Objectives:
1) Discussion of two fundamentally different types of adapted states:  those achieved by design and those achieved by (gradual) evolution by natural selection
2) Discussion of the "adaptive landscape" and "selection constraint" concepts as they pertain to the previous objective.
3) Discussion of seed dispersal in plants, its importance in plant life history.
4) Survey of different mechanisms for seed dispersal in plants, with a greater emphasis on wind dispersal.
5) Discussion of surface area and mass in the determination of terminal velocity, and the role of terminal velocity in wind dispersal.
6) Discussion of physical/biological constraints on wing size and seed mass in the seeds of plants.
7) Discussion of genetic/developmental constraints, as it pertains to Tipuana tipu.
7) Conduct class-wide experiment using wing size and seed mass as predictors of terminal velocity.
8) Evaluation of "winning designs":  would they be achievable by evolution by natural selection?

This lab was inspired in part by Richard Dawkins' book, Climbing Mount Improbable, and it makes liberal use of prose taken from an exercise written by Keith Karolyi at Reed University.

Objectives 1 and 2 above are the most important ones, but they are so philosophically slippery that I find it best to lay a concrete foundation with the biological realities of seed dispersal, which will offer us the context for discussion of the two main objectives.

Seed dispersal:  general
The dispersal of offspring is an important life history stage for almost all organisms.  For most seed plants, dispersal represents one of the two points in the life cycle in which plants move from location to location (those roots make it pretty hard to walk).  Offspring that are dispersed some distance from the parent plant are usually at an advantage, relative to offspring that pop up right underneath the seed parent.  Being "right underneath mom" probably also means having to deal with her shade and having to compete with all your siblings.  If mom has a lot of parasites, you'll probably get them, too.  The general idea here is that selection favors plants that disperse seeds further, and as a result, adaptations have arisen via natural selection for having seeds dispersed at a distance away from the seed parent.

For the time being, let's imagine natural selection operating on the basis of :  "the further away a plant sends seeds (on average-- naturally there will be a lot of variation in the actual distance dispersed), the more successful it is."

Survey of modes of seed dispersal
Major modes of seed dispersal include ballistic, animals, water, and wind.  In ballistic seed dispersal, seeds are "shot" out of a pod-like structure--sometimes explosively-- as the pod opens.  A lot of legumes (bean family Fabaceae members) use this method of dispersal, and it requires a special adaptations in the structure of the pod, such as a weak "seam" the splits open very gradually as the pod dries out.  Animals carry seeds either in their guts or on their fur.  The existence of edible fruit on a plant due entirely to its need to attract animals that will eat the fruit and disperse the seeds somewhere else.  Plants that live close to the water (like coconuts) tend to just drop their seeds straight down.  What is important is that the seeds float so that the movement of the water will carry the seeds to new locations.  With wind dispersal, plants are typically dropping their seeds (or releasing them when agitated) and allowing the movement of air to carry them away from the seed parent.

This lab focuses on plant adaptations for wind dispersal.  Note that this is not an obscure phenomenon.  In temperate-zone plant communities, 30-40% of tree species show adaptations for wind dispersal of diaspores (Howe and Smallwood 1982).  Most of the adaptations observed for wind-dispersed species appear to functionally reduce the terminal velocity (maximum rate of descent) of the seed.  As terminal velocity decreases, the opportunity for horizontal travel (i.e., away from the parent) increases.  The terminal velocity for maple samaras (the one-winged spinning seeds of maple trees) has been empirically shown to be proportional to the square root of wing-loading (seed weight divided by wing area).

Fortunately for us, the morphological features that influence distance of of wind-dispersed seeds really boil down to just two:  "wing" area (presuming that the seed is using a wing-like structure for slowing the rate of descent) and seed mass.  For wind dispersal, increased seed weight has a negative impact on the distance dispersed.  Not surprisingly, wing area has the opposite effect.  The theoretical extreme would therfore predict that a weightless seed or one with an infinitely large wing could stay afloat forever (terminal velocity of zero), but this is obviously impossible in reality.  We can, however, predict that as far as natural selection on seed dispersal is concerned, smaller seeds and larger wings are favorable.

Biological reality steps in when the question is asked:  "why don't seeds continually evolve to become more and more miniscule, and their wings to become larger and larger?"  You must remember that seed dispersal is only one element in the overall life history of the plant.  In order to achieve high Darwinian fitness, the plant must be successful in all aspects of the life cycle.  Making a seed that is too small, for example, would mean taking nutrition away from the seedling that is to develop from that seed.  In order to make a seed-wing that is enormous would require so much of a parent's resources that it may have to produce fewer seeds.

Besides that, having a seed with a zero terminal velocity and infinite dispersal distance may not be the best thing either.  I can imagine that there could be disadvantages to dispersing too far, such as risking leaving the habitat in which the seed is adapted to live.  Perhaps there is some "optimal" distance for seed dispersal that is greater than zero and smaller than infinity.

Local example:  Tipuana tipu, a native of South America
If you are from practially anywhere else in the U.S. besides Southern Cal, it's likely that you are already familiar with maple trees, which are  common in the deciduous forests of the east coast and midwest.  Maples have seeds (diaspores actually, if you want to use the correct term) that are samaras, which "helicopter down" as they fall from the seed parent tree.  Around our campus, we don't have maples, but we do have a tree that produces samara-like diaspores, the Tipuana tipu.

The really neat thing about this tree is that although its seeds look superfically like a maple's, it is a legume and almost certainly the morphology of its diaspore is derived from an ancestral state that was a regular bean pod. If you look at the seed pods shortly after fertilization, it looks pretty much like a Chinese pea pod (snow pea), though through subsequent development, it matures a single seed at the base and a thickened edge to a wing that is otherwise and very thin. Like samaras, this pod "helicopters" as it descends at a nearly horizontal attitude, spinning with the thickened edge in "front" of the spinning wing.  By the way, this shape is a pretty important part of the adaptation, since the horizontal attitude it causes is really what slows down the terminal velocity.  If it were to fall in a vertical-wing position, it's rate of descent would be the same as a wingless seed.

The next part of this lab will entail your fashioning wind-dispersed diaspores out of tape and construction paper, each diaspore carrying a Skittle® for a "seed."  You will have more or less free-license to design your diaspore(s).  As you are thinking about how you are going to approach this engineering problem, remember that in order for the wing surface area to have a slowing effect, it must stay in as close to a horizontal attitude as possible.

The Experiment
As a class, we will travel to a site on campus where your instructor can climb up to a dizzying height and drop the seeds that you have designed and built.  Someone will record descent times.  We will then return to the lab and determine the weight and surface area for each individual samara for which we have recorded a distance.  If there is a willing voluteer in the class that has a scanner, we may be able to generate some very accurate measures of wing area.  The information for each diaspore (mean distance, weight, length, and width) will be entered into a data set, which we will analyze as a class-wide exercise.  The data and graphs will be posted on a web page and you should be printing them off and including them with a critical evaluation in your laboratory journals.

The Philosophy Lesson:  "true design" vs. "apparent design"
In all probability, there will be three approaches that students in this lab will use in designing their customized diaspores.  First, there is "improvement on an existing natural design," such as taking a diaspore morphology that already floats well-- like a samara-- and copying it while adding modifications that might even further diminish terminal velocity.  Second-- which is quite similar to the first-- is "improvement on an existing design that is not found in nature," basically mimicking a human invention, like a hang-glider or a blimp, with modifications to suit the materials available.  The third approach would be to cook up a brand new way of making a low-velocity diaspore using the materials provided.

But no matter which approach you use and regardless of your design's success (your brilliantly designed diaspore may in fact drop like a stone), the way in which your diaspore morphology came to be is different from the way real diaspore evolution takes place, and this difference can be summed up in one word:  teleology.  From the start, you knew what needed to be done (minimize terminal velocity), and you directed your planning toward meeting that end.  The diaspore you made was "purposefully designed."

Natural selection operates on a fundamentally different mechanism that starts with variation on an existing morphology.  The differences among "this generation's diaspore morphology" are filtered, presumably with trees that have poorly dispersing diaspores (higher terminal velocity) having lower reproductive success relative to trees that disperse seeds more distantly (lower terminal velocity), and if this trait is inherited, the next generation's diaspores will have on average a lower terminal velocity.  The variation itself, arising through mutation and genetic recombination, is not directional.  "Good" variants that reduce terminal velocity are actually far less common than variants that cause the diaspores to fall more quickly.  If the population is to evolve toward "diaspores with lower terminal velocity," it must occur over many generations of this process of sorting through random variation and selecting favorable variants.

Question:  is this example of evolution by natural selection a random process or a directional process?  Directional!  If diaspore evolution occurred in a random direction, the result would almost certainly be poorer dispersal by faster-falling diaspores.  But is there a "purpose" to this directionality?  [Here's where we get into the fine semantics of evolutionary epistemology.]  There is no purpose to the incremental evolutionary movement toward morphologies with lower terminal velocities.  I say this because the word "purpose" implies "intent," and there is no intention on the part of the plant population to systematically slow diaspore descent rates.  Evolution occurs without purpose, yet in a direction toward greater levels of adaptation!

Both purposeful design and natural selection can yield highly functional diaspores.  If you are faced with a natural diaspore (like that of Tipuana tipu), is it possible to distinguish between competing hypotheses of A) it came about by natural selection and B) it came about by Intelligent Design?

Richard Dawkins in his book Climbing Mount Improbable suggests that hypothesis A could be refuted by showing that the adapted morphology would not be achievable via a "thousand little steps" in morphological evolution from its ancestral state.  Each of the thousand micro-steps would have to be a functional improvement resulting from natural selection's sorting through random variation, as described in a paragraph above.  If you were to find that there is no way for natural selection to have achieved the highly adapted morphology, then it figures that the Intelligent Design hypothesis looks all the stronger.

But is it possible to refute hypothesis B?  Well, to definitively disprove the intervention by an Intelligent Designer is not possible, at least not by any scientific methodology.  If, however, it turns out that every single one out of the millions of highly adapted states that we see in the biological world is explainable via the "thousand little steps," then what?  This is, in fact, the state of affairs in biology, and in his book, Dawkins describes some of the apparent "masterpieces" of the biological world (including the use of sonar by bats, the vertebrate eye, the spider's web, the life cycle of the fig wasp) and makes a compelling case that the natural history of each of these awe-inspiring marvels is entirely consistent with arisal through the "thousand steps" scenario.

Getting back to our lab now, we shall be taking the five designs with the longest "hang times" and embark on a hypothetical discussion that will follow the basic tune of:  if this diaspore was a naturally occurring one, would it be explainable via the "thousand steps," or would it constitute the kind of disproof that creationists are so desperately in need of?

Literature Cited
Green, D. S. 1980.  The terminal velocity and dispersal of spinning samaras.  American Journal of Botany 67:  1212-1224.
Guries, R. P. and E. V. Nordheim.  1984, Flight characteristics and dispersal potential of maple samaras.  Forest Science 30:  434-440.
Howe, H. F. and J. Smallwood. 1982.  Ecology of seed dispersal.  Annual Review of Ecology and Systematics 13:  201-228.