Hi Gipo-

I don't have any references at hand (but I must have some somewhere), but I admit that I don't like the idea of evaluating FEEDING preference by taking into account DIGESTION-dependent processes:

Of course, the amount of feces correlates with the amount food eaten which, in turn, correlates with preference, palatability or whatever - however, different food sources can significantly differ in how well they are digested: the better they are digested, the less feces are produced. Thus, a highly preferred food source that is very well digested (actually, this is probably the case, as we expect a strong relationship between ingestion rate and digestion, although we don't know how the feedback works) will produce only little feces - this would heavily underestimate the preference for this food source...

There is better ways of estimating preference! (and, btw, take care that you don't use ANOVA or such for statistical analysis, but chose monte carlo-techniques)

Cheers,

and good luck,

Martin

Thanks Martin, as usual your answer is very useful and interesting.

If you find some papers please share it. I'm reading now your paper on nutrition of terrestrial isopods (with evolutionary point of view).

I also think that some species are efficient consumer while other are good assimilator. Maybe in the first (and proved) case also the number rate (per day) of faecal pellets can be a measure of preference. Just preference I mean... otherwise, in your opinion and with your experience, what is the best method to evaluate food choice?

Thanks again, ciao!

Gipo

PS See you in Budapest!

Gipo - the very best and most accurate way is still to offer different food sources of exactly known dry mass (i.e., you have to dry and weigh the food, and then re-wetten (which, of course, affects the microbial biofilm and thus, potentially, the palatability and digestibilita); after the experiment you dry the remaining food sources again (by exactly the same means) and weigh again: the difference is the amount of each food eaten.

If this method is not feasible, an alternative is to measure the area of each food source eaten (accuracy depends on the differences in thickness and mass per area unit among different food sources) (e.g. mm2), when you use pieces of exactly known size (circles or rectangles)...

In any case, I suggest to present data as % of each food source eaten of the total amount eaten: i.e., you determine for each experimental unit (individual or group of individuals) how many percent of the total amount eaten was chosen from the different food sources, and then calculate the mean/median and the variation...

Hope that helps...

Martin

p.s.: as you know, I am not working on isopods anymore - hence, there is no way for me to attend an isopod symposium...:-(

Thanks Martin... I was thinking at the second method you've proposed.

With the software ImageJ it is possible to calculate exactly the areas (before and after) of the food we can offer them.

So, I'm right: your answers are always useful and interesting!

Ciao,

Gipo

PS It's a real pity... you are one of the best member of our isopod family. We'll miss you a lot. :-(

Sofia... It was a good choice to ask!