Testing a placebo blocker

Oooo!  This really is a hard experiment to design.  Placebo effect is real, and this xkcd cartoon posted nearby at the Curious Wavefunction hints at some of the complications. Warning: science nerds only; ordinary people may not find this funny. 

Inconclusive apple pollination experiment

Poor experimental design is the reason that the results of the apple pollination experiment were inconclusive. However, some pollination did occur because some young fruit are developing on both varieties of columnar apples. So whether the one variety is self-fertile remains a question. Yes, it could have been netted, and self-pollinated by hand, but then if not self-fertile it would have produced no apples at all. Botanical curiosity will go unsatisfied, but a few apples will be produced. As these are brand new trees, although of good size (6-7 feet tall) only a few apples will be allowed to grow.  The new trees have produced a very nice crown of leaves and if we can keep the Japanese beetles from defoliating them, they should get off to a nice start. Some nice bridal veil will serve nicely in that regard.  Mrs. Phactor's pear tree also has quite a few developing fruit, and it's about time it got going with the fruit production.  Perhaps the basal pruning administered to the non-producing apple tree as a lesson had some effect. The mild wet spring has also been producing some excellent lettuces, and just about the time the crop becomes over-whelming, they will begin to bolt and that will be that until August when it's time to replant. Herbivores have helped some.  The bibb lettuces planted in a parsley bed, and then fenced to no particular benefit, were just getting big enough to be harvested when they were eaten to the ground, although their tastiness spared the parsley from devistation. That kind of gluttony usually indicates a woodchuck rather than a bunny, so our watchfulness must increase, but no confirmatory sightings as yet. Well, that got rather off topic quickly.

Filthy dirt

Experimental work with plants takes some planning and some time.  Nothing magic about plant growth so you just have to wait, but time to get a glasshouse experiment going.  And so an energetic student ready to get underway with the next experiment asks, "Where do I find the dirt?"  Thus they innocently gave voice to a common misconception.  TPP answers, "You find dirt under your finger nails, and in certain kinds of books, movies, and websites."  "Perhaps you mean soil?"  Soil and dirt are not the same at all.  "Does that mean my research proposal has to be changed?"  "Yes."  Careful consideration has to be given to the soil used for experiments.  Sometimes you can use a commercial potting mix, or just make one, but a potting mix is not at all like the soil these plants naturally grow in.  Soil is more than just a whole bunch of eensie weensie rock particles and organic matter, it also contains microorganisms to the tune of about 8-10 million per cubic centimeter (for the non-metric, a volume about the size of a sugar cube).  Not only that, but the plants growing in a particular area can greatly alter the soil microflora to their benefit and perhaps the detriment of others, and in particular that includes invasive plants.  This produces some problems in the experimental design, but let's let that go for now.  Fortunately your mentors were thinking way ahead and soil (sod actually) was dug 2 months ago and set aside in anticipation of this experiment; soil from our research prairie, soil affected by an invasive plant, and soil from an adjacent grassy meadow, so a very practical consideration is next.  Those buckets with great big old chunks of sod need to be screened to remove rocks and root systems and in the process reducing the big old chunks into a much finer texture.  It's a great student activity for a cold winter afternoon.  So, no dirty thoughts, just become one with the soil.  Wonder how many hits this title will generate of people looking for real "dirt"?  We'll see if the data for click ins and outs, i.e., zero time spent reading the blog goes up.  Something like this happened a week ago when the daily traffic at TPP jumped up by a factor of 3 for no particular apparent reason producing quite a blip in the data.

Experimental design

Field work is generally very time consuming, but for many of us that is our laboratory. This week the effort is to add a new wrinkle to a long term experiment. Central to all of this is a hemiparasitic plant, green and photosynthetic, but an obligate parasite on surrounding plants. So if parasitic, why be green; and if green, why be photosynthetic? Part of the answer is nutrients, particularly nitrogen, in a limited environment, which is what this hemiparasite obtains from its hosts. In the process it further limits their growth, but the additional light may improve its photosynthesis. So shade was added, nutrients were added, and the hemiparasite removed (several prior posts have bemoaned this treatment), and of course, all the possible combinations of these three treatments. After three years there have been changes, and now the invertebrate population is being sampled to see under what conditions their numbers, species diversity, and ecological roles have changed. Sounds like fun doesn't it? But sticky traps are a devil to deal with because they are really, really sticky. So soon an answer may be forthcoming. Predictions: the hemiparasite's presence with increase diversity; additional nutrients will reduce diversity. If only data weren't so hard to get, this would all be fun.

Science in theory and practice

My current field work deals with the ecology of a hemiparasitic plant, which is both green and photosynthetic and an obligate parasite. To try and determine why its both green and parasitic, and test our evolutionary stable strategy hypothesis, which differs from the switch-hitter and way-station hypotheses, my collaborator and I designed an experiment. The simplest and most logical treatment was eradication of the hemiparasite.

Hey, sounds so simple "in theory", but in practice its turning out to be a pain, in the back. The hemiparasite is a very successful on our prairie study site, i.e., numerous, and what with other treatments and all combinations, together with replication, the subject of this research must be eradicated from over 50 one-square-meter quadrats. And that alone is not enough because of the root zone (it's a root parasite), so a 50 cm wide buffer zone must be cleared of the hemiparasite around the study quadrat. That's a total of 4 square meters per quadrat! Under good conditions and working quite hard, two of us can do 3-4 quadrats an hour. This amounts to an awful lot of dirty, down-on-hands-and-knees work just for this one treatment. The picture shows the high-tech equipment used in plant ecology, and the recently removed hemiparasitic plant.

This is the problem with field work. You make sound, logical decisions, and get started and it's like grabbing a tiger by the tail. You can't let go or you lose everything, your whole investment. And it's a five year experiment so the treatments must be checked and redone every year.

But that's the way science goes, and what it takes to demonstrate if your ideas are correct and the old ideas wrong. The effort and time explain why so few long term field experiments are done. And I do not recommend such work to young colleagues who lack tenure.

And it's not the work I mind, but just that it's at ground level, and it's much further down, and back up again, than it used to be. Of course, the solution is to have a squad of young backs while I sit back in my research director's chair and point to the next quadrat. Any volunteers?