LOGIN

RSS Facebook Twitter YouTube
GLOSSARY       

SEARCHGLOSSARY

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

PROFILESEARCH

racoll

Forum Replies Created

Viewing 15 posts - 1 through 15 (of 18 total)
  • Author
    Posts

  • racoll
    Participant

    Thanks Matt!

    If anyone wants a copy of the paper they can download it from goo.gl/2kRSCP

    in reply to: Species coding systems in the Knowledge Base #350037

    racoll
    Participant

    retro_gk’s suggestions sound very sensible to me.

    in reply to: Tapatalk #350006

    racoll
    Participant

    No worries chaps. Thanks for the quick reply.

    in reply to: Species coding systems in the Knowledge Base #349987

    racoll
    Participant


    @mikev
    said:

    Having a small number of well-defined lists is not really more complicated than having a single list.

    Yes, the L and C number systems are easy, and successful, but this is only two lists. I’d be concerned that when there are a dozen or more separate lists, it may not be so easy to keep track of them mentally? Maybe it’ll be no problem, as long as the contents of each list is obvious (see below).

    Matt said 
     

    If we do go with that, it will have to be SF-family-number at the very least because SF-vernacular name-number doesn’t work for me at all.

    Yes, the vernacular names are ugly, but at least they’re stable! A “loach” will always be a “loach” to a hobbyist, whatever happens to the taxonomy.

    As all you guys are probably aware, there is currently a couple of big NSF funded cypriniform taxonomy/systematics projects underway, which will likely result in further big changes to the systematics of cypriniform fishes. Therefore, there’s likely to be significant instability in the family level classifications. Not saying don’t do this, but just be aware!

     

    in reply to: Species coding systems in the Knowledge Base #349982

    racoll
    Participant

    @mikev said:
    The problem is that randomly generated codes are going to be very difficult to remember.

    Yes, I agree, but is the alternative much better? Is trying to remember numbers from multiple, parallel naming systems easy? You’d have SF-loach-008, SF-goby-008, SF-cichlid-008, SF, SF-barb-008, SF-rainbowfish-008, etc, etc, etc.

    But is this not why we have common names, which are inherently easier to remember? Wouldn’t, say, Schistura sp. 5 just be remembered as the “super convict loach” when in speech, or in the LFS? The majority of hobbyists really struggle with scientific names anyway, so I can’t see a huge difference there. Not many people, I imagine, go to the LFS looking for Schistura cf. balteata, but they’ll know what a “sumo loach” is.

    If you want complete simplicity, why not just have SF001, SF002, SF003 etc for all the undescribed species of all groups?

     

    in reply to: Species coding systems in the Knowledge Base #349958

    racoll
    Participant

    It seems to me that the problem is finding a system that is (1) easy to understand by the average hobbyist; and (2) makes SF’s life as easy as possible when the taxonomy changes or species/data need to added. Perhaps a compromise is the best you can hope for?

    I think the system of Pangio sp. “SF1” and Schistura sp. “SF1” is easiest for hobbyists, but a right pain when taxonomy changes and you have to deal with two sp. “SF1” in the same genus, and then a genus left with only a sp. “SF4”. Again, the current Pangio sp. “PAN01” system will break when species move between genera. The system of Pangio sp. “SF-loach-2”, Schistura sp. “SF-loach-3” and Pangio sp. “SF-loach-14”, I think could be confusing for hobbyists, as given time the numbers would simultaneously look as though they had logical structure (i.e. the early ones arranged in sequence), but then be essentially meaningless when later ones are added all over the place. A system of Pangio sp. “SF-g8N” and Schistura sp. “SF-Yj4” , would be nice and manageable for you at SF, with no naming conventions to deal with in reshuffles, but would look like just plain nonsense to normal users …

    Difficult decisions, and probably not one that should be rushed into!

     

    in reply to: Species coding systems in the Knowledge Base #349942

    racoll
    Participant

    A free statistics package such as “R” would be my first choice, but I’m sure it can be done in Excel.

    I can help you do this if you are interested.

    Also, you might be interested in the blog of Prof. Rod Page (Univ. Glasgow), who is one of the leading scientists working in biodiversity informatics. He talks about this kind of thing a lot.

    http://iphylo.blogspot.com

     

    in reply to: Species coding systems in the Knowledge Base #349933

    racoll
    Participant

    Yes, you are right mikev. They don’t absolutely have to be randomised, as long as they are unique and stable (persistent). That’s what’s important.

    However, randomised descriptors do have the advantage of dealing with a situation arising where a previously undescribed species gets described. For example, in the Knowledge Base you would have a list of: Pangio sp. “SF1”, Pangio sp. “SF3”, Pangio sp. “SF4” (i.e. number two is now missing); that might look a bit weird and confusing. It also copes with species moving into different families.

    Randomising also stops people reading into the numbers, and thinking that, for example, another site’s Pangio sp. 1 is the same as your Pangio sp. “SF1”.

    Five digits is probably overkill. Three case sensitive alphanumeric characters should be more than enough.

    Hoping this is helping, not hindering!

     

     

    in reply to: Species coding systems in the Knowledge Base #349923

    racoll
    Participant

    As someone who works with biodiversity informatics, what you need are randomised unique identifiers, able to be databased and synonymised easily. Something like a DOI, but for species. The idea would replicate the unique binomial. This number could then be used as part of the URL and can permanently resolve to a web page if linked or searched for from another site, even after being described with a proper name, or found to belong in a different genus or family.

    In reality it could look like an output from a URL shortener service, e.g. Pangio sp. “sf-zJg7o”.

    I understand that this probably has zero appeal to the hobbyist, but it will future-proof the site very effectively if combined with a solid database.

    in reply to: Wild Zebra Danios? #347104

    racoll
    Participant

    QUOTE
    Are there transgenic fish existing in natural habitats already Rupert?

    Not that I know of. I think the study just wanted to know what the potential capacity was for transgenic danio to survive in the USA.

    Let me know if you want the paper.

    in reply to: Wild Zebra Danios? #347081

    racoll
    Participant

    jfb, you’ll be wanting:

    Cortemeglia, C. and Beitinger, T.L. (2006) Projected US distributions of transgenic and wildtype zebra danios, Danio rerio, based on temperature tolerance data. Journal of Thermal Biology 31: 422–428.

    http://dx.doi.org/10.1016/j.jtherbio.2006.01.011

    in reply to: Dna Barcoding Aquarium Cyprinids #346923

    racoll
    Participant

    QUOTE
    Were the ‘regular’ D. choprae you used from the aquarium trade?

    Correct, almost every fish in the study was sourced from the aquarium trade.

    in reply to: Dna Barcoding Aquarium Cyprinids #346909

    racoll
    Participant

    QUOTE (Matt @ Jan 24 2012, 09:30 PM) <{POST_SNAPBACK}>
    Is aff. choprae the ‘northern glowlight danio’ that was in the trade a few years back?

    Yes, the larger grey one.
    Image

    There was also, surprisingly, a heck of a lot of molecular diversity in “regular” D. choprae. Unfortunately neither my specimens or my images were particularly good for this species, so I could not tell if any of these splits were backed up by morphological differences. Seems strange given their relatively restricted distribution.

    in reply to: Dna Barcoding Aquarium Cyprinids #346899

    racoll
    Participant

    QUOTE
    Do you have locality details for the three Puntius specimens pictured?

    Sadly not. They were bought from retailers. Maybe some of the Indian members can help out here?

    QUOTE
    it appears I’ve some editing to do and at least one additional profile to add!

    You could also consider adding a few more. For example:

    I found Puntius hexazona was well differentiated from P. pentazona in terms of DNA, and can readily be distinguished by the spot at the end of the dorsal fin. Funnily, they didn’t seem that closely related, not grouping together in either mtDNA or nDNA trees (again, however, an explicit phylogenetic analysis was not carried out).

    The striped and spotted Danio kyathit were very different in terms of DNA, and this was consistent with them being different species. There were other morphological differences too. The name D. kyathit applies to the spotted one. I referred to the striped one as D. aff. kyathit.

    The fishes I called D. aff. choprae and D. aff. dangila were also different in terms of DNA and morphological characters. They are likely different species too.

    There are more, and I can update when these come up in discussion.

    in reply to: Dna Barcoding Aquarium Cyprinids #346865

    racoll
    Participant

    Thanks Matt!

    QUOTE (“Matt”)
    Is the other form of P. gelius the larger one? Think we talked about the possibility of it being P. canius?

    I found three types of “P. gelius“. Two from the trade, and another kindly donated to me by Heok Hee. I don’t know where he got that one from, but I didn’t find any others like it in the trade.

    This is it Heok Hee’s. The specimens matched published P. gelius characters, but it is a darker and larger fish than the “regular” P. gelius. The difference in DNA is consistent with it being a separate species from the regular one.
    Image

    This is the “regular” P. gelius from the trade, and the one that is called P. canius in your P. gelius profile. This one is most consistent with the description of P. gelius.
    Image

    Now this one has completely different DNA to both those above, and does not appear to be closely related to either (I didn’t do a phylogeny, so this remains to be tested more thoroughly). It is common in the trade, and is the one labelled P. gelius in your profile. It is not P. gelius, and is distinct in numerous morphological characters: most notably, a lack of spots on pelvics, and the spot on the anal fin extending well onto the body. I believe this is undescribed, although P. ornatus Vishwanath & Laisram is a possibility. I referred to it as P. aff. gelius.
    Image

    Now, which of these is P. canius? That species is shown to have distinct spots on the pelvics, as illustrated by Hamilton in McClelland (1839). That rules out the last one. Puntius canius is described as a smaller fish (than P. gelius) with a reddish hue; I did not record this colouration in any my specimens, and it may be a seasonal breeding effect. Bordoloi and Baishya (2006) report this colouration from specimens of “P. ornatus” from Assam, and the specimens they picture appear similar, but I am not sure these are P. ornatus as described. The description of that species being not being terribly informative, however. There are certainly more forms of this complex out there, and it will require a more study before we can get to the bottom of it (especially regarding the real identity of the elusive P. canius)

    QUOTE
    Still haven’t been able to bring myself to change the Trigonostigma profiles to Rasbora.

    That would be a personal choice I think. I was asked about this on PC. Seems taxonomists are ignoring the work of Tang et al. (2010), so it might be better to stick with Liao’s genera? I don’t personally have any favouritism to either lab group, or with molecules vs. morphology being superior. It is my philosophy that seeing as there’s conflict between the data, then it is best to be conservative until this issue is resolved by a study directly focussing on that. Treating them all as Rasbora seems to me like the safest option for now. I am probably in a minority with that kind of opinion, however.

Viewing 15 posts - 1 through 15 (of 18 total)

« Previous Entries