Документ взят из кэша поисковой машины. Адрес оригинального документа : http://zmmu.msu.ru/personal/pavlinov/doc/gerbillids.pdf Дата изменения: Fri Nov 25 12:43:12 2005 Дата индексирования: Tue Oct 2 03:32:49 2012 Кодировка:

Proceedings of the 8th International Symposium on African Small Mammals Paris, July 1999 Communications presentees 8 Symposium international sur les Petits mammiferes africains. Paris, juillet 1999

African Small Mammals
Petits mammiferes africains
Scientific Editors / Editeurs scientffiques

Christians Denys, Laurent Granjon et Alain Poulet

-

Editions
INSTITUT DE RECHERCHE POUR LE DEVELOPPEMENT

collection Colloques et seminaires Paris, 2001


Curren t concepts of Gerbillid phylogeny and classification
Igor J. A. Pavlinov

1 Introduction
GerbiUids constitute a moderately small group of about 20 genera (PAVLINOV et al, 1990; MUSSER and CARLETON, 1993). This group attracted close attention of taxonomists in the 70s and 80s, when several hypotheses about its relations with other muroids and principal phylogenetic divisions were elaborated (CHALENE et al., 1977; PAVLINOV, 1982; TONG, 1989; PAVLINOV et al., 1990). All these studies were morphological and considered a very limited data set, mainly dental and bullar morphology. One of the most surprising matters in this respect is that very few biochemical and cytogenetical works have been made with gerbillids, when compared with their "neighbors" in the classification (murids, arvicolids, etc.). I do not mean that morphological data exhausted themselves and are now outdated. Instead, they still provide quite interesting information about gerbillid affinities and phylogeny. So, below I shall consider some focal points in the phylogeny and taxonomy of gerbillids, as they are seen by a morphologist.


142 T

African Small Mammals / Petits mammiferes africains

I The protoconulid story
On the basis of dental morphology, gerbillids have been continuously placed within the Palaearctic cricetids or considered as an independent family or subfamily. However, in the 70s, a hypothesis of their possible relation to Miocene myocricetodontines was proposed by paleontologists (JAEGER, 1977; CHALTNE et al., 1977). That was a really sudden idea, because the genus Myocricetodon with its closest allies possessing at least some accessory murid cusplets on lower molars was traditionally considered as a member of Murinae (JACOBS, 1978). Contrary to this, no recent gerbillid genus was known at that time to possess such a murid feature; only the Miocene Protatera, presumably related to taterillines, has been reported to possess a protoconulid on the first lower molar (JAEGER, 1977). However, at the present moment, the protoconulid, one of the murid cusplets, was found in several recent gerbillid genera (fig. 1) as in Ammodillus (PAVUNOV, 1981) or in Asian Tatera and South African Desmodillus (PAVLINOV, 1985). In all these taxa, this small cusplet is clearly visible on the unworn first lower molar and disappears with tooth wear. The presence of a protoconulid is clearly a primitive condition for gerbillids. The more advanced morphotypes of the anterior part of M] are of two principal types, horseshoe and rhomboid ones (PAVLINOV, 1984). The first type is a characteristic feature of African taterillines, while the second is typical for Gerbillus, Meriones and their allies. In Desmodillus, it is possible to trace, in a set of teeth with various degrees of wear, the interrelations among these morphotypes. The protoconulid first becomes united with elements of the anteroconid to turn into the horseshoe type, and then the latter turns into the rhomboid type. If we take this ontogenetic transformation series as a kind of recapitulation, it could be supposed that the rhomboid type is a derivative of the horseshoe type, and the latter is a derivative of the type with an isolated protoconulid (PAVLINOV, 1984). In this connection, one can note that the horseshoe type also occurs sometimes in Gerbillus (PETTER, 1956; PAVLINOV, 1984) for which the rhomboid type is typical. Following the above arguments, it is possible to consider such a condition as a kind of atavism. If this is


I. J. A. PAVLINOV - Current concepts of Gerbillid phytogeny and classification

143 V

I Figure 1 Dental crown pattern of minimally worn Mi in some gerbillids and their allies to show the presence of protoconulid and anteroconid (both shown in black). A) Myocricetodon, B) Protatera, C) Tatera indica, 0) Desmodillus, E) Ammodillus, F) Gerbillurus, Q) Gerbillus. Arrows indicate the position of the protoconulid. A and are after JAEGER (1977); others are from PAVLINOV (1984).

correct, then it can be supposed that the horseshoe type of anteroconid was a characteristic feature of direct ancestors of Gerbillus. And mis can be taken as an indirect evidence that the protoconulid occurred in the ancestors of many gerbillid genera, and not only of those still retaining it. One may conclude that the most recent ancestors of gerbillids did have an accessory "murid" cusplet, namely a protoconulid, on the first lower molar. On the basis of this conclusion alone, it seems clear that cricetids cannot be ancestors of gerbillids. The hypothesis of gerbillid affinities to myocricetodontin.es looks more plausible in the light of the anteroconid morphology.


144 T

African Small Mammals / Petits mammiferes africains

However, it seems premature to unite these taxa in the same family, as CHALIN E et al., (1977) suggested . On the one hand , the myocricetodontines show no evolutionary trends in dental morphology common to "real " gerbillids. On the other hand, they still possessed a fairly "murid " mandible (as it was drawn in JAEGER, 1977), which suggests their auditory bulla was murid-like as well, that is without significant pneumatization. As to the gerbillids themselves, the anteroconid morphology may witness about the following. The Ethiopian Ammodillus occupies an isolated position, as it is the only recent representative of the family possessing a separate protoconulid that does not join with the anteroconid proper. The South African Desmodillus and Gerbillurus are very similar with respect to dental morphology. They both display a primitive horseshoe pattern of anteroconid and, by this, could be related phylogenetically to the taterilline lineage. Finally, the Asian Tatera indica displays a more primitive anteroconid morphology as compared to its African relatives.

I Mastoid morphology
The next structure to be discussed is the morphology of the mastoid portion of the auditory bulla. LAY (1972), PAVLINOV (1980), and PAVLINOV et al. (1990) showed that gerbillid taxa differ from each others by both the degree and, more importantly, the type of mastoid pneumatization (fig. 2). Omitting the details, pneumatization of a primitive solid mastoid can be considered as resulting from the penetration of tympanic cavity into the mastoid. Two principal pathways of this penetration are supposed to occur in the gerbillids, i.e. anterior and ventral. In the first case, the tympanic cavity penetrates into the mastoid through the dorsal part of the bulla, just above of the ossicles and before the mastoid. This process produces morphotypes to E. The ventral penetration occurs through the posterior part of the bulla, just ventral to the mastoid. To it, morphotypes F to H belong.


I. J. A. PAVUNOV - Current concepts of Gerbillid phytogeny and classification

145 T

I Figure 2
Principal morphotypes (A-H) of the mastoid in Gerbillidae (after PAVUNOV et al., 1990). Schematic parasagittal section. Arrows indicate pathways of penetration of the tympanic cavity into the mastoid. Dashed line indicates an imaginary anterior border between tympanic and mastoid cavities (no bony septa separates them), dotted area indicates part of the lateral mastoid wall knitted with the parafloccular fossa. Figures designate respective bony elements: 1,3) semicircular channels; 2) inner mastoid septa; 4) mastoid septa; 5) caudal septa; 6) tympano-mastoid septa.


146 V

African Small Mammals / Petits mammiferes africains

I Figure 3 Transformation series indicating supposed evolutionary relations between mastoid morphotypes in Gerbillidae. Morphotype designations are the same as on fig . 2.

There is also a secondary ventral penetration occurring within the socalled anterior type. The resulting pattern of mastoid chambers may be superficially very similar to those displayed by primary ventral penetration (compare morphotypes E and F). However, the bony elements separating the respective mastoid chambers are not really homologous: the diagonal septa includes the horizontal semicircular channel in morphotype E and does not include it in morphotype F. Thus, one may suppose that each of these two pathways is quite constrained morphogenetically. As such it seems highly improbable for one of them to "switch" into the other. On this basis, it is possible to reconstruct transformation series of gerbillid mastoid morphotypes (fig. 3). It is then reasonable to hypothesize that the type of mastoid pneumatization can be taken as the evidence of phylogenetic affinities of respective gerbillid taxa. The most interesting points in this respect are the following. Among the taterillines, Taterillus is the most primitive (morphotype A) and the Asian Tatera is the most advanced (morphotype C).


I. J. A. PAVLINOV - Current concepts of Gerbillid phytogeny and classification

147 T

African Taterona and Gerbilliscus belong to different lineages, each displaying the initial stage of anterior and ventral penetrations (morphotypes and F, respectively). As with dental morphology, Desmodillus and Gerbillurus sensu lato are again very similar in mastoid structure (morphotypes and C, respectively). They belong to the same lineage and differ only by the degre e of pneumatization , the leas t being in the subgenus Progerbillurus (morphotype B). Desmodilliscus (morphotype G) does not belong to the same lineage as Desmodillus, so they cannot be considered as close relatives. Instead, Desmodilliscus resembles in this respect to Gerbillus and Meriones and their allies (all possess morphotypes F or G). Mastoid pattern in the genus Pachyuromys (morphotype H) is peculiar. It belongs to the ventral type of pneumatization but can be derived from a very primitive condition only.

I The cladogram
The resulting cladogram (fig. 4) has been produced (PAVLINOV, 1982; PAVLINOV et ah, 1990) without using any computer program. It reflects relations among recent genera derived mainly from the data on dental and bullar morphology. The following peculiarities are to be stressed. Taking Gerbillidae as a family, two subfamilies can be defined, Taterillinae and Gerbillinae sensu stricto. The former contains also Desmodillus and Gerbillurus as closest relatives, the latter - the remainder genera, except probably Ammodillus. At the tribal level, two presumably monophyletic taxa are recognized in Taterillinae and four in Gerbillinae sensu stricto. Their contents are evident from the cladogram, so my additional comment might only be that Desmodilliscus and Pachyuromys both constitute monotypical tribes of their own. Notice also that both the middle ear and dental morphology place Sekeetamys with Rhombomyini rather than with Gerbillini sensu stricto. (PAVLINOV et al, 1990). At nearly the same time, a paleontologist, TONG (1989), published a phylogenetic tree of Gerbillidae which appeared to be very similar


148

African Small Mammals / Petits mammiferes africains

Tatera Taterillus Gerbllliscus Gerbillurus Desmodillus Ammodillus GerbUlus « Monodia » Mlcrodillus Desmodilliscus Pachyuromys Sekeetamys Meriones Brachiones Psammomys Rhombomys

I Figure 4 Cladistic relations among recent genera of Gerbillidae (after PAVLINOV et al., 1990) derived mainly from dental and bullar morphology.

to that of mine (PAVLINOV, 1982; PAVLINOV etal., 1990). The principal differences are the following. The genera Desmodillus, Ammodillus and Desmodilliscus are placed by TONG (1989) together among the gerbillines proper. Based upon the morphological data described here, this suggestion seems to be incorrect. The patterns of bullar morphology in Desmodillus and Desmodilliscus belong to different types of pneumatization; these genera also have distinct dentitions. The position of Ammodillus remains unclear. Th e genu s Pachyuromys is place d by TONG (1989 ) clos e to Brachiones, which is improbable: the latter is just a specialized relative of Meriones (PAVLINOV etal., 1990). Among the questions regarding within-gerbillid affinities, of special interest are the relations among taterillines. Usually, Tatera proper, Gerbilliscus and Taterona axe all considered congeneric, while Taterillus is given full generic rank. However, my data indicate that the Asian branch of mis group represented by Tatera sensu stricto may be quite isolated from the African branch including the remainding taterillines. It is characterized by peculiar combination of a more


I. J. A. PAVLINOV - Current concepts of Gerbillid phylogeny and classification

149 T

archaic dental pattern (initial stage of horseshoe anteroconid development) and a more advanced mastoid type. Contrary to this, Gerbilliscus and Taterona (together with Taterillus) possess the most advanced horseshoe anteroconid and a more primitive mastoid. The taxonomic conclusion might be that the genus Tatera is to be devoted to the Asian species only, while African taterans are to be included in a separate genus whose valid name should be Gerbilliscus sensu lato.

References
CHALINE J. , MEI N P. PAVLINOV I.J., 1981 --

andPETTERF., 1977 -- Les grandes lignes d'une classification evolutive des Muroidea. Mammalia, 41:245-252.
JACOBS L.L., 1978 --

Systematic position of the gerbillid genus Ammodillus Thomas, 1904. Zool. Zh., 60:472-474. In Russian.
PAVLINOV I.J., 1982 --

Fossil rodents (Rhizomyidae and Muridae) from Neogene Siwalik deposits, Pakistan. Bull. Mus. N.Arizona, 52:1-103 .
JAEGER J.-J., 1977 --

Phylogeny and classification of the subfamily Gerbillinae. Bull. Mosc. Soc. Natur., 87:19-31 . In Russian.
PAVUNOV I.J., 1984 --

Les rongeurs du Miocene moyen et superieur du Maghreb. Paleovertebr., 8:3-166 .
LAY D.L., 1972 -- The anatomy, physiology, functional significance and evolution of specialized hearing organs of gerbilline rodents. J. Morphol., 138:41-120.

Evolution of dental crown pattern in Gerbillinae. Arch. Zool. Mus. Moscow Univ., 22:93-134. In Russian.
PAVLINOV I.J., 1985 --

Contributions to gerbillid dental morphology and phylogeny (Rodentia: Gerbillinae). Zool. Zh., 64:574-582. In Russian.
PAVUNOV I.J., DUBROVSKY A.Y., POTAPOVA E.G. and ROSSOUM O O.L.,

MUSSER G.G.
and CARLETON M.D., 1993 --

Family Muridae. In WILSO N O.E., REEDER D.M. (eds): Mammal Species of the world: A Taxonomic and Geographical Reference. Smithsonian Institution Press, Washington & London: 501-756.
PAVUNOV I.J., 198fJ --

1990 -- Gerbillids of the World. Nauka Publ., Moscow. In Russian.
PETTERF. , 1956 --

Evolution du dessin de la surface d'usure des molaires de Gerbillus, Meriones, Pachyuromys et Sekeetamys. Mammalia, 20:419-426.
TONGH. , 1989 --

Evolution and taxonomic significance of the osseous middle ear in the subfamily Gerbillinae. Bull. Mosc. Soc. Natur., 85:20-33. In Russian.

Origine et evolution des Gerbillidae (Mammalia, Rodentia) en Afrique du Nord. Mem. Soc. Geol. France,

n.ser., 155:1-120.