Yes, I am very sorry for the rush at the end, and I really would have liked some more time for both of us to describe what we observed.
Long and short of it is that at this point we have data but we have not done any analyses to see if there is any significant differences in the growth, reproduction or mortality of the corals, fish, echinoderms, gastropods, sponges, octocorals, zoanthids, corallimorphs, algae or other species compared to the control.
What we presented were some interesting things - such as what Frank mentioned... finding various micro and turf-algae, copepods, and other microbes in tanks triple washed with autoclaved substrate, purified water and salt. We saw clear trends in some of the salt brands in terms of turf and cyanobacteria continuing to thrive in some salts and not in others. We saw very clear differences in coralline coverage and foraminiferan growth and variations in successional species. I mistakenly kept saying bryozoan rather than foraminiferan in the talk...my bad. We also saw, in some cases, very heavy undeisrable algal growth month after month even after removing it after each water change despite very low measurable inorganic nutrients, and very little undesirable algae growth in some cases where inorganic nutrients were quite measurable. The difference, as Habib Sekha rightfully pointed out, is probably the unmeasurable organic components, but then the question is why the difference in the dynamics of nutrients varied given the same treatments.
Because the study literally ended the day before the conference, there was no way to analyze the results. I will say that, at least from being familiar with the data and the appearance of the tanks and the survival and appearance of the species that there are certainly some salts I would prefer to use in my tank and some that I probably wouldn't use again, even though I have used them for many years without any obvious negative effects. I have a feeling that the reason these differences are not observed by those using these salts, myself included, is that the complexity of the reef tank community is able to mitigate the good and bad aspects of the salts that became apparent in a more controlled environment.
BUT, and I STRESS BUT, the results once analyzed might very well show there are no significant differences between the salts and the control. It appears from a cursory look at the data that at least for some species there will be a significant difference. If there is, then I think it will be interesting to go back and run replicates of those tanks and species and look more carefully at what factors might be at work to cause the differences, either better or worse.
Anecdotally from our observations and the overall appearance and experience with the nine salts tested - in other words, if the average person came into the lab and had to pick a tank they would want to take home - they would probably throw Tank J (Instant Ocean) away because of the continuous heavy cyanobacterial growth with every batch, and would probably choose Tank F (
Red Sea) because of the heavy coralline growth, booming amphipod population (I still have to count all of them!!!), tight ball of Chaetomorpha, and overall apparent health of the species. Yet, some of the other tanks had reproduction that exceeded Tank F in some species, and concurrently had less that other salts, too. The control was not without mortality, either, and had relatively low algal growth (corallines and turfs/cyano) but was the first to show strong foram growth.
Basically, there are a lot of things that happened, lots of data, pluses and minuses in each tank, and until we do the analyses, there really can't be an informed answer as to what salts are great or not so great.
Frank, I appreciate your offer. Rob Toonen and I spoke quite a bit last night as to how to approch the statstics, and I would like to talk with you about it as well. There are likely limitations to how much we can actually say and I need to think about how best to approach and get as much information out of what we did as possible. Some things are straightforward, others not so easy, and some may indicate something significant that will require more testing and replication.
The good thing is we have samples for further analyses and I would finally end this post by saying this is far from the definitive study. I hope it will be a platform that can be used for further work, comparison and discussion.
