#24-6 The Lost Art of Scientific Tinkering
In which the E@L supplements the scientific method with good old trial and error
Scientific research has come a long way. In the pre-industrial era, the study of the natural world was considered “Natural Philosophy”. Its practitioners conducted their work either by observing and the world and trying to explain it (think Newton or Darwin), or by tinkering with stuff until they got it right (think Galileo). It wasn’t until the mid 19th century that scientists began to conduct carefully controlled experiments and applied mathematical analysis (think Mendelian genetics). Since then, all scientific progress has depended on ever-more sophisticated statistical analysis of data.
As an ecological scientist, I learned and taught my students to conduct carefully defined research projects. Each study had to have at least one specific hypothesis that could be tested and either refuted or accepted. Organisms had to be randomly selected, and experiments had to include control groups and multiple replicates in order to rule out chance probabilities and unknown factors. Multiple variables could be explored if each was accounted for in the statistical analysis. This usually resulted in a highly robust scientific study, even if it didn’t support the hypothesis being tested. But even now, that’s not how all scientists work. Some just tinker until they get it right.
Giant Shrimp
Dr. Ling Shao-Wen was a pioneer in the development of aquaculture. Born in China in 1907, he studied there, and received his PhD from Cornell University. Following WWII, he was the director of the Chinese National Fisheries Institute, and in 1949 was appointed by the United Nations Food and Agricultural Organization as the first Regional Officer for Southeast Asia. During the 1960s he worked in Thailand, Malaysia, and Sri Lanka, where he developed aquaculture methods for the giant freshwater prawn, Macrobrachium rosenbergii. In the late 1970s, he became an adjunct Professor at the University of Washington, where he taught an aquaculture class in which I was a student. His lectures were published in a book along with many of his own drawings [i].
One day in class he told us how he had discovered the lifecycle of the giant freshwater prawn. As he began his talk, he dipped a chalkboard eraser into a dish of water and powdered chalk and swiped it across the board; a dark streak appeared, which slowly dried to reveal a white stripe.
Dr. Ling told us that he had been trying to culture the prawns in his laboratory for many weeks, without success. At each pause in his talk, he made another wet swipe on the chalkboard; this one slowly turned into a white shrimp claw.
He began his research by placing male and female prawns in a tank filled with fresh water, allowing them to mate, and then raising the larvae. The adults thrived in his tanks, did their reproductive duty, and produced eggs.
More dark swipes on the board, which turned into antennae.
After a few days, the eggs hatched into larvae, but the larvae died within hours. Every day, Dr. Ling repeated this experiment, trying to find a way to keep the larvae alive, but they kept dying. There must be something missing, he thought, perhaps some food item. One day while eating lunch, Dr. Ling added a bit of rice to the tank, but the larvae still died.
Two swipes on the chalkboard became eyestalks.
The next day he added some egg, but the larvae still died.
Now, legs and an abdomen appeared.
The next day he added some fish; the larvae lived a little longer, but still died. The next day he added a few drops of soy sauce. To his surprise, the larvae survived for a day, but still died.
Large wide swipes of the eraser morphed into the body of a prawn.
The next day he added more soy sauce, and the larvae lived several days. After several repetitions of this, Dr. Ling concluded that the larvae must require salt water to survive. That implied that the eggs or larvae must drift downstream toward the ocean, where they would encounter brackish water. Later as juveniles, they made their way back upstream, where they lived as adults.
As he revealed his epiphany, he made a few final strokes with the eraser, and the form of the giant prawn appeared before our eyes.
Discovering this secret about the prawn’s lifestyle allowed Dr. Ling to cultivate the prawns through their full life cycle in the laboratory and became the foundation for the world-wide aquaculture of giant prawns. And his revelatory story impressed his students forever. Lucky for me, I just happened to have a new camera with me and was able to capture this moment on film.
Dr. Ling didn’t know statistics, and he didn’t conduct random, controlled, experiments with replicates. Most of what he achieved was by tinkering. Trying a bit of this, and a bit of that, and using his own judgement to determine if the result was due to something he did. And then building on that, piece by piece. Very few of us (scientists) work that way anymore. The requirements for publishing research, and for obtaining grants, mean that every step must be spelled out in advance and justified statistically. Otherwise, your research will be shredded by reviewers, and you might not get funding to do more. But we often tinker in our laboratories; sometimes it’s the only way to develop a hunch into a real idea.
From the Lab to the Engine Room
Lately I’ve been spending a lot of time tinkering in my engine room. That’s a bit of a stretch though because I don’t actually have an engine room, just an engine, in a sailboat, with barely enough room around it to work. Sailboats are complicated beasts. They include all sorts of complicated rigging and sails with specialized names, and the bigger they get, the more complicated they are. Mine includes lots of pumps and plumbing, wiring and winches, sails and switches, cleats and clutches, and various other gizmos, all of which need to be understood, examined periodically, and maintained or replaced. Unlike a car, if something on my boat breaks down, I can’t walk away from it and call a repair shop. A boat is a life-support system, and a breakdown could be life-threatening. So, it pays to understand how every part of it works. I have only ever bought used boats (because I can’t afford a new one), and I know from the outset that things will need repairing, so I spend a lot of my time finding out what they are. As I discover problems, and repair them, I learn details about how the parts work that I would never know if the boat were new.
Since this is my first boat with a diesel engine, and that is the most expensive part of the boat (after the hull), I took a diesel repair class to learn how it works. Then I undertook the task of inspecting, repairing, or replacing every part of it that I could get my hands on. Although it was “made” by Universal, it is actually a Japanese-built Kubota tractor engine, so most of the bolts are metric. Most of the exterior components, however, like the fuel pump, alternator, and heat exchanger, are American made so use English (SAE) type bolts. It is maddening to discover that simple jobs, such as adjusting the V-belt tension, require three different wrenches: 12 mm, ½ inch, and 5/8 inch. Especially when I am in a contorted boat-yoga position and can’t reach my toolbox.
By now you are wondering what this has to do with ecology or prawns. Truth is, I find that learning about a diesel engine, and refurbishing it, engages the same mental pathways, and requires the same skills and strategies that I used as a professional ecologist. The major concepts are similar. As an ecologist or a sailor, I start by looking at a system, which could be the reproductive biology of the king crab, or a diesel engine, and learning all the technical jargon that I need just to describe the thing I am studying. Whether it is a sperm injector or a fuel injector, I need to be able to identify the individual parts of a system. Then I try to understand how it works. As I learn more about it, I begin to see problems, or gaps in my knowledge. Then I begin to ask questions. Much like Dr. Ling did.
How does a larval king crab select a place to settle down? How does my engine know when to circulate cooling water? In the laboratory, I can hatch the crab larvae and observe them swimming in a tank. In the boat, I can fire up the engine and watch as it operates and engages the cooling system. I read what other people have written, what they know, and what they suspect, but don’t know. Then, I narrow my focus to examine specific aspects of the system I am studying, whether diesel engine or crab behavior.
Bento Boxes and Thermostats
In 1996, I spent a year in Nemuro, Japan, working in the laboratory of Dr. Jiro Kittaka, another aquaculture pioneer, where we were trying to develop cultivation methods for king crabs. From previous observations, I knew the crab larvae only occurred on specific substrata, usually other living organisms such as hydroids, bryozoans, mussels, or worm tubes, but they were never found on bare sand. The commonality among preferred substrata is that they are all highly branched, or structured. Do the larvae end up in these locations by accident, or as a result of predation, or do they make a specific choice? I decided to provide several alternative habitats for the crabs to settle on.
Designing an experiment in the laboratory was a kind of tinkering. How could I determine if crabs chose a substrate by accident or on purpose? What would prevent them from changing location after settlement? What type of apparatus would help me answer these questions? Perusing the local department store, I found little plastic bento boxes for storing food. These gave me an idea; I won’t give my lunch to my larvae, but maybe I can put my larvae in lunch boxes. In the laboratory, I placed several boxes in small 10-liter tanks, with a different substrate in each box: sand, gravel, or a 10 cm square of aquarium filter mesh (as a proxy for hyroids). Some tanks had only one type of substrata, and others had all three. Then I came up with a testable hypothesis: I predicted that my larvae would choose one of three possible substrata. If the larvae settled on a substrate and liked it, I reasoned, they would stay. If not, they would have to swim to another box. They could not move from one substrate to another accidentally or randomly.
In my boat, I have been tinkering with the engine. I knew my engine should be responding to feedback from its thermostat, but I wasn’t sure if it was. I thought my temperature gauge was reading low, even though the engine was quite warm. Based on this, I hypothesized that either my thermostat or the temperature gauge was not operating correctly. To test this, I replaced the old thermostat, along with the sensor that sends temperature information to the gauge and the alarm.
In the lab, every day for two weeks I checked the position of all my crab larvae. In tanks with only filters or gravel, all the larvae had settled on the substrata by day 2, whereas in the tanks with only sand, 60% of the larvae were still swimming after 10 days. In tanks with multiple substrata, 70% of the larvae were on the aquarium filters, about 20% chose the gravel, and a few were still swimming. None of them settled in the box with bare sand. To me, that was a clear message. The larvae settled among complex substrata immediately but tried to avoid settling on sand. Where multiple substrata were available, larvae specifically selected those with more complex structure, and rejected bare sand. This makes sense because sand provides them with no shelter from predators. The crabs had evolved to select habitats that would give them the greatest chance of survival. I did not know yet whether their choice was based on purely tactile sensation, or whether there were other cues involved, but I would discover that later. When we finally published our work in 1998, it was the first experimental study on king crab settling behavior [ii].
In the boat, I learned that the new thermostat and sensors were now sending correct information to the gauge and engaging the cooling system of my diesel engine properly. This mechanical system has also “evolved” since it was first invented a century ago, to operate smoothly with simple electro-mechanical feedback loops. I did not know yet whether the thermostat or the temperature sensor was at fault, but I would discover that later, too.
Sometimes science is all about randomized, controlled trials with double-blind experiments. But sometimes it is just about tinkering. Finding out what works, and what doesn’t. Like Dr. Ling.
From Science to Sailboats
As a retired scientist, I miss going out in the field with my students, working on boats or in the laboratory, and seeing and discovering new things every day. This is especially true in winter when the weather keeps me from exploring outdoors. But my scientific drive, my need to know how things work, is still active. So now I’m sating my scientific appetite with another activity – exploring the functionality of a 25-year-old sailboat and its associated parts.
These pursuits keep my mind and body engaged in intellectual as well as physical activities. I no longer need to accumulate publications for my resume as emblems of professional competence; instead, my accumulation of bruises and busted knuckles from boat maintenance gives me cred at my local yacht club. I used to lie awake at night worrying about how to solve some problem in my research; now I spend those sleepless hours trying to solve some problem in my boat.
Soon it will be time to climb up out of the dark engine room into the summer sunlight, hoist the sails, and start exploring the water again. Then I can tinker with the sails all day. Eventually I’ll get it right.
This issue of Ecologist at Large is available to all readers. However, if you would like to support my work with a one-off contribution, click “Buy me a coffee” below.
[i] Ling, S-W. 1977. Aquaculture in Southeast Asia, A Historical Overview. Washington Sea Grant and University of Washington Press. 108 pp.
[ii] Stevens, B. G., and J. Kittaka. 1998. Postlarval settling behavior, substrate preference, and time to metamorphosis for the red king crab (Paralithodes camtschaticus). Mar. Ecol. Prog. Ser. 167:197-206.
Thanks for this thoughtful piece. Tinkering is vital and I don’t do any at the moment. Must have a go at something.