HMW 7 - Family text: Dipodidae (Jerboas)

Family text: 

Class Mammalia

Order Rodentia

Suborder Myomorpha

Superfamily Dipodoidea

Family DIPODIDAE (Jerboas)


  • Small to medium-sized, bipedal and characterized by short forelimbs and long strong hindlimbs and long tails, often ending with black and white brush of long hairs.
  • 10–50 cm.
  • Palearctic and Afrotropical regions.
  • Terrestrial species, desert, semi-desert and steppe habitats.
  • 13 genera, 35 species, 91 taxa.
  • 1 species Vulnerable; none Extinct since 1600.



The superfamily Dipodoidea (Rodentia, Myomorpha) is the sister group of Muroidea. As noted by J. Michaux and colleagues in 2001, this close relationship, based on morphological and molecular data, confirmed the suprafamilial Myodonta concept, associating both superfamilies. As summarized by V. S. Lebedev and colleagues in 2013 and M. E. Holden and G. G. Musser in 2005, birch mice, jumping mice, and jerboas have been traditionally recognized in a single family Dipodidae or placed in up to six families within superfamily Dipodoidea. Four of them included morphologically specialized bipedal, arid-dwelling jerboas (Cardiocraniinae, pygmy jerboas; Euchoreutinae, Long-eared Jerboa; Dipodinae, three-toed jerboas; and Allactaginae, five-toed jerboas), and the other two were represented by more generalized quadrupedal taxa (Zapodinae, jumping mice, and Sminthinae, birch mice). Despite important effort from morphologists, the taxonomy and phylogeny of Dipodoidea remains controversial. More particularly, the family-level classification has long been a matter of debate, with the number of recognized families ranging from one to six. This lack of consensus on dipodoid taxonomy is linked by the fact that, until recently, phylogenetic relationships among the main lineages were not unambiguously established.

Traditional classifications based mainly on morphological similarities included two related families: Dipodidae, including all jerboas, and Zapodidae, including jumping mice and birch mice. This morphology-based system reflected the evolution of locomotory adaptations, with subfamilies (or families) corresponding to grades of evolutionary development from primitive quadrupedal to specialized bipedal locomotion. Other studies noted that a simple dichotomy between bipedal and non-bipedal taxa was inadequate to explain significant morphological variation within the superfamily. The use of traits that are not directly associated with locomotion appeared to be more reasonable to develop phylogenetic studies less subject to homoplasy. This approach was performed in 1992 by G. I. Shenbrot in a cladistic analysis based on characteristics of dentition, male reproductive systems, and auditory bullae. This study did not reveal any synapomorphies to support the monophyly of the bipedal taxa and proposed to divide Dipodoidea into four families: Allactagidae, Dipodidae (including Paradipus and Cardiocraniinae), Sminthidae (with Euchoreutinae), and Zapodidae. Following the same strategy, recent paleontological studies proposed another classification with three main families: Zapodidae (containing Sminthinae/Sicistinae and Zapodinae), Allactagidae (containing Allactaginae and Euchoreutinae), and Dipodidae (including Cardiocraniinae, Dipodinae, and the extinct Lophocricetinae, even though this last subfamily was included within Cardiocraniinae by some authors). These last two classifications suggested that the evolution of Dipodoidea was a complex process involving independent and parallel locomotory, trophic, and substrate adaptations.

Recent genetic studies based on several nuclear and mitochondrial markers resulted a new taxonomic revision and supported the recognition of only three families: Sminthidae (synonym of Sicistinae, 14 species), Zapodidae (five species), and Dipodidae, which included Cardiocraniinae (seven species), Euchoreutinae (one species), Dipodinae (eleven species), and Allactaginae (16 species) as subfamilies. This arrangement was chosen to emphasize the monophyly of all bipedal taxa within Dipodoidea, which appears extremely clear with molecular markers. It is also interesting that the number of species of Dipodidae is still highly debated and that several cryptic species probably still need to be identified. This is particularly the case for the subfamily Allactaginae. The use of new molecular and morphological markers and better sampling will help to decipher the correct number of species existing for this still poorly known rodent group.

Concerning their phylogeny, these recent genetic studies also highlighted the basal position of Sminthidae, followed by Zapodidae. This last family appears therefore as the sister group of the monophyletic jerboas. This result was already proposed in earlier morphological studies that generally considered Sminthidae as representing the most primitive dipodids morphologically. In contrast, this pattern contradicts the view that the non-bipedal birch mice (Sminthinae) and jumping mice (Zapodinae) constitute a monophyletic group, as proposed by several other morphologists and paleontologists. Indeed, two Oligocene lineages, Heosminthus-Plesiosminthus and Sinosminthus-Parasminthus, were generally considered to be the ancestral groups of Zapodinae-Sminthinae and Allactaginae-Dipodinae, respectively. Nevertheless, analyses performed by Lebedev and colleagues in 2013 clearly rejected the monophyly of the former association, thus suggesting that fossil data should be revised to determine which of the Paleogene taxa represent stem-groups of the three, but not two, main dipodoid clades.

Another important result based on recent morphological and molecular analyses concerns the monophyly of bipedal Dipodoidea, including Cardiocraniinae and Euchoreutinae. Although this group comprises highly divergent lineages from morphological or genetic points of view, its cohesiveness is notably supported by a stable karyotype (2n = 48) shared by all genera except Salpingotus and Stylodipus. The monophyly of bipedal Dipodoidea was often questioned, mainly due to the controversial phylogenetic position of the Long-eared Jerboa (Euchoreutes naso). It was generally considered as a single survivor of some ancient lineage. Some morphologists even suggested that Euchoreutes was sister to all other bipedal taxa based on intermediate characteristics between specialized jerboas (Dipodinae and Allactaginae) and Sminthidae (e.g. generalized zygoma and less specialized pelvis and hindlimb). Shenbrot in 1992 considered Euchoreutes as an independent derivative from an ancestral dipodoid stock that shared true synapomorphies only with Sicista. Following this hypothesis, similarity between Euchoreutes and other jerboas should be treated as a result of parallel (or convergent) evolution. Nevertheless, recent reanalyses of morphological data performed by Lebedev and colleagues indicated that such a position of the Long-eared Jerboa close to Sicista was not well supported. This was corroborated by molecular data that definitively supported the monophyly of Euchoreutinae + Dipodinae + Allactaginae, even if the morphological resemblance among these subfamilies was also a result of parallel evolution of bipedality in arid open and flat landscapes.

The last remaining question concerns the relationships among Euchoreutinae, Dipodinae, and Allactaginae. Some authors tend to propose that Euchoreutes is more related to Allactaginae; however, existing data do not contain sufficient information to confirm this hypothesis. Their relationship is now regarded as an unresolved trichotomy, which could be partly explained by the fact that the three lineages probably diverged in rapid succession within a relatively short time span. Indeed, the earliest fossils attributed to the Euchoreutinae were found from the early Miocene (16–20 million years ago), which is not significantly later than the earliest record for Allactaginae. Therefore, the split between the three jerboa lineages should have occurred quite quickly during the early Miocene.

The position of Cardiocraniinae within Dipodidae also has been a subject of strong debate. Previous paleontological studies suggested close affinities between Dipodinae and Cardiocraniinae. According to the hypothetical paleontological scenario, all extant jerboas descended from two ancestral lineages that had separated from each other by the end of Oligocene and evolved later into Allactaginae and Lophocricetinae. The now-extinct Lophocricetinae was regarded as the immediate ancestor of Cardiocraniinae and, indirectly, Dipodinae. The latter taxon was believed to have originated, in turn, from some early Cardiocraniinae. Nevertheless, recent molecular studies show that Cardiocraniinae is the sister group to all other jerboas and that the lineage leading to the pygmy jerboas was the first to split from the common ancestor of bipedal taxa. Therefore, Dipodinae is presently considered as more closely related to Allactaginae and would correspond to a lineage that appeared late in the jerboa’s evolution.

The monophyly of Cardiocraniinae has also been strongly debated. Indeed, different studies of the morphology of genital organs or cytological characteristics have shown that Cardiocranius and Salpingotus—the two modern genera of this subfamily—were substantially differentiated. Nevertheless, recent molecular studies confirmed the monophyly of Cardiocraniinae, even if the two genera were highly differentiated. This result confirms morphological results obtained with more comprehensive data sets. The first representatives of Cardiocraniinae are found in the early late Miocene (9·7–11·1 million years ago) or even late mid-Miocene (11·1–12·5 million years ago).

Considering phylogenetic relationships within the subfamily Dipodinae, morphological and genetic characteristics confirm its monophyly and converge to place the genus Dipus basal compared with the other genera Stylodipus, Jaculus, and Eremodipus. These latter genera are considered to be sister taxa. The only incongruence between morphology and genetics concerns the monotypic genus Paradipus that is characterized by a unique combination of autopomorphic and shared derived features. Considering different morphological studies, Paradipus was considered to be close to Eremodipus and Jaculus, a separate lineage just distantly related to other Dipodinae, or a sister group to Cardiocraniinae, indicating extensive parallelism between Paradipus and other three-toed jerboas. Molecular data support the second hypothesis, and Paradipus should be regarded as the most divergent lineage in Dipodinae.

Concerning the Allactaginae, Moore and colleagues demonstrated on the basis of molecular data, that the phylogenetic position of Scarturus was situated within the genus Paralactaga. Thus, the name Paralactaga coined by C. C. Young in 1927 became the junior synonym of Scarturus, named by Gloger in 1841. Therefore, it seems reasonable to replace the genus name Paralactaga by the genus name Scarturus, where three subgenera can be recognized: Scarturus, Paralactaga, and Microallactaga.

Moreover, Lebedev and colleagues, on the basis of molecular studies using several nuclear and mitochondrial markers, evidenced a lack of monophyly for the genus Allactaga sensu lato. Indeed, the morphologically distinct genus Pygeretmus appears as a sister group with the genus Scarturus, while the genus Allactodipus is grouped with the Great Jerboa (Allactaga major) and the Svertzov’s Jerboa (A. severtzovi). Therefore, Allactaga sensu lato should be subdivided into three separate genera by elevating their taxonomic rank from subgeneric to generic: Allactaga sensu stricto, Scarturus, and Orientallactaga. From a morphological point of view, this hypothesis cannot be rejected. Indeed, among the morphological characteristics considered, only the shape of glans penis can be counted as a potential synapomorphy for Allactaga sensu lato. The paleontological record indicates that the first unequivocal allactagines appeared in the lower Miocene 16–20 million years ago, while the earliest fossils attributed to the genus Scarturus (subgenus Paralactaga) appeared 7·5–8·7 million years ago. The genera Protalactaga, Proalactaga, and Scarturus (subgenus Paralactaga), based on interpretation by most paleontologists, represent stages of evolutionary development of molar crown patterns (grades) rather than real phylogenetic clades. Therefore, the roots of different lineages within the genus Allactaga sensu lato could be significantly deeper than it is usually believed.

According to Lebedev and colleagues, it is commonly accepted that Pygeretmus is a rather recent (latest part of the Miocene) derivative of Allactaga, sensu lato, which evolved toward a more specialized herbivore diet. In contrast to that, Allactodipus was hypothesized to be an early offshoot of the ancestral allactagine stock, thus representing a separate evolutionary lineage. Molecular data strongly contradict this view, indicating instead that the Allactodipus lineage originated at about the same time as major clades within Allactaga. Allactodipus shares some similarity in pelvis shape with the Great Jerboa and Severtzov’s Jerboa (Allactaga severtzovi). Although this finding is in line with the pattern inferred from genetic data, the phylogenetic value of this condition remains unclear.

Another important question is whether bipedality was acquired only once or evolved independently in several dipodoid lineages. Available paleontological data do not provide direct support for either of these hypotheses. According to Lebedev and colleagues, the most parsimonious scenario, based on molecular studies, suggests that the last common ancestor of extant jerboas was bipedal. Moreover, the molecular phylogeny, while at odds with most morphological characteristics, is compatible with the general trajectory of locomotory evolution in dipodoids leading from non-ricochetal Sicista via ricochetal but quadrupedal zapodines to primitively bipedal cardiocraniines and finally to highly specialized Dipodinae and Allactaginae. Thus, it appears that each stage of progressive specialization might have been attained just once. Nevertheless, although available genetic data do not support the hypothesis of independent acquisition of key morphological adaptations to saltatory locomotion in different branches of Dipodoidea, adaptive parallelisms could have played certain roles in its locomotory evolution.

Concerning the evolutionary history of Dipodoidea, the first occurrences of this superfamily in the fossil record are from North America with Elymys (?Zapodidae, middle Eocene, 40–45 million years ago) and Simimys (Simimyidae, middle to late Eocene, 38–40 million years ago). In Asia, the oldest dipodoid representatives are Heosminthus (Zapodidae or Dipodidae depending on the studies, middle Eocene to late Oligocene, 28–40 million years ago) and Sinosminthus (Zapodidae, middle Eocene to middle Miocene, 15–40 million years ago). Whether these genera belong to extant taxa or represent extinct sister groups of Dipodoidea has yet to be determined. A recent study based on a single genetic marker proposed that the diversification of modern dipodoids took place during the middle Eocene (40–45 million years ago). Combining a time-calibrated phylogeny with a compilation of the fossil record further suggested that diversification events and distributional expansions were mostly influenced by new ecological opportunities triggered by an increasing aridity and the development of open habitats. Indeed, birch mice (Sminthidae) were shown to diversify during the warming period of the Oligocene–Miocene (24 million years ago), while jumping mice (Zapodidae) and jerboas (Dipodidae) were assumed to have radiated during the global cooling following the mid-Miocene climatic optimum (15 million years ago). In another study based on important molecular markers, the origin of modern Dipodoidea was estimated to have occurred during the early Oligocene (34 million years ago), which provides a different evolutionary history. A more recent study by J. Pisano and colleagues in 2015 based on five coding genes and 34 Dipodoidea species suggested that Dipodoidea and Muroidea diverged in the late Paleocene (57 million years ago). They also confirmed that modern Dipodoidea originated about 32 million years ago. This estimation appears to be the most realistic because the numbers of species and molecular markers were much more important compared with previous studies.

Furthermore, fossil calibrations used to estimate times of divergence were also more complete and precise. According to this last study, modern Dipodoidea would have diversified rapidly after their appearance during the early Oligocene (32 million years ago) in the proto-Himalayan region. At that period, global temperatures decreased significantly, and the Antarctic ice-sheets expanded rapidly. In the Palearctic, this severe cooling resulted in the development of open grassland that induced a great worldwide mammalian faunal turnover (called the Mongolian Remodeling in Asia). The Oligocene fauna was then reorganized, and faunas dominated by lagomorphs and rodents (e.g. Dipodoidea, Cricetidae, or other rodent taxa) replaced the perissodactyl-dominated fauna of the Eocene. The ancestral dipodoid group would have split about 32 million years ago into two distinct lineages, leading to the ancestors of birch mice (Sminthidae) in the Himalaya–Tibetan Plateau region and to the common ancestors of jumping mice and jerboas in the Gobi and Taklamakan desert regions. These diversification events would have been directly linked to the collision between India and Asia, which happened 40 million years ago and caused significant uplift episodes of Himalaya.

About 15–17 million years ago, many subfamilies had differentiated or undergone radiations: radiation of Sminthidae and Zapodidae, divergence of Euchoreutinae, and split between Allactaginae and Dipodinae. Biogeographical analyses indicated that these diversification events happened in the Gobi and Taklamakan desert regions, except for Sminthidae that diversified in the Himalaya–Tibetan Plateau region. That period was associated with the most intense orogenic phase of Himalaya, which, together with the mid-Miocene Climatic Optimum, favored year-round aridity in Central Asia and led to the formation of many deserts in this region. Aridity and establishment of open-land habitats were also suspected to have triggered the early evolutionary history of the Dipodoidea. The most recent common ancestor of modern Dipodinae would have also diversified in the Gobi and Taklamakan desert regions about 11·1 million years ago. While ancestors of the basal Comb-toed Jerboa (Paradipus ctenodactylus) would have moved to Central Asia, ancestors of all other Dipodinae would have split into two distinct groups about 8·6 million years ago. Actually, at that period, a significant increase in elevation of the Tibetan Plateau enhanced aridity in Central Asia. This aridification event would have driven the diversification of Dipodinae—species of which now prefer arid habitats.

While ancestors of Lichtenstein’s Jerboa (Eremodipus lichtensteini) would have settled in their native region of Central Asia, ancestors of the modern Greater Egyptian Jerboa (Jaculus orientalis) would have moved to North Africa after their differentiation about four million years ago. Besides, their arrival in Africa is congruent with those of other rodents that likely followed the same migration routes from Asia for a few million years. Common ancestors of the Lesser Egyptian Jerboa (Jaculus jaculus) and Blanford’s Jerboa (Jaculus blanfordi) would have expanded their distributions from Central Asia to North Africa about four million years ago. Given the simultaneous diversification of this ancestral group and decreasing temperatures associated with the onset of the Pleistocene glaciations, aridification and development of open landscapes would have once again triggered the diversification of these rodents and probably triggered the split into two lineages about 3·4 million years ago, leading to the Lesser Egyptian Jerboa in North Africa and Blanford’s Jerboa in Central Asia. Z. Boratyński and colleagues in 2012 noted the co-existence of two cryptic species within the Lesser Egyptian Jerboas living in North Africa. They were named Jaculus jaculus and J. deserti. Nevertheless, a new study by Shenbrot and colleagues in 2016 described the presence of these two species in Israel, eastern Egypt, and the Sinai Peninsula, and according to a comparison of available type specimens from these regions, they proposed to rename J. deserti to J. hirtipes (African Hammada Jerboa), following taxonomic norms. Comparisons of geographical and habitat differences of the two species revealed high niche divergence between them, slightly higher in the sympatric North African populations than in the parapatric populations of Israel and Sinai Peninsula. Moreover, genetic mitochondrial and nuclear markers clearly found an important level of genetic divergence among the populations, corresponding to values generally observed among differentiated mammalian species. Both species probably originated near the upper Pliocene–lower Pleistocene boundary (2–4 million years ago). Their differentiation could be explained by vicariant (separation) events, which happened during climate fluctuations that characterized the end of the Tertiary era. They diversified during subsequent periods of aridification occurring in North Africa. A. Ben Faleh and colleagues in 2012 dated these diversification events to 0·23–1·1 million years ago, suggesting that the middle Pleistocene climatic change and its environmental consequences affected the evolutionary history of these two African jerboas. Interestingly, the expansion of the Lesser Egyptian Jerboa to its current distribution probably predated that of the African Hammada Jerboa and is estimated at 42,000–98,000 years ago (versus 19,000–45,000 years ago for the African Hammada Jerboa).

About 7–8 million years ago, the expansion of C4 grasses to the detriment of C3 plants favored the replacement of many woodland-adapted mammals by more open-habitat representatives. Dipus and Stylodipus spp. probably would have been favored through this transition from C3-dominated and C4-dominated plant cover, given their preference for open-land habitats. Then, ancestors of modern Northern Three-toed Jerboas (Dipus sagitta) settled in Central Asia, as did the common ancestors of all modern Stylodipus spp. Ancestors of the Mongolian Three-toed Jerboa (Stylodipus andrewsi) would have then moved to Mongolia while the common ancestors of the Thick-tailed Three-toed Jerboa (Stylodipus telum) and the Dzungarian Three-toed Jerboa (Stylodipus sungorus) first extended their distributions to Mongolia. By vicariance, these ancestors diversified about 1·4 million years ago, leading to the Thick-tailed Three-toed Jerboa in Central Asia and the Dzungarian Three-toed Jerboa in Mongolia.

In summary, the diversification of Dipodoidea can be associated with major uplift episodes that occurred in Central Asia and the Himalaya–Tibetan Plateau region due to the collision of India with Asia. These events also induced diversification events in many other groups. Other important diversification events (e.g. divergence between Zapodidae and Dipodidae in Central Asia) took placed during the Eocene–Oligocene transition when the global temperature decreased significantly and faunas dominated by lagomorphs and rodents replaced Eocene perissodactyl-dominated faunas. All of these climatic and geological disruptions in the Central Asia and Himalaya–Tibetan Plateau region modified landscapes and offered new habitats that favored diversification events, thus triggering the evolutionary history of Dipodoidea.


Morphological Aspects

Jerboas are small or medium-sized and bipedal, and they are characterized by short forelimbs and long strong hindlimbs. The forelimbs are not used for locomotion but are used to gather food and dig burrows. The foot bones are often fused into a single long cannon bone, which is advantageous for jumping. Jerboas are particularly adapted to running fast in desert habitats. They move either by jumping or walking on their hindlegs. Their tails often end with black and white brushes of long hairs and are generally longer than their head and body lengths to balance movements when they jump. They also use their tails as props when they sit upright. In contrast, tails among pygmy jerboas (Cardiocraniinae) and Lesser Fat-tailed Jerboas (Pygeretmus platyurus) are relatively short and fat and without hairs brushes. Jerboas have large heads, with wide and flat muzzles and big eyes. The family is also characterized by long vibrissae and auricles (external ear pinnae) that vary, depending on the subfamily, from relatively short for three-toed jerboas (Dipodinae) to very long for the Long-eared Jerboa (Euchoreutes naso, Euchoreutinae). Cardiocraniinae have particular short and tubuliform auricles. Long ears enhance hearing used to avoid nocturnal predators and to hunt insects, like moths, during the night (e.g. Long-eared Jerboa). Long ears also are probably used for thermoregulation during hot weather. Other morphological characteristics of jerboas is the absence or near absence of necks and their compact bodies. In some subfamilies, neck vertebrae are even fused (e.g. Cardiocraniinae and Dipodinae).

Jerboas have dense light-brown fur, with lighter sand-colored fur on their backs and white fur on their ventral surfaces, usually matching the environments in which they live. Birch mice, jumping mice, and five-toed jerboas are characterized by short claws, in contrast to three-toed jerboas that have long and narrow claws. Hindfeet of jerboas have 3–5 toes. For species with more than three toes, the first and fifth toes are much shorter than the three middle ones. When running, three-toed and five-toed jerboas have different adaptations. All toes on the hindlimbs of three-toed jerboas are in contact with the soft substrate surface. They also have often brushes of long hairs on sides of their toes to stabilize the foot in soft sand. In contrast, only the tall compact callus on top of the central toe of a five-toed jerboa has contact with the hard substrate. The second and fourth toes are shorter and act as shock absorbers.

Skulls of dipodids have enlarged infraorbital foramina. Like other bipedal animals, the foramen magna (hole at the base of the skull) of jerboas are forward-shifted, which allows two-legged locomotion. Skulls do not have well-developed zygomatic plates and are characterized by a sciurognathous lower jaw. The angular process of the lower jaw is characterized by a thin bone that is often perforated. Auditory bullae are small and simple (one chamber) in species of Sminthidae, Zapodidae, Allactaginae, and the most primitive Dipodinae (Dipus and Stylodipus), and enlarged and subdivided into several chambers in Euchoreutinae, Cardiocraniinae, and advanced Dipodinae (Paradipus, Jaculus, and Eremodipus). The dental formula of dipodoids is I 1/1, C 0/0, P 0–1/0, M 3/3 (×2) = 16 or 18. The large upper incisors are smooth in all subfamilies of Dipodidae. Lower premolars are normally absent but are present in rare individuals (less than 1%) of Great Jerboas and Northern Three-toed Jerboas. C. Charles and L. Viriot in 2007 found a rare case of a supernumerary molar (M4) in one Greater Egyptian Jerboa. Upper first premolars (P1) are well-developed in Euchoreutes, Cardiocraniinae, Dipus, Mongolian Three-toed Jerboas, Allactaga, Orientallactaga, Allactodipus, and Scarturus and absent in Pygeretmus, Thick-tailed Three-toed Jerboas, Dzungarian Three-toed Jerboas, Jaculus, Eremodipus, and Paradipus. Crown heights of unworn molars are significantly less that their length in Cardiocraniinae and Dipus, about equal to their length in Allactaga, Orientallactaga, Scarturus, Stylodipus, and Jaculus, or significantly more that their length in Allactodipus, Pygeretmus, Eremodipus, and Paradipus. Masticatory surfaces of low-crowned molars are usually tubercular (bunodont molars), except for Euchoreutes in which molars are ratchet-like (low-crowned molars with tuberculous surfaces and tuberculas having sharp-pointed tops). Species with medium-crowned molars have terraced masticatory surfaces, and species with high-crowned molars are characterized by flat masticatory surfaces. For most species, the emerging root and the end of crown growth occur at early stages of postnatal development (less than one month). Only in the Comb-toed Jerboa do roots start to develop and crowns stop growing in height at 1–1·5 years of age.

Karyologically, numbers of chromosomes of Dipodidae vary from 2n = 46 in species of Salpingotus to 2n = 58 in species of Stylodipus. Interestingly, Allactaginae, Euchoreutinae, Cardiocraniinae, and a part of Dipodinae—the Northern Three-toed Jerboa, Lichtenstein's Jerboa, the Greater Egyptian Jerboa, Blanford’s Jerboa, the Lesser Egyptian Jerboa, the African Hammada Jerboa, and the Comb-toed Jerboa—are characterized by homogeneous chromosomal numbers, with a constant value of 48 chromosomes.

Research has focused on the morphology and shape of the male sex organ to classify dipodoids. In particular, studies have focused on the general shape of the glans penis, its roughness, and its lobe structure; presence of large stylet-shaped, forward-directed thorns or a single-vertex, backward-directed aciculae; and shape of the baculum (os penis). These studies have highlighted a closer relationship between Sminthidae and Zapodidae compared with Dipodidae, which appears more distant using shapes and sizes of bacula. It was less clear with other morphological characteristics, where shapes often overlapped among families and subfamilies. Nevertheless, some of them, like the general shape of the glans penis, helped to confirm the monophyly of different genera of jerboas such as Cardiocranius, Salpingotus, Paradipus, Allactodipus, and Pygeretmus.



Jerboas are distributed in deserts, semi-deserts, and steppes of North Africa and Eurasia. These areas are characterized by moving sands, clay depressions, rocky-gravel plateaus, and dry mountain slopes. Most species are habitat-specific, but some are less selective and can be found in diverse environments. Five-toed jerboas prefer areas with relatively hard soils. Sand-dwelling specialists (or psammophilous species) include species of Paradipus, Eremodipus, or Salpingotus. Others dipodids are habitat generalists (Dipus, Stylodipus, and Lesser Egyptian Jerboa) or specialize on hard clay substrates (Blanford’s Jerboa and Greater Egyptian Jerboa). The Northern Three-toed Jerboa in its western and north-eastern distribution inhabits sand massifs at different stages of sand stabilization (from moving non-stabilized sands to pine forests on sand dunes) in desert, semi-desert, and steppe zones, but it always needs patches of pure sand without grass or forb vegetation. The Long-eared Jerboa inhabits flat sandy and sandy-gravel terraces in true and extra-arid deserts, but it avoids non-stabilized sands. The Five-toed Pygmy Jerboa (Cardiocranius paradoxus) inhabits gravel plains in lower parts of mountain foothills with mat-grass vegetation in semi-deserts and deserts. In contrast, other species such as the Euphrates Jerboa (Scarturus euphraticus) prefer to live in steppe and semi-deserts or subtropical and tropical dry lowland grasslands. It cannot tolerate desert habitats and is highly sensitive to habitat change, mainly due to agricultural extensions in Near East and Middle East regions.

Jerboas sleep during day in individual burrows with entrances closed by soil plugs. This protects them against the heat of the day. They leave their burrows at night when ambient temperatures are cooler. Pygmy jerboas build plugs with their tails, but all other species of jerboas use the muzzles. Activity patterns are strongly nocturnal for the genera Cardiocranius, Eremodipus, Euchoreutes, Salpingotus, and Allactaga. In contrast, the Northern Three-toed Jerboa is nocturnal but also crepuscular. Its aboveground activity starts at sunset in summer or 20–90 minutes after sunset in spring and autumn. In spring and summer, activity lasts 7–9 hours and ends at sunrise. In autumn, individuals are active only during the first one-half of the night, and length of their activity can decrease to only one hour. For Cardiocranius, activity lasts 4–6 hours.


General Habits

All jerboas are bipedal and also move by running and jumping. They can run very fast for long distances, depending on the species. Maximum running speeds of jerboas are 40–50 km/h for the Great Jerboa, 36 km/h for Severtzov’s Jerboa, 33 km/h for Blanford’s Jerboa, 32 km/h for the Comb-toed Jerboa, 29 km/h for the Northern Three-toed Jerboa, 30 km/h for the Small Five-toed Jerboa, 31 km/h for Bobrinski’s Jerboa (Allactodipus bobrinskii), 29 km/h for the Northern Three-toed Jerboa, 28 km/h for the Dwarf Fat-tailed Jerboa (Pygeretmus pumilio), 26 km/h for Lichtenstein’s Jerboa, 19 km/h for the Lesser Fat-tailed Jerboa, and 6 km/h for the Thick-tailed Pygmy Jerboa (Salpingotus crassicauda). Big jerboas such as the Comb-toed Jerboa, Bobrinski’s Jerboa, and the Small Five-toed Jerboa can bound up to 3 m in length and 1 m in height when predators chase them. Maximum lengths of jumps are 302 cm for the Comb-toed Jerboa, 291 cm for Severtzov’s Jerboa, 283 cm for Blanford’s Jerboa, 211 cm for Bobrinski’s Jerboa, 202 cm for the Small Five-toed Jerboa, 200 cm for the Northern Three-toed Jerboa, 196 cm for Lichtenstein’s Jerboa, 95 cm for the Dwarf Fat-tailed Jerboa, and 54 cm for the Lesser Fat-tailed Jerboa.

Jerboas never move in a straight line, and they change direction constantly, moving in a zig-zag pattern to confuse their numerous enemies. They can run for hundreds of meters to escape danger. Another defensive tactic, notably observed in the Euphrates Jerboa, is energetically jumping off their hindlegs into the air in huge bound. Some species (e.g. Small Five-toed Jerboa and Northern Three-toed Jerboa) also escape danger by jumping into and climbing through the shrub canopy. Most jerboas also use shelter burrows to escape. Densities of shelter burrows and frequencies of their use are highest in the Greater Fat-tailed Jerboa (Pygeretmus shitkovi), the Dwarf Fat-tailed Jerboa, Vinogradov’s Jerboa (Scarturus vinogradovi), and the Thick-tailed Three-toed Jerboa. Maximum speed for pygmy jerboas (e.g. Five-toed Pygmy Jerboa and Pallid Pygmy Jerboa, Salpingotus pallidus) is much slower than other jerboas and does not exceed 9 km/h. These species generally move using small jumps with maximum lengths of 3–5 cm during normal foraging, but they can jump as high as 20–30 cm straight up into the air when disturbed. During these movements, hindfeet work simultaneously, and forelimbs never touch the substrate. Pygmy jerboas cannot run for long distances, so they try to escape predators by hiding and lying on the ground under shrubs or bushes; they never use shelter burrows to escape.

Three types of locomotion are identified in jerboas. First, synchronous ricochet jumps involve push offs and touchdowns with both hindfeet in synchrony; distal ends of left and right hindfeet are placed on the line perpendicular to direction of movement. This type of locomotion is typical for all modes of movement of Cardiocranius and Salpingotus and slow movement of Dipus and Allactodipus. Second, asynchronous ricochet jumps involve push offs and touchdowns with hindfeet working consecutively; in this case, push off by one hindfoot is accompanied by synchronous carrying forward of the second hindfoot. Resulting distances between supporting points of left and right hindfeet (“acceleration step”) are 4–5 cm in the Lesser Fat-tailed Jerboa and 16–18 cm in Severtzov’s Jerboa. This type of locomotion is typical for fast running by most species of Allactaginae and Dipodinae because it provides additional acceleration to the bipedal gallop. Third, bipedal running (or pacing) with alternating support by left and right hindfeet. This type of locomotion is typical for slow movement by most species of Allactaginae and Dipodinae and for all manners of movement by Paradipus.

Jerboas make their nests in burrows, and these can be complex, with different side-chambers. They can construct four types of burrows: day burrows used to sleep during the day in summer and where females keep their offspring during the reproduction season; shelter burrows used to escape predators during the night; temporary summer shelter burrows used for cover when foraging during the day; and wintering burrows used for hibernation. Day and hibernation burrows are similarly constructed. They differ mostly in depth and length, with a greater depth in hibernation burrows. Several chambers (up to four) are constructed at different depths and are typical of wintering burrows. Indeed, during hibernation, jerboas change chambers to find optimal temperature for hibernation. Three-toed jerboas mostly use forelimbs to dig their burrows in sandy soils. In contrast, five-toed jerboas use their incisors to dig the hard soils where they generally live, and they use their forelimbs and snout to excavate soil from the tunnel. Burrows of three-toed jerboas are often complex. They can have 2–3 tunnels, 1–2 main chambers, 1–2 additional chambers, and 2–3 emergency exits.

When a jerboa is in its burrow, exits of the tunnel are closed with soil plugs and are almost invisible. These plugs have different purposes for jerboas: hiding from predators and protection from hot summer air. In some cases in the nesting burrows, additional soil plugs are constructed, separating the nest chamber from other parts of the burrow. Burrows of five-toed jerboas are generally simpler and usually have two tunnels. The first tunnel is parallel to ground surface, with the initial part filled by soil at one end; its main entrance is at the opposite end; and an emergency exit is near the internal end of the initial tunnel but not completely dug up to the soil surface. The second tunnel starts from the middle part of the first tunnel and goes down to the nest chamber. Nest chambers are constructed from dry vegetation or, in the case of desert jerboas, camel hairs. In a similar way, day burrows are generally more complex for three-toed jerboas and pygmy jerboas compared with those of five-toed jerboas. Generally, day and night shelter burrows of five-toed and several three-toed jerboas are simple short tunnels with open holes used to escape predatory attacks. During the rainy season, they generally dig their burrows in small hills to reduce the risk of flooding.

Most species of birch mice, jumping mice, and jerboas hibernate for at least one-half of the year. They survive using fat that they build up during summer. Duration of hibernation depends on the geographical region and species. During hibernation periods, their body temperatures can decrease to 2–3°C, and they can lose more than 50% of their weight. This period is therefore highly sensitive for many dipodid species, and their mortality rates are high. For jerboas, the hibernation period is also related to their reproductive cycle. For species having a single breeding period and living in cold Asian deserts (e.g. Five-toed Pygmy Jerboa), hibernation can be extremely long from late August to May–June. For species with two breeding periods in spring and autumn, hibernation can be less than six months.

Most species of jerboas, including the subtropical desert species, the Greater Egyptian Jerboa, are obligate hibernators. Some species with wide distributions (e.g. Small Five-toed Jerboa, Dwarf Fat-tailed Jerboa, and Northern Three-toed Jerboa) are facultative hibernators. They always hibernate in northern parts of their distributions, but in southern parts, they hibernate in cold winters and do not hibernate in warm winters. Hibernation times can vary depending on latitude. For example, populations of Lichtenstein’s Jerboa in southern parts of its distribution hibernate 3·5 months from end of November to early March, but in northern parts, hibernation lasts 5·5–6 months from September to early April. Generally, hibernation periods start in September–October depending on the genera and their distributions. The end of hibernation is more variable and depends of climatic conditions. Species of several genera, such as Salpingotus or Stylodipus, are not known to hibernate.

Dipodids have numerous predators that vary depending on where they occur. The most common predators are mid-sized mammals such as foxes (e.g. Vulpes vulpes, V. corsac, and V. rueppellii), jackals, wild cats (e.g. Caracal caracal, Felis silvestris, and F. margarita), weasels (e.g. Martes foina, Vormela peregusnaMustela eversmanii, M. sibirica, and M. erminea); different birds of prey such as owls (e.g. Tyto alba, Bubo bubo, and Athene noctua), hawks and falcons (e.g. Buteo lagopus, B. vulpinus, Aquila heliaca, and Falco cherrug); and snakes (e.g. saw-scaled vipers, Echis spp.; Elaphe dione; Rhagerhis moilensis; and Spalerosophis diadema).



Little is known about communication systems of dipodids. Many species of dipodids use dust baths, and even if they do not have well-developed skin glands for scent marking, dust bathing is often considered a kind of chemical communication. Other specific marking behavior, such as touching substrate with the anogenital area, has been observed in the Long-eared Jerboa. Good hearing of dipodids suggests they may use sounds and vibrations to communicate with one another, even if they are generally silent. Tactile communication probably exists between mates and between mothers and their young.


Food and Feeding

Most dipodids are omnivorous, with diets of vegetation, fruits, fungi, insects, and relatively soft seeds. Some species are mostly insectivorous, particularly eating beetles or moths. Dipodids do not store food. Some species of jerboas, such as the Great Jerboa, Severtzov’s Jerboa, and the Siberian Jerboa (Orientallactaga sibirica), are omnivorous, feeding on what they find in arid habitats of Central Asia. They mostly eat seeds, fruits, fungi, and insects, but they also eat plant parts such as green stems and leaves, roots, and bulbs. The Northern Three-toed Jerboa has a relatively generalized diet composed of seeds and flowers (36–73%, average 51%), green plant material (16–54%, average 35%), roots and bulbs (0–40%, average 10%), and insects (1–11%, average 4%). Composition of plant species in diets of Northern Three-toed Jerboas varies geographically and seasonally. Locally, Northern Three-toed Jerboas include 15–40 species of shrubs, grasses, and forbs in their diets.

Diets of other species of jerboas range from near-pure insectivory where plant material is less than 5% of the diet (e.g. Long-eared Jerboa) to granivory–insectivory where seeds and insects are eaten in roughly equal amounts (e.g. Thick-tailed Pygmy Jerboa and Kozlov’s Pygmy Jerboa, Salpingotus kozlovi), near-pure granivory where mainly seeds of Stipa sp. (Poaceae) are eaten (e.g. Five-toed Pygmy Jerboa), granivory–folivory (e.g. Thick-tailed Three-toed Jerboa, Small Five-toed Jerboa, and Lichtenstein’s Jerboa), and folivory (e.g. Bobrinski’s Jerboa, Pygeretmus sp., and Comb-toed Jerboa).

Hunting strategies of insectivorous jerboas are sometimes impressive. Long-eared Jerboas and Siberian Jerboas can precisely locate flying insects such as moths, using their long and flexible auricles, and then catch them after fast vertical jumps. Kozlov’s Pygmy Jerboa uses its large auricles to detect the presence of insect larvae in sand dunes, and it uses its nose and long vibrissae to dig in the sand until it catches its prey. The Comb-toed Jerboa forages by climbing in Haloxylon (Amaranthaceae) shrubs, using its forelimbs and incisors to cut green stems at heights up to 1·5 m.



Dipodids generally give birth to litters of 2–9 young after gestations of 19–42 days. They breed 1–3 times/year, depending on the species. Jerboas are polygamous, and their breeding biology can be divided into three groups. Species in the first group breed only one time in spring or summer (warm season). This group includes Lichtenstein’s Jerboa, Kozlov’s Pygmy Jerboa, the Lesser Fat-tailed Jerboa, the Thick-tailed Three-toed Jerboa, and the Mongolian Three-toed Jerboa. Species in the second group have one and, less regularly, two litters during the warm period and include the Great Jerboa, the Balikun Jerboa (Orientallactaga balikunica), the Gobi Jerboa (Orientallactaga bullata), the Siberian Jerboa, the Thick-tailed Pygmy Jerboa, Heptner’s Pygmy Jerboa (Salpingotus heptneri), the Pallid Pygmy Jerboa, the Long-eared Jerboa, and the Lesser Fat-tailed Jerboa. Five-toed Pygmy Jerboas very rarely have two litters during spring or summer. Litters of species in these two groups have 2–9 young, with an average of five young. Their sexual maturity occurs about a year later after the first hibernation, and mating usually occurs after awakening from hibernation. Species in the third group breed 2–3 times a year. Overwintering individuals in this group can produce two litters in spring and early summer without interruption (only one in the case of the Greater Fat-tailed Jerboa) and one in autumn after an interruption in breeding in late summer. This group includes the Small Five-toed Jerboa, Severtzov’s Jerboa, Bobrinski’s Jerboa, the Northern Three-toed Jerboa, the Dwarf Fat-tailed Jerboa, and the Greater Fat-tailed Jerboa. Litters in this group have 1–8 young, with an average of four young. Juveniles born in spring can produce one litter in the following autumn.

Gestation is only known for several species. It is generally long, varying from 19–20 days for the Thick-tailed Three-toed Jerboa to 25–42 days for the Four-toed Jerboa (Scarturus tetradactylus) and intermediate at 25–30 days for the Northern Three-toed Jerboa, 25–35 days for the Gobi Jerboa and the Long-eared Jerboa, 28–30 days for Severtzov’s Jerboa, and 30 days for Bobrinski’s Jerboa. Weaning seems to be long for jerboas. For example, for species such as the Small Five-toed Jerboa and the Great Jerboa, young are nursed and cared for during their first 30–45 days. A plausible explanation of such long parental care is that the motor skills needed for bipedal locomotion must be completely developed before young can leave their burrows. As soon as they leave their nests and after just few nights outside, they become independent of their mothers. An interesting social behavior among a mother and her young has been recorded for at least in one species, the Thick-tailed Three-toed Jerboa. Upon existing their burrow and for some days after, young in a “train” formation: the first young holds onto the base of its mother’s tail, the second young holds onto the tail of the first young, and so on. The same order of young is retained when the "train" returns to their burrow.


Movements, Home range and Social organization

Dipodids are typically solitary, and every individual has its own burrow for sleeping and hibernating. Antagonistic behaviors vary among species. In many species of jerboas, individuals avoid contact but fight with each other in overlapping home range areas. Some species of pygmy jerboas, such as the Thick-tailed Pygmy Jerboa, can be territorial and aggressive toward neighbors. Home range sizes of jerboas vary among species and depend on the size of the species and the habitat productivity in which it lives. Sizes also vary geographically and with changes in density. Smallest home ranges (0·3–0·5 ha) are typical for species of Cardiocraniinae, the Thick-tailed Three-toed Jerboa, Kozlov’s Pygmy Jerboa, and the Greater Fat-tailed Jerboa. The largest home ranges (up to 28 ha) are observed for the Great Jerboa, the Comb-toed Jerboa, and Lichtenstein’s Jerboa. For the latter species, foraging occurs in small patches of 4–15 m2 distributed throughout the home range, and during one night, individuals can visit 3–5 patches.

Home ranges of Northern Three-toed Jerboas in relatively productive habitats are subdivided into two functional parts: the core with living burrows and foraging areas and the peripheral area that is searched when needed. In less productive habitats, all of the home range is used with equal intensity. Home range sizes of Northern Three-toed Jerboas vary from 4 ha to 19–22 ha for males and 2·9 ha to 12–15 ha for females. At low and moderate densities, home ranges of females are isolated, whereas home ranges of males overlap with home ranges of other males and females. At high densities, all home ranges widely overlap, regardless of individual’s sex. Nightly movements total 1–2 km for females and 4–8 km, sometimes up to 11 km, for males and subadults.

In most species of dipodids, home ranges of males are 1·5–4 times larger than those of females. Usually, the home range of one male overlaps the home ranges of several females (except for the Gobi Jerboa where the home range of one male overlaps the home range of only one female), and home ranges of females are non- or marginally overlapping. Home ranges of males are also non- or marginally overlapping in the Long-eared Jerboa, the Thick-tailed Pygmy Jerboa, the Thick-tailed Three-toed Jerboa, Severtzov’s Jerboa, the Gobi Jerboa, and the Dwarf Fat-tailed Jerboa, but widely overlap in the Comb-toed Jerboa, the Northern Three-toed Jerboa, the Small Five-toed Jerboa, and the Balikun Jerboa. In some species (e.g. Small Five-toed Jerboa and Dwarf Fat-tailed Jerboa) at high densities, all home ranges widely overlap.


Relationship with Humans

Although dipodids have important roles in many ecosystems in Eurasia, they have very little significance to humans, except some species are kept as pets (e.g. pygmy jerboas) and eaten. Jerboas are traditional food of Bedouin and Tuareg people in the Sahara Desert and Middle East. The Euphrates Jerboa is hunted and eaten by humans in some regions of Turkey and Jordan and used as food for captive falcons. Some species of jerboas can become abundant pests in arid and desert environments in western North Africa. Jerboas are common pests of watermelon and melon crops in southern Russia, Kazakhstan, and Central Asia.

Recent studies in Mongolia found the bacteria Yersinia pestis, the responsible agent of the sylvatic plague, in different jerboas such as the Siberian Jerboa and the Five-toed Pygmy Jerboa. Other studies have found this bacteria in several other species: the Great Jerboa, the Small Five-toed Jerboa, the Lesser Fat-tailed Jerboa, the Dwarf Fat-tailed Jerboa, the Northern Three-toed Jerboa, the Thick-tailed Three-toed Jerboa, Lichtenstein’s Jerboa, and the Comb-toed Jerboa. These results suggest that jerboas can promote the expansion of an epizootic but they cannot be significant reservoirs for the plague because the bacteria causes their death in less than 24 hours. Many other pathogens have also been detected in dipodids, such as tularemia (Great Jerboa), Omsk hemorrhagic fever (Great Jerboa), Q fever (Small Five-toed Jerboa), toxoplasmosis (Small Five-toed Jerboa), tick spirochetosis (Small Five-toed Jerboa), cutaneous leishmaniasis (Severtzov’s Jerboa and Small Five-toed Jerboa), and brucellosis (Dwarf Fat-tailed Jerboa).


Status and Conservation

Four species of jerboas have special conservation status on The IUCN Red List. The Four-toed Jerboa is classified as Vulnerable, and the Euphrates Jerboa, Vinogradov’s Jerboa, and the Greater Fat-tailed Jerboa are classified as Near Threatened. The Iranian Jerboa (Allactaga firouzi), the Five-toed Pygmy Jerboa, Thaler's Jerboa (Jaculus thaleri), the Balochistan Pygmy Jerboa (Salpingotus michaelis), Heptner’s Pygmy Jerboa, and the Pallid Pygmy Jerboa are classified as Data Deficient on The IUCN Red List, making assessment of their conservation status very difficult. New data on the biology and distributions of these six species would improve the understanding of their conservation needs.

The Four-toed Jerboa has very specific habitat requirements and a restricted distribution along the coast of Egypt and Libya. It is known from less than ten locations where availability and quality of its habitat has significantly decreased. Conservation of its remaining habitat is essential to its long-term survival. The Euphrates Jerboa is mainly threatened by agricultural expansion in Jordan, Syria, and Turkey. According to The IUCN Red List, it has declined in Jordan by about 50% over the past 20 years because of agricultural expansion, and it is also considered to be edible in some regions of Turkey and Jordan. Vinogradov’s Jerboa requires specific habitat types, and its area of occupancy may be quite small. It is mainly threatened by expansion of agriculture in Kazakhstan, Uzbekistan, Tajikistan, and Kyrgyzstan. The Greater Fat-tailed Jerboa is mostly threatened by habitat loss caused climate change, leading to the expansion of steppe habitat.



Ben Faleh, Granjon, Tatard, Boratyński et al. (2012), Boratyński et al. (2012), Charles & Viriot (2007), Fokin (1978), Gloger (1841), Holden & Musser (2005), Lebedev et al. (2013), Michaux et al. (2001), Moore et al. (2015), Ognev (1948), Pavlinov & Shenbrot (1983), Pisano et al. (2015), Potapova (1998), Shenbrot (1984, 1992), Shenbrot & Potapova (1982), Shenbrot et al. (2016), Vinogradov (1937), Young (1927), Zazhigin & Lopatin (2000, 2001), Zazhigin et al. (2002).