is possible, then, that the negative reviewers may have mixed bitter oranges into their cakes.
  
• It's not entirely clear to me why this cake is so sticky. I tried to compare it to cake recipes that contain butter, which only revealed to me that either a) there is no such thing as a standardized cake recipe (and therefore no comparison is possible), or b) my search skills stink. At any rate, my guess is that it's a combination of the number of eggs (more than most recipes, it seems, though not excessively so) and the amount of butter used (zero). While almonds contain a lot of fat, I think they're still bound up in the ground particles, and thus are not useful for lubricating the cake. On the flip side, the protein in the almonds is similarly bound up, so it should not contribute to making the cake any stickier.

KOSHER FOR PASSOVER?

Well, I'm not going to get into details about Jewish dietary laws, because I know almost nothing about them. From what little I understand, though, I believe chemical leaveners are acceptable for Passover, and the only offending ingredient in most baking powders is cornstarch. If that's the case, then it's easy to make this recipe kosher by mixing your own baking powder at home.

The standard formula for homemade baking powder is two parts cream of tartar to one part baking soda by volume. Cornstarch is sometimes added (same quantity as baking soda) to absorb moisture during storage, but if you're going to make and use your own powder right away, this is a non-issue. For this recipe, the quantities are:

1 tsp	Cream of tartar
½ tsp	Baking soda

Sift together the two ingredients. This will make 1½ tsp of baking powder, which at first appears less than what I had specified above; however, please realize that the original quantity assumes usage of commercially-made baking powder, which has added bulk from cornstarch whereas this homemade powder doesn't.

This is a single-action baking powder, which is activated by mixing with a liquid. All the baking powders I've seen in stores are double-acting, which are activated again by the high temperature inside an oven. To make sure the leavening action from this baking powder doesn't fizzle out, make sure to place the cake into the oven as soon as possible after the batter is mixed.

Of course, feel free to correct me if I'm wrong.

CLEMENTINES, MANDARINS, & TANGERINES

So what's the difference between a clementine orange, a mandarin orange, and a tangerine anyway? Well, to start with, the tangerine is a type of mandarin orange, not a different variety of citrus fruit. While the clementine is commonly thought to be an accidental cross between a mandarin and a sweet orange discovered in the Algeria by Father Clement (Vincent Rodier) in the early 1900s, some scholars maintain that "it is indistinguishable from and probably identical to" the mandarin orange common in the Guangzhou (Canton) province of China. Now, I'm no expert so don't take my word as gospel, but it seems the short answer is: they're all mandarins.

The satsuma is yet another member of the mandarin family, though I don't think I've seen it in the stores around here lately. This may well mean that, in terms of eating and cooking, the question of distinguishing between different mandarin breeds will become moot as the clementine increases its market share and displaces other mandarins. Specifically with regard to this recipe, considering that a large-orange variant of it exists, I'd venture to say that all mandarin types can be used interchangeably for it.

YIELDS

I did not record this, but if my memory is correct, I started with 388 grams of raw almonds. After removing the skins and toasting them (and eating a few along the way, I admit), I weighed them again and got 358 g of almonds and 14 g of skins. Did I eat that many? Almonds with skins weigh, roughly, around 1.7 to 2 grams each, which calculates out to a loss of between 8 to 10 almonds. Except I don't think I ate that many. Really, I don't. My guess is that the difference between the original quantity, what I ate, and the final weight is the result of moisture lost during toasting, but I don't have enough data to be sure. For future planning, though, I think that the weight ratio of these blanched/toasted almonds and skins (358:14) is a good place to start.

For this cake, I added a second batch of almonds to bring the blanched/toasted almond weight up to 375 g, although as noted above, exactly quantities are probably not imperative for this recipe.

Update 2009-03-03: From my homemade almond butter post, 286 grams of skin-on almonds yield 280 grams after roasting. If whole almonds and blanched almonds evaporate similarly, we should expect an after-roast weight of 380 g for a start weight of 388 g. From above, the blanched almonds and skins add up to 372 g, meaning an unaccounted-for loss of 8 grams. Taking into consideration how much an individual almond weighs, this calculates to a loss of 4 or 5 almonds—which actually sounds about how many I snacked on, so the data appears correct.

Read more inside...