Leptorycteropus, PATTERSON, 1975

Lehmann, Thomas, 2009, Phylogeny and systematics of the Orycteropodidae (Mammalia, Tubulidentata), Zoological Journal of the Linnean Society 155 (3), pp. 649-702 : 676-678

GENUS LEPTORYCTEROPUS PATTERSON, 1975

Emended diagnosis: Medium-sized Orycteropodidae with the following autapomorphies: mandibular symphysis extended backwards, starting between P 1 and P 2; pubis oriented medioventrally. It can also be distinguished from the other Orycteropodidae by the combination of the following characters: presence of canines; absence of palatine groove; weak deltoid crest, not projected laterally, olecranon fossa triangular in shape, and slender distal epiphysis on the humerus; sharp oblique rim on the radius; sciatic notch positioned at the level of the acetabulum; sacrum formed by five sacral vertebrae; articular facet for the sesamoid bone of the gastrocnemius muscle at the level of the diaphysis on the femur; absence of a falciform process, as well as presence of an uninterrupted rim of bone between the tibial tuberosity and the fibula on the proximal epiphysis of the tibio-fibula; and a long tibial crest.

Remarks: Different species of aardvark have been described to possess canines: A. depereti , A. mauritanicus , aff. A. pottieri , and L. guilielmi . Presumed canines have also been observed in the craniums of juvenile individuals of O. afer , but they consist of small rounded dentine masses, and do not pierce the gum. However, as suggested by Lehmann et al. (2005: 126): ‘[A.] depereti and [A.] mauritanicus display supernumerary premolars and not canines’, as can be observed in the extant form. I consider that a real canine in Orycteropodidae is a large antemolar tooth in front of, and separated from by a diastema, a decreasing (in size) premolar row. Additionally, the maxilla (and, respectively, mandible) shows a slight bulge on the vestibular side of this canine. The general size of Leptorycteropus is estimated according to the dimension of the pelvis, which is the sole complete post-cranial element of the hypodigm. The length of the ilion in L. guilielmi is about 60% that of its counterpart in O. afer .

Type species: Leptorycteropus guilielmi Patterson, 1975 .

Geographic distribution: Lake Turkana basin (Lothagam Hill, Kenya).

Temporal distribution: Late Miocene, between 7.4 and 6.5 Mya.

Remarks: This genus is monospecific and known from only one site (Lothagam) so far.

SPECIES LEPTORYCTEROPUS GUILIELMI PATTERSON, 1975

Diagnosis: Same as for the genus.

Holotype: KNM-KP 419, partial skeleton; housed at the NMK, Nairobi.

Remarks: The holotype bears a reference number associated with the site of Kanapoi (KP), but it has actually been found in Lothagam (fossil usually labelled LT).

Type locality and age: Lothagam 1, Lothagam Hill , Lake Turkana basin ( Kenya). The holotype has been found in the Member B, which corresponds to the Lower Nawata Formation dated to between 7.4 and 6.5- Mya ( Leakey & Harris, 2003) .

Main occurrences: Type locality.

Additional material: Only some fragments of a femur (KMN LT 28573) have been additionally found in the Lower Nawata Formation of the Lothagam site ( Milledge, 2003), and these are housed at the NMK, Nairobi.

Discussion: Patterson (1975: 201) insisted that L. guilielmi ‘was the most generalized member of the family so far known’, and was primitive in various features. Indeed, this author suggested that no osteological structures of the cranium could be associated with a specialized mymecophagous diet. Patterson (1975) quoted, for instance, the well-developed temporal fossa, post-orbital process, and jugal. Furthermore, he considered that the facial region was short, and that ‘a rather large canine is present, together with a full complement of cheek teeth’, which is the common eutherian condition ( Patterson, 1975: 223). First, it is difficult to estimate the development of the temporal fossa on the very fragmentary material. Then, the reduction of the snout and mandible length cannot be confirmed. In fact, in all Orycteropodidae , the teeth are mainly restricted to the rear of the maxilla and mandible. The foremost tooth (canine for aff. A. pottieri or premolar for the other species) is then preceded by a long diastema, until the tip of the premaxilla and mandible. The maxilla and mandible of L. guilielmi are broken in front of the canine, so that it is not possible to tell the extent of the diastema and whether this species possessed frontal teeth. On the other hand, the mandible presents a symphysis extended backwards that almost reaches the P 2. This can be interpreted as ‘an extensive, firm symphysis’ ( Patterson, 1975: 223), consistent with active canines, or as a normal-sized symphysis shifted backwards because the snout is actually short. Note that in aff. A. pottieri the symphysis starts just behind the canine, whereas in Orycteropus , as well as in the new genus, the symphysis is situated well in front of the foremost premolar. The presence of both a canine and a firm symphysis is necessary but not sufficient to conclude that L. guilielmi used his canines actively. Moreover, in mammals, canines are not only used for the food intake process, but can also be a sexual character (although this is unlikely to be the case here, with respect to the feeble sexual dimorphism shown by the extant aardvark). Therefore, it seems more sensible to merely hypothesize that this species was not a specialized myrmecophagous animal.

The post-cranial elements, on the other hand, reveal more about the habits of L. guilielmi . For instance, Patterson (1975: 223) noticed that: ‘The limb bones indicate an animal capable of digging but not highly specialized for it’. Indeed, L. guilielmi is unique among Orycteropodidae for the structure of its humerus, in particular for the absence of a welldeveloped deltoid crest and the presence of a strong ventral pectoral crest (see also the discussion for A. abundulafus ). Moreover, if the femur is rather similar to the one in other taxa (but also see remarks), its pelvis shows some peculiarities. For instance, the reduction of the iliac and ischial bones, as well as the medioventral orientation of the pubis, implies a different distribution of the muscular masses. The back of L. guilielmi might have been less powerful than in the extant species, which reinforces the image of a less fossorial animal. This might also induce a less arched back, as suggested in the reconstruction made by Mauricio Anton ( Milledge, 2003: fig. 8.12). Despite its comparatively recent age, L. guilielmi shows characters that would be expected from a primitive Orycteropodidae . These hypothesis must, however, be confirmed on more complete Leptorycteropus material.

Remarks: The femur KNM-KP 419 F of the holotype is consolidated in part with plaster. In particular, the dorsal surface of the diaphysis has been reconstructed with a strong ridge. However, this structure is not consistent with the existing parts of the bone, and must thus be considered incorrect. Furthermore, a revision of the holotype confirms that the element KNM-KP 419 L is not a metacarpal bone, as suggested by Patterson (1975), but is rather the distal epiphysis of the right metatarsal III.