Journal of the proceedings of the Linnean Society, Zoology 04

Onthe Characters, Principles of Divisionand Primary Groups ofthe Class Mammalia 1 Sclater, Philip Lutley, M.A., F.L.S.. In mostothermammalsparticularteethhave special forms forspecial us

LIST OF PAPERS.

Page

Cobbold,T Spencer, M.D., F.L.S

Descriptionof a New Form ofNaked-EyedMedusa

(Thauman-tiasachroa), withbriefhistologicaldetails 38

Couch, Jonathan, F.L.S &c

Note onthe OccurrenceofPkyllosoma communeontheCoast of

Note onasingular caseofColouringof the HumanHair 41

Owen,Professor, F.R.S.,V.P.L.S &c

Onthe Characters, Principles of Divisionand Primary Groups ofthe Class Mammalia 1

Sclater, Philip Lutley, M.A., F.L.S &c

Onthe general GeographicalDistributionoftheClassAves 130

OntheZoology ofNewGuinea 149

Smith,Frederick, Assistant in theZoologicalDepartment intheBritishMuseum.

Catalogue of theHymenopterous Insects collected at Sarawak,Borneo ; Mount Ophir, Malacca; and at Singapore, by

JOURNAL OF THE PROCEEDINGS

LINNEAN SOCIETY OF LONDON.

On the Characters, Principles of Division, and Primary Groups

of theClassMammalia ByProfessorOwen,F.B.S., P.L.S.,Superintendent ofthe Natural History Departments in theBritish Museum.

[Bead February 17thandApril21st,1857.]

The class Mammalia, the most highly organized of the animal

been thelast class of animals introducedon thisplanet, and not

tohave attained plenary development until the tertiary division

ofgeological time

covering ofhair,and (withtwoexceptions) byteats ormammae

—

whence thename oftheclass* AllMammalspossess mammary

glands, andsuckletheiryoung: theembryoor foetusisdeveloped

lungs, composedof a highly vascularand minutelycellular ture throughout, and suspended freely in a thoracic cavity sepa-rated by a muscularandtendinous septumor diaphragm fromthe

struc-abdomen.

* From mamma, a pap ThePlatypus and Echidnaare the only known

exceptionsto this rule. The Mareisanapparent one,fromthepudendal

posi-tion of thenipples. ThefoetalCetaceashowtuftsof haironthemuzzle

2 PROFESSOR OWEN ON THE CHARACTERS, ETC.

and twoauricles,andhavewarmblood: they breathe quickly; butinspiration is performedchieflyby the agencyof the diaphragm and theinspiredairacts onlyonthe capillariesof thepulmonarycircidation

The blood-discs are smaller than in Reptiles, and, save in theCamel-tribe, arecircular Theright auriculo-ventricular valve is

membranous, at least neverentirely fleshy; and the aorta bends

over the left,never over theright, bronchial tube The primarybranches of theaorta are given offnotimmediatelyafter, butata

little distance from, its origin,and there is less constancyin theorder oftheirorigin thaninBirds: the phrenicarteries,thecoeliacaxis, and the superior mesenteric artery are always branches ofthe abdominal aorta, which terminates by dividing beyond thekidneys intotheiliac arteries,from whichspringboth the femoral

and ischiadic branches: the caudal orsacro-median artery,which

in some long-tailed Mammals assumes the character of the tinued trunkoftheaorta,neverdistributes arteries tothe kidneys

con-orthe legs, as inBirds The kidneys are nourished,andderivethe material oftheir secretion,exclusivelyfromthearterialsystem.Theirveins are simple,commencing by minute capillaries in the

parenchyma and terminating generally byasingle trunk on eachside intheabdominal venacava: theyneveranastomose with themesentericveins

Thekidneys are relativelysmallerand present amore compactfigure than in the other vertebrate classes; theirparenchyma is

divided into a cortical and medullary portion, and the secretingtubuli terminate in a dilatation of the excretory duct, called the

pelvis

The liver is generally divided into a greaternumberoflobesthan in Birds The portal system is formed by veins derivedexclusivelyfrom the spleen andchylopoieticviscera The cysticduct,whenitexists, alwaysjoins thehepatic, anddoes not enter

condyletothe base ofthe zygomaticprocess,andnotto the

tym-panic element of the temporal bone; the base ofthe coronoid

and the alveolar processes The jawsof Mammalswithfew ceptions are provided with teeth, which are arranged inasinglerow; they are always lodged in sockets, and never anchylosedwith the snhstance of the jaw The tongue is fleshy,well-deve-loped, with the apexmore orless free The posteriornares areprotected bya soft palate, and the larynx by an epiglottis: therings of the trachea are generally cartilaginous and incompletebehind: thereisnoinferior larynx Theoesophagusis continuedwithout partial dilatations to the stomach, which varies in its

ex-structure according tothe nature of the food, orthe quantity of

nutrimenttobe extracted therefrom

The truevertebrae of Mammaliahavetheirbodiesossified from

three centres, and present for alongeror shorterperiod of life adiscoid epiphysis at each extremity They are articulated by

concentric ligaments with interposed glairy fluid forming whatare calledthe intervertebralsubstances; thearticulating surfacesare generallyflattened, but sometimes,as in the neck of certainRuminants,theyare concave behind and convex in front: such

a vertebra, however, may be distinguished from a vertebra of

aReptile, with asimilar ball-and-socket structure of thearticularsurfaces,evenwhen foundinafossilstate,and whenthetestofthearticulating mediumcannot beapplied,bythecomplete anchylosis

or confluence of the annular with the central part orbody, and

bythe large relative size of thecanal forthe spinal chord The

cervical vertebrae, with one or two exceptions, are seven innumber, neither more nor less: the Monotremes, which are theinstances commonly opposed to other generalizations, form no

exception to this rule The lumbarvertebrae are moreconstant

and usually more numerous than in other classes of vertebrateanimals The atlas is articulated byconcave articular processes

to two convexcondyles,whichare developedfromtheex-occipitalelements of the last cranial vertebra The tympanic element ofthe temporal bone is restricted in functiontothe service of the

organ of hearing, and never enters into the articulation of thelower jaw The olfactory nerves escape from thecranial cavitythrough numerous foramina of a cribriform plate The opticforaminaare always distinctfrom oneanother

The scapula is generally an expanded plateofbone; the

cora-coid, with two (monotrematous) exceptions, appears as a smallprocess of the scapula The sternum consists of a narrow and

usually simple seriesofbones: thesternal portions of theribsaregenerallycartilaginousandfixed tothevertebral portions without

4 PROFESSOR OWEN ON THE CHARACTERS, ETC.

theinterposition ofadistinctarticulation: therearenogristlyor

bony abdominalribs orabdominalsternum Thepubicandischial

arches are generally complete,and united together by bonyfluence on the sternal aspect, so that the interspace ofthe twopelvic arches is converted into two holes,calledforamina obtura-toria or thyroidea The sclerotic coat of the eye is a fibrousmembrane, and never contains bony plates In the quantity of

con-aqueous humour and the convexity of the lens Mammals aregenerally intermediate between Birds and Fishes The organofhearing is characterized by the full developmentof the cochleawith a lamina spiralis: there are three distinct ossicles in the

exter-nally; the meatus auditorius externus often commences with

a complicated external ear,having a distinct cartilaginous basis

The external apertures of the organ of smell are provided with

moveable cartilages and muscles,and the extent of the internalorgan is increased byaccessorycavities or sinuseswhich commu-

nicate with the passages including the turbinated bones

There arefew charactersofthe osseous system common, andatthe same time peculiar, to the class Mammalia The following

The cancellous texture of mammalian bone is of a finer and

more delicate structure than in Reptiles,and forms a closer

net-workthanin Birds Themicroscopicradiating cellsare relativelysmallerandapproachmorenearlytothe spheroidform; but boththese histological characters areliable to mislead,if unsupported

by more obviousand constant ones, in the interpretation of a

fossil.

Dental characters.— The Mammalia, like Reptilia and Pisces,include a few genera and species that are devoid of teeth; thetrue ant-eaters {Myrmecophaga),thescaly ant-eatersorpangolins(Ifanis), and the spiny monotrematous ant-eater (Echidna), are

examplesof strictlyedentulousMammals The Ornithorhynchus

has hornyteeth, and thewhales (Balcena andHalienopterd) havetransitoryembryonic calcifiedteeth, succeededby whalebonesub-stitutes inthe upperjaw ThefemaleNarwhal seems tobeeden-tulous, but has the germs of two tusks in the substance of the

upperjaw-bones; oneofthese becomesdevelopedintoalarge and

conspicuousweaponin themale Narwhal, whence thename ofits

genus Monodon.

The examplesofexcessivenumberofteeth are presented,intheorder Bruta, by the priodont Armadillo, which has ninety-eightteeth: and in the Cetaceous order by the Cachalot,which has

upwards of sixty teeth, though mostof them areconfined tothelower jaw; by the common Porpoise, which has between eighty

andninetyteeth: bytheGangeticDolphin,whichhasone hundred and twenty teeth; and bythe true Dolphins (Delphinus), which

havefrom one hundredtoone hundred andninetyteeth,yielding

"When the teeth are in excessivenumber, as intheArmadillos

and Dolphinsabovecited,they are small, equal,orsub-equal, and

usually of a simple conicalform

In mostothermammalsparticularteethhave special forms forspecial uses; thus,thefront teeth, frombeingcommonly adapted

toeffect the first coarse division of the food, have been calledcuttersorincisors; and the back teeth, whichcomplete its com-

minution, grinders ormolars; large conical pointedteeth situatedbehind the incisors, and adapted, by being nearer the insertion

ofthe bitiugmuscles,to actwith greaterforce, are called holders,tearers, laniaries, or more commonly canines, from being welldevelopedinthe Dog and other Carnivora

It ispeculiar tothe classMammaliatohave teeth implantedinsockets by twoormorefangs; but this canonlyhappen to teeth

of limitedgrowth,andgenerally characterizes the molarsandmolars: perpetually growing teeth require the base to be keptsimple andwidely excavated forthepersistent pulp In no mam-

pre-miferous animal does anchylosisof the tooth with the jawtute anormal mode of attachment Each tooth has its peculiarsocket, to which it firmly adheres by the close co-adaptation of

consti-periosteumtothe organizedcement whichinveststhe fangorfangs

of thetooth

Trueteeth implanted insockets are confined, intheMammalian

class,tothemaxillary, premaxillary,and mandibularorlower

max-illary bones, and form a single row in each They mayprojectonlyfromthe premaxillary bones,as intheNarwhal; oronlyfrom

the lower maxillary bone,asinZiphius; orbelimited tothe

supe-riorandinferior mamillariesandnot presentinthepremaxillaries,

as in the true Euminantia and most Bruta (Sloths, Armadillos,Orycteropes) In most Mammals teeth are situated in all the

bones above mentioned

The teeth ofthe Mammalia usuallyconsistof hard unvasculardentine, defended at the crown by an investment of enamel, and everywhere surrounded bya coat ofcement

Thecoronalcementisofextremetenuityin Man, Quadrumana

andtheterrestrialCarnivora; it is thicker inthe Herbivora,

espe-ciallyin thecomplexgrinders oftheElephant

Vertical foldsofenamel and cement penetrate the crownof thetooth in the ruminating and manyother Ungulata, and in most

Rodents,characterizingby their variousforms the genera of thoseorders

the tooth-matrix does not develope the germ ofa second tooth,destinedtosucceed oneintowhich the matrix has beenconverted;such a tooth, therefore,when completed and worn down, is notreplaced The Sperm Whales,Dolphins,andPorpoises are limited

tothis simple provision of teeth In the Armadillos and Sloths,thewantofgenerative power,as itmay be called,inthematrix is

compensated bythepersistenceofthe matrix, and bytheruptedgrowtli ofthe teeth

uninter-In mostotherMammalia,the matrix of thefirst-developedtoothgives origin tothe germ of a second tooth,which sometimes dis-

placesthefirst, sometimestakes itsplaceby the sideofthe tooth

from which ithas originated

All those teethwhich r.redisplacedby theirprogenyare called

'temporary,' deciduous, ormilk-teeth; themode anddirection inwhich theyare displacedandsucceeded,viz.fromabovedownwards

in the upper, from below upwards in thelower,jaw, inboth jawsvertically—are the same as in the Crocodile; but the process is

never repeatedmorethan once inany mammaliananimal Asiderable proportion of the dental series is thus changed; the

con-THE 7

to the jaws of the adult, as the 'temporary' teeth were adapted

tothose oftheyounganimal

Those permanentteeth, which assumeplacesnot previously cupied by deciduousones,are always the most posterior in theirposition,andgenerallythemost complexin their form The term

oc-'molar' or'truemolar' is restricted to these teeth The teeth

between them and the canines are called'premolars;' theypush

out the milk-teeth that precede them, andareusuallyof smaller

sizeandsimpler form thanthe true molars

Thus the classMammalia, inregard tothetimes of formation

and the succession of the teeth,may be divided into two groups,monophyodonts*,orthosethatgenerate asingle set ofteeth; and

thedipliyodonts\, orthose that generate two setsof teeth But

this dentalcharacter is not so associatedwith other organic racters as to indicatenatural orequivalentsubclasses

acquire fixed individual characters, receive specialdenominations,

andcan be determined fromspecies to species This zation oftheteethis eminentlysignificative of the high grade oforganization ofthe animals manifestingit.

individuali-Originally, indeed, the names 'incisors,' 'canines,' and '

mo-lars,' were given to the teeth, in Man and certain Mammals,as

inReptilesandFishes, inreference merelytotheshapeandoffices

indicated by these names; but they are now used as arbitrary

signs,in amorefixedanddeterminatesense In someCarnivora,

e. g the front-teeth havebroad tuberculate summits,adapted for

nipping and bruising, while the principal back-teeth are shapedforcutting,and work uponeach other like the blades of scissors

Thefront-teeth intheElephantprojectfromtheupperjaw,in theform, size and direction oflong pointed horns In short, shape

and size are the least constant of dental characters in the

Mam-malia; andthe homologousteeth are determined,likeotherparts,

by their relative position, by their connexions, and by theirdevelopment

Those teethwhichareimplantedin thepremaxillarybones,and

in the corresponding part of the lowerjaw, are called 'incisors,'

whateverbe theirshape orsize. Thetoothinthe maxillary bone,

which is situated at or near to the suturewith thepremaxillary,

is the 'canine,' as is also that tooth in the lower jaw, which, inopposing it, passes in front of the upper one's crown when the

* juoj'os,once; <pv<i), Igenerate; ddoiis,tooth

8 PBOFESSOE OWEN ON THE CHAEACTEES,ETC.

mouthis closed The other teeth of thefirst set are the'

deci-duous molars; the teethwhich displace and succeedthem

verti-callyarethe 'premolars; ' themoreposterior teeth,whichare notdisplacedhyvertical successors, arethe'molars'properlysocalled

I havebeenled, chiefly b}rthe stateof thedentition inmost oftheearlyforms of both carnivorous and herbivorous Mammalia,

which flourished during the eocene tertiary periods, to regard

3 incisors, 1 canine,and 7 succeedingteeth, on each side of bothjaws, as thetypeformulaof diphyodont dentition

Threeof theseventeethmay be'premolars,'andfourmaybe true'molars;' orthere may be four premolars, and three true molars.This difference,as Ihave elsewhere shown,forms a character of asecondarygroupor orderinthemammalianclass* The essentialnature of thedistinctionisasfollows: truemolarsareabackward

continuation of thefirst seriesofteeth; they are developedinthe

same primary groove of the foetal gum; they are 'permanent

becausetheyarenot pushed out bysuccessionalteeth—the'molars,' called'dents de remplacement' by Cuvier Seven teethdevelopedin the primarygrooveis, therefore,thetypicalnumber

pre-of first teeth, beyond the canines If, as in Didelpliys, theanteriorthree develope tooth-germs,which come to perfection in

a 'secondarygroove,' there are then 3 deciduous teeth, 3 molars,and 4 true molars: if, as in Gymnura, the anterior four

pre-of the 'primary' teeth develope tooth-germs, which grow in asecondary groove, there are then4 deciduousteeth,4premolars,

and3 true molars Thefirsttrue molar of the marsupialisthusseen tobe the homologueofthelastmilk-molar of the placental

exist-ing quadrupeds which retain the typical number and kinds ofteeth In a young Hog of ten months, the first premolar, p.1,

and the first molar,m.1, are in place and use together with thethree deciduous molars,d 2, d 3, and d. 4; the second molar,

on.2,hasjust begun to cutthegum; p.2,p.3, and^>.4,togetherwith m.3,aremore orlessincomplete,andwillbe foundconcealed

in their closedalveolif.

The lastdeciduous molar, d.4,has thesamerelativesuperiority

ofsizetod.3 andd 2, whichm 3 bearstom 2andm.1; andthe

* Outlines of aClassificationoftheMammalia,Trans.Zool Soc vol ii p.330

Or THE CLASS MAMMALIA. 9

crowns ofp.3 andp.4areofamoresimpleformthan those of themilk-teeth,which theyaredestined to succeed Whenthe milk-teeth are shed,andthepermanentonesareallin place, theirkindsare indicated, inthegenus Sns,hythe followingformula:

which signifiesthatthereareoneachsideof both upper andlower

jaws3 incisors, 1canine, 4premolars, and3 molars,makingin all

44teeth,each toothbeing distinguishedbyitsappropriate symbol,

e g.,p 1 to p.4,m 1tom.3 Thisnumberof teethisneverpassedinthe placentalDiphyodont series

sur-When the premolars and the molars are below this typicalnumber, the absent teeth are missing from the fore part of thepremolarseries,and fromthe backpart ofthe molarseries The

most constant teeth are the fourth premolar and the first true

molar; andthesebeing known bytheir orderand mode of lopment, the homologies of the remaining molars and premolarsare determinedby countingthe molarsfrombefore backwards, e.g.'one,' 'two,' 'three,' and the premolarsfrom behindforicards,

deve-'four,' 'three,' 'two,' 'one.' The incisors are counted fromthe

median Hue, commonly the foremost part, of both upper and

lower jaws, outwards and backwards The first incisor of theright sideisthehomotype,transversely, of the contiguousincisor

of the left side in the same jaw, and vertically, of its opposingtooth in the oppositejaw; andso with regardto thecanines, pre-molars,andmolars; just as the right arm isthehomotypeofthe

left arminitsownsegment, andalsooftheright legof a ingsegment It suffices, therefore, toreckon and namethe teeth

succeed-ofoneside of eitherjawin aspecieswith thetypical number and

kinds ofteeth, e.g thefirst,second, andthirdincisors,—thefirst,

second, third, and fourth premolars,—the first, second,and thirdmolars; and ofonesideofbothjaws inanycase

I havebeenindixced to dwellthuslongonthe dental characters

of the class Mammalia, because they have not been clearly oraccurately defined in any systematic or elementarywork onzoo-logy,althoughan accurate formula andnotation of the teeth are

ofmoreuseandvaluein characterizinggenerain this thanin any

otherclassofanimals

Inextproceedto reviewbrieflytheprincipal primarydivisions

Natural History have adopted different characters, drawn fromdifferentsystems oforgans, fortheprimarygroupsor divisions of

Aristotlechose the locomotive system,anddividedhisZootoka

—the equivalent of the Linnean Mammalia —into three

sec-tions:—1st,Dipoda,orbipeds; 2nd,Tetrapoda,orquadrupeds;

which includesallthe classsavetheHuman-kind andthe

Whale-tribe, is subdivided into thosewith claws, and those with hoofs

The unguiculatequadrupeds areagain subdivided accordingtothenature of theirteeth; the ungidate quadrupeds, accordingto thedivisions oftheir hoofs, as e.g.into Polyschidce, ormultungulates,

Discliiclce, or bisulcates, and AscJiidce, or solidungulates Ineedscarcely remark that this, in most respects admirable, system,

wouldhavecommandedgreaterattention,and been nowrecognized

asmoremanifestlythebasisof latersystems,haditsimmortalthormoretechnicallyexpressedhis appreciationofthelaw ofthesubordinationof characters; buthe applies toeach ofhisgroups,

au-whatevertheirvalue,thesame denomination,viz.genos, orgenus

Ray,with a less philosophical appreciation of the extent and

nature oftheclass ZootokaorMammalia, arrangeshisequivalentgroup of"Viviparous Four-footed Animals" chiefly on the Ari-stotelian characters; the primarydivision being into Ungulate

and Unghicttlate, and the subdivisions being based on

locomo-tive anddental characters

Linnaeus, restoring the classMammaliato its Aristotelian

inte-grity,primarily subdividesit intoTJngtjiculata, TTngtjlata, and

Mtjtica, the latterbeing the 'Apoda'of Aristotle: the secondarygroupsorordersare foundedchieflyonmodificationsofthe dentalsystem

Cuvier, adopting the same threefold primary division ofthe

class,subdivides itintobetter and more naturally defined orders,according to various characters derived from the dental, theosseous, generative, andthelocomotive systems

Illiger,inprimarilydividingtheMammaliaintothosewithfree,

andthose with fettered limbs—the 'pedes exserti distincti,' trasted with the'pedesretractiobvoluti,'— made a more unequal

con-andless naturalpartition than thethreefoldoneofAristotle; theSeals and the Whales balance all the rest of the class in theIlligeriansystem Thesubdivisions,also,of these primary groups,based exclusivelyon characters oflocomotion,havemetwithlittle

acceptancebeyond some ofthe schools ofGermany.

De Blainville appears first, 1816, to have adopted a character

from the reproductive system for the primary division of the

MAMMALIA. 11

delphes.' His ordersare in themainareturnto theLinneau syT

stem and nomenclature, with some peculiar views, as e.g.of the

quadrumanous orprimatialaffinityof theSloths,whichhave nevergained acceptance But his system indicates a clearer apprecia-tion orstrongerconviction of the valueofthe character of parityand imparityin the number of toes of the Ungulata, first sug-gested by Cuvier*, than was subsequently entertained by theoriginator oftheidea

Thepositionofthe marsupialand monotrematous quadrupedsatthebottom of theclassMammalia, and the higher value assigned

tothe group whichtheyconstituted,thanthat inthe'Eegne

Ani-mal' ofCuvier, were ideas also in closerconformity with nature

They were, however, but surmises, unstistained by anatomical

knowledge; and, as such, failed to carry conviction, or gain ceptance Nor was ituntilcomparativeanatomy had shown thatthe Marsupials and Monotremesagreedin differingfrom allother

ac-mammals intheabsenceof a placenta,andofthe greatcommissure

ofthe brain, in certain bird-like characters ofthe heartf,and from

allotherdiphyodont Mammals in alessnumberof premolars,and

a greater number of true molars,— depending essentiallyon theretention of a milk-tooth (m.4), which is displacedand changed

inthe placental diphyodonts,—thatthe trueaffinitiesofthephidandornithodelphid mammals to each other, and their trueposition inthe classMammalia, were finallyrecognized

didel-Inthe 'SystemaVertebratorum,' communicated in 1840to the

Linnean Societybythataccomplished and indefatigable zoologistPrince Charles LucienBonaparte, theprimarysubdivision of the

is adopted; and the first division or series Blacentalia is divided,agreeably withM.Jourdan's distribution ofMammaliain

Ineducahilia, the latter including the orders Bruta, Cheiroptera,Insectivora and Bodentia, with the common character of 'cere-

brum unilobum.' This Iregard as the most important

improve-mentinthe classification of theMammalia, which has been posed since the establishment of the natural character of theimplacentalorovo-viviparousdivision

pro-Cuvier had earlynoticed the relation of the Australian

mam-mals, as asmall collateral series,to the unguiculate mammals of

* OssemensFossiles, 4to ed. 1812,p.9; torn iii ed.1822,p 72.

t OntheClassification of theMarsupialia,Zoological Transactions, vol.ii.

the rest oftheworld, "some," hewrites,"correspondingwith theCarnaria, some with the Hodentia, and others again with the

It does not appear, however, that Cuvier meant to do more

thanindicate certain relations of analogy; just as therelationofthe pedimanous and frugivorous Marsupials to the pedimanous

(Tliyla-cimis) totheWolf, of the Flying Petaurist totheFlyingSquirrel,

of the Koala to the phytiphagous Sunbear, of the Opossums tothe Shrews, and of the Echidna to the Anteater, &c, had been

pointed out by myself My esteemed friend and colleague Mr Waterhouse,whilst admitting thejustness of some of thesecom-parisons, appended a timely warning, in a valuable note in hiscomprehensiveandexcellent history of theMarsupialia-f, againstthe mistake to which the young zoologistmight beliable, of con-cluding the analogicalgroups of the MarsupialiaandPlacentaliathusindicated to be of equalrank andvalue Ihave always par-ticipated inthisconvictionofthelower valueofthe Implacentalia

as compared with the Placentalia; and have used those terms

merely as useful collective or general signs of certain tions of structure, whichareassociatedwith thedevelopment and non-developmentof the placenta

modifica-In like manner, when indicating the highest generalization towhich I had arrived after comparisons of the dentition of the

signifying respectively the single and double set of teeth lopedin differentgroups oftheclass,Ihavebeencareful toguard

deve-myselffrombeing misunderstood,assupposingthatthe

monophyo-* RegneAnimal,ed.1829,vol i p.174

t Natural History of the Mammalia, 8vo 1815, part i p 14. I must

remark, however, that in stating "byProf.Owen andsome other naturalists,

the presentsection (Marsitpiata)isrankedasasubclass,"thereader,fromthepeculiarlyextendedsignificationgiventotheterm'Marsupiata,'mightbe mis-

led. TheMarsupialia form oneof the orders ofmysubclassImplacentalia.See thearticles'Marsupialia' and'Monotremata,'inthe "Cyclopaedia ofAna-tomy,"vol hi.1841

dontMonotremata,Bruta,andCetacea,formed anequivalentgroup

with the diphyodont bulkof the Mammalia, or that the binarygroups,definedbythissingle dental character,werenatural ones.Nothing more than a passing allusion seems needed to the

system of classifying the Mammalia on the modifications of theplacenta, originally proposed by Sir Everard Home*, and sincereproduced and modified byafewother naturalists Thegroup,

e.g associatedbythe character ofthediscoidplacenta,is aslittle

natural as that which would be composed on the basis of the

diphyodontdentition,ortheunguieulate feet Theassociation ofthe JRodentia and Insectivora with the Quadrumana, as in thelatestmodificationofthe placentarysystemf, isnot likely tocom-

mand acceptance The diffused placenta, as in the Mare, poise, Peccari, Rhinoceros, and Camel, would lead to an equallyheterogeneous assemblage In two well-defined minor groups,

Por-e.g the true Camivora and the true Muminantia, there existcharacteristic modifications of the placenta, viz the zonular and

cotyledonal respectively; but though thezonular type iscommon

to the Camivora, it is not peculiar to them; it is that of theplacenta in the Hyrax and theElephant, amongst the Ungulata

Solikewise the cotyledonal typecharacterizesthe placenta oftheSloth amongtheBruta

Primary DivisionsoftlieMammalia — Thequestionorproblem

of the trulynatural and equivalent primary groups of the class

beenpresenttomy mind whengatheringany newfactsinthe

ana-tomyoftheMammalia,duringdissectionsoftherarerformswhich

have died atthe ZoologicalGardens, oron otheropportunities

Thepeculiarvalueof the leadingmodificationsof themammalian

brain, inregardto theirassociationwith concurrent modifications

inother important systems of organs,was illustrated in detail in

theHunterian Course of Lectures onthe Comparative Anatomy

ofthe Nervous System, delivered by me at the Eoyal College of

Surgeonsin 1842 The ideaswhich were broached or suggested,during the deliveryof that covirse,I have tested byevery subse-

quent acquisition of anatomical knowledge, and now feel myselfjustified in submitting to thejudgement of theLinneanSociety,with aviewto publication,the followingfourfoldprimarydivision

of themammalianclass,basedupon thefourleadingmodifications

ofcerebral structure in thatclass

* Lectureson Comparative Anatomy,vol iii 4to p.445

Thebrainis thatpartofthe organization which, byits superiordevelopment, distinguishes the Mammalia from all the inferiorclassesofVebtebbata; andit is thatorganwhich Inowpropose

to showtobe theonethatbyitsmodificationsmarksthebestand mostnaturalprimarydivisions oftheclass

partiallyconnected togetherbythe'fornix' and'anterior

commis-sure: in the rest of theclassa partcalled 'corpus callosum' is

added,whichcompletes the connectingor'commissural'apparatus

Withtheabsence ofthisgreat superaddedcommissure* isciated a remarkable modification of the mode of developmentofthe offspring, which involves manyother modifications; amongst which are the presence of the bones called 'marsupial,' and the

asso-non-developmentofthe deciduousbody concernedinthe

nourish-mentoftheprogenybeforebirth, called'placenta; theyoungin

all this 'implacental' division beiug brought forth prematurely,

ascomparedwith therest ofthe class

This first and lowest primarygroup, or subclass, of Mammalia

signi-fying the comparativelyloose or disconnected stateofthecerebralhemispheres The sizeof thesehemispheres(fig 1,a)issuchthattheyleave exposed the olfactoryganglions(a), the cerebellum(c),

and more orlessoftheoptic lobes (b) ; their surface isgenerallysmooth; theanfractuosities,whenpresent, arefew and simple

The nextwell-markedstage in the developmentof the brainis

wherethe corpus callosum (indicated infig 2,bythe dottedlines

d, d) is present, but connects cerebral hemispheres as little vancedinbulkoroutwardcharacter as intheprecedingsubclass;thecerebrum (a) leavingboth theolfactory lobes(a) andcerebel-

ad-lum (c) exposed, and being commonly smooth, or with few and

simple convolutions in a very small proportion,composedof thelargest members of the group The mammals so characterizedconstitutethe subclass Lissencephala* (fig. 2)

Inthis subclassthetestesare eitherpermanentlyortemporarilyconcealed in the abdomen: there is a common external genito-urinary aperture in most; two precaval veins ('superior' or

'anterior venae cava?') terminate in theright auricle The mosal in most, and the tympanic in many, retain their primitiveseparation as distinct bones The orbits have not anentirerim

squa-* "On the Structure of the Brain in Marsupial Animals,"Philos.Trans

1837,p 87.

OF THE CLASS MAMMALIA. 15

of bone Besides these more general characters by which theLissencephala,in commonwith the Lyencephala, resemble Birds

andBeptiles,therearemanyother remarkableindicationsoftheir

affinityto the Oviparous Vertebratain particular orders orgenera

Fig 2.—Brainof Beaver.Fig.1.—BrainofOpossum

of the subclass Such, e.g., are the cloaca, convoluted trachea,supernumerary cervical vertebrae and their floating ribs, in the3-toed Sloth; the irritabilityofthemuscularfibre,andpersistence

of contractilepowerintheSlothsand someotherBruta; thelong,slender,beak-like edentulous jaws and gizzard of the Anteatersthe imbricatedscalesof the equally edentulous Pangolins,which

haveboth gizzard and gastricglandslikethe proventricularones

in birds; the dermal bony armourof the Armadillos likethat ofloricated Saurians; the quills of the Porcupine and Hedgehog

the proventriculus of the Dormouse and Beaver; the prevalence

ofdisproportionatedevelopmentofthe hind-limbsintheHodentia;

coupled, in the Jerboa, with confluence of the three chieftarsals into onebone, as in birds; the keeled sternum and wings

meta-of the Bats; the aptitude of the Cheiroptera, Insectivora, andcertainHodentiatofall,like Beptiles, intoa stateoftrue torpidity,associatedwith a corresponding faculty ofthe heart to circulatecarbonized orblack blood:—these, andthe likeindicationsof co-

affinity with the Lyencephala to the Oviparous air-breathing

me

16 PE0FESS0E OWEN ON THE CHAEACTEES, ETC.

intheelevation of differentgroupsofthe Lissencephalatoa higherplace in theMammalian series,andin their respective association,through somesinglecharacter,with better-brainedorders,according

to Mammalogical systems which, at different times, have been

proposedby zoologists ofdeserved reputation Such, e.g.,astheassociationof the long-clawedBrutawith the Ungulata*,and ofthe shorter-clawed Shrews, Moles and Hedgehogs,as well astheBats,with the Camivora\; of the Sloths with the Quadrumana^;

oftheBatswith thesamehigh order§; andofthe Insectivoraand

Bodentia in immediatesequenceaftertheLinnean 'Primates,' as

in the latest published 'System of Mammalogy,' from a

distin-guishedFrenchauthor||.

* Macleay, Linn Trans,vol xvi. (1833); Gray, Dr.J E.,MammaliaintheBritishMuseum,12mo.1843,p xii.

t Cuvier,RegneAnimal,1829,p.110

% DeBlainville,Osteographie,4to fasc 1 p.47 (1839)

§ Linnaeus,SystemaNaturae

|| Prof. Gervais, Zoologie et Paleontologie Francaise, 4to. 1852, p. 194

This scheme is avowedly anadoption of that proposed by Professor

Milne-Edwards, in the first volume ofthe 3rd seriesof the 'Annalesdes Sciences

Naturelles,'1844,in a paper entitled 'Considerations sur quelques Principes

relatifs h la Classification Naturelle des Animaux,'&c.; in referring towhich,

M.Gervaisstates hisconviction thatMilne-Edwards," amishors de douteles

rapports desRongeursavec les premiersMammiferes."—Annalesdes SciencesNaturelles,ser hi vol i p.251 The high andjustly-earned reputation of

both these naturalists renders it incumbentonme tostate thedoubts with

respect to the actual affinity of the Rodentia to the Quadrumana whichremained on mymindafter an attentiveperusal of the arguments urged byMilne-Edwards Thefirstof theseargumentsisbaseduponanallegedresem-blance of placental structure, expressed by the term " a placentadiscoide,"

apphed as a character to the Bimana,Quadrumana,Cheiroptera, Insectivora

andRodentia,collectively.

Thedegree of resemblanceinoutwardform, betweentheplacenta of theRat

or Hare, on theone hand, and theMycetesandMacacusontheother,seems

to me to bemore thancounterbalanced bythedifferenceofstructure. The

pedunculateandcotyloid placenta of theRatconsistsoffoetalpartsexclusively

the maternalareolarportionis as distinctfromit as it is in the cotyledon of the

Ruminant, andisapersistentstructureof theuterus. Thediscoid placenta ofthe Monkeyincludes a large proportion ofmaternal cellular structure,whichcomes away withthefoetalportion Thedifference inthe organic interblending

of thecirculatoryorgans ofmother andoffspring,between the Rodentia andQuadrumana,isofmuch morerealimportance thanthe degreeofsuperficial similarity Stillmoresignificant, inregardtogeneticgroundsofaffinity, isthegreatdifference in thedevelopment and function of the vitelhcleor umbilical

sac in the foetalmembranesof thetwoorders. But,asregardsoutwardform,the cotyloid placenta of theMuridcediffersmorefromthe thin,expanded and

subdivided placenta of the Hare,thanitdoesfromthatof the Marmoset

The third leading modificationof the Mammalian cerebrum is

suchanincrease inits relative size, that it extends over moreor

of form, that there are two distinct discoid placentae in Callithrix as in

Cercopithecus,Macacus and Semnopithecus; whilstinMycetes,as in

Troglo-dytes,thereisbut one suchplacenta.

The structure of the discoid placentain thePteropus,likethat of the Eat,

moreresembles that of thefoetalportion of the cotyledon in theCowthanthat

of the ceUulo-vascularspongyplacenta of the Quadrumana; andthis difference,

with the more important oneof the larger umbilical sac,appears tome togreatlyoutweighthe degree ofresemblanceinmere outward formof the placenta

Anyargumentinfavour of theaffinityofthe Cheiropteratothe Quadrumana,

basedonthat degree of resemblance,mustbeaffectedbythe prevalence of the

doublediscoid placentainthe Quadrumana SinceHunterfirstma.deknown

thatmodification*in a speciesofMacacus,which,from a comparison oftho

foetus now preserved in the Museum of the Royal College of Surgeons, I

behove to be the 'Wrinkled Baboon' of Shaw {Macacus rhesus, Desm.),

ProfessorBreschet has describedandfigured thetwoseparate discoidplacentae

in the smallSouthAmerican Squirrel-monkey(Callithrix sciureus,Kuhl),in

the Green Monkey {Cercopithecus sabceus, Desm.), and in the Long-nosedMonkey (Semnopithecus nasicus). Yet thiswell-marked modification of the

cellulo-vascular placenta is not constant in the Quadrumana,oreven in the

primary groupsof theorder. In the Platyrhines, e.g.,the Howler (Mycetes

seniculus, Kuhl) ha3 a single placenta, and amongst the Catarliines, I have

ascertainedthat, in theChimpanzee (Troglodytesniger) the placentais single,

as intheHumansubject.

The five flat placentallobes, virtually as distinct as ifthey were separate

placenta?, intheHare,resemblemorethesubdividedplacentaeof the Sloththan

thesinglehemispheroidpedunculate placenta of the Rat, or theflattened circular

placenta of theHowler Monkey In short, the observeddifferences ofform

intheplacentaeoftheRodentia, Insectivora, CheiropteraandQuadrumanaby

no meansjustifytheuse ofonegeneraltermasapplicabletothewholef.Thesecondargumentfortheassociationof the Insectivora, CheiropteraandRodentiawith the Quadrumana is takenfrom alleged conformity of cerebral

mypaperonthe'Brainsof the Marsupialia' (Phil.Trans 1837),Iharedescribed

and figured (pi v p.93) the brain of a Beaver (see fig 2, p.15) and that of

a small Monkey (Midasrufimanus,fig 3, p 19),showingthe absence of

cere-bralconvolutionsinboth Asthe cerebralhemisphereshavesincebeenshown

tobe equallysmoothinotherHapalidceof Isidore Geoffroy,inthePottoLemur%

(Perodicticus,Bennett),inMicrocebus§,andwithfewandfeeble tracesof

con-* AnimalEconomy,4to.1780

f Annales des Sciences Nat.torn cit p 96.

X Bijdragetot de Kennis van den Pottovan Bosnian, 4to. 1851,V der

Hoeven

§ ComptesRendusde l'Acad desSciences, Janvier19,1852

—

18 PROFESSOR OWEN ON THE CHARACTERS,ETC.

lessofthe cerebellum; andgenerallymoreorlessover theolfactorylobes Save in very few exceptional cases of the smaller andinferiorforms ofQuadrumana (fig. 3),thesuperficiesisfolded into

moreor less numerous gyri orconvolutions,— whence the name

Gyrencephala*, which Ipropose forthe third subclass of

Mam-malia (fig. 4)

Inthissubclassweshalllook in vain forthosemarks ofaffinity

to the Ovipara,which have been instancedinthepreceding

sub-classes Thetestes are,indeed, concealed,and through anobvious

volutionsiii Stenops tardigradus (Trolik,Reck d'Anatomiecomparee surle

genre Stenops,in N.Verhand.der lsteKlasse Koninkl Nederl.Inst.dam, Oct 1843); there is, to that extent, in the Quadrumanous order, a

Amster-superficial resemblance to thenon-convolutedbrains of theRodentiaandsectivora; but it is attended bythat moreimportant difference in theformandproportions of thecerebralhemispheres, ofwhichIexpressmyestimateby

In-thesystemofClassificationproposedinthepresent paper

The smooth hemispheresofthebrainof theMidas (fig 3.a) "extend,as in

mostof theQuadrumana,over the greater part of the cerebellum(c)"(Phd Trans

1837,p.93); itresembles,in short,thebrain of theHumanembryobeforethecerebralsurfacebeginstobe folded; whereashithe Insectivora,inthe Beaver,

andeven intheCapybara,inwhichthereareafewshallowaiifractuosities,the

cerebralhemispheresleavethe cerebellum quiteexposed

Withregardtothealleged contrast betweenthe brains of theRodentiaand

Carnivora, in the breadth of the anterior andmiddle part of the cerebral

hemispheres, acomparisonof the brains of theBeaver andCoatimoiidi,andofthe Porcupine andthe Civet Cat,leaves me entirelyunable to appreciate the

forceof the remark

Thethirdargumentfor the high position of the Rodentia, Cheiropteraand

InsectivoraintheMammalian scale, isdeducedfrom some particularsoftheir osteology,and principallyfrom thecommonpresence of the clavicle in them,

ascontrastedwith itsconstant absenceinthe Carnivora and Ungulata The

clavicle is present in all Quadrumana,butit isnot a peculiarcharacteristicofthe higher forms of the Mammalian class It is muchmoreconstantinthe

classof Birds andKeptiles: it is present in theMonotremes,inMarsupials,

and in most Bruta An affinity of the Insectivora and ofthe claviculate

Rodentia witha lower vertebrate type, might therefore be inferred from the

clavicle, at leastwith as much reason,aswiththe Apes andMan Astothe

shapeof the articularcavityfor the mandible, theRodentia differ morefrom

the Quadrumanain thisparticular thantheCarnivora do; whilst,inrespect

of the size, form, and persistent individuality of the tympanic bone, the

Rodentiaplainly show then" more essential relations to the oviparous typetheCarnivoraresembling theQuadrumanaintheearlycoalescenceofthe petro-

tympanicwith thesquamosalelements of thetemporalbone

Sucharesomeof the considerationswhich have inducedmeto set a different

valuethanM.Grervaisdoes,ontheargumentsadduced byProf.Milne-Edwards

infavour ofanassociation of theRodentia with theQuadrumana,ina highlyplacedprimarygroupof theMammalianclass.

asmanifested bythe bulk of some, bythe destructivemastery ofothers,by the addressandagility ofathird order And, through

the superior psychologicalfaculties— anadaptive intelligence dominating over blind instinct— which are associated with thehigher development of the brain, the Gh/rencephala afford thosespecieswhich have ever formed the most cherished companions and servitors,andthemostvaluable sources of wealthandpower,

pre-toMankind.

In Man the brain presents an ascensive step indevelopment,higher and more strongly marked than that by which the pre-ceding subclass was distinguished from the one below it. Not

only do thecerebralhemispheres (figs.5&6,a) overlap the

olfac-torylobes andcerebellum, but they extendin advance of the one,andfurtherback thanthe other (fig.6, c) Theirposterior deve-lopment is so marked, that anatomists have assigned to that partthe characterofa third lobe; it ispeculiar tothegenus Homo, and

equally peculiaristhe'posterior liornofthe lateral ventricle,'and

the 'hippocampus minor,'whichcharacterizethehindlobeof eachhemisphere Thesu-

perficial grey matter FiS"^ Negro

of the cerebrum,

throughthe number

anddepthofthe

con-volutions, attains its

in Man.

Peculiar mental

powers are

associ-ated with this

high-estformofbrain,and

am led toregard the

genus Homo, as not

merelya

representa-tiveof a distinct

or-der,butofa distinct

subclassof the

Mam-malia*, for which I

propose the nameof

'

(fig. 6)

"With this

preli-minary definition of the organic characters, which appear to

* Not beingable to appreciate,or conceive of the distinction between thepsychical plisenomena of aChimpanzeeandof aBoschisman,orofan Aztecwitharrested brain-growth,asbeing of a natureso essential as topreclude acomparisonbetween them,or as being otherthan adifferenceofdegree,I cannot shutmy

eyes to the significance of that all-pervading similitude of structure—every

tooth,every bone,strictlyhomologous,—whichmakesthedetermination of the

difference betweenHomoandPithecus the anatomist'sdifficulty. And,

there-fore, with every respect for the Authorof the "Records of Creation" (8vo,

1816, pp 18-21), I follow Linnaeus and Cuvierin regarding mankind as a

legitimate subjectofzoologicalcomparisonandclassification.

Fig.6.—Side view, Negro

guide to a conception of tlie most natural primary groups ofthe classMammalia, I next proceed to define the groups of se-

condary importance, or the subdivisions of the foregoing

sub-classes

In the Lyencephalous Mammalia some have the'optic lobes'simple, others partly subdivided, or complicated by accessoryganglions, whence they are called ' bigeminal bodies.' The

Lyencephalawith simple optic lobes are 'edentulous' orwithout

calcified teeth, are devoid of external ears, scrotum, nipples,and marsupial pouch: they are true 'testiconda;' they have

a coracoid bone extending fromthe scapula to the sternum,andalso an epicoracoid and episternum, as inLizards; they are un-guiculate and pentadactyle, with a supplementary tarsal bone

supporting aperforated spur in the male Theorderso terizediscalled'Monotremata,'inreference tothesingle excre-toryand generative outlet,which, however, isby no means pecu-

insectivo-rous, and arestrictlylimitedto AustraliaandTasmania

orduplicature of the abdominalintegument,whichin themalesis

everted,formingapendulous bagcontaining thetestes; andinthefemales isinverted,formingahidden pouchcontaining the nipples

and usually sheltering the young foracertain periodafter theirbirth: they have the marsupial hones in common with the Mo-

notremes; a much-varied dentition, especially as regards the

numberofincisors,but usually including4true molars; andnever

more than3 premolars*: the angle of the lowerjaw is more or

lessinvertedf

"With the exception ofone genus, Didelphys,which isAmerican, andanothergenus Ouscus, whichisMalayan, alltheknown exist-

ing Marsupials belong to Australia, Tasmania,and New Guinea

Thegrazingand browsing Kangaroosare rarelyseenabroadin full

daylight, save indark rainy weather MostoftheMarsupialiaarenocturnal Zoologicalwanderers in Australia, viewing its plainsandscanningitsscrubs by broaddaylight,arestruckbytheseem-ing absence of mammalianlife ; but during thebrieftwilightand dawn, or by the light of the moon, numerous forms are seen to

* " Outlines of aClassification ofthe Marsupialia," Trans Zool Soc.vol ii.1839

f Forother Osteological and Dentalcharacteristics of the Marsupialia,see

thepaper abovecited,andthat "Onthe Osteology of the Marsupialia," Trans

emerge fromtheir hiding-places andillustrate the variety ofsupial lifewithwhich manyparts of the continent ahound "We

mar-mayassociatewith theirlow position in themammalianscale theprevalent habit amongst the Marsupialia oflimiting theexercise

of thefacultiesof activelifetotheperiodwhentheyare shielded

bythe obscurity ofnight

The Lissencephalaor smooth-brained Placentals form a group whichIconsiderasequivalenttotheLyencephalaorImplacentals;

and which includes the following orders, Hodentia, Insectivora,Cheiroptera andBruta The Bodentiaare characterized by twolarge and long curved incisors in eachjaw, separated by a wideintervalfrom^the molars; andthese teeth are so constructed,and

thejawisso articulated, as to serve inthe reductionofthe foodtosmall particles byactsofrapidandcontinued gnawing,whencethe

name of theorder Theorbits are not separated from the poral fossa?. Thetestespass periodically fromthe abdomen into

tem-a temporary scrotum, andare associatedwith prostatic and

vesi-cular glands Theplacentais commonlydiscoid,butissometimes

a circularmass(Cavy),or flattenedand divided intothreeormore

lobes (Lepus) The Beaver and Capybara are now the giants ofthe order,which chiefly consists of small, numerous,prolificanddiversifiedunguiculategenera, subsistingwholly orinpartonvege-table food SomeEodents,e.g.theLemmings,perforinremarkablemigrations, the imprdse to which, unchecked by dangers or any surmountable obstacles, seems tobe mechanical Many Eodentsbuild very artificial nests, and a few manifest their constructiveiustinct in association In allthese inferior psychicalmanifesta-tions we are reminded of Birds Many Eodents hibernate like

Beptiles They are distributedoverall continents

The transition from the Marsupials tothe Eodentsismade by

by an equally easy step, through the smaller Opossums to the

Insectivora. This term is givento the order of small brained Mammals, the molar teeth of which are bristled withcusps, and are associated with canines and incisors: they areunguiculate, plantigrade, and pentadactyle, and they have com-plete clavicles The testes passperiodicallyfrom the abdomen

smooth-into a temporary scrotum, and are associated with large

pro-static and vesicular glands: like most other Lissencephala, theInsectivora have a discoid or cup-shaped placenta Their placeandoffice in South America and Australiaarefulfilled by Marsu-

THE The orderCheiroptera,with the exceptionofthe modification

of theirdigitsforsupporting thelarge websthat serve aswings,repeat the chief characters of theInsectivora; but a few of thelarger species are frugivorous and have corresponding modifica-tions of the teeth and stomach The mamma? are pectoral inposition,andthe penisispendulous inallCheiroptera The most

remarkable examples of periodically torpid Mammals are to be

found in the terrestrialandvolant Insectivora Thefrugivorous

Bats differ much in dentition from the true Cheiroptera, and woidd seem toconduct through the Colugos or FlyingLemurs,directly to the Quadrumanous order The Cheiroptera are cos-mopolitan

The order Bruta, called Edentata by Cuvier, includes two

generawhich are devoidof teeth; therest possess those organs,which, however, have no true enamel, are never displaced by asecond series, and are very rarelyimplanted in the premaxillarybones All the species have very long and strong claws The

ischiumas well asthe ilium uniteswiththe sacrum; theorbit is

not divided from the temporalfossa Ihavealready adverted tothe illustrationof affinity tothe oviparousVertebratawhich theThree-toed Sloths afford by the supernumerary cervical vertebraesupporting false ribs and bytheconvolution of the windpipe inthe thorax; and Imay add that the unusual number —three and twenty pairs—of ribs, forming a very long- dorsal,with a shortlumbar, region of the spine in the Two-toed Sloth,recallsalacer-

tine structure The same tendencyto an inferior type is shown

bythe abdominaltestes,thesingle cloacal outlet,thelowcerebraldevelopment, the absence of medullarycanals in the long bones

in theSloths, and by the greattenacityof lifeandlong-enduring

irritability of the muscular fibre, in both the Sloths and eaters*

Ant-The order Bruta is but scantily represented at the presentperiod One genus,Manis or Pangolin, is common toAsia and

Africa; the Orycteropus is peculiar to SouthAfrica; the restof

* Thislatter vital character attracted the notice of theearliest observers ofthese animals ThusMarcgraveandPiso narrate of the Sloth:—"Cormotum

suumvalidissime retinebat,postquam exeniptum erat ecorpore per

semiho-rium:—exemptocordecseteris visceribus,multdpostseniovebatetpedeslente

contrahebatsicutdormiturienssolet." Buffon,whoquotes theabovefromthe'Historia Naturalis Brasilia?,' p. 322, well remarks, "Par ces rapports, ce

quadrupedeserapproche non seulement dela tortue,dontilala lenteur,mais

encore des autresreptiles etde tousceuxquin'ont pasuncentredusentiment

—

24 PROFESSOR OWEN ON THE CHARACTERS, ETC.

the order, consisting of the generaMyrmecophaga, or true eaters, Dasypusor Armadillos, and Brady pus or Sloths, are con-finedtoSouthAmerica

Ant-Having defined the orders or subdivisions ofthetwo foregoingsubclasses, I may remark that the Lyencepbala cannot be re-

garded as equivalent merely to one of the orders, say Bodentia,

of the Lissencepbala, without undervaluing the anatomical racterswhichare soremarkable anddistinct inthe marsupial and monotrematousanimals Theanatomicalpeculiaritiesof the eden-tulousLyencepbala*appear to me to be,at least, of ordinal im-portance In these deductions I hold the mean between those

cha-who,with GreoffroySt Hilaire,woidd make oftheMonotremataadistinct classofanimals, orwith De Blainville, adistinct subclass(Ornithodelplies) ofMammalsf,andthosewho, withCuvier,would

Mr Waterhouse,a family of the Marsupiata% In like manner,whilst I regard theLyencepbala (Marsupiata of Waterhouse) asforming a groupofhigher rank than an order, Ido not consider

it as forming an equivalent primary group tothat formed byall

theplacental Mammalia.

It appears tome thatthe true value of the Lyencepbalaorplacentaliaisthatofoneoffourprimarydivisions or subclasses oftheMammalia; thatitstrue equivalencyiswith the Lissencepbala,

Im-and that all it's analogical relations are tobefound moretruly inthat smooth-brainedsubclass thanin thePlacentalia atlarge

ThefollowingTableexemplifiesthecorrespondenceofthegroups

in the Lyencephalous and Lissencephalousseries:

—

Bhizophaga§ BurrowingBodentia

Poephaga§ Dipodidceand Leporidce.Petaurus Pteromys

Plialangistidce Sciuridce and prehensile-tailed

arboreal Rodents

Phascolarctos Bradypus

Perameles and 3Iyrmecobius Erinaceidae

Chceropus Macroscelis.

* SeemyarticleMonotremata,inthe Cyclopaedia ofAnatomy,partxxvi.1841

t Osteographie,fasciculepremier,4to, 1839,p 47.

% Nat.Hist,ofMammalia,parti. 1845,p 18.

§ See the* Classificationof the Marsupialia,'inthe Zoological Transactions,

of the class mammalia. 25

Dideijyhysand Phascogale . Soricidce

Dasyuridce Centetes, Gymnura.

Echidna 3Ianis

Ornithorhynchus Orycteropus

The classificationproposed by M.Gervais,alreadycited (p.16),

in which the Rodentia, Cheiroptera, andInsectivora are associated

inthesamehighprimary groupwith theQuadrumana and Bimana,

isavowedly adopted fromthatpreviouslyproposedbyProf

theLinneanterms maywellbe retained; viz. Mutilata, Ungulata

and UnguicuJata, themaimed,thehoofed, andtheclawed series

These characters can only be applied to the Gyrencephaloussubclass; i e. they do not indicate natural groups, save in thatsection of the Mammalia To associate the Lyencephala and

Lissencephalawith the unguiculate Gyreneephala into onegreat

primary group, as in the Mammalian systems of Ray,Linnaeus

andCuvier,is amisapplication ofa solitarycharacter akinto thatwhich would have founded aprimary division on the discoid pla-centaorthediphyodontdentition No onehasproposedto asso-ciate the unguiculate Bird orLizard with the unguiculate Ape

andit isbut alittlelessviolation of naturalaffinitiesto associate

(unguiculate) division ofthe Mammalian class

The threeprimary divisions ofthe Gyreneephala areof highervalue than the ordinal divisions of the Lissencephala; just asthose orders are of higher value than the representative families

oftheMarsupials

The Mutilata,orthe maimed Mammalswith folded brains, are

so called because theirhind-limbs seem, as itwere, to have been amputated; they possess only the pectoral pair of limbs, and

thesein theformoffins: thehindendofthetrunk expandsinto

abroad, horizontally flattened,caudalfin. Theyhave large brainswith many and deep convolutions, are naked, and have neitherneck, scrotum, norexternalears

26 PROFESSOR OWEN ON THE CHARACTERS, ETC.

The firstorder, calledCetacea,in this division are eithertulousormonophyodont, andwithteethofonekindand usually ofsimple form Theyaretesticondaandhaveno'vesiculaeseminales.'

nostrils—single or double—are onthe top of the head, andcalledspiracles or "blow-holes." They are marine, and, for the mostpart, range the unfathomableocean; though with certain geogra-phical limits as respects species They feed on fishesor marineanimals

Thesecondorder, called Sirenia, haveteethofdifferent kinds,incisorswhich are preceded by milk-teeth,and molars with flat-

tenedorridged crowns, adaptedforvegetable food Thenostrilsaretwo,situated at theupperpartofthe snout; thelips arebesetwith stiffbristles; themammae are pectoral; thetestes areabdo-minal, as inthe Cetacea,butare associatedwithvesiculaeseminales

The Sirenia exist nearcoasts orascend large rivers; browsing on

fuci,waterplants orthe grassof theshore There ismuch intheorganizationofthis orderthat indicates itsaffinityto membersofthe succeedingdivision

In the Ungulata the four limbs arepresent, butthatportionofthe toe which touches the ground is incased in a hoof, which

blunts its sensibility and deprives the foot of prehensile power

Withthe limbs restricted to support and locomotion, the

Ungu-latahave noclavicles: thefore-leg remains constantlyinthestate

of pronation,andthey feedonvegetables

A particular order, or suborder, of this group is indicated bycertain South Americangenera,e.g Toxodon andJVesodon*, withlong,curved,rootless teeth,havinga partialinvestmentofenamel,

and with certain peculiarities of cranial structure: the name Toxodontia is proposed forthis order, alltherepresentatives of

whichare extinct

Asecond remarkableorder,most ofthe membersofwhichhave,

also, passed away, ischaracterized by twoincisors in the form of

long tusks; inone genus(DiiwtJierium)projectingfromtheunderjaw, in anothergenus (J?lej)Jias) fromtheupperjaw, andinsome

ofthespeciesofathirdgenus(JSLastodoii),frombothjaws Thereare nocanines; the molarsarefew, largeandtransverselyridgedthe ridges sometimes few and mammillate, often numerous and

with every intermediate gradation Thenoseisprolongedinto acylindrical trunk, flexible in all directions, highly sensitive, and

terminated by aprehensile appendage likea finger: on this organ

isfounded the name Proboscidia given to theorder The feetarepentadactyle, but are indicated onlyby divisions ofthe hoof;the testes are abdominal; the placentais annular*; themamma?

are pectoral

Both the present and preceding orders of Ungulata may becalled aberrant: the dentition of theToxodon, and several parti-culars ofthe organizationof the Elephant,indicate an affinity tothe Eodentia; thecranium oftheToxodon,like thatofthe Dino-there,resemblesthat oftheSirenia initsremarkablemodifications

The typicalUngulate quadrupeds are divided, according tothe

Abtio-DACTTLAf Intheperissodactyle orodd-toedUngulata—odd-toed

at least in regard totbe hind-foot,—the dorso-lumbar vertebrae

differ in number in different species, but are never fewer than twenty-two; thefemurhasathirdtrochanter; andthemedullaryarterydoesnot penetrate the fore-partofits shaft Thefore-part

of the astragalus is divided into two very unequal facets The

some disproportionately so, and the digit is symmetrical: the

sameapplies tothe ectocuneiformand the digitwhichit supports

inthehind-foot Ifthe speciesbe horned, thehorn issingle; or,

ifthere be two, theyareplacedonthemedianlineofthehead,one

behind theother,eachbeing thusan oddhorn Thenasalsexpandposteriorly Thereisawell-developedpost-tympanic processwhich

is separatedbythe truemastoidfromtheparoccipital intheHorse,but unites with the lower part of theparoccipital intheTapir,and seemstotaketheplace ofthe mastoidintheRhinoceros and Hyrax.

Thehinderhalf,or a largerproportionofthepalatines enters intothe formation of the posterior nares,the oblique aperture ofwhich

commencesinadvanceeitherofthelastmolar,or,asinmost, of thepenultimateone Thepterygoid process hasabroadandthick base,and is perforated lengthwise by the ectocarotid The crown of

from one tothreeofthe hinder premolarsisascomplexasthose ofthe molarsJ: that of the last lower milk-molar is commonly bi-

lobed Tothese osteological anddental charactersmaybeadded

some important modifications of internal structure, as, e.g.thesimple form of the stomach and the capacious and sacculated

* Besides the annularplacenta there is a subcircularvillouspatchat eachpole of the chorionic bag, bywhich it derived additional attachment to the

uterus, inthe Elephant

t From 7rept<Tcro(>dicTv\os, qui digitos habet impaves numero; andapriov, par, ScuctvXos, digitus.

caecum, which equally evince the mutual affinities of the toed or perissodactyle hoofed quadrupeds, and theirclaims to beregarded as a natural group of the Ungulata The placenta is

odd-replaced by a diffusedvascular villosityof the chorion in all therecent generaof this order,excepting the little Hyrax, in which

there is a localised annular placenta, as in the Elephant But

the diffused placenta occurs in some genera of the next group,

showingtheinapplicabilityof thatcharacter to exactclassification

Manyextinct genera, e.g.Coryphodon, PUolophus, Lqphiodon, pirotherimn, Palceotherium, Ancitherium, Hipparion, Acerotherium,Elasmotherium, &c, have been discovered,which once linked to-

Ta-gether the now broken series of Perissodactyles, represented by

theexistinggenera Rhinoceros, Hyrax, Tapirus, and Equus.

In the even-toedor 'artiodactyle' Ungulates, thedorso-lumbarvertebras are the same in number, as a general rule, in all the

species,being nineteen The recognition of this important racter appears to have been impeded by the variable number

cha-of moveable ribs in different species of the Artiodactyles, thedorsal vertebra?, which those ribs characterize, being fifteen in

of this distinction has been exaggerated owing to the common

conception of the ribs as special bones distinct from the

ver-tebrae, and their non-recognition as parts of avertebra lent to the neurapophyses and other autogenouselements The

equiva-vertebral formulae of the Artiodactyle skeletons show that thedifference in the numberof the so-calleddorsal and lumbarver-tebrae does not affect the number of the entire dorso-lumbarseries: thus, the Indian "Wild Boar has d. 13, I. 6=19; the

Domestic Hog and the Peccari have d.\4>, 1.5=19; the popotamus has d.15,I.4=19; theGnu and Aurochs haved.14,

Hip-1.5=19; theOx and mostofthe trueRuminantshaved.13,1.6=

19; the aberrant Ruminants have d. 12, I. 7=19 The naturalcharacterandtrueaffinities ofthe Artiodactyle grouparefurtherillustratedbythe absence of thethirdtrochanterinthe femur,and

bytheplace of perforationof the medullaryarteryattheforeand upperpart of theshaft,as intheHippopotamus,theHog, and most

oftheRuminants The fore partof the astragalusis dividedintotwoequalorsub-equalfacets: the osmagnum doesnot exceed, or

islessthan,the unciformeinsize,inthecarpus; andtheform is less, or not larger, than the cuboid, in the tarsus The

ectocunei-digitanswering to thethird inthepentadactyle foot is

unsymme-symmetrical

pair If tlie species be horned, the horns form one pair or twopairs; theyare never developedsingly,of symmetrical form,from

the median line Thepost-tympanic does not project downward

distinctly fromthe mastoid,nor supersedeitin anyArtiodactyle;

andtheparoccipitalalways exceeds both those processes inlength

The bonypalateextends furtherbackthan inthe Perissodactyles;thehinder aperture of thenasalpassages ismoreverticaland com- mencesposterior to thelast molartooth Thebase of theptery-goid process is not perforated by the ectocarotid artery The

crowns of the premolars aresmallerand less complex than those

of the true molars, usually representing half of such crown

Thelast milk-molaris trilobed

Tothese osteologicaland dental characters may beadded some

importantmodifications ofinternal structure, as, e.g.thecomplex formofthestomachintheHippopotamus,Peccari,and Ruminants;

thecomparatively small andsimple caecumandthe spirallyfoldedcoloninallArtiodactyles,which equally indicatethemutual affini-

tiesoftheeven-toed hoofed quadrupeds,andtheir claimstobe

re-gardedas anatural group ofthe Ungulata The placenta is

dif-fusedinthe Camel-tribeand non-ruminants; iscotyledonal inthetrue Ruminants Manyextinct genera, e.g.Chceropotcwms, An-

thracotherium,Uyopotamus,Entelodon, DicJwdon,Merycopotamus,Xipltodon, DicJiobime,AnoplotTierium, Microtlieriwn,&c,havebeendiscovered, which once linked together the now brokenseries ofArtiodactyles, representedbytheexistinggenera, Hippopotamus,Sus, DicotyJes, Camelus, Auc7ie?iia,Moschus, CameJopardalis, Cer-vus, Antilope, Ovis,and Bos

A well-marked, and at the present dayvery extensive dinategroup of theArtiodactyles, is called Ruminantia,in refer-

subor-encetothe second mastication towhich the food is subject afterhavingbeenswallowed; the act of rumination requiring a pecu-

liarly complicated form of stomach The Ruminants have thecloven foot,' i e. twohoofed digits on each foot formingasym-metricalpair, asbythe cleavage of a singlehoof; inmost speciestwosmallsupplementaryhoofedtoes are added The metacarpals

ofthetwofunctional toes coalesce toforma single'cannon-bone,'

as do the corresponding metatarsals The Camel-tribe have the

upper incisorsreduced to asingle pair; in therest of the

Rumi-nants theupperincisors arereplacedbya callous pad Thelowercaninesarecontiguous,and, save inthe Camel-tribe, similar tothe

six lower incisors, forming part of the same terminal series of

interval Thetrue molars have theirgrinding surfacemarked by two doublecrescents, the convexityof whichis turned inwardsintheupper and outwards intheunderjaw.

Manyfossil Artiodactyles, with similar molars,appear to havedifferedfrom theRuminants chiefly byretaining structures whichare transitoryand embryonicin mostexisting Ruminants, as, e.g

upper incisors and canines*, first premolars, and separate carpalandmetatarsalbones; these are among the lostlinks thatonceconnectedmoreintimatelythe Ruminantswith theHog and Hippopotamus.

meta-The Pachyderms in the Cuvierian systemincluded allthe

non-ruminant hoofed beasts; theywere divided bythe great French

anatomist into the Proboscidia, Solidungula, and Pachydermataordlnaria, thelatteragain being subdivided according to the odd

evidenceto showthatthe rightprogressionoftheaffinities oftheUngtdatawas broken bytheinterpositionof the Horse and otherPerissodactyles betweenthe non-ruminant*oromnivorous andru-minantArtiodactyles;and that toohighavaluehad beenassigned

to theRuminantia by making them equivalent to all the other

Ungulatescollectively%.

* Inanew-bornDromedary (Camelus Dromedarius,L.),whichperishedm

the birthattheLondonZoologicalGardens, the followingwasthestateof the

dentition. In the upperjawtherewere six deciduousincisors (3—3),whichwerecalcined,andpresented a larger proportionalsize than any rudimentsofthose teeth thathave beennoticedinordinaryRuminants, andtheyleavecon-spicuousalveoli inthe premaxillaries: the deciduous canineandfirstfunctionalmilk-molar(d 2) weresmall,thelatterwith a simple crown; thesecond {d 3)

andthird(d 4)molarswerelarge,bilobed,andeach lobewasbicrescentic. In

thelowerjawthe six incisors and two canines form a semicircular series ofnearlyequalteeth,with overlapping leaf-shaped crowns, the deciduous canines

more resembling the incisors than the permanent ones do: the functional

molarsarebut twoinnumber, oneach side ; thefirst is small,simple,conical,

compressed, notched behind; the second is very large and three-lobed, eachlobe beingbicrescentic,andthelastthelargest. Onlythesummitsof thecres-

cents of themolarteethhadpierced thegum (Catal.of Osteology,Mus Roy

Coll.ofSurgeons,vol ii p.577,4to,1853)

f QuarterlyJournalofthe GeologicalSociety,December1S47

X Since thecommunicationofmypaperontheclassificationandaffinitiesofthe hoofed animals to the Geological Society, Nov 3, 1847, in which the

grounds forthe division of the Ungulata into two orders, according to theparity or imparity of thedigits, asproposed in my'Odontography,'aregiven

in detail, the ideahasbeenventilated andmoreor less adoptedbyM Pomel(Comptes Rendus de l'Acad des Sciences,June19,1848),and byM.Gervais(ZoologieetPaleontologie Franchise,p 42). Thelatterexperiencedpalaeonto-

OF THE CLASS MAMMALIA. 31

Thethird division of the Gyrencephala enjoy a higher degree ofthe sense of touchthrough the greaternumber andmobility of the

digits, andthe smaller extent towhichtheyare coveredby horny

matter This substanceforms a single plate, in the shape of aclaw ornafl, which is applied to onlyone of the surfaces of theextremityofthedigit,leaving the other,usuallythelower, surfacepossessed of its tactile faculty; whence the name Unguiculata,applied to this group, which, however, is more restricted and

natural than the group to which Linnaeus extended the term.All the speciesare 'diphyodont,' and the teeth have a simple in-

vestment ofenamel

The first order, Carnivora, includes the beasts of prey, perly so called With the exception of a fewSeals, the incisors

teeth, and usually exhibiting a full and perfect development aslethalweapons; themolars graduate from atrenchantto a tuber-culate form,inproportionasthe dietdeviates from onestrictlyof

fleshtooneofamoremiscellaneouskind Theclavicleisrudimental

orabsent; theinnermostdigitisoftenrudimentalorabsent; theyhave novesicuke seminales; theteats areabdominal;the placenta

is zonular The Carnivoraare divided,accordingto modifications

ofthelimbs, into 'phmigrades,' 'plantigrades,' and'digitigrades.'

Inthe Phmigrades (Walrus, Seal-tribe) both fore and hind feet

are short,and expandedinto broad,webbedpaddles for swimming,

thehinder ones being fettered by continuation of integument tothe tail. In the Plantigrades (Bear-tribe) the whole or nearlythewhole of the hindfootformsasole, and rests onthe ground

Inthe Digitigrades (Cat-tribe,Dog-tribe,&c.) only the toestouchthe ground, the heelbeingmuchraised

Ithasbeen usualto place the Plantigrades atthe head of theCarnivora, apparently because the higher order, Quadrumana, is

plantigrade; butthe affinities of the Bear,as evidenced bynal structure,e.g the renal and genital organs, are closer totheSeal-tribe*; thebroaderandflatterpentadactylefootof theplanti-theminto 'Pachydermes herbivores' and 'Pachydermes omnivores,' respect- ivelyequivalenttomyPerissodactylaandArtiodactyla,whichlattertermsM.Pomel adopts M Grervais writes: "Les pachydermes omnivores se lient

inter-d'unemanieresiintimeaux Ruminants par lesChevrotainset les Chameaux

qu'il estdevenuimpossibledeseparer,commeordredifferentdeceluides

Rumi-nants l'ensemble decesPachydermes,autrefoisconfondusaveclesPachydermes

herbivores."

—

Op.cit.Expl dePlanchexxxvi.p 6,4to,1854

* 'Catalogue of the PhysiologicalSeries,' Mus.R.Coll.of Surgeons, 4to vol

Mr Waterhouse,

grade isnearerinform tothe nipper of the Sealthanisthe more

perfect digitigrade, retractile-clawed,longand narrowhindfoot ofthefeline quadruped,whichisthe highestand mosttypicaloftheCarnivora

The nextperfectionwhich issuperinduceduponthe unguiculatelimbissucha modification inthesize,shape, position,anddirection

of the innermostdigit,thatitcanbe opposed,as athumb,totheother

digits,thusconstitutingwhatisproperlytermed a'hand.' Those

Unguiculateswhich havebothforeandhind limbs so modified, or

at leastthe hiudlimbs,form the orderQuadrtthana Theyhave

|^| incisors*, and ^broad tuberculate molarst; perfectclavicles,

pectoralmammae,vesicularandprostatic glands,asimpleor slightly

bifid uterus, and a discoid, sometimes double, placentaJ. The

structural division The Strepsirhines are those with curved ortwistedterminalnostrils,with much modified incisors, commonly

|^|; premolars^ or |=Hin number, andmolars with sharp

tuber-cles; thesecond digit of the hind limb has a claw Thisgroupincludes the Galagos, Pottos, Aye-Ayes, Loris, Indris, and the

whence the group diverges in one direction to the continent ofAfrica,inthe other to the Indian Archipelago The Platyrhinesarethosewith thenostrilssubterminal and wideapart; premolars

?5j in number, the molars with blunt tubercles; the thumbs ofthe fore-hands not opposableorwanting;the tailinmostprehen-

sile ; theyare peculiar to South America The Catarhines havethe nostrils oblique and approximated below, and openingabove

and behind the muzzle: the premolars are ^J in number; the

Old World, and, saveasingle species onthe rock ofGibraltar, toAfricaand Asia The highest organized family of Catarhines is tailless, and offers in the Oraug and Chimpanzee the nearestapproachtothehumantype

undersideoftheramus,alittle inadvanceof the angle of the lowerjawinthe

Ursidcn,remarks:—"Thesamecharacteris alsofoundinmanySeals(Phocida),

whichin severalotherrespectsappeartoapproachthebears."—Proc.Zool.Soc.Sept 1839

* Withfew exceptionsintheanomalousLemuridce

t Reducedto|^|intheMarmosets (Hapale,Mydas).

% Amongthe Platyrhines, the placentais single inMycetes,doublein

Calli-thrix: amongthe Catarhines, the placentaisdoubleinMacacus,Cercopithecus,

OJF T1IE CLASSMAMMALIA 33

Thestructural modifications iu the genus Homo, —thesole

re-presentative of the Archencephala,— moreespeciallyof the lowerlimb,by whichthe erect statureandbipedalgait are maintained,aresuchastoclaimforManordinal distinctionon merelyexternalzoological characters But as Ihave already argued, his psycho-logical powers, in association with his extraordinarilydevelopedbrain, entitle the group which he represents to equivalent rank

with the other primarydivisions of the classMammalia founded

on cerebral characters In this primary group Man forms but onegenus, Homo, and that genus but one order, calledBimana,

onaccount of the opposablethumb being restrictedtotheupperpair of limbs The testes are scrotal; their serous sacdoes not

communicate with the abdomen; they are associated with

vesi-cularandprostatic glands Thepenis is pendulous,andthe pucehasa fraenum The mammae are pectoral Theplacentais

pre-asingle, subcircular, cellulo-vascular, discoidbody

Man has onlya partial covering of hair, which is not merelyprotective of thehead, but is ornamental anddistinctive of sex

The dentition of thegenus Homo is reduced to thirty-two teeth

by the suppression of the outer incisor and the first two molars of thetypicalseries oneachside of bothjaws, the dentalformula being:

pre-—

]—

l

2—2 3—3 qo

All the teeth are of equal length,and there is no break in the

series; theyaresubservientinMannot onlytoalimentation, but

tobeauty andtospeech

as in Quadrumana,but applied flat to the ground; the legbearsverticallyonthefoot; theheelisexpanded beneath; thetoes areshort, butwith the innermost longer and much larger than the

rest, forminga 'hallux' orgreattoe,which isplacedon the sameline with, andcannot be opposed to, the othertoes; thepelvis is

short, broad, and wide, keeping well apart the thighs; and the

neck ofthefemuris long,andformsan openanglewith the shaft,

increasingthe basis of supportfor the trunk The wholebralcolumn, with its slight alternate curves, and thewell-poised,short,but capacious subglobular skull, are in like harmony withthe requirements of the erect position The widely-separatedshoulders, with broad scapulas and complete clavicles, give afavourable position to the upper limbs, now liberated from theserviceof locomotion, with complex joints forrotatoryaswell as

flexile movements, and terminated by a hand of matchless fection of structure,the fit instrumentfor executing the behests

per-ofa rational intelligence anda freewill Hereby, thoughnaked,

Mancanclothe himself, andrival all nativevestmentsinwarmth

andbeauty; though defenceless, Mancan armhimselfwith everyvariety of weapon, and become the most terribly destructive ofanimals Thus he fulfils his destiny as the supreme master of

this earth,andof the lower Creation

Inthese endeavours to comprehend how Naturehasassociatedtogether her mammalianforms, theweary student quits his taskwith a conviction that, after all, he has been rewardedwith but

an imperfect viewof such natural association The mammalian

class has existed, probablyfrom the triassic, certainly from thelower oolitic period; and has changed its generic and specific

formsmore thanonceinthelonglapse ofages,duringwhich

life-workhas been transacted on this planetby animals of thathighgrade of organization Not anyof themammaliangenera of thesecondary periods occurinthe tertiaryones No genus foundinthe older eocenes (plastic and septarial clays,&c.) has been dis-

covered inthenewereocenes Extremely feweocenegenera occur

in miocene strata, and none in thepliocene Many miocene nera of Mammalia are peculiar to that division of the tertiary

ge-series Speciesindistinguishablefromexisting ones begin to pear onlyin the newer pliocene beds Whilst some groups, as

ap-e.g the Perissodactyles and omnivorousArtiodactyles, have been

gradually dying out, other groups, as e.g the true Ruminants,

have been augmentingingeneraandspecies

In many existing genera of different orders there is a more

specialized structure, a greater deviation from the general type,than intheansweringgenera of themiocene andeoceneperiods;such later and less typical Mammalia domore effective work bytheir more adaptivelymodified structures The Ruminants, e.g

more effectuallydigestandassimilate grass, and form out ofit amore nutritive and sapid kind of meat, than did the antecedent

moretypicalorless specializednon-ruminant Herbivora

The monodactyle Horseisabetterandswifterbeast of draught

and burthen than its tridactyle predecessor the miocene rioncouldhave been Thenearer to a Tapir or a Rhinocerosinstructure, the furtherwillan equine animal beleftfromthe goal

Uippa-in contending with a modern Racer The generaFelis and

Ma-and

OP TUE

short strongjaws,become,thereby,morepowerfullyandeffectively

destructive than the eoceneHyamodon with its typical dentitionand three carnassial teeth on each side of its concomitantlypro-longedjaws couldhave been

Much additional and much truer insight has, doubtless, been

gained into the natural grouping of the Mammalia since

palae-ontology has expandedour survey of the class; but ourracterized groupsdo butreflectcertainmental conceptions,which

best-cha-must necessarily relate to incomplete knowledge, andthat as quired at a given period oftime Thus the orderwhich Cuvier

the debrisof agroup, knownata subsequent periodtobeamore

naturalorder

We cannot avoid recognizing, in the scheme which I now

submit, the inequality Avhich reigns amongst the groups, which

our present anatomicalknowledgeleadsusto place in onelineorparallel series as orders I do not mean mere inequality as re-

spectsthe number andvariety ofthefamilies, genera, andspecies

of such orders, because the paucity or multitude of instancesmanifesting a given modification or grade of structure in noessentialdegreeaffects the valueofsuch grade or modification

The orderMonotremata is not the less ordinally distinct from

the Marsupialia, becauseit consists ofbut two genera,thanistheorderBimana fromthat of Quadrumana,because itincludes only

a single genus So likewise the anatomical peculiarities of theProboscidia, Sirenia, andToxodontia call,at least,forthose generalterms, toadmit of the convenient expression of general proposi-tions respectingthem; and someofthese general propositions are

of a value as great as the organic characters of more expandedorders

There are residuaryor aberrant forms in some of the orders,which,tothe systematistdisagreeably,compelmodificationsof thecharacters thatwouldapplytothe majority of suchorders Thefly-

ingLemurs(Galeopitlieci),the rodentLemurs(Cheiromys), the slow

Lemurs(Boris, Otolienus), forbidanygeneralization as toteeth or

nailsinthe Quadrumana,whilstthey continueassociatedwith thatorderbythe characterofthe hinderthumb;which,bytheway, theypossess incommonwith the pedimanous Marsupials The large,volant,frugivorous Bats(Pteropus)are equallyopposedtothe ap-plicationof acommondental charactertotheCheiroptera Theyareassociated with theinsectivorous Bats on account of the common

externalformarising out of the modification of theirlocomotive

organs for flight, just as the Dugongs and Manatees are ciated with the Cetacea onaccountof theirresemblancetoFishesarisingout of the samemodificationof thelocomotive system for

asso-anaquaticexistence Theherbivorous Cetacea arenow separated

from thepiscivorous Cetacea asa distinct order; andwith almost

as good reasonwe might separate the frugivorous from the

in-sectivorous Cheiroptera; thecasesarevery nearlyparallel

Nature, in short,is not so rigid a systematistasMan Therearepeculiar conditions of existencewhich she is pleasedshallbe

enjoyedbypeculiarlymodifiedmammals;thesepeculiaritiesbreak

throughtherulesofstructurewhichgovern the majorityof speciesexistingand subsisting under the more generalconditions of ex-istence, towhichthelarger groups ofMammaliaarerespectivelyadjusted

Oneclass oforgans seems to govern one order,another class

another order; the dental system,which is so diversified in theMarsupialia and JBruta, is as remarkablefor its degree of con-stancyin the Rodentia andImectioora Bat, as a general rule,

thecharactersfrom the dental, locomotive, andplacentalsystemsaremoreclosely correlated in the Gyrencephalous ordersthan inthoseintheinferior subclasses ofthe Mammalia.

Inthe subjoined tabularviewof the classificationof the

Mam-malia, the groupsbelowtheranksoforders areinsertedmerelyasillustrations of thoseorders, not as equivalent subdivisions, or asthemost natural subdivisions of those orders, intowhich it hasnotbeen theaimofthe present paperto enter