NovAMERICANttatesMUSEUM PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY CENTRAL PARK WEST AT 79TH STREET NEW YORK, N.Y. 10024 U.S.A. NUMBER 2635 OCTOBER 12, 1977

RICHARD G. VAN GELDER Mamnrnalian Hybrids and Generic Limits -.I.I.III- :.IW i.IiLI., 1-, -,I,,.:...... ,.. :,7. ...I I.I,d.,..I:, i;;I. .,..I..wI .II.1II., .,. ;.".,- .,1, I: ;.. ...AI.I,,,,- ,.:!.AI,.,.1.I..II., ,-..I-,.,-,,,,-,, ;,I-,-..,..,I .II.I.IIIo.---,.II.,.III.1..-..-," .,.,-..I.I..rI,I..".,.,, ...... ,%I...1 .-.:.:.,, ;.. .II II.I....I.-.I.,I.IIII...... i.I.,:1 ',..'..., .... ,.,I-iI ,.;4, ,,a., .,: ...... ,.I...I.II.I%...-..I--.i..II.."I..,II.,,.,,.,... .I:,,I-.'f;::.,.,.III,,-;I.,...I ..;I .i.. .I....IIII ;.7,,-.. ...II.iII;.I.II...,.:,.I.,...II..,,. ,.. :.,..,wII.,.:,II.:I-.-.1-..II.I....'.I:II.III.I....i11i....I..II ..,.':,.-1-,-1...I-..II.I.III". ,-...I1, 1.I1.I11. 11..Ir-.I.i-l.....I-IIII...I-.I..-I.,w.;I-.,I...,IIII.-I-i,. .,iI...,,:1.. rI..III II,II..I.IiI..... ;,.II--II.;..-.,,,1.:.,:.,,.I. rI... ..I....I.I;.II:.,,,III:.i...IIII.: ..'--,...d..I..,I. .,.I...... I I..I-....I..I.IiII.II%, I4iI1. .II,I:...I.I...I...., ,. I.I. ..ii.I..IIIII., ,,..-II. .I.I .A; ,....1:..,I.:Iw.:I.-.I ,:I:,..I-I I-II II.:I. .I ...I.III.,. -I..-.I.Ii .. .I-I. I..r'IIII;1w..I.I-I-.I,.,.;.II-IIII....'...., e., .IIIIIII, :.IiII.I.II.I.-.....I..,..IIII..-.rIIII.-...I .....'::.II.,..,.:. .Ii.I ,6,..I:I ..IIII.1,..I-,,I...II.- .II.1I. I.,,-II. .1411,.I .Ii. ..,I:I.,-, .I.:..11I .I..1 .-,I.I.-I.. .I.... ;,.,..-I. I...I.I-. ,I.I. II:, .-..I:iI'-... 1711-II. ,I.,..,, .I. Ii ,I.I..I.,1,:;I.I..-1.I,..1I.--I1...I.:.-.,,..I.'- .I...... ",j.I II .I II.,.:,Iiz.-.1I-.,.1.-II...I....I.I...... ,. .IIcII.III.-I.,,,.o .:,,...t, ..-.I.. .. I..iI.- II.. .III..I.-,.,Ij-1.:,.f,,,...'...-.,.,1. .,III. t.., II .I-t.I.. .,.:.;..I-..Il,I.. I,...,.,, .....II..II.I,. ,I .I.. II,.I.,.I"..-..II%II..," " I.IIII.I...I.I...... II:,II I.I.I .;I. II- ..-.-II. ", ....I.1I.IL,II- .I. II..I..I1...... I.,.,. ::r,..-.I..I,.:;,., ....,..;.II.II ,.III...... ,..I.-..I...,,,II.-.I:.I..-,.-,: .-,,,i,.I-I..,-I.II,.,,,I,...:I.T .I ".1.-.I.'L ,I.II,.I. .I.,".IIII-.,.-. .II.IhF...iI..4,'.,.1--....,.I..I...II,,I-.,m ,.. II.I.Ix-, .-II,I., ,.". .I.. iI;I ,I .I.Ii..III.II,,I.11II I. .,...z.I,... I-.I..,;...I...III..I..%.Ial.'".,,I..I,.,.,-,I -p. .,.I...... z..IIIA:,.,,.I.IIII,..,,;.,I,.I..II..,;I.,, ,I.-....141111 1...4..:..I.."..III-,.-,.-.....III..-.I...-., .II -II.I.I.I.I,I.,.Ii.IIII:.,...... I.11.11..IIIIL..I-..: .,,.,.I-I. ...lI. .,"..III.14 1.., I,,I.l..I.-.., .-..II .L .-.I.II;I,..,,:I.I.II II,,..,....,II,. I. ..I..I .;.I-.I I... 1.I IIII.:..3- ..I.. ..I.I.II...... ,.....I", ...I. .,...I...... III III.j.. II'. iI.IIIw.II.:,I-...... ,.. .I,.I-II.r:, -,.. -.i ., II.1.11,..., :III ,I I:,..I; ..II. ....II.,. III.IIII,,..,..i..,.::1..-.I..- ..I.".. 1,n.,,,.,.,.I. I.aI..I.I.II.II".-;I.'.I--...I:.I..I1.I.,.,..,,.,...,I..-...I liIII. .I...III.II.I..-.r1.:.... -.I.I.I.,.4iIL..IIII.I1. 17. .I;.-...,I,....I.7.. -.III.I ....IIIII.II.II..;.II"-.II..I..,:.,..-, I. ..I-I-j..-IIII.I1,I..I.II,:..:,a,- '.,L.0,..,.,...:. ..I,.I..I. I.I-...1.,.cIII.III.I ;I.,.. '.:;-%.1-...:I.: ,-I .--.."."I IIII..IIj.I..,.I.I II-,I,,:..II,.. ;p.I.,.,.r,.,....II.,.II.I.I-I1, ..II I,.,I....,..).... .II.I..,"III..4.I..oI,.-11III.v,-.II;I...II....::..,I.1,".;IzI. ,.""..IIII .I-."..I.,,,;. .11 I,- ..:...7. ;I.,.. ,.,I III..,,II.;..,I.I,II-IIi,.I.II.1...... ,,I.",..I...I1. ,:. '.I. ..1.. Ir,--I --.....I.III. ,I,,. ..I., I.I...I, I.I.II II.-.III-.-.I -.,.. II III.I...i.,,-.,"II..,..III.;.II.I.I11II.II.,.i ;.'i-I.. -..,-I...... IIIII.II..I.,....I.....--,."I:,,;..,:;:::..I,..-.--- -...IliI,..II.I I.I. I:-II...I.II.- I.I.I;, :.;I.I..,...,.;..I,.,."II,,-. ,.I -III. I..I.I.;.iLI-,;I ,I.11I., ;..-,II..I..I.I,,I.I11.I.I IIII.I .iI.I,.II.7,.'.I.I...:4., .I..I..I.,.II.III...I. I.i .:.-...i I.,. .,.I....II,,.. .,.,II..,...I.II...... e..II I...I .. "I...II.;I,.-..I.-I ..,...r.I.4. I1. ..I..,.-L.I I9I...II.,II-II,II;...,,I..-I..-i:I..I,."II...I....,.....;...I....,.,II,! ,."..."I1. -IcI.II.I...-..I,.IIII IIIIII. ,I17 ..I. k.II, .-I-1,.I,,, ,,,,,:,"I;,I ..I I.- .I.....,I:1..I.I-: ,.,I.2,,,.I..I.II,.I..,II,I- 1:.-II..I.I..I-7 .IIII. ..rIIIII i,...: " ,. ..m,I.--..I,..; :,,I.I I.l ..11..,II".;I..,1.,,...%.I .,.! ....I..I:..-III.I- ..II-IIII..,I:-qII.-..1.%;:.."II1.,-..-I....I..-.. ,I-, II.." ..i :IiI,II -,,I,I.I,I....I.II,.,j".d.;Ii.....II..1,-I...,Ir,I....I .I.rIIIII.I.II..I.II,.I,.II,..,..;I..-IIII.-...,I."II%.II.. -.,I ...1.IIII"II III.I.I.,..I:;.-I ,.II-. ,..rI.I.-I%.'...II...II.I11..." ;l,." %-,II.I.I. .I. I....-.I.-.. :-...,. I.IIII..,..",....1,;II:I.I I..II.I,.F:...., .II.I I.I.:I,I I.... .II..,II..,.I.1,.."...I..-;..IIII-II; .,-.I ,II.,II,...I..::..I".I..1..IIIII-..,I1. 11..,,-..,.,...I.-I.III..I% ...I.o...'-.II.I-I -1 .;:.., ....-..x,'"'':I.IiI.....e.' .I..I1..,..I:r .;,..I-III,I.II II.....1.I..;I.i.:;...L .. -.1I.II,...1::-;.-II .I..II,I..I.:.I.IIII...,,I.:II.,,Z.,-II..I..I11.I,II,...-!,.:, ,.IIII.-I IIiI. I;II.".,.1..:. .I.. .,I-IIIII,...... --..II I.III.III"III..II,--11..:1IIIIII.I I..: ,.;...... I.q.;,.. 1..iIIII',."11. .II..I.I-.II..I...-,L'...Ij...,.I....,.--,.-.,i,...II.I-.;., ..III .II...I.,I. i.. "II.1.,.I ...;...1:, -I., ;-,II.611 1.I..I.I...I...I."...-I,.IIII.4-I-...,I ...,. ..I ': .I, I.I;II-,;.I..-.III. I...... I;I,-1.II.11I..,..I..,.III1.,II:;I.nI,II.I..-7,I..I I..,.,III." -IIII.,...r-I. -.:1, ..,.;,.I.,I....II11I. .-. I...I.-.I.I..I..,.I.11I...II....II. ..,;..-...,t,. e.II II.III.I,,III....,.,.., II,I;.I.1..II.:III..I.,.I1.7.,,I,I.I.I..IIIIIiIII...1....,.. ,..I I.IIII...I:,-..I..I., .II.z.., 1.I..I,I,I....-I..I .I.,. 1.III.,i::.;l.."I Z.,II"... .1.I"I IIII.II.... iI. 11,.-...II.I.j4....,I."I..I-..,I..- .....:.-.,;I11II.-II,.,,,.,,..I-.I.II.II,.i..-II,!..I I.II11I...I':,".,II...I.,,.I,I. .I.-..;", ,.7 ,II.I ,.II.-,...I".I-.-.:, Ij.Ij,II.,w.:.,.:I.IIII.,:,..IIn.,.,..-.I-I- ,I..-..I.-.I.I..I I.I.....I.;-ILII-II I.-oII.:.....-.,,."I....,,.,.- ,.., .I..,-III IIII...".:,:i1.-I... .,, -.-1.,, ".II., ..:., II;IIII.IJi.-11I.,,.,..I.I:...1,I4..II,..:I.:I..,-1",....,..I:.-.-.r. I.-I-I.:,.iI,.I.,.I.,..I,.-.I..,1..I-..--I. ..7,... I.,"IIIIII..!I.IIvII,..,I..I.III..I,.. .. -I:.I,.. -,.1, II..IIII.I .t.I,-. ,II...i- ..I.I.i,. ,.I .II.I.,:.... ,..I.: ;zII., ,;I.- I.IIIIIIm.11 ,-.III-I,i ,.",. .;!..: .I.I..I-....C:I-I.IL,t,II-I.,o.,:; ....-I....I- -.I.. ,I,-,I ..I,I-II.41.I1I:,... .III:11Io.:.I,-.Ir....,I...._;I.I...,,",,..I..III.II...,Ip-.I:.IIII...,:". ....I 3..III...k,III,,.,-III III...I,..1... ,.,I. '.II ...I .-I...!-I,:., ,I.III,..%,. .II1,IwI.II.. .141-.,..I- ,.. III.. ..LI.I.,.I .I.I.,..k..I.....,I.,I.--:-..-III.4, .II 7....I...II .; ;.I.I.-III.; :.I...,..I...... I..II .11.III III-.II-:,.I. .I.I.IL.I"I.-.III.. ..III.. II,... .I.iI::...i..... II...... I.,..I.,".:-I.III-I-....-..",.II,:.-w..-I.,"',iI..,1;I .-.j.- i. ,i...I.-I.I I..I..I.III.i.-1.".I...,ll,,I.II-.II.;,.1;..II.,..'.. ..I-.IIIiI1,,.4..I..I.,.....; .,,-I....1; I.II.I.II ,-..,.. .I.I1.I.I.,....-I,I:..:I..II...I.,.4.,,:L.II:.I..I...I..,.I:..,.II-.I. I..III11I.:IIII,,I.,:p.IIIII?.I. ..;...... "-11m ...:I,,4I.L.1.1I-.;.I.I.II,I.II.I..I.IIII.,: ,; ,...II.I,IIII IIII ..1.II.III.II--II...I"--".-.I.4..-I...I I. I-.I .c..i.L.I...:..I.I..I.IwI-..I..,.I.I,.,...,11I..;.I.:.I.a.. -III,.I.,mI... .11-- .:II .I.r:. .I-..I.I...-.. ,11 .. .z.I..- ,...i I....II....III.i--.I,."..I.IIIIiI7 ,...I.II.I; .ILII.,::. .:I.I.:.,.'tII1. :.I. I-I...I,..1-.IIi,I.I.I.,.,,I,.I.11I.."I..II.'....I IIL.-1I.,.I.III..II.II.II.I.II..II.....,.:2 .1 II.II..I.;1.iII.II.1,.,Z.IIIII., ..,.'.I:;., .,... -.I I.i..III..-I. .IIIIIII..II....-I ..I.,:,i.I-.II.. ..:,.I,.I.II.II.II.-,...:II,,.:.II-,.II I.II-.7,--,.,.II.IIII....,...-:,..,I....I"-.L,I.'..., II..I1:I.I-..,IIIII.1..:.I.:.I".I,..-.I...."'.-..,i,,.,"II.I-I.I.:,II" .I..r,..I.I.....iIjI...I11....I.1. ..I,.,...I ,,-..:,...;:.,"..I,:.,;..-,,I..-.-I..I ...II1.II.I-...-,-.I..,....I -III.I....,,:. ,.,"I..I. I.III..i, .,..II I.I, .l ..II...I.II,II: ..I.1.I14 .I.,-.,I,II.I.I i.II-I.IIiI.I... .I.!.,.I. ..I.. ..:.I-II..,...I I..II.IpII:,. ,,.-I-...III,,-. .II:1:iI,Im...I,II-,,..II1 ..1.I.-.I .I.i-I.,I.II:.. ."nI: ,II-"...I.--",,.I;,I.1.I I..%.IIF,I,....,',..:I`-.I.:....:7..-II.I.Ir...... I. I:"..-. .'-.-I .I;II...I...., ,-...I I.I.,,. ..I.1..I-I.,,,...2..,11..I.,..:.,!.:..I..jI.,. :,I. -I.I.II.;.", ...:II.,....I. -.III.,....,,,...-..I,. .I,I-I.. .%LII ...I.LII", I....I.I.,j,.....,.;.,,,,..,.,,;I...,..1,,. III...... ,zI,....I..-II.1,I.1....7,...I.;I.." .,..II- II-.v...... I".,,.. ::I.IIII ,i.,II.-.. ...,.:..:,,,t,liL. ..-, .,.,....II,.II ...... ,,.II.I,,.-.-.1 ,Z:.,....4 ,.-,III"."II...... I,..I.,..I..-.I.. ,..I. .IIII.I.,.II-I.,.I...,;mI..I.I .....,I...-..Ii:.I.:.-II..-. .I....I.;...I..I.,..I...;III-,...I;...III.I:I -,I..:1.I. ,.II,!,.: j. ,.....,III.;:. .I.;, ..r..I.. ...II.,...,I..,,.,,.I....III;t,:II ...II...,.-.,.i,I,I.,.."I. :.I -1III.II .I.I.I.I..-..jkII-..1,,II...;,....II.f,I ..I.IIII.IZ,II.....;.,;;;I,,I-,..14,.1,.,,,I.;.:)..--,II.:, ..-,... III.I.I.-...I7. .I.i .,I..I..I,..1.I:...-. ...%I..... p, ,-I -? I.,..w,:..I ,.:. ...II. ..I..I.. ., ,.r.I.:1..r....I:...a....:-. .1,II:iI.,.-,I. ..-.I...1,..,',II....I11.I::.I.-..,.,IIII,I:, L'.,.. ,-..I.,IrI.I-;,;...,.:II.I..II.III,". I!II.11-...... II1II.I .-.I-.I'.. .II ,III.III4,I..I I, 1'...I.. .,..,-I.I.I,i.I.I ...I.I.II..I III.II11I..I...I..iI,IIiI':I.11,1...;.,.,-..".,I.:-l'.II.I--I. ,....I.I... .I..,I....,.-..I.,I-:--:,.,i..I-..I.11.I:.-.I. I...,...... :. ,i i,,,, "..,-.. ,,-.,II,.I.. .-:t;.I...t-.II.iIIt,..,.I..:III,,.,.!II.I..i.I,I.-.iI,,..11,....,I II-....,,-I.-..I,...:i..II..IIIIII..i.:..;.I-..-',. :. ,. .4 .,4.,., '..II6... .I..III.II.-:-.I.,.:IIII.,.I-..I-,-;I--.:,IIe....:.I...I;.I.,.mI...,,I...,I,LII...... i,::., I.IIli%I.,.i.".II.II..IIITIII,.-.....I...,I%I:.II..`.-;.!.,.III-I.. I.I.III..IIII,.II..:4 .,.-1... .1II1I.I .I. ..i..... II..I..,I...II,I.oI .,I.,;II.-I...%,. I.,!,.,11.1.4 i-;.I .1. ,I.-...... I.:..I--II.,.';.IpI.I::....I.;.` ..,..,I.I.I-;,L',:,1.:.rI.I ;"; ,. .:. ,I..'.IIII ....IIII. ;II ..-I,.i1..I-. .I..III..I--1.,.IIII.-IIIIII.I--..,...I.,..:.I.11.I.;..I..;IIIII. I.i-I.-I I.....II.. ,I. .., ..,, .-.IIIII.I. -I..I L'.IIII I,. ,-., -..... I.i ;t.1 I.:, 'I' ,..I.-F.II;11-II.I....,..1....I -II....;.:.,..I..,I...,:..II.-.Im.,-..%....,,'. -1...II11 I..-.,I-,.I.. --.,:;. ,.,III." 1,..I...I:" III:.. .I:1.;:,.I ,,.,.,.. ,.-.II-;.':-p.i...I.I I.-.1...,iI.II..:...,II. ,I.I..IIIII II.I-.II.....,"....:-1.;,.. ,I... ,,;.;.;.I,.- .I.I,.II-..1.i;-I.I. :--,.1...-II...!.,12;:..-,IIr.11I.:.I,I-I,:.IdI..;.,I.I II..II,;;,,.,..,,,..-I"IIII.;...,,.IIIIIi..1..II :,. I,-,.I.;. -..I,.1,I...... i:,..-I.,.I..1,It....I a..:,....:..II". II...;I..; I-I;t...lq .It.;II...7,II-...IL..I.III.II..I..I- ,..1..:,II....::...-.I..:" r:.I .,..II.I1.-.., .,I. .IIIII ,.I.I...I.I. -II,.I..1II- 1:;..': .. ,II I.II II;III.I,,.-.-IIIII..1..-..i.;.,.1-...... --:. I.I.,mI-"..11.I...,:,..,:I:. II;-I.,.II...I..I,II-.:,.1.-II....IIz,; .-.,.-,.,,. -..-.. -.II.III..j-I.-.."..'II.,.,.I.II.':.,.IIII. .";I III-,,,.,I,.., .II.I I-.IL..II.Ie.I.,....l..iI,I. . ..., ,.:..I.I.I .II .I.I.;I..I,., .L'..III".,.,,:1..,,--., I-..I..II ....IIIII;II....iI..,.II.I.I.II:. .IIIII;..1. ..I!...I.,I..., :,.-1 ,II.:.I.,.III..IIi.,-IIII II,...-1-7I.1,I..I .:..III,....,II I::..r.I.k.I...... 5..I-I..I:1,.1.II.",....I;.I,.i.-I.IIi. III.. .-!,,,,-.I ,,I. .I;...iI11I....-,IIIL,,,I....-IIII%.,.-....II..I.I,.,.cII.:1I-I1, I.Y'-. II..I:;,.;;,,I, 1: I,.,.-I...I.,.Ii ..Ii,.-I:.F.I..I....I ,.;I..I.-,.I- I.II.".,.II..II;I.,I..,:,.I.-IiI.I.v .,..i.1 I .;.II.,;,I1..I..:...,.-, .. ,,"., 11"..I.. ,.II.In.II.,-.I..I14."..I..,..II...,.I.II.I .I..I,I..I:liiI---.-II...-I..,;.I.11.I", ,,,,I.;..z,I,ll.,-,11.:,4.tII-,,.',.. ..-.....I. 12I..,,II:.,:,:....%II ..I1. :1;II ,.' ,,,.!,II ,.I"III.1. ..:..I.I.II,.II...I.I.I.II-1I.11II.%III.. II..I-,;I.I,:.I...II.:,.,.II."I.e",I..IIIII.7-I,.III..I..I.I. ..I... .,I-.1.,I.-., .IIII..;.I,;,..I..,,II.4.,.!.Ii I...... ,I,,....,,, -;,.i" .IIII,:..11I-I..III,..I .II.I...'I,,. ,,. 1.,.-I-. AII..I.. IIIi II.-II..III.I71. I-...I. ..71. .-. .,. ,.IIII -., ,-.It1. .I". .IIII ... III.III,II.11.4,,.II...II..I,.!,..I.,,...-;I,-7..I:.II-II.II... ,I, ...-...,IIII..,I,:- M.II.. I..II:.I I..,r .-I.:.-, ..I..III II..I. ,iI,. 11,I- I...,.I.LIII.I-..-, i.- .,,,- ,III- ;,I -I..I-..:I.-II.,,.,.i.,.I.I-....,.I, ,,. .-,-I...I.I.II.....,., I.. .I.:II.I .,N;r11 ....:It,II .IIIII11 r,.I.: ..,.I.IIIII.III....III,I... .. ;.. ..I.I.n-..-....II.'i.-.,I-,4IIIq,...I. .11 .IIr.-: I.: ..:, 1.... :,,i,I..I-,.,I.III.II.II....,,";..,II.,.,.-I:. .Ir.)- I-.;-.'.I".I.I;7.zI.IIII.-I. ...I.II.%.IIIII..,..I I,, 1....11 .IIIII.II I.I.I ,.I..r.x,:i.I.I..I. .".",.1,.I.. .I-I-;,.I..ITI.I.II.-.I.I,,. .I...., -IIII..,....I.-.I I:IIqI.;I.....I.,.I:.iII.I.L'1,..I..z:. ..i ,...... --I...I.. II:. ,III....I,I.I, I...II,,IIII.1,.-II.II.::m.. ,;,.I.1.IIIII. .I.II17...I..I..I,.I .:..II.., .I.,I,.,..14 ---,...I.IIi:.Ib,, ,I.II.III..I .IIIII.I.III.-I--,,:I.:.1.I:11-I...I.1.IIII.II.I...I -'iII.II .I.-.II.I.."I.-I.,-..IIIII:11...- ,,I. ,..,.-I"-1 ,.I :-. ...III...,...... ,.,--.,..II.:III..II...1.I I.II.I,,I.I-III.II1.I.IIII..1:I",I.I.III, .II.,-.I.! ..11 ,.. .I.---. I-I.,.I)II ;III.I.I, ,,II.III ...III-I-I I.. I,., ,..:,..11.;.I...I.,IIII.I ILI.II.-...-.:..II; I..,.II....-,,. '...II.1-.,.-,II,II.I-..,w .:.,.I. -III.; III".,1.I.I..IIII.:.,-, .,-..:..II .I,,,....,-.,-- ,. ,...,. I"I.... .I..,.;I ..71.1..:-.c-I1z' :..- .. III"..-I-, ...t. .I.II,I. II...I.,,,,:.:....L.I-1 ,.mI..,I...I-II--I.II...II.,.I;I.I...LI. ;..II,%I...,e.,.I.I-..I:....:II.llt.I. I1,.;.-II I.... ,.7.1 I.I.I,..21..I.....I-.;...... I,..--I.....- .] ...,,.IIII..II., .,.II.,.I...I. .;C.,I.,11I.,.I...,III1.I. ..IIIzI....I.I..I.I...,.,,.III..II.,.-.w.IItI. .,,...,,. 1,.I,:,II,, I:.I:...i.III...... ;I.:II...III.,. ,7...I.I, ..'...-I.I.I.,.I, ....,.-:: ..LI ,I.I.I..,L,I:7.-III -:III.IIII...-IIII ...I,.I.I..I..I...I.I.-:-1I..II ..".I....:.I..I.I.-:-I.F..-I .I .;,,i, :. I,, I-,..III... .,.. ..I..,.,..I. I.I.I -.;I.III:.:I1. ..,.., .,,. .I... .II1.,I1 ,.i-.. -.I.I,.I,;"..:-..;.....,.,z. -Ie.....:..,....I.I.,III.,.., 141111.-1.-.,IIII:...... ,:II AMERICAN MUSEUM Novitates PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY CENTRAL PARK WEST AT 79TH STREET, NEW YORK, N.Y. 10024 Number 2635, pp. 1-25, table 1 October 12, 1977

Mammalian Hybrids and Generic Limits

RICHARD G. VAN GELDER'

ABSTRACT A taxonomic level, the genus, is analyzed (Papio, Macaca, Cercocebus, Mandrillus, Cynopi- from the perspective of reproduction isolation. It thecus, Theropithecus, Erythrocebus, Allenopi- is concluded that if species are, by definition, thecus, Comopithecus); Grampus (Steno, Tur- reproductively isolated, then species in one genus siops); Canis (Alopex, Urocyon, Fennecus, should not be capable of breeding with species in Vulpes, Cerdocyon, Dusicyon); (Thalarc- other genera. A survey of reported mammalian tos, Selenarctos); Melursus (Helarctos): Felis intergeneric hybrids and the application of repro- (); Arctocephalus (Zalophus); Tayassu ductive criteria to the definition of the category (Dicotyles); Cervus (Dama, Axis, Elaphurus); Bos of genus led to the recommendation of a number (Bibos, Bison); Tragelaphus (Taurotragus, Bo- of taxonomic changes in the content of genera. ocercus); Cephalophus (Sylvicapra); Alcelaphus These include the following genera, with syno- (Damaliscus, Beatragus); Connochaetes (Gorgon); nyms in parentheses: Macropus (Thylogale, Kobus (Adenota, Onotragus); Capra (Ovis, Am- Osphranter, Megaleia, Wallabia); Cercopithecus motragus). INTRODUCTION The genus is a taxonomic level higher than genetic concept of species, taxa of this category that of a species. It is generally stated that the were defined mainly on morphological grounds, boundaries of the genus are arbitrary because the essentially: a species was composed of individuals criteria for them are wholly subjective. In this that resembled one another more than they re- way the genus differs from the species, which is sembled any other individuals or groups (or pop- presumed to have an objective basis for its con- ulations). The amount of difference required to struction. This conceptual objectivity is founded distinguish one species from another was not es- on the biological and evolutionary idea that, in tablished in this definition, and the determina- order to maintain itself in nature, a species must tion of species boundaries remained an arbitrary be reproductively isolated from its close relatives. construct of the taxonomist. Essentially the same Prior to the development of this biological or definition was applied to the genus and to other 'Curator, Department of Mammalogy, the American Museum of Natural History.

Copyright © The American Museum of Natural History 1977 ISSN 0003-0082 / Price $1.90 2 AMERICAN MUSEUM NOVITATES NO. 2635 higher categories, each using the next lower cate- "In entomology there is sentiment in some quar- gory in its definition. ters for setting an upper limit (perhaps 40) to the The biological concept of the species (Simp- number of species allowed in a single genus" son, 1961, pointed out that the other concepts (Ross, 1975), or the admonition that the mor- are also biological) grew from the morphological phological gap between genera should be in in- one. The morphological similarities that charac- verse ratio to the size of the group (Mayr, Lins- terized species were genetically determined and ley, and Usinger, 1953), size here being taken as maintained by gene flow within the species. Re- the number of species. Despite its limited or non- productive isolation prevented dilution or altera- existent objectivity, the idea of groups of species tion of the species morph. The biological species arranged in genera is a useful one, so much so concept, based on reproduction, has been at- that Simpson (1945) said "the genus is, on the tacked because it is inapplicable in some in- whole, the most definite and permanent unit of stances (e.g., paleontology) or still permits arbi- modern classification, to such an extent that the trariness in the case of forms that are geographi- genus may be considered the basic unit of practi- cally isolated, such as insular ones, and therefore cal and morphological taxonomy, although the are reproductively isolated. In practice, however, species is the basic unit of theoretical and geneti- the biological concept is used both by paleon- cal taxonomy." tologists and neontologists who deal with geo- A genus is, by definition, composed of one graphically or chronologically isolated popula- or more species, and it therefore follows that for tions. On the basis of the morphological distinc- a genus to exist, its species must exist. In part tions between the populations, or actually this portion of the definition derives its basis between samples representing these populations, from nomenclature, in that a species name must the taxonomist estimates whether or not, in his include its generic name. This historical artifact, opinion, the differences are so great that inter- however, has no particular bearing on the con- breeding would be impossible if the two came cept of the genus composed of species. For a into contact. For the majority of species of species to form, a high degree of reproductive , knowledge of reproductive habits is isolation must develop, either through some kind lacking, even within a population. There are of isolation that permits a genetic change that some taxonomists who, because of this, would results in incompatibility, or through a genetic deny the biological concept of the species and change in a number of individuals within the return to the purely morphological concept for population that permits them to interbreed species (see, for example, Blackwelder, 1967). among themselves and not with others. Sibling It is not the purpose of the present paper to species, which could be the result of the latter explore the concept of the species, which has a mechanism, exist, but it is argued that some literature of its own. The idea of the species as a other form of isolation-ecological, behavioral, distinct unit that is maintained by its own supply chronological, or physical (geographic)-must of genes, and which is not influenced by the have preceded the genetic differentiation. Mor- genes of other taxa flowing into it, is one that I phological differences are scant in sibling species. accept as fundamental to the understanding of An essential part of the species definition is evolution and to the evolution of organisms. I that if the populations involved are different think that most mammalian taxonomists accept species and sympatric or contiguous, they must the concept of a reproductively isolated species, maintain their genetic integrity in nature. If the even if their material does not permit the utiliza- populations are separated by geographic or eco- tion of reproductive data for determination of logic barriers, their. potential for successful inter- species. breeding becomes the criterion for determining The genus, however, is thought to be an arbi- their status as species. "Species are groups of ac- trary division, a purely construction, tually or potentially interbreeding natural popu- which may have no real basis in nature. The idea lations, which are reproductively isolated from that the genus is constructed for human conven- other such groups" (Mayr, 1942). Thus, repro- ience is clearly evident from such statements as, ductive incompatibility in nature is practical as 1977 VAN GELDER: HYBRIDS 3 an objective test only for sympatric and syn- greater than between contiguous populations chronous species, or allopatric species in a zone (subspecies) he may well regard the allopatric of contact. For wholly allopatric populations, forms as separate species. Should the mam- however, laboratory testing is possible and ac- malogist, in a case of this sort, have the oppor- ceptable to establish the potentiality of inter- tunity to cause members of the two taxa to be breeding. For allochronous populations, testing is brought together and if they did interbreed and generally not possible. produce viable offspring, he would then have the Hybrids between species of free-living mam- opportunity to determine their potentiality for mals are rare. In part this may be due to the reproduction in drawing his taxonomic conclu- difficulty of finding and identifying hybrids, but sion. In general, should the offspring be sterile or also probably is a result of actual rarity of occur- show reduced fertility, probably two species rence, or reduced viability of the hybrids. More would be recognized. If the offspring were fer- often when two presumptive species are found in tile, the two allopatric populations probably contact and there is interbreeding, it is deter- would be recognized as only subspecifically dis- mined that there is fairly continuous gene flow tinct. In this example, if course, the potential for and lack of sterility in the hybrids. The two reproduction has been determined under artifi- parental "species" are then considered to be only cial, not natural conditions, between populations subspecies, rather than species, and the "" that are not in contact in nature and therefore do zone to represent an area of intergradation be- not have the opportunity to interbreed. Were the tween two subspecies. same artificial experiments to be performed with Often, when two closely related species come two species that occur together in nature, and into contact, there tends to be a reinforcement which are not known to interbreed, the produc- of the characters that distinguish each. This tion of viable, fertile hybrids under artificial con- "character displacement" may be more common ditions would not deny the specific integrity of than hybridization. When hybridization does both species in the wild, and would tend to con- occur between two species that are sympatric or firm it. Despite genetic compatibility, the species contiguous in distribution, and if they are valid maintain their gene pools by not interbreeding in species, the hybrids are expected to be sterile, or nature. Since genetic, morphological, physiologi- at least to have greatly reduced fertility. Thus, cal, or physical barriers to potential reproduction while there may be a flow of genetic material would have been ruled out by the artificial hy- from both parental species into the hybrid popu- bridization, other mechanisms such as behavior lation, there is a general lack of flow from one or timing would seem to be the ones that keep species to the other, conforming with the species these species from diluting each other's genes in definition. natural situations. What happens in nature is the fundamental Thus, at the species level there may be pairs of criterion for determining the species when two species that cannot interbreed as well as pairs nominal taxa are in contact and have the physical that can, but do not, interbreed. The latter are opportunity to interbreed. In the cases of popu- those that have been shown to be genetically lations that are geographically, ecologically, or compatible under artificial conditions, but otherwise physically isolated, and for which which, nevertheless, are reproductively separated there are no reproductive data even from non- in nature. The usual morphological criteria would natural conditions, the taxonomist usually judges seem to play no role in these cases. Sibling the relationship of the two taxa on the basis of species, almost indistinguishable morphologi- the amount and kind of morphological difference cally, are reproductively incompatible. Species evident between the two. If the differences are that are seemingly quite distinct, such as equal to, or less than, those to be found between (Leo leo) and (Leo pardus), are capable contiguous subspecies of the taxon, he will gener- of producing viable offspring in captivity, al- ally consider the isolates to be potentially capa- though they are not known to, or likely to in the ble of interbreeding and regard them as sub- wild, where in many places they are sympatric. species. If, however, the amount of difference is Evolutionary proximity is implied by inter- 4 AMERICAN MUSEUM NOVITATES NO. 2635 specific hybrids. Presumably two populations nition is the same as that of a species. Surely we that become geographically separated could go should question our criteria for genera when we their own evolutionary ways and yet maintain have rather similar (morphologically) species that the potential for interbreeding for a long time, or cannot interbreed and highly distinctive (mor- until such time as their morphological or physio- phologically) entities that we call genera but logical differences became so great that the which are still evolutionarily so proximate that physical production of a viable hybrid was im- they are potentially and actually capable of inter- possible. They could also, once separated, diverge breeding. so rapidly morphologically or physiologically, To a large extent our concepts of genera are that reproductive incompatibility could be at- based on tradition and comfortable familiarity, tained in a relatively short period. The element rather than science. The retention of the genus of time is often an unspoken thought in consider- Bison, for example, is largely traditional. I have ation of taxa, but is not necessarily relevant in spoken with many mammalogists about the fact these considerations. Sibling species could con- that not only are the two "species" B. bison and ceivably maintain their identity for the same B. bonasus interfertile in both sexes, but mem- period of time during which allopatric popu- bers of the "genus" Bison can also hybridize and lations could show remarkable divergence that produce viable offspring with various members of would lead to their recognition as distinct genera the "genus" Bos, including the gayal (B. fron- with sufficient physiological or morphological talis), the (B. grunniens), and western deviation to make hybridization impossible. (B. taurus). All agreed that two species of bison However, the question that is the subject of this probably should not be recognized, but only paper is the situation when morphological diver- Haltenorth (1963) followed by Koopman (1967) gence has taken place to the point where the two seem to have adopted this. Virtually all mam- populations are considered different genera and malogists agree verbally that "Bison" and "Bos" hybridization can still occur. are congeneric. Tradition weighs so heavily, how- Are intergeneric hybrids possible? Certainly in ever, that I cannot find "Bos bison" regularly the literature of mammalogy (Gray, 1972) there used anywhere in the recent literature and even I, are a number of cases of intergeneric hybrids. although I had thought of them as being con- The question here is whether the taxonomic posi- generic long ago, am guilty of using Bison, as a tion of the species involved in these instances of lapsus mentali (Van Gelder, 1969). hybridization should be reconsidered. If popula- Few mammalogists have dealt with the ques- tions that actually or potentially do not inter- tion of the taxonomic rank of intergeneric hy- breed are considered different species, does brids. Simpson (1961) commented "Even with- it not follow that genera, which are composed out direct knowledge of genetic constitution, of species, must neither actually nor potentially successful hybridization is practically conclusive interbreed? The duality of philosophy con- proof of chromosomal equivalence and second- cerning the species and the higher categories has arily of the homologous nature of the structural been pointed out (Simpson, 1961): that the similaries." He was discussing, however, homolo- species is "real" while the genus is a human arti- gies between species and subspecies, but went on fact. Nevertheless, the genus and all the other in a footnote: "Supposedly intergeneric hybridi- higher categories have within them the funda- zation, usually with sterile offspring, is possible mental evolutionary reality of the species. Thus, among , for instance in mammals, the the upper limit for the species (reproductive in- artificial crosses, Bos X Bison, X Asinus, compatibility) should also be contained in the and Ursus X Thalarctos. In my opinion, however, definition of the genus, and at least represent its this might better be taken as basis for uniting the lower limit. This would imply that intergeneric nominal genera. I would not now give generic hybrids should not be possible by so defining the rank to Bison, Asinus, or Thalarctos." In his genus. The arbitrariness of the definition of the 1945 classification, Simpson gave generic status genus exists in its width and upper reaches, not to Bison and Thalarctos, and listed Asinus as a at its interface with the species, where its defi- probable subgenus of Equus. 1 977 VAN GELDER: HYBRIDS 5

Buettner-Janusch (1966) utilized the informa- ACKNOWLEDGMENTS tion on hybrids to suggest combining the pri- Mc- mate genera Papio and Macaca, both in a I am especially indebted to Drs. Malcolm broad sense. "If which are presumed to Kenna, Ian Tattersall, John Buettner-Janusch, be members of separate genera mate and produce Karl F. Koopman, Guy G. Musser, Ms. Marie sterile offspring, they should be Lawrence, and Dr. Sydney Anderson for their living though comments and criticisms, and to Mr. Henry Ricci considered members of the same genus." for providing original data on a Canis X Vulpes I have previously raised the question of ge- interpretation neric status between the antelopes Taurotragus cross. Responsibility for errors of and Tragelaphus (Van Gelder, 1977), combin- and for the conclusions is my own. ing these two as well as Boocercus in the genus Tragelaphus on the basis of hybridization: INTERGENERIC HYBRIDS "it would seem that if two taxa have achieved Gray (1972) listed no records of intergeneric the level worthy of generic separation, genetic crosses in the orders (as listed by Simpson, 1945) incompatibility should have occurred, and that Monotremata, Insectivora, Dermoptera, Chirop- intergeneric hybrids should, by definition, be tera, Edentata, Pholidota, Tubulidentata, Probos- impossible." cidea, Hyracoidea, or Sirenia.

MATERIALS AND METHODS MARSUPIALIA Annie Gray's most useful compilations of The intergeneric crosses within Marsupialia are sources on mammalian hybrids (1954, 1966, known only from the family Macropodidae. 1972) have served to provide most of my data on Ride (1970) recognized both Macropus and intergeneric hybrids. In utilizing this work, I Megaleia as genera for kangaroos, as did Van have, wherever possible, gone to original sources Deusen and Jones (1967). Simpson (1945) con- cited by Gray to attempt to determine for myself sidered Megaleia a synonym ofMacropus. Crosses the validity of the claimed hybridization. Unless in captivity between Macropus fuliginosus and otherwise mentioned, Gray (1972) is the initial Megaleia rufa produced hybrids that were sterile reference for the source of hybridization data (Gray, 1972). Hybrids between Macropus gi- and original references. The birth of a living off- ganteus and Megaleia rufa have also been pro- spring was taken as the criterion for successful duced in captivity. hybridization. Where there are doubts concerning Gray (1972) reported, under the name Wal- the data, suitable comment follows. The genera labia elegans Lambert, a cross with Macropus mentioned here are those used by Gray (1972). giganteus with hybrids produced in captivity. In making recommendations for the allo- Ride (1970) listed the former species, Wallabia cations of genera, I have selected the earliest elegans, whose hybridization with Macropus was valid generic name. In concluding that two or reported by Calaby and Poole (1971) and Rich- more genera should be merged, I have not re- ardson, Czuppon, and Sharman (1971), as Macro- quired that more than one species of one genus pus parryi and did not place this species in the be capable of hybridizing with more than one of genus Wallabia, which he reserved only for W. the other. An alternative to this, of course, bicolor. would be to transfer only the species that are Simpson (1945) recognized Protemnodon known to hybridize, leaving the residue in its Owen, 1873, as the generic name for the large original genus. By presenting data on the actual wallabies, listing Wallabia auct. as a synonym. species involved in intergeneric crosses, I have Van Deusen and Jones (1967) recognized Wal- given the reader the opportunity to follow the labia Trouessart (1911) with a single species, pre- latter course. Persons doing this, however, should sumably the same as Ride (1970) did, for W. take care to check the nomenclature involved in bicolor. Since there are known hybrids between the transfer, especially to ascertain whether the Wallabia bicolor and Macropus agilis of Ride type species of a genus is being transferred. (1970), Wallabia would seem to be congeneric 6 AMERICAN MUSEUM NOVITATES NO. 2635 with Macropus even in Ride's restricted sense of Gould, 1842, Megaleia Gistel, 1848, and Wallabia Wallabia. Trouessart, 191 1, as synonyms. Gray (1972) also listed crosses of Megaleia rufa with Osphranter robustus and with Osphran- ter antilopinus cervinus. Both of these repre- PRIMATES sentatives of Osphranter were included by Ride In the primates there are many intergeneric (1970) in the genus Macropus, and Simpson hybrids recorded, virtually all from the family (1945) also considered Osphranter a synonym of Cercopithecidae, suggesting that generic splitting Macropus. has been extensive in this group. Simpson (1945) According to Gray (1972) a hybrid was born recognized 10 genera in the subfamily Cerco- in the Frankfurt Zoo from a cross between pithecinae: Macaca, Cynopithecus, Cercocebus, Thylogale thetis and Wallabia rufogrisea frutica. Papio, Comopithecus,Mandrillus, Theropithecus, The latter was considered by Ride (1970) to be Cercopithecus, Allenopithecus, and Erythroce- in the genus Macropus, while he did recognize T bus. The more recent arrangement of Napier and thetis as one of three Australian species in the Napier (1967) was less divided, with A lienopithe- genus Thylogale. This cross would question the cus included in Cercopithecus and Comopithecus validity of the genus Thylogale, especially in light in Papio. of the interspecific crosses known between T. Anderson (1967) recognized only six of the thetis and one of the other species, T. stimatica, 10 genera listed by Simpson: Cercocebus, Cer- in the genus Thylogale. copithecus (including Allenopithecus and Eryth- The intergeneric hybrids known for marsu- rocebus), Cynopithecus, Macaca, Papio (in- pials, then, suggest that the inclusive Macropus of cluding Comopithecus and Mandrillus), and Simpson (1945) with Megaleia as a synonym was Theropithecus. correct. Wallabia, which was recognized by Ride Allenopithecus nigroviridis has been crossed (1970) only for W. bicolor, would not only be with Cercopithecus aethiops and produced a vi- affected by the cross of Thylogale thetis with W. able offspring. rufogrisea frutica, which Ride placed in Macro- Cercocebus atys was successfully crossed with pus, but the genetic compatibility of W. bicolor Macaca fascicularis as well as with Macaca neme- withMacropus agilis also questions the validity of strina, and Cercocebus torquatus has been hybrid- Wallabia as a genus. The cross between Thylogale ized with Macaca fascicularis and with Macaca thetis and W. rufogrisea (=Macropus) questions nemestrina. Similarly, Cercocebus has been suc- the status of Thylogale as a correct generic ap- cessfully crossed with Mandrillus: a living off- pelation for the pademelon T. thetis, if not for spring was born, but did not survive, from a cross the genus Thylogale as a whole. between C torquatus and Mandrillus leucophae- The generic synonymies for each of the hy- us, and a hybrid between C torquatus and M. brids and the earliest name according to priority sphinx survived for four months before dying of are: diarrhea. Cercopithecus aethiops has been reported to Megaleia Gistel, 1848 X Macropus Shaw, 1790= cross successfully with Macaca mulatta, M. radi- Macropus Shaw, 1790. ata, and M. sinica. A hybrid between Cercopith- Osphranter, Gould, 1842 X Macropus Shaw, ecus (probably cephus) and Erythrocebus patas 1790=Macropus Shaw, 1790. was produced in 1960. Cercopithecus pygery- Megaleia Gistel, 1848 X Osphranter Gould, thrus has been successfully crossed with both 1842=Osphranter Gould, 1842. Macaca radiata and Macaca sinica, and Cercopith- Wallabia Trouessart, 1911 X Thylogale Gray, ecus sabaeus with Macaca mulatta. 1838=Thylogale Gray, 1837. Cynopithecus niger has produced living off- Wallabia Trouessart, 1911 X Macropus Shaw, spring when crossed with Macaca fascicularis, 1790=Macropus Shaw, 1790. Macaca mauius, and Macaca nemestrina. Fooden (1976) considered the proper names for C. niger Recommendation. Genus Macropus Shaw, and M. maurus to be Macaca nigra and Macaca 1790, with Thylogale Gray, 1837, Osphranter maura, respectively. 1977 VAN GELDER: HYBRIDS 7

In addition to the crosses between the kinds Allenopithecus Lang, 1923 X Cercopithecus Lin- ofMacaca already mentioned, Macaca fascicularis neaus, 1758=Cercopithecus Linnaeus, 1758. has been crossed successfully both with Mandril- Cynopithecus Geoffroy St.-Hilaire, 1835 X Ma- lus leucophaeus and Mandrillus sphinx. Macaca caca Lacepede, 1799=Macaca Lacepede, 1799. fascicularis is supposed also to have been success- Erythrocebus Trouessart, 1897 X Cercopithecus fully crossed with Papio cynocephalus. Macaca Linnaeus, 1758=Cercopithecus Linnaeus, 1758 maurus, when crossed with Mandrillus leucophae- Mandrillus Ritgen, 1824 X Macaca Lacepede, us, produced a hybrid that died soon after birth. 1799-Macaca Lacepede, 1799. Macaca mulatta was successfully crossed with Theropithecus Geoffroy St.-Hilaire, 1843 X Papio hamadryas, and Macaca nemestrina with P. Papio Erxleben, 1777=Papio Erxleben, 1777. hamadryas. No conception seems to have re- Mandrillus Ritgen, 1824 X Cercocebus Geoffroy sulted from copulations between Macaca neme- St.-Hilaire, 1812=Cercocebus Geoffroy St.- strina and P. anubis, and a hybrid between Hilaire, 1812. Macaca nemestrina and P. ursinus was born dead. Cercocebus Geoffroy St.-Hilaire, 1812 X Macaca Macaca nemestrina is also reported to have pro- Lacepede, 1799=Macaca Lacepede, 1799. duced a hybrid with Mandrillus leucophaeus, the Macaca Lacepede, 1799 X Papio Erxleben, offspring dying soon after birth. 1777=Papio Erxleben, 1777. Mandrillus leucophaeus has successfully hy- Macaca Lacepede, 1799 X Cercopithecus Lin- bridized with Papio hamadryas as well as with naeus, 1758=Cercopithecus Linnaeus, 1758. Papio papio, and M. sphinx with Papio anubis Comopithecus Allen, 1925 X Theropithecus and P. hamadryas. Comopithecus as a generic Geoffroy St.-Hilaire, 1 843=Theropithecus name for the hamadryas baboon is not generally Geoffroy St.-Hilaire, 1843. used today, and crosses between P. hamadryas and P. anubis occur in nature in a narrow zone According to Buettner-Janusch (1966) and where the two species are in contact. This should Napier and Napier (1967) the first valid use of lead to consideration of them only as subspecies. the name Papio was by Miller (1776) for the Papio hamadryas has also produced viable hy- mandrill. Delson and Napier (1976) cited Mul- brids when crossed with Papio ursinus and with ler's (1773) use of Papio (as a subgenus) that Theropithecus gelada (Anon., 1975). seemingly included a macaque, a squirrel It is clear that the 10 genera of cercopith- monkey, and a drill or mandrill. Napier and ecines that Simpson (1945) recognized are far Napier (1967) and Dandelot (1971) utilized more closely related than would be indicated by Papio Erxleben (1777) for baboons even though their separation at the generic level. I have not some of these authors acknowledge that it is dealt with the numerous interspecific hybrids wrong. But because they recognize Mandrillus as produced between the various species within the a genus and the type of Papio Muller (1776) is nominal genera, for which Gray (1972) is the supposedly a mandrill, they believe that stability best source. To show the connecting lineages, we is enhanced by retaining Papio Erxleben for what find the following successful crosses: are popularly called baboons, as opposed to drills and mandrills. Delson and Napier (1976) have Cercopithecus X Allenopithecus to the Commission on X Erythrocebus appealed International X Zoological Nomenclature for a decision regarding Macaca these names. two alternatives: Macaca X Cynopithecus They suggest Thiat X Cercocebus Papio Muller (1773) be retained for the mandrills X Papio and that Chaeropithecus Gervais (1839) be used X Comopithecus for the savannah baboons; or that Papio Erxleben Comopithecus X (1777) be the name of the savannah baboons and that Mandrillus Ritgen (1824) be assigned to the Theropithecus drills and mandrills. The proposal synonymizing The generic synonymies for each of these hy- "genera" of mandrills with baboons obviates this brids and the earliest name according to priority problem at the generic level, as was pointed out are: by Buettner-Janusch (1966), with Papio Muller 8 AMERICAN MUSEUM NOVITATES NO. 2635

(1776 or 1773) serving as the generic name for Attempts to produce hybrids from artificial both baboons and mandrills. insemination of Oryctolagus with Sylvilagus have The congeneracy of Macaca and Papio not not been successful. only had been suggested by Buettner-Janusch (1966), but he also suggested that Cercocebus RODENTIA might be united with these. Chiarelli (1961) had summarized hybrids and noted these in- Considering the large number of genera in the tergeneric crosses, but made no recommendation order Rodentia, it is surprising that there are not of taxonomic changes. The prime novel sug- more cases of hybridization known. In part this gestion I am presenting here is the union of Cer- may be because rodents, especially the smaller copithecus with the all-inclusive Papio. Since ones, have not been especially desired by zoos, Gray (1972) referred to a number of crosses be- where most of the hybridization of other species tween species of Cercopithecus and various has taken place. Alleged crosses between Arvicola species ofMacaca, and to crosses between species and Mesocricetus, Cricetus and Mesocricetus, of Macaca and various baboons, drills, and man- Mesocricetus and Mus, Mesocricetus and Rattus, drills, the union of all of the extant "genera" Microtus and Mus, and Mus and Jaculus were all that were included by Simpson (1945) in the either unsuccessful or are of doubtful validity, subfamily Cercopithecinae into a single genus according to Gray (1972). The same is true of seems warranted. The earliest name for this genus reports of interordinal crosses between Rattus would be Cercopithecus Linnaeus, 1758. Al- norvegicus and Oryctolagus cuniculus. though Dandelot Gray (1972) cited Gunning's (1901) report of (1971) retained five genera of hybridizing Aethomys chrysophilus with Rattus cercopithecines for Africa (Papio, Theropithecus, norvegicus ("Mus chrysophilus" and "Mus decu- Macaca, Cercocebus, and Cercopithecus), he mans"). The status of Aethomys as a genus is commented: "As regards the mangabeys, which complicated, and Ellerman (1941) stated that are sometimes regarded as African representatives "This genus is very difficult to classify, being one of the Asiatic macaques, it can be assumed that of the rather numerous African 'borderline' they form the link between Papio (sensu lato) genera, overlapping to a certain extent Rattus on and Cercopithecus (sensu lato)." the one hand, and Arvicanthis on the other." Recommendation. Genus Cercopithecus Lin- Davis (1971) indicated that the content of the naeus, 1758, with Papio Miiller, 1776 or 1773, genus is still a matter for study, and some of the and Papio Erxleben, 1777, Macaca Lacepede, subgenera are a "mixed bag." Missone (1969) 1799, Cercocebus Geoffroy St.-Hilaire, 1812, thought that Aethomys was "possibly an assem- Mandrillus Ritgen, 1824, Cynopithecus Geoffroy blage of different things." St.-Hilaire, 1835, Theropithecus Geoffroy St.- Another cross reported by Gunning (1901) in- Hilaire, 1843, Erythrocebus Trouessart, 1897, volved Mus musculus and Rhabdomys pumilio. Allenopithecus Lang, 1923, and Comopithecus He stated that "The result of a cross between the Allen, 1925, as synonyms. ordinary albino Mus musculus with the Striped Mouse (Arvicanthis pumilio) is a peculiarly LAGOMORPHA cream-coloured, not striped specimen, which Some hybrids between genera of lagomorphs looks very much like a cream-albino Mouse with have been reported (Gray, 1972), but all of them black eyes, which stand out very distinct against were questioned. Lepus europaeus X Oryctolagus the cream-coloured fur." From what is known of cuniculus hybridization is doubtful because all the genetics of the coat color of Mus, it seems artificial insemination attempts have resulted in highly unlikely that the cream-colored animals little more than fertilization and development no would be the result of this cross, and before ac- further than blastocysts. Alleged crosses between ceptance as a valid hybrid, confirmation is other species ofLepus and Oryctolagus cuniculus needed. are generally doubted by Gray, and all artificial Gray (1972) also reported, on the basis of per- insemination experiments have been unsuccessful sonal communication from E. Schlesser in 1952, in producing offspring. a cross between Rattus (Mastomys) natalensis 1977 VAN GELDER: HYBRIDS 9 and Mus musculus: "Seven hybrid young (3 dd There exists, however, no previous evidence of and 4 99) have been reported. They were dark successful crossing between genera or even grey in colour, like R. natalensis. The hybrids species of dolphins, although that fact in itself showed remarkable growtn. They were wild like does not exclude the chance of its occurrence." the father, and escaped before fertility could be Fraser pointed out that, if the specimens he ascertained." Rattus natalensis is currently examined had not resulted from hybridization placed in the genus Praomys, subgenus Mastomys between Tursiops and Grampus, he might have (Missone 1969, 1971). considered "merging the two genera" because of The distinction of Rattus and/or Praomys the gradation in morphology. He rejected the from Mus is based on characters of the tooth idea that they were not intergrades because row. Ellerman (1941) pointed out that these do "they are so strikingly distinct that, in the pres- not hold up completely in separating Rattus and ent state of our knowledge, such a course would Mus. Missone's key (1971) pairs Praomys and not be justified." Fraser considered and rejected Mus quite closely. Size alone will not distinguish the idea of naming the three specimens as new Rattus and Mus, as there are species of the genera and species, although he found that under former that are smaller than some larger species the criteria in use in taxonomy in 1940, one of ofMus. In the case of the supposed hybridization the three specimens could represent a distinct between R. natalensis and Mus, in the absence of species of Tursiops and the other two would rank any confirming data, because no one seems to as new genera. have duplicated this cross, and in view of some At the time Fraser wrote, no other inter- question about Schlesser's report, as evinced by generic hybrids, or even interspecific ones were the following account, I prefer to defer accept- known in the cetaceans. In 1974, however, Dohl, ance of this cross until there are better data. Norris, and Kang reported a viable hybrid from a Also based on personal communication from female Steno bredanensis and a male Tursiops Schlesser, Gray (1972) mentioned a cross be- truncatus. Steno had been placed in the family lieved to be between Mus (Leggada) sp. and Mus Stenidae (Fraser, 1966) while Tursiops is in the musculus: "In the Belgian Congo, a 9 albino family Delphinidae. Hershkovitz (1966) did not mouse produced offspring believed to be sired by recognize the family Stenidae, although Rice a local wild mouse (Leggada)." Leggada is gener- (1967) did and suggested, on the basis of infor- ally not recognized as more than a synonym of mation from Fraser, that "it may be necessary to Mus (Ellerman, 1941; Missone, 1971). place each of these genera [Sotalia, Sousa, and Recommendation. Because of the content of Steno] in a separate family." However, by 1968, the genera involved and even the parentage of the Rice and Scheffer had put Steno back into the alleged hybrids is open to considerable question Delphinidae. Mead (1975) on the other hand, and requires further study and confirmation, no kept Steno in the family Stenidae. Mead's ana- taxonomic allocations are recommended. tomical studies of the nasal passages and facial complex of Steno, however, led him to point out that it was similar to Tursiops. He also noted the anatomical proximity of Grampus to Tursiops. The intermediate characters of three dolphins The generic synonymies for each of the hy- stranded on the western coast of Ireland in 1933 brids and the earliest name according to priority led Fraser (1940) to suspect that they might be are: hybrids between Tursiops truncatus and Gram- pus griseus. He pointed out that both genera Tursiops Gervais, 1855 X Grampus Gray, 1828= "are of equal size, about 12 feet, both have a Grampus Gray, 1828. common area of distribution and, so far as is Tursiops Gervais, 1855 X Steno Gray, 1846= known, the breeding times coincide. The bottle- Steno Gray, 1846 nosed dolphin is known to be gregarious, and the effect of fertile hybridization would possibly be Recommendation. Genus Grampus Gray, evident in succeeding generations to a varying ex- 1828, with Steno Gray, 1846, and Tursiops Ger- tent, in different individuals within the school. vais, 1855 as synonyms. Those who question the 10 AMERICAN MUSEUM NOVITATES NO. 2635 recognition of the three specimens from the Irish mented that hybridization between Vulpes coast (Fraser, 1940) as hybrids could recognize vulpes and Canis familiaris had not been sus- only the Tursiops X Steno hybrid, for which the tained; in one case that was checked cyto- earliest name is Steno Gray, 1846. logically, the offspring proved to be C. familiaris. Sanjeeva Raj (1953) reported hybrids from a fe- male Vulpes bengalensis crossed with a domestic Among the carnivores there are a large num- of Alsatian X bull-terrier parentage that was ber of intergeneric hybrids, some of which are Dalmation in color. Photographs of the hybrids widely known but are often disregarded as indi- unfortunately do not help to substantiate this cations of taxonomic proximity. case. In the family , hybridization between In 1975 (Anon.) the birth of pups from a Alopex lagopus and Vulpes vulpes (including Canis latrans X Vulpes vulpes was reported at Vulpes fulva) is known and some F1 offspring are Bridgeton, New Jersey. Mr. Henry Ricci, Curator fertile (Gray, 1972). Youngman (1975) used the of the Cohanzick Zoo at Bridgeton, has kindly combination Vulpes lagopus for Arctic foxes, provided me with the details: A male concluding that they were congeneric, thus fol- born April 5, 1971, was donated to the zoo in lowing Bobrinskii, Kuznetsov, and Kuzyakin August of that year by a returning soldier. It was (1944). Stains (1967) also had questioned the at first housed adjacent to a young female Amer- generic status of Alopex on the basis of its hy- ican red fox (Vulpes vulpes) that had been born bridization with Vulpes. Based on numerical classi- about the same time as the coyote. Eight months fication analysis, Clutton-Brock, Corbet, and Hills later the two animals were put in the same cage (1976) maintained Alope-x as a separate, mono- and were compatible. During the first and second typic genus, although they acknowledged that, in weeks of February 1973, the animals were seen comparison with the other genera of canids they to mate and on April 9, 1973, the vixen gave recognize, Alopex is "the least distinctive, and birth to two male pups. They were left with the the skull and teeth show a very high degree of female who began to ignore them on April 12. similarity to certain species of Vulpes especially One was found dead that morning. The second V. corsac." Todd (1970), who analyzed canid pup was removed and hand fed, but died the phylogeny on the basis of karyotypes, stated that evening of April 12 "probably due to congestion despite the distinctive chromosome complement and exposure" according to Dr. Ernest Zirkle, of Alopex, "there appears to be no good reason the zoo's veterinarian. beyond custom for sustaining such a [generic] Judging from photographs I have seen of the distinction. All evidence, both morphological ... parents, there seems no reason to doubt their and through hybridization studies ... indicate a identification, nor does the appearance of the close relationship betweenAlopex and V. vulpes. cage suggest entry by a local wild Vulpes male. Although hybrid sterility characterizes the prod- The pups were preserved and are in the posses- ucts of this cross, the fact that a viable somatic sion of Mr. Ricci and Dr. Zirkle. is produced argues for no greater separa- According to their numerical analysis, Clutton- tion than specific." Brock, Corbet, and Hills (1976) found Vulpes and Recognition of the genus Thos for is Canis to warrant generic separation on the mor- generally not granted today, and in any event, phological-numeric criteria of their study. They hybridization between Canis familiaris as well as do point out, however, that determining the Canis latrans with jackals placed in the genus limits within the Canis/Vulpes/Dusicyon com- Thos have occurred with fertile offspring of both plex is difficult. "If Dusicyon were merged with sexes produced. Canis, it would be difficult to argue that Vulpes Crosses of domestic with other canids should not be treated likewise." Langguth (1975) (and even with Felis) are often alleged, but many did propose the inclusion of some of the South are questionable because adequate documenta- American canids generally known under the tion on parentage is not given. Gray (1972) com- generic name Dusicyon in the genus Canis. 1977 VAN GELDER: HYBRIDS 11

Krieg (1925) reported a hybrid between Canis dent from Clutton-Brock, Corbet, and Hills, familiaris and a pampas fox for which he used especially figures 4 and 5. Hershkovitz (1961) the name "Pseudalopex (Canis) agarae" [sic]. had also regarded Cerdocyon as being congeneric The specific name azarae has been used for at with Dusicyon. least eight taxa of South American canids, and it Pocock (1941) mentioned a Cuon X Canis is difficult to ascertain which species actually aureus cross in a Madras menagerie, but without may have been involved in the hybridization. further data, this hybrid is questioned. Gray (1972) questionably assumed that Cer- No intergeneric hybrids in the family Pro- docyon thous was the mother; there is equal cyonidae have been reported. In the family Mus- likelihood that Dusicyon gymnocercus (=cul- telidae, the only report of an intergeneric hybrid paeus?) may have been involved. According is that of Ackermann (1898) cited by Gray to Cabrera and Yepes (1940), there is anecdotal (1972) between Martes martes and Mustela information that Cerdocyon thous X Canis famil- putorius putorius. However, adequate docu- iaris hybrids were used by indigenous hunters in mentation for this is lacking, and considering the Guianas. Although Langguth (1970, 1975) that these are widely kept fur-bearers for which believed that Cerdocyon should be maintained as other crosses have been attempted, additional ac- a distinct genus, Clutton-Brock, Corbet, and Hills counts would be expected to have been reported. (1976) regarded it as a member of an expanded Based on the hybridization of some of the genus Dusicyon. The arrangements of South Ursidae reported by Gray (1972) the status of American canids by Langguth in 1970 and 1975, most of the genera seems open to question. The and Clutton-Brock, Corbet, and Hills in 1976 are polar , formerly considered Thalarctos, has presented in table 1. been shown to be capable of crossing with Ursus If Kreig's (1925) hybrid actually involved Cer- arctos and the offspring are fertile. The American docyon, the logical conclusion would be a com- black bear, formerly considered the genus bination of Langguth's (1975) merging of Euarctos and still sometimes placed in its own Dusicyon with Canis, and Clutton-Brock, Corbet, subgenus, has produced viable offspring when and Hills's (1976) placement of Cerdocyon in crossed with Ursus arctos. One such hybrid was Dusicyon with all in Canis. The proximity of successfully backcrossed to a grizzly bear (U. a. Dusicyon culpaeus (including culpaeolus) to hornibilis). Ursus arctos has also been crossed with Canis (adustus, mesomelas, and simensis) is evi- Selenarctos thibetanus and produced a viable off-

TABLE 1 Comparison of Three Classifications of South American Canids Clutton-Brock, Corbet, and Hills, 1976 Langguth, 1975 Langguth, 1970 Vulpes cinereoargenteus Urocyon cinereoargenteus Urocyon cinereoargenteus Dusicyon australis Canis (Dusicyon) australis Dusicyon (Dusicyon) australis Dusicyon culpaeus Canis (Pseudalopex) culpaeus Dusicyon (Pseudalopex) culpaeus Dusicyon culpaeolus [= D. culpaeus andensis] Dusicyon gymnocercus Canis (Pseudalopex) gymnocercus Dusicyon (Pseudalopex) gymnocercus Dusicyon inca [= D. culpaeus andensis] Dusicyon griseus Canis (Pseudalopex) griseus Dusicyon (Pseudalopex) griseus Dusicyon fulvipes [= D. griseus fulvipes] Dusicyon sechurae Canis (Pseudalopex) sechurae Dusicyon (Pseudalopex) sechurae Dusicyon vetulus Lycalopex vetulus Lycalopex vetulus Dusicyon thous Cerdocyon thous Cerdocyon (Cerdocyon) thous Dusicyon microtis A telocynus microtis Cerdocyon (Atelocynus) microtis Chrysocyon brachyurus Chrysocyon brachyurus Chrysocyon brachyurus Speothos venaticus Speothos venaticus Cerdocyon (Speothos) venaticus 12 AMERICAN MUSEUM NOVITATES NO. 2635 spring (Gray, 1972; Duplaix-Hall, 1975). Asakura leopards, and between jaguars and leopards has (1969) reported mating of Helarctos malayanus been accomplished. Some of the offspring are with Melursus ursinus and the birth of a single fertile in these crosses, and the ability to hy- cub 95 days later. Thus of the six genera of bridize bespeaks the genetic proximity of these that were recognized by Simpson in 1945, only species and questions their separation into genera one, Tremarctos, has not been known to hybrid- or even subgenera. The hybridization of Felis ize with some other genus of bear: catus with rufa (Gray, 1972) similarly creates doubt as to the generic separability of the Ursus X Thalarctos nominal genus Lynx. X Selenarctos Of more taxonomic importance is the hy- Helarctos X Melursus bridization of Panthera pardus with Felis concolor reported in Gray (1972). The produc- Simpson (1961) acknowledged that the hy- tion of viable offspring from this cross implies bridization between Thalarctos and Ursus indi- that Panthera and Felis are congeneric or that cated congeneracy. However, the generic name Felis concolor should be placed in Panthera-a Thalarctos is still widely used for the polar bear, consideration that would be regarded as unac- especially in Europe. The validity of Selenarctos ceptable on the morphological grounds that are as a genus was also questioned by Ellerman and supposed to distinguish Panthera. Morrison-Scott (1951). Bobrinskii, Kuznetsov, I have used the name Panthera for the "large" and Kuzyakin (1944) referred both Selenarctos here because that is the name that Gray and Thalarctos to Ursus. The record of a hybrid (1972) used; however, the nomenclature is cur- between Melursus and Helarctos also points to rently sub judice (and presumably has been for a the genetic proximity of these bears. Simpson quarter of a century) by the International Com- (1945) stated that all of the nominal genera of mission for Zoological Nomenclature. As pre- living bears were "very closely allied." Thenius viously expressed (Van Gelder, 1975), my belief (1976) found Tremarctos to be subfamilialy dis- that the correct name for the large cats, if one is tinct from the other living bears. to be used at all, is Leo. There are no intergeneric crosses reported In the pinnipeds, hybridization has occurred from the viverrids or the hyaenids. between Arctocephalus pusillus and Zalophus The felids have, at times, been placed in many californianus (Gray, 1972). These two genera are genera and subgenera, very nearly one for each sometimes placed in separate subfamilies (Stains, species. Current usage is still variable, and ranges 1967), which classification the hybridization from recognition of only two genera, Acinonyx would question. Another hybrid between for the cheetah and Felis for all other living cats, Zalophus and some other otariid, either Arcto- to finer distinctions, the most common being the cephalus pusillus or Otaria byronil, has also been consideration that, in addition to the cheetah as reported (Gray, 1972). The most parsimonious a distinct genus, Felis be split into the "large" consideration would be to regard it as a Zalophus cats under the name Panthera or Leo and the X Arctocephalus cross, even though the zoo- "small" cats with Felis as the generic name. Neo- keepers suspected Otaria as the sire. The alleged felis is often distinguished generically as well, hybridization between Callorhinus ursinus and and recognition of the snow as Uncia and Otaria byronii which Gray (1972) cited from the as Lynx is not uncommon. Simpson Ackermann (1898) does not seem likely on the (1945) listed three genera: Acinonyx, Panthera geographic grounds of its basis and may represent (with six subgenera, one for each species), and an error in updating synonymy. The statement in Felis (with 16 possible subgenera for about 29 Ackermann is based on an observation attributed species). to Steller, and might refer to Eumetopias, rather Hybridization (Gray, 1972) between lions than Otaria, and Callorhinus. However, it would (subgenus Leo) and (subgenus Tigris), lions require further substantiation before I accept it. and leopards (subgenus Panthera), lions and In the phocids, a newbom but dead hybrid jaguars (subgenus Jaguarius), between tigers and between Halichoerus grypus and Phoca hispida 1977 VAN GELDER: HYBRIDS 13 was found in the Stockholm zoo in 1929 (Gray, recommendation of a broadly construed genus 1972). Canis seems to represent a balanced appraisal of Simpson (1945) recognized five genera of the genus from the separate conclusions of the otariids (Arctocephalus, Callorhinus, Zalophus, authorities, Langguth, Clutton-Brock, Corbet, Eumetopias, and Otaria). Scheffer (1958) sepa- and Hills, together with the genetic information rated Neophoca from Zalophus as an additional provided by the known hybridizations. genus. Winge (1941), however, had a much Ursus Linnaeus, 1758, with Thalarctos Gray, broader view, recognizing only Arctocephalus 1825 and Selenarctos Heude, 1901 as synonyms. and Otaria in what is usually considered the Melursus Meyer, 1793 with Helarctos Hors- family Otariidae. field, 1825 as a synonym. The generic synonymies for each of the carni- Felis Linnaeus, 1758 with Panthera Oken, voran hybrids accepted as valid by me, and the 1816 (=Leo Brehm, 1829) as a synonym. earliest name according to priority are: Arctocephalus E. Geoffroy St.-Hilaire and F. Alopex Kaup, 1829 X Vulpes Bowdich, 1821= Cuvier, 1826 with Zalophus Gill, 1866 as a syn- Vulpes Bowdich, 1821. onym. Vulpes Bowdich, 1821 X Canis Linnaeus, 1758= Canis Linnaeus, 1758. PERISSODACTYLA Dusicyon H. Smith, 1839 X Canis Linnaeus, Other than in the family , no hybrids 1758=Canis Linnaeus 1758. are reported for the perissodactyls. Each of the Thalarctos Gray, 1825 X Ursus Linnaeus, 1758= seven species of generally recognized equids has Ursus Linnaeus, 1758. been shown to be capable of hybridizing with Selenarctos Heude, 1901 X Ursus Linnaeus, virtually each of the other species with which 1758=Ursus Linnaeus, 1758. mating has been attempted. The offspring are al- Helarctos Horsfield, 1825 X Melursus Meyer, most always sterile. Various subgenera have been 1793=Melursus Meyer, 1793. suggested for many of the species, and sometimes Panthera Oken, 1816 X Felis Linnaeus, 1758= have been used as genera (Asinus, Hippotragus, Felis Linnaeus, 1758. Dolichohippus, Onager, Hemionus). The hybridi- Zalophus Gill, 1866 X Arctocephalus E. Geof- zation data would suggest that these taxonomic froy St.-Hilaire and F. Cuvier, 1826=Arcto- categories are questionable at the generic level. cephalus E. Geoffroy St.-Hilaire and F. Cuvier, 1826. ARTIODACTYLA Recommendation. Canis Linnaeus, 1758, in- No intergeneric hybrids are reported from the cluding Alopex Kaup, 1829, Urocyon Baird, families Suidae, Hippopotamidae, , 1857, Fennecus Desmarest, 1804, Vulpes Bow- Tragulidae, or Giraffidae. dich, 1921, Cerdocyon Hamilton-Smith, 1839, In the family Tayassuidae, there are three liv- and Dusicyon Hamilton Smith, 1839. This ing species, the third being the newly discovered recommendation represents a combination of South American Catagonus, a genus formerly taxonomic conclusions mainly by Langguth known only from fossils (Wetzel et al., 1975). (1970,1975) and Clutton-Brock, Corbet, and Hills Simpson (1945) recognized Catagonus, which he (1976) and the data from hybridization. The lat- knew only as a fossil, and put the other two ter recognized the genus Vulpes to include Ur- living peccaries in the single genus Tayassu. ocyon and Fennecus, to which may be added Woodburne (1968) concluded that the two living Alopex on the basis of its hybridization with species of peccaries then known were generically Vulpes. Langguth (1975) concluded that Dusi- distinct, with Tayassu pecari for the white-lipped cyon (species australis, griseus, sechurae, and and Dicotyles tajacu for the collared peccary. culpaeus) was more properly placed within the The generic distinction of these two peccaries genus Canis, but continued to recognize Cerdo- was initially followed by Wetzel et al. (1975), cyon. Clutton-Brock, Corbet, and Hills (1976) in- but later Wetzel (1976) indicated that he con- cluded Cerdocyon in Dusicyon. The present sidered Tayassu and Dicotyles congeneric. This 14 AMERICAN MUSEUM NOVITATES NO. 2635

conclusion would seem to be in keeping with the Elaphurus, would comprise only a single genus, information (Gray, 1972) that Tayassu and Cervus. Except for Elaphurus, Haltenorth (1963) Pecari (=Dicotyles) had successfully been hybrid- included all of these in Cervus, recognizing Dama ized at the London Zoo. and Axis as subgenera, as well as a number of In the family Cervidae, the number of genera other subgenera. Hybridization between Cerpus has been reduced since 1945 when Simpson elaphus and Elaphurus davidianus has been re- listed 17. His arrangement included a number of ported (Gray, 1972), and fertile young of both suprageneric divisions, and no subgeneric ones. sexes have been backcrossed to C. elaphus. Comparison of Simpson's (1945) arrangement Crosses between a number of the nominal sub- with that of Haltenorth (1963) is given below. genera (see Gray, 1972) are known. The 17 genera of Simpson are arranged by Two reported hybridizations involving Cerpus Haltenorth in 11 genera, all of the generic lump- (as construed by Haltenorth, 1963) would seem ing taking place within Simpson's subfamilies to bear on higher categories of the deer. Simpson Cervinae and Odocoileinae. The data available (1945) considered a major division to be between from intergeneric hybridizations not only bear the Cervinae and the Odocoileinae. He put Cap- out Haltenorth's conclusions, but suggest addi- reolus in its own tribe, Capreolini, in the Odo- tional generic lumping. coileinae. Haltenorth, too, allied Capreolus On the basis of hybridization, it would appear with the Odocoilinae [sic], and in a sense placed that the subfamily Cervinae, as construed by them even closer to Odocoileus in that he did not Simpson (1945) with Dama, Axis, Cervus, and erect a subfamily for Capreolus, as he did for

Simpson, 1945 Haltenorth, 1963 Family Cervidae Family Cervidae' Subfamily Moschinae Subfamily Moschinae Genus Moschus Genus Moschus Subfamily Muntiacinae Subfamily Munticinae Tribe Muntiacini Genus Mun tiacus Genus Muntiacus Genus Elaphodus Genus Elaphodus Subfamily Cervinae Subfamily Cervinae Genus Dama Subgenus Dama, Genus Cervus Genus Axis Subgenus Axis, Genus Cervus Genus Cervus Genus Cervus, Subgenus Cervus Genus Elaphurus Genus Elaphurus Subfamily Odocoileinae Subfamily Odocoilinae Tribe Odocoileini Genus Odocoileus Genus Odocoileus Genus Mazama Genus Mazama Genus Hippocamelus Subgenus Hippocamelus, Genus Odocoileus Genus Blastocerus Subgenus Odocoileus, Genus Odocoileus Genus Ozotoceras Subgenus Blastoceros, Genus Odocoileus Genus Pudu Subgenus Pudu, Genus Mazama Tribe Alcini Subfamily Alcinae Genus A Ices Genus AIces Tribe Rangiferini Subfamily Rangiferinae Genus Rangifer Genus Rangifer Tribe Hydropotini Subfamily Hydropotinae Genus Hydropotes Genus Hydropotes Tribe Capreolini Genus Capreolus Genus Capreolus, Subfamily Odocoilinae 1977 VAN GELDER: HYBRIDS 15 each of Simpson's other tribes in this group. Both Axis (=Cervus) hybridization is made, it is best to Capreolus and Odocoileus were reputed (Gray, regard the record as questionable. Gray's other 1972) to have hybridized with species now in- citation of a Capreolus cross, an interfamilial one cluded in the genus Cervus, and in the subfamily with Ovis aries, is not regarded as valid. Cervinae of Simpson (1945). The alleged hybrid between Alces and Cervus Gray (1972) reported two instances of hybrid- canadensis (Gray, 1972; Anon., 193 1) is undocu- ization of Axis with Odocoileus. One of these, mented by a specimen and is not regarded as Axis axis X Odocoileus virginianus, she cited valid. Of even greater doubt is a reputed inter- from Lantz (1910). Lantz (1910) stated merely familial Alces X Bos taurus cross (Gray, 1972). that "the chital or spotted deer (Cervus axis) of The higher categories of the family India and Ceylon . . . has been crossed with the have received varied treatments. Simpson (1945) Virginia deer." Subsequently, he attributed this arranged them in five subfamilies: , information to Caton (1877). If Caton's stock Cephalophinae, Hippotraginae, Antilopinae, and actually did come from Ceylon, the animals in- , and split these into tribes. Haltenorth volved in the hybridization could have been (1963) put the bovids into 10 subfamilies with either Axis (Hyelaphus) porcinus or Axis axis 22 tribes, and reduced the number of genera ceylonensis. From Caton's descriptions, it is not from Simpson's 54 to 42 by the inclusion of possible to distinguish the two, nor is it possible Bibos in Bos, and Anoa and Synceros in Bubalus, to avoid the conclusion that his "Ceylon deer" Philantomba and Sylvicapra in Cephalophus, and "Acapulco deer" (which have been regarded Gorgon in Connochaetes, Beatragus in Dame- as Odocoileus virginianus) were actually the same liscus, Adenota and Onotragus in Kobus, Rhyn- species. Again and again Caton indicated how chotragus in Madoqua, Nesotragus in Neotragus, similar the two were in antlers, coloration, tail, and Taurotragus in Tragelaphus. Haltenorth con- and other characters. Inasmuch as it does not sidered Tragelaphus De Blainville, 1816 as the seem possible to determine accurately what appropriate name for what Simpson called Strep- species may have been involved in these crosses, I siceros Frisch, 1775. Koopman's (1967) arrange- deem it best to regard these data as too dubious ment of Simpson's five families was essentially for consideration. What seems to be a second ac- that of Simpson, but he followed Haltenorth ex- count of an Axis X Odocoileus cross in Gray cept for retaining Sylvicapra and Taurotragus as (1972) is cited from Przibram (1910) whose in- valid genera. formation is derived also from Caton (1877). The data from hybridization confirm some of Gray (1972) cited a reference to the hybrid- Haltenorth's generic amalgamations. The banteng ization of Capreolus capreolus with Axis (Hy- (Bibos banteng) and the gaur and gayal (Bibos elaphus) porcinus. She derived her data from frontalis and/or gaurus) not only hybridize with Rorig (1903) who reported that two hybrids one another in captivity, but also with members were born at the Dresden Zoological Garden in of the genus Bos including B. indicus, B. taurus, May 1862. The fawns, one of each sex, lived only and B. (Poephagus) grunniens. Some degree of for eight days. Rorig gave as his reference "Z. fertility exists in some offspring of these crosses. [oologische] G. [arten] 2.65", but no reference Haltenorth (1963) recognized Bibos as a sub- to this cross appears there, nor does there seem genus of Bos, with two species, B. javanicus (the to be any such account from the Dresden banteng) and B. gaurus (the gaur). Zoological Garden within a year or two of that The genus Bison of Simpson (1945) and time. If this case of hybridization actually in- Haltenorth (1963) is also capable of hybridizing volved the two genera as stated, then a question with the "genus" Bibos, as an offspring resulted of the alliance of Capreolus with the Odocoil- from a cross between a Bison bull and a female eines would be raised. Flerov (1952), inciden- Bos indicus X Bibos gaurus hybrid. This triple tally, allied Capreolus with Cervus (with Rusa, hybrid was also fertile when backcrossed to Axis, and Dama as subgenera), Alces, and Bison (Gray, 1972). Bison has also been success- Elaphurus as members of his subfamily Cervinae. fully crossed with another hybrid, Bos (Poe- However, until substantiation of Capreolus X phagus) grunniens X Bibos frontalis (Gray, 16 AMERICAN MUSEUM NOVITATES NO. 2635

1972). Fertile offspring have resulted from Connochaetes and Gorgon. Haltenorth (1963) crosses of Bison with Bos indicus as well as with divided this group into two tribes, Alcelaphini Bos taurus. Haltenorth (1963) considered Bison with Damaliscus and Alcelaphus, regarding Bea- bison and Bison bonasus to be conspecific, and tragus hunteri as no more than a subspecies of listed them under the latter name. Hybrids be- Damaliscus lunatus, and Connochaetini with a tween these two nominal species are fertile in single genus, Connochaetes, divided into sub- both sexes. genera Connochaetes and Gorgon. Koopman Gray (1972) listed a number of crosses involv- (1967) followed this generic arrangement, but ing Bos taurus and/or B. indicus with Alces, recognized three species in Damaliscus, evidently Bubalus, Cervus, Ovibus, Ovis, and Taurotragus. considering hunteri a valid species. Ansell (1971) None of these is believed to have been successful likewise recognized hunteri, dorcas, and lunatus in producing offspring. as the species of Damaliscus. Although Halten- Haltenorth (1963) included Taurotragus and orth (1963) recognized only a single species of Boocercus in the genus Tragelaphus, regarding Alcelaphus, Koopman (1967) and Ansell (1971) them both as belonging to the subgenus Tauro- both recognized two species, A. buselaphus and tragus. Koopman (1967) did not follow Halten- A. lichtensteinL orth in this, leaving Taurotragus as a genus. Hybridization between the red hartebeest, Anseil (1971) regarded Boocercus as only a sub- Alcelaphus buselaphus, and the blesbok, Dama- genus within Tragelaphus, largely on the basis of liscus dorcas phillipsi, was reported by Kettlitz the hybridization of Boocercus eurycerus with (1967). Although these hybrids were presumably Tragelaphus spekei (Tijskens, 1968), but in re- sterile, they had grown to maturity. Selous gard to a Taurotragus oryx X Tragelaphus strep- (1893) reported on a specimen that was thought siceros hybrid (Bolineau, 1933), Ansell required to be a hybrid between the red hartebeest (A. knowledge of the fertility of the hybrid before buselaphus caama) and the tsessebe (Damaliscus making a generic change. I have commented else- lunatus lunatus). where (Van Gelder, 1977) on this cross, and it Gorgon as a genus for the black wildebeest is has recently been repeated at the San Diego Zoo usually not recognized today, although some re- (Jorge, Butler and Benirschke, 1976). Female gard it as a valid subgenus. Viable and fertile hy- hybrids of the Boocercus X Tragelaphus spekei brids have been produced between the black cross were fertile (Koulischer, Tijskens and Mor- wildebeest and blue (Connochaetes taurinus) ac- telmans, 1973). Taurotragus oryx has also been cording to documentation in Ansell (1971). successfully crossed with Tragelaphus spekei In 1945 Simpson recognized five genera in his (Anon, 1975). tribe Reduncini: Kobus, Adenota, Onotragus, Neither Koopman (1967) nor Ansell (1971) Redunca, and Pelea. Haltenorth (1963) put Pelea followed Haltenorth's (1963) inclusion of Sylvi- in its own tribe, Peleini, while Ansell (1971) capra in Cephalophus. They did include the third raised it to subfamilial rank, Peleinae. Ansell genus of duiker that Simpson recognized in the (1971) and Haltenorth (1963) retained Redunca subfamily Cephalophinae, Philantomba, as a syn- as a genus in the subfamily Reduncinae (Ansell) onym of Cephalophus. Sylvicapra, however, has or tribe Reduncini (Haltenorth) along with the hybridized with Cephalophus nigrifrons (Anon., single genus Kobus. Haltenorth (1963) recog- 1965). Another hybrid (Bigalke, 1932) died nized four species of Kobus, each of which he shortly after birth and had an abnormal lower put in its own subgenus: K. (Kobus) ellipsi- jaw. The presumptive parents of this cross were prymnus, including defassa; K. (Adenota) kob, Sylvicapra grimmia and Cephalophus natalensis. including vardoni; K. (Hydrotragus) leche; and K. Lydekker and Blaine (1914) had earlier consid- (Onototragus) megaceros. Ansell (1971) did not ered Sylvicapra and Cephalophus as congeneric. recognize any subgenera of Kobus, and consid- Simpson (1945) recognized five genera in his ered K. vardoni a valid species. Of Simpson's tribe Alcelaphini: three hartebeests, Damaliscus, genera and tialtenorth's subgenera, Gray (1972) Alcelaphus, and Beatragus; and two wildebeests, recorded hybrids between Kobus and Adenota, 1 977 VAN GELDER: HYBRIDS 17

Kobus and Onototragus, and Hydrotragus and been noted, and the amount of reliance on ex- Onototragus. A fertile hybrid of a Kobus X ternal characteristics required to differentiate Onototragus cross was successfully mated with them is well indicated by Ellerman and Morri- an Adenota. Ansell (1971) pointed out that Hal- son-Scott's (195 1) table of differences. tenorth's use of the name Onototragus Heller, The generic synonymies for each of the artio- 1913, for K. megaceros was an error; Heller's dactyl intergeneric hybrids I deem valid, and the name was an erroneous spelling of Onotragus, for earliest name according to priority are: which the type is K. leche. Hydrotragus Fitzinger (1866) may be an earlier name for Onotragus Tayassu Fischer, 1814 X Pecari Reichenbach, (Ellerman, Morrison-Scott and Hayman, 1953). 1835 or Dicotyles Cuvier, 1817=Tayassu Simpson (1945) erected the tribe Caprini for Fischer, 1814. five genera: Ammotragus, Capra, and Ovis, Hemi- Dama Frisch, 1775 (invalid) or Platyceros Zim- tragus, and Pseudois. Haltenorth (1963) followed mermann, 1780 X Axis Smith, 1827=Platy- ceros Zimmermann, 1780. this classification. Ansell (1971) who was con- Axis H. Smith, 1827 X Cervus Linnaeus, 1758= cerned only with African genera, pointed out Cervus Linnaeus, 1758. that Ammotragus had not only been hybridized Elaphurus Milne-Edwards, 1866 X Cervus Lin- with domestic Capra hircus, but that the off- naeus, 1758=Cervus Linnaeus, 1758. spring were fertile when backcrossed with a male Bibos Hodgson, 1837 X Bos Linnaeus, 1758=Bos Capra ibex. This led Ansell to regard Ammo- Linnaeus, 1758. tragus as a subgenus of Capra, as Capra (Ammo- Bison H. Smith, 1827 X Bos Linnaeus, 1758=Bos tragus) lervia. Attempted hybridization between Linnaeus, 1758. Ammotragus and Ovis aries either from natural Taurotragus Wagner, 1855 X Tragelaphus De (captive) matings or artificial insemination was Blainville, 181 6=Tragelaphus De Blainville, not successful in producing an offspring (Gray, 1816. 1972). Nadler, Hoffmann and (1974), on Boocercus Thomas 1902 X Tragelaphus De Blain- the basis of chromosomal homologies, postulated ville, 1816=Tragelaphus De Blainville, 1816. that Ammotragus was closer to the ancestral lin- Sylvicapra Ogilby, 1837 XCephalophus H. Smith, eage of Ovis than to Capra. Manwell and Baker 1827=Cephalophus H. Smith, 1827. (1975), however, found that biochemical analy- Damaliscus Sclater and Thomas, 1894 X Alce- ses of , , and aoudad haemoglobins laphus De Blainville, 1816=Alcelaphus De "suggest that A. lervia is no closer to Ovis than to Blainville, 1816. Capra; indeed, the nature of the ca chain duplica- Gorgon Gray, 1850 X Connochaetes Lichtenstein, tion indicates that there is more in common be- 1814=Connochaetes Lichtenstein, 1814. tween A. lervia and C. hircus than between A. Adenota Gray, 1847 X Kobus A. Smith, 1840= lervia and 0. aries." Kobus A. Smith, 1840. Hybrids between (Capra) and sheep Onotragus Gray, 1872 (=Hydrotragus Fitzinger, (Ovis) have long been reported, but seldom ade- 1866) X Onototragus Haltenorth, 1963 (nec quately documented. Between the time of Gray's Heller, 1913)=Hydrotragus Fitzinger, 1866. 1954 compilation and that of 1972, hybridiza- Onototragus Haltenorth, 1963 (nec Heller, 1913) tion between Capra hircus and Ovis aries with the X Kobus A. Smith, 1840=Kobus A. Smith, production of viable offspring that were fertile 1850. when artificially inseminated with Ovis semen Ammotragus Blyth, 1840 X Capra Linnaeus, was reported (see Gray, 1972, for extensive 1758=Capra Linnaeus, 1758. bibliography). An additional instance of hybrid- Ovis Linnaeus, 1759 X Capra Linnaeus, 1758= ization between Ovis aries and Capra ibex was Capra Linnaeus, 1758 (page priority). believed to have involved C. hircus, rather than C. ibex. Capra hircus and C. ibex are interfertile, Recommendation. Tayassu Fischer, 1814, in any event (Gray, 1972). with Dicotyles Cuvier, 1817, and its synonym The similarities of sheep and goats have long Pecari Reichenbach, 1835, as synonyms. 18 AMERICAN MUSEUM NOVITATES NO. 2635

Cervus Linnaeus, 1758, with Dama Frisch, strong subjective components, as the selection of 1775 (=Platyceros Zimmermann, 1780), Axis H. criteria for analysis or the evaluation of primitive Smith, 1827, and Elaphurus Milne Edwards, and advanced, or what is shared, are still prod- 1866, as generic synonyms. ucts of the investigator's personal and scientific Bos Linnaeus, 1758, with Bibos Hodgson, objectives. 1837, and Bison H. Smith, 1817, as synonyms. The application of reproductive data to higher Tragelaphus De Blainville, 1816, with Tauro- categories, as proposed here, has largely been ig- tragus Wagner, 1855, and Boocercus Thomas, nored. In part, perhaps, this stems from the con- 1902, as synonyms. cept that the criterion of reproductive isolation Cephalophus H. Smith, 1827, with Sylvicapra for the species is unspokenly accepted as a com- Ogilby, 1837, as a synonym. ponent of the definition of the genus and higher Alcelaphus De Blainville, 1816 with Damalis- categories. When investigators have faced the cus Sclater and Thomas, 1894, and Beatragus question of intergeneric hybridization of mam- Heller, 1912 (following Haltenorth, 1963, and mals, they have usually concluded that the pro- Ansell, 1971) as generic synonyms. duction of a viable offspring from supposedly Connochaetes Lichtenstein, 1814, with Gor- different genera is or might be sufficient grounds gon Gray, 1850, as a synonym. for lumping the genera (Buettner-Janusch, 1966; Kobus A. Smith, 1840, with Adenota Gray, Simpson, 1961; Stains, 1967; Ansell, 1971; Van 1847, and Onotragus Gray, 1872 (=Hydrotragus Gelder, 1977). To my knowledge, the present Fitzinger, 1866) as synonyms. paper is the first attempt to consider all known Capra Linnaeus, 1758, with Ovis Linnaeus, intergeneric mammals in hybrids in the context 1758, and Ammotragus Blyth, 1840, as syno- of the validity of the genera and to apply the nyms. concept of reproductive incompatibility to the definition of the genus. DISCUSSION AND CONCLUSIONS Basically, the logic of a reproductively iso- lated and self-contained genus seems inescapable A trend in taxonomy for more than a century if one accepts the concept of the reproductively has been to increase the objectivity by which isolated species. The genus neither can nor taxa are distinguished. For the most part this should be of less dimension than the species, and trend has been concerned with the species and if the parameters of the species are ultimately infraspecies categories. The utilization of a established by its reproductive capabilities, then broader spectrum of data than morphology the genus, too, must be so proscribed. The alone-the "new systematics" and the application greatest extent of reproductive compatibility of reproductive compatibility or incompatibility allowed between species is generally the produc- to the definition and concept of the species-has tion of sterile offspring. The production of fertile been a major component in an objective ap- offspring in nature is usually sufficient grounds proach to taxonomy. In addition, the establish- for merging the parental stocks into a single ment of statistical and population parameters for species with their recognition only as subspecies. species and subspecies, and the utilization of Similarly, for allopatric species, captive hybridi- computers to group data and taxa are further zation with fertile offspring may be used to con- indications of this trend. The attempt to apply sider the parental stocks conspecific. If these are objective criteria to taxonomic categories higher the reproductive limits of species it would seem than species is quite recent, stemming in part to follow that the genus must be reproductively from the clustering evaluations that can be pro- at least, if not more, separable, and that crosses duced by computers. Cladistic analysis, similarly, between genera be wholly incapable of producing is another approach to attempt to reduce the a live offspring. subjectivity of the investigator and is applied pri- The utilization of reproductive data in defin- marily to taxonomic categories higher than ing genera, especially as applied in this paper, is species. All of these techniques, however, have open to a number of criticisms. In some cases, 1977 VAN GELDER: HYBRIDS 19 especially older ones, the identity of the genera of the genus incorporates the concept of the re- involved in the cross may be erroneous, or the productively isolated species, one must assume parentage of the hybrid may somehow be sus- that reproductive isolation is also a component pect. Although I have tried to eliminate those of the definition of a genus. that I or others (e.g., Gray, 1972) viewed with That Mayr, Linsley and Usinger (1953) and reservation, some of the data remaining may still others (e.g., Ross, 1975) regard the genus as arbi- be erroneous, leading to incorrect taxonomic trary is evident in their guidelines for the size of conclusions. genera, essentially "lumping" monotypic species No systematic attempts at hybridization for and "splitting" large genera. Ross (1975) stated the purpose of determining generic status have his contention that a genus should not contain been made. Most of the data are the result of more than 40 species. These suggestions contain fortuitous matings in zoos or laboratories. As a the premise that the genus is not a "real" evolu- result, the data for any group of mammals are tionary component but only a human artifact not uniform, and while there may be evidence developed for convenience in handling species. that species of genera A and B may hybridize, Schaefer (1976) discussed the "reality" of higher the status of species in genus C in, for example, categories from an evolutionary viewpoint of the same tribe or subfamily may not be resolved. adaptive levels. Objection to the use of hybridization data on The generic groupings proposed in this paper these grounds is the same as has been raised by unintentionally fit within the criteria of Ross for some (e.g., Blackwelder, 1967) against the use of genera. The broadly construed genus Cercopith- the "biological" species concept. However, to ecus contains exactly 40 species; the most species ignore data when they exist by demanding equiv- the genus Canis would have is 36; and all other alent data for all taxa seems not only grossly genera would have fewer. The idea of an arbi- unscientific, but if applied uniformly would stul- trary 40 as the limit on the size of a genus is tify taxonomy and science as a whole. neither biological nor evolutionary, and not even For the most part, the genus has been ac- representative of convenience. As of 1967 (data cepted and utilized as an arbitrary category. from Anderson and Jones, 1967, except as Mayr, Linsley and Usinger (1953) stated "An ob- noted) there were 12 genera (about 1% of the jective criterion for generic rank does not exist genera of living mammals) with 40 or more equivalent, let us say to reproductive isolation as species: Crocidura, 160; Sorex, 40;Pteropus, 65; a species criterion. It is therefore impossible to Rhinolophus, 71 (reduced to 68 in Koopman and give an objective definition of the genus." Al- Jones, 1970); Hipposideros, 43 (increased to 44 though they did not deal with reproductive com- in Koopman and Jones, 1970); Myotis, 68; patibility between genera, Mayr, Linsley and Pipistrellus, 50; Tadarida, 49 (reduced to 45 in Usinger (1953), who recognized and fostered the Koopman and Jones, 1970); Oryzomys, 58; "biologic species' concept, tied the genus firmly Peromyscus, 57; Microtus, 58; and Rattus, 120. to the species in their definition" "A genus is a Thus, some 20 percent of the species of living systematic category including one species or a mammals are considered to exist in genera that group of species of presumably common phylo- would be regarded by Ross (1975) as too large. genetic origin, which is separated from other A criticism of the generic groupings proposed similar units by a decided gap." (Italics theirs.) here is that there would be a loss of the "infor- They regarded the species as having the essential mation content" of the genera. Presumably this property of reproductive isolation, while they implies an intellectual concurrence of the exteht clearly indicate that the essence of the genus is of a particular genus, a contention that lacks morphological distinctness. They did not deal merit either in theory or in practice. If the genus with the lower limits of the genus to consider is arbitrary, then the use of a generic name has what happens at the interface between the information content no greater than that of the criteria of reproductive isolation with morpho- species that is the type of the genus. If one talks logic distinctness. However, since their definition of the genus Peromyscus, does that automatically 20 AMERICAN MUSEUM NOVITATES NO. 2635 include or exclude Baiomys or Ochrotomys? dimorphic bovids, cervids, and cercopithecids. In does a discussion of Papio include Mandrillus, or practice, of course, most mammalogists would does mention of Macaca include Cynopithecus? agree that many species in these families may Because the category of genus is flexible and generally be identified far more readily by exam- somewhat arbitrary and therefore subject to ination of adult males than adult females, and by varied interpretation, its ability to convey infor- their external characters more readily than by mation concerning its content is fundamentally skulls or teeth. Females of these groups, on the limited. The application of reproductive data, as other hand, are often much more difficult to proposed in this paper should, in fact, serve to re- identify, mainly because the criteria for "gen- duce the arbitrariness of the concept of the genus eric" distinction are largely if not entirely based and thereby enhance the stability of its informa- on dimorphic male attributes. tion content. It is generally conceded that taxa of a given The idea of a stable genus with a fixed infor- category in different phyla of animals are not mation content exists largely only as an element equivalent, the most uniform perhaps being the of taxonomic tradition and experience, and in species (Schaefer, 1976). Even within a group practice one finds constant fluctuation. In the 14 such as the class Mammalia, the higher categories years between the publication of Allen's check- in one order may not be the equivalent of those list of African mammals (1939) and that of Eller- in others. In this context, it is particularly inter- man, Morrison-Scott and Hayman for southern esting to note that essentially all of the genera Africa (1953), 51 of 210 (24%) generic names of involved in hybridization are of animals that are mammals from southern Africa were changed, largely diurnal. In these the species-isolating three for nomenclatural reasons, the remainder mechanisms seem to be based importantly on on taxonomic grounds (Van Gelder, 1975). In visual identification, and a high degree of social the 22 years from the publication of Simpson complexity and sexual dimorphism has devel- (1945) to that of Anderson and Jones (1967) the oped. The ready discernment and conspicuous- number of Recent genera increased from 932 to ness of these identifying characteristics seems to 1004, about 8 percent. However, a sample of the have biased taxonomic judgment to grant these first 100 generic names listed by Simpson com- species higher taxonomic status than they may pared with Anderson and Jones reveals changes merit on actual evolutionary grounds or in rela- involving 39 percent of them. Thus, changing tion to the taxonomic levels of other groups of concepts and content of mammalian genera not mammals. Conversely, the genera of mammals only occurs, but does so at a rate of two or three with 40 or more species that I listed on p. 19, a year. Obviously, taxonomists note, absorb, and mainly are genera that are nocturnal and in utilize these refinements and modifications with- which there is little sexual dimorphism. Vision out undue stress, confusion, or dissension. A probably plays a minor role in species-identifica- desire to maintain the status quo in the face of tion or social behavior in these, and in the ab- new evidence and analysis is unscientific and un- sence of marked external differences, taxon- professional. omists have tended not to split them into genera In the absence of reproductive data, most as readily as they have divided other groups. genera will continue to be distinguished on mor- I have avoided consideration of how taxon- phological evidence. It is noteworthy that, in omists of groups other than mammals deal with estimating the "gap" between species or species- intergeneric hybrids or the level of the genus. groups in formulating genera, excessive attention However, it is noteworthy that many mammalo- seems to have been paid to sexual dimorphism gists regard the taxonomy of -a mainly and conspicuous external features. If taxon- diurnal and extensively dimorphic class-as being omists have consciously or subconsciously over-divided at higher categories. Naturally guessed whether or not they thought the amount occurring intergeneric hybrids are far more of morphological difference they observed was common in birds than in mammals (Gray, 1958; beyond the level of reproductive compatibility, Cockrum, 1952; and Short, 1969). they seem to have erred particularly in the highly A lingering problem that remains from the use 1 977 VAN GELDER: HYBRIDS 21 of data on reproductive compatibility to delimit From the evolutionary viewpoint, when genera is the status of subgenera. By strict inter- species are grouped into a genus they are usually pretation, subgenera as well as genera should not considered to have reached a new adaptive zone. be capable of hybridization, if the species re- From the presumed isolation provided by specia- mains as the taxon distinguished by reproductive tion, a species can then, without further genetic isolation. The utility of the subgenus is unques- mixture with its near relatives, evolve to exploit a tioned, however, as a formal expression of new environment or niche. When the morpho- morphologically similar and related species. logical changes in adaptation to this new field There are other terms, of course, such as ailo- reach a stage where the appearance or difference species, superspecies, or species group which, in is greater than between most of the other species one form or another, express the idea of a closer in the group, the species may be considered to be proximation of the included taxa to one another a genus. than to other species or groups within the same In dealing with sympatric or contiguous genus. It seems appropriate now to apply the re- species that do not interbreed in nature but do striction of reproductive isolation to subgenera, hybridize (even with fertile offspring) in artificial as well, even though this will undoubtedly make situations, such evidence is construed as further more difficult the transition of the taxonomic support for the specific validity of each of the changes proposed here. I have already stated that species. Allopatric or allochronous species that I believe that a genus, by definition, should in- can be hybridized with fertile offspring (under clude species that are incapable of breeding with artificial conditions, of course) are generally con- those of any other genus. We already have the sidered to represent only subspecies because they criterion that offspring, if any, of crosses be- are potentially capable of interbreeding. Most of tween species should be sterile or have greatly the mammals that had been considered genera reduced fertility. These definitions leave little but were known to hybridize in nature have gen- space for an objective definition of other cate- erally by now been relegated to congeneric gories in the interstices between the level of the status. Genera that are capable of hybridization species and that of the genus. and the production of a living offspring under In actual practice, the taxonomic changes sug- artificial conditions form a case that is not unlike gested here that are determined by reproductive that of allopatric species or even sympatric data are not novel. Most of them have been pro- species that do not interbreed in nature, but do posed by other taxonomists in the past, largely in captivity. For intergeneric hybrids, fertility of on morphological similarities. Resistance to their the offspring need not be a criterion for evalua- acceptance has been based less on the taxonomic tion of the status of the taxa. Sterility would in- or evolutionary consideration than on the desire dicate only that the parents of an intergeneric to retain a familiar appellation, especially in non- cross are good species; fertility would suggest taxonomic fields. The retention of Thalarctos for that they are conspecific. I have somewhat arbi- the polar bear, and of Mandrillus for the man- trarily selected the production of a living offspring drills in much of the popular and applied litera- as the criterion for acceptability in an intergeneric ture are cases in point. cross. It seems to me that if the chromosomes The definition of the genus has received far of two taxa are compatible enough to develop a less attention than that of the species. Although fetus to term, then the parents would seem to be chapters and even volumes are devoted to the more closely related than generic separation concept of the species, the genus has usually re- would suggest. The viability of the offspring after ceived little more than a few lines or a paragraph birth has not been considered, partially because or two. Winge (1941) is one of the few mam- there is no way of determining how survival malogists who attempted to consider the genus as might be related to the animal's genetics. Survival a uniform taxonomic level within the class, and of captive-born animals is often low, even when interestingly, his genera often anticipate the data intertaxa crosses are not involved. from intergeneric hybrids. He did not, however, The relationships of reproductive capability to define the genus. taxonomic levels, as I construe them, are: 22 AMERICAN MUSEUM NOVITATES NO. 2635

Taxon Can interbreed Can interbreed Cannot brid duiker. 18th Rept. Direct. Veteri- and pro- but produce inter- nary Serv. Animal Indust., S. Africa, duce fer- sterile off- breed part 2, pp. 1093-1094. N. V. tile off- spring Blackwelder, Richard E. spring 1967. Taxonomy. A text and reference book. New York, John Wiley & Sons, Inc. Subspecies Yes No No Bobrinskii, N., B. Kuznetzov, and A. Kuzyakin Species, 1944. Mammals of U.S.S.R. Moscow, Soviet- Sympatric No Some Yes skaya Nauka [In Russian]. Species, Bolineau, P. Allopatric No Some Yes 1933. Hybridations d'antilopides. Terre et Vie, Species, vol. 3, pp. 690-691. Sibling No ? Yes Buettner-Janusch, John Genera No No Yes 1966. A problem in evolutionary systematics: nomenclature and classification of baboons, genus Papio. Folia Primatol., LITERATURE CITED vol. 4, pp. 288-308. Ackermann, Karl Cabrera, Angel, and Jose Yepes 1898. Thierbastarde. Abhandl. Ber. 43, Ver. f. 1940. Mamiferos sud-americanos (vida, cos- Naturk. zu Kassel, vol. 62, pp. 38-79. tumbres y descripcion). Buenos Aires, Allen, Glover M. Compania Argentina de Editores. 1939. A checklist of African mammals. Bull. Calaby, J. H., and W. E. Poole Mus. Comp. Zool., Harvard College, 1971. Keeping kangaroos in captivity. Inter- vol. 83, pp. 1-763. natl. Zoo Yearbook, vol. 11, pp. 5-12. Anderson, Sydney Caton, John Dean 1967. Primates. In Anderson, Sydney, and J. 1877. The antelopes and deer of America. Knox Jones, Jr. (eds.), Recent mam- New York, Hurd and Houghton. mals of the world. A synopsis of Chiarelli, B. families. New York, Ronald Press Co., 1961. Ibridolgia e systematica in Primati I. pp. 151-177. Raccolta di data. Atti Assoc. Genetica Anderson, Sydney, and J. Knox Jones, Jr. (eds.) Italhana, vol. 6, pp. 213-220. 1967. Recent mammals of the world. A syn- Clutton-Brock, Juliet, Gordon B. Corbet, and opsis of families. New York, Ronald Michael Hills Press Co. 1976. A review of the Family Canidae, with a Anon. classification by numerical methods. 1931. Moose-elk in Montana. California Fish Bull. British Museum (Nat. Hist.), and Game, vol. 17, p. 198. Zool., vol. 29, pp. 117-199. 1965. [Duiker hybrid (Cephalophus nigrifrons Cockrum, E. Lendell x Sylvicapra grimmia), Pretoria, S. A. ]. 1952. A check-list and bibliography of hybrid Internatl. Zoo Yearbook, vol. 5, p. birds in North America north of 347. Mexico. Wilson Bull. vol. 64, pp. 1975. [Coyote/Fox hybrid]. Internatl. Zoo 140-159. Yearbook, vol. 15, p. 372. Dandelot, P. Ansell, W. F. H. 1971. Order Primates. In Meester, J., and H. 1971. Artiodactyla. In Meester, J., and H. W. W. Setzer (eds.), The mammals of Setzer (eds.), The mammals of Africa. Africa. An identification manual. Wash- An identification manual. Washington, ington, D. C., Smithsonian Institution D.C., Smithsonian Institution Press, Press, part 3, pp. 1-45. part 15, pp. 1-84. Davis, D. H. S. Asakarua, Shigenharu 1971. Genus Aethomys. In Meester, J., and H. 1969. A note on a bear hybrid, Melursus W. Setzer (eds.) The mammals of ursinus X Helarctos malayanus, at Africa. An identification manual. Wash- Tama Zoo, Tokyo. Internatl. Zoo Year- ington, D. C., Smithsonian Institution book, vol. 9, p. 88. Press, part 6.6, pp. 1-5. Bigalke, R. Delson, E., and P. H. Napier 1932. Anatomical studies, no. 37: On a hy- 1976. Request for the determination of the 1977 VAN GELDER: HYBRIDS 23

generic names of the baboon and man- bibliography. Commonwealth Bur. drill (Mammalia: Primates, Cercopithe- Animal Breeding and Genet., Edin- cidae). Z.N.(S.) 2093. Bull. Zool. burgh, Tech. Commun. no. 10 (re- Nomenclature, vol. 33, pp. 46-60. vised), pp. 1-262. Dohl, Thomas P., Kenneth S. Norris, and Ingrid Gunning, J. W. B. Kang 1901. Experiments in hybridity at Pretoria. 1974. A porpoise hybrid, Tursiops X Steno. Zoologist, vol. 5 (ser. 4, no. 57), no. Jour. ., vol. 55, pp. 217-222. 723, p. 321. Ellerman, J. R. Haltenorth, Th. 1941. The families and genera of living ro- 1963. Klassifikation der Saugetiere: Artio- dents. London, British Museum (Nat- dactyla, vol. 1, no. 18, pp. 1-167. ural History). Handbuch der Zoologie, vol. 8, no. 32. Ellerman, J. R., and T. C. S. Morrison-Scott Berlin, Walter de Gruyter and Co. 1951. Checklist of Palaearctic and Indian Hershkovitz, Philip mammals 1758 to 1946. London, 1961. On the South American small-eared British Museum (Natural History). zorro A telocynus microtis Sclater Ellerman, J. R., T. C. S. Morrison-Scott, (Canidae). Fieldiana: Zool. vol. 39, pp. and R. W. Hayman 505-523. 1953. Southern African mammals 1758 to 1966. Catalog of living whales. Bull. U. S. 195 1: a reclassification. London, Brit- Natl. Mus., vol. 246, pp. 1-259. ish Museum (Natural History). Jorge, W., Sandra Butler, and K. Benirschke Flerov, K. K. 1976. Studies on a male eland x kudu hybrid. 1952. Musk deer and deer. Fauna of U.S.S.R., Jour. Reprod. Fert., vol. 46, pp. 13-16. vol. 1, no. 2, pp. 1-257. Moscow, Acad- Kettlitz, W. K. emy of Sciences U.S.S.R. (translation 1967. The blesbok (Damaliscus dorcas phil- by A. Biron and Z. S. Cole, 1960, lipsi) with special reference to the herd Office Tech. Serv., U. S. Dept. Com- in the Percy Fyfe Nature Reserve. merce, Washington, D. C.). Fauna and Flora, Pretoria, no. 18, pp. Fooden, Jack 36-46. 1976. Provisional classification and key to liv- Koopman, Karl F. ing species of macaques (Primates: 1967. Artiodactyls. In Anderson, Sydney, and Macaca). Folia Primatol., vol. 25, pp. J. Knox Jones, Jr., (eds.), Recent mam- 225-236. mals of the world. A synopsis of Fraser, F. C. families. New York, Ronald Press Co., 1940. Three anomalous dolphins from Black- pp. 385-406. sod Bay, Ireland. Proc. Royal Irish Koopman, Karl F., and J. Knox Jones, Jr. Acad., sec. B. vol. 45, pp. 413-455. 1970. Classification of bats. In Slaughter, Bob 1966. Comments on the Delphinoidea. In H., and Dan W. Walton (eds.), About Norris, K. S. (ed.), Whales, dolphins bats. Fondren Science Series II, Dallas, and porpoises. Berkeley, Univ. Cali- Southern Methodist Univ. Press, pp. fornia Press, pp. 7-31. 22-28. Gray, Annie P. Koulischer, L., J. Tijskens, and J. Mortelmans 1954. Mammalian hybrids. A check-list with 1973. Chromosome studies of a fertile mam- bibliography. Commonwealth Bur. malian hybrid: the offspring of the Animal Breeding and Genet., Edin- cross bongo x sitatunga (Bovoidea). burgh, Tech. Commun. no. 10, pp. Chromosoma (Berl.), vol. 41, pp. 1-144. 265-270. 1958. hybrids. A check list with bibliog- Krieg, Hans raphy. Commonwealth Bur. Animal 1925. Biologische Reisestudien in Sudam- Breeding and Genet. Edinburgh, Tech. erika. VII. Notiz uber einen bastard Comm. no. 13, pp. 1-390. zwischen hund und pampafuchs 1966. Mammalian hybrids. Supplementary (Pseudalopex (Canis) agarae), nebst bibliography of Tech. Commun. no. 10, bemerkungen fiber die systematik Commonwealth Bur. Animal Breeding der argentinisch-chilensichen fiichse. and Genet., Edinburgh, pp. 1-33. Zeitschr. Morph. Okol. Tiere, vol. 4, 1972. Mammalian hybrids. A check-list with pp. 702-710. 24 AMERICAN MUSEUM NOVITATES NO. 2635

Langguth, Alfredo mit einer ausfiurlichen Erklarung, vol. 1. 1970. Una nueva clasificacion de los canidos Nurnberg, Gabriel Nicolaus Raspe. sudamericanos. Actas IV Congreso 1776. Des Ritters Carl von Linne, koniglish Latinoamericano de Zoologia, vol. 1, Schwedischen leibarztes, volistandigen pp. 129-143. Natursystems, Supplements -und Regis- 1975. Ecology and evolution in the South ter-Band uber alle sechs Theile oder American canids. In Fox, Michael W. Classen des Thierreichs. Nurnberg: (ed.), The wild canids: their system- Gabriel Nicolaus Raspe. atics, behavioral ecology and evolution. Nadler, C. G., R. S. Hoffmann, and A. Woolf New York, Van Nostrand Reinhold, pp. 1974. G-band patterns, chromosomal homol- 192-206. ogies, and evolutionary relationships Lantz, David E. among wild sheep, goats, and aoudads 1910. Raising deer and other large game ani- (Mammalia, Artiodactyla). Experientia, mals in the United States. U. S. Dept. vol. 30, pp. 744-746. Agr. Biol. Surv. Bull. no. 36, pp. 1-62. Napier, J. R., and P. H. Napier Lydekker, R., and Gilbert Blaine 1967. A handbook of living primates. Mor- 1914. Catalogue of the mammals in phology, ecology and behavior of non- the British Museum (Natural History), human primates. New York, Academic vol. 3, pp. 1-295. London, Trustees of Press. the British Museum. Pocock, R. I. Manwell, C., and C. M. Ann Baker 1941. The fauna of British India, including 1975. Ammotragus lervia: progenitor of the Ceylon and Burma. Mammalia. vol. II. domesticated sheep or specialized off- London, Taylor and Francis, Ltd. shoot of Caprine evolution? Experientia, Przibram, Hans vol. 31, pp. 1370-1371. 1910. IV Bastardierung. Experimental-Zool- Mayr, Ernst ogie, Leipzig, Franz Deuticke, vol. 3. 1942. Systematics and the origin of species Rich, Dale W. from the viewpoint of a zoologist. New 1967. Cetaceans. In Anderson, Sydney, and J. York, Columbia Univ. Press. Knox Jones, Jr. (eds.). Recent mam- Mayr, Ernst, E. Gorton Linsley, and Robert L. mals of the world. A synopsis of Usinger families. New York, Ronald Press Co., 1953. Methods and principles of systematic pp. 291-324. zoology. New York, McGraw-Hill Book Rice, D. W., and V. B. Scheffer Co., Inc. 1968. A list of the marine mammals of the Mead, James G. world. U.S. Fish Wildlife Serv., Spec. 1975. Anatomy of the external nasal passages Sci. Rept., Fisheries, no. 579, pp. 1-16. and facial complex in the Delphinidae Richardson, B. J., A. B. Czuppon, and G. B. (Mammalia: Cetacea). Smithsonian Sharman Contrib. Zool. vol. 207, pp. 1-72. 1971. Inheritance of glucose-6-phosphate de- Misonne, Xavier hydrogenase variation in kangaroos. 1969. African and Indo-Australian Muridae. Nature, vol. 230, pp. 154-155. Evolutionary trends. Ann. Mus. Roy. Ride, W. D. L. Afrique Central, Ser. 8, no. 172, pp. 1970. A guide to the native mammals of Aus- 1-219. tralia. Melbourne, Oxford Univ. Press. 1971. Order Rodentia, main text. In Meester, Rorig, Adolph J., and H. W. Setzer (eds.). The mam- 1903. Ueber Saugetier-Bastarde. Zoologische mals of Africa. An identification Garten, vol. 44, pp. 247-254. manual. Washington, D. C. Smithsonian Ross, Herbert H. Institution Press, part 6, pp. 1-39. 1975. Biological systematics. Reading, Mass., Miuller, Philipp Ludwig Statius Addison-Wesley Publishing Co. 1773. Des Ritters Carl von Linne, koniglich Sanjeeva Raj, P. J. Schwedischen leibarztes, vollstandiges 1953. A cross between the domestic dog Natursystem nach der zw6lften latein- (Canis familiaris) and the fox (Vulpes ischen Ausgabe und nach Anleitung des bengalensis). Jour. Bombay Nat. Hist. hollandischen houttuynischen Werks Soc., vol. 51, pp. 724-726. 1977 VAN GELDER: HYBRIDS 25

Schaefer, C. W. eny. Jour. Theoretic. Biol., vol. 26, pp. 1976. The reality of the higher taxonomic 445-480. categories. Zeitschr. Zool. Syst. Van Deusen, Hobart M., and J. Knox Jones, Jr. Evolut.-forsch., vol. 14, pp. 1-10. 1967. Marsupials. In Anderson, Sydney, and Scheffer, Victor B. J. Knox Jones, Jr. (eds.), Recent mam- 1958. Seals, sea lions, and walruses. A review mals of the world. A synopsis of of the pinnipedia. Stanford, Stanford families. New York, Ronald Press Co., Univ. Press. pp. 61-86. Selous, F. C. Van Gelder, Richard G. 1893. Exhibition of, and remarks upon, the 1969. Biology of mammals. New York, skull of an antelope, believed to be a Charles Scribner's Sons. hybrid between the sassaby and the 1975. [Comments on Pan and Panthera]. Bull. hartebeest. Proc. Zool. Soc. London for Zool. Nomenclature, vol. 32, no. 2, pp. 1893, pp. 1-2. 69-73. Short, Lester L. 1977. An eland X kudu hybrid, and the 1969. Taxonomic aspects of avian hybridiza- content of the genus Tragelaphus. Lam- tion. Auk, vol. 86, pp. 84-105. mergeyer, no. 23, pp. 1-6. Simpson, George Gaylord Wetzel, Ralph M. 1945. The principles of classification and a 1976. The Chacoan peccary, Catagonus wag- classification of mammals. Bull. Amer. neri (Rusconi). Abstract Papers Pre- Mus. Nat. Hist., vol. 85, pp. 1-350. sented 56th Ann. Meeting, Amer. Soc. 1961. Principles of animal taxonomy. New Mammal., Lubbock Texas, p. 19. York, Columbia Univ. Press. Wetzel, Ralph M., Robert E. Dubos, Robert L. Stains, Howard J. Martin, and Philip Myers 1967. Carnivores and pinnipeds. In Anderson, 1975. Catagonus, an 'extinct' peccary, alive in Sydney, and J. Knox Jones, Jr. (eds.), Paraguay. Science, vol. 189, pp. 379- Recent mammals of the world. A syn- 381. opsis of families. New York, Ronald Winge, Herluf Press Co., pp. 325-354. 1941. The interrelationships of the mam- Thenius, E. malian genera. Copenhagen, C. A. 1976. Zur stammesgeschichtlichen Herkunst Reitzels Forlag, 3 vol. (Translated by E. von Tremarctos (Ursidae, Mammalia). Deichmann and G. M. Allen). Zeitschr. Saugetierk. vol. 41, pp. Woodburne, Michael 0. 109-114. 1968. The cranial myology and osteology of Tijskens, J. Dicotyles tajacu, the collared peccary, 1968. Preliminary notes on the F1 bongo an- and its bearing on classification. Mem. telopes x sitatunga hybrids Taurotragus Southern Calif. Acad. Sci., vol. 7, pp. eurycerus X Thagelaphus spekei. Inter- 1-48. natl. Zoo Yearbook, vol. 8, pp. 137- Youngman, Phillip M. 139. 1975. Mammals of the Yukon Territory. Natl. Todd, N. B. Mus. Nat. Sci., Natl. Mus. Canada, Publ. 1970. Karyotypic fissioning and canid phylog- Zool., no. 10, pp. 1-192.