In my previous article entitled "The Sunday Scientist," I described my trials and tribulations with the first lab assignment in Dr. Ron's online sand bed ecology class. It should come as no surprise that the other lab assignments were equally challenging. I know that for both students and instructor alike, this class required a substantial investment of time and effort. I, for one, feel the time was well spent and the effort worthwhile. So little has been done to quantify the processes occurring in captive marine environments that any contribution to this knowledge base has value. I would like to encourage my fellow reefkeepers, especially the experienced aquarists, to repeat our experiments in their own tanks. The larger the sample size, the more conclusions we can draw from the data. No matter how long you've kept a reef tank, the results of some of the tests may surprise you, and the wonder of discovery is a reward unto itself.

While our first class assignment was to define the physical parameters of our sand bed by determining the grain size distribution, (this is the infamous "Sand Castles 1A" you always heard they offered in college), the second lab assignment was designed to determine the environment's chemical parameters. The plan was to use a syringe with a fine needle to extract water from the sand bed, and test that water for ammonia, nitrite, nitrate, phosphate, pH, dissolved oxygen, sulfide and copper. Samples were to be collected from the overlying water, and then at two-centimeter increments down to the bottom of the bed. Since my sand bed's depth averages ten centimeters, I needed six water samples for each test. Each chemical was to be measured in three different places in the sand bed. Eight chemicals, six samples and three replicants worked out to 144 water samples ranging from three to ten milliliters. (Thank goodness I didn't go with a six-inch sand bed!) While the sheer number of samples seemed daunting, using a syringe and needle to extract water from the sand bed didn't sound like it would be that big a deal. (If only I had a dime for every time I've thought THAT since I've been in the hobby!) I acquired a selection of syringes and needles from a pharmacist friend and went to work.

It Ain't Easy

After my dismal failure at one-handed sand coring from my first lab assignment, I decided to practice the needle and syringe method of water extraction first in the refugium, where I was free to use both hands. Once again, I was confounded by the really fine sand in my system. While it makes a great sand bed, it was impossible to suck water through it with a needle. With smaller needles, I couldn't pull hard enough on the large syringe to get the water to flow in. Larger gauge needles clogged with sand. Even if I had been able to find a syringe and needle combination that would work, I still would have had to snorkel in the tank to get both hands to the sand bed. (And, yes, for those men in the audience, I did ask my husband to give the syringe a try. While he can easily open any jar for me, he couldn't get water into the syringe either).

I learned the lesson, "If at first you don't succeed, use a bigger tool" from watching my dad work in the garage when I was a kid. So, my husband and I came up with the brilliant plan to use a hand pump, like the one used to bleed brakes on motorcycles, to create enough suction to get the water into the syringe. A couple of hours of work with rigid airline tubing, air hose, pieces of syringes, and assorted necessities such as tape and superglue, yielded a device capable of exerting 45 pounds/square inch of suction on a piece of rigid airline tubing with a needle attached. While it was truly an impressive device (of which MacGyver would have been proud), it was nonetheless utterly useless. Once again, I found myself sitting at the kitchen table with my head in my hands. (Just in case you were curious, the way the real scientists do it is by taking core samples, flash freezing them, cutting them into pieces and then extracting the water with a centrifuge. But, alas, after purchasing all the test kits for class there was no money left in my science budget for such things, so a less expensive solution had to be found.)

A suggestion from a friend helped lead me to the ultimate solution to the problem. I ended up using a siphon made from the smallest diameter rigid airline tubing I could find. I made marks on it every two centimeters and stuffed a wad of cotton into the end of the tubing. I attached an air hose to the opposite end of the tube.

My high-tech water collection device. The cotton stuffed into the end of the tube slowed the flow of water and kept sand out of the sample.

By pushing the rigid airline into the sand to whatever depth I needed, I used the natural siphon effect to draw water slowly into the tube. The wad of cotton helped slow the flow of water and keep sand out of the tube. I used the air hose like a straw to apply suction when needed, or provide backpressure if the water was entering the tubing too quickly. It was important to collect the water samples as slowly as possible. If they collected too quickly, I would have been collecting water from levels above or below my target level. Each milliliter took two minutes to collect. My rough calculations say that I extracted approximately 600 milliliters of water at two minutes per milliliter for a grand total of 20 hours. I spent another five hours actually performing the tests.

The tests themselves were a challenge, as any of you who has tried reading colormetric tests knows. I carried my trays of test tubes from room to room, from lights to windows trying to discern ever-so-slight differences in shades of pink, blue and yellow. If it sounds to you like this process was less than exact, you are absolutely right. (Fortunately for me, analyzing the data and compensating for errors were Dr. Ron's problems!) In all, I know it took me a good two weeks to complete the chemical analysis.

I found it easiest to first collect all the water samples for each test, then perform the test.

The chemical analysis of my sand bed yielded a couple of interesting results. First, conventional wisdom held that the lower level of my sand bed would be an anaerobic zone. However, my tests showed a value of 5ppm dissolved oxygen at the very bottom of the bed, a depth of 10 cm. Second, I found only small, isolated patches of hydrogen sulfide that tended to be relatively close to the sand bed's surface. I had expected to find a layer of anoxic sediments with hydrogen sulfide at the bottom of my sand bed. I was pleased to learn that the ammonia, nitrite and nitrate levels indicated that my sand bed was functioning as expected with regard to the nitrogen cycle. In fact, it seems to work just like the aquarium books say it does; imagine that!

The Fun Stuff

Finally! With the mechanical and chemical stuff out of the way, we could get down to the business of counting animals. For the first week, we conducted a microscopic safari. Several of my classmates had not attended Dr. Ron's class in invertebrate zoology, so they didn't really know how to identify the animals we were going to be counting. We all needed to agree on our identification of the animals if our counts were to be compared; hence, the safari. I used a plastic bulb-tipped pipette to extract small quantities of sand from my tank and searched for critters. The idea was to compile a photographic catalog of the animals with at least a genus-level identification. The result was the geekiest scavenger hunt ever, with everyone submitting photos of worms, pods, mollusks and even mites. Dr. Ron provided several animal identification keys so we could put a name on all of the animals we found.

With an identification guide to work with, the real work started. In order to estimate the population of the various species of animals living in my substrate, I had to collect and count the animals in a defined area. The first sample I used was a one-inch diameter circle. To get the proper sample area, I cut a piece of clear tubing from a package of underwater epoxy, placed it into the sand about an inch deep, then carefully siphoned the top one centimeter of sand from within the tube into a glass beaker. I used an eyedropper to transfer small amounts of sand and water into a Petri dish. Only one layer of sand grains could be put into the Petri dish at a time, otherwise animals would be obscured by the substrate. (I had twelve Petri dishes to count in my first sample.)

I had made a one-centimeter grid on a square of clear plastic and had numbered each square. The Petri dish was placed on top of the grid on the lighted stage of my dissecting microscope.

Using thirty times magnification, I examined each grid square and tallied the animals. My first sample took three and a half hours to count. I found three hundred larger animals including nematodes, copepods, ostracodes and segmented worms. I counted at least five hundred large ciliates. I had to ignore the small ciliates and sessile animals such as forams or I could have been counting all day!

One of the major challenges of counting the animals was their stubborn refusal to stay in one square. Copepods were the worst! They bounced around the Petri dish like fleas on a hot plate. I had to keep switching between the ten-times and thirty-times magnification to try to ensure I hadn't counted any copepods twice. I'm sure my conservatism meant I ended up underestimating the number of copepods in my samples. (Blame it on too many years as an accountant).

The lab assignment called for a total of five counts. In addition to my main tank, I had done some of the prior lab work on my refugium, so I wanted to include a couple of counts from it for comparison. Taking six more samples at three and a half hours each was a little more than my schedule would allow, so I reduced my sample diameter from 25.4mm (1 inch) to 15mm. Even with the smaller sample size, I was still finding an average of 156 larger animals and it took about two and a half hours to get through each count. In our reading assignments for class, we discovered that in real field surveys of benthic communities, the sample sizes were one square meter. (Of course, the folks doing such a survey would not have tried counting the animals alive.) Fortunately for my animals, most were returned unscathed to their homes. I must confess, however, a few unfortunate individuals did not survive my clumsy attempts to pin them under a cover slip so I could take their picture. (And you thought it was hard to get a two-year old to hold still for a family portrait!)

Here are a few of the most interesting animals I found in my sand bed.

This mite is a relative of the spider.
Here we have a newly settled snail.		This worm is budding off another worm. The arrow shows the point where a new head and palps are forming.

For those curious types, these pictures were taken through my Olympus BH-2 microscope with an Olympus E-10 digital camera. I found the animals with the dissecting scope, then used an eyedropper to transfer them to a depression slide. Most of these pictures were done at between 100X - 400X magnification.

The results of the counts from my display tank are as follows:

After averaging the samples and converting the area, my average density was 919,922 animals per square meter (excluding ciliates). I was impressed!

What Does It All Mean?

I know that many of you are curious to know what conclusions (if any) can be drawn from the data collected in our class. There are those of you who will ask, "What possible conclusions can be reached with such a small and imprecise set of data?" I wondered that myself, and, like you, will have to wait to see what Dr. Ron can make of it. I do believe, however, that if we had data from one hundred tanks, instead of just ten, the data would be much more useful. So, how about it… are you game?