Sciency Answer: Seed cleaning and sprouting

Today's question is from Keith Long, brought on by a comment this post on his blog:

My question is about Rhodochiton astrosanguineum seeds.
When you buy these they're very small seeds with the husk (of each seed) removed. Yet the germination rate is very poor. Either that or the the type of people who buy them clearly are incapable of following their instructions!
Yet once you have a plant, and have collected the seeds, the germination rate is near 100%. Clearly, I don't mess about removing the individual seed husk, I just put the seeds in. The seeds in their husk are nearly the size of a chilli seed.
Why are they husked? Why does this lead to such a low rate of germination? And merely out of interest, how on earth do they do it without losing the minute seeds?

The factor here is almost certainly not the husk covering the seed, but the freshness of the seeds. Many seeds rapidly loose their viability, and need to be sown right away, while others simply start taking longer to germinate the longer they sit around dry. Based on my poking around, it appears that very fresh Rhodochiton seeds germinate rapidly and easily, while older seeds will still sprout, it just takes significantly longer.

Which makes one wonder... why would seeds do that?

A gardener, of course, wants every seed to sprout as soon as it is planted. But in the wild, plants need to be more careful. If every seed sprouts right away, one flood or late frost can wipe out the entire next generation. So most wild plants have various tricks to ensure seeds don't germinate all at once, or germinate at the best possible time. For a plant like Rhodochiton, fresh seed that falls on moist soil will sprout right away, getting a quick start on the next generation. But any seeds get a chance to sit dry for a while drop into a deeper dormancy and hang around without sprouting, acting as a sort of insurance policy to make sure there are still seeds around if something happens to those that have already germinated -- much as gardeners usually don't plant the whole packet at once in case of damping off.
How readily or uniformly seeds sprout often depends on the climate they evolved in. Plants from desert areas with erratic rainfall are notoriously hard to get good germination from, instead one seed at a time will sprout over a very long period -- extreme insurance for a difficult, erratic climate. Plants from wetter, more predictable climates tend to have seeds that sprout more uniformly.
Plants that have been grown for a long time by humans almost always develop quick and uniform germination because without even trying to, we tend to select the individuals that sprout first. If you sow 100 seeds, and 10 sprout in a week, most people just prick out those ten and forget about the other 90, even though they may have eventually sprouted. Those quick germinating seeds will go on to have more rapidly germinating offspring, and so on, until they all sprout at once like most familiar annuals and vegetables.

To get to your other questions about removing the husks, seed companies usually remove them for a number of reasons: It looks neater and tidier in the seed packet, the cracks of crevices of the husks can offer ideal little hiding places for fungi. In some cases, it also allows the seed producers to get a good look at the seed itself and separate out small, shriveled seeds that are unlikely to germinate.
Cleaning off all the husks and chaff of seeds can be rather a pain. I worked for a while for the Ornamental Plant Germplasm Center, and spent quite a bit of time cleaning seeds. The first step is usually to gently rub the seed heads between to rubber blocks, which crushes and breaks up the seed husks. You can then separate the chaff from the seeds a number of ways. A fine sieve will let fine seed fall through but keep big chunks of chaff behind. We also had a cool machine which was basically a big plastic tube with a fan in it which allowed us to blow off the light chaff but leave denser seeds behind. Various other shaking, blowing, and sieving machines are used to rapidly get all the seeds in one pile and the other stuff in another. It is kind of cool, but when you are working with many different species as we were, you have to figure out the best machine and setting for each species, which can lead to a frustrating amount of trial and error.

Have a question? Get a sciency answer! Just e-mail me: engeizuki at gmail dot com