Kryptobaatar

Kielan-Jaworowska, Zofia & Lancaster, Terry E., 2004, A new reconstruction of multituberculate endocranial casts and encephalization quotient of Kryptobaatar, Acta Palaeontologica Polonica 49 (2), pp. 177-188 : 179-181

Endocast of Kryptobaatar

Introductory remarks.—Although Kryptobaatar is the most common mammalian genus in rocks of the Djadokhta Formation and Ukhaa Tolgod beds ( Kielan−Jaworowska 1970, 1980, 1998; Rougier et al. 1996; Kielan−Jaworowska and Hurum 1997, 2001; Wible and Rougier 2000) and in the Bayan Mandahu Formation in China ( Smith et al. 2001), its endocast has remained unknown.

Kielan−Jaworowska and Dashzeveg (1978) described a new multituberculate genus and species Tugrigbaatar saichanensis , based on a nearly complete skull with dentaries and fragments of the postcranial skeleton (GI PST 8−2) from Tögrög beds (equivalent of the Djadokhta Formation). Wible and Rougier (2000) claimed that Tugrigbaatar saichanensis Kielan−Jaworowska and Dashzeveg, 1978 is a junior synonym of Kryptobaatar dashzevegi Kielan−Jaworowska, 1970 . Kielan−Jaworowska et al. (2000) accepted the congenerity, but not the conspecifity of the two taxa. Smith et al. (2001) described another species of Kryptobaatar , designated K. mandahuensis from the Bayan Mandahu Formation in China. Re−study of the individual variation among the specimens of Kryptobaatar dashzevegi housed in ZPAL and GI PST, and a comparison with the holotype of Tugrigbaatar saichanensis convinced Kielan−Jaworowska et al. (2003) that the differences between T. saichanensis and K. dashzevegi fall within the range of variation of K. dashzevegi . They agreed with Wible and Rougier (2000) that the two taxa are not only congeneric, but also conspecific, which we follow here.

Endocast from Tögrög (GI PST 8−2, Figs. 1 , 2 , 3B ).—The endocast of GI PST 8−2, as originally preserved, was figured by Kielan−Jaworowska and Dashzeveg (1978: pl. 2: 1b); this specimen shows exposure of the partial endocast, including the superior cistern and cerebral hemispheres. After the paper by Kielan−Jaworowska and Dashzeveg (1978) was published, the first author (Z. K−J) removed a part of the cranial roof preserved in GI PST 8−2, as well as the bone surrounding the right paraflocculus and exposed nearly complete dorsal, lateral, and posterior aspects of the endocast ( Figs. 1 and 2 ). Although the sand that filled the braincase after the death of the animal was extremely coarse, as it may be seen from the photographs published herein, details of the endocast are well visible.

The skull length of GI PST 8−2 has been estimated by Kielan−Jaworowska and Dashzeveg (1978) as about 30 mm; by comparison with the skull of ZPAL MgM−I/41, the length of which was estimated as 34 mm, we now think that 34 mm is also a more probable length for GI PST 8−2. The full length of the exposed endocast (which is slightly deformed) is about 17 mm, which is one half of the estimated skull length. The most characteristic feature of the endocast is the size of the impression of the olfactory bulbs, which are narrow and strongly elongated, about 6 mm long (which amounts to 35% of the endocranial cast length). The olfactory bulbs as a whole are narrow, their greatest width being at about one third of the length from the posterior; from that level they taper anteriorly and slightly posteriorly. In lateral view the olfactory bulbs are relatively shallow, especially anteriorly. The bulbs slightly diverge posteriorly and there is a single, median fusiform structure of unknown significance, interposed between the posterior part of the olfactory bulbs and the anterior part of the cerebral hemispheres. A similar, but a better preserved structure occurs in numerous endocasts of Chulsanbaatar ( Kielan−Jaworowska 1983) . Because of the distortion of the endocast ( Fig. 1A ), the cerebral part of this structure has been moved to the left with respect to the part lying at the end of the olfactory bulbs (but see reconstruction in Fig. 3B ).

The cerebral hemispheres (measured in longitudinal projection) are about 8.8 mm long. They diverge slightly anteriorly, leaving a small, convex sub−triangular body (posterior part of the fusiform structure described above), which reaches the transverse furrow separating the cerebral hemispheres from the olfactory bulbs. Posteriorly the hemispheres diverge more strongly and embrace the bulbous roughly triangular structure that we call the superior cistern (the former “vermis”). The superior cistern is about 4 mm long in dorsal view and about 4.8 mm long in lateral view. Its maximum width is 5.3 mm. The casts of the transverse and sigmoid sinuses are partly preserved. A very large paraflocculus extends laterally (with slight posterior deviation) and ventrally as extension of the superior cistern (preserved on the endocast) and originate apparently from the vermis or other posterior cerebellar lobules, lying at the rear of the cerebral hemisphere.

Other endocasts of Kryptobaatar and individual variation.—In the collections of K. dashzevegi housed in ZPAL and GI PST, the endocast has been preserved only in GI PST 8−2. However, in ZPAL MgM−1/41 (see Kielan−Jaworowska and Gambaryan 1994: fig. 1), the endocranial cavity has been exposed, showing a very deep subarcuate fossa that extends into the base of the zygomatic arch.

Fragments of the endocast of K. dashzevegi have been figured by Wible and Rougier (2000: fig. 8, left stereo−photo) for PSS−MAE 101, but were not included in the right explanatory drawing to that figure. In this specimen the posterior part of the endocast is clearly seen, with the superior cistern deeply inserted between the cerebral hemispheres, which strongly diverge posteriorly. The size and proportions of the exposed part of the endocast are identical as in GI PST 8−2, figured here in Fig. 1 .

The size of the olfactory bulbs is subject to individual variation in Chulsanbaatar vulgaris ( Kielan−Jaworowska 1983: pl. 1, see also our Fig. 3A ) and Nemegtbaatar gobiensis ( Kielan−Jaworowska et al. 1986: fig. 18B and wax model in fig. 30). Therefore, it is not unlikely that the size of olfactory vascular nasal foramina foramen magnum condyle 5 mm

bulbs also varied in Kryptobaatar dashzevegi . Our conclusions about the olfactory bulbs and olfaction are based on the endocast, in which the bulbs may be significantly larger than the brain structures in this area.

Another specimen of Kryptobaatar that exposes the endocast belongs to K. mandahuensis . Smith et al. (2001: pl. 1: 1) figured the dorsal aspect of IMM 96BM−II/3, in which the posterior part of the cranial roof is damaged, showing a partial superior cistern and diverging cerebral hemispheres. In this specimen, again, the structure of the posterior part of the endocast appears similar to that in GI PST 8−2.

Inner ear fragments.—The multituberculate inner ear is well known, due mostly to the meticulous work of Fox and Meng (1997, see also references to earlier literature therein), based on CT−scan, X−radiographic, and SEM study of unidentified Bug Creek petrosals, and Hurum (1998) based on serial sections and enlarged models of Nemegtbaatar and Chulsanbaatar .

Our contribution to the knowledge on inner ear structure is modest and concerns recording the preservation of fragments of semicircular canals and ampullae in GI PST 8−2 around the right paraflocculus ( Figs. 1B, C and 2 ). In front of the middle part of the paraflocculus two osseous ampullae adhering to each other are clearly seen ( Fig. 2A ). The more proximal ampulla (of the anterior semicircular canal) adheres to the postero−lateral end of the cerebral hemisphere. The anterior semicircular canal has not been preserved; it extended in a groove between the posterior end of the cerebral hemisphere and paraflocculus, as in Chulsanbaatar ( Hurum 1998: fig. 15). The more distal ampulla (of the lateral semicircular canal) is placed postero−laterally with respect to the ampulla of the anterior semicircular canal. A part of the lateral semicircular canal has been preserved, extending from the ampulla postero−laterally, above the lateral part of the paraflocculus. The ampulla of the posterior semicircular canal has also been preserved and is seen in the posterior view of the endocast ( Figs. 1C , 2B ) below the paraflocculus, but the canal has not been found.

Hurum (1998) based his studies upon the model of the inner ear prepared by Kielan−Jaworowska et al. (1986) during their study on serial sectioning of the skull of Nemegtbaatar . On the basis of the model, Hurum (1998: figs. 6 and 7) made a reconstruction of the inner ear of Nemegtbaatar (fig. 17B in his paper). In that reconstruction the ampullae are less sharply delimited from the canals than, for example, in the extant guinea pig (see Hurum 1998: fig. 17D). The ampullae preserved in GI PST 8−2 and the whole system of semicircular canals, as inferred from the preserved fragments resemble the reconstruction of multituberculate inner ear made by Fox and Meng (1997: fig. 2C).

Comparison with endocasts of other multituberculates.— The endocast of Kryptobaatar dashzevegi ( Figs. 1 , 2 , 3B ) differs from those of other multituberculates in having relatively longer olfactory bulbs. In Chulsanbaatar ( Fig. 3A ) and Nemegtbaatar Kielan−Jaworowska et al. (1986 : fig. 20a) the olfactory bulbs are shorter, and their length varies at about 1/4 of the total endocast length. The olfactory bulbs appear to be relatively narrower in K. dashzevegi than in Chulsanbaatar and Nemegtbaatar . The shape of the olfactory bulbs in K. dashzevegi is probably most similar to that of Ptilodus montanus (USNM 9710) figured by Krause and Kielan−Jaworowska (1993: pl. 1: 1), although the greatest width of the bulbs in K. dashzevegi is situated more posteriorly.

Another difference with multituberculate endocasts of other taxa concerns the shape and size of the paraflocculi, which, however, may not fill completely the volume of the subarcuate fossa (Sánchez−Villagra 2002). In Chulsanbaatar , Nemegtbaatar , and apparently also in Ptilodus (if Simpson’s 1937 reconstruction is in this respect correct) the paraflocculi are rounded in all the views (ball−shaped). In K. dashzevegi , in contrast, the paraflocculus is narrower longitudinally, extending laterally (with slight posterior deviation) and ventrally. The paraflocculus in GI PST 8−2 shows division into larger proximal and smaller distal parts; the distal part might correspond to the vessels referred to for Nemegtbaatar ( Kielan−Jaworowska et al. 1986: fig. 32) as “posttemporal recess vessels” (= arteria diploetica magna, see Rougier et al. 1992). However, as the sediment that filled the subarcuate fossa in GI PST 8−2 is coarse−grained, the reconstruction of these vessels was not possible, and we refer to them in Figs. 1 and 2 simply as vessels. In living mammals the flocculus−paraflocculus system is related to oculomotor and other sensorimotor coordination, and correlated with this function, it is involved in the control of such coordination by higher mental processes ( Altman and Bayer 1997).

Figures

Fig. 1. Kryptobaatar dashzevegi, GI PST 8−2. A. Skull showing endocast in dorsal view. B. The same in posterior view. C. Posterior part of the same skull in right lateral view. A1, B1, C1, stereo−photographs; A2, B2, C2, explanatory drawings of the same, slightly reconstructed and reduced.

Fig. 2. Kryptobaatar dashzevegi, GI PST 8−2. A. Right paraflocculus in oblique ventro−lateral view, showing anterior ampulla, lateral ampulla, and lateral semicircular canal. B. The same in posterior view, showing posterior ampulla. Both stereo−photographs, coated with ammonium chloride.

Fig. 3. Reconstruction of two Late Cretaceous multituberculate endocasts superimposed on the outlines of the skull (rendered to the same length) in dorsal view. A. Chulsanbaatar vulgaris. B. Kryptobaatar dashzevegi. A, modified from Kielan−Jaworowska et al. (1986); B, original. The outline of the skull in B, based on reconstruction of Wible and Rougier (2000:fig. 30), and data from the collection of K. dashzevegi, housed in ZPAL; the nasal vascular foramina are based in the holotype specimen ZPAL MgM−I/21.