Coconut palms grow some of the biggest seeds on the planet (coconuts), and the tiny black specks in very good real vanilla ice cream are clumps of some of the smallest, seeds from the fruit of the vanilla orchid (the vanilla “bean”). Both palms and orchids are in the large clade of plants called monocots. About a sixth of flowering plant species are monocots, and among them are several noteworthy botanical record-holders and important food plants, all subject to biological factors pushing the size of their seeds to the extremes.
People seem innately fascinated by biological extremes—what is biggest, smallest, fastest, slowest, oldest, smelliest, most dangerous or most of whatever other adjective is of interest. The satisfaction of parameterizing our known world with this trivia motivates us to leaf through The Guinness Book of World Records and stand in awe of the General Sherman tree, and perhaps it similarly inspired Christ to invoke extreme seed size in the “Mustard Seed Parable” (Matthew 13:31-32):
“He set another parable before them, saying, ‘The Kingdom of Heaven is like a grain of mustard seed, which a man took, and sowed in his field; which indeed is smaller than all seeds. But when it is grown, it is greater than the herbs, and becomes a tree, so that the birds of the air come and lodge in its branches.’”
We’ll forgive his smudging of botanical fact into allegory. After all, though mustard seeds are not the smallest seeds on Earth (and grow into decent-sized weeds, not trees), they are pretty small, and it’s unlikely that his audience of farmers in the Holy Land were familiar with tropical epiphytic orchids. The range of seed size in the modern flora should capture our interest beyond metaphor, however, as it happens to be extraordinary and involves some interesting evolutionary history, and seed size affects many aspects of a plant’s life. Seed size in the modern flora spans 11.5 orders of magnitude by mass (Moles et al. 2005), and the entire range is encompassed by a single group of plants that consists of about a sixth of modern flowering plant species.
Both the species with the largest seeds (the double coconut palm, Lodoicea maldivica, family Arecaceae) and the species with the smallest seeds (various epiphytic orchids, Orchidaceae) in the world are monocots, members of the large, monophyletic (derived from a single common ancestor) Monocotyledon clade, which split from the rest of the angiosperms and began diversifying relatively early in angiosperm (flowering plant) evolution. Technically, the double coconut and the orchid species with the tiniest of the microscopic orchid seeds are beyond the usual scope of this blog, as they’re not food plants.
Another very large-seeded palm, however, the coconut (Cocos nucifera), and another orchid with exceedingly small seeds, the vanilla orchid (Vanilla planifolia), could very well be in your pantry right now (or will be soon, if you decide to bake the Extreme Monocot Cookies from the recipe below) and are certainly representative of the tail ends of the seed size range. Boasting the biggest and smallest seeds alone would make the monocots noteworthy. Monocots also, however, are directly or indirectly (as animal forage) responsible for a large fraction (most?) of the calories consumed by people worldwide. Before we speculate about what forces may have pushed seed sizes in monocots, we should take a moment to further appreciate the clade.
The monocots (and a sort of stream-of-consciousness account of some botanical record holders)
In addition to the single (“mono”) cotyledon (embryonic leaf) that gives the clade its name, most of the 60,000+ monocot species are characterized by a list of features that you probably had to memorize at some point for a test (summarized here), including, in part, parallel leaf veins, flower parts in groups of threes, pollen with a single pore, unique vasculature and lack of true secondary growth (described by Katherine in her asparagus post), and adventitious roots (which are mechanically impressive, as Katherine explains in her post about leeks). The monocots of particular interest here are vanilla, which is one of the 20,000+ species of orchids (family Orchidaceae; order Asparagales), and the palms (family Arecaceae; order Arecales), which provide coconuts, dates, palm oil, and numerous tropical fruits, include acaí. Monocots also include the 15,000+ species in the grass family (Poaceae; Poales), which covers a large fraction of the planet in pasture and provides the world with grains, sugarcane, and bamboo. Other edible monocots are on the phylogeny below.
Monocots incidentally also include the smallest flowering plant species, the aquatic duckweed Wolffia (Araceae), which also makes the smallest flowers, which mature into the smallest fruit, but these smallest fruits do not contain the smallest seeds. That distinction goes, as I have mentioned and will explain below, to the orchids. The Wolffia fruit is a thin husk that barely coats the single seed inside it, which is about the same size as a grain of table salt (sodium chloride). Orchid fruits, in contrast, are pods that can contain up to multiple millions of seeds, each of which can be smaller than a grain of salt and weigh less than a microgram. That is, tiny seeds in not-so-tiny fruits.
In The Private Life of Plants, David Attenborough claims that, at 34 square feet, the world’s largest undivided leaves belong to a giant Bornean arum (I think he’s talking about Alocasia robusta), which is awesomely in the same plant family as the miniscule Wolffia (and therefore also a monocot). “Undivided” means that the leaf lacks the feathery dissections like a palm leaf or division into leaflets, like in a clover. As much as I admire Attenborough, though, the A. robusta leaf is probably not the biggest. The undivided circular leaves of the giant water lily (Victoria amazonica; Nymphaceacea), a basal angiosperm, can have a 10-foot diameter (78 square feet in surface area) and are therefore larger than the arum leaves, as noted by botanist Wayne Armstrong on his fabulous page on botanical trivia. Interestingly, both the very large size of the waterlily pad and the very tiny body of duckweed may be construed as alternative adaptations to living in water, so maybe on terra firme the point goes to Attenborough.
Not to be easily dismissed, though, the arum family, Araceae, also boasts the largest unbranched inflorescence (flower stalk), on the Bornean titan arum. Its Latin name, Amorphophallus titanium, means “giant mis-shapen penis,” a reference to that inflorescence. This guy is responsible for that name (I think). The titan arum is also in the running for the stinkiest flower. Its inflorescence strongly smells of rotting carrion to attract flies that pollinate the flowers as they’re searching for the source of the smell. More importantly, though, the arum family gives us taro root (Colocasia esculenta), an important staple food throughout the tropics. The largest branched inflorescence? From a palm. The largest single flower, though, Rafflesia arnoldii, is a eudicot, but like the titan arum, it also lives in the rainforests of Borneo and strongly reeks of carrion to attract flies. Being on an island (Borneo) might have something to do with the large size of these flower structures, although that’s just rampant speculation on my part. Island life often evolutionarily pushes taxa to be exceptionally larger or smaller than their mainland relatives (see David Quammen’s The Song of the Dodo for a great discussion of the biological consequences of island habitats). Such appears to be the case for the double coconut palm, which has a very limited natural range in the Seychelles and is much larger than its closest relatives.
Back to leaves for a moment, though. The matter of the species with the smallest leaf is complicated, but 80 feet in length, the biggest dissected leaves, along with those largest seeds, are from palms (family Arecaceae; those 80-footers are from an African palm, Raphia regalis). Those super large palm seeds, though, don’t house particularly large plant embryos. According to Armstrong, the biggest embryo is in the seed of the legume Mora oleifera, a eudicot. The Mora seed is probably the largest eudicot seed, but with a weigh around a kilogram, it’s an order of magnitude lighter than the double coconut, whose mass can top out around 20kg. The ecological factors pushing seed size evolution likely similarly affect all major clades of flowering plants, and most flowering plant species are eudicots, not monocots. Therefore, it seems remarkable to me that the monocots hold so many botanical size records, including seed size records. I’d love to know what makes them so seemingly structurally labile.
Seed size variation
So why are orchid seeds so small, and coconut seeds so big? The job of the seed is to help maximize the chances that it will land in a good place for the embryo within it to grow, to protect the plant embryo until conditions are right for germination, and to help the newborn seedling establish. All of these components of that job can push seed size toward being large or small. As Katherine noted in her excellent post about black-eyed peas (seeds), it would seem that the interests of the embryo should determine seed characteristics. In most species, however, a mother plant produces more than one seed, and she must hedge her bets across all her offspring, and seed size is under maternal control. The seeds of most species store a certain amount of nutritive tissue (endosperm or cotyledon, depending on the taxon) to support the seedling after germination until it can photosynthesize on its own. Producing seeds, therefore, is energetically and nutritionally expensive for the mother plant, and she only has a certain amount of energy and nutrients to allocate to seed production. She can therefore either invest a large fraction of that allocated amount into each of a few large seeds, or she can invest a smaller fraction into each of several smaller seeds. There is cool math that describes this seed size vs. number tradeoff with reasonably well supported predictions (Leishman et al. 2000) about how the mother plant should allocate her reproductive effort across her seeds under different scenarios. Across all species, however, there are a few trends in the relationship between seed size and a plant’s habitat and life history (Leishman et al. 2000, Moles et al. 2005).
Bigger plants tend to produce bigger seeds. This trend seems to be especially true among closely-related species. Think about the seeds of familiar fruit species in the rose family (Rosaceae): strawberry plants are herbaceous and have very small seeds; roses, Saskatoon berries, raspberries and blackberries grow on woody shrubs or vines, and their seeds are a bit bigger than those of strawberries; and the biggest seeds are in the tree species within the family (almonds, peaches, plums, nectarines, apricots, cherries, apples, pears, loquats, quince, medlar). The palms (family Arecaceae) are the only tree-sized woody monocots, so large plant size may be one factor pushing large seed size in coconuts. Incidentally, coconut fruit and seed structure is fairly uncomplicated and is well explained here.
Seeds tend to be larger for species that germinate under stressful conditions: for example, in shade, or in saline or nutrient-poor soils, or in dry places. Several plant families, for example, have species that became mangrove trees, living in coastal areas in wet soils of various salinities. In every case, seed size is bigger in the mangrove species than in its closest relatives (Moles et al. 2005). Under stressful conditions, a seedling may require more time or support until it can produce enough of its own food, so the seed supplies a large amount of nutritive reserve, which increases seed size. Double coconuts sprout exclusively on granitic outcrops on their islands, and coconuts, of course, inhabit sandy tropical beaches. The former probably involves shade and perhaps drought stress, and the latter sounds pretty appealing to me, but only because I’m not a seedling, so maybe environmental stress has helped push seed size in coconuts.
Orchids and some other plant taxa have arrived at an alternative arrangement for fueling their seedlings: taking advantage of fungus. Instead of relying on nutrients packed into a seed, the embryo becomes a parasite on a fungus, which in turn gets its nutrients and energy from decaying organic matter or by being a parasite itself upon other plants. After seedling establishment these mycotrophic (fungus parasitizing) plants may become partially or entirely parasitic on another plant, and they become independent of their fungal symbiont, or they may maintain the relationship, often repaying the fungus with sugar from photosynthesis. Many plants form symbiotic relationships with soil fungus, in which the fungus receives photosynthate in exchange for soil nutrients, but only a few plants forego seed nutrient reserves and rely on fungal support for seedling establishment. So, without the burden of nutritive tissue, seeds can get very small, and a mother plant can make millions of such seeds. The epiphytic orchids have made this a fine art, as is well explained here and here, but some parasitic eudicots do it, too, if not quite so spectacularly.
Dispersal strategy affects seed size, too. Seeds transported by wind tend to be smaller than those dispersed by animals or those that, well, don’t fall far from the tree (mother plant). This is true in orchids. The tiniest dust-like orchid seeds are wind-dispersed, but the seeds of the vanilla orchid (a beautiful neotropical vine) are relatively big for an orchid and are coated in a sticky resin, which adheres itself to the coats of animals or detritus after the fruit (vanilla “bean” or “pod”) splits upon maturity. That sticky resin in the fermented and dried vanilla fruit is the origin of the delicious vanilla flavor. The tiny black specks in good vanilla ice cream or other desserts are clumps of seeds scraped from the vanilla pod. Coconuts, on the other hand, are dispersed by floating across the ocean. Part of their large size and morphology obviously helps them make the voyage. Double coconuts, though, don’t disperse across the ocean. They stay close to home, which is apparently exactly where mother wants them.
The cookie recipe below, modified from Heidi Swanson’s animal cracker recipe, combines coconut and vanilla and only uses plant ingredients from monocot species (grains, sugar, coconut, vanilla). I recently made these for the party favor bags for my daughter’s birthday party, but, ironically, I used a cookie cutter in the shape of an oak leaf, a eudicot.
Extreme Monocot Cookies
1 cup whole wheat pastry flour or oat flour
¾ cup unsweetened finely shredded coconut, ground in a blender or food processor into coarse flour
¼ cup extra-virgin coconut oil, softened
¼ to ½ cup sugar (from sugar cane, not sugar beets), as you wish
¼ teaspoon salt
1 large egg, lightly beaten
½ teaspoon vanilla extract, or seeds scraped from 1 inch of vanilla bean, if you’re feeling indulgent
turbinado or large-crystal sugar for sprinkling (optional)
Whisk the flour and shredded coconut together in a medium bowl. Set aside. In a separate medium bowl, beat the coconut oil with the sugar and salt until it’s smooth . Beat in the egg and vanilla until everything is uniform in appearance. Add the flour mixture and stir just until incorporated. Turn the dough out onto the counter-top, knead it once or twice and gather it into a ball. Cut the dough in half, flatten each piece, wrap and refrigerate for at least an hour.
When you are ready to bake the cookies, preheat the oven to 350F degrees. Place the racks in the middle and line a couple baking sheets with parchment paper. On a floured work surface roll the dough out 1/8-inch thick. If the dough cracks, let it sit and warm up for a couple more minutes. Stamp out shapes with floured cookie cutters and place the cookies an inch apart on the baking sheets, sprinkle with a bit of the turbinado sugar, if using. Bake until the cookies are just beginning to color at the edges, 7-8 minutes. Remove from the oven and cool the cookies on racks.
Leishman, M. R., I. J. Wright, A. T. Moles, and M. Westoby. 2000. The evolutionary ecology of seed size. Seeds: The ecology of regeneration in plant communities, 2nd edition. Pages 31-57. CAB International 2000. Ed. M. Fenner.
Moles, A. T., et al. 2005. A brief history of seed size. Science 307: 576-580.