CHAPTER V.FOSSIL RODENTIA.

Families of Rodents represented by Fossil Remains—State of the “Record of the Rocks”—THESCIURIDÆ—Sciurine Genera now Extinct—No Fossil ANOMALURIDÆand HAPLODONTIDÆ—ISCHYROMYIDÆ—Pseudotomus hians—Gymnoptychus—CASTORIDÆ—Mr. Allen’s CASTOROIDIDÆ—THEMYOXIDÆ—No Fossil LOPHIOMYIDÆ—THEMURIDÆ—THESPALACIDÆ—THEGEOMYIDÆ—THEDIPODIDÆ—THETHERIDOMYIDÆ—THEOCTODONTIDÆ—THEHYSTRICIDÆ—THECHINCHILLIDÆ—THEDASYPROCTIDÆ—THECAVIIDÆ—THELEPORIDÆ—THELAGOMYIDÆ—Mesotheriumcristatum—Difficulties concerning it—Mr. Alston’s Suggestion—THEHEBETIDENTATA—Teeth—Skull—Skeleton—Conclusions regarding it—Table of Rodent Families—Concluding Remarks.

THEmajority of the preceding families are more or less clearly represented by fossil remains, either in the younger strata of the earth’s crust, or in those cave-deposits of comparatively recent date which have furnished so many interesting relics of the Mammals of former days. It must be remarked, however, that while a considerable number of fossil Rodents have been named and described by palæontologists, the materials upon which many of them have been established are very imperfect; in a great number of cases isolated molar or even incisor teeth furnish the sole evidence of the existence of creatures which were manifestly Rodents, but of which the other characters are rather difficult to divine from such scanty material. Still, imperfect as may be “the record of the rocks” in this as in other instances, it is in some parts sufficiently complete to enable us to trace back the existence of many forms of gnawing Mammals through a long period of geological time.

Of the SCIURIDÆa considerable number of fossil species have been recorded. Species of the generaSciurus,Arctomys, andSpermophilus, some of them identical with those still existing, have left their remains in Post-Tertiary deposits and in bone-caves in various parts of Europe; while species belonging to the first two genera, and to the American genusTamias, have been detected in similar situations in North America. A few forms referred to the same genera go down still lower in the series of geological formations. True Squirrels are recorded from Miocene and Upper Eocene deposits in France, and a single species from the Tertiaries (probably Miocene) of Colorado; Marmots from Pliocene and Miocene beds in the South of France, and from a Pliocene deposit in Nebraska; and a Spermophile from the Miocene of Weisenau in Germany.

Besides these examples of known types, several fossils have been obtained both in Europe and America, which are regarded as indicating genera distinct from any now living.Plesiarctomys Gervaisiiis founded on a fragment of jaw with four molars, obtained from Upper Eocene beds near Apt, Vaucluse. In its characters it appears to be intermediate between Squirrels and Marmots.Pseudosciurus suevicus, from the Upper Eocene (Bohnerz) of Würtemberg, seems to differ from all other Sciuridæ in the form of the molar teeth of the lower jaw, which are somewhat elongated, and have four tubercles arranged in two pairs, each pair being connected by a ridge. From the Tertiary deposits of the western territories of the United States, Professors Cope, Marsh, and Leidy have described several Sciurine Rodents as belonging to genera now extinct: thusParamyshas five species;Sciuravus(perhaps identical with the preceding), three;Heliscomys,Mysops,Colonymys,Taxymys, andTillomys, one or two species each.

Of the ANOMALURIDÆand HAPLODONTIDÆno fossil remains are known. On the other hand, a North American fossil Rodent, described by Dr. Leidy under the name ofIschyromys typus, is regarded by Mr. Alston as the type of a distinct family, theIschyromyidæ, nearly allied to the Sciuridæ, but also showing an affinity to the Beaver in some of its characters. The specimen described and figured by Dr. Leidy was obtained by Dr. Hayden from Miocene deposits in the “Bad Lands” of Wyoming. It was originally referred to the Sciuridæ, with which it agrees in its dentition, but is distinguished by its large infra-orbital opening, the presence of a sagittal crest, and the absence of post-orbital processes. The parietal region of the skull is much narrowed, and in this, as in the large size of the infra-orbital opening,Ischyromysresembles the Musk Rat.

Two other forms must be referred to here. Under the name ofPseudotomus hians, Professor Cope has described the remains of an animal which he believed to have been about the size of an Agouti, and originally thought to belong to the order Edentata. Subsequently he referred it to the Sciuridæ; but both Mr. Alston and Mr. Allen think that it may belong to the family Ischyromyidæ. In somerespects the skull resembles that ofArctomys, but it has the same contraction between the orbits asIschyromysandFiber. The incisor teeth are separated, and Professor Cope believes that the animal had only three molars on each side in each jaw. A still more doubtful member of the family is Professor Cope’s genusGymnoptychus, which includes four species, all said to be from the “Tertiary of the Plains.” In this genus there are five molars above and four below on each side, as inIschyromysand the Sciuridæ; but these teeth show two crescents on the inner side in the upper, on the outer side in the lower jaw, and each crescent gives origin to a cross-ridge running to the opposite margin of the tooth.

The CASTORIDÆ, including at present only a single species common to the northern parts of both hemispheres, are represented by several peculiar fossil forms. Remains of the Common Beaver (Castor fiber) are not uncommon in peat bogs and other late superficial deposits both in Europe and America; and, according to Sir R. Owen, in association with those of the Rhinoceros, Mammoth, and Mastodon, even in the Fluvio-marine Crag (Newer Pliocene) of Norfolk. In Belgium its bones have been found in caves. Among the Mammals from the Upper Tertiaries of the Sivalik Hills, Messrs. Falconer and Cautley record a Beaver distinct from the existing species, although nearly allied to it. The skull of a great Beaver, one-fifth larger than that of the living species, was obtained many years ago by M. Fischer from sandy deposits on the shores of the Sea of Azov, and, as it differed in some peculiarities of the teeth fromCastor fiber, was described by him as constituting a distinct genus under the name ofTrogontherium Cuvieri. It is now regarded as a true Beaver, and namedCastor Trogontherium. The British species, described and figured by Sir R. Owen from the Norfolk Forest bed under the name ofTrogontherium Cuvieri, is, however, quite distinct, and belongs to the genusDiobroticus, characterised by having the third upper and first lower molar teeth with four enamel folds, and the rest only with two, most of the folds soon becoming isolated as the teeth wear down. This animal must have been nearly twice the size of the European Beaver.

At a still earlier period—namely, in the Miocene—the family Castoridæ was represented, both in Europe and America, by some small species, nearly agreeing with the Beavers in dentition, but differing in the characters of certain bones of the skull. These form the genusSteneofiber. The largest (S. viciacensis), from the Miocene of the Allier, was about half the size of the Beaver; another (S. sansaniensis), from the fresh-water limestone of Sansan, was about as large as a Rat; an American species (S. nebrascensis), from the “Bad Lands” of Dakota, was rather smaller than a Marmot, and presented some resemblance to the Agoutis in the characters of the teeth; and a fourth species (S. pansus) occurs in the Santa Fé marls.Eucastor tortus, a species rather smaller than a Marmot, is described by Dr. Leidy as very nearly related to the Beaver. Its remains were found in loose sands of the Niobrara River, Wyoming.ChalicomysandPalæomysare genera doubtfully placed here. Their species occur in the Miocene and Pliocene of Europe.

Some bones of a gigantic Rodent, indicating an animal as large as a full-grown Black Bear, obtained from Quaternary and Alluvial deposits of various parts of the United States, have been described under the name ofCastoroides ohioensis, Mr. J. W. Foster, its first describer, having an idea that it was a great Beaver. It has generally been known as the “Fossil Beaver” of North America, but several authors have entertained doubts of the correctness of this designation, and Mr. Allen has lately made it the type of a special family, CASTOROIDIDÆ, which he regards as most nearly related to the Chinchillidæ. In the general aspect of the skull it resembles the Beaver, but in several details of structure approaches the Viscacha; while the structure of the molar teeth, which consist of a series of plates of dentine, completely enclosed by enamel, and held together by a thin coating of cement, occurs elsewhere only in the Chinchillidæ, and in the last molar of the Capybara.

Dormice as well as Squirrels disported themselves in the Tertiary woods and thickets of Europe, and remains of several species of MYOXIDÆoccur in various deposits in France, Switzerland, and elsewhere, from the Upper Eocene onwards.Myoxus glis, the Garden Dormouse, has been identified with some doubt from the caves of Lunel Viel; and this is also probably the species occurring in the Belgian bone-caves, and described asMyoxus priscusby Dr. Schmerling. A species a little larger than the Dormouse occurs in Russian caves, and has received the name ofMyoxus fossilisfrom M. Fischer; and the most striking species of all is also a Post-Pliocene form, namely, the gigantic Dormouse of Malta (M. melitensis). This animal, which seems to have been about the size of aGuinea-pig, must have been excessively abundant in Malta, for its describer, Professor Leith Adams, says that “its remains are met with in abundance throughout the cavern and fissure deposits, up even to the superficial alluvium now in course of formation.” From older times we have evidence of the existence of a Dormouse, about the size of the common species, at the time of the deposition of the gypsum of Montmartre (Upper Eocene), in which a well-preserved skeleton of the animal has been found. The same deposit has furnished traces of a second rather larger species. The Miocene of Switzerland and of Sansan has also yielded species ofMyoxus; and Professor Hermann von Meyer has recorded a Dormouse from the Miocene of Weisenau, under the name ofBrachymys ornatus.

No fossil LOPHIOMYIDÆhave yet been detected, but the great family MURIDÆhas left abundant evidence of its former existence. Species of the generaMus,Arvicola,Myodes, andCricetus, identical in many cases with those now living, have been obtained frequently in Post-Pliocene deposits and in bone-caves in Europe. Lemmings (Myodes lemmusandtorquatus) are recorded from English caves. The genusMusis also represented by several species in the Miocene deposits of France, and in the Sivalik beds investigated by Falconer and Cautley. The Miocene of Sansan has furnished a form which has been doubtfully regarded as a Gerbille, and namedMeriones Laurillardi. In the same and other deposits of the same age in South-eastern France several species of an extinct genus (Cricetodon) have been obtained. Their dentition resembles that of the Hamster, but the first molars in both jaws have a tubercle less; the largest species (C. sansaniensis) rather exceeded the Hamster in size, while the smallest was less than a Mouse. Associated with some of these are two doubtful forms,DecticusandElomys, the latter considered by M. Aymard, its describer, to be allied toHydromys. The American fossil Muridæ are for the most part either species of the genusHesperomys, or nearly related to it. Twelve species of that genus were obtained by Dr. Lund from the Brazilian bone-caves, but of these eight were identified by him with species still existing. In North America two species of a nearly-allied genus (Eumys) have been obtained from Miocene deposits; and the bone-caves of Pennsylvania furnish the remains of a species ofNeotoma(N. magister), hardly distinguishable from the Florida Rat.

ARhizomysfrom the Sivalik deposits of North-western India is the only recorded fossil representative of the SPALACIDÆ; and of the GEOMYIDÆthe only known species are aGeomysfrom the Pliocene of Nebraska, nearly allied to, if not identical with, the livingG. bursarius; and one from the “Tertiaries of the Plains,” described by Professor Cope asColotaxis cristatus, which, however, has only three molars in the lower jaw.

The DIPODIDÆare still more scantily represented. A Jerboa described by M. Fischer from Post-Pliocene deposits, probably of Tartary, is very nearly allied to the livingDipus platurus, but has shorter toes and broader cannon-bones. The genusDipoïdes, from the “Bohnerz” of Würtemberg, is founded on a single tooth, and its position in this family is very doubtful.

On the other hand, some fossil allies of the Dipodidæ and Geomyidæ constitute a distinct family, for which Mr. Alston proposed the name of THERIDOMYIDÆ, from that of one of its genera,Theridomys. In this genus, of which six species are recorded from the Eocene and Miocene deposits of France, there are four rooted molars in each series, and each of these has several enamel folds, some of which are converted into isolated loops as the crown is worn away. The best known species isTheridomys platiceps, from the Miocene of Caylus. InArchæomys chinchilloidesthere are still four molars, but these present a very different structure; they are rootless, and have the enamel folds extending diagonally across the crown, so that they are composed of a series of plates, thus presenting a certain amount of resemblance to the Chinchillas, which American familyArchæomyswas at one time supposed to represent in Europe. In fact, in the structure of their molar teeth, both the above genera approach American types; but in other characters, especially the form of the lower jaw, they appear to have been decidedly Mouse-like, and Mr. Alston regards them as most nearly related to the Dipodidæ, with which they are joined by a third form referred to the familyIssiodoromys, a genus sometimes placed with the Jerboas. The teeth in this genus are of the same number as in the preceding, but the molars are much simpler, each of them exhibiting one large re-entering fold of enamel, which causes the surface of the tooth to present two heart-shaped lobes. This structure is not dissimilar to that prevailing in some Dipodidæ, and especially inPedetes, but it was formerly thought to indicate a relationship to the Cavies, and accordingly the best-known species has received the name ofIssiodoromys pseudanæma(Anæmabeing a sub-genus of Cavies).This species occurs abundantly in the Miocene lacustrine limestone near Issoire. A second species (I. minor) has been detected in the Upper Eocene of Lamandine-haute.

Of the OCTODONTIDÆ, an essentially American family at the present day, nearly all the recorded fossil forms are also American. Species ofEchinomys,Loncheres, andPhyllomyswere obtained by Dr. Lund from the Brazilian bone-caves, which also furnished him with the remains of a Coypu (Myopotamus antiquus), and of an allied form,Carterodon sulcidens, distinguished by its having broad incisors with longitudinal furrows and raised ridges. The latter has since been found living in South America. Another species, allied toEchinomys, is named by LundLonchophorus fossilis. The superficial deposits of South America have yielded the remains of two species ofCtenomys, one of which is believed to be identical with a recent species. As several species of this family now live in Africa, the occurrence in the eastern hemisphere of fossil forms belonging to it would not be surprising, but the few that have been referred to it are of very doubtful nature. M. Lartet obtained some isolated teeth from the Miocene of Sansan, which he described under the name ofMyopotamus sansaniensis; and one or two other types (Aulacodon,Adelomys), from Upper Eocene and Miocene beds, are of very uncertain position.

Of the HYSTRICIDÆ, or Porcupines, remains have been obtained in both hemispheres. In the Old World traces of true Porcupines (Hystrix) are recorded from the Valley of the Arno, from the Sivaliks, the Pliocene deposits of the Auvergne, from Pikermi, and, on very doubtful evidence, from the Upper Eocene of Lamandine-basse; whilst Dr. Leidy has described two teeth from the Pliocene deposits of Dakota, as belonging to a species (Hystrix venustus) allied to the European Porcupine. This determination, if confirmed, would be of great interest, as no true Porcupine now occurs in America. Of the American type, two species ofSphingurushave been obtained from the Brazilian bone-caves; and Professor Cope records a species of the North American genusErythizonfrom a similar cave in Pennsylvania.

The CHINCHILLIDÆhave left but scanty traces of their former existence.Lagostomus brasiliensisis from the Brazilian bone-caves; andMegamys patagoniensisfrom the Eocene sandstone of Patagonia. The latter species is founded upon a tibia and rotula, which on comparison seemed to approach most nearly to those of the Rodents of this family, and if the determination be correct it was probably one of the largest species of the order, as the tibia measures about a foot long.AmblyrhizaandLoxomylus, are two genera described by Professor Cope from bone-caves in Anguilla Island, West Indies.

The DASYPROCTIDÆhave but few fossil representatives, and the undoubted ones are all from the bone-caves of Brazil, which furnished Dr. Lund with two Agoutis and two Pacas. Of the former, one is described asDasyprocta capreolus; the second is allied to the livingD. caudata. The two species ofCœlogenysare extinct. Some teeth, found in Tertiary deposits of the Puy-de-Dôme, have been referred toDasyprocta, but this determination is excessively doubtful.Diobroticus schmerlingifrom Belgian caves has been placed with the Castoridæ.

Of the CAVIIDÆ, Dr. Lund obtained three species of the genusCavia, and two ofHydrochœrus, from Brazilian bone-caves. Of the latter, one was allied to the existing Capybara; the other was a gigantic species, measuring about five feet in length. Dr. Leidy has described a species (Hydrochœrus æsopi) from teeth found in Post-Pliocene deposits in South Carolina; and the Pampean deposits of the same age furnished M. D’Orbigny with the remains of a Cavy (Cavia antiqua) which, however, is doubtfully distinct from the Patagonian species.

The remains of species of the family LEPORIDÆare very abundant in some Post-Pliocene cave deposits on both sides of the Atlantic, and in several cases the species are evidently identical with those now living. Besides these, species of the genusLepushave been found in Pliocene and Miocene beds in France. In North America three extinct Leporine genera have been recognised, differing fromLepusin certain peculiarities of the molar teeth:—Palæolagus, with three species, from the Miocene of Dakota and Colorado;Panolax, from the Pliocene marls of Santa Fé; andPraotherium, from a bone-cave in Pennsylvania. The last-named genus has the crowns of the molars transversely oval, and without the enamel-band or crest which is seen on the surface of the teeth of other Hares.

The LAGOMYIDÆare known in a fossil state chiefly from Post-Pliocene deposits, and the bone breccias of caves in various parts of Europe. In Post-Pliocene times the genusLagomysseems to have been very generally distributed over the South of Europe; and the earliest appearance of the genusis in the Pliocene, three species having been described from deposits of that age at Œningen and Montpellier. The family is, however, carried further back in time by the genusTitanomys, in which the molars differ but slightly in structure from those ofLagomys, but there are only four of them in each series, both above and below. Two species of this genus have been recorded from Miocene deposits in France and Germany.

SIDE VIEW OF SKULL AND LOWER JAW OF MESOTHERIUM CRISTATUM.

DENTITION OF MESOTHERIUM CRISTATUM.(A) Upper Jaw; (B) Lower Jaw; (C) Incisors.

We have thus passed very briefly in review the fossil Rodents belonging to the two great sections of the order to which all its living species are to be referred; and it will be seen that while a knowledge of their existence is necessary to complete the history of the order, they present none of those peculiar characters which lend such interest to the fossil members of many other orders. There is, however, one fossil South American type to which we have yet to refer, as, by the curious combination of characters which it presents, it has long been somewhat of a puzzle to palæontologists, and although generally placed among the Rodents, its peculiarities are such that Mr. Alston found himself compelled to establish a third primary section of the order for its reception. According to M. Bravard, the first discoverer of this peculiar type, the Pliocene deposits of the Pampas of La Plata contain the remains of three species belonging to it; but the bones which have been sent to Europe, and which represent most parts of the skeleton, seem all to belong to a single species, which has been very fully described by M. Serres under the name ofMesotherium cristatum.[55]What distinguishes it at once from all other known Rodents is the presencein the lower jawof four incisor teeth, the second pair being very small and placed immediately behind the outer edge of the broad middle pair. The latter are peculiarly widened and compressed from front to back in both jaws, and their summits, instead of being worn to a sharp chisel-like edge as in ordinary Rodents, show an elongated ring of enamel surrounding a slightly depressed surface. Hence Mr. Alston denominated this section HEBETIDENTATA, or BLUNT-TOOTHEDRODENTS. The enamel in all the incisors is continuous round the tooth. The molar teeth are rootless and curved, the convex side being directed outwards, contrary to what occurs in other Rodents. They are surrounded by enamel, and show re-entering folds which differ in the two jaws. Their number on each side is five in the upper and four in the lower jaw. The skull is massive, with enormously-developed sagittal and occipital crests, the latter of which run forward so far as to join the zygomatic arches; and these crests rise so high that the upper surface of the actual brain-case is entirely concealed by themwhen the skull is looked at from the side. The lower jaw in its characters presents some resemblance to the same part in the Leporidæ; but it has the condyle for its articulation with the skull transverse, and fitting into a cavity of corresponding direction, a character which occurs in no other Rodent. Of the remainder of the skeleton we need only state that the animal possessed perfect clavicles; that the shoulder-blade and humerus somewhat resemble those of the Beaver; that the fibula articulated with the heel-bone; and that both front and hind limbs possessed five toes, some of which, judging from the form of the terminal joint, were probably furnished with hoof-like claws.

Thus, as regards its affinities in the order Rodentia,Mesotheriumpresents resemblances in its lower jaw (as also in some peculiarities of the skull), and in the articulation of the heel with the shank, to the Hares; while in the shortness of the incisors and some other cranial peculiarities, the form of the shoulder-blades, and the probably hoof-like character of the claws, we may notice an approach to the Cavies, which are also South American forms, and especially to the Capybara, which it probably resembled in its habits, although, if the evidence of the Beaver-like shoulder-blade and humerus be taken into account, it would appear to have been still more aquatic.

On the other hand, the resemblance to certain other Mammalia, and especially to some aberrant Ungulates, is unmistakable. The number of incisor teeth is the same as inHyrax, and in these teeth there is also a certain amount of resemblance to the curious genusToxodon, in which the incisors are four in the upper and six in the lower jaw, and worn away in somewhat the same fashion. InToxodonalso, the convexity of the curve of the molars is turned outwards. Certain other characters ofMesotherium—such as the mode of articulation of the lower jaw, and the peculiar connection of some of the caudal vertebræ with the ischiatic bones—present resemblances to the Edentata. As Mr. Alston says, “It appears to have been a survivor, to Pliocene times, of a much earlier type, which represented an era at which the Rodents were not yet clearly marked off from their allies. In fact,Mesotheriumseems to continueintothe order Glires that line of affinity which Professor Flower has pointed out as extending from the typical Ungulates throughHyracodon,Homalodontotherium,Nesodon, andToxodon.”

The general relationships ofMesotheriumto the other Rodents, and of these among themselves, are represented by Mr. Alston in a diagrammatic form, from which the following scheme, which will serve also as a table of the families, is derived:—

Fossil Rodents⇒LARGER IMAGE

⇒LARGER IMAGE

It seems quite clear, even from the above brief sketch of the history of the Rodentia in time, that, except in the case ofMesotherium, the fossil remains of animals belonging to this order furnish us with no important information bearing upon their alliances and possible origin. They make their earliest appearances, so far as we know, in deposits of Eocene age; and the earliest forms the remains of which are sufficient to give us a clear insight into their nature, are manifestly members of families, and often nearly allied to species still extant in the regions where their traces are now found. Thus in Eocene and Miocene deposits, we have representatives of the familiesSciuridæ,Castoridæ,Myoxidæ,Muridæ,Geomyidæ,Chinchillidæ, andLeporidæ, already differentiated as at the present day, so far as the evidence goes; and it is clear that we must go much further back in time to seek the earliest appearance of the Rodent type, whether it branched off directly from the Marsupial series, or passed, as would seem to be indicated byMesotherium, through a sequence of forms more or less related to the Ungulates.

Nor does the geographical distribution of the animals lead to any more definite conclusions. Certain families and even sub-families are of very wide range, the Muridæ and Sciuridæ especially being represented nearly all over the world, while the Hystricidæ and Leporidæ are also spread over very large areas, occurring in both hemispheres. Certain groups, such as the Sciuridæ of both sub-families, the Castoridæ, the Murine, and Arvicoline sub-families of Muridæ, the Leporidæ, and the Lagomyidæ, may be said to have a circumpolar distribution in the northern hemisphere, nearly allied and sometimes identical species being found in the more northern parts of both the Old and the New World, but mixed with other forms peculiar to the regions, especially as we advance southwards. On the other hand, the Myoxidæ are peculiar to the eastern hemisphere, as are also the Spalacidæ and the Dipodidæ (with the exception ofZapus, which is considered by Dr. Coues to form a distinct family), and all the sub-families of Muridæ, except those above mentioned as having a circumpolar range. The Myomorpha may in fact be looked upon as an Old World group, the Geomyidæ being the only exclusively American family; while the Hystricomorpha as a whole may be regarded as American, certain aberrant forms of the Octodontidæ inhabiting various parts of Africa and the Old World Porcupines being the sole representatives of that great section outside the western continent. Considering these facts, we may regard the Sciuromorpha and the Duplicidentata as originally polar types, or at all events as having an equal claim to an origin in the northern regions of either continent; while the Myomorpha, with their multitudinous forms spreading over all parts of the Old World, and having a much scantier representation in America, probably originated in the eastern hemisphere, and spread by a northern passage into the New World; and the Hystricomorpha would seem to have originated in South America, where they display the greatest variety of forms.

W. S. DALLAS.

Back to Index Next