Pleco growth rates.....

[Hitch]

Hey

So I was randomly eading some forums and I noticed that there are usually A LOT of questions asked about "growth rates", "time to get to max size" etc.

So I thought I would try to compile a spread sheet of the growth rates of various plecos.

I am aware that these numbers will vary depending on many environmental factors...but that is when statistical analysis comes in...lol.

So please help me in this little project and report your growths..

for example: my peco X was 1.5" when I got it, now 1 year later it is 3" or something like that.

When I analyze the data...I will be doing statistical analysis....so the more data points the better.

Thanks

[MatsP]

A very interesting project, and I'm looking forward to seeing some good data from it.

Having said that, I don't think this is a particularly practical project, really, even if you do get enough information to get some statistically viable data - which probably means about 10 "hits" on each species that you list. There are around 420 registered keepers of Loricariidae (I'm not guaranteeing that number, but I went through the list, removed all Corydoras and Synodontis by a "search and replace" type operation, and then removed everything else by hand - I may have missed one or two, or removed one or two Loricariidae). If we remove Otos and close relatives (at least the large genus in this group), we're left with around 400. Of these 142 has at least ten keepers (I agree that many of the fish with less keepers will still have more than one fish, so you may be able to get useful data for some more fish).

It is much harder to figure out how many of those 150-200 fishes that have registered keepers will answer your question. But my guess would be about 10-20. They may well have more than one fish - so let's say the average is 2.5 fish per user. That's 25-50 data points. But not all of those data points will be much use, because you have only one or two data items for a species. I'd say if you are really lucky, you get some 10-15 fish that you get statistically viable data for. Some of those will of course be the most popular fish, such as common pleco, common bristlenoze, Hypancistrus zebra, Panaque nigrolineatus, and such. and perhaps a couple of other random species.

And of course, you need the SAME people to come back some months later with a new measurement of the same fish.

Now, we also have "rough estimates" for many of those:
Common pleco (Pterygoplichtys spp): at least 6 inches, from 2" to 8" in a year.
Common bristlenose: 1" to 3" in a year, then relatively slowly until it reaches max size.
Hypancistrus zebra: Approx 1" per year until fully grown. Other Hypancistrus species are pretty similar.
Panaque nigrolineatus: Slowly - 1" per year at most when relatively small.
Some others that I know from personal experience:
Small Panaque species: Less than 1" per year until fully grown.
Sturisoma: About 3"-4" first year, then relatively slowly.
Hemiancistrus spp.: Medium growth rate when young, around 2" per year.

Now, that probably covers 95% of all questions about growth rates, and it took me a few minutes to type this summary.


--
Mats

[Suckermouth]

Out of curiosity, what statistical test will you be using? Usually it's better to know these things beforehand so you know what and how to collect data...

[Hitch]

Thanks for the input guys.

Let me start by saying I am aware of the magnitude of this project and definitely not practical for every single pleco out there. And to say its ambitious is an understatement.

Its just something that I always had questions about, I see a lot of people keep journals of their fish (specially spawns) that illustrate their growth rates and such so I am trying to set out to "organize" this data.

When I was keeping my aquatic turtles, I have met many people that use the shell growth rate as a marker for health and what they consider "good" development. And that is when I was first exposed to the idea.

As for the analysis themselves....I am still working out the details.

But so far what I am thinking is using tests like ANOVA or in simpler cases a student's t test to identify "groups of development"...and these groups will differ by strong environmental factors (using the L114 growth in Athens and Manchester for example...we would obviously identify them as two sep groups and cannot be put in the same category).

This would then lead to the identification of dependent factors in each of those groups, and identify other data points that have the same "blue print" of dependent factors.

Once we have enough data points for each of the groups, I am thinking of using a time series to plot out the expected growth curve.

So for example. I have 10 data points for the size of pleco X at the age of 1 year. ANOVA would help me in identifying which of these data points are statistically different from the others. Lets say for the sake of simplicity these 10 data points were divided into 2 groups (A and B). Then I would sample some points from A and some points from B and try to figure out the environment and feeding and such that makes them different (using the L114 example again, Manchester uses soft, tannin stained waters where as Athens used hard water..which contributes to the difference in growth size by the age of 3 years). Once I have identified the main "causes" of these variations, I would be able to find other data points that have the same "causes" at different ages. For example, for group A, I would be able to find more data points showing the growth at 6 months or 2 years etc. With this, I would then be able to use a Time series and generate a growth curve.

I am not a statistician, so there are still a lot of things to be worked out.

As for the time frame...I am thinking of 0 month, 2, 4, 6, 8, 10 months, 1 year, 1.5 years, 2 years etc.

[Suckermouth]

Tell me about it, statistics can be confusing. I should know stats better than I do...

In general, you should know the specific questions you want to ask before you gather the data. One, this streamlines data collection, but it also helps so you waste less time taking data that is irrelevant and helps to prevent not taking any data you didn't think of gathering. This also goes for your treatments for your ANOVA, as the treatment groups are relevant data you'll want to collect.

So for example. I have 10 data points for the size of pl*co X at the age of 1 year. ANOVA would help me in identifying which of these data points are statistically different from the others.

Pardon me if this my assumption is incorrect, but these few sentences already suggest that you might not have a clear idea of what you're doing.

First, it suggests to me that you don't know how the statistical tests you want to use work. ANOVA/t-test are used to test whether two treatment groups are different, not how individual data points are different. That is, you will be testing something like, for example, how 10 data points for the size of pleco species X at age 1 differ from 10 data points for the size of pleco species Y at age 1.

Next, knowing that the sizes of a fish are different doesn't get at growth rate unless you know they started at exactly the same size when put in their respective treatments. Of course, you were simply making an example so you may have known this.

Another issue is the fact you seem to be considering pleco age. This is impossible for wild-caught fishes as we don't know when exactly they hatched. I am not sure how accurately age can approximated, but actual age may not be data that can be obtained for many species.

Anyway, I'm a little unsure of your actual question. Are you trying to find how growth rates differ between species? Or are you trying to find out how environmental variables affect growth in loricariids? This is relevant to choosing the data to gather and the correct statistical tests for analysis.

If it's the former, then basically asking for lengths at different time points and species should be enough. Species would be treatments and growth rate would be your measurement variables.

If it's the latter, a simple way to do this is to first focus on one species rather than looking at many species. Ancistrus cf. cirrhosus is a great candidate. It is captive bred by many people, so many people may know the EXACT ages of their fish. Ask for a recording of data such as length of fish at a certain age X, which will be your dependent data. As your independent data, you can take qualitative data such as presence/absence of food types, presence/absence of wood in the tank, etc. and quantitative data such as water quality parameters, number of young in the batch, "population" density in the aquarium, etc. Since the question seems to be about environmental parameters, I'm not sure you actually need multiple time points unless you're interested as to how growth rate changes with age.

Assuming bristlenoses all hatch at about the same size, the size at age X can be used as a proxy for growth rate. Use correlations to test if change in length is correlated to measurement variables, and ANOVA to test for the qualitative variables (ie. presence or absence of bloodworms in the diet, presence or absence of zuchinni in diet). Feel free to use a p-value correction (ie. Bonferroni p-value correction) since the above experiment could require quite a few statistical tests, although some people have argued against p-value corrections.

I'm also jumping ahead and ASSUMING the data fits assumptions of normality and equal variance that is necessary for most tests, including ANOVA/t-test... Technically you'd want to test your data with those assumptions before using ANOVA so you can use a nonparametric test if necessary. Statistics, am I right?

Of course, environmental variables do not necessarily affect other loricariids as they would in A. cf. cirrhosus, but this kind of experiment would still be pretty interesting and a good set-up to focus on what kind of data is more relevant for loricariids and which ones might be thrown out if you expand to look at other loricariids.

[Hitch]

lol..ya....I usually just do my experiments and then get someone els to do the stats or at least tell me what I need to use....

had to crack open my second year stats text book when I first thought about this.

But yes.....I see the flaws in the observational procedure.....and I do need to look into ANOVA more...(hated ANOVA when I first studied it)...lol.

anywho...what I am trying to accomplish with this is to have a table that would illustrate the growth at various ages under various conditions. Not really a comparison of the rate of various species but more of how factors like hardness and pH of water, temp, age affect the rate. And as a by product, when I look into a new pleco to keep...I could look at it and be like ok..my pleco is 3" right now, I should be able to get it to 7" by this time under these conditions...kinda thing. (thats why I thought of ANOVA.....since each of these conditions can be considered a treatment...but obviously I am wrong...).

yes...I was also assuming that this will fit a normal.....

thanks for the tips.

[Bas Pels]

oblem with growth rates is that these are depending from the circumstances. In december 2003 I got 5 Pt gibbyceps, ~5 cm which I put into a 4 meter tank. After a year they were 17 cm, after 2 years 30. Now, after 6 years, they measure some 40 cm. 2 are in this 4 meter tank, the others have their 1000 liter tank of themselves

HOWEVER had I put them into a 1 meter tank (together) I'm quite certain they would still measure 20 cm or less

[oli]

Hi everyone,
An amateur breeder told me that: At a higher temp. they will grow faster, BUT they will not get so old.
Anyone a reply on this??

[MatsP]

Yes, temperature affects the metabolic rate of all cold-blooded animals, and the metabolic rate will in turn affect both life-span and growth rate. And it only "works" to a certain extent - the digestive system of colder-water fishes will not be be able to process the food quickly enough at higher temperatures, which will lead to the fish slowly starving to death if they are kept too high temperature.

--
mats

[Bas Pels]

these effects may be much larger then most people realize

Your average, common chemical reaction goes twice as fast for every 10 C the temperature increases - I'm talking about the oxydation of iron, for instance.

However, in an organism, the chemical processes are all steered by enzymes - which can be very, very temperature sensitive. Assuming they function all at optimal temperature, the conversion rate normally increases twofold for every 3 C

Thus, if you keep a fish, such as a goldfish, which might manage it, 6 degrees above the optimal temperature, it geriates 4 times as fast as it normally would have - the lifespan of opver 20 years reduced to a meagre 5 years

Thre same goes for other fish we might care more about

I have a tank of cich lids and lifebearers from the Rio Tamasopo, that is a north Mexican river, flowing into the Rio Panuco. This tank is now 20 C, instead of the normal advised 24 C

Apart from saving money, this also increases the fishes lifespan

[MatsP]

Of course, the growth achieved will still be relative to the food supplied and water quality. If the nitrate level is high, the fish will grow slower than under equivalent circumstances where the nitrate level is lower. So it's not as simple as "raise the temperature by x degrees and the fish will grow faster". You need to ensure that water changes, feeding, filtration is at ideal levels. I've read about people who do 80% water changes each day to keep the nitrate level down.

--
Mats

[2wheelsx2]

I've read about people who do 80% water changes each day to keep the nitrate level down.

That's the discus keeping philosophy, up the temperature, stuff the food into the fish, keep it BB and change 80-90% of the water every day. Me, I'd rather plant the tank and change water once a week.

[sunfish]

This an interesting question, but I don't think you can answer it by sending out questionnaires to hobbyists. There are literally hundreds of factors influencing growth, and due to the limited number of data you will get you won't be able to say which of the number of differences was responsible for the different growth rate. Plus, there is a very basic problem, and that's to to with the measuring itself. It can be very difficult to get an exact measurement for a live fish, especially if it's shy. For species that grow very slowly measurement errors could riun the data completely.

I think this question can only be answered under controlled conditions, i.e. by running an experiment. Take a batch of, say, 60 BNs of the same age and put them in 60 identical tanks. Then decide on 2 factors (e.g. temperature and water hardness). Have three levels for each factor (low, medium, high). Assign 10 fish to each of the six treatment groups. Keep the fish under otherwise identical conditions for a year and record their growth. This will, hopefully, give you statistically viable data.

Something like that would be the minimum amout of work you will have to do, and this will only give you a very limited amout of data (no information on a possible interaction between temperature and hardness can be gained here). And even this is not at all practical to do at home. That's what bachelor students are for.

[Hitch]

HEY! Im a bachelor student......

but ya, I know what you mean. I am in the middle of refining the "experiment".

[sunfish]

Been there, done that (repeatedly). I have completed a Bachelor thesis, a master thesis and am about to hand in my PhD thesis.

Are you planning to do that for your BSc? It should be doable (depending on how long you have to complete it), you just need a lot of resources (space, mainly). And of course a supervisor.

[MatsP]

If possible, I'd prefer a set of circumstances where you vary for example nitrate, rather than hardness. I personally do not think hardness has much to do with growth rate.

--
Mats

[Bas Pels]

for Ancistrus? certainly

However, I can imagine fish from hard water growing less, with much disforming, in too soft water: these fish are used to get their calcium from the water, and if the water does not contain much, they might be in trouble

But whether the DH is 10 or 20, I don't think that will matter (much)

[MatsP]

for Ancistrus? certainly

However, I can imagine fish from hard water growing less, with much disforming, in too soft water: these fish are used to get their calcium from the water, and if the water does not contain much, they might be in trouble

But whether the DH is 10 or 20, I don't think that will matter (much)

Yes, but the whole discussion is about "Pleco growth rates", and loricariidae are restricted to South America, which is to a vast majority soft areas - don't confuse us even more by dragging in fish from other regions... Otherwise, yes, I agree. And I believe some species of for example corydoras will not be able to reproduce in hard water, because the fish relies on the soft water to dissolve the calcium in the egg-shell - if the water is too hard, this will not happen (or at least not soon enough for the fry to be released).

--
Mats