Dietary Experiment with my Pyxie Froglet

[Goomba]

I have had Ham, my Pyxie Froglet, for a month now. He was about 3/4 of an inch when i got him, a total baby. Now he is a tad over 2 inches SVL. He eats crickets like crazy, and yesterday he got his first pinky (I posted the video on here).

Anyways, I wanted to hear some thoughts on what I am considering doing.
I am thinking about getting 50 frozen fuzzy mice for the bugger (since he still takes prekilled, unlike Goomba). I wanted to see if a staple of mice coupled with powdered suppliments every other feeding and decent heat providing accelerated metabolic rates will lead to faster growth.

I've done alot of research on corneal lipids, etc, and I think that alot of it is crap, or at least circumstantial. I believe that if ample water and heat is provided for digestion, calorically dense mice can serve as a hardy meal. I've seen many healthy Pyxies on a staple of mice, so I figured I'd give it a shot for a few month. Goomba will remain on Roaches, mice, and occasional crickets. But I think having this fresh slate would give a valuable insight into how these frogs grow and the impact of diet, etc.

I am thinking of starting the experiment in a week or two, till then I'll plump him up with dusted crickets. When I start, I want the feeding schedule to go like this:

2 weeks: Feed one fuzzy a week, with a couple dusted crickets.

3 weeks: Feed 2 fuzzy mice a week. Dust every other mouse

4-8 weeks: Feed 3 fuzzy mice a week. Dust 1/3 of the mice.

And if he gets big enough, increase the frequency of meals per week, etc.


What do you folks think?

 

[skinheaddave]

In principle, it is a good idea. There is, however one major flaw.

You have only one subject in your experiment (and the potential of a control group of one, I suppose). That means that regardless of the results, you will be unable to draw any significant conclusions -- except to say what worked for your particular subjects.

Cheers,
Dave

 

[Goomba]

Haha, this is very true. I am not doing an indepth imperical study, rather I will base it by eye and in comparisson to Goombas growth as a froglet. I wish I had mroe froglets, hehe.

 

[JLDomestics]

I have seen corneal lipidosis in horned frogs and african bullfrogs that were fed too many mice. I have also seen corneal lipidosis in tree frogs that were on a diet of only crickets. It seems that if the crickets are consistently fed a high calcium diet and then they are dusted with calcium before being fed, after a while the calcium intake becomes too high and corneal lipidosis results. This cricket diet can be corrected by not feeding the crickets such a high calcium feed and not dusting with calcium. However, a diet of mice can not be corrected as mice are consistently high in calcium! Surely you can feed your frog mice but maybe just keep an eye on how much calcium it is getting in its diet as well. If you are feeding mice, then you don't need to feed calcium dusted crickets. You should gutload your crickets with carrots or potatoes or something that is fruit/vegetable and that doesn't contain large amounts of calcium. It should also be noted that the majority of animals that develop corneal lipidosis are those that are kept in unsanitary and unacceptable conditions, so they were more prone to stress.

 

[AnthrpicDecadnc]

In principle, it is a good idea. There is, however one major flaw.

You have only one subject in your experiment (and the potential of a control group of one, I suppose). That means that regardless of the results, you will be unable to draw any significant conclusions -- except to say what worked for your particular subjects.

Cheers,
Dave

well, he doesnt have a control group, as there is no other frog to compar his results too..

you'll get results but you cant compare them to anything to make a good analysis. score another pixie froglet, and try it out. it would be an exremely rough comparison, as every animal is different. don't you think it would be better to try it on a population of pixies? that way the results would be more... general i guess is the word?

 

[nickbachman]

I have also seen corneal lipidosis in tree frogs that were on a diet of only crickets.

I think everyone's seen that picture.

 

[skinheaddave]

well, he doesnt have a control group, as there is no other frog to compar his results too..

As I understand it, he has "Ham" and "Goomba," with the later serving as the control -- sort of.

Cheers,
Dave

 

[JLDomestics]

I think everyone's seen that picture.

Im curious as to what picture you are referring to? I was talking about actual frogs that I was able to hold.

 

[AnthrpicDecadnc]

As I understand it, he has "Ham" and "Goomba," with the later serving as the control -- sort of.

Cheers,
Dave

oooh. whoops, i still suggest trying out the experiment on a larger group. or maybe you can find other keepers to try it out and some to use as a control and see what you get for results.

 

[skinheaddave]

or maybe you can find other keepers to try it out and some to use as a control and see what you get for results.

This is an area of interest to me -- the ability to split up an experiment between "labs" and still draw conclusions as a whole. Tradition dictates that you lose some degree of standardization as soon as you do this and so cannot draw statistical conclusions as readily. That being said, we do have methods of dealing with variations in experimental factors (ANOVA analysis can work with appropriate blocking, for example). The key would be not just to increase the overall number of experimental subjects (which would effectively only constitute smooshing together several iterations of a study) but to actually facilitate the reduction in the size of the group needed in a single "lab." I don't know how easily this can be done -- but I suspect the key would be in lowering the detail of the data and accepting a lower insightfulness of any conclusions.

Cheers,
Dave

 

[nickbachman]

Im curious as to what picture you are referring to? I was talking about actual frogs that I was able to hold.

Context clues hint that I'm talking about a picture of a cricket-fed tree frog with corneal lipids.

 

[JLDomestics]

Context clues hint that I'm talking about a picture of a cricket-fed tree frog with corneal lipids.

Still waiting for a link bud.

 

[Goomba]

 

[nickbachman]

Binngggoooo.

 

[JLDomestics]

Cool, never seen that picture before.

 

[AnthrpicDecadnc]

This is an area of interest to me -- the ability to split up an experiment between "labs" and still draw conclusions as a whole. Tradition dictates that you lose some degree of standardization as soon as you do this and so cannot draw statistical conclusions as readily. That being said, we do have methods of dealing with variations in experimental factors (ANOVA analysis can work with appropriate blocking, for example). The key would be not just to increase the overall number of experimental subjects (which would effectively only constitute smooshing together several iterations of a study) but to actually facilitate the reduction in the size of the group needed in a single "lab." I don't know how easily this can be done -- but I suspect the key would be in lowering the detail of the data and accepting a lower insightfulness of any conclusions.

Cheers,
Dave

could always find the mean results and use those as a general outcome. It would at least tell if it boosted any growth

 

[skinheaddave]

could always find the mean results and use those as a general outcome. It would at least tell if it boosted any growth

This is just a variation on the "smooshing together of iterations" I mentioned in my previous post. Once could take mean results from different studies, but if each study itself has no statistical power behind it, the results are simply a bigger form of useless. Assuming the results from each unit are supported then this kind of comparison might be useful -- though listed results might be better than averaged results since the latter would tend to dismiss the variation between "labs."

The neat part would not be taking three studies, each with 100 subjects, and comparing the results (each of which should have enough statistical power to demonstrate any moderately strong effect) -- the neat part would be taking 100 studies, each with three subjects, and being able to defend it statistically. I suspect that this would only be possible for the strongest effects in the simplest of experiments.

Cheers,
Dave

 

[Goomba]

Hah, you chaps are taking this thread way to seriously, as I said, perhaps "experiment" was the wrong term. I should have said something along the lines of "groundless shits and giggles", I don't have the money for a controlled test, but no worries, I dont think I'll go through with this proposed diet.