Return to search

African dwarf-crocodile scale-counts evaluated as supporters of Osteolaemus tetraspis osborni (Reptilia, Crocodylia, Crocodylidae)

The Convention on International Trade in Endangered Species of Wild Fauna and Flora (C.I.T.E.S.) has always recognized Osteolaemus tetraspis osborni (Schmidt, 1919) as a geographic subspecies of the African dwarf-crocodile, taxonomically distinguished from the subspecies Osteolaemus tetraspis tetraspis Cope, 1861. These two subspecies are primarily based on anatomical evidence that is not visible on a whole living animal. The subspecies distinction is the internal degree of boniness of the septum between the two external nostrils. This character must be examined by dissection of the animal’s head. Therefore it is desirable that some reliable and diagnostic kind of externally visible identification character (such as coloration, measuring head proportions, or counting scales) should be searched for. Otherwise, any individual Osteolaemus tetraspis animal must be partly dissected to see what subspecies it is. In this thesis, approximately a dozen different external characters (mostly counting scales or rows of scales) are independently tested and compared with each other in an effort to determine which of them is the most useful as an identification tool for separating the Osteolaemus t. osborni subspecies from the O. t. tetraspis taxon. If none of these scalation details are found to be 100% reliable as a predictor of the appropriate geographic taxon, then some additional factors become relevant, such as their comparative difficulties in physically performing the investigation. Some kinds of scale-counts are easier to define and record than others. The new sample is 68 African dwarf-crocodiles, of which 53 are geographically in the COAST (O. t. tetraspis) taxon, and 15 are from the INTERIOR (O. t. osborni). The data was collected by field-work in numerous African nations, and the expedition sampled and travelled across the C.I.T.E.S. subspecies boundary between these two taxa in several ways. In comparison with earlier published reports about scalation characters in the Osteolaemus tetraspis species, this new sample is relatively large, and has remarkably good precision about locality data. Further, this expedition (the Zoer Osteolaemus Project, 2006-2007) exercised extreme care to apply clear definitions for the various characters (scale-counts, row-counts, etc.). Based on good samples, and on the high quality scalation data collected in 2006-2007 in Africa, it is concluded that none (zero) of these dorsal and ventral skin characters (with the possible exception of the two kinds of ventral scale-counts, for which the INTERIOR sample was only 3 animals) yield satisfactory results for discriminating subspecies of dwarfcrocodiles in the general part of the continent that was examined. All of these many kinds of scalation features fail to 100% reliably predict the correct subspecies (COAST or INTERIOR) outcome. Some characters never (0,00%) work, while other characters work part of the time, but not one gives the level of predictability that is required by C.I.T.E.S. for quick and accurate taxonomic identification in Osteolaemus tetraspis with these two subspecies as regulated taxa. All of the literature scalation characters required detailed special examination procedures, usually involving physically handling the animal. This thesis invents and recommends a new kind of character which has the advantage that just looking at the dorsal surface of the crocodile is often sufficient, and sometimes it can be done through a telescope. For this “permit-free” scalation character, properly taken photographs can include the normal (tourist at the zoo) view from the side and slightly above. It is not necessary to look straight down at the animal, as long as the number of transverse rows on the base of the neck, and along the length of the body, and on the proximal half of the tail can be counted. Neither the old nor the new characters work 100% of the time, and although the permitfree count of transverse rows on the dorsum is slightly more predictive than any of the others, even this new method yields remarkably poor results. However, given a large enough sample (for example bushmeat or stuffed animals or flat hornback hides passing through a local market over a long period of time), it is predicted that an interesting indication of subspecies identification could, in a statistical way, possibly be obtained. In the final analysis, though, African dwarf-crocodile scale-counts do not appear to support the model of two taxa in Osteolaemus tetraspis as a species, and therefore they do not support the inclusion of Osteolaemus tetraspis osborni as a regulated taxon on the C.I.T.E.S. list of crocodilians. AFRIKAANS : Het CITES verdrag (“Convention on International Trade in Endangered Species of Wild Fauna and Flora”) heeft Osteolaemus tetraspis osborni (Schmidt, 1919) altijd erkent als een geografische ondersoort van de Afrikaanse dwergkrokodil, die onderscheiden kan worden van Osteolaemus tetraspis tetraspis Cope, 1861. Deze twee ondersoorten zijn primair gebaseerd op anatomisch bewijs dat niet zichtbaar is op een puntgaaf en levend dier. Het onderscheid tussen de ondersoorten is de interne graad van het aanwezig bot van de afscheiding tussen de twee uitwendige neusgaten. Dit karakter zal moeten worden onderzocht middels een autopsie op het hoofd van het dier. Daarom is het wenselijk dat er gezocht wordt naar een aantal betrouwbare en diagnostieke vormen van uitwendig zichtbare identificatie karakters (zoals kleuring, het meten van hoofdproporties, of het tellen van schubben). Zo niet, dan zal ieder individueel Osteolaemus tetraspis dier gedeeltelijk moeten worden ontleed om te zien welke ondersoort het betreft. In deze scriptie zijn ongeveer een twaalftal verschillende uitwendige karakters (met name het tellen van schubben of rijen van schubben) onafhankelijk van elkaar getest en onderling vergeleken in een poging om vast te stellen welke van hen het meest bruikbaar is als een middel ter identificatie voor het onderscheiden van de Osteolaemus t. osborni ondersoort en de O. t. tetraspis taxon. Wanneer geen van deze schubbenkleed details als 100% betrouwbaar wordt beschouwd als voorspeller van de van toepassing zijnde taxon, dan worden een aantal aanvullende factoren relevant, zoals de moeilijkheidsgraad van de fysieke uitvoering van het onderzoek. Sommige schubbentellingen zijn eenvoudiger te definiëren en te registreren dan andere. De nieuwe sample bestaat uit 68 Afrikaanse dwergkrokodillen, waarvan er 53 geografisch gezien in de KUST (O. t. tetraspis) taxon zitten en er 15 zijn van het BINNENLAND (O. t. osborni). De gegevens zijn verzameld middels veldwerk in een groot aantal Afrikaanse landen en de expeditie verzamelde op verscheidene manieren gegevens van deze taxa aan beide zijden van de C.I.T.E.S. ondersoorten grens. In vergelijking met eerder gepubliceerde rapporten over schubbenkleed karakters in de soort Osteolaemus tetraspis, is deze nieuwe sample relatief groot en heeft het opvallend nauwkeurige plaatsgegevens. Verder heeft deze expeditie (het Zoer Osteolaemus Project, 2006-2007) voor zeer duidelijke definities gezorgd voor de verschillende karakters (schubbentellingen, rij-tellingen, enz.). Op basis van goede samples en de hoge kwaliteit van de in 2006-2007 in Afrika verzamelde schubbenkleed gegevens, kan worden geconcludeerd dat geen (nul) van deze rug en buik huidkarakters (met mogelijk uitzondering van de twee schubbentelllingen op de buik, waarvan de BINNENLAND sample slechts 3 dieren was) bevredigende resultaten geeft om ondersoorten in dit breedvoerig onderzochte deel van het continent te kunnen onderscheiden. Al deze schubbenkleed kenmerken schieten te kort om 100% betrouwbaar de juiste ondersoort te voorspellen (KUST of BINNENLAND). Sommige karakters werken nooit (0,00%), andere gedeeltelijk, maar geen enkele geeft het door C.I.T.E.S. vereiste niveau van voorspelbaarheid voor een snelle en accurate identificatie van Osteolaemus tetraspis met deze twee ondersoorten als gereguleerde taxa. Alle schubbenkleed karakters uit de literatuur vereisten gedetailleerde en speciale onderzoeksprocedures, waaronder vaak het fysiek hanteren van het dier. Deze scriptie vindt uit en beveelt aan een nieuw karakter, welke het voordeel heeft dat enkel het kijken naar de bovenkant van de krokodil voldoende is en soms met behulp van een verrekijker kan gebeuren. Voor deze “vergunning-vrije” schubbenkleed karakter kunnen fatsoenlijk genomen foto’s het normale (toerist in een dierentuin) aanzicht weergeven van de zijkant en lichtelijk van boven. Het is niet nodig om recht boven het dier naar beneden te kijken, zolang het aantal rijen van de basis van de nek in de lengterichting van het lijf tot ongeveer de helft van de staart kunnen worden geteld. Noch de oude, noch de nieuwe karakters werken in 100% van de gevallen en ondanks dat de “vergunning-vrije” telling van dwars op de bovenkant van het dier lopende rijen een iets betere voorspelling geeft dan alle andere, heeft zelfs deze nieuwe methode opvallend slechte resultaten. Echter, met een sample die groot genoeg is (bijvoorbeeld “bushmeat” of opgezette dieren of huiden, die over een lange periode op lokale markten worden verhandeld) ligt het in de lijn der verwachting dat er een interessante indicatie voor de identificatie van ondersoorten (op een statistische manier) kan worden verkregen. Al met al kan worden gezegd dat schubbentellingen in Afrikaanse dwergkrokodillen niet het model van twee taxa in Osteolaemus tetraspis ondersteunen en daarmee Osteolaemus tetraspis osborni niet erkennen als een gereguleerd taxon in de C.I.T.E.S. lijst van krokodillen. / Dissertation (MSc)--University of Pretoria, 2010. / Animal and Wildlife Sciences / unrestricted

Identiferoai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:up/oai:repository.up.ac.za:2263/25795
Date24 June 2011
CreatorsDe Boer, Rogier
ContributorsVan Hoven, Wouter, deboer.rogier@gmail.com
PublisherUniversity of Pretoria
Source SetsSouth African National ETD Portal
Detected LanguageDutch
TypeDissertation
Rights© 2010, University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria.

Page generated in 0.0031 seconds