Do specimens attributed to Lystrosaurus murrayi and L . declivis ( Triassic Therapsida ) represent one species ?

HOW TO CITE: Thackeray JF. Do specimens attributed to Lystrosaurus murrayi and L. declivis (Triassic Therapsida) represent one species? S Afr J Sci. 2018;114(3/4), Art. #a0258, 2 pages. http://dx.doi. org/10.17159/sajs.2018/a0258 Lystrosaurus murrayi and L. declivis are two dicynodont (Therapsida, or mammal-like reptile) species commonly represented in Triassic deposits which are referred to as the ‘Lystrosaurus Assemblage Zone’ of the South African Karoo, postdating the Permo-Triassic boundary at 252 million years ago. The holotypes of these two species are curated in the Natural History Museum in London. Unfortunately, both are fragmentary (personal observation). Despite the fragmentary condition of the holotypes, Cluver1, Cluver and King2 and Brink3 accepted the view that at least two species of Lystrosaurus (L. murrayi and L. declivis) could be distinguished.


Two hypotheses
Thackeray's hypothesis addressed in this paper (H T ) is that specimens attributed to one or other of the two Triassic species are conspecific.H BB is the alternative hypothesis (reflected for example by Botha-Brink et al. 6 ) which is that two distinct Triassic species of Lystrosaurus can be recognised.
In order to test these two hypotheses, attention was given to the maximum basal skull length (BSL) dimensions in a total of 196 specimens of Lystrosaurus in Triassic deposits. 6Two other species of Lystrosaurus (L.maccaigi and L. curvatus), both of which occur in Permian deposits, were not considered here but deserve attention in future studies.
A summary of measurements obtained for BSL for specimens examined by Botha-Brink et al. 6 and attributed by them to two Triassic species of Lystrosaurus, is given in Table 1.The results presented in Table 1 can be used to determine whether H T is supported.From Table 1, the following observations are made: 1.The mean BSL dimensions in specimens attributed to L. murrayi (106.7 mm) is only slightly smaller than that for specimens attributed to L. declivis (118.1 mm).
2. The means are not significantly different (Students t-test, p=0.05).
3. The standard deviations are almost identical (31.1 and 30.0 mm).
4. The coefficients of variation are almost identical (29.1 and 25.2).
Two important additional observations are: 5. Specimens attributed to the two species are both represented in the Lystrosaurus Assemblage Zone, with L. declivis apparently occurring only slightly later than L. murrayi, which apparently occurs first at the time of the Permo-Triassic boundary, 252 million years ago. 6Notably, the specimens are coeval in most of the Lystrosaurus Assemblage Zone.
6. Specimens attributed to L. murrayi and L. declivis both display an 'early breeding pattern', which is associated with relatively low percentages of large individuals in the Lystrosaurus assemblages in Triassic deposits. 6 can be concluded that H T is supported by all six of these observations.
Further analyses should be undertaken on measurements in addition to BSL to test H T , using a morphometric analysis of the kind developed by Thackeray et al. 7 , updated by Thackeray 8 and discussed by Thackeray and Schrein 9 in the context of a probabilistic definition of a species, related to sigma taxonomy 10,11  the general tendency to use alpha taxonomy 12 whereby specimens are classified in terms of discrete taxa, assuming clear boundaries between them.
The six observations presented here, in support of H T , are consistent with the view held by Broom 4 who stated that Owen's specimen attributed by him to L. declivis was 'probably' the same species represented by the holotype of L. murrayi described by Huxley in 1859.The six observations presented here are also consistent with those of Cosgriff et al. 13 Thackeray et al. 5 suggested that much of the variability in specimens attributed to L. murrayi and L. declivis may be a result of sexual dimorphism (associated in part with 'bossing' on crania above the orbits), or ontogeny within one species.It was indicated that bosses were likely to occur primarily in male specimens, as reflected by the following statement: 'Where supraorbital bosses are present, these occur mainly in relatively large individuals.We consider the presence of supraorbital bosses in almost 50% of large individuals and the absence of such bosses in about 50% of similar-sized individuals attributed to the same species, as a potential indication of differences between adult males and females' 5 .
If only one species of Lystrosaurus species is represented in Triassic deposits of the South African Karoo, as hypothesised through H T , the nomen 'murrayi' would have precedence over 'declivis' according to rules of nomenclature, as Huxley's specimen was described in 1859, a year earlier than that reported formally by Owen.A possibility to be considered is that L. murrayi and L. declivis are components within a chronospecies, recognising also the importance of hybridisation in a diversity of modern taxa. 9

Table 1 :
Comparison of basal skull length measurements for Lystrosaurus murrayi and L. declivis