56.9,33N
Article XLI.- PRELIMINARY DIAGNOSIS OF AN APPARENTLY
NEW FAMILY OF INSECTIVORES.
BY H. E. ANTHONY.
PLATE XXIII.
In July, 1916, while searching for remains of fossil mammals in Porto Rico, inaccordance with the plan for a natural history survey of the Island, amostinteresting type of insectivore was discovered. A hint of its presence had been given in 1915 whenfragments of two mandibles were sent in with thematerial collected by Dr. Franz Boas,' but this material was too incom-plete to give any clue as to the exact nature of the animal. Thefirst satis-factoryspecimen, apracticallycomplete cranium, was found by the author's wife, Edith I. Anthony, in the Cueva Clara, near Morovis, July 19, 1916. Later,July 29, an abundance of material was secured in a cave more or less well known as the Cathedral Cave (Cueva Catedral) about two miles from Cueva Clara. Owing to the time that will be needed to thoroughly work up theresults of this expedition, I am putting this insectivore on record in a preliminary paper with the intention of writing more fully at a later period.
Theauthordesirestoexpresshis appreciationof theassistanceand valu-able counsel given him in the study of this animal by Dr. W. D. Matthew and Dr. W. K. Gregory of the Department of VertebratePalaeontology and by Dr. J. A. Allen, Curator of the Department of Mammalogy and Orni-thology.
Nesophontes2 editha gen. et sp. nov.
Type, No. 14174, Dept. Vert. Pal., from the Cueva Catedral, near Morovis, Porto Rico, July 29, 1916; collector, H. E. Anthony. The type skull is nearly
perfect,being broken onlyattheextreme tip of the premaxillaries and in the audi-toryregion. The bone is denseand hard although brittle from age as is to be ex-pected.
The skull isnarrow and elongate, tapering graduallyin width from theregion
ofthe squamosals to the end of the nasals. The entiresuperior outline is in
practi-cally the sameplane, the frontals andparietalsbeing butslightlyraisedabove the nasals and occipital crest. Thebraincase is subeylindrical andsomewhat inflated
1 For reportonthismaterialseeAnn. NewYorkAcad.Sci.,Vol.XXVII,p.193.
2Nesophontes.- From v~aos, island and
46vr-qs,
aslayer; edithae,inhonorof theauthor'swife,EdithI. Anthony, who found the first skull and directed attentiontothe presenceof the animal. 725
Bulletin American Museum of Natural History. [Vol.XXXV,
just back of the orbits. A very low sagittal crest is present and the lambdoidal crest is well developed. The interorbital region shows no noticeable constriction, having
parallelmargins and a rather flat transverse aspect. The nasal sutures cannot be tracedbutthe rostrum is long and tube-like, being higher than wide. The zygomatic arch isincomplete,lacking the jugal, but the zygomatic roots on both the squamosal andthe maxillary are present as distinct processes. There is a pair of welldeveloped
infraorbital foramina. The foramen magnum is proportionally very large. The interpterygoid fossa is well developed, being long and deep, but thepterygoid fossae are reduced to obsolescence. The dentition is essentially of a primitive type and isrepresented by the formula, I., 3; C., 1; P., 3; M., 3. Theincisors of a topo-type (thetype lacks three incisors) areranged innearlyparallel rows, onlyslightly
convergentanteriorly, which means that there is a noticeable diastema between the first incisors ofeach side. These teeth are slightly procumbent,especiallythefirst,
which is the largest. All aresimplein structurebeing sharpandsubconical. The canine istwo-rooted, longanddagger-like, almoststraight,and concave on the inner side. Thefirstand secondpremolarsaresimple and subequal; thethird is
submolari-form and triangularin cross-section with a high sharp cusp on the outermargin. The three triangularmolarsall have sharp, welldevelopedcusps,the first twobeing
of equal size, the third rather smaller.' The tooth rows along the inner margins are practically parallel fromthecanine backward. The glenoid fossa is deep and
extensive. The auditoryregion isincompletein everyskull nowavailable andthe presentevidence is tooinadequateto statethatthisformhas or has not anosseous auditalbulla. Thepalate is shallowlyconcave and there is apair ofsmall, sepa-ratedincisiveforamina.
No mandiblewasfound in unquestionable associationwith the type skull buta
lowerjawselected as of proper size to matchthe craniumgivesthefollowing char-acters. The horizontal ramus of the mandible is curvedevenly belowthroughout itsentirelength and theascendingramusis broadandheavy. Thecoronoid process isprominently developed andthere is aconspicuousfalciformangularprocess. The
condylehas an unusual aspect due to theobliqueventralpositionofthearticulating
surface. The insertion areas for thetemporalandpterygoidmuscles are well defined. The lowerdentitionissimilartothe upper(I.,3; C.,1; P., 3; M., 3)withthe excep-tion that the last premolar is simple, not submolariform; there is probablyno diastema between the first incisors of each side, and the molars are of typical tuberculo-sectorial form. The teeth of the lower rowarewell in contactwith one
another. There are twoanterior dental foramina belowpm2andpm4respectively.
Measurements.- Greatestlengthofcranium, 41mm.; greatest breadth (across
zygomatic roots onsquamosals,17 mm.; greatest breadth of braincase, 14.5mm.; interorbital breadth, 8.5 mm.; widthof rostrum(backofcanines),6.5 mm.; length
ofmaxillarymolarseries, 12.5mm.; greatest transverse extent ofiml,3mm.;
pala-talwidth at pm4, 5 mm.; length of mandibular molar series, 13.5 mm.; greatest length of mandible, 28 mm.
Thereappearsto be agreatdifference in size betweensexes, unless later
research establishes more-than onespecies, the male presumably being the
larger, and if thisassumption proves correct,thetypeis a female. Thetype
1Thedetails of the dentitionwillbemorefullytreated in thepapertofollow. 726
Anthony, New Family of Insectivores.
skull shows evidence of full maturity. Among the topotype material considerable individual variation is shown, some of the variation being correlated with the size difference and
consequently
may bedue to sexual (lifferentiation. Such a variation is the much greater developmenitof the lainbdoidal crest in a large skull fully fifty per cent larger than the type. The teeth do not apparently change much with wear, sharp cuYsps being prominentin every instance.Nesophontes has no evident close relationships with any known genus. Its characters are so strikingly different andof a nature so important that full justice may be done them onlv by the erection of a new fainily. The PortoRican animal is a true insectivore, thus removing at once from con-siderationanymember of the CarnivoraorCreodonta to which groups some ofthe charactersmightprompt oneatfirstto turn.
Amongthe Insectivora the charactersofthedentitionalone aresufficient to exclude the entire section of the Zalambdodonta 1 while for the same
reasontheLeptictideemaynotinclude Nesophontes. All theZalambdodonta havespecialized ormodifiedincisors andcaninescontrasting with the primi-tive conditionof these teetlhinNesophontes andalso have areduced
meta-cone,thecuspthat inNesophontesishighandV-shaped.
From the Erinaceidae Nesophontes is excluded by having tritubercular instead of quadritubercularuppermolars, bynormalorprimitive instead of specialized incisors and canines, by the reduced number ofpremolars, 3 in Nesophontes,4intheErinaceidae, and by the loss of thezygomaticareh.
Compared with the Tupaiidae, Nesophontes differs in the possession of molars with V-shaped metacone only, instead of both metacone and para-cone, whiletheincisors and canines of theTupaiidie aremuch more special-ized thaninNesophontes. In addition the latter lacksthe postorbital ring and greatly expanded braincaseof the Tupaiidae.
Detailed characters well separate Nesophontes from the Eocene insecti-vores (figured in the 'Carnivora and Insectivora of the Bridger Basin' by Dr.W.D.Matthew)and will be fully setforth in a later paper.
Nesophontes may notbe included with the Talpidaw because of the unin-flated condition of the auditory andbasicranial regions, so expandedin the
Talpidw,
the primitive form of the skull, as opposed to the flattened and swollen braincase in the Talpidw, and the far less specialized character of the incisors, canines, premolars and molars, which latter lack the long crowns, sharp V-shaped paracones andinwar(dly
projecting inetacones foundin the Talpidw.IForconveniiencetheZalambdodonitaistakentoincludethefamiliesCentetidae, Potamogalidae and Solenodontidw,the Necrolestidae and Chrysochloridae beingso verydifferentin structurethat theydonot enterinto consideration.
Bulletin American Museum of Natural History. [Vol. XXXV.
From the Soricidae, Nesophontes is sufficiently differentiated by the non-falcate form of the upperincisors, by the normal or primitive form of the lower incisors, by the normalformof the canines, by the unreduced premolar series, by the lack of the postero-internal spurs found on the shrew molar,and by thelarge size of the last upper molar. The general contour of the skull is muchless specialized than that of the shrew skull. However, the affinities of Nesophontes appear, on the whole, to connect the Nesophontid'T more closely to the Soricidaethan to other families.
Nesophontide fam. nov.
To summarize, Nesophontes represents a distinct family, apparently hithertounknown, characterized by the combination of thefollowing import-ant structures.
Nearly complete, primitive dentition in parallel, closed rows. I., 3-3;
1-1 3-3 3-3.
C.,
1_1; P., 33-3;
M., 3_3;
= 40. Canines in normalposition, normally
developed, two-rooted. Incisors smaller than canines with median dias-tema above. Upper molars triangular in cross-section, asymmetrically cusped, with V-shaped metacone but no paracone. Lower molars of extremely primitive tuberculo-sectorial type.
Zygomatic archincomplete, malar absent, zygomatic roots peg-like. Skullelongate, narrow, braincase but little expanded.
Basicranial regionmoderately elongate antero-posteriorly. Palate completely ossified.
Pterygoid fossee much reduced.
Mandibular condyleslightly oblique tothe transverse plane, the articu-lating surface directed downward.
Judging from associated forms Nesophontes must have lived at least up into the late Pleistocene asmingled with the bones of this insectivore were found the remains ofseveral genera of bats some of which areliving on the island today, the species even being identical.
BULLETINA. M. N.H. VOL.XXXV, PLATEXXIII.
_~~~~F
.4-~~~~~~F m.S-...7__
NESOPHONTES EDITHME sp. nov.
Figs. 1, 2 and 4,natural size. Fig. 3enlarged to about twice natural size,