on the origin of allopatric primate species · structure in most primate societies, in particular...

Biodiversity Journal, 2016, 7 (1): 117–198

On the origin of allopatric primate species

Marc G.M. van Roosmalen* & Tomas van Roosmalen

¹MVRS - Marc van Roosmalen Stichting, Leiden, The Netherlands*Corresponding author, e-mail: [email protected]

ABSTRACT Here we present a theory on the origin of allopatric primate species that follows - at least inNeotropical primates - the irreversible trend to albinotic skin and coat color, called “meta-chromic bleaching”. It explains why primates constitute such an exceptionally diverse,species-rich, and colorful Order in the Class Mammalia. The theory is in tune with theprinciple of evolutionary change in tegumentary colors called “metachromism”, a hypothesispropounded by the late Philip Hershkovitz. Metachromism holds the evolutionary change inhair, skin, and eye melanins following an orderly and irreversible sequence that ends in lossof pigment becoming albinotic, cream to silvery or white. In about all extant sociable Neo-tropical monkeys we identified an irreversible trend according to which metachromic varietiesdepart from the saturated eumelanin (agouti, black or blackish brown) archetypic form andthen speciate into allopatric taxa following the trend to albinotic skin and coat color. Speci-ation goes either along the eumelanin pathway (from gray to silvery to cream to white), orthe pheomelanin pathway (from red to orange to yellow to white), or a combination of thetwo. The theory represents a new and original evolutionary concept that seems to act indef-initely in a non-adaptive way in the population dynamics of male-hierarchic societies of allsociable primates that defend a common territory. We have successfully tested the theory inall 19 extant Neotropical monkey genera. Our theory suggests the trend to allopatry amongmetachromic varieties in a social group or population to be the principal behavioral factorthat empowers metachromic processes in sociable Neotropical monkeys. It may well representthe principal mechanism behind speciation, radiation, niche separation, and phylogeographyin all sociable primates that hold male-defended territories. We urge field biologists who studyprimate distributions, demography and phylogeography in the Old World to take our theoryto the test in the equally colorful Catarrhini.

KEY WORDS Neotropical primates; phylogeography; metachromic bleaching; speciation; radiation.

Received 20.12.2015; accepted 09.02.2016; printed 30.03.2016

MONOGRAPH

INTRODUCTION

We could ask ourselves (Darwin, 1859): “Whyprimates constitute by far the most diverse, species-rich and colorful Order in the Class Mammalia?Do primate diversity, metachromism and meta-chromic processes relate directly to sexual selection?Or, rather to its generally complex, hierarchically

organized social structure and male territoriality?If not sexual selection, what could be the principalfactor(s) in primate social behavior to be heldresponsible for metachromic processes, speciation,radiation, niche separation, and phylogeography?”Inspired by Alfred Russel Wallace whose concept

of the “Origin of Species” was laid down in a paper

he sent for review to Charles Darwin, here weintroduce a new and original theory about speciesevolution taking place in particular in sociableterritorial primates. Our theory “On the tendency ofmetachromic varieties in sociable primates todepart indefinitely from the agouti archetype andevolve in advanced eumelanin, pheomelanin toalbinotic bleached allopatric taxa” is equally rootedin life-long fieldwork on socio-ecology of all Neo-tropical monkey genera, both in captivity and in thewild. It closely follows the principle of evolutionarychange in tegumentary colors called “meta-chromism”, a hypothesis propounded by PhilipHershkovitz (1968; 1977). Metachromism holds theevolutionary change in hair, skin, and eye melaninsfollowing an orderly and irreversible sequence thatends in loss of pigment through which a taxon of agiven genus or phylogenetic clade eventually be-comes albinotic, cream to white. Individual haircolor or the entire coat changes from agouti (char-acterized by alternating blackish-brown and reddishbands on the terminal half of the hair) to uniformlyblackish-brown, and thereafter to gray, and even-tually to white or colorless, called the eumelaninpathway; or, it changes from agouti to uniformlyreddish to orange to yellow to cream, and even-tually white, called the pheomelanin pathway. Theprocess itself is called saturation, which means thechange from the primitive agouti pattern of the hair,or part of the pelage, or the entire coat, to a satur-ated eumelanin (blackish) or saturated pheomelanin(reddish) coloration. The dilution, or gradual reduc-tion in the amount of pigment deposited in thegrowing hair, and disappearance of pigmentarycolors is called bleaching (Fig. 1). In the color ofthe skin and iris of the eye, it follows the eumelaninpathway (brown to drab, to gray, or blue), and thenit is termed depigmentation. Metachromism appliesto all mammalian species. It is thought to also occurin bird feathers.Our theory suggests that among social groups

or populations of advanced intelligent, sociallyorganized, male-territorial mammals, in particularprimates, phenotypical varieties (mutants) thatshow slightly bleached eumelanin or pheomelanincolored skin or coat characteristics do arise indef-initely. Their melanocytes (skin cells that producethe black pigment melanin) are smaller, and for thator any other reason produce less melanin. In gen-eral, the tendency of these metachromic varieties is

neotonic, taking place locally (e.g., naked muzzle,bald head, euchromic blaze/forehead or part of thecoat, depilation of skin) or all over the body. Socialstructure in most primate societies, in particularthose of the more advanced monkeys and apes, ishierarchically organized, whereas male over femaledominance is the rule, with very few exceptions(e.g., spider monkeys in the Neotropics and pygmychimpanzees in Central Africa adopted a matri-archal social system, in which males patrol anddefend a common territory, and females are allowedto transfer to neighboring social groups). Socialselection is the recognition of and preference for theparental (or foster parental) phenotype in societalgrouping and mating. Social selection for color orcolor pattern through assortative mating tends tostabilize within a chromatic range recognized andaccepted by free-ranging but chromotypicallyimprinted members of the social group. Slightlydepilated or somewhat eumelanin or pheomelaninbleached individuals deviant from the sociallyselected skin or hair color pattern, in particularwhen it is detected in adolescent to subadult males,may be discriminated against by high-ranking(alpha)-males. For that reason alone they can bepushed into the periphery of the group. Dependingon the primate taxon or genus, such individualyoung males may also be expelled from the parental

MARC G.M. VAN ROOSMALEN & TOMAS VAN ROOSMALEN118

Figure 1. Bleaching from saturated eumelanin and saturatedpheomelanin fields to white or colorless. Gradual reductionin the amount of pigment deposited in the growing hair re-sults in apparent change from blackish through brown, drabgray to white or colorless in the eumelanin pathway, andfrom reddish through orange, yellow, cream to white in thepheomelanin pathway. Switching from the eumelanin to thepheomelanin pathway occurs in saturation but not in blea-ching (modified from Hershkovitz, 1977).

group, and then become social ‘outcasts’. Peri-pheral or outcast males do suffer on a daily basefrom less and shorter access to the group’s prefer-ential, comparatively more nutritious food sources.They may join one another for reasons of socialcomfort and during ranging or foraging they tendto hang out together at the periphery of the group.Eventually, they may decide to leave the pack asall-male parties and roam around in much largerareas than just the home range or territory of theparental group. They then may attract young fe-males from neighboring social groups. Together,they may seek some hitherto overlooked, ‘empty’or little-used living space in an attempt to settledown and start their own family or social group. Incase the taxon or genus it belongs to shows ter-ritorial behavior - which is the case in almost allNeotropical monkeys - these emigrants will be sub-sequently pushed out from neighboring territoriesas well. Consequently, they will die from starvation,parasite load and/or diseases forthcoming thedietary constraints they are suffering from. Or, as amatter of luck, in the end they may find some livingspace that is not (yet) occupied, most likely atconsiderable distance from the taxon’s core distri-bution. Sometimes, such emigrant parties can beforced to survive in peripheral habitat that has to beconsidered marginal for that species to occur in.In extremely rare cases, such parties might evenmanage to circumvent a certain geographical barrierand beyond it find for the species appropriate hab-itat, where their specific ecological niche is notoccupied, as such involving a range extension. Incase that habitat is already occupied by a closelyrelated species, a battle for life will take place andthe best fitted taxon will drive the other to extinc-tion, the red-handed tamarin Saguinus midas that isreplacing the bare-face tamarin S. bicolor. Onlyover geological spans of time, for example after avicariance has taken place, suitable habitat mayopen up where the taxon’s niche is not occupied byanother primate. One may imagine that along thesepaths small reproductively isolated founder-colon-ies that contain somewhat bleached and/or depilatedindividuals may establish themselves there. For thesake of survival alone they would unselectivelyinterbreed or hybridize. Inbreeding then may relaxstabilizing forces and stimulate or accelerate meta-chromic and other degenerative (= non-adaptive)processes. The more metachromic advanced each

119On the origin of allopatric primate species

successively isolated breeding colony is, or thefarther it has moved from the center of that taxon’sdispersion, the nearer it will come to the end of itsmetachromic evolution. And, the narrower will beits range of chromic fitness (e.g., prey and predatorcamouflage). This degenerative process, though,may be counterbalanced if under strong naturalselection newly diverged forms, that evolve in a, forthe original species marginal or new habitat, nicheor landscape, at the same time selectively becomebetter fitted, more cooperative, more inventive, orsmarter in the adaptation process. This may happenevery time founder-colonies successfully travelacross existing geographical barriers, such as rivers,watersheds, mountain ranges, or open areas witharid scrub vegetation. Completion of the processesof metachromic bleaching, depigmentation, or de-pilation, whether taking place single or combined,eventually will result in extinction of the race orspecies. Unless the founder-colony or population intime does find and manage to occupy hithertoempty, but suitable habitat. Or: if it adapts to a dif-ferent ecological niche, where skin and coat colordo not have survival value by lack of competitionfrom closely related species. Dead-end, isolated,peripheral, or new habitats may be occupied bymetachromic dead-end populations, such as hashappened over and over in the Neotropics in ad-vanced albinotic callitrichids, uakaris, sakis, titis,capuchins, howlers, spider, woolly, and woollyspider monkeys.

RESULTS

In non-territorial, peaceable Dwarf MarmosetsCallibella humilis neotony and euchromism areclearly demonstrated as infants are overall muchlighter colored than adults, showing a tendency toalbinotic. Their overall pelage is light brown, theirtail alternately light and dark-brown banded, andtheir face flesh-colored with a circumference oflong, bright white hairs. From three months on, theypass through a complete metachromic metamorph-osis. Their overall coat turns into saturated eu-melanin, the muzzle of their faces into pinkish, andtheir semicrescent ocular rings or eyebrows intowhite (Fig. 2). This natural process may be relatedto slightly smaller melanocytes (skin cells thatproduce the black pigment melanin) producingoverall less melanin.

Callibella stands at the base of the phylogenetictree from which all extant Amazonian marmosets,Cebuella and Mico, have derived (Van Roosmalen& Van Roosmalen, 2003). It finds itself at the vergeof extinction, for it occupies the niche of exudategouging - that is feeding on resins ouzing out oflittle holes they themselves have gnawed in the barkof certain gum trees and climbers. That niche isfilled in by the advanced, larger, highly territorialAmazonian marmoset genus Mico (its distributionsare shown in figures 5–7). We believe that theseaggressive, over twice as big callitrichid monkeyshave displaced the non-territorial dwarf marmosetand taken over its specific feeding niche all overits former, much larger range - the entire interfluvedelineated by the Rios Madeira, Amazonas andTapajós. The genus Callibella is thought to haveevolved there in the late-Pliocene to early-Pleisto-

cene landscape that was dominated by lacustrineseasonally inundated clear-water igapó wetlands.Being peaceable monkeys that like their neighbor’scompany instead of attacking or trying to kill themapparently has not been an evolutionary successamong primates (Van Roosmalen, 2013a, b; 2015).Contrastingly, the pygmy marmoset Cebuella

that derived from prototypic Callibella nowadaysoccupies the entire western Amazon Basin. Webelieve it was so successful because Cebuella,being allopatric with Callibella and Mico, couldoccupy the ecological feeding niche of exudategouger west of the Rio Madeira. There, it did nothave to face competition from other callitrichidsover exudate food sources. Indeed, Amazoniantamarins (genus Saguinus) that range west of theRio Madeira (Figs. 7–13) lack the elongated tuskedmandibular second incisors needed for tapping sap

Figure 2. Ontogeny in Black-crowned Dwarf Marmosets Callibella humilis Van Roosmalen & Van Roosmalen (2003) based on photos of captive and wild individuals (Van Roosmalen et al., 1998).


121

from tree barks. As such, tamarins do not directlycompete with pygmy marmosets over gum.

Instead, tamarins of different taxa are reportedto parasitize on pygmy marmosets by licking theresins from tap holes made by the latter.

Cebuella pygmaea being overall agouti coloredis clearly the most archetypic among the two extanttaxa of pygmy marmoset. Distributed north of theUpper Amazon River (Rio Solimões/Río Marañón)and specialized in exudate gouging, the species (orits precursor) seems to have adapted to seasonallywhite-water inundated floodplain forest (várzea)habitat. Somewhat pheomelanin bleached colon-izers of ancestral C. pygmaea (having an orangecolored tail and breast, progressively bleached yel-low to white belly, yellow-white mustache, nakedpink-colored muzzle and circumocular rings) fol-lowing the trend to allopatry once must have man-

aged to traverse the Amazon River proper, on floa-ting várzea islands and/or passively through ri-verbend cut-offs (oxbow lakes). By lack ofcompetitors the nearest to albinotic taxon Cebuellaniveiventris - the form that derived from allopatricarchetypic C. pygmaea - could then have extendedits range from the Rios Javarí and Juruá east as faras the Rio Madeira and south of the Amazon Riveras far as the Bolivian Amazon (Fig. 3). There, itsecondarily adapted to never inundating terra firmehigh forest. Nowadays, it is found there, especiallyat edges of treefall clearings and in secondarygrowth. As C. niveiventris is fully allopatric with C.pygmaea and, moreover, shows completely dif-ferent habitat preferences, we here propose to at-tribute both taxa full-species status naming themC. pygmaea and C. niveiventris. During our system-atic surveys of primate distributions and diversity

Figure 3. Present-day distributions are here depicted for Black-crowned Dwarf Marmosets genus Callibella and PygmyMarmosets genus Cebuella, representing the smallest monkeys in the world. The current distribution of the monotypic genusCallibella perhaps has to be considered the smallest of any primate in the Neotropics.


carried out in the matrix terra firme hinterlandstretching out behind the floodplain of white-waterrivers (i.e., the Rios Javarí, Juruá, Purús andMadeira), we were not able to detect any pheno-typical difference between individuals sightedat any point along these far-apart rivers. It mayindicate that in highly territorial monkeys likepygmy marmosets that occupy large distributionsdelineated by some of the largest tributaries of theAmazon River, phenotypical characters of skin andpelage coloration and/or local hair growth or de-pilation seem to have stabilized. In other words, webelieve that within a given monkey’s distributionsomething like a gradient of slightly differingphenotypes or color morphs, or geographic races,in reality does not exist. These and other observa-tions from the larger field have led us to attributefull-species status to monkey taxa such as C.pygmaea and C. niveiventris that we ourselves haveconfirmed to be phenotypically stable throughouttheir (sometimes very large) range. Here, we would like to propose a new species

concept: ecospecies. This species concept is furthercorroborated by the here introduced evolutiontheory that aims to explain the origin of allopatricprimate species. We define ecospecies as follows:“An ecospecies is a genetically isolated populationor group of populations of a kind that does notundergo any gene flow from other populations ofone or more closely related kinds, and that demon-strates a stabilized, well-defined phenotype over itsentire range, in which it occupies and defends aspecific ecological feeding niche against any out-side competitor”. This definition of a primatespecies avoids the confusing, rather arbitrary dis-tinction between species and subspecies (or race),for it adds sociobiological factors to geographical,geomorphological and phytosociological ones thatact on the evolutionary process of primate speci-ation and radiation. Following this concept, for in-stance, an enclave population of Callibella humilisthat we found living year-round in the seasonallyinundated floodplain forest (igapó) along bothbanks of the Rio Atininga - genetically isolatedfrom the main population occuring in terra firmeforest at least one-hundred km to the north - shouldbe given its own species name and treated as such.Or, in case the ranges of two saddle-back tamarinsof the Saguinus fuscicollis Clade, hitherto taxonom-ically treated as subspecies, are separated by a con-

tact zone, where territorial behavior effectively im-pedes gene flow through hybridizing, both popula-tions should be attributed full-species status.The callitrichids Goeldi’s Monkey Callimico

and Black-crowned Dwarf Marmoset Callibella dorepresent the only monospecific (= monotypic)primate genera in the Neotropics. Callimico livesin the upper Amazon Basin region of Bolivia,Brazil, Peru, Colombia, and Ecuador (Fig. 4).Goeldi’s monkey coat coloration is saturated eu-melanin, blackish or blackish-brown. It forages indense scrubby undergrowth of low mixed forestswith discontinuous canopies and in so-called‘tabocais’ (low forest dominated by bamboo) atlevels of less than five meters. Social groups consist-ing of monogamous pairs with single offspringcount on average six individuals. Groups live inpatches of suitable habitat, often separated by milesof unsuitable vegetation. Goeldi’s monkeys arevertical clingers and leapers able to leap horizontaldistances of up to four meters between branches. Asthey are peaceable monkeys not showing any formof territoriality, Goeldi’s monkeys often associatein mixed species groups with different species oftamarin Saguinus (Mittermeier et al., 2013). Thefact that this primitive little monkey, just like thedwarf marmoset Callibella, remained archetypic inits blackish agouti coat coloration, is peaceable, isnot showing any territorial behavior towards itsneighbors, is occupying a unique feeding niche(foraging on the ground for fungi and invertebrates,and for fruits at low levels of a discontinuouscanopy), and over geological time-span did notdiverge into more than one taxon, strongly supportsour doctrine that attributes speciation and radiationin male-territorial Neotropical primates primarilyto the trend to allopatry as expressed in meta-chromic bleaching.As shown in the schematic distribution map of

all known Amazonian marmosets (Fig. 5), each in-terfluve in the area delineated by the most effectiveriverine barriers - Rio Amazonas in the north, RioMadeira in the west, Rio Guaporé in the south, andRios Tapajós-Juruena and Xingú in the east- isinhabited by a different taxon of Mico, whichspecies phylogeographically and phylogeneticallyradiated away from an ancestral, archetypic agouti-colored form much resembling the extant speciesM. melanurus (Van Roosmalen et al., 2000) fromthe upper Rio Aripuanã basin - the taxon with the

122 MARC G.M. VAN ROOSMALEN & TOMAS VAN ROOSMALEN

southernmost distribution of all Amazonian mar-mosets to be placed at the base of Mico’s phylo-genetic tree.Four monophyletic cladistic Groups or Clades

are distinguished: the Bare-ear M. argentatus Clade,the (Tufted-ear or) Tassel-ear M. humeralifer Clade,the White-mantle (white-hip) M. melanurus Clade,and the Orange-leg M. marcai Clade (Figs. 6, 7).Within each Clade, the evolutionary pathwaytowards advanced metachromic bleached (andultimately albinotic) taxa can be plausibly retraced.Albinotic forms in dead-end distributions mayeventually go extinct (i.e., M. chrysoleucos in theM. humeralifer Clade; the new Mico species thatoccurs between the Rios Teles-Pires and Ronuro,M. leucippe, and M. argentatus in the M. argentatusClade; M. acariensis and M. saterei in the M.melanurus Clade; and M. manicorensis in the M.marcai Clade). In territorial sociable primates the

principle of metachromic bleaching that seems tofuel the trend to allopatry is an irreversible, seem-ingly non-adaptive evolutionary pattern. The meta-chromic pathway followed within the M. argentatusClade is a predominantly pheomelanin one, withfirst the nearest to archetypical, dark orange-coloredtaxon M. emiliae from the Rio Irirí. From M. emiliaediverged in southward direction the moderatelybleached new species that we identified to occurbetween the Rios Ronuro and Teles-Pires, andnorthward the advanced albinotic taxa M. leucippe(all white with a pink face) and M. argentatus (allwhite with a black tail). The latter occupy dead-enddistributions, as they are pressed at their northernlimit against the untraversable Rios Tapajós andAmazonas, respectively. Within the tufted-ear ortassel-ear Clade of Mico the metachromic patternfollowed the eumelanin pathway, from the darkestagouti-colored taxon M. mauesi going straight into

123

Figure 4. Present-day distribution of monotypic Goeldi’s Monkey Callimico goeldii.


the overall whitish and gray M. humeralifer in theClade’s northernmost dead-end distribution (delin-eated by the Amazon and Tapajós Rivers). Alongthe pheomelanin pathway it diverged into theoverall orange and white colored golden-whitetassel-ear marmoset chrysoleucos, the species thatoccupies the westernmost dead-end distributiondelineated by the Rios Madeira and Aripuanã.Within the white-mantle (white-hip) Clade of Micothe pathway followed goes from the nearest ar-chetypic agouti-colored taxon M. melanurus innorthern direction to the advanced pheomelaninbleached half-way albinotic taxa M. intermedius,M. acariensis, and M. saterei. And within the fourthGroup of Mico, the orange-leg M. marcai Clade, thepathway followed in northern direction starts fromthe metachromic nearest to archetypic taxon M.marcai diverging into the advanced euchromic toalmost albinotic taxon M. manicorensis, and inwestern direction proto-marcai evolved into theslightly but progressively bleached taxa M. nigri-ceps and M. rondoni, all three occupying dead-enddistributions delineated by the untraversable RioMadeira (after the Amazon River proper the secondstrongest river barrier in the entire Amazon Basin).As all interfluves occupied by a different Mico

species show dead-end distributions delineated byuntraversable rivers at their northern and westernlimits, each species represents a different stagealong the eumelanin or pheomelanin pathway thatis frozen in time, but at the end of its metachromicevolution it invariably will turn into albinotic (Figs.6, 7). Once arrived there, such primate taxa will in-evitably go extinct, unless a founder-colony man-ages in time to cross the geographic (riverine)barrier by means of a river bend cut-off, by hoppingon várzea forested floating islands, or by circum-venting a geographical barrier. According to thedoctrine, the evolutionary rate of metachromism isprimarily controlled for by the trend to allopatry,and secondarily by environmental and geneticfactors which may accelerate, retard, or terminatemetachromic processes, or hold them in dynamicequilibrium, but cannot alter, reverse, or deflectthem from their course. Hypothetically, growth andspread of a founder-colony of a certain Amazonianmarmoset across a certain interfluve delineated byrivers, entails social selection. Effective selectionstabilizes the mean chromotype of the colony at acolor tone or grade inbetween that of the founders

and that of the albinotic ones towards which allmonkeys tend. Amazonian marmosets (genus Mico)represent an advanced stock of callitrichids thatevolved as late as the Pleistocene, south of theAmazon River and east of the Rio Madeira, froman ancestral stock of the Callithrix ouistitis occur-ring in Central and SE Brazil (Van Roosmalen &Van Roosmalen, 2003). About 1.5 MYA, a majorvicariance took place - the break-through by theproto-Madeira River of the continental watershedrunning across the Chapada dos Parecis in Rondo-nia (Grabert, 1991). Thereafter, the entire area southof the Amazon and east of the Madeira drasticallyreversed its drainage pattern. Former rivers thatsince the beginning of the Pliocene had been drain-ing the extensive clear-water wetlands in north-south direction, dried up. New rivers (mostly of theblack-water type) arose and began to drain the areain opposite direction, from south to north. Most ofthese rivers emptied out in the Rio Madeira, somedirectly in the Amazon River. Founder-colonies atdifferent phenotypic stages of metachromic bleach-ing that derived from archetypic M. melanurus -pushed by the trend to allopatry - subsequentlyinvaded and inhabited the newly formed interfluvialterra firme ‘islands’ that new rivers had beencreating. These newly available lands offered themtheir preferred habitat of terra firme rainforest, inwhich they filled the niche of exudate gouging,which niche east of the proto-Madeira River washitherto exclusively occupied by the much smallerand peaceable, non-territorial dwarf marmosetCallibella humilis (Van Roosmalen & Van Roos-malen, 2003). Ever since, Callibella humilis seemsto have lost the battle against the aggressivelyexpanding Mico newcomers. Our assumption is thatthe dwarf marmoset has been locally driven toextinction almost all over its former range since thegenus evolved in the late Pliocene. Presently, theblack-crowned dwarf marmoset hangs on along thewestbank of the lower Rio Aripuanã. As a com-mensal, it takes refuge on the terras pretas (human-made black-earth farmland) from the deadly attacksof the local Amazonian marmoset Mico manicoren-sis (Van Roosmalen et al., 2000; Van Roosmalen &Van Roosmalen, 2003). This example may welldemonstrate that a specific ecological niche such asthat of specialized gougers and feeders of gum(exudates) in a certain habitat (e.g., primary rainforest) can only and exclusively be occupied by a


Figure 5. Schematic distributions as delineated by (for Amazonian Marmosets) untraversable rivers drawn for all known Amazonian Marmosets belonging to the genera Callibella, Cebuella, and Mico.

On the origin of allopatric primate species 125

single taxon that defends it, in this case evenbeyond generic bounds.Figures 5–7 demonstrate that about all inter-

fluves occupied by a single species of Mico showdead-end distributions delineated by for rainforesthabitat-specialists untraversable rivers at theirnorthern and western limits. At their southern limits,all distributions invariably show a mostly narrowopen-end, where a contact zone between twoadjacent distributions must exist. Hybridizationbetween Amazonian marmosets, though, has neverbeen seen or reported in the wild. This may well beattributed to strong social and sexual selection.Indeed, all Amazonian marmosets of the genusMico developed hypertrophied external genitalia ineach gender that are physically greatly differingamong related taxa (Van Roosmalen et al., 2000)(Figs. 8–11).

We ourselves have kept, raised and bred with anumber of Amazonian marmosets, both in free-ranging and captive conditions. Expressive andoften violent territorial behavior of all members ofa social group, aside of species-specific sexualdisplay of external genitalia, pheromones and scent-marking of one another’s coat, has always preven-ted our marmosets from hybridizing (interspecificcross-breeding). For instance, we kept breedingsocial groups of all three taxa of the tassel-ear M.humeralifer Clade (i.e., M. humeralifer, M. mauesi,and M. chrysoleucos). To avoid one group fromwiping out the other, we had to keep differentspecies in separate cages, whereas we let only onegroup of M. chrysoleucos free-ranging in the forestthat surrounded the compound. Even so, adultswere still seen trying to grab and bite one anotherthrough the fine-mess wire. From our unique exper-

126

Figure 6. Geographical distributions delineated by rivers drawn in one map for all known Amazonian Marmosets that belong to the marmoset genera Callibella, Cebuella, and Mico.

MARC G.M. VAN ROOSMALEN & TOMAS VAN ROOSMALEN

Figure 7. Radiation and metachromic diversification following eumelanin and pheomelanin pathways of metachromicbleaching depicted for all recognized phylogenetic Clades of Amazonian (Mico), Dwarf (Callibella), and Pygmy Marmosets(Cebuella) depicted to scale.


ience having kept all kinds of marmosets (untiltoday, not a single zoo in the world has any Micoon exhibit) and other callitrichids, both in captivityand free-ranging in a tropical rainforest environ-ment, we believe that where adjacent distributionsof two species of Mico are not defined by an untra-versable river, a sharp-lined contact zone mustexist, where cross- breeding never takes place. Thisassumption concurs with the principle of meta-chromic bleaching being irreversible. In theory,only through cross-breeding with a darker, overallmore saturated eumelanin taxon the metachromicpathway to albinotic could be reversed, something,however, that will never happen in the wild.As all Mico do display strong interspecific

territorial behavior - each group defending its livingspace by means of (often ritualized) territorial

boundary conflicts - within a given contact zonecross-breeding will not take place between neigh-boring groups of different but related ecospecies, asdistance is maintained by regularly performedboundary conflicts. This way, any gene flowbetween phenotypically different populations isimpeded. In phylogeographic terms, the fartherradiated away from the origin of a Clade’s disper-sion - that of the nearest to archetypic species withina monophyletic Clade - the more progressivelybleached the species will become. Partly or fullyalbinotic taxa, therefore, often occur in or near theClade’s dead-end distributions.

In figures 12–15, we have visualized the phylo-geographic distributions, radiation, and supposedpathways of metachromic bleaching of all knownTamarin Monkeys genus Saguinus. We have di-

128

Figures 8–11. In all taxa of Mico, both males and females evolved hypertrophied, species-specific, in anatomical respect verydifferently shaped external genitalia. Fig. 8: male M. manicorensis sexually displaying; Fig. 9: exposed pudenda in adultfemale M. acariensis; Fig. 10: pudenda with 2 cm long vaginal lips in M. saterei; Fig. 11: pudenda in M. acariensis. Thisfeature supports our view that all taxa of Mico should be considered different species and not just metachromic color morphs.


vided them up in the following monophyleticGroups or Clades: the Saddle-back Tamarins of theS. fuscicollis Clade (Fig. 13); the Black-mantleWhite- mouth Tamarins of the S. nigricollis Cladein one map combined with the Mustached Tamarinsof the S. mystax Clade, the Red-chested MustachedS. labiatus Clade, and the Emperor Mustached S.imperator Clade (Fig. 14); and the Bare-face Tamar-ins of the S. midas, S. bicolor and S. geoffroyiClades (Fig. 15). To complete the callitrichidpicture, we have visualized the distributions of theLion Tamarins genus Leontopithecus, and the TrueMarmosets or Ouistitis genus Callithrix, from SEBrazil (Fig. 16).In geological history, speciation and radiation

within the Saddle-back Tamarins of the S. fuscicol-lis Clade (Figure 13) went along two pheomelaninpathways of metachromic bleaching: one sub-Claderadiated south of the Amazon River from east to

west, from the most saturated eumelanin, nearest toarchetypic taxon S. mura (green distribution) to thecompletely albinotic all-white taxon S. melanoleu-cus (blue distribution) via the taxa S. avilapiresi,S. fuscicollis, and S. cruzlimai. The bleachingprocess took first place in the head parts - muzzleand blaze - and, after having traversed the Rio Juruáback to its right bank, the metachromic bleachingprocess completed from the overall orange-coloredtaxon S. cruzlimai into the fully albinotic taxon S.melanoleucus. Another radiation took place from S.mura directly into S. weddelli, and, after havingtraversed the Rio Purús, into the overall light-browncolored taxon S. primitivus - both with a fully al-binotic blaze and muzzle/mouth. A second sub-Clade of saddle-back tamarins radiated from thePeruvian Amazon in eastern direction, from thesaturated eumelanin nearest to archetypic taxon S.leucogenys (light blue distribution) into the slightly

129

Figure 12. Distributions of all Neotropical Tamarin Monkeys, genus Saguinus, compared with those of all Amazonian Marmosets.


bleached taxa S. illigeri and S. nigrifrons, and aftercrossing the upper Amazon River (where it is calledRío Marañón) northward into the progressivelybleached taxa S. lagonotus, S. fuscus, and S. tri-partitus, the latter three taxa being distributed northof the Amazon River in the Ecuadorian, Colombian,and Brazilian Amazon.Within the Black-mantle White-mouth Tamarins

of the S. nigricollis Clade (Fig. 14) that is dis-tributed only north of the Amazon River in theBrazilian, Ecuadorian and Colombian Amazon, thenearest to archetypic saturated eumelanin taxon isS. nigricollis. It radiated northwestward and di-verged into the slightly bleached taxa S. graellsi andS. hernandezi. The S. nigricollis Clade is sympatricwith the saddle-back tamarins of the taxa S.lagonotus, S. tripartitus and S. fuscus (Fig.13).However, they occupy different ecological nichesand therefore can be seen traveling and foraging inmixed species associations, with the larger-sizedblack-mantle tamarins in the lead and stayinghigher up in the canopy of the terra firme rain forest.Within the Emperor Mustached Tamarins of the S.imperator Clade both extant taxa are alreadyprogressively bleached, the grayish taxon S. sub-griscescens slightly more so than S. imperator. Inthe upper Rio Purús region there must exist anarrow contact zone between the two taxa along thesouthernmost open-end distribution of S. imperator.Within the Red-chested Mustached Tamarins of theS. labiatus Clade, Saguinus labiatus occupies thesouthernmost distribution and represents the nearestto archetypic taxon with a dark red chest andthin-lined white mustache. It radiated north of theRio Ipixuna and diverged into the advanced orange-chested taxon S. rufiventer that has a more bleachedwhite mustache and head-stripe. The third taxon ofthe S. labiatus Clade is S. thomasi the precursor ofwhich once must have traversed the Rio Solimões.It might have been replaced later by S. inustus northof the Rio Solimões as far west as the Rio Japurá.Saguinus thomasi nowadays only occupies thelower Rios Solimões/Japurá interfluve. It representsthe most progressively pheomelanin bleached taxonof the S. labiatus Clade in its light orange-coloredchest and the broad-lined triangular white mus-tache. Within the mustached tamarins of the S.mystax Clade, the more saturated eumelanin,nearest to archetypic form is represented by thetaxon S. mystax that is distributed west of the Rio

Juruá. After traversing the Rio Juruá, the Clade hasradiated eastward while further bleaching along thepheomelanin pathway into the orange-crownedtaxon S. pileatus, and along the eumelanin pathwaydiverging directly from S. mystax into S. pluto. Thelatter taxon is overall more grayish and has a dis-tinctive albinotic spot around the base of the tail. Inthe lower Rios Juruá/Purús interfluve we havesighted S. pluto ranging always in mixed-speciesassociation with the smaller saddle-back tamarinS. avilapiresi, with S. pluto always in the lead andS. avilapiresi rushing behind and below the groupof S. pluto in the lower strata of high forest, alwaysin a hurry feeding on S. pluto’s left-over food items.A hypothetical pathway of allopatric speciation,

radiation and metachromic bleaching followed bythe Bare-face Tamarins of the S. midas, S. bicolorand S. geoffroyi sub-Clades may have had its originin the Guianas (Fig. 15). An all-black, saturatedeumelanin archetypic precursor of S. midas mayonce have traversed the lower Rio Amazonas andspeciated allopatrically into the black-handed taxonS. niger. Or vice-versa (archetypic black-handedS. niger may once have traversed the lower RioAmazonas and speciated allopatrically into the red-handed S. midas). The same or another all-blackprecursor of S. midas may have traversed the RioNegro and allopatrically speciated into the taxon S.inustus that is all-black with a white-mottled face.Saguinus inustus nowadays occupies the entireinterfluve between the Rio Negro in the north, andthe Rios Solimões, Japurá and Caquetá in the south.A founder-colony of a predecessor of S. inustusdriven by the trend to allopatry may then haveventured from the taxon’s westernmost distributioninto the NW Colombian Río Magdalena basin.Once having inhabited the Río Magdalena basin, itmay have diverged along a pheomelanin pathwayinto the extant taxon S. leucopus that has a white-hairy facial circumference similar to S. inustus.Saguinus leucopus then may have radiated furtherinto the progressively pheomelanin bleached,almost euchromic taxon S. oedipus, and fromthere into the near-albinotic taxon S. geoffroyi thatis distributed from extreme NW Colombia intoPanama, as such the farthest away from the centerof dispersion of the Bare-face Tamarin Clade. Withrespect to the three derived euchromic taxa of theS. bicolor Clade, as we have mentioned elsewhere,these taxa find themselves in the process of being


rigorously displaced from their respective territoriesby the now sympatric archetypic saturated eu-melanin red-handed tamarin S. midas. All three taxa(i.e., S. bicolor, S. martinsi, and S. ochraceus) findthemselves pushed with the back against the untra-versable Rio Negro and/or Rio Amazonas (Fig. 15).At present, the red-handed tamarin S. midas iswrapping up the last stage of its range extensiontowards the south to the cost of all three Bare-faceTamarins of the S. bicolor subClade. This battleover a specific ecological (feeding) niche, in whichtwo sympatric, closely related primate taxa areinvolved, will inevitably lead to the extinction ofthe most euchromic among the two, that is theBare-face Tamarins of the S. bicolor sub-Clade:the taxa S. bicolor, S. martinsi, and S. ochraceus(Fig. 15).The eumelanin S. midas sub-Clade might have

originated in the Guianas north of the watershedwith the northeastern Amazon formed by theTumac-Humac Mountains and the open wet savan-nas of Roraima and Pará. A predecessor of the S.midas sub-Clade, perhaps the extant S. midas itself,once may have circumvented the watershedbetween the Guianas and Brazil by traversing theParú Savanna, whereafter it may have penetratedfar southwards into the northeastern quadrant of theBrazilian Amazon. We assume that before somevicariance took place this vast territory or a largepart of it was inhabited by precursors of the closelyrelated Bare-face Tamarins of the S. bicolor sub-Clade. Apparently, as the two sub-Clades do occupythe same ecological niche, (proto)-midas sub-sequently has displaced (proto)-bicolor over mostof its former range. This battle is still being foughtover between S. midas and each taxon of the S.bicolor sub-Clade, but it seems to come close to itsend. The process of replacement is accelerated bydeforestation and other human disturbance such asroad-building that has taken place north of the rap-idly expanding megacity of Manaus. This ongoingstory clearly demonstrates interspecific intolerancein closely related territorial monkeys that occupyand exploit the same ecological niche. It inevitablyleads to displacement, or sooner or later extermina-tion of the more progressively bleached (eu-chromic) taxon. This kind of replacements may takeplace after a geographic barrier has been success-fully overtaken by the more saturated eumelanin(more adaptive and/or aggressive?) of two related

131

taxa. Or: after a vicariance has removed a hithertogene-flow impeding geographic barrier inbetweenthe distributions of two or more closely relatedspecies. Vicariance (from Latin vicarius) means a pro-

cess by which the geographic range of an individualtaxon, or an entire biota, is split into discontinuousparts by the formation of a physical barrier to geneflow or dispersion.Today, the S. bicolor sub-Clade only inhabits a

20–30 km narrow strip of terra firme rain forestalongside the southernmost edge of the Pre-Cam-brian Guayanan Shield.The three bicolor taxa are so to speak pushed

with the back against rivers that happen to be thewidest and most difficult to traverse on the entireSouth-American continent: the Rios Negro andAmazonas. The three extant taxa of Bare-faceTamarins each occupy what is called a “dead-enddistribution”. The distribution of the half-brown,half- white taxon S. bicolormeasures not more than20-30 x 200 km, delineated in the west and southby the Rios Cuieiras, Negro, Amazonas, and Urubú.Bicolor’s neighbor to the east - the almost fullybleached, ochraceous colored taxon S. ochraceus -occupies the interfluve between the Rios Urubú andUatumã. To the east of its distribution, the pheo-melanin, light orange-colored taxon S. martinsioccupies the lower interfluve between the RiosUatumã and Nhamundá (Fig. 15). Disputedly, anow extinct precursor of the Bare-face S. bicolorsub-Clade that once ranged somewhere to the northof the Amazon River, may have driven the threeextant taxa of the S. bicolor subClade - each at adifferent stage of metachromic bleaching - into thesmall interfluvial dead-end distributions, that theyoccupy today. The saturated eumelanin (blackish-brown) red-handed tamarin S. midas that later ex-panded its range to the south, is now simultaneouslyinvading the three remaining adjacent interfluvialstronghold territories of the S. bicolor sub-Clade. A sharp-line contact zone drawn between S.

midas and S. bicolor territory has been notified byus in the early 1990s to run at 28–30 km north ofand parallel to the Negro and Amazon Rivers.While running a halfway house for orphaned mon-keys situated right at the edge of the contact zone,we have repeatedly witnessed different socialgroups of S. midas raiding resident family groupsof S. bicolor. These incidents invariably ended up


Figure 13. Distributions, allopatric speciation, radiation, and supposed pathways of metachromic bleaching in all knownSaddle- back Tamarins of the Saguinus fuscicollis Clade. Figure 14. Idem, in the more robust, larger-sized Black-mantleWhite-mouth Tamarins of the S. nigricollis Clade, the Emperor Mustached Tamarins of the S. imperator Clade, the Red-chested Mustached Tamarins of the S. labiatus Clade, and the Mustached Tamarins of the S. mystax Clade.


Figure 15. Distributions, allopatric speciation, radiation and supposed pathways of metachromic bleaching in all extantBare-face Tamarins that belong to the Saguinus midas, S. bicolor, and S. geoffroyi sub-Clades. Figure 16. Distributions,allopatric speciation, radiation and supposed pathways of metachromic bleaching in all known True (or Atlantic Forest)Marmosets (genus Callithrix) and Lion Tamarins (genus Leontopithecus) from SE Brazil.


in the defensive, less aggressive (more sensitive?)S. bicolor bitten to death. Now, about twenty yearslater, S. midas has extended its range at least fivekm further to the south to the cost of S. bicolor oc-cupied territory. As S. midas is more opportunisticand flexible in its habitat preferences - venturingalso into secondary growth and edge habitats suchas roadsides - it rapidly penetrates into S. bicolorterritory, at some places (e.g., Ducke Reserve)already reaching the outskirts of Manaus. Runninga rehabilitation center for orphaned monkeys, wesometimes received whole families of S. bicolorthat were rescued from isolated pockets of forest inurbanized areas. After some time spent in quaren-taine, we used to put them in large cages built onpoles in the middle of the rain forest about thirtykm north of Manaus in an attempt to reintroduce thespecies where we assumed it had occurred not longbefore. One day before releasing a wild-caughtsocial group of 8 S. bicolor, we found them allbitten to death inside the cage that was fenced withgalvanized small-meshed wire. The only animal leftalive in the cage was a wild adult S. midas thatapparently had not found back the little hole in thewire through which he and some other family mem-bers had entered the cage that very morning. On theother hand, a hand-tame S. midas infant that weraised free around the compound at the time, oneday was ‘kidnapped’ and adopted by the wild S.midas group that roamed around in the project area.

Within the True (Atlantic Forest) Marmosets orOuistitis genus Callithrix we distinguish twomonophyletic Clades: the Ca. penicillata Clade andthe Ca. aurita Clade (Fig. 16). Within the firstmonophyletic Clade we consider Ca. penicillata thenearest to archetypic, most saturated eumelanintaxon that occupies the largest distribution (darkgreen area). From there, it radiated in northerndirection and diverged into the overall progressivelybleached taxon Ca. jacchus that has fully albinoticear-tufts. In eastern direction, from it derived andradiated away the taxa Ca. kuhlii and Ca. geoffroyithat are progressively bleached euchromic to al-binotic in their mantle and head parts (except theblack ear-tufts). Their dead-end distributions arepressed against the Atlantic coast. Interestingly, Ca.kuhlii’s range fully overlaps with that of Leonto-pithecus chrysomelas. The Ca. aurita Clade has Ca.aurita representing the nearest to archetypic, overallmetachromic agouti taxon that ranges allopatric

with the lion tamarins (Leontopithecus) in theAtlantic forest of SE Brazil. From there derived thenear albinotic taxon Ca. flaviceps that occupies asmall area in SE Minas Gerais, allopatric with Ca.geoffroyi (Mittermeier et al., 2013).As for the Lion Tamarin genus Leontopithecus,

we consider the overall saturated eumelanin, almostall-black taxon L. chrysopygus the nearest to ar-chetypic lion tamarin. From it derived in southeast-ern direction the taxon L. caissara that followed ametachromic pathway of pheomelanin bleachingin its bright orange-colored dorsal parts whilemaintaining the saturated eumelanin black tail,arms, legs, mantle and head of L. chrysopygus. Itssmall range in coastal Paraná State represents thesouthernmost distribution of any callitrichid. FromL. chrysopygus derived in northeastern directionalong a pathway of pheomelanin bleaching the twoother taxa, L. chrysomelas and L. rosalia. Leonto-pithecus chrysomelas bleached in the orange coloredlower arms and legs, and in the light orange tocream-white head and mantle maintaining the restof its body saturated eumelanin. Leontopithecusrosalia, in turn, is evenly light orange-colored overits whole body, with the tail becoming almost al-binotic. Both taxa occupy small dead-end distribu-tions in the Atlantic forest along the coast of SEBrazil (Mittermeier et al., 2013).In a further attempt to falsify the principle of

metachromic bleaching and the crucial role webelieve it plays in allopatric speciation of (at least)Neotropical monkeys, we now will proceed toexamine currently known distributions, allopatricspeciation and radiation, and the pathways ofmetachromic bleaching supposedly followed in allother male-territorial Neotropical monkey genera(i.e., Callicebus, Saimiri, Cacajao, Chiropotes,Pithecia, Lagothrix, Ateles, Brachyteles, Alouatta,Cebus, Sapajus, and Aotus).Titi Monkeys of the genus Callicebus are strongly

territorial in behavior, a family marking its territoryvocally - a pair calling in duet, or a whole familycalling in chorus. In the Amazon, a single taxon ofthe Collared Titi Cal. torquatus Group may occurin sympatry with a single titi of any of the otherNon-collared Titi cladistic Groups, once the formertitis are only found high up in the canopy of primaryterra firme rain forest. Collared titis occupy adifferent, more frugivorous feeding niche than thetitis that lack the white collar. The latter prefer the


lower strata and edges of terra firme rain forest,secondary growth, and savanna forest, being overallmore omnivorous in their diet that contains alsoyoung leaves and insects, in addition to pulpy fruits(Hershkovitz, 1988; Hershkovitz, 1990; Mitter-meier et al., 2013).In figures 17–20, we show the distributions of

all known Titi Monkeys genus Callicebus. Withinthe titi monkeys five phylogenetic cladistic Groupsor Clades are recognized: Cal. personatus (south-eastern Brazilian taxa), Cal. torquatus (Amazoniancollared taxa), Cal. moloch, Cal. cupreus and Cal.donacophilus (Amazonian non-collared taxa) (VanRoosmalen et al., 2002). Within each titi Clade theirreversible pathway of metachromic bleachingtowards partly or fully albinotic, from saturatedeumelanin and saturated pheomelanin fields towhite or colorless, is clearly demonstrated. Thefarther radiated away from the prototypic agouti orsaturated eumelanin (black or dark brown) taxon -Cal. melanochir in the Cal. personatus Group, Cal.medemi in the Cal. torquatus Group, Cal. cineras-cens in the Cal. molochGroup, Cal. brunneus in theCal. cupreus Group, and Cal. modestus in the Cal.donacophilus Group - the more its pelage turns intoorange, yellowish or cream to white, first in certainparts of the body, and eventually all over its coat.Near-albinotic forms in dead-end distributions (e.g.,Cal. pallescens, Rio Xingú titi, Rio Mamurú titi)are doomed to eventually go extinct, as meta-chromism with the trend to allopatry as the drivingbehavioral factor is an irreversible, initially seem-ingly non - adaptive evolutionary pattern in allterritorial monkeys. As shown in the maps, in theAmazon all distributions of titis without a whitecollar are occupied by just a single taxon and aredelineated by rivers that function as (for titis thatcannot swim) strong geographic barriers. Narrowcontact zones between adjacent interfluvial distribu-tions surely do exist, usually near the headwaters, butnowhere interbreeding or hybridization between thetwo neighboring taxa has been reported to take place.Our extensive primate surveys carried out

throughout the entire Amazon Basin have revealedthat, in general, a given monkey taxon looksphenotypically identical throughout its entire range.In contact zones or across opposite banks of riversthat demographically separate two phylogeneticallyrelated taxa, we have noticed interspecific boundaryconflicts and vocal battles to occur regularly, in

particular performed by social groupings of titis,howling monkeys and spider monkeys. In at leastone contact zone between two differently lookingtitis we have been able to perceive the ‘trend toallopatry’ put in motion in metachromic bleachedindividuals that were deviant from the commonlyseen phenotype. At the far northeastern corner ofthe distribution of Hoffmann’s Titi Monkey Cal.hoffmannsi a small founder-population of an al-binotic all-creamwhite form, that we provisionallynamed the “Rio Mamurú titi”, apparently has beenpushed into a dead-end distribution between theright bank of the lower Rio Mamurú, the for titisinhospitable várzeas (seasonally white-water inund-ated floodplain forest) along the right bank ofthe Rio Amazonas, and the parapatric distributionof Cal. hoffmannsi to the east and south as far asthe lower Rio Tapajós (Fig. 17). Only mtDNA se-quences may determine what taxonomic status weshould allocate to this new, fully euchromic taxon:‘color morph’ or ‘taxon in the making’. A colormorph of Cal. hoffmannsi, a subspecies to be namedCal. hoffmannsi mamuruensis, or a valid newspecies to be described as Cal. mamuruensis? Asmentioned before, here our ecospecies conceptcould be applied in case the population has beenconfirmed to be allopatric and genetically isolated(not allowing any gene flow) from the taxon itderived from, or when the enclave population hassuccessfully adapted to a different ecological niche- in this case turning itself into a várzea versus terrafirme rainforest habitat specialist. Our ecospeciesconcept (hereafter named ESC) in combination withthe phylogenetic species concept (PSC) is, at leastin the field, more practical, less arbitrary, and betterdefined, in particular when used for the purpose ofspecies and biodiversity conservation. The ESCwould put an end to the academic discussion aboutthe arbitrary and controversial subspecies/raceconcept.As Groves (2001a; 2001b; 2004; 2005) points

out: “There is no official taxonomy”. The numerousconcepts as to what is and what is not a species arecontroversial, and every named species is itselfnothing more than a hypothesis. Our understandingof the systematics of the primates is constantlygrowing, not only through the discovery of newspecies but also with new information brought tobear from diverse fields such as morphology, cyto-genetics, molecular genetics, paleontology, biogeo-


graphy, physiology and behaviour - contributingto test the hypothesis that a certain organism is aspecies distinct from another. Distinct in whatsense? An individual is distinct, a population isdistinct, but when and in what way is it a distinctspecies?Among the Titi Monkeys of the Cal. moloch

cladistic Group (Fig. 17), the all-agouti dark-tailedtaxon Cal. cinerascens, ranging along the east bankof the Rio Aripuanã and between the right bank ofthe lower Rio Madeira and the left bank of theRio Canumã, seems to represent the nearest to ar-chetypic, most original or ancestral titi from whichall other taxa of the Cal. moloch Group havederived. Phylogeographically, the current central-southern Amazonian distribution of Cal. cineras-cens is thought to represent the center of dispersionof the Cal. moloch Clade. In other words, the upperAripuanã region in Rondonia may be considered the

cradle of Cal. moloch Clade’s evolution and disper-sion. From there, all taxa of the Cal. moloch Cladehave diverged, radiating away in all (but southern)directions. A distinct metachromic trend to saturatedpheomelanin (orange beard and sideburns) and al-binotic (cream to white beard, sideburns, tail and/orwhole body) can be seen, which means that themost progressively bleached taxa that demograph-ically radiated the farthest away from the arche-type’s origin of dispersion tend to euchromic oralbinotic (i.e., Cal. moloch east of the Rio Tapajós,and Cal. hoffmannsi in the northernmost dead-end distribution delineated by the untraversableAmazon and Tapajós Rivers). The supposed meta-chromic pathway taken is as follows: Cal. cineras-cens radiated first in northern direction, somefounder-colony traversed the Paraná do Urariá,followed the pheomelanin pathway, and diverged

Figure 17. Distributions, allopatric speciation, radiation and supposed pathways of metachromic bleaching in all known Amazonian Non-collared Titi Monkeys genus Callicebus.


into taxon Cal. baptista (which has dark orange-colored beard, sideburns, lower extremities, andbelly). Radiating eastwards, it diverged into Cal.hoffmannsi (its forehead, beard, sideburns, hands,feet, and belly bleached light gray to cream-white).After Cal. hoffmannsi happened to traverse the RioTapajós, most likely where it is called Rio Juruena,it diverged into the advanced pheomelanin bleachedto albinotic taxon Cal. moloch that now occupies alarge distribution east of the Rio Tapajós and southof the Amazon River. Callicebus hoffmannsi alsodiverged along the upper course of the Rio Tapajósinto the recently described taxon Cal. vieirai (ran-ging between the Rios Juruena and Teles Pires),which is near-albinotic.When a taxon is occupying a given interfluvial

distribution delineated by hard to traverse riverbarriers, it has irreversibly changed its pelage orparts of its coat (e.g., beard, sideburns, ear-tufts,

forehead, tail, hands, feet) following the eumelaninpathway from agouti or saturated eumelanin to al-binotic (cream or white), via black, brown, drab,and gray, and/or the pheomelanin pathway via red,orange, and yellow, or a combination of the twopathways in different parts of the body or coat.The trend to albinotic in the Cal. moloch Clade iscompleted near its northernmost dead-end distribu-tion in the all-cream to white new form that wehappened to identify along the right bank of the RioMamurú (Fig. 21). It must have derived from dark-tailed but cream-bearded and -bellied Cal. hoff-mannsi. Following the trend to allopatry, this colormorph (or ecospecies or ‘taxon in the making’?) ispushed with the back against the várzeas (white-water floodplain forests) and right bank of theuntraversable Rio Amazonas. Callicebus molochClade’s westernmost distribution is represented bythe advanced pheomelanin bleached (bright orange

Figure 18. Distributions, allopatric speciation, radiation and supposed pathways of metachromic bleaching in all knownAmazonian Collared Titi Monkeys of the Callicebus torquatus Clade with two sub-Clades: Cal. torquatus and Cal. lugens.


belly, beard and sideburns) to albinotic (white fore-head, hands, feet and tail tip) taxon Cal. bernhardiwhich, in turn, is pushed with the back against thealso untraversable Madeira River.Among the Titi Monkeys of the Cal. cupreus

cladistic Group (Fig. 22), we consider Cal. brun-neus the nearest to archetypic taxon. Centrally dis-tributed, the overall agouti colored taxon Cal.brunneus radiated in northwestern direction via the

progressively pheomelanin bleached taxa Cal. du-bius and Cal. discolor into the most pheomelaninbleached (light orange tail base, beard, sideburns,belly and inner limbs) to albinotic (snowwhite tail,hands, feet and front/blaze) white-fronted taxonCal. ornatus that is distributed north of the Amazonin the Colombian Amazon. From Cal. brunneussouthwards diverged the advanced pheomelaninbleached taxon Cal. aureipalatii in the Clade’s

Figure 19. Schematic map of the distributions of the SE Brazilian or Atlantic Forest Titi Monkeys of the Callicebus personatus Clade, which are separated by (for these taxa) untraversable rivers.


southernmost distribution (the Bolivian Amazonbetween the Rios Madre de Dios and Beni). FromCal. brunneus radiated away in northern directionfirst the slightly pheomelanin bleached taxonCal. cupreus that now occupies a large interfluvialarea west of the Rio Purús and south of the RioSolimões. After an ancestral founder-colony of Cal.cupreus managed to traverse the Rio Purús to theeast, it diverged into the advanced pheomelaninbleached all-orange, but white-tailed taxa Cal. ca-ligatus and Cal. stephennashi in the northeastern-most dead-end part of Cal. cupreus Group’sdistribution, as the Rio Madeira represents thesecond strongest riverine barrier on the South-American continent. Last but not least, also fromCal. cupreus derived in northwestern direction therecently described, advanced pheomelanin bleachedtaxon Cal. caquetensis that at present occupies

a small, not yet fully identified area north of theAmazon and Caquetá Rivers in the Colombianlowland Amazon, allopatric with and south of thedistribution of the white-fronted titi Cal. ornatus(Fig. 17).Among the Titi Monkeys of the Cal. donaco-

philus cladistic Group (Fig. 17), we consider theoverall agouti-colored taxon Cal. modestus themost original, nearest to archetypic taxon. It occu-pies the Clade’s northernmost distribution deline-ated by the Rios Beni and Mamoré. It radiatedsouthwards into the slightly pheomelanin bleachedorange-brown taxon ollalae. Following a eumelaninbleaching pathway, Cal. modestus also radiated insoutheastern direction, first into the near-albinotictaxon Cal. donacophilus, and from there into thefully albinotic taxon Cal. pallescens. The latternowadays occupies the southernmost dead-end

Figure 20. Distributions, allopatric speciation, radiation and supposed pathways of metachromic bleaching in all known SE Brazilian or Atlantic Forest Titi Monkeys of the Callicebus personatus Clade.


Figure 21. Distributions of Callicebus baptista, Ca. hoffmannsi, and the “Rio Mamurú titi” - the latter perhaps to be con-sidered a new taxon or one ‘in the making’. This satellite image shows the location of an enclave population of fully albinotictiti monkeys that we have found to exist along the right bank of the Rio Mamurú. This population is on the verge of extinctionas it is pushed with the back against for titis inhospitable habitat - the seasonally inundated floodplain forest (várzea) alongthe Rio Amazonas and the outskirts of the rapidly expanding town of Parintins in the north, and lands occupied by Ca. hoff-mannsi stretching to the east as far as the Rio Tapajós. The species Ca. baptista belonging to the Ca. moloch Clade originallyranged only north of Paraná do Canumã, P. do Urariá and P. do Ramos, east of the lower Rio Madeira, south of the RioAmazonas and west of the Paraná do Ramos. South of this narrow distribution evolved the species Ca. hoffmannsi, whichoccupies a large distribution between Rio Canumã in the west, Rio Tapajós in the east, and Rio Amazonas in the north, eastof Paraná do Ramos and Rio Mamurú. Baptist’s Titi is much more colorful being dark to bright red on the ventral parts andlower limbs, having a red beard and red sideburns, whereas the rest of its body is grayish to blackish agouti. Hoffmann’sTiti is basically two-colored grayish and yellowish-white to almost white, its sideburns and beard being light cream-white.However, we spotted the Ca. baptista titis also along the west bank of the Rio Uíra-Curupá, hence it once must have tra-versed the Paraná do Ramos west of the town of Parintins, forming an enclave population there after it displaced Hoffmann’stitis from the interfluve delineated by the lower Rio Uíra-Curupá and Rio Andirá. We also spotted advanced metachromicbleached, near-albinotic, pale yellowish to all-white ‘color morphs’ being phenotypically most related to Ca. hoffmannsialong the Rio Mamurú, the next river to the east, and classic yellowish- white and gray Ca. hoffmannsi with black tailsalong both banks of the middle and upper Rio Andirá. These observations may confirm a case of what is called parapatry.The two valid species Ca. hoffmannsi and Ca. baptista that are allopatric for the greater part of their distributions - phylo-geographically separated from one another by untraversable water bodies - exclude one another where Ca. baptista happenedto traverse a riverine barrier and subsequently replaced the local Ca.hoffmannsi population. There, both taxa live parapatric,meaning in adjacent ‘patrias’ not separated by geographic barriers, where gene flow in theory is possible, but in reality doesnot occur. A plausible explanation would be that the two taxa have already diverged too far from one another. One couldonly speculate about the future of the fully albinotic form seen along the right bank of the Rio Mamurú. It may represent afounder-colony or population of metachromic progressively bleached individuals that have been driven into parapatry bythe Ca. hoffmannsi populations found to the east and south of Rio Mamurú as far as the Rio Tapajós. The Rio Mamurú titiseventually might go extinct, unless they manage to adapt to (for titis) inappropriate habitat - the extensive várzeas along theright bank of the Rio Amazonas. If the founder-colony, following the trend to allopatry, would successfully adapt to theecological niche of várzea, then a new taxon could derive from Ca. hoffmannsi. Through inbreeding, the currently adoptedeuchromic coat coloration would stabilize phenotypically across the entire population of that new taxon in a relatively shortperiod of time. The hypothetical evolutionary path would then go from a metachromic fully bleached, near-albinotic colormorph in a dead-end distribution to a new taxon belonging to the monophyletic Ca. moloch Clade. In that case, we wouldhave to name the Rio Mamurú Titi Monkey Ca. mamuruensis.


distribution of the Cal. donacophilus Clade, penet-rating far into the arid Chaco of Paraguay and thepampas of Argentina. A new species of titi, recentlycollected by the Brazilian ornithologist MarceloVasconcellos in the Chiquitanos area along the RioParaguay in the Pantanal of Mato Grosso do Sul(for which taxon we identified the holotype in thezoological collection of the AMNH, in 1977 collec-ted by George Schaller and misidentified as Cal.donacophilus), represents the easternmost distrib-uted taxon of the Cal. donacophilus Clade. Exceptfor its dark gray ears (white in Cal. donacophilus),the Chiquitanos titi is overall more pheomelaninbleached towards albinotic than Cal. donacophilus,but less so compared to Cal. pallescens. Further-more, from Cal. modestus derived in northwesterndirection the advanced pheomelanin bleached near-albinotic taxon Cal. oenanthe that is nowadaysfound isolated in a small area in the east-PeruvianAmazon, south of the Río Marañón.Among the Collared Titis of the Cal. torquatus

cladistic Group (Fig. 18), we consider the saturatedeumelanin black-handed taxon Cal. medemi withthe westernmost distribution north of the AmazonRiver the nearest to archetypic form from whichderived the all-black but yellow-handed titi fromthe southbank of the Rio Negro - a newly identified,as yet to be described taxon - and from that taxonderived the all-black, dorsally slightly reddish-tinged taxon Cal. lugens with the northernmostdistribution of the Cal. lugens sub-Clade. Fromarchetypic Cal. medemi south of the Rio Caquetáderived, first in eastern direction the dorsally pheo-melanin bleached taxon Cal. lucifer. Some ancestralfounder-colony of the new Rio Negro southbankspecies then must have managed to traverse thelower Rio Solimões somewhere between the mouthof the Rio Purús and that of the Rio Madeira. Fromthere, collared titis could radiate away back inwestern direction, though south of the AmazonRiver, into the further pheomelanin bleached,overall reddish-brown colored white-handed taxaCal. torquatus and Cal. purinus, and, after tra-vers-ing the Rio Juruá, into the advanced pheo-melanin bleached red-handed red-fronted taxonCal. regulus.Within the SE Brazilian Titi Monkeys of the

Cal. personatus Clade (Figs. 19–20) the nearest toarchetypic, most saturated eumelanin taxon is Ca.melanochir. It ranges along the Atlantic coast south

of the Rio Paraguaçú in the center of dispersion ofthe personatus Clade. From Cal. melanochir de-rived in northern direction along the pheomelaninpathway the advanced pheomelanin bleached (all-orange colored) taxon Cal. barbarabrownae, and,in a small dead-end distribution delineated by theuntraversable lower Rio São Francisco in the northand the Atlantic Ocean in the east derived thealmost fully bleached, near-albinotic taxon Cal.coimbrai. Radiating in southern direction, ancestralCal. melanochir diverged along the eumelaninpathway into the orange-tailed, but overall darkbrown-colored taxon Cal. nigrifrons, and along thepheomelanin pathway into the advanced pheo-melanin bleached, all-orange colored and near-al-binotic taxon Cal. personatus.Within the Squirrel Monkeys genus Saimiri

(Fig. 22), we phylogeographically distinguish twomonophyletic Clades: Sa. sciureus - including theCentral-American Sa. oerstedii sub-Clade - and Sa.boliviensis - including the Bare - ear Sa. ustus sub-Clade (Hershkovitz, 1984). It is inferred that thegenus Saimiri evolved relatively recently, withcrown lineages diverging as late as the Pleistocene(ca. 1.5 MYA) and other major Clades divergingbetween 0.9-1.1 MYA. Concurring with Chiou etal. (2011), we include Sa. oerstedii in the mono-phyletic Sa. sciureus Clade that originated in theGuianas. North of the Amazon, it radiated in west-ern direction and diverged first into Sa. cassiquiar-ensis, a taxon that is nowadays distributed acrossthe entire Rio Negro basin, its distribution in thesouth delineated by the Rio Japurá/Caquetá and thelower Rio Solimões. From Sa. cassiquiarensis di-verged in northern direction the advanced bleached,least colorful taxon Sa. albigena that ranges allo-patric (north of the Río Guaviare) in the southwest-ernmost part of the Río Orinoco basin. Inconcurrence with Chiou et al. (2011), who foundevidence for monophyly in the Sa. sciureus and Sa.oerstedii Groups, we suggest that from Sa. albigenaor some ancestral precursor of it derived and radi-ated away in northwestern direction the advancedpheomelanin bleached taxa Sa. oerstedii and Sa.citrinellus. These now range in Sa. sciureus Clade’sdisjunct northwesternmost dead-end distribution -along the Pacific coast of Panama and Costa Rica.Along a different metachromic pathway derivedfrom Sa. cassiquiarensis in southern direction theadvanced bleached taxon Sa. macrodon. Its distri-


bution is delineated by the Ríos Guaviare and Apa-poris in Colombia, and the Rio Japurá in Brazil, andsouth of the Amazon by the upper Río Marañon inthe west, and the Rio Juruá in the east. Within itslarge distribution, Sa. macrodon is excluded fromthe Ríos Huallaga/Ucayali interfluve in the Per-uvian Amazon that is occupied by Sa. peruviensis.In the Guianas, Sa. sciureus once managed to tra-verse the lower Amazon River to the south. As it isa riverbank marsh and mangrove forest specialist,Sa. sciureusmust have colonized the south bank ofthe Amazon after reaching it on floating islandscovered with várzea or mangrove vegetation. FromSa. sciureus south of the Amazon subsequently de-rived the recently described, advanced bleachednear-albinotic taxon Sa. collinsi that is confined toMarajó Island - the Sa. sciureus Clade’s eastern-most dead-end distribution delineated by the At-lantic Ocean, and the Amazon and Pará Rivers.

The second monophyletic Clade of SquirrelMonkeys, the Sa. boliviensis Clade, has origin-ated in the extensive white-water floodplain forest(várzea) near the confluence of the Japurá andSolimões Rivers. The lower Japurá/Solimões inter-fluve does not contain any terra firme. It is season-ally flooded over 6–8 months. Here lives the nearestto archetypic, saturated eumelanin taxon of the Sa.boliviensis Clade, Sa.vanzolinii. It is overall agoutiand black colored, representing the only extantsquirrel monkey with an all-black tail. A somewhatbleached Sa. vanzolinii founder-colony once musthave reached (swimming or on a floating várzea is-land) the south bank of the Rio Solimões east of itsconfluence with the Rio Juruá. There evolved fromit the somewhat pheomelanin bleached, orange toyellowish taxon Sa. boliviensis. It then occupiedeast of the Rio Juruá the entire area delineatedby the lower Purús, upper Madeira and Guaporé

Figure 22. Distributions, allopatric speciation, radiation and supposed pathways of metachromic bleaching in all known Squirrel Monkeys genus Saimiri divided up in the S. sciureus and S. boliviensis Clade.


Rivers, whereas in the south it extended its rangefar into the Peruvian and Bolivian Amazon. After afounder-colony of somewhat bleached Sa. bolivien-sis happened to traverse the easternmost river bar-rier, it diverged into Sa. ustus - the least colorful,most eumelanin bleached taxon of the Sa. bolivien-sis Clade. Its distribution is confined by the AmazonRiver in the north, the Rio Xingú in the east (sep-arating the distributions of Sa. ustus and Sa.sciureus), and the Rio Guaporé in the south. Fur-thermore, from boliviensis in its southwesternmostdistribution in the Peruvian Amazon derived theprogressively pheomelanin bleached near-albinotic,most colorful taxon Sa. peruviensis. It occupies adead-end distribution - the interfluve delineated bythe Ríos Huallaga and Ucayal i - as it is surroundedby Sa. macrodon occupied territory.Historically followed metachromic and phylo-

geographic pathways, intraspecifically pushedahead by the trend to allopatry and the principle ofmetachromic bleaching, within a genus or mono-phyletic Group of primates may be traced backmost expressively, when we examine the distribu-tions, speciation and radiation of all extant UakariMonkeys genus Cacajao (Figs. 23, 24). This ex-clusively Amazonian genus contains two mono-phyletic Groups or Clades: the Black-headedUakaris of the Cac. melanocephalusGroup, and theBald-headed Uakaris of the Cac. calvus Group(Hershkovitz, 1987a). Among Uakaris, the sup-posedly nearest to archetypic (prototypic) ancestralform is represented by the extant Black-headedUakaris, more in particular by the saturated eu-melanin, all-black taxon Cac. ayresi - the north-easternmost distributed among all Uakaris.Uakaris are the only monkeys in the Neotropics

that lost a functional tail. All other genera evolvedeither a long pendulous, short-hairy to bushy tailthat is in the first place designed to use for balancewhile moving through the tree tops; or, a long pre-hensile tail that is used as a fifth limb during verticalclimbing and walking on top of or brachiating un-derneath twigs and thin branches in the peripheryof tree tops (where the fruits are hanging). Onlyafter observing Black-headed Uakaris Cac. hosomiin the wild along the Rio Cauaburí and in Pico daNeblina National Park, we came to understand whythe region drained by the Rio Negro has to be con-sidered the center of dispersion for all uakaris,in other words the cradle of evolution of the genus

Cacajao. Simultaneously, we came to understandthe very reason why uakaris have lost a functionaltail, whereas in all other canopy-dwelling monkeysfrom the Amazon it seems to be a fifth limb of vitalimportance.Across the entire upper Rio Negro basin the type

of vegetation that dominates the landscape is a veryimpoverished sort of thin-stemmed savanna forest.It stands on poorly drained, highly acidic white-sand soils that are deposited on top of an imper-meable, several meters thick layer of coarserounded pebbles. This type of forest is called“caatinga-do-Rio-Negro”, for it resembles much thearid dry seasonally deciduous vegetation in largeparts of the Brazilian northeast. It seasonally floodsduring the long rainy season, but also throughoutthe year on a daily base during heavy rainstorms.Physiognomically, this forest type resembles two-storey mangrove forest, as most of its trees usepneumatophores (aerial roots) and stilt-roots tocope with frequent flooding conditions. Phytoso-ciologically, the ‘caatinga-do-Rio-Negro’ is dom-inated by trees belonging to families likeEuphorbiaceae and Apocynaceae, known for theiroften toxic latex and plant parts, most in particularfull-grown seeds. Surprisingly, this forest lackshemi-epiphytic climbing shrubs, vines, and twiners.Over geological times Black-headed uakaris seemto have co-evolved with this natural environmentthrough specializing themselves in the depredationof immature seeds. From early maturation on, theseeds are often loaded with toxic alkaloids andsecondary compounds. Uakaris have guts that arespecially adapted to neutralize these toxins. Theircanines are oversized and wedge-shaped withrazorblade sharp edges, as such adapted to open upthe toughest-husked fruits and kernels (endocarps)around. Their incisors are procumbent and used toscoop out the seed content (endosperm) from anyendocarp or pericarp. Uakaris are full-fashionedseed predators to such length that, if one offers auakari a juicy pear or apple, the monkey will in-stantly bite the pulpy pome in half with its powerfulcanines. Then, it will pick the tiny seeds from thecentral part, discard the pulp, and delicately splitthe tiny seeds one by one with their canines. In theend, it has its procumbent incisors scoop out the en-dosperm from the seed coat. Black-headed uakarisdo occupy very large home ranges. They restlesslytravel or forage in very large multi-male dominated


social groups that may contain over two-hundredmonkeys. Since their preferred habitat ‘caatinga-do-Rio-Negro’ basically lacks climbing shrubs, thetree tops are not interlinked by vines, twiners andclimbing hemi-epiphytes as they are in primaryterra firme rain forest elsewhere in the Amazon. Bylack of a walkway through the tree tops, Black-headed uakaris co-evolutionarily have adaptedto this ancient impoverished, physiognomicallydiscontinuous and frequently inundated forest typeby developing the locomotor pattern of so-called‘vertical clinging and leaping’. A traveling or for-aging troop of Black-headed uakaris much re-sembles Madagascar indris, Indri indri (Gmelin,1788) that also make enormous leaps, catapultingthemselves for - and upwards by means of theirstrong muscular upper legs. Like indris in Mada-gascar, black-headed uakaris lost most of a func-tional tail while adapting to this type of locomotion.The few cm long tail provided with a tuft is onlyused for intragroup communication. Black-headeduakaris can curl it upwards and wave it sidewayslike dogs would do with a largely amputated tail.Black-headed uakaris of the species Cac. ho-

somi and Cac. ayresi, distributed north of the RioNegro, east of the Cassiquiare and west of the RioDemeni, and Cac. melanocephalus from south ofthe Río Orinoco, west of the Cassiquiare and northof the lower Rio Solimões and Rio Japurá/RíoCaquetá, have a pitch-black face, a black, forwarddirected hair-tuft on the forehead, and a short black-ish, red or orange-tinged tail. Black-headed uakarisfrom the Rio Içana basin, being distributed in-between the upper Río Orinoco and the lower RioUaupés, show a black upper back and pheomelaninbleached, orange to blond bleached lower back.Perhaps, for that reason they should be taxonom-ically treated as a valid species (we here suggestCac. ouakary).The Bald-headed Uakaris of the Cac. calvus

cladistic Group, which range south of the Amazon/Solimões and Japurá Rivers, have a bald head,bright-red bare face, blue-gray eyes, a shaggy pheo-melanin bleached, near-albinotic coat, and a rudi-mentary tail that is shorter and even less functionalthan the tail of Black-headed Uakaris (Figs. 23, 24).The Cac. calvus Group contains five taxa whichaccording to our phylogenetic ecospecies concept(ESC) should be all given valid species status: 1)Cac. calvus living exclusively in the white-water

floodplain forests (várzeas) between the lowerJapurá and Solimões Rivers, being cream-whitewith pheomelanin bleached, orange-brown ventralparts; 2) Cac. novaesi occuring in disjunct pocketsalong both banks of the lower and middle Rio Juruáas far upstream as its confluence with the Rio Ta-rauacá, its coat being pheomelanin bleached, orangebrown-colored, but albinotic from the back of thehead to mid-dorsum; 3) Cac. rubicundus, the pheo-melanin bleached, bright orange-colored (except forthe albinotic cream-white back of the head andneck) bald-headed Cac. uakaris that occurs in thewhite-water floodplain forests (várzeas) along theleft bank of the upper Rio Solimões in the central-westernmost Brazilian Amazon; 4) Cac. ucayalii,its coat overall saturated pheomelanin, darkbrown to orangish colored, ranging in the PeruvianAmazon along the right bank of the Ucayali Riverin the white-water inundated floodplain forest(várzea) as well as adjacent terra firme rain forest;5) a form newly identified by us in the year 2000,its coat near-albinotic, advanced euchromicbleached to all-white. We provisionally name thisnew taxon the “Rio Pauiní Bald-headed Uakari”Cacajao sp., for it is only found in the várzeas alongthe upper Rio Pauiní, a left-bank tributary of the RioPurús (Figs. 23, 24).The ‘trend to allopatry in metachromic varieties

of sociable, but territorial primates’ applies to theevolutionary path along which a certain primaterace, species, monophyletic clade, or genus has ex-tended its geographic range in the geological past.As a founder - colony or - population at the outerlimit of a taxon’s current range represents an ex-tremely narrow gene pool, through inbreedingcertain phenotypic characters like partial depilationof the skin, or skin/coat coloration will be rein-forced in the beginning and therefore advance morerapidly.Through the process of metachromism (= evol-

utionary change in tegumentary or hair/skin color-ation), with the ‘trend to allopatry’ in metachromicbleached individuals as the principal behavioraldriving force, speciation, radiation, and phylogeo-graphy can be plausibly retraced and explained forin all extant Neotropical primates. According to theprinciple of metachromic bleaching, primate taxa atthe base of a phylogenetic tree or clade being thenearest to archetypic, prototypic, primitive, or ori-ginal, in general are agouti or saturated eumelanin,


Figure 23. Schematic map of the distributions of Uakari Monkeys of the Cacajao melanocephalus and C. calvus Cladesdivided up by (for them) untraversable rivers. Figure 24. Distributions, allopatric speciation, radiation and supposed pathwaysof metachromic bleaching in all known Uakari Monkeys genus Cacajao.


which means the least colorful, agouti, black, ordark brown colored. Among Uakari Monkeys genusCacajao, the origin or center of dispersion is sup-posed to be located in the northeasternmost partof the Brazilian Amazon, south of the watershedbetween the Rio Negro and Río Orinoco basins, anarea delineated by the Rios Demeni and Aracá(Figs. 23, 24; Fig. 27). Within this interfluvethe landscape is dominated by ‘caatinga-do-Rio-Negro’, the most impoverished habitat type ima-ginable, but preferred by uakaris of the Cac. mela-nocephalus Clade. Here lives the saturated eu-melanin, least bleached taxon of the Black-headedUakaris, the recently described Cac. ayresi (Boubliet al., 2008). Its coat is all-black and dark-browncolored. It may well represent the proto- or ar-chetypic uakari from which all other uakaris havederived. From Cac. ayresi in western direction firstdiverged along the pheomelanin pathway taxonCac. hosomi. It is distributed between the RioMarauiá, the upper Rio Negro, and the CassiquiareChannel (we have confirmed its presence in Pico daNeblina National Park and along both banks of theRio Cauaburí). After an ancestral founder-colonytraversed the Rio Cassiquiare - the channel thatconnects the Rio Negro basin with that of the RíoOrinoco in Venezuela - Cac. hosomi diverged intoan intermediately pheomelanin bleached taxon thatdiffers from classic Cac. melanocephalus in theblack shoulders, dark-red legs and tail. If thisphenotype, which is thought to represent a colormorph of Cac. melanocephalus, turns out to occurthroughout the entire distribution delineated in thenorth by the Río Orinoco and in the south by theRio Uaupés, one should consider it a new taxon tobe named the “Rio Içana Black-headed Uakari”Cac. ouakary. After an ancestral founder-colony ofthe latter managed to traverse the Rio Uaupés, ithas diverged into the progressively pheomelaninbleached blond- backed black-headed uakari taxonCac. melanocephalus. Subsequently, blond-backedCac. melanocephalus have occupied the entireinterfluve south of the Rio Negro, eastwards as faras Archipelago de Anavilhanas located about fortykm west of Manaus, and to the west far into theColombian Amazon, and south as far as the northbank of the Rio Japurá (Río Caquetá in Colombia).We suppose that once upon a time a founder-colonyof slightly bald-headed, advanced pheomelaninbleached ancestral Cac. melanocephalus, being

pushed out of its westernmost dead-end distributionin the Colombian Amazon, may have managed totraverse the upper reaches of the Río Caquetá. Itthen could extend its range southwards, eventuallyreaching the Río Marañón (as the upper AmazonRiver is called in Peru). A founder - colony ofan advanced pheomelanin bleached, bald-headedancestral form must then have traversed the RíoUcayali. It subsequently occupied terra firme andvárzea forests in the interfluve between the RíoUcayali in the west, the Río Marañón in the north,and the Rio Javarí in the east. Nowadays, this inter-fluve is inhabited by the bald-headed dark reddish-brown taxon Cac. ucayalii that belongs to thebald-headed Cac. calvus Clade. Disjunct from Cac.ucayalii’s distribution and ranging farther to the eastderived taxon Cac. rubicundus, a progressivelypheomelanin bleached bright- orange colored bald-headed uakari. It is fully adapted to várzea habitatfound in abundance along the left bank of the upperRio Solimões. From Cac. rubicundus going farthereastwards, but disjunct from its distribution, alongthe same (left) bank of the Solimões/Amazon Riverthe almost fully albinotic taxon Cac. calvus isfound. It fully adapted to white - water inundatedfloodplain forest (várzea) - the only availablehabitat in this for the Cac. calvus Clade dead-enddistribution situated inbetween the banks of theJapurá and Solimões Rivers. Directly from Cac.rubicundus to the south of Cac. calvus’ distributionderived the bald-headed taxon Cac. novaesi thatranges along both banks of the Rio Juruá as farsouth as the confluence with the Rio Tarauacá andRio Envira. This taxon is near-albinotic from theback of the head to beyond the mid-dorsum, andprogressively pheomelanin bleached light orange-brown on the lateral and ventral parts of the body.It ranges in the várzeas of the floodplain, butwe have also spotted large troops foraging forimmature seeds in the adjacent terra firme rainforest. In 2000, we identified a fifth taxon of bald-

headed uakari, the completely white, fully al-binotic taxon that we named “Rio Pauiní Bald-headed Uakari” Cacajao sp. It lives along the southbank of the Rio Pauiní, a left-bank tributary ofthe upper Rio Purús. It represents the southern-most distributed and the farthest pheomelaninbleached most albinotic taxon of all extant uakaris.It lacks the pheomelanin orange-brown to orange


ventral parts seen in the other near-albinotic taxaCac. novaesi and Cac. calvus (Figs. 23, 24).Analyzing metachromic skin and coat characters

as linear and irreversible progressions within Neo-tropical primate genera and their monophyleticClades does add substantially to the reconstructionof biogeographic divergence events and phylo-genetic relationships over a wide range of Neo-tropical primate taxa, in particular those that defendtheir living space or ecological (feeding) nichethrough male-dominated, hierarchically organizedsocieties. So it does to the Bearded Sakis genusChiropotes (Figs. 25, 26) even if we have confirmedin the field that social groups of (at least) theGuianan taxon Ch. sagulatus do freely fuse andfission on a regular base with neighboring socialgroups. The genus Chiropotes clearly shows sexualdimorphism in the larger, more robust males thatalso grow bigger beards and frontal hair lobes ontheir heads (Hershkovitz, 1985). During foragingand resting, a large social group of bearded sakis,similar to woolly monkeys, consists of severalpolygamous dominant males each taking care of his‘harem’. The high-ranking males tend to stick to thecenter of the foraging troop, whereas lower rankingmales with or without harems are pushed closer tothe periphery of the foraging troop. This way, adultmales do avoid confrontations, for their impress-ively large wedge-shaped canines designed to crackhard-husked fruits and kernels in order to get to theseed pulp would be lethal if used in fights. But ad-olescent, subadult, and, we assume, also behavior-ally or phenotypically deviant individual males maywell be pushed into the periphery of the foragingand ranging troop. More than once, we haveencountered a solitary male, or a couple of malestraveling at high speed through the canopy in anapparently fixed direction, leaving us no means todetermine if these monkeys only temporarily hadlost contact with the troop, or if they were expelledfrom the parental troop, or if they were representingsubtly deviant young males that had been forced toleave the pack and search for new living groundssomewhere beyond the limits of the group’s homerange. Only through long-term field studies onewould be able to obtain clear answers to this sort ofquestions.Within the Bearded Sakis genus Chiropotes

we distinguish two monophyletic Groups: the Ch.satanas and the monotypic Ch. albinasus Group.

The Ch. satanas Clade consists of five taxa, amongwhich the nominate species Ch. satanas representsthe saturated eumelanin, all-black, nearest to ar-chetypic bearded saki. Its distribution in the NEBrazilian state of Maranhão is assumed to representthe cradle of evolution or center of dispersion forthe genus. An equally all-black form that we re-cently identified west of the headwaters of the RioXingú (e.g., Rios Ronuro, Batovi and VondenSteinen) may either represent an enclave populationthat became disjunct from that of Ch. satanas(ranging east of the Rio Pará/lower Rio Tocantins),or a new taxon of the Ch. satanas Clade that stillhas to be collected and described. From Ch. satanasdiverged in western direction the slightly eumelaninbleached, overall light-brown colored taxon Ch.utahicki. It occupies the entire interfluve delin-eated by the Rios Amazonas/Anapú/Tocantins-Araguaia/Xingú. An ancestral founder-colony ofsomewhat pheomelanin bleached, red to orange-brown backed Ch. utahicki once must have man-aged to traverse the lower Rio Amazonas, fromwhich then derived taxon Ch. sagulatus. Thisspecies occupies the entire area north of the AmazonRiver and east of the Rio Branco, including most ofthe Guianas east of the Essequibo River. This taxonis absent from most of Amapá state, French Guianaand also from a wide coastal belt of the Guianas. Afounder-colony of ancestral sagulatus once musthave traversed the Rio Branco and radiated inwestern direction diverging into the advancedeumelanin bleached taxon Ch. israelita. Thisspecies is characterized by the albinotic (whiteinstead of pink) genitals and the light-grayish tobrownish coat color of the trunk. Chiropotes israel-ita ranges west of the Rio Branco as far north as theRío Orinoco in Venezuela. It seems to be parapatricwith Black-headed Uakaris, as Chiropotes is a seed-predating terra firme rainforest specialist, and Cac.ayresi and Cac. hosomi are ‘caatinga-do-Rio-Negro’-habitat specialists. The Rios Marauiá andCauaburí seem to divide their distributions. Our ex-tensive surveys in the Rios Demeni/Aracá inter-fluve did not reveal the occurrence of Ch. israelita,as the landscape is dominated by ‘caatinga-do-Rio-Negro’ habitat (Figs. 26, 27).The second monophyletic Group of Bearded

Sakis is that of monotypic Ch. albinasus. The Red-nosed Bearded Saki is very different from the Ch.satanas Clade, not just in metachromic sense. Its


Figure 25. The hitherto recognized taxa of Bearded Sakis genus Chiropotes (above) and Uakaris genus Cacajao (below), all depicted in one plate (Courtesy of Stephen Nash).


vocalizations are very different, the beard and tailare shorter-haired, and the genitals of each genderare brightly red-colored as is the muzzle (the sci-entific name Ch. albinasus - Latin for “white nose”-relates to the taxonomist, who may never have seenthe monkey he described alive. Furthermore, groupsize in Ch. albinasus is much larger than that of anyof the taxa belonging to the Ch. satanas Clade,ranging on average from 30–80 individuals. WhereCh. albinasus occurs in sympatry with woolly mon-keys (i.e., west of the Rio Tapajós-Juruena, east ofthe Rio Madeira, and north of the Rio Jí-Paraná),they are often seen in mixed species associations.Red-nosed saki groupings mixed with woolly mon-keys (Lagothrix cana), tufted capuchins (Sapajusapella) and/or white-fronted slender capuchins(Cebus unicolor) may contain as many as 150 mon-keys.

In figure 28, we have depicted the distributions,allopatric speciation, radiation, and supposedlyfollowed pathways of metachromic bleaching in allknown Saki Monkeys genus Pithecia. Sakis occurexclusively in the rain forests of lowland Amazoniaand the Guayanan Shield (Hershkovitz, 1987b;Mittermeier et al., 2013). Within the genus Pitheciawe distinguish three monophyletic cladistic Groups:P. monachus, P. pithecia, and P. hirsuta (Fig. 29).Within the P. monachus Clade allopatric specia-

tion is thought to have followed evolutionary path-ways of metachromic bleaching with P. monachusrepresenting the nearest to archetypic precursor ofall extant sakis. Both sexes have an overall satur-ated eumelanin, slightly bleached silky coat, exceptfor the cream-white hands and feet. Taxon P. mon-achus ranges along both sides of the Amazon up-stream from its confluence with the Rios Juruá and

Figure 26. Distributions, allopatric speciation, radiation, and supposedly followed eumelanin pathwaysof metachromic bleaching in all known Bearded Sakis genus Chiropotes.


Japurá, large rivers delineating its distribution in theeast and north. The species is sexually dimorphic,not in size but in metachromic pelage characters ofthe head. Both sexes have a slightly bleached maskthat is light brown in males and cream-white in fe-males. It surrounds a black face with yellow tocream eyebrows and malar stripes. Forehead andcheeks are covered with short, forward directedhairs resembling much that of members of the P.pithecia Group. From P. monachus diverged innorthwestern direction taxon P. milleri, supposedlyafter a metachromic deviant founder-colony ofancestral monachus traversed the Río Caquetá.Pithecia milleri nowadays occupies a small part ofthe Colombian Amazon that is confined by the RíosCaguán and Putumayo. Both sexes are overall eu-melanin bleached, more so in females. The foreheadis covered with long, forward directed hairs forming

a kind of hood that is yellowish in males and cream-white in females. The black muzzle is contrastedwith the advanced euchromic malar and lip stripes.From P. milleri derived the taxon P. napensis aftera founder-colony of P. milleri traversed the RíoPutumayo in southern direction. Pithecia napen-sis occupies a small area in the Colombian andEcuadorian Amazon delineated by the Río Putu-mayo in the north and the Río Napo in the south. InP. napensis both sexes are progressively pheo-melanin bleached in the yellowish to orange breast,more so in males that also differ in the silverygrayish lower part of a well-defined mask and in thealbinotic hood. After a founder-colony of ancestralP. napensis once traversed the Río Napo to thesouth, the progressively pheomelanin bleachedtaxon P. aequatorialis diverged. It occupies a largearea in the Ecuadorian and Peruvian Amazon delin-

Figure 27. Map showing distributions of the Bearded Saki taxa Chiropotes sagulatus and C. israelita, and the parapatric Black-headed Uakaris that occur north of the Amazon and Negro Rivers.


eated in the north by the Río Napo and in the southby the Río Tigre. Pithecia aequatorialis, in particu-lar in the metachromic characters of the male’s head(fully albinotic mask) and (orange) breast pelage,represents the most advanced pheomelanin bleachedtaxon in the P. monachus Clade. Its dead-end distri-bution at the end of the phylogeographic radiationof the P. monachus Clade is confined at all butwestern (Andean Mountain range) sides by P. mon-achus occupied territory. We may ponder aboutwhat would be the result of any hypothetical hy-bridization between P. aequatorialis females and P.monachus males at the contact zone that shouldexist in the species’ westernmost distribution. Evenif the offspring would remain fertile, it would neverresult in parapatric speciation. In concurrence withour theory, deviant young males with metachromicgenes from P. aequatorialis would be expelled by

the dominant male(s) of the P. monachus parentalgroup, back to P. aequatorialis territory.Within the P. pithecia Clade we consider P.

lotichiusi with the overall darkest agouti (in fe-males) and saturated eumelanin black (in males)pelage the nearest to archetypic taxon. This taxonis only found in the easternmost part of the inter-fluvial peninsula between the lower Solimões andNegro Rivers, from opposite the city of Manaus asfar west as the towns of Manacapurú and NovoAirão. In the past, P. lotichiusi may have occupieda much larger distribution, for no untraversablegeographic barriers exist when going further westinto the Rios Solimões/Negro interfluve. If so, theP. pithecia Clade may have monophyleticallyderived from the P. monachus Clade, when that ra-diated to the east. A founder-colony of slightlypheomelanin bleached ancestral P. monachus may

Figure 28. Distributions, allopatric speciation, radiation, and supposedly followed metachromic pathways of bleaching in all known Saki Monkeys genus Pithecia.


have traversed the lower Rio Japurá and thereafterdiverged into the allopatric taxon P. lothichiusi. Thelatter then extended its range to the east. During oneof the late-Pleistocene glacials, when ocean levelsdropped over up to 120 m, a founder-colony of P.lotichiusi could well have traversed the lowerRio Negro and then reached the north bank of theAmazon. This way, it may have diverged into theallopatric Golden-faced Saki taxon P. chryso-cephala. Nowadays, Golden-faced sakis range fromthe Rio Branco as far east as the Rio Trombetas.After a founder-colony of ancestral P. chryso-cephala once traversed the Rio Trombetas, taxonP. pithecia may have diverged. Pithecia pitheciathen expanded its range in northwestern directionacross the states of Roraima, Pará and Amapá, and

across the Guianas into Venezuela as far west as thelower Río Orinoco. It may have circumventedeither side of the watershed formed by the TumacHumac Mountains. Within the sexual dimorphic P.pithecia Clade, females are progressively pheo-melanin bleached orange to yellowish brown, whe-reas males are all-black with a progressively pheo-melanin bleached to albinotic mask. In the Braziliantaxa P. lotichiusi and P. chrysocephala the mask thatconsists of short, stiff, forward directed hairs isgolden to orange-yellow colored. In the Guiananwhite-faced saki P. pithecia the mask is albinotic,white with orange-colored cheeks in males fromGuyana and Suriname, and overall white in malesfrom French Guiana.Sakis of the P. monachus and P. pithecia Clades

Figure 29. Among the Saki Monkeys genus Pithecia three monophyletic cladistic Groups or Clades are distinguished: theP. monachus Group containing four taxa (P. monachus, P. milleri, P. napensis, and P. aequatorialis), the P. pithecia Groupcontaining three taxa (P. lotichiusi, P. chrysocephala, and P. pithecia), and the P. hirsuta Group containing four taxa(P. hirsuta, P. irrorata, P. vanzolinii, and P. albicans). We only recognize sexual dimorphism as expressed in metachromiccharacters in the monachus and pithecia Clades (Courtesy of Stephen Nash).


distinguish themselves locomotorily from sakis ofthe third clade - the P. hirsuta Clade. A specificlocomotor pattern called “vertical leaping andclinging” is performed during foraging and travel-ing in their preferred habitat, which is the discon-tinuous lower canopy and understory of terra firmerain forest. As these sakis have to leap from treetrunk to tree trunk, they are commonly known as“flying monkeys”. In contrast, saki taxa of the P.hirsuta Clade prefer the middle to upper strata ofprimary rain and seasonally inundated floodplainforests, which strata are interconnected by thick-stemmed vines and hemi-epiphytic climbingshrubs. For that preferred habitat they have adopteda different locomotor pattern, that of horizontal

leaping, and quadrupedal running or hopping acrossthick horizontal branches and boughs. A significantdifference in limb proportions between taxa belong-ing to each of the two Clades has been measured,with those of the P. pithecia Group being longerrelative to trunk length (Hershkovitz, 1987a;1987b). Another important feature in which the P.hirsuta Clade distinguishes itself from the P. mon-achus and P. pithecia Clades is mean group sizeand sexual dimorphism. Social groups of taxa be-longing to the P. hirsuta Clade are larger and multi-male structured, instead of the extended familygroup that contains only one or sometimes twoadult males in taxa belonging to the other Clades.Moreover, contrary to what recent taxonomies

Figure 30. Satellite image taken from the region, where the várzea floodplain of the Rio Solimões borders on that of the RioPurús. Behind each floodplain are located black-water backwater lakes (rias), such as Lago Coarí, Lago Uauaçú, and LagoAyapuá. A red line indicates where parapatric bald-faced saki Pithecia hirsuta is encroaching onto buffy saki P. albicansterritory. (Below) Portraits of different adult males of Gray’s saki P. hirsuta. (Above, left) White-masked mutant male P.hirsuta that was seen roaming around alone far into P. albicans territory north of Lago Uauaçú. (Above, right) Adult malebuffy saki P. albicans; note the black face with the showy albinotic eyebrows and white long-haired hood.


(merely based on museum collections) suggest, wewere not able to recognize metachromic sexualdimorphism in any taxon of the P. hirsuta Clade. Inthe field, we failed to distinguish gender amonggroup members of P. hirsuta, P. irrorata, and P.albicans. Nor could we, in captivity, determine theirgender without up-close examining the saki mon-key’s concealed genitals.Within the Bare-faced Sakis of the P. hirsuta

Clade we suggest the least eumelanin bleachedoverall blackish-gray taxon P. hirsuta to be thenearest to archetypic taxon. It may well have de-rived from a founder-colony of proto-monachusthat once traversed the Rio Juruá in eastern direc-tion. The following pathways of metachromic blea-ching and allopatric speciation are recognized.From P. hirsuta that occupies the entire interfluvedelineated by the Juruá, Solimões and Madeira Ri-vers, diverged and radiated away in eastern direc-tion taxon P. irrorata after a founder-colony ofprogressively bleached P. hirsuta traversed or cir-cumvented the Rio Madeira (most likely at its upperreaches) during one of the late-Pleistocene glacials.Nowadays, taxon P. irrorata occupies the entireinterfluve delineated by the Madeira, Amazonasand Tapajós-Juruena Rivers. Its overall coat isadvanced eumelanin bleached in comparison withthat of P. hirsuta, and albinotic in the distal half ofthe hood, the hands and feet. Its tail is less bushy,the hairs more curly. Pithecia irrorata has an almostbare face, and its forehead is only halfway coveredby an albinotic hood that does not conceal thecheeks and temples. As a result, the monkey’sprofile looks more pronounced. Metachromic skinand fur characters of the head that play such animportant role in the taxonomy of monkeys likePithecia, Sapajus and Ateles are often poorly pre-served in museum specimens. Hence, the confusionin most hitherto elaborated taxonomic reviews ofthese genera. Zoological collections all over theworld have lumped misidentified taxa, such as P.hirsuta and P. irrorata, under the latter. Some lead-ing taxonomists even attribute sexual dimorphismto the Bare-faced Sakis. From P. hirsuta to the westdiverged taxon P. vanzolinii, after a progressivelybleached founder-colony of P. hirsuta traversed theRio Envira. Pithecia vanzolinii is now confined tothe headwaters of the Rio Juruá. It differs in the al-binotic lower limbs and ventral parts that contrastmuch with the blackish-gray dorsal parts and tail.

From P. hirsuta to the north derived the overallnear-albinotic taxon P. albicans that is pheomelaninbleached orangish-yellow only on the lower limbs.Buffy Sakis P. albicans occupy the northernmostdead-end distribution of the P. hirsuta Clade, whichis delineated by the untraversable lower SolimõesRiver in the north, the lower Juruá River in the west,and the lower Purús River in the east. Buffy Sakisare parapatric with the more opportunistic Gray’sSakis P. hirsuta, from which they once derived. Atits southern limit, its distribution shows an open endrunning across the Rio Tapauá axis. After it tra-versed the Rio Tapauá to the north, Gray’s SakiP. hirsuta was, and still is expanding its rangenorthwards to the cost of the Buffy Saki P. albicans.This example may well demonstrate that progress-ively bleached to albinotic primate taxa that occupydead-end distributions will eventually go extinct.East of the Rio Coarí and north of the Rio Tapauá -a left-bank tributary of the Rio Purús - we have con-firmed the sympatric occurrence of the taxa P. al-bicans and P. hirsuta, with P. hirsuta advancingonto P. albicans as far north as Lago Ayapuá (Fig.30). North of the Ayapuá contact zone in territoryexclusively occupied by P. albicans, we once spot-ted and photographed a solitary young male, itshead pelage resembling that of male White-facedSaki P. pithecia from the Guianas (Fig. 30). We as-sume that this male was a progressively bleacheddeviant color morph of taxon P. hirsuta that wasexpelled from or forced to leave its parental group.It may have ventured into adjacent P. albicansterritory north of Lago Uauaçú. As we have oftenseen P. hirsuta groups opportunistically penetratingfar into white-water floodplain forest (várzea), thismetachromic deviant near-albinotic, sexuallydimorphic mutant male of taxon P. hirsuta in theorycould become the founding father of a new taxon.This could happen after this young male wouldhave attracted one or a few P. albicans females toform a small reproductive family group. It thenwould have to survive making a year-round livingin the extensive várzeas found along the south bankof the Rio Solimões. We have never seen any saki,uakari or other seed-predating monkey occupyingthat ecological feeding niche in the várzeas thatfringe the right bank of the middle Rio Solimões.Perhaps, this hypothetical scenario may also explainhow metachromic sexual dimorphism in primatescould have evolved.


In figure 31, we have visualized the distribu-tions, allopatric speciation, radiation and sup-posedly followed pathways of metachromic bleach-ing in all known Woolly Monkeys, genus Lago-thrix. Woolly monkeys are exclusive matrix terrafirme rainforest dwellers that under normal circum-stances will never enter white-water floodplainforest (várzea). For that reason alone, the distribu-tion of Lagothrix is greatly determined by riverinebarriers. Within the genus only one monophyleticClade is recognized. We consider the saturated eu-melanin, metachromic least bleached Poeppig’sWoolly Monkey taxon La. poeppigiiwith its overallblack to dark chestnut-brown coat the nearest to ar-chetypic woolly monkey. In the north, La. poeppi-gii’s distribution is confined by the Amazon River,in the east by the Rio Juruá that is also fringed withextensive várzeas, and in the south and west by the

foothills of the Andean Mountain range. From La.poeppigii derived in western direction the PeruvianYellow-tailed Woolly Monkey La. flavicauda,which has (disputedly) been upgraded to its owngenus Oreonax. It occurs in parapatry with La.poeppigii, but genetically isolated from it, as it livesin high-altitude Andean cloud forest. With its al-binotic lower half of the circumocular rings, facialmuzzle, chin and pheomelanin bleached yellow tailthe taxon is following a pheomelanin pathwaytowards albinotic. From a founder-colony of some-what eumelanin bleached La. poeppigii that tra-versed or circumvented the upper Rio Juruá andthen radiated to the east and north, the darkbrownto black headed taxon La. tschudii derived. Its coatis overall dark gray-brown colored, becomingblackish on all five limbs. It occupies the entireinterfluve delineated by the Juruá, Solimões-

Figure 31. Distributions, allopatric speciation, radiation, and supposed eumelanin pathways of metachromic bleaching in all known Woolly Monkeys genus Lagothrix.


Amazonas and Madeira Rivers. From La. tschudiiin eastern direction diverged the Black-headed orGeoffroy’s Gray Woolly Monkey taxon La. cana,its entire coat progressively eumelanin bleached,light-gray colored, with a dark-gray to black head.Only as recent as the late- Pleistocene or earlyHolocene, an advanced eumelanin bleachedfounder-colony of La. tschudiimust have traversedor circumvented the upper Madeira River north ofthe Rio Jí-Paraná (also known as Rio Machado) ineastern direction. It then extended its range bypassing the geographic barrier formed by the ex-tensive Tenharim Savanna in Rondonia alongsideits southern border. This way, it could enter theinterfluve delineated by the Madeira, Amazonasand Tapajós Rivers. Circumventing the extensiveTenharim Savanna, taxon La. cana apparentlymissed the narrow entrance to the north that existsbetween the upper Rio Jí-Paraná and the RioRoosevelt. This could well explain why woollymonkeys are absent from the entire RiosMadeira/Aripuanã interfluve north of the RioMarmelos. The relatively recent occupation by La.cana of the entire interfluve delineated by theMadeira, Aripuanã, Amazonas and Tapajós Riversis near to its completion. Taxon La. cana’s currentnorthernmost distribution gets to a halt at the lat-itude running across the upper reaches of theAbacaxis and Andirá Rivers, not much south of theuntraversable Rio Amazonas. We assume that onlywhen La. cana invaded all smaller interfluves eastof the Rio Aripuanã and west of the Rio Tapajós, itbegan to displace the All-black Woolly Monkey thatin the far geological past evolved in the area east ofthe (proto)-Madeira River. This newly identifiedwoolly monkey still has to be collected and de-scribed. We here provisionally allocate the commonname “Rio Aripuanã Black Woolly Monkey” to thisfully saturated eumelanin, all-black taxon. Ap-parently, as it occupies the same ecological nicheas newcomer La. cana, the Rio Aripuanã BlackWoolly Monkey finds itself on the verge of ex-tinction. It is smaller, lives in small, socially lesscomplex family groups, and its coat is in meta-chromic respect the most primitive or archetypic. Itlives in sympatry with La. cana, but only hangs onin a small enclave distribution situated between thelower to middle Rio Aripuanã and the Rio Acarí. Itmay well represent the ancient, most original, ar-chetypic taxon of all Woolly Monkeys genus Lago-

thrix that evolved in the Late-Pliocene east of theproto-Madeira River, fully isolated from the rest ofthe Amazon.Woolly monkeys also radiated into the north-

western Amazon, most likely after a founder-colonyof taxon La. poeppigii circumvented or traversedthe upper Amazon River in Peru (where it is calledRío Marañón). Two progressively eumelaninbleached forms that derived from La. poeppigiionce must have occupied the Colombian Amazon:the euchromic light-gray Colombian Woolly Mon-key taxon La. lugens that occurs at high altitudes inthe foothills of the South-Colombian Andes and inthe upper Río Magdalena valley, and the Brown orHumboldt’s Woolly Monkey taxon La. lagotricha.The coat of taxon La. lugens is eumelanin bleachedcharcoal to light-gray colored, but lacks any mix-ture with brown. On the head, a mid-dorsal stripeand a rim across the eyebrows are advancedbleached to euchromic. Mean body size and weightin La. lugens are the largest among all extant woollymonkeys. Lagotricha’s coat is progressively eu-melanin bleached light-brown colored, except forthe blackish hands and feet. Its head is light-browncolored, with a slightly bleached yellowish eyebrowrim and sideburns aside of the blackish-brown face.Taxon La. lagothricha ranges across the Colom-bian, Venezuelan and NW Brazilian Amazon.Most interestingly, we confirmed the small

distribution of a newly identified, advanced pheo-melanin bleached, overall orange-colored taxon inthe upper reaches of the Rio Jutaí. A founder-colonyof advanced pheomelanin bleached La. poeppigiimutants pushed out of La. poeppigii territory mustonce have successfully adapted to white-waterseasonally inundated floodplain forest (várzea)located between the east bank of the upper Rio Jutaíand the west bank of the Rio Juruá, near the townof Eirunepé. We were not able to determine theexact range of the Rio Jutaí Woolly Monkey, for thearea is inhabited by uncontacted Amerindians of theKorubo tribe (so-called “caçeteiros”) that areknown to kill any non-indigenous intruder.We encountered in the zoological collection of

the Brazilian Museu Goeldi (MPEG, Belém-PA) anoverall orange-colored stuffed juvenile specimenthat was deposited without collecting data. Thisvery animal is depicted in Da Cruz Lima’s 1945Mammals of Amazônia. We here provisionallyname it the “Rio Jutaí Orange Woolly Monkey”.


In addition, we found an albinotic overallcream-colored taxon that we provisionally namedthe “Rio Javarí Fair Woolly Monkey” Lagothrix sp.It resembles much Humboldt’s Woolly Monkeytaxon La. lagotricha, but its pelage is longer, softerand silky, besides being overall advanced eu-chromic to cream-white colored. It has long-hairedwhite sideburns alongside a pitch-black face,muzzle and chin. A near-albinotic ancestral founder-colony must once have been driven out of La.poeppigii territory somewhere near the northern-most border of its distribution. This colony musthave been forced to make a living in the white-water floodplain forests (várzeas) that stretch outalong the south bank of the Rio Solimões (near thetown of Tabatinga) all the way to the left-bankvárzeas of the lower Rio Javarí. Under normal cir-cumstances this type of habitat should be conside-

red inappropriate for woolly monkeys to guaranteea durable and sustainable living. This seems to beanother case where a progressively bleached, near-albinotic founder-colony of La. poeppigii has beendriven into a (for woolly monkeys) marginal habitat- seasonally white-water inundated floodplain forest(várzea). According to our theory of allopatricprimate speciation, albinotic fair woolly monkeysmust have diverged this way from archetypic, sat-urated eumelanin, dark brown coated La. poeppigii.Apparently, it has survived until today in geo-graphic sympatry, but ecological parapatry (inhab-iting adjacent but different habitats) with taxon La.poeppigii, the species it derived from. In 2002, thesecond author, while at Colombia University, NY,ran the mtDNA sequences of the Rio Javarí FairWoolly Monkey using earlier preserved DNA-samples. He found 4% divergence from sympatric

Figure 32. Distributions, allopatric speciation, radiation, and supposed pathways of metachromic bleaching followed inall known Spider Monkeys genus Ateles that occur in the Amazon and along the Pacific coast of Ecuador and Colombia.


La. poeppigii and over 7% from the allopatric taxonLa. lagotricha. The AMNH holds three well-preserved skins of the Rio Javarí Fair Woolly Mon-key Lagotrix sp., which were collected by the OlallaBrothers in 1927 along the south bank of the RioSolimões, somewhat upstream from the town ofTabatinga. All three specimens are misidentified asLa. lagotricha (Humboldt, 1812).For Spider Monkeys genus Ateles, allopatric

speciation, radiation, and phylogeography alongdifferent pathways of metachromic bleaching aredepicted in figures 32, 33. Four monophyleticcladistic Groups or Clades are recognized: A. pan-iscus, A. chamek, A. belzebuth, and A. geoffroyi.Spider monkeys have evolved during the Pliocenein the Guayanan Shield, most likely from a pre-cursor of the most ancient of the four extant mono-phyletic cladistic Groups, the A. paniscus Clade.The Red-faced Black Spider Monkey A. paniscus

from the Guianas represents the nearest to ar-chetypic extant taxon within the genus. This as-sumption is based on some unique primitivecharacters that are not seen in other spider monkeys.Here we mention: the presence of a vestigial thumbor, if lacking, at least the metacarpal of the first digitthat is maintained in the hand; its incapacity ofusing the tip of the prehensile tail in picking andmanipulating small objects like food items; theoverall long-haired coat, in particular around thebase of the tail and in the forward directed hairtufton the forehead that resembles a cap; the overallsaturated eumelanin black coat without any sign ofearly eumelanin bleaching; the advanced pheo-melanin bleached bright-red bare face lackingwhiskers; the frequent occurrence of albinotic blue-colored eyes; the albinotic cream-white colored,hypertrophied, pendulous clitoris in females andcream-white protruded anus in both sexes, whereas

Figure 33. Phylogeographic distribution, allopatric speciation, radiation and metachromic diversification in all known SpiderMonkeys (Ateles) that occur from the Pacific coast of W Ecuador and NW Colombia far into C America as far N Mexico.


the clitoris is long, flattened and lacking the muscu-lature to erect during foreplay and copulation (VanRoosmalen, 1985a). Moreover, spider-monkey mat-riarchal social organization is markedly expressedin (leading) female’s body size, which in A. pan-iscus may exceed that of males; and in the per-manent fusion-fission social structure centeredaround alpha-females that lead foraging parties onday ranges. As such, complete gatherings of alltwenty or so members of a social grouping willnever happen (Van Roosmalen, 1985a). This spe-cific type of social organization that is uniqueamong Neotropical primates may well be related tothe specific phytosociological composition, pheno-logy and physiognomy of the more ancient, moreheterogeneous type of primary terra firme rainforest that evolved uniquely and without majorinterruptions during the last 60–70 million years onthe Guayanan as well as on the Brazilian Shield.Here, available food sources are generally widelydispersed, and rarely clumped at any time of theyear. Maturation of nutritious large-seeded fruits -A. paniscus is a mature-fruit specialist frugivore -is slower and species-specifically stretched out overlonger periods of time (Van Roosmalen, 1985b).Mast-fruiting, as commonly seen in tropical rain-forests on other continents, is a phenomenon thatdoes not exist in this ecosystem. Hence, the earlyevolution of semi-brachiation (brachiation with thehelp of a prehensile tail) as the principal locomotorpattern, and the fusion-fission type of social struc-ture during traveling and foraging took place in an-cestral spider monkeys as the principal adaptationof a large-bodied monkey to a well-defined ecolo-gical feeding niche, in a biome that took over 60million years to develop. It may well explain whythe A. paniscus Clade did not speciate and radiateany further, as the distribution of extant A. paniscusis still confined to the larger part of the GuayananShield.Most plausibly somewhere in the late-Pliocene,

from an agouti or saturated eumelanin all-black an-cestor of A. paniscus derived the phylogeneticallydistantly related, nearest to archetypic Black SpiderMonkey taxon chamek of the A. chamek Clade. Itis distributed south of the Amazon as far south asthe Brazilian Shield (in Rondonia and Mato Grossostates). North of the Amazon, the Brown SpiderMonkey taxon A. brunneus that ranges in N Colom-bia (in an area confined by the Sierra Nevada

Mountains), may represent the least eumelaninbleached, nearest to archetypic taxon of the A.belzebuth Clade. Moreover, in the Pacific coastalforests of Ecuador and Colombia is found the sat-urated eumelanin Brown-headed Spider Monkeytaxon A. fusciceps (formerly A. fusciceps fusciceps).Along the Pacific coast of N Colombia and SPanama is found the all-black but dark red-belliedColombian Black Spider Monkey taxon A. rufiventris(formerly A. fusciceps robustus). All-black Brown-headed Spider Monkey taxon A. fusciceps maytherefore represent the nearest to archetypic taxonof the A. geoffroyi Clade (Fig. 32).Within the A. chamekClade, nominate A. chamek

represents the nearest to archetypic taxon. It is sat-urated eumelanin in its overall black coat color andblackish or slightly bleached pinkish circumocularrings and/or facial muzzle, and in the forwarddirected black hairtuft on the forehead. It rangesacross a large part of the Amazon basin delineatedby the Amazon River in the north, the Andes Moun-tains in the west, the highlands of the BrazilianShield in the south, and the Purús and GuaporéRivers in the east. Like the other taxa of the A.chamek Clade, the Black-faced Black Spider Mon-key A. chamek is only found in patches of terrafirme rain forest close to major waterbodies, suchas lakes, rivers, and creeks. It frequents in particularseasonally inundated marsh forest and black- andclear-water floodplain forest called igapó. We havenever spotted spider monkeys belonging to the A.chamek Clade in matrix primary rainforest of thehinterland at distances of over ten km from anymajor waterbody. There, spider monkeys of the A.chamek Clade are commonly replaced by woollymonkeys (Lagothrix) that occupy the same feedingniche in primary terra firme rain forest. All taxaof the A. chamek Clade do laterally migrate to thenearest igapó floodplain forest of clear- and black-water rivers during the 2–3 months lasting fruitingseason, which coincides with the peak of the flood.From A. chamek diverged and radiated away ineastern direction the Rio Purús Black Spider Mon-key that we identified to be new to science. Thistaxon ranges in the interfluve between the Purúsand Madeira Rivers, south of the Rio Ipixunaand north of the Río Tahuamanu in the BolivianAmazon, a left-bank tributary of the upper RioMadeira. The Rio Purús Black Spider MonkeyAtheles sp. is having a near-albinotic cream to pink


colored muzzle, chin, and ears, and a triangularpatch of short, backward directed black hairs on theforehead instead of a cap. After a founder-colony ofthe Rio Purús Black Spider Monkey traversed theRio Madeira to the east, the Long-limbed BlackSpider Monkey A. longimembris diverged. Thistaxon was already identified as a distinct species byDa Cruz Lima (1945) based on two specimens thatwere collected by Leo E. Miller along the upper RioJí-Paraná in Mato Grosso during the first part of the1914 Roosevelt-Rondon Expedition. It was first de-scribed as Ateles longimembris by Allen (1914).Holotype and paratype of A. longimembris depos-ited in the zoological collection of the AMNH underNo. 36909 were later misidentified as A. chamekand therefore not included in Kellogg & Goldman’s(1944) revision of the Spider Monkeys genus Ate-les. The Latin name that Allen (1914) attributed tothis taxon relates to the “excessively long tail andlimbs, the tail length very nearly twice the length ofhead and body”. Aside of its elongated and slenderlimbs, taxon A. longimembris is further character-ized by the pitch-black face and ears, except for apale cream-white albinotic triangular patch on thenose, and a wide triangular patch on the foreheadthat is barely covered with sparse backward direc-ted, stiff, black hairs. Another character of thistaxon is the relatively robust incisors and caninesthat look oversized so that the lips seem unable toconceal them. This feature gives adult Long-limbedBlack Spider Monkeys taxon A. longimembris abulldog-like appearance. Moreover, its loud orlong-distance calls that are so typical for otherspider monkeys do not carry far. They sound likebird whistles blowing in the wind. The distributionof A. longimembris is confined by the Rio Madeirain the west, the lower Amazon River in the north,the Rio Tapajós-Juruena in the east and the RioJí-Paraná in the south. From a founder-colony of A.longimembris that once traversed the Rio Tapajós-Juruena to the east derived the White-whiskeredBlack Spider Monkey A. marginatus. It is all-blackand only euchromic in the small triangular foreheadpatch or blaze formed by backwards directed whitehairs. However, we have seen also adult free-ran-ging A. marginatus that had black forehead patches.This taxon occupies the interfluve delineated by theRios Tapajós and Teles-Pires in the west, the lowerAmazon River in the north, the Rios Tocantins andAraguaia in the east, and the upper Rio Teles-Pires

or Rio Minisuiá-Miçú (both right-bank tributariesof the upper Rio Tapajós) in the south. After asomewhat eumelanin bleached founder-colony ofA. marginatus once traversed the upper Rio Teles-Pires south of the A. marginatus distribution, a newtaxon diverged that we name the Upper Rio XingúWhite-whiskered Brown Spider Monkey. Its coat ischestnut-brown dorsally, and lighter brown on theventral parts. The snow-white semi-crescent blazeis much larger than in A. marginatus. It widensabove the eyes into long sidewards directed streaks.This newly identified taxon distinguishes itself alsofrom taxon A. marginatus in the long whitewhiskers that run from below the eyes across thelips and chin. Moreover, facial skin is pink to flesh-colored in the circumocular rings, muzzle, lips andchin. Within the monophyletic A. chamek Clade theWhite-whiskered Brown Spider Monkey from theUpper Rio Xingú represents the furthermost eu-melanin bleached taxon that, in accordance with ourtheory, metachromically and phylogeographicallyradiated farthest away from archetypic Black-facedBlack Spider Monkey taxon A. chamek.Within the A. belzebuth Clade, we recognize the

dorsally saturated eumelanin darkbrown BrownSpider Monkey taxon A. brunneus as the nearest toarchetypic taxon. Belly and inner limbs are eu-melanin bleached light- brown colored. The trian-gular forehead patch formed by backward directedhairs is only slightly bleached brownish-blackcolored. Taxon A. brunneus is found in N Colom-bia, between the Cauca and Magdalena Rivers.It is taxonomically treated as a subspecies of A.hybridus. In the far geological past, the A. belzebuthClade could well have derived from the archetypic,saturated eumelanin, all-black taxon (A. geoffroyi)A. fusciceps (formerly A. fusciceps fusciceps) of theA. geoffroyi Clade that occurs west of the AndesMountains in the Pacific coastal forests of Ecuadorand Colombia. An ancestral founder-colony of A.fusciceps once may have circumvented the SierraNevada north of it and diverged into ancestral A.brunneus in the western part of the lower Río Mag-dalena valley. After a progressively eumelaninbleached founder-colony of A. brunneus traversedthe Río Magdalena to the east, the light-brown andsilvery-white colored Variegated Spider Monkeytaxon A. hybridus could have derived. It rangesfrom the northern Colombian Río Magdalena Basininto the southwesternmost corner of Venezuela, in


the foothills of the Sierra Nevada mountain range(near the city of Mérida). Inner parts of limbs, bellyand the small triangular forehead patch are silverywhite in taxon A. hybridus, whereas the rest of thecoat is light-brown colored. An advanced eu-melanin bleached founder-colony of A. hybridusonce may have circumvented the Sierra NevadaMountains to the east and reached the headwa-ters of some of the Río Orinoco’s tributaries inVenezuela’s Amazonas state. It then diverged intothe furthermost pheomelanin bleached White-bellied Spider Monkey taxon A. belzebuth. It rangesfrom north of the Rio Negro and west of the RioBranco into the Venezuelan State of Amazonas westof the Río Orinoco, and also far into the lowlandAmazon of Colombia. Upper parts, head and dorsalcoat of White-bellied Spider Monkeys taxon A.belzebuth are light-brown, but their pelage onventral parts and inner sides of limbs are silverywhite, often pheomelanin bleached yellow toorange-colored. The skin of muzzle and chin is palebrown to pinkish colored. The triangular foreheadpatch or blaze is light brown, and the eyebrows,whiskers, and throat are silvery. From a founder-colony of A. belzebuth that once traversed the upperRío Caquetá derived in southwestern direction thesouthernmost distributed taxon of the A. belzebuthClade, A. variegatus. This taxon occurs in the NPeruvian, SW Colombian and eastern part of theEcuadorian Amazon, east of the Andes Mountainsand north of the Amazon River (where the river iscalled Río Marañón). Its coat is dorsally eumelaninblackish to dark gray, and ventrally euchromic toalbinotic silvery-white, except for the dark grayishhands and feet. The legs are silvery white, as are thewhiskers and the large blaze or triangular patch onthe forehead. Advanced pheomelanin bleachedcolor traits (yellow and orange) as seen in A. belze-buth are lacking in A. variegatus. In accordancewith our theory and the principle of metachromicbleaching, within the belzebuth Clade the most eu-chromic taxon, A. variegatus, has phylogeograph-ically radiated the farthest away from the dark-brownish colored, nearest to archetypic taxon A.brunneus (Fig. 32).Within the A. geoffroyi Clade (Fig. 33), we re-

cognize the saturated eumelanin ColombianBlack Spider Monkey taxon A. (fusciceps) rufiventris(formerly A. fusciceps robustus) from the Pacificcoastal forests of Colombia and South Panama west

of the Andes Mountains as the nearest to archetypictaxon of the A. geoffroyi Clade. Its coat is glossypitch-black, whereas color morphs of this taxonshow a saturated pheomelanin dark red coloredbelly and genital area. Fur on the forehead isslightly brownish tinged. From taxon A. rufiventrisderived in southern direction the Brown-headedBlack Spider Monkey, the nominate taxon A. (fus-ciceps) fusciceps from the Pacific coastal forests ofEcuador and Colombia. It is slightly eumelaninbleached blackish-gray on the belly, brownish blackabove, with a yellow-brown anterior crown, gradingfrom brown to black on the nape. It often has awhite mustache and beard. Taxa A. fusciceps and A.rufiventris stand at the base of the monophyleticCentral-America Spider-Monkey A. geoffroyiClade,which radiated away in northwestern directionacross the Isthmus of Panama into Central Americaas far north as Mexico. From the Colombian BlackSpider Monkey A. rufiventris derived the advancedeuchromic, near-albinotic (except for the saturatedeumelanin feet, hands, lower arms and distal part ofthe tail) taxon A. (geoffroyi) grisescens. However,the validity of this taxon is doubtful, for it has neverbeen seen in the wild. It is thought to occupy adead-end distribution along the Pacific coast fromthe Río Tuyra valley in SE Panama into the Cor-dillera de Baudó in NW Colombia. To the east, itsdistribution is confined by territory occupied by theColombian Black Spider Monkey A. rufiventris.From A. rufiventris diverged in western directionthe advanced pheomelanin bleached Ornate SpiderMonkey taxon A. (geoffroyi) panamensis. It isargued that the form A. panamensis is a juniorsynonym of A. ornatus. Taxon A. panamensis/ornatus has a golden brown, dark red to orangecolored back, with saturated eumelanin black pe-lage on the top of the head, outer sides of legs,hands, feet and distal part of the tail. It is distributedthroughout Panama (from Chiriquí Province as faras E of the Canal Zone) and C+E Costa Rica. FromA. ornatus (or A. panamensis) derived in southerndirection the advanced pheomelanin bleachedAzuero Spider Monkey taxon A. azuerensis. Itsback is grayish-brown, somewhat darker than theunderside. Outer surfaces of the limbs are black,the top of the head and neck are (brownish)-black.Its distribution is delineated by the PanamanianPacific coast in the south and east. From the Or-nate Spider Monkey taxon A. ornatus derived in


northern direction into Nicaragua the advancedpheomelanin bleached, near-euchromic Geoffroyi’sSpider Monkey taxon A. geoffroyi. It is silvery tobrownish-gray on the back, upper arms, and thighs.Its coat (except for the black head, elbows, knees,upper arms, lower legs, hands and feet) is overallorangish and cream-white colored. Its face is black,often with flesh-colored ‘spectacles’ aroundthe eyes. From A. ornatus radiated away, first inwestern direction and from coastal Costa Ricanorthwards into Nicaragua, the advanced pheo-melanin bleached Black-browed Spider Monkeytaxon A. frontatus.With its orange, black and whitecoat A. frontatus is the most colorful taxon of theentire A. geoffroyi Clade. From taxon A. frontatusderived the overall most euchromic bleachedMexican Spider Monkey taxon A. vellerosus. Itsdorsal surfaces range from black to light brown, andcontrast strongly with its lighter abdomen and innerlimbs. Flesh-colored skin is often present aroundthe eyes. It occupies the entire northwestern part ofthe Isthmus containing El Salvador, Honduras(along the N coast into the lowlands of LaMosquitia), Guatemala (including the highlands)and E & SE Mexico. From taxon A. vellerosus tothe north derived the near-albinotic Yucatán SpiderMonkey taxon A. yucatanensis. It is characterizedby the overall advanced eumelanin bleached, lightbrown and white colored coat. Its fur is brownish-black on the head, neck, and shoulders, grading intolighter brown on the lower back and hips andcontrasting with its silvery-white underside, innerlimbs, and sideburns. Ateles (geoffroyi) yucatan-ensis occupies a large distribution containing NEGuatemala, all of Belize, and SE Mexico (YucatánPeninsula). The near-albinotic taxa A. vellerosusand A. yucatanensis that occupy dead-end distribu-tions confined by untraversable geographic barriersin the northernmost range of the A. geoffroyi Clade,phenotypically do resemble taxon A. grisescens(from SW Panama) that occupies the southernmostdistribution of the A. geoffroyi Clade within theIsthmus. These taxa are equally confined to phylo-geographic dead-end distributions, therefore fullyconcurring with our theory that pretends to unveiland retrace allopatric primate speciation and radi-ation along phylogeographic pathways of meta-chromic bleaching.Woolly Spider Monkeys or Muriquis genus

Brachyteles (family Atelidae) from SE Brazil are

disputedly the largest among New World monkeys(adults weighing up to 11–12 kg). It is estimatedthat alouattines (howling monkeys) and atelines(woolly, spider, and woolly spider monkeys) splitabout 16 MYA and that the ancestor of Muriquis(Brachyteles) and Woolly Monkeys (Lagothrix)separated about 10 MYA from the lineage thatwould eventually lead to the Spider Monkeys (Ate-les). Amazonian Lagothrix and Atlantic Forest Bra-chyteles are therefore considered to be sister groups(Mittermeier et al., 2013). Two species of Muriquisare recognized: the Northern Muriqui B. hypox-anthus, and the Southern Muriqui B. arachnoides(Fig. 34). Taxon arachnoides is distributed in SEBrazil, through the coastal Serra do Mar in thestates of Rio de Janeiro, São Paulo, and (the NEof) Paraná. Its northern limits are the Serra daMantiqueira and the Rios Paraíba and Paraíba doSul. Taxon B. hypoxanthus historically rangedthrough the Atlantic Forest in the states of Bahia,Espírito Santo, Minas Gerais, and Rio de Janeiro,excluding only lowland forests in the extreme S ofBahia and N Espírito Santo. The northern limitof its distribution was probably the Rio Jequiriçáor the right bank of the Rio Paraguaçú, whereasthe southern limit most likely was the Serra daMantiqueira, in S Minas Gerais state. There, itmeets the distribution of the Southern Muriquitaxon B. arachnoides. Sexual dimorphism is absentin Muriquis. The Southern Muriqui has a predom-inantly beige, with light, or dark brown or lightgray-brown colored coat. It retains the black pig-mentation of the face, palms, and soles of the feetfrom infancy into adulthood. Adults of both sexesdevelop only minor depigmentation in small pinkor white spots in the pubic region and sometimeson the face. The Northern Muriqui taxon B. hypox-anthus has a uniformly beige colored pelage, withlight or dark brown or light gray-brown colora-tion. At birth, the face is black, but at sexual matur-ity face and genitals lose their pigmentationand become spotty pink or flesh-colored (Fig.34). Northern Muriquis have a vestigial thumb,which character differentiates them from SouthernMuriquis that lack the thumb. The Southern Muriquiseems to be nearer to archetypic woolly spider mon-keys than the Northern Muriqui, for the latter isoverall progressively pheomelanin bleached near-albinotic in the head characters (white eyebrows,sideburns, and beard), and also in the advanced de-


pigmentation of the face, in particular the spottyflesh- colored muzzle (Mittermeier et al., 2013).For Amazonian Howling Monkeys, genus Al-

ouatta, allopatric speciation, radiation, and phylo-geography along eumelanin and pheomelaninpathways of metachromic bleaching are depicted infigures 35, 36). Two monophyletic cladistic Groupsor Clades are recognized: Al. belzebul and Al.seniculus (Mittermeier et al., 2013). Within the Al.belzebul Clade, distributed south of the Amazon,saturated eumelanin all-black howling monkeys ofthe Amazon Black Howler taxon Al. nigerrimarange between the Tapajós and Madeira Rivers. A

founder-pair or colony of somewhat bleached Al.nigerrima howlers once must have traversed thelower Rio Madeira, most likely lifting on floatinglogs or on drifting islands covered with chavascal(low type of várzea) forest. Presently, this howleralso inhabits almost the entire interfluve delineatedby the Rios Amazonas, Purús and Ipixuna, an areathat was formerly occupied by the advanced pheo-melanin bleached yellowish-orange colored PurusRed Howler taxon Al. puruensis (belonging to theAl. seniculus Clade). We have spotted Al. nigerrimahowlers in the várzea near Carreiro (opposite thecity of Manaus) and, also, as far south as the Rios

Figure 34. Metachromic bleaching in the Woolly Spider Monkey or Muriqui genus Brachyteles from the Atlantic coast of SE Brazil.


Igapó-Açú and Tupana - black-water rivers thatempty out into the Rio Madeirinha (a white-waterleft-bank tributary of the Rio Madeira). It seemsthat the overall orange-colored resident howler Al.puruensis and the all-black invasive Al. nigerrimahowler do co-exist locally. However, the two taxado not mix nor interbreed. While conducting acanoe survey during the peak of the flood season inthe vast igapó floodplain along the Rio Igapó-Açúand Igarapé Cujubim, we have heard and seen thetwo taxa belonging to different Clades (Al. belzebuland Al. seniculus, respectively) in the same generalarea. The more frugivorous Al. nigerrima howlerwas only seen in the middle of seasonally floodedigapó forest during the peak of fruiting, whereas themore folivorous puruensis howler stayed back inthe adjacent primary terra firme rain forest. Weassume that Al. puruensis does so by lack of its

elsewhere preferred habitat - seasonally white-water inundated floodplain forest (várzea). It there-fore seems that monkey taxa belonging to differentmonophyletic clades locally can co-exist, but onlyif they occupy different feeding niches, and withinthe local landscape parapatric or partly overlappinghabitats. As these two howler taxa are consideredvalid species, they seem to have sufficiently di-verged from one another to impede interbreeding inthe contact or overlap zone.Representing the nearest to archetypic taxon of

the Al. belzebul Clade, the range of the AmazonBlack Howler taxon Al. nigerrimamay well be con-sidered the center of the Clade’s dispersion. Fromhere, the other taxa diverged in eastern direction.From Al. nigerrima derived east of the Rio Tapajósthe pheomelanin bleached Spix’s Howler Al. dis-color. It has an overall dark-brown to mahogany-

Figure 35. Distributions, allopatric speciation, radiation, and phylogeography along different pathways of metachromic bleaching depicted for all known Amazonian Howlers genus Alouatta.


Figure 36. Metachromic variation, radiation, and phylogeography along eumelanin and pheomelanin pathways of metachromic bleaching depicted for all known Amazonian Howlers of the Alouatta seniculus and Al. belzebul Clades.


red coat and a rufous-chestnut dorsal band, hands,feet, and tip of the tail. A further pheomelaninbleached founder-colony of Al. discolor must oncehave traversed its eastern distributional limit -the Rio Xingú, Irirí, or Santa Helena (left-banktributary of the Rio Teles-Pires in Mato Grosso).From Spix’s Howler Al. discolor derived the darkbrown colored Red-handed Howler Al. belzebul.This species is progressively pheomelanin bleachedin the reddish-brown to yellow hands, feet, tail tip,forehead and back. It is distributed south of theAmazon, east of the Rio Xingú-Irirí, in the states ofPará (including Mexiana, Caviana, and MarajóIslands in the Amazon estuary), Maranhão,Tocantins, and Mato Grosso. West of the Atlanticcoast of NE Brazil in the states of Rio Grande doNorte, Paraíba, Pernambuco, and Alagoas, arefound enclave populations isolated from what isthought the taxon’s former distribution, which musthave been continuous through the states of Cearáand Piauí to the Amazonian population. From theRed-handed Howler Al. belzebul derived the ad-vanced pheomelanin bleached Maranhão Red-handed Howler Al. ululata. This species distributedin NE Brazil occurs in remnant forest patches of dry

forest scrub called caatinga. It enters also thecoastal mangrove forests of northern Maranhão,Piauí, and Ceará. The Maranhão Red-handedHowler Al. ululata radiated farthest away from thearchetypic overall black Central Amazon BlackHowler Al. nigerrima. It is sexually dichromatic.The male is black with rufous to reddish-brownhands, feet, tip of the tail and flanks. The female isyellowish-brown with sparse grayish hairs, givingit an overall olivaceous appearance (Fig. 36).Within the Al. seniculus Clade we consider the

overall dark reddish-brown Ursine Red Howlerarctoidea from N Venezuela east of Lake Mara-caibo, and from the coast (including the Islands ofTrinidad and Tobago) extending S through thellanos to the Río Orinoco, the nearest to archetypictaxon for the monophyletic Al. seniculus sub-Clade,which is distributed north of the Amazon. Bothsexes have a coat that is dark reddish-brown on thebody, contrasting with a darker brown to blackishhead, shoulders, limbs, and proximal part of the tail.Male Ursine Red Howlers often have a blackishbeard, limbs and tail. From a founder-colony thatonce traversed the Orinoco River to the east, has de-rived the advanced pheomelanin bleached GuiananRed Howler Al. macconnelli. This taxon rangeseast of the Río Orinoco throughout the Guianas, NBrazil (east of the Rio Negro and north of the RioAmazonas, including Gurupá Island in the Amazonestuary), and S Venezuela (between the Cassiquiareand Orinoco Rivers). The Guianan Red Howler’scoat is uniformly dark rufous-brown, the back ispheomelanin bleached yellowish to golden-brownwith a dark dorsal stripe, and arms to elbows andlegs to thighs are orangish-red. Distal part of the tailis pale-yellow (Figs. 35, 36).From a founder-colony of further pheomelanin

bleached Al. macconnelli that once traversed eitherthe upper Rio Negro in the Colombian Amazon, orthe Orinoco River at its headwaters, has derived theoverall orange-colored Colombian Red Howler Al.seniculus. This taxon is now distributed north of theAmazon across E Ecuador and E Peru (east of theRío Huallaga), Colombia, NW Venezuela, and theBrazilian Amazon inbetween the Rio Solimões andRio Negro. The Colombian Red Howler Al. sen-iculus is overall golden-toned to coppery-red onthe body, contrasting with the maroon head,shoulders, limbs, and proximal part of the tail. MaleColombian Red Howlers are much bigger than fe-

Figure 37. Adult male Guianan Red Howler Alouatta mac-connelli. It was photographed while pulling off a juvenilethat tried to seek protection from the human intruder, some-where along the black-water Rio Jauaperí, a north-banktributary of the Rio Negro (Courtesy of David Lemmon).


males. Within the monophyletic Al. seniculus sub-Clade the bright orange-red Colombian Red HowlerAl. seniculus is phylogeograpically the most ad-vanced pheomelanin bleached taxon. It radiated thefarthest away from the nearest to archetypic satur-ated eumelanin, overall dark brown colored UrsineRed Howler Al. arctoidea, as such fully concurringwith our theory. With respect to the monophyleticAl. sara sub-Clade of the Al. seniculus cladisticGroup that is largely distributed south of theAmazon, we consider the Juruá Red Howler juarathe nearest to archetypic, least eumelanin bleached(dark brown) taxon. It ranges in the W BrazilianAmazon south of the Rio Solimões, and in theRio Juruá Basin, extending west into the PeruvianAmazon. It is not sexually dichromatic. Its coat isgenerally dark reddish-brown, with the middle ofthe back lighter orange-rufous colored, and limbsand tail base dark rufous to black. The tail is paler,more golden from middle to tip. From a pheo-melanin bleached founder- colony that once tra-versed the Rio Juruá to the east derived the PurúsRed Howler taxon Al. puruensis. It is distributedacross the entire Rios Juruá/Purús interfluve as fareast as the middle Rio Madeira. From there, it ex-tended its range across the upper Rio Aripuanã asfar east as the Rio Teles-Pires, and south as faras the Rio Abunã (which forms Bolivia’s northernborder). The Purús Red Howler Al. puruensis issexually dichromatic. Males are dark rufous or red-brown with a golden upper dorsum and shoulders,whereas females are golden-orange with distal por-tions of limbs, tail base, and beard dark rufous.From a progressively pheomelanin bleachedfounder-colony of the Purús Red Howler thatonce traversed the Río Abunã, has derived the quitedistinct Bolivian Red Howler taxon sara. It is dis-tributed across the Bolivian Amazon including theentire Río Beni Basin, and east as far as the RiosMamoré/Guaporé interfluve. The Bolivian RedHowler’s coat is brick-red above, with limbs, head,and proximal part of the tail darker, more rufouscolored. It represents the most advanced pheo-melanin bleached taxon of the Al. sara sub-Clade(Fig. 36). It occupies a dead-end distribution in thesouth bordering the drier savanna and Chaco area(Fig. 35). Going further southwards begins the dis-tribution of the Paraguayan Howler Al. caraya.For extra-Amazonian Howling Monkeys genus

Alouatta, allopatric speciation, radiation, and phylo-

geography along eumelanin and pheomelaninpathways of metachromic bleaching are depicted infigures 38, 39. Four non-Amazonian monophyleticClades are recognized: the Brazilian Brown HowlerAl. guariba, the Paraguayan Howler Al. caraya, theCentral American Mantled Howler Al. palliata, andthe Mexican Black Howler Al. pigra (Mittermeieret al., 2013). The Brown Howler Al. guariba Cladeconsists of two populations that may represent dif-ferent valid taxa or species: the Northern BrownHowler Al. guariba and the Southern BrownHowler clamitans. Taxon Al. guariba ranges in theAtlantic Forest from the Rio Paraguaçú, BahiaState, along the coast south as far as Rio Paraiba inRio de Janeiro State. Inland, it extends into MinasGerais State. The Southern Brown Howler Al. clam-itans is distributed in the Atlantic Forest south ofRios Doce and Jequitinhonha, south as far as RioGrande do Sul State. Taxon Al. guariba is not sexu-ally dichromatic and both sexes are red-fawn, withfemales usually somewhat duller in color. Taxon Al.clamitans is generally dark reddish-brown, withmales often being lighter colored than females.Males from São Paulo are orange-red to red-brownwith a red belly, whereas males from Santa Catarinaand Rio Grande do Sul are bright red-orange,having dark brown feet. Females are overall darkbrown or blackish. The Northern Brown Howler Al.guariba is the lesser metachromic bleached. TaxonAl. clamitans derived from it, the males progress-ively following the pathway of pheomelaninbleaching. The further south it ranges, the morethe male’s overall coat color tends to red-orange orbright orange (Fig. 39).The Paraguayan Howler Al. caraya forms a

monotypic Clade. It is a sister species to the Amazo-nian red howlers of the Al. seniculus Clade. It di-verged from a common ancestor about 4 MYA. TheParaguayan Howler Al. caraya is distributed acrossC Brazil, south of the states of Pará, Tocantins,Maranhão, and Piauí, west into the Pantanal, southinto Paraguay, E and SE Bolivia, and maybe alsointo NW Uruguay. Much of its range is in the‘cerrado’ of central Brazil and semi-arid ‘caatinga’forest scrub in NE Brazil, where it uses gallery andriparian forest and patches of seasonal (semi)-deciduous ‘cerradão’ (a type of savanna forest).Adults of Al. caraya are sexually dichromatic,but both sexes are blond at birth. Mature males aregenerally uniformly black. Females and young of


either sex are pale grayish-yellow to golden-brown.Male Paraguayan Howlers Al. caraya from Bahiaand Goiás are black, but those from Mato Grossoand Paraná are black with a brown back and hindparts. Males from São Paulo and Minas GeraisStates are brown-black, with yellowish hands, feet,belly, and tail tip. In all male individuals the faceis invariably dark, the fur is stiff and lengthy, andthe beard is prominent. The scrotum is rust-redcolored.The Pacific Coastal and Central American

Mantled Howler Al. palliata is, based on geo-graphic distribution, divided in five taxa that couldwell represent distinct valid species: Al. palliatafrom NE Guatemala, ranging east to E Costa Ricaor W Panama; Al. aequatorialis from the southerndistributional limits of Al. palliata ranging through

the Darién into W Colombia, W Ecuador, and southas far as NW Peru; Al. mexicana ranging from S toSE Mexico and Guatemala following the southern-most distribution of the Central American BlackHowler Al. pigra; Al. coibensis from Coiba andJicarón Islands in SW Panama; and Al. trabeatafrom the Azuero Peninsula in SW Panama (Fig. 38).The coat of the Mantled Howler is smooth, veryshort and upright, being silky black with a mantleof longer, gold or yellowish-brown fur along theflanks. Adult males have a white scrotum.The Central American Black Howler Al. pigra

is monotypic. It is distributed across SE Mexico,Belize, and N to C Guatemala. Fur of Al. pigra isnotably long, soft, and dense. Adults are not sexu-ally dichromatic. They are overall black with tracesof brown on the shoulders, cheeks, and back. The

Figure 38. Distributions, allopatric speciation, radiation, and pathways of metachromic bleaching followed in all extra-Amazonian Howling Monkeys genus Alouatta. Two Clades occur south of the Amazon: Al. guariba from the E BrazilianAtlantic forest, and Al. caraya from the ‘cerrado’ and ‘cerradão’ of the Central Brazilian Plateau. Along the Pacific coast ofEcuador and Colombia, far into Central America, occur the Al. palliata and Al. pigra Clades.


Figure 39. Pelage color variation, radiation and metachromic bleaching along eumelanin and pheomelanin pathways depictedfor all extra-Amazonian Howling Monkeys genus Alouatta of the Brazilian Al. guariba and Al. caraya Clades, and theCentral-American Al. pigra and Al. palliata Clades.


Central American Black Howler Al. pigra is consi-dered the most saturated eumelanin, least bleached,nearest to archetypic form of the Al. pigra and Al.palliata Clades. It is also by far the largest howlingmonkey. The Al. palliata Clade is believed to havediverged from ancestral Al. pigra about 3 MYA(Mittermeier et al., 2013). Taxon Al. palliata issympatric with taxon Al. pigra in Tabasco State,Mexico and in a small part of Guatemala. From theGolden-mantled Howler Al. palliata radiatingnorthwards derived the Mexican Howler Al. mexi-cana, and radiating southwards the South PacificBlackish Howler Al. aequatorialis, ranging farsouth into the Pacific coastal forests of Colombiaand Ecuador. There, it is sympatric with the Colom-bian Red Howler Al. seniculus. From Al. aequato-rialis in SW Panama derived the Azuero PeninsulaHowler Al. trabeata and the Coiba Island HowlerAl. coibensis (Fig. 38).During his long-term fieldwork on the ecology

of all eight monkey species that occur in theGuianas, the senior author has repeatedly watchedthe basic principles of allopatric primate speciationat work. At his study site situated in pristine primaryterra firme rain forest in Central Suriname, localpopulations of the Guianan Red Howler Al. mac-connelli (Fig. 37), the most territorial among allextant howling monkeys when measured by the sizeof the hyoid bone, had passed beyond the howler’soptimal densities (Van Roosmalen, 2013a; 2015).This was measured by the high frequency of dawnchorus and vocal battles of neighboring groupsthroughout the day and nighttime in the proximityof territorial boundaries. One day, a subadult malegot pushed out of his parental group that rangedclose to the campsite area. For several days afterbeing expelled, this young howler male got repe-atedly involved in vocal battles with neighboringgroups that subsequently chased him out of theirrespective territories. Weeks later, far away from thecampsite, a boundary conflict took place thatseemed never ending. The researcher rushed overto the spot. He arrived just in time to witness thisvery subadult male being attacked by the leader ofa resident group in the company of his harem. Thewhole group chased the young male into an isolatedtree close to where he could watch the scene. Thesubadult male was in the company of a female hepresumably had attracted (‘stolen’) from some re-sident group that had chased him out earlier. In an

attempt to escape from his attackers, the howlermale almost fell out of the canopy. He just couldget hold on a thick branch and was hanging under-neath it only secured by the grip of his hands andtail. Then, they all began to bite in his hands andtail tip. With a scream, he let loose and camecrashing over forty meters down to the forest floor,hitting it at a hair width away from the researcher’shead. The monkey looked dead, his motionlessbody covered with blood. After a few minutes,however, he got back on his feet and slowly clim-bed up a small tree. Back in the canopy, he sat nextto his mate that had been watching the show from adistance. The pair was never seen again within theborders of the 400-ha study area. Some time later,vocal battling recommenced. It came from the samedirection, sounding only much farther away. Inretrospect, we assume that the couple survived andin the long run found a place to settle down, start afamily, and defend a small territory squeezed inbe-tween the territories of some resident howler groupsfar away from their respective parental groups. Onemay speculate that the howler pair, driven by thetrend to allopatry, also may have survived by ven-turing into some ‘empty’, marginal, or for howlersunfamiliar habitat. Or, in case the male was expel-led from his parental group for his skin or (part of)coat color being somewhat lighter, he could havejoined other outcast males that were discriminatedupon and pushed out of their parental groups forother mutant metachromic deviances of skin and/orcoat characters. For the sake of survival alone, suchhealthy young individuals may join efforts to stayalive. Together, they may turn into potential foun-der-colonies venturing into new lands, where theycan thrive and reproduce unrestrictedly. At least aslong as those lands, in turn, do not reach the taxon’soptimal population density. By the time they do so,the generally accepted phenotype of that new para-patric or allopatric taxon or (eco)-species will havebeen stabilized while showing whatever features offurther metachromic bleaching and/or depilation.Living on an island in the Coppename River at

about ten km from his field site, the senior authorhas repeatedly witnessed the coming and going ofsmall groups of potential howler-founders to andfro Foengoe Island after having been forcefully pu-shed out from some mainland territory by the rulinggroup male(s). Pushed against the riverbank, theyapparently did overcome their natural fear of water


and then swam toward the island. For some time,such immigrants tried to make a living on theisland. Until it became clear to them they weretrapped on an island too small to sustain a howlergroup year-round. Occasionally, such groups werespotted later while ranging along the opposite river-bank. We assume they had traversed the river swim-ming. Interestingly, a female howler that was raisedas a pet and then set free to range across the 30-haisland, was eager to join any howlers coming ontothe island. Sadly, when the immigrants eventuallyswam back to the mainland in search of new lands,the female stayed back on the island. Perhaps, shedid so for lack of sufficient bonding or for fear ofswimming across the river.Capuchin Monkeys (genera Cebus and Sapajus)

have diverged from Squirrel Monkeys (Saimiri)about 15 MYA. They formed distinct monophyletic

Clades that diverged during the Late Miocene toEarly Pliocene, about 6.2 MYA. The Clades diver-sified during the Plio-Pleistocene era into twogroups: Gracile or Untufted Capuchins (genusCebus) in what is today the western Amazon, about2.1 MYA, and Robust or Tufted Capuchins (genusSapajus) in what are today SE Brazil, E Paraguay,and N Argentina, beginning about 2.7 MYA (Mit-termeier et al., 2013). Gracile Capuchins genusCebus are separated into the following five cladi-stics Groups or Clades: Humboldt’s White-frontedCapuchin Ce. albifrons with four+ taxa (Ce. albi-frons, Ce. yuracus, Ce. unicolor, and Ce. cuscinus),Guianan Weeper Capuchin Ce. olivaceuswith threetaxa (Ce. brunneus, Ce. olivaceus, and Ce. casta-neus), White-faced Capuchin capucinus with twotaxa (Colombian White-faced Capuchin Ce. capu-cinus and Panamanian White-faced Capuchin Ce.


Figure 40. Distributions, allopatric speciation, radiation, and pathways of metachromic bleaching in all hitherto recognizedGracile Capuchins of the five distinguished phylogeographic Clades: Cebus capucinus, C. olivaceus, C. versicolor, and C.albifrons. The fifth Clade C. aequatorialis is monotypic.


Figure 41. Metachromic diversification along eumelanin and pheomelanin pathways of metachromic bleaching (arrowedlines), speciation and radiation in all hitherto recognized Gracile Capuchins (Cebus) of the five distinguished phylogeo-graphic Clades: Cebus olivaceus, C. versicolor, Cebus capucinus, and C. albifrons. Cebus aequatorialis from W Ecuadorand NW Peru is monotypic, but may have derived from ancestral C. yuracus that once traversed the Andes Mts.

imitator), and Varied White-fronted Capuchin Ce.versicolor with four taxa (Ce. versicolor, Ce. leu-cocephalus, Ce. cesarae, and Ce. malitiosus). Therecently discovered Ka’apor Capuchin Ce. kaaporiranging S of the lower Rio Amazonas is geo-graphically closest related to the Guianan WeeperCapuchin (i.e., taxon Ce. castaneus ranging alongthe left/north bank of the lower Rio Amazonas)and, therefore, may form a sister Clade to it (Figs.40, 41).Within the Ce. capucinus Clade we consider

the Colombian White-faced taxon Ce. capucinus(ranging from E Panama, through W Colombiasouth as far as NW Ecuador), the nearest to arche-typic, least metachromic bleached form. Its body,crown, limbs, and tail are black. The chest is white,extending forward to the face and front of the crownand upward to the shoulders and upper arms. TheGorgona White-faced Capuchin Ce. curtus curtusis a small and relatively short-tailed subspeciesfrom Gorgona Island sitting on the ColombianPacific coast. From taxon Ce. capucinus derivedCe. imitator, the taxon that ranges from N Hon-duras, C and W Nicaragua, Costa Rica southinto W Panama. It resembles much the typicalColombian White-faced Capuchin Ce. capucinus,but females have elongated frontal tufts with abrownish tinge.Within the Ce. olivaceus Clade we consider the

Venezuelan Brown Capuchin Ce. brunneus from NVenezuela east of the Sierra de Perijá and along theCoastal Range, including the island of Trinidad(where it is possibly introduced), the nearest to ar-chetypic, least metachromic bleached form. Its pel-age is thick and long, the upperparts are generallydarker along the middle of the back than on thesides, the hairs are dusky basally, with a broad zoneof chestnut in the middle, and black at the tips. Faceand sides of the head are pale yellowish gray. Thecrown has a broad V-shaped patch of long hairs,narrowing to a point in front of which a narrowblack line runs forward to the nose. Chin and lowerparts of cheeks are grayish or fulvous white towhitish. Underparts are blackish brown, with tipsof the hairs hazel. The throat is lighter than the chestand belly. Upper arms are maize yellow. Outerforearms are blackish with yellowish tips, insideforearms are much darker. Hands are blackish,hindfeet are nearly black. Tail is colored as back.From ancestral Ce. brunneus derived the Guianan

Weeper Capuchin Ce. olivaceus that is restricted tothe Venezuelan Amazon Basin in forests of the Gu-ayanan Shield, from the upper Río Orinoco east tothe left bank of the Rio Essequibo in W Guyana. Itspelage is overall dark brown or reddish with black-agouti banding on flanks, limbs, and tail. The faceis naked and pink. Cheeks are buffy-white. It differsfrom Ce. brunneus in the advanced bleached albi-notic head and upper arms and the wider V-shapedblack crown cap.

From ancestral Ce. olivaceus derived the Chest-nut Weeper Capuchin Ce. castaneus. This taxonranges from the Rio Essequibo E through Surinameand French Guiana into N Brazil, where its distri-bution is delineated by the Rios Negro, Branco, andCatrimani in the W, Rio Amazonas in the S, and theAtlantic coast in the E (it also inhabits Caviana andMexiana Islands in Amazon’s estuary). It differsfrom Ce. olivaceus in the narrower black triangleon the crown and the pelage of the head being ove-rall yellowish-white, but reddish-chestnut above theear and nape, in the advanced pheomelanin ble-ached reddish-chestnut upperparts of the body andlimbs, and pale yellow shoulders and fronts of armsabove the elbows. A founder-colony of the ChestnutWeeper Capuchin Ce. (olivaceus) castaneus mustonce have traversed the lower Rio Amazonas. Fromit derived the Ka’apor Capuchin Ce. kaapori thatranges in NE Brazil south of the lower AmazonRiver (NE Pará and NW Maranhão). This taxon ischaracterized by a longer body in comparison toother Cebus species. It is grayish agouti-brown, andlighter on the flanks. Face, shoulders, mantle, andtail tip are silvery-gray, the limbs are agouti, and thehands and feet dark brown or black. The crown hasa triangular black cap that extends to a dark stripedown the nose. The pelage of the Ka’apor Capuchinis overall advanced eumelanin bleached to nearlyalbinotic, as such much contrasting with the satu-rated eumelanin blackish crown cap, hands and feet.Being phylogeographically farthest radiated awayfrom the center of dispersion (NW Venezuela) ofthe Ce. olivaceus Clade, and occupying a dead-enddistribution, where it also has to compete with theGuianan Brown Capuchin Sapajus apella (Figs.42–44), Ce. kaapori is clearly the most progressi-vely bleached, near-albinotic taxon within the Ce.olivaceus Clade (Fig. 41).Within the Ce. versicolor Clade we consider the

Varied White-fronted Capuchin Ce. versicolor the


nearest to archetypic, least metachromic bleachedtaxon. It is distributed in N Colombia in the middleRío Magdalena Basin. It is the darkest among theClade’s four taxa, though a rather pale form withred tones on the mid-dorsal region and foreparts ofthe limbs, generally contrasting with the rest ofthe body (Fig. 41). From Ce. versicolor derivedtowards the NE the Sierra de Perijá White-frontedCapuchin Ce. leucocephalus that ranges in N Co-

lombia from the W slope of the Cordillera OrientalE to the Ríos Zulia and Catatumbo Basins and NWVenezuela (Zulia State). This taxon is progressivelybleached, near-albinotic in the head, chest, andshoulder parts. From Ce. versicolor derived towardsthe N first the Río Cesar White-fronted CapuchinCe. cesarae, ranging in N Colombia, in the RíoCesar Valley, W into the S and E slopes of the SierraNevada de Santa Marta. From taxon Ce. cesarae,


Figure 42. Phylogeography, allopatric speciation, radiation, and pathways of metachromic bleaching followed in all hithertorecognized Robust (Tufted) Capuchins of the three distinguished phylogenetic Clades: Sapajus nigritus, S. apella, and S.macrocephalus. From an ancestral saturated eumelanin (all-black) form of the S. nigritus Clade, quite recently (an estimated400,000 YA) radiated away into the Amazon the pheomelanin bleached species S. apella (including the insular taxon S. a.margaritae), and S. macrocephalus, the latter with four different taxa/’species-in-the-making’ - from SE to N: juruanus,pallidus, maranonis, and fatuellus).


Figure 43. Radiation and metachromic diversification in the Sapajus nigritus and S. apella Clades of Robust or TuftedCapuchins. From an ancestral saturated eumelanin form of S. nigritus the species Sapajus apella and S. macrocephalusradiated away into the Amazon with different taxa ‘in-the-making’.

in turn, derived the Santa Marta White-frontedCapuchin Ce. malitiosus that is only known fromthe NW base of Sierra de Santa Marta. It may rangealso throughout the lower W and N slopes of theSierra Nevada in N Colombia. The two taxa are thepalest among the N Colombian and VenezuelanWhite-fronted Capuchins. Taxon Ce. cesarae isbuffy in the head and throat parts and pheomelaninbleached orangish in the cap, middle of the back,forearms, and forelegs, as such contrasting with thesides of back and trunk. Taxon Ce. malitiosus isadvanced eumelanin bleached in the silvery to cin-namon-brown chest and belly, and a contrasting al-binotic area of the front extending well over theupper surfaces of the shoulders and inner sides ofupper arms (Fig. 41).Within the Ce. albifrons Clade we consider the

Marañón White-fronted Capuchin Ce. yuracus thenearest to archetypic, overall least metachromicbleached taxon (Fig. 41). This taxon is distributednorth of the Amazon River in S Colombia, EEcuador, NE Peru, and presumably W Brazilbetween the Rios Içá and Solimões. It is gray-fronted on the forehead, sides of the face, chest,and outer sides of the arms. Its general color isochreous-brown, contrasting sharply with the gray-ish to buffy outer sides of forelimbs, and with thepale silvery to orangish underparts. The cap isblack, with a median line running down inbetweenthe eyes. The tail is brown like the back, but palertowards the tip. From ancestral Ce. yuracus derivedfirst Humboldt’s White-fronted Capuchin, the nom-inate taxon Ce. albifrons that is widely distributedacross the upper Amazon Basin of S Venezuela, Sand E Colombia (occurring north of the Rio Amazo-nas and the Río Içá-Putumayo, N as far as the RíoMeta, and in the lowlands W of the Orinoco, andNW Brazil (N of the Rio Solimões, and W of theRios Negro and Branco, as far north as the Rio Ur-aricoeira). Humboldt’s White-fronted Capuchin Ce.albifrons is overall pale grayish-brown, darker onthe limbs. Hands and feet are yellowish-brown. Thetail is ashy above, whitish below, and brownish-black towards the tip. The front is creamy white,and there is a cap of short dark fur on the crown thatis rounded in the front and well demarcated fromthe light-colored forehead. The face is naked andpinkish, flesh-colored. From Ce. yuracus derivedalso Spix’s White- fronted Capuchin Ce. unicolor,most likely after a founder-colony of ancestral Ce.

yuracus traversed the upper reaches of the RíoUcayali in E Peru. It is nowadays widely distributedin the upper Brazilian Amazon Basin, south of theAmazon River and west of the Rio Tapajós, through-out the northern parts of Mato Grosso and Rondo-nia States, and throughout the Rios Madeira, Purús,Juruá, and Javarí Basins as far west as theRío Ucayali. Cebus unicolor is uniformly brightochreous or grayish-brown with darker grayish-brown flanks and mid-back, with a yellowish orcream-white front and reddish-yellow to reddishlimbs and tail. From ancestral Ce. unicolor derivedlater the Shock-headed Capuchin Ce. cuscinus thatis believed to range from the right bank of the upperreaches of the Rio Purús in SE Peru, W into theCuzco Department including the upper Río Madrede Dios, and S and E as far as the Río TambopataBasin, also extending into NW Bolivia. Taxon Ce.cuscinus has a longer, silkier fur than Ce. unicolorand is less brightly colored. Its limbs are brownerand contrast less with the back. The cap is large,distinct, and dark brown. The forearms are orange-rufous on the outside, darker on the wrists andhands. Underparts are ochreous- orange and silvery,becoming buff on the chest. The fronts of theshoulders and inner sides of the upper arms arewhitish. The tail is brown, somewhat paler towardsthe tip. The male has a broad pale frontal regionsharply defining the dark-brown cap. Overall,Ce. cuscinus is the most advanced pheomelaninbleached taxon of the Ce. albifrons Clade. It is theform that radiated away farthest from the center ofthis Clade’s dispersion (Fig. 40). The EcuadorianWhite-fronted Capuchin Ce. aequatorialis is mono-typic. It may form a sister Clade to the GracileCapuchins from the upper Amazon Basin. AncestralMarañón White-fronted Capuchin Ce. yuracus oncemust have traversed the Andes Mountains some-where at the upper reaches of the Río Marañón andthen diverged into Ce. aequatorialis. Cebus aequat-orialis is distributed in Ecuador and NW Peru, inthe lowlands west of the Andes (Fig. 34). Its upper-parts are pale cinnamon rufous, darker along themidline of the back. Front and sides of the head areyellowish white, with a narrow black transverse lineon the forehead forming the cap, from which anarrow median black line descends to the nose.Hands and feet are a little darker, more brownishthan the arms and legs. The chest is lighter than thebelly (Fig. 41).


During long-term fieldwork in Central Suri-name, the first author spotted a few times by chancesmall parties consisting of phenotypically deviantcream-white, long-haired, fluffy-coated malesof the Guianan Weeper Capuchin Ce. (olivaceus)castaneus. Such all-male parties seemed to rangerandomly while travelling at high speed through thevast landscape of pristine matrix lowland rain forestin the middle of which his study area was situated.It is located at more than one-hundred km north ofKaiser Mountains, a hilly country of which the foot-hills seem to form the Guianan Weeper Capuchin’score distribution. This region that provides thismonkey with its preferred habitat - ‘mountainsavanna forest’- was found to sustain a very largepopulation of this elsewhere in the Guianas ex-tremely rare taxon Ce. (olivaceus) castaneus.Mountain savanna forest is typified by an under-story that is dominated by the majestic ‘bergi-maripa’ palm Attalea speciosa Mart. (ArecalesArecaceae). Above 400 m altitude, this palm tree islocally so abundant that one gets the impression towalk through a monocultural plantation of theAfrican oil-palm Elaeis guineensis Jacq. (ArecalesArecaceae). The large fruits of Attalea speciosaconstitute the Guianan Weeper Capuchin’s principaldaily food throughout most of the year. Mountainsavanna forest above 400 m altitude, therefore, mayfunction as a ‘keystone habitat’ to the GuiananWeeper Capuchin, hence the high population density.Two decades later, while conducting biod-

iversity surveys in Pico da Neblina National Parksituated in the extreme northwestern corner of theBrazilian Amazon, the authors spotted a populationof near-albinotic Weeper Capuchins that werecharacterized by a very dense, fluffy, overall long-haired, cream-white bleached fur. Their coat fea-tures looked very similar to that of the all-maleparties that were seen sporadically passing throughthe Voltzberg study area. The Pico da Neblinapopulation of weeper capuchins was spotted in alow type of cloud forest scrub that grows at highaltitudes of 2,000 to 2,500 m. To the astonishmentof the researchers, the capuchins were seen spend-ing part of the daytime on the ground in the middleof open tepuí (sandstone table-mountain) ‘rocksavanna’. They were seen foraging for inverteb-rates, mostly snails, other organisms endemic totepuí mountain tops, in addition to vegetable matter(e.g., roots, tubers and pseudobulbs of all sorts of

terrestrial bromeliads and orchids). In retrospect,our sighting may be explained for as follows. In thepast, a founder-colony of near-albinotic GuiananWeeper Capuchins, driven by the ‘trend to allo-patry’ out of the center of dispersion of archetypicCe. (olivaceus) castaneus, may have traversed theupper Rio Branco and then reached the Pico daNeblina area. The latter is situated somewhat southof the Río Cassiquiare, the channel that runsthrough the watershed connecting the basin of theRio Negro with that of the Río Orinoco. The fullybleached euchromic, long-haired, soft-coatedweeper capuchins that were seen foraging in tepuícloud forest and open rock-savanna at 500–1,000m below the 3,004 m Pico da Neblina summit muchresembled the near-albinotic, fluffy-coated Ce.(olivaceus) castaneus from Kaiser Mountains,Central Suriname. If the Pico da Neblina populationturns out to represent a new taxon or one in-the-making, the ‘Neblina Weeper Capuchin’ wouldoccupy a dead-end distribution in the southwestern-most corner of the Ce. olivaceus Clade’s range,the farthest away from the supposed center ofthe Clade’s dispersion (the Guianas or VenezuelanCoastal Range). The upper Rio Negro forms the di-vision between the distributions of the GuiananWeeper Capuchin Ce. olivaceus Clade and theHumboldt’s White-fronted Capuchin Ce. albifronsClade (Fig. 39). This example from the field is inline with our theory of allopatric speciation in male-defended territorial primates such as Cebus. The‘Neblina Weeper Capuchin’ may have radiatedaway from the Ce. olivaceus Clade’s center ofdispersion in the Guianas following a pathway ofmetachromic bleaching driven by the trend to allo-patry in phenotypically deviant euchromic, long andfluffy-haired males. Interestingly, the mechanismof allopatric speciation and radiation of a mono-phyletic clade of monkeys like that of Humboldt’sWeeper Capuchins at first sight seems non-ad-aptive, at least in strict Darwinian sense, for it issolely based on discriminatory behavior performedexclusively by high-ranking males. The genes forwarm, long and fluffy-haired coats are simplyretained in the genes of these capuchin ‘founder-colonies’. Such a feature would therefore not apriori be the result of adaptive processes of naturalselection. Its warm coat only secondarily happenedto have survival value. It only turned adaptive whenthese gracile capuchins had to adapt in a short


period of time to a new habitat or feeding niche thatwould not have suited the species they derivedfrom. Following this rational, one may speculateabout a similar metachromic pathway that our hom-inid ancestors about 6 MYA must have followedwhen exchanging the canopy of tropical rain forestfor a landscape of arid, open savanna scrub. Or asimilar pathway of metachromic bleaching towardsalbinotic (from a black to yellow or white skincolor) and/or depilation of the body that differenthominids followed between 100,000 and 50,000years ago, when the trend to allopatry (male dis-criminatory behavior) forced them to leave thecenter of hominid dispersion and the cradle ofhominid evolution - C and N Africa - to make aharsh living of nomadic big-game hunting/gath-ering in (for hominids) clime - and habitat - wisenew, marginal, unsuitable, or inhospitable land-scapes of Central Europe, the Middle-East andSE Asia.Capuchin Monkeys of the genera Cebus and

Sapajus formed distinct monophyletic Clades thatdiverged during the Late Miocene to Early Plio-cene, about 6.2 MYA. During the Plio-Pleistoceneera the Clades diversified into two groups: Gracileor Untufted Capuchins genus Cebus, about 2.1MYA in what is nowadays the western Amazon, andRobust or Tufted Capuchins genus Sapajus, begin-ning about 2.7 MYA in what are today SE Brazil, EParaguay, and N Argentina. There is strong evid-ence from molecular genetic studies that RobustCapuchins (genus Sapajus) spent most of their evol-utionary history in the Atlantic Forest of SE Brazil,NE Argentina, and E Paraguay. And that the currentwide-ranging sympatry of Robust and GracileCapuchins across the larger part of the AmazonBasin is the result of a single, rapid, Late-Pleisto-cene invasion of Robust Capuchins from the At-lantic Forest, first into the ‘Cerrado’ and ‘Cerradão’of C and NE Brazil, and only recently (about 0.4MYA) from central South America north into theAmazon Basin and the Guianas (Mittermeier et al.,2013). Though widespread throughout the AmazonBasin and the Guayanan Shield, the genetic differ-entiation of the Amazonian Robust Capuchins islimited. The fact that the phenotypic diversity of theAmazonian Robust Capuchins is not mirrored by acorresponding genetic diversity strongly supportsour theory of allopatric primate speciation. A num-ber of the 16 taxa that are overall recognized in

different taxonomic arrangements (e.g., Groves,2001a; Silva Jr., 2001; Silva Jr., 2002) may wellrepresent taxa ‘in-the-making’. Here, we followSilva Jr. (2001) in recognizing only two species: theGuianan Brown Capuchin Sap. apella with threesubspecies distributed in the eastern Amazon andthe Guianas, and the Large-headed Capuchin Sap.macrocephalus with four subspecies that are dis-tributed across the western Amazon. These taxaform two monophyletic Clades in which littlegenetic differentiation is shown. In contrast, thenon- Amazonian species recognized by Silva Jr. aregenetically distinct forming the monophyletic Sap.nigritus Clade (Figs. 39, 40). Among the six speciesof the extra-Amazonian Sap. nigritus Clade weconsider the Black-horned Capuchin Sap. nigritusthe nearest to archetypic, less bleached species(Figs. 42, 43). Its southernmost populations repres-enting the darkest, overall most saturated eumelaninform may well be a distinct taxon named Sap.cucullatus by Spix in 1823. The Black-hornedCapuchin is the most S occurring of all robustcapuchins. It is distributed in SE Brazil, S of theRios Doce and Grande, extending S through theAtlantic Forest, and taxon Sap. cucullatus furthersouth E of the Rio Paraná into Rio Grande doSul State and NE Argentina. The Black-hornedCapuchin is a large-sized species with horn-liketufts on either side of the head at the temples. Its furis overall very dark brown or grayish in nigritus,and black in Sap. cucullatus, often with slightlypheomelanin bleached, reddish or yellow-fawncolored underparts. A black to dark-grayish crown(with tufts in adults) contrasts much with the lightcolored face. The tail is black. From Sap. nigritusderived the monotypic Crested Capuchin Sap. ro-bustus after a founder-colony of Sap. nigritus tra-versed the Rio Doce to the north. It is distributedin SE Brazil from the Rio Jequitinhonha in BahiaState S to the Rios Doce and Suaçuí Grande in Es-pírito Santo State and E Minas Gerais State, E ofthe Serra do Espinhaço. This taxon is very darkwood-brown or blackish above and on the limbs,with a faint dorsal stripe. The underparts are pheo-melanin bleached red or yellowish, whereas fore-arms, hands, lower legs, and feet are deep darkbrown to black. Its face is dark grayish, with somewhite hairs on the forehead and temples. The crowntufts are tall and conical in shape. From a founder-colony of the northern Black-horned Capuchin Sap.


nigritus that once traversed the Rio Jequitinhonhato the north, derived the Yellow-breasted CapuchinSap. xanthosternos. Yellow-breasted Capuchinstend to be much darker in overall color in thesouthwestern part (N Minas Gerais State), whereasthey are pale in the northern part of this taxon’s dis-tribution. The monotypic taxon Sap. xanthosternosis further distributed in CE Brazil, S and E of theRio São Francisco, south to the Rio Jequitinhonha(in S Bahia State). It is generally pheomelaninbleached brindled reddish above with a sharplymarked, golden-red underside. Tail and limbs re-mained saturated eumelanin black. Its crown doesnot contrast with the body, the cap is black, andthe face and temples are fawn. It has small back-ward pointing tufts. From Yellow-breasted Sap.xanthosternos derived the monotypic BlondCapuchin Sap. flavius, which was described bySchreber in 1774. Until it was collected in 2005, theBlond Capuchin was only known from an earlyillustration. Before colonial times, it must have beendistributed in Coastal NE Brazil from S Rio Grandedo Norte State through Paraíba State into NEPernambuco. This taxon may extend its range to theleft bank of the Rio São Francisco in Alagoas State.The Blond Capuchin is small, distinctive, anduntufted. Its body and limbs are uniformly ad-vanced pheomelanin bleached golden-yellow,whereas its lower-body parts are slightly darkergolden-yellow. Hands and feet are black, whereasthe tail is uniformly golden-blond, but darker on thedorsal side than the rest of the body. It further has arectangular snow-white cap on the front of the head,to just above the ears, and a furless, pendulousthroat flap. Face and forehead are near-albinotic,cream to pinkish colored, the eyes are brown.Sapajus flavius occupies degraded Coastal AtlanticForest and Montrichardia linifera (Arruda) Schott(Araceae) swamp in Pernambuco State, and‘caatinga’ scrub in W Rio Grande do Norte State.Being advanced pheomelanin bleached to near-albinotic, taxon flavius occupies a dead-end distri-bution. It therefore fully concurs with our theory onthe origin of allopatric speciation. The theorysuggests that a founder-colony of progressivelypheomelanin bleached Sap. xanthosternos once wasforced to make a living in the (for Robust Capuchins)marginal or unsuitable habitat of swamps and lowxerophytic, spiny scrub of profusely branchedbushy vegetation up to 8–10 m in height, mixed

with prickly succulent cacti, and spiny, rigid-leavedbromeliads. Blond Capuchins are reported to useeven sand dunes and mangroves.From the northern form of the Black-horned

Capuchin derived to the W the Hooded CapuchinSap. cay, and to the N the Bearded Capuchin Sap.libidinosus (Fig. 42). The monotypic taxon cay isdistributed in SE Bolivia, N Argentina, SW Brazil- W of the Rio Paraná through Mato Grosso Stateinto SW Goiás and Mato Grosso do Sul - andParaguay (E of the Río Paraguay as far as the RíoParaná). The Hooded Capuchin Sap. cay is a small,short- limbed species without sexual dimorphism,typified mainly by its prominent dark dorsal stripe.Sapajus cay is very variable in color, but generallyrather pale. Its crown is pale to blackish-brown,with two small hornlike tufts. Dorsal parts of thebody (shoulders, front of the upper arms, saddle,rump, and thighs) are grayish-brown. Forearms,hands, wrists, lower legs, and feet are blackish.Eyes, nose, and mouth are surrounded by whitehairs. It has a small white beard, and a dark lineextends down from the ears to under the chin. Fromthe Black-horned Capuchin derived to the north themonotypic Bearded Capuchin Sap. libidinosus. Thistaxon is distributed in C and NE Brazil, W and Nof the Rio São Francisco into Maranhão State, andin the W of Piauí State, and E to C Rio Grandedo Norte, NW Paraíba, W Pernambuco, and WAlagoas; to the W it extends to the Rio Araguaia,and its southern limit is the north bank of the RioGrande in Minas Gerais. To the west, the BeardedCapuchin taxon Sap. libidinosus is replaced by Sap.apella, to the east by Sap. flavius, and to the southof the Rio São Francisco by Sap. xanthosternos.Sapajus nigritus occurs just south of the RioGrande. Some hybridization between Sap. libidi-nosus and Sap. nigritus is reported to occur inthe western part of Minas Gerais. The BeardedCapuchin Sap. libidinosus is comparatively smalland does not show sexual dimorphism. It differsfrom all other Robust Capuchins by the rusty-redhair on the back of the neck, the dark-brown preau-ricular stripe running down the side of the face infront of the ears, and the orange-yellow throat anddorsal parts of the body, flanks, outer part of arms,and proximal two-thirds of the tail. Forearms aredark, and the lower back and outer surface of thighsare grayish-brown, mixed with some reddish hairs.The crown is black, with rounded, sometimesbushy, black tufts.


Here, we recognize only two Amazonian Robustor Tufted Capuchins (genus Sapajus): the mono-typic Guianan Brown Capuchin Sap. apella thatis distributed in the eastern Amazon and in theGuianas, and the Large-headed Capuchin Sap. mac-rocephalus with a number of forms/morphs/sub-species that are distributed throughout the westernAmazon as far north as the Magdalena Valley in NColombia (Fig. 42). Taxon Sap. apella is found inthe rain forests of the Amazon Basin of Brazil N ofthe lower Rios Negro and Amazonas, E of the RioBranco, extending N to the southeastern part of theOrinoco Delta in Venezuela and the Guianas. Itsdistributional limits in the S, SE, and E are definedby the extent of the Amazon rain forest, in the S andE of Maranhão State marking the transition zone toxeric deciduous forest and ‘caatinga’ scrub. In theWest, its distribution is limited by the interfluve of

the Rios Negro and Solimões and the Rio MadeiraBasin. The Guianan Brown Capuchin species Sap.apella is relatively large and heavily built, with abroad head, flat face, and short limbs. Its coatis long and coarse, with all five extremities darkercolored than the rest of the body. It is generallygray-fawn to dark brown above, with a yellowishor red underside. The lower limbs and tail are black,and there is a variably distinct dorsal stripe. Theface and temples are light gray-brown. The crowntuft is black and forms short tufts above the ears (thecharacteristic ‘horns’). The crown cap extendsdown the cheeks forming ‘sideburns’ that oftenmeet below the chin. There is no sexual dimorph-ism, but males are slightly heavier and often overallmuch darker colored. The Margarita Island Capuchintaxon Sap. apella margaritae that is endemic to Islade Margarita off the Caribbean coast of Venezuela


Figure 44. Phylogeography, allopatric speciation, and metachromic bleaching in all Amazonian Robust Capuchins disputedly divided up in Sapajus apella and S. macrocephalus, the latter with different taxa ’in-the-making’.

distinguishes itself from the nominate GuiananBrown Capuchin by longer dark sideburns in frontof the ears, and progressively bleached, pale-yellowor straw colored, near-albinotic upper arms andshoulders. The thighs and rump are pale yellow-brown, and flanks, lower back, and upper chest arepale brown, becoming paler from the upper backto the neck. The face is grayish, tinged pink onthe cheeks and chin. The black cap extends in a “V”to between the eyes, with small round tufts abovethe eyes.The monotypic Large-headed Capuchin taxon

Sap. macrocephalus is distributed in the westernAmazon Basin, but its taxonomy and distributionallimits are poorly defined. According to Silva Jr.(2001) this species includes the forms/morphs/sub-species Sap. fatuellus from the upper MagdalenaValley, Colombia, Sap. maranonis from Rio Ham-

burgo, Peru, Sap. pallidus from the Río Beni, C+NBolivia, and Sap. juruanus from the Rio Juruá,Brazil. Preliminary genetic studies in 2012 failed toindicate that Sap. apella and Sap. macrocephaluswere distinct taxa. Large-headed Capuchins aredistributed across the upper Amazon Basin in EColombia, north as far as the Río Arauca on theborder with Venezuela, E Ecuador, E Peru, WBrazil, and C and N Bolivia (S at least as far as theupper Río Beni). Their overall coat color is gray-brown or ochreous to dark brown above, with adark dorsal stripe, and yellow-fawn or red-goldbelow. Sides of the neck are lighter, upper arms arepale yellowish, and legs are black with yellow-fawnor red-gold below. Adults have high, pointed crowntufts that resemble horns, which become reducedwith age. There is often a gray-white stripe runningfrom eye to ear. Four forms of the Large-headed


Figure 45. Phylogeography, allopatric speciation, radiation, and metachromic diversification in all hitherto recognized taxa of Night Monkeys, genus Aotus.

Capuchin have been distinguished (Fig. 44). TheColombian form Sap. fatuellus is bright brownabove and red below, having a prominent dorsalstripe. Its face is almost naked and dark-purplish toflesh-colored. The Peruvian form Sap. maranonisis uniformly dark chestnut-brown above, becomingmore reddish towards the flanks, and deep yellow-brown below. Its legs, tail, and (sometimes) fore-arms are black. Its cap is distinctly black, whereastemples and sides of the crown are often white. Ithas a crescent-shaped whitish patch above each eye.There are no crown tufts or they are minimal. TheBrazilian form Sap. juruanus is reddish-brownabove with a very distinct blackish dorsal stripe.The throat and upper chest are blackish or palereddish-buff, and limbs and tail are dark brown orblack. The Bolivian form Sap. pallidus from southof the Río Madre de Dios has also been referred toas a subspecies of Sap. libidinosus, but such tax-onomy would be conflicting with our theory on al-lopatric speciation, for Sap. libidinosus fromCE Brazil is overall more advanced pheomelaninbleached in comparison with Sap. pallidus. Both theColombian morph/taxon Sap. fatuellus of theLarge-headed Capuchin Sap. macrocephalus andthe insular Margarita Island Brown Capuchin Sap.apella margaritae are in their overall advancedpheomelanin bleached coat coloration clearly fol-lowing the metachromic pathway to albinotic, andtherefore fully concur with our theory of allopatricspeciation (Fig. 44).Night Monkeys or Douroucoulis genus Aotus rep-

resent a very old lineage that is generally placed ina family of its own - Aotidae. The molecular geneticevidence classifies them as a subfamily of the Ce-bidae. There is also morphological evidence toplace Aotus in the Pitheciidae. There are generallytwo Groups distinguished: the “Gray-neckedGroup” (characterized by grayish to brownishagouti sides of the neck and body), which occursnorth of the Amazon River, and the “Red-neckedGroup” (characterized by partly or entirely orangeor yellowish sides of the neck and chest, muchcontrasting with the grayish to brownish-agouticolored sides of the body), which occurs south ofthe Amazon River (Mittermeier et al., 2013). Re-cently, up to eleven species have been recognized,of which at least seven in the Gray-necked Group:the Lemurine Night Monkey Ao. lemurinus, thePanamanian Night Monkey Ao. zonalis, Brumback’s

Night Monkey Ao. brumbacki, the Gray-leggedNight monkey Ao. griseimembra, Spix’s NightMonkey Ao. vociferans, Humboldt’s Night MonkeyAo. trivirgatus, and Hernández-Camacho’s NightMonkey Ao. jorgehernandezi (Figs. 45, 46). In theRed-necked Group are recognized four species: theAndean Night Monkey Ao. miconax, Ma’s NightMonkey Ao. nancymaae, the Black-headed NightMonkey Ao. nigriceps, and Azara’s Night MonkeyA. azarae (Figs. 45–47). Sexual dimorphism innight monkeys is absent. They are also not sexuallydichromatic in coloration and facial markings. Thecoat is in metachromic sense primitive, archetypicsaturated eumelanin, grayish to grayish-tan with apheomelanin bleached, lighter tan or yellowishunderside. In Red-necked species, ventral surfacesof neck, chest, abdomen, and inner sides of armsand legs are orangish or russet colored. The faceshave white patches over eyes, topped by blackstripes, and a triangular black patch running fromthe center of the forehead down between the eyes.Black stripes are also extending from the lateral sideof each eye to the forehead, varying in width anddarkness, and may or may not converge posteriorlywith the central stripe. Tails are generally agouti-brown, distally black-tipped. Night Monkeys mostlikely descended from a diurnal haplorrhine. Theyonly are secondarily nocturnal and have retainedtheir color vision.Within the Gray-necked Clade Ao. lemurinus is

the nearest to archetypic, saturated eumelanin, lessbleached taxon. It is a montane species of theColombian Andes range, at elevations above 1,000-1,500 m, in the upper Río Cauca Valley and on theslopes of the Cordillera Oriental (but not in theMagdalena Valley that is occupied by the Gray-legged Night Monkey Ao. griseimembra), extend-ing its range S into Ecuador through the humid sub-tropical forests of the Cordillera Oriental. TheLemurine Night Monkey is rather shaggy and long-haired, with the upperparts of the body often eu-melanin grayish to buffy-agouti, with a poorlydefined brownish medial dorsal band. The under-side of the body is pheomelanin bleached yellowishto pale orange. Inner and outer sides of limbs areentirely grayish-agouti, or the inner sides have ayellowish to pale orange tone extending from thechest and belly to the mid-arm or mid-leg. Handsand feet are dark. Temporal stripes may be separ-ated or united behind the head. From ancestral Ao.


lemurinus derived the Gray-legged Night Monkeytaxon Ao. griseimembra. It is distributed in NColombia and NW Venezuela. It occurs in the RíoMagdalena Valley and northern lowland forests ofColombia (including the Sierra Nevada de SantaMarta and the Ríos Sinu and San Jorge basins),extending into Venezuela in the vicinity of LakeMaracaibo. It is grayish to brownish-agouti on theside of the neck. Upperparts are grayish to buffy;chest, belly, and inner surfaces of the legs are brown-ish or yellowish to pale orange. Pelage is relativelyshort. Hands and feet are light-brown. From taxonAo. griseimembra derived to the NW the monotypicPanamanian Night Monkey zonalis. This taxon isdistributed in NW Colombia in the Pacific low-lands, S towards the Ecuadorian border, and W intomost of Panama; it is absent from SW Panama(Chiriquí). Its overall coat color is brownish in theCanal Zone and Colombia, but it grades into palerand grayer tones along the upper Río Tuira, EPanama. From Ao. zonalis derived Hernández-Camacho’s Night Monkey Ao. jorgehernandezi.This monotypic taxon is believed to occur in the(sub)-montane tropical forests on the westernslopes and foothills of the W Colombian Andes(in Quindío and Riseralda). It is advanced bleachedto albinotic in the head and ventral parts. Its facehas two discrete supraocular white patches separ-ated by a broad black frontal stripe. Moreover,subocular white bands of fur are separated by a thinblack malar stripe on each side of the head. Ventralparts of the arms from the wrists running up into thechest and belly are of a thick white fur (Fig. 39).From the Gray-necked Night Monkey Clade’snearest to archetypic taxon Ao. lemurinus derivedto the SE first Brumback’s Night Monkey Ao.brumbacki. This monotypic taxon is distributed inNC Colombia in the eastern part of Boyacá De-partment, E to the highlands of Meta (to at least1,500 m above sea level). Its coat is dorsallygrayish-buffy agouti colored with a dark brownmid-dorsal zone. Ventral parts extending to theelbows, knees, and lower throat are pale orange.Sides of the neck are entirely grayish or brownishagouti, like the flanks and outer sides of the arms.The head shows well-marked, thin, brownish-blacktemporal stripes. The white above the eyes is yel-lowish, and the white on the face extends to thechin. From Ao. brumbacki derived first to the SSpix’s Night Monkey Ao. vociferans. This mono-

typic taxon is widespread in the upper AmazonBasin, extending from NW Brazil (W of the Negro,Uaupés, and Amazonas-Solimões Rivers) into SEColombia (S of the Río Tomo, Orinoco Basin), andS into the Ecuadorian Amazon and NE Peru (as farsouth as the north bank of the Marañón-AmazonasRiver). It occurs also S of the Rio Solimões in asmall area on the lower Rio Purús. Spix’s NightMonkey’s coat is brown-toned above, with an over-all white, slightly orange tinged underside, extend-ing to the wrists, ankles, and chin. Hands and feetare black. The proximal one-third to one-half of theventral side of the tail is reddish or grayish- red, therest is black. The crown stripes on the head are thickand brownish, with white fur above the eyes con-fined to two small patches grading into the agouti-colored crown. The temporal stripes are unitedbehind, and the malar stripe can be well defined toabsent. The face is white, except for the chin. FromAo. vociferans derived to the N and E the mono-typic Humboldt’s Night Monkey taxon Ao. trivir-gatus. It is widespread across N Brazil, N ofthe Rios Negro and Amazonas and W of the RioTrombetas, N into SC Venezuela and E Colombia.Sides of the neck are grayish-agouti to mainlybrownish-agouti colored. Upper parts of the bodyare grayish to buffy-agouti. The inner sides of thelimbs, extending to the wrists and ankles, are sim-ilar in color to the orange-buffy of chest and belly.The face has triradiate brown stripes. It is rathergrayish in comparison with the usual white of otherNight Monkeys. Hands and feet are dark-brown.Taxon Ao. trivirgatus can be distinguished from allother Night Monkeys by its parallel temporal stripeson the head and the lack of an interscapular whorlor crest (Figs. 45, 46).Within the Red-necked Clade of Night Mon-

keys, the Black-headed Night Monkey Ao. nigri-ceps is the nearest to archetypic, less pheomelaninbleached taxon (Fig. 47). This monotypic species isdistributed in the Brazilian Amazon, S of the RioAmazonas-Solimões and W of the Rio Tapajós-Juruena, as far south as the right bank of the RioGuaporé and the left bank of the Río Madre de Diosin N Bolivia. It occurs also in SE Peru, west to theRío Huallaga, and north as far as the Río Cush-abatay. Its coat is iron-gray above and brownish-agouti on the dorsum. The underside is orangecolored with white tones, extending to the neck,throat, chin, and sides of the jaw and also to the


Figure 46. Radiation and metachromic diversification in the Gray-necked Night Monkey Group (Aotus), following eumelaninand pheomelanin pathways of metachromic bleaching, in particular in the head, proximal half of the tail, and ventral partsof the body.


Figure 47. Radiation and metachromic diversification in the Red-necked Night Monkey Group (Aotus), following eu-melanin and pheomelanin pathways of metachromic bleaching, in particular in the head, tail, and ventral parts of thebody.


inner surfaces of the wrists and ankles. The cap isblack, the face stripes are broad, and it has distinctareas of white on the face.From the Black-headed Night Monkey Ao. nigri-

ceps derived to the W Ma’s Night Monkey Ao.nancymaae. This monotypic taxon ranges in WBrazil (S of the Rio Solimões from the Rio Javarías far east as the Rio Jandiatuba) and NE Peru (fromthe Rio Javarí W to the Río Huallaga). This taxonis also found in an enclave between the lower RíosTigre and Pastaza. The upper parts of its coat aregrayish-agouti, with a dark mid-dorsal zone and apale orange underside, extending up the sides of theneck and inner limbs. The proximal part of the tailis orange, with a blackish stripe above; the under-side is blackish. Its face is grayish-white, the crownstripes are narrow and dark brown colored, and thesides of the throat and jaw are colored like the body(Fig. 41). From Ao. nancymaae derived to the Wthe Andean Night Monkey taxon Ao. miconax. Thismonotypic night monkey is endemic to Peru. It isconfined to a small area S of the Río Marañón andW of the Río Huallaga. It inhabits the primary andsecondary humid, lower-montane cloud forests inthe Andes at elevations of 800–2,800 m.Upper sides of its coat are light gray with a

brownish tint, often quite infused with red-brown.Its underside is pale orange, extending forward asfar as the chin and on the inner sides of the limbs.Outer surface of the body is overall brownish tobuffy-agouti. The tail is bushy, its upper side isblackish, its lower side reddish-orange. Head partsand throat are advanced bleached to near-albinotic.From the nearest to archetypic Red-necked Black-headed Night Monkey Ao. nigriceps derived in op-posite direction (to the S and E) Azara’s NightMonkey species Ao. azarae. Three subspecies ofAo. azarae are recognized: the nominate taxonAo. azarae, distributed in SC Brazil, S Bolivia,Paraguay, and N Argentina; taxon Ao. boliviensis,distributed in SE Peru and Bolivia east of theAndes; taxon Ao. infulatus, distributed in Brazil, Sof the Rio Amazonas (but with a small enclave inthe SE tip of Amapá State), including Marajó andCaviana Islands, extending east into MaranhãoState as far as the Rio Parnaíba, S along the westbank of the Rio Tocantins to the Pantanal of MatoGrosso. Taxon Ao. azarae infulatus’s western limitis the Rio Tapajós-Juruena. Azara’s Night MonkeyAo. azarae is highly variable. It generally has an in-

terscapular whorl. Taxon Ao. azarae has a long,thick, and shaggy fur that is grayish to pale buffy-agouti above and pale whitish-orange below. Facialstripes are narrow. The basal hairs of the distal ¼of the tail are orange. Taxon Ao. boliviensis has arelatively short fur, with an olive tone above andcontrastingly grayer on the limbs. The facial stripesare very narrow except where the middle one ex-pands on the crown; the black temporal stripe in thistaxon is poorly defined, the black malar stripe isfaint or absent, and there is usually a whitish bandbetween the eyes and temporal stripe. There is aconspicuous whorl between the shoulder blades.The third taxon Ao. infulatus, the “Feline NightMonkey”, is very similar to subspecies Ao. bolivi-ensis, but the white on the face is more prominent.There is no whitish band between the eyes and thetemporal stripe as there is in Ao. boliviensis. Thetemporal stripes are black, well defined, and con-tinuous with the malar stripe. The tail is reddishthroughout its length except for the black tip. Theorange color of the underparts extends to or abovethe ventral one-half of the sides of the neck. Thecolor of the throat varies from orange, with theanterior one-half grayish-agouti to entirely orangecolored (Fig. 47).Our theory suggests that the trend to allopatry

in Neotropical primates resulted from a specifickind of social selection. That the discrimination ofsomewhat deviant mutant young males by high-ranking males, which push them toward the peri-phery of the parental group’s range, has been thetrue driver behind metachromic bleaching onthe evolutionary path along which a certain race,species, phylogenetic clade, or genus has extendedits geographic range in the past. As any founder-colony or population at the limit of a taxon’s currentrange will represent a narrow gene pool, through in-breeding certain phenotypic characters (e.g., localdepilation of the skin, change of coloration of theskin, pelage or parts of it) will initially be reinforcedand advance more rapidly within the population.Through the process of metachromism (changinghair and skin color) with the trend to allopatry asthe behavioral driving force, speciation, radiation,and phylogeography can be retraced and wellexplained for in all extant Neotropical primates.According to the principle of metachromic bleach-ing, extant primate taxa at the base of a phylogen-etic tree or clade are in general agouti or saturated


eumelanin colored. They are the least colorful,black(ish) or dark brown toned, and therefore con-sidered to be nearest to the ancestral, archetypic,primitive or original form.Geographic variation and diversification in

color patterns of the coat among Neotropical mon-keys demonstrates with unusual clarity the unilat-eral direction and irreversibility of processes thatlead to progressively metachromic bleached andultimately (near)-albinotic allopatric forms, irre-spective of environmental factors (Figs. 1–47). Theessentially behavioral and genetic driving forcesbehind metachromic processes, though, have neverbeen studied. They are generally considered enig-matic. The reason may be that they seem to disobeycommonly accepted Darwinian rules of evolution.Different from birds, in territorial (Neotropical)monkeys metachromic changes in coat color towardbleaching or albinotic and/or all sorts of local hairgrowth or loss of hair (depilation) do not seem toplay an essential role in sexual display and mateselection. Consequently, they may seem to be non-adaptive. In the wild only rarely one is able towitness how exactly processes of metachromicbleaching do work out. For instance: when some-what bleached or depilated deviant young males arebeing pushed from the center into the periphery ofa ranging or foraging group. Or: when ‘outcast’males do join in all-male parties. Or: when suchparties set out to look beyond the horizon, for meresurvival willing to overtake any habitat delimitationor geographic barrier found on their ‘path to allo-patry’. These crucial data will only come availablewhen fieldworkers, like we did, do live for pro-longed periods of time among undisturbed primatepopulations in pristine tropical forest environment.As very few primatologists have done so, at least inthe Neotropics, and sample sizes are consequentlytoo small to be published and divulged, it is im-possible for us to add more references then our ownon the matter. Even though, living over more thana decade in permanent intimate contact with pristinenature, both in the Brazilian Amazon and in the over-all even better preserved Guayanan Shield, led usto believe that high-ranking males pushing slightlybleached and/or depilated young males to the peri-phery of a group’s range, or sometimes beyond itsboundaries, could plausibly be the true and prin-cipal motor or driver behind allopatric speciationand radiation of taxa in nearly all Neotropical

primate genera - Pygmy Marmosets (Cebuella),Tamarins (Saguinus), Amazonian Marmosets(Mico), True Marmosets (Callithrix), Lion Tamar-ins (Leontopithecus), Sakis (Pithecia), BeardedSakis (Chiropotes), Uakaris (Cacajao), Titi Mon-keys (Callicebus), Night Monkeys (Aotus), SquirrelMonkeys (Saimiri), Gracile/Untufted CapuchinMonkeys (Cebus), Robust/Tufted Capuchin Mon-keys (Sapajus), Howling Monkeys (Alouatta),Woolly Monkeys (Lagothrix), Spider Monkeys(Ateles), and Woolly Spider Monkeys (Brachyteles).Interestingly, but concurring with our theory (forthose monkeys that do not defend a common ter-ritory), metachromic bleaching did not take placein peaceably living monkeys like the archetypicagouti and saturated eumelanin colored Black-crowned Dwarf Marmosets Callibella humilis, anewly identified monotypic genus of diminutivecallitrichid monkeys (Figs. 2, 3). Nor did it takeplace in saturated eumelanin all-black Goeldi’sMonkeys (Callimico goeldii) - the only other mono-typic primate genus in the Neotropics that does notbehave territorial in any sense and therefore doesnot defend a common living space against the neigh-bors of its own kind (Fig. 4). Their external featuresshowing archetypic agouti and saturated eumelanincoat coloration without any sign of metachromicbleaching are in full accordance with their geneticsthat put them at the base of their respective phylo-genetic trees. It further corroborates our theory onthe origin of allopatric speciation in primates andthe principle of metachromic bleaching, for DwarfMarmosets and Goeldi’s Monkeys are equally so-ciable, peaceable little monkeys that do not demon-strate any rate of territorial defense. The primitiveagouti and saturated eumelanin (blackish-brown)Black-crowned Dwarf Marmoset stands at the baseof the phylogenetic tree of all Amazonian marmo-sets (Van Roosmalen & Van Roosmalen, 2003). Itrepresents the nearest to ancestral, archetypic mar-moset from which all extant, advanced and highlyterritorial Amazonian marmosets (genus Mico) andpygmy marmosets (genus Cebuella) have derivedin the Late Pleistocene.Our theory is firmly rooted in over 30-year field-

work on primates, both in the Guianas and in theentire lowland Amazon Basin. From the very be-ginning we have given special attention to issueslike socio-ecology, ecological feeding niches, ter-ritorial behavior, distributions, and phylogeography.


Simultaneously, we have kept, raised, bred, rehab-ilitated, and reintroduced back into the wild entirefamilies or social groupings of a multitude of mon-key taxa representing about all hitherto known Neo-tropical primate genera. Many unique, extremelyrare or sometimes once- in-a-lifetime observationsthat we gathered in pristine tropical rainforest en-vironment as well as in captivity (the bulk of itnever published inherent to ‘insignificant’ samplesizes) now do add up to the validity of our theory.It basically helps us to better understand the com-plex distribution patterns, phylogeography, diversi-fication, speciation, and radiation in Neotropicalprimates. Most likely, the theory applies to all theworld’s primates (including man), as long as thetaxa exhibit social groupings that defend a commonliving space, home range, or territory. The fact thatonly two Neotropical primate genera (Callimicoand Callibella) are monotypic strongly supportsour theory, as it does not apply to peaceable, non-territorial social primates. By boat, canoe, and onfoot we have surveyed entire basins of a number ofmajor tributaries of the mighty Amazon River tostudy primate diversity and distributions across theentire Amazon Basin, including also large parts ofthe Brazilian and Guyanan Shields. We have testedand empirically come to fully validate Alfred Rus-sel Wallace’s river-barrier hypothesis that he firstlaid down in his 1852 account On the Monkeys ofthe Amazon, and later in his 1876 paper “The Geo-graphical Distributions of Animals”. Herein, Wal-lace points at the larger rivers that he sailed as theprincipal evolutionary cause of the Amazon’s richextant primate diversity and complex biogeography,since many rivers effectively block off gene flowbetween populations along opposite riverbanks(genetic isolation). As the Amazon still representsa largely pristine and vast natural realm that is (notyet) drastically and irreversibly modified by humaninterference, no better place to study and retraceevolutionary processes that may have acted uponprimates and other mammals since the Pliocene era,no matter on which continent. Moreover, mostrivers that in the course of millions of years haveplayed a significant role in the demography ofAmazonian primates - the majority of which cannotswim or fly- remain acting as such. Therefore,distributions of primate taxa in the Amazon, ifcorrectly studied, documented, and taxonomicallytreated, do follow a more transparent and rational

overall pattern in comparison with those of the OldWorld. In SE Asia, instead of rivers, the oceanplayed an equally important role in the island bio-geography of mammals. And in Africa (includingMadagascar), the landscape with its complex anddiffuse mosaic of vegetation types and habitatsseems to have played a more determinant role thanrivers in primate distributions. Moreover, massivehuman disturbance has long irreversibly changedthe landscape of the Old World. This may haveobscured to some extent the principal factors thatinfluenced and determined distributions and phylo-geography in catarrhine primates, most importantlythe hominins.

CONCLUSIONS

Here we discuss the above proposed doctrine onthe origin of allopatric primate species and the prin-ciple of metachromic bleaching among Neotropicalprimates as a conclusive socio-ecological answer tothe question: why primates are such a highly diver-sified, species-rich, and colorful order in the ClassMammalia. The Order Primates contains a worldtotal of 73 genera, 414 IUCN-recognized species,and 612+ known taxa of which roughly one thirdare found in the Neotropics (Mittermeier et al.,2013) (see also Table 1). Globally, only the rodents(Order Rodentia) outnumber the Order Primates.However, compared to primates, rodents are by farnot that diversified. They are mostly opportunists,not very sociable, and not particularly colorful.While studying color variation in callitrichid mon-keys, Hershkovitz (1968; 1977) proposed the“Theory of Metachromism.”. He attributed evolu-tionary change in mammalian tegumentary colorsto social, sexual, and predatory selection, as itseems to be the case in birds. He argued that thehighly ‘visually’ adapted primates may be predis-posed to select mates based on coat color and hairadornments. However, primates generally do notsexually display their skin and coat colors, or hairdresses, except for a few genera in the Old World(e.g., Theropithecus, Mandrillus). Instead, somedisplay their genitals, like both sexes of AmazonianMarmosets (genus Mico) do. Or, both sexes ofBearded Sakis (Chiropotes), female Spider Mon-keys (Ateles) or male Woolly Spider Monkeys(Brachyteles) do. In that case, their genitals are


mostly hypertrophied (e.g., Mico, Chiropotes, Bra-chyteles, Pan). Hershkovitz’s key hypothesis ofmetachromism, which is tested in tamarins (genusSaguinus) and confirmed for many of its predictionsby Jacobs et al. (1995), concerns the orderly, irre-versible loss of pigment within chromogeneticfields. Its key concept is that genetic drift togetherwith social selection could fix phenotypes departingfrom primitive agouti or saturated eumelanin(blackish-brown) fields by various degrees ofso-called “metachromic bleaching”. Thus, an al-binotic (nearly white) coat would represent the endpoint of geographic variation in a series of near-al-lopatric forms (color morphs) deriving ultimatelyfrom an agouti-colored or saturated eumelanin pig-mented ancestral form. Using metachromism, wehave demonstrated that most Amazonian monkeygenera are monophyletic and composed of two ormore major phylogenetic Groups or Clades. Wefound only two genera (i.e., Callibella and Cal-limico) to be monotypic. Contrary to Hershkovitz,who followed the Darwinian fallacy of adaptiveevolution by linking evolutionary change in mam-malian tegumentary colors (‘bleaching’) to social,sexual, and predatory selection, we suggest to at-tribute metachromic diversification in extant socialand territorial primates exclusively and uniquely to“male social selection”. We propose the “trend toallopatry in somewhat metachromic bleachedand/or depilated varieties” to be the principalmechanism and driver behind speciation, radiation,and phylogeography in group-living Neotropicalmonkeys that defend the group’s living space. Itarguably applies also to any group-living territorialprimate worldwide, including our own species andits ancestors (be it hominids or hominins). For allnineteen genera of Neotropical primates we havepresented distribution maps of all known extanttaxa and indicated the geographic barriers (rivers,lakes, mountain ranges, seasonally inundated flood-plain forests, open scrub areas, etc.) delineatingeach taxon’s distribution. We have also elaboratedthe phylogeography and radiation within eachmonophyletic cladistic Group or Clade and relatedthem to the irreversible patterns of metachromicbleaching. Through the process of metachromism(changing hair and skin colors) with the trend to al-lopatry as the behavioral driving force, speciation,radiation, and phylogeography can be well retracedand explained for in all extant Neotropical primates.

According to the principle of metachromism, prim-ate taxa at the base of a phylogenetic tree or cladeare in general agouti or saturated eumelanin (blackor blackish-brown) colored - that is the least color-ful. Within that Clade they are considered thenearest to ancestral, archetypic, primitive, or ori-ginal taxon. Based on metachromic skin and furcharacters, without a single exception, we were ableto retrace phylogeographic pathways of speciationand radiation that were plausibly followed in theevolutionary history of each monophyletic Clade.In all cases we could confirm the trend to allopatryfollowing irreversible eumelanin and pheomelaninpathways of metachromic bleaching. The farther ataxon radiated away from the origin or center ofthe Clade’s dispersion, the more progressively eu-chromic or bleached and eventually albinotic itscoat/pelage, or part of it, will become.The great majority of primates are sociable,

group-living animals. Group sizes vary from nuc-lear families (4–7 individuals) to troops of mixedage and sex classes containing 15 to over 200 indi-viduals. The far majority of the world’s primate so-cieties are socially structured in a hierarchic wayand based on male dominance and ranking. Maledefense of the group and its living space within apopulation benefits from male social selection.Even in matriarchally organized social groups, suchas those of spider monkeys (Ateles) and pygmychimpanzees or bonobos (Pan), males associate inall-male parties to jointly patrol and defend thegroup’s territory or living space. In social conflictsamong males over ranking, inferior males as wellas mutant males that show somewhat different,deviant phenotypic characters (such as a slightlybleached pelage here or there or depilated skin incertain body parts) will be pushed into the peripheryof the group during ranging and foraging. We haveseen this happening, both in the wild and in semi-free ranging conditions, in particular in socialgroups and societies of monkeys like Lagothrix,Ateles, Cebus, Sapajus, Saimiri, Cacajao, or Chiro-potes. Depending on the species, such young malesalso happen to be expelled from the parental group.We have witnessed this in wild and semi-freeranging populations of Alouatta, Callicebus, Mico,Cebus, Sapajus, and Pithecia. Either way, thechances of outcast males to survive and pass ontheir mutant genes are utterly slim. If this wouldhappen in other mammals - being comparatively


Alouatta Lacépède, 1799

Alouatta arctoidea Cabrera, 1940Alouatta belzebul (Linnaeus, 1766)Alouatta caraya (Humboldt, 1812)Alouatta discolor (Spix, 1823)Alouatta guariba guariba (Humboldt, 1812)Alouatta guariba clamitans Cabrera, 1940Alouatta macconnelli Elliot, 1910Alouatta nigerrima Lönnberg, 1941Alouatta palliata palliata (Gray, 1848)Alouatta palliata aequatorialis Festa, 1903Alouatta palliata coibensis Thomas, 1902Alouatta palliata mexicanaMerriam, 1902Alouatta palliata trabeata Lawrence, 1933Alouatta pigra Lawrence, 1933Alouatta sara Elliot, 1910Alouatta seniculus seniculus (Linnaeus, 1766)Alouatta seniculus juara Elliot, 1910Alouatta seniculus puruensis Lönnberg, 1941Alouatta ululata Elliot, 1912

Aotus Illiger, 1811

Aotus azarae azarae (Humboldt, 1812)Aotus azarae boliviensis Elliot, 1907Aotus azarae infulatus (Kuhl, 1820)Aotus brumbacki Hershkovitz, 1983Aotus griseimembra Elliot, 1912Aotus jorgehernandezi Defler et Bueno, 2007Aotus lemurinus I. Geoffroy Saint-Hilaire, 1843Aotus miconax Thomas, 1927Aotus nancymaae Hershkovitz, 1983Aotus nigriceps Dollman, 1909Aotus trivirgatus (Humboldt, 1811)Aotus vociferans (Spix, 1823)Aotus zonalis Goldman, 1914

Ateles É. Geoffroy Saint-Hilaire, 1806

Ateles belzebuth É. Geoffroy Saint-Hilaire, 1806Ateles chamek (Humboldt, 1812)Ateles fusciceps fusciceps Gray, 1865Ateles fusciceps rufiventris Sclater, 1872Ateles geoffroyi geoffroyi Kuhl, 1820Ateles geoffroyi azuerensis (Bole, 1937)Ateles geoffroyi frontatus (Gray, 1842)Ateles geoffroyi grisescens Gray, 1865Ateles geoffroyi ornatus (Gray, 1870)Ateles geoffroyi vellerosus Gray, 1865Ateles geoffroyi yucatanensis Kellogg et Goldman, 1944Ateles (hybridus) hybridus I. Geoffroy Saint-Hilaire, 1829Ateles (hybridus) brunneus Gray, 1870Ateles longimembrisAllen, 1914Ateles marginatus É. Geoffroy Saint-Hilaire, 1809Ateles paniscus (Linnaeus, 1758)

Brachyteles Spix, 1823

Brachyteles arachnoides (É. Geoffroy Saint-Hilaire, 1806)Brachyteles hypoxanthus (Kuhl, 1820)

Cacajao Lesson, 1840

Cacajao (calvus) calvus (I. Geoffroy Saint-Hilaire, 1847)Cacajao (calvus) novaesi Hershkovitz, 1987Cacajao (calvus) rubicundus (I. Geoffroy Saint-Hilaire

et Deville, 1848)Cacajao (calvus) ucayalii (Thomas, 1928)Cacajao (melanocephalus) melanocephalus

(Humboldt, 1812)Cacajao (melanocephalus) ayresi Boubli, Silva,

Hrbek, Pontual et Farias, 2008 Cacajao (melanocephalus) hosomi Boubli, Silva,

Hrbek, Pontual et Farias, 2008 Cacajao ouakary (Spix, 1823)

Callibella van Roosmalen M.G.M. et van Roosmalen T., 2003

Callibella humilis (van Roosmalen M.G.M., van Roosmalen T., Mittermeier et de Fonseca, 1998)

Callicebus Thomas, 1903

Callicebus aureipalatiiWallace, Gómez, Felton A. et Felton A.M., 2006

Callicebus baptista Lönnberg, 1939Callicebus barbarabrownae Hershkovitz, 1990Callicebus bernhardi van Roosmalen M.G.M., van

Roosmalen T. et Mittermeier, 2002Callicebus brunneus (Wagner, 1842)Callicebus caligatus (Wagner, 1842)Callicebus caquetensis Defler, Bueno et Garcia, 2010Callicebus cinerascens (Spix, 1823)Callicebus coimbrai Kobayashi et Langguth, 1999Callicebus cupreus (Spix, 1823)Callicebus donacophilus (d’Orbigny, 1836)Callicebus dubius Hershkovitz, 1988 Callicebus hoffmannsi Thomas (1908)Callicebus lucifer Thomas, 1914Callicebus lugens (Humboldt, 1812)Callicebus medemi Hershkovitz, 1963Callicebus melanochir (Wied-Neuwied, 1820)Callicebus modestus Lönnberg, 1939Callicebus moloch (Hoffmannsegg, 1807)Callicebus nigrifrons (Spix, 1823)Callicebus oenanthe Thomas, 1924Callicebus olallae Lönnberg, 1939Callicebus ornatus (Gray, 1866)Callicebus pallescens Thomas, 1907Callicebus personatus (É. Geoffroy Saint-Hilaire, 1812)Callicebus purinus Thomas, 1927Callicebus regulus Thomas, 1927


Table 1/1. References of scientific descriptions of all known Neotropical primates (present paper).

Callicebus stephennashi van Roosmalen M.G.M., van Roosmalen T. et Mittermeier, 2002

Callicebus torquatus (Hoffmannsegg, 1807Callicebus vieirai Gualda-Barros, Nascimento et Amaral, 2012

Callimico Miranda Ribeiro, 1912

Callimico goeldii (Thomas, 1904)

Callithrix Erxleben, 1777

Callithrix aurita (É. Geoffroy Saint-Hilaire, 1812)Callithrix flaviceps (Thomas, 1903)Callithrix geoffroyi (Humboldt, 1812)Callithrix jacchus (Linnaeus, 1758)Callithrix kuhlii Coimbra-Filho, 1985Callithrix penicillata (É. Geoffroy Saint-Hilaire, 1812)

Cebuella Gray, 1865

Cebuella (pygmaea) pygmaea (Spix, 1823)Cebuella (pygmaea) niveiventris Lönnberg, 1940

Cebus Erxleben, 1777

Cebus aequatorialisAllen, 1914Cebus albifrons (Humboldt, 1812)Cebus brunneus Allen, 1914Cebus capucinus capucinus (Linnaeus, 1758)Cebus capucinus curtus Bangs, 1905Cebus cesarae Hershkovitz, 1949Cebus cuscinus Thomas, 1901Cebus imitator Thomas, 1903Cebus kaapori Queiroz, 1992Cebus leucocephalus Gray, 1865Cebus malitiosus Elliot, 1909Cebus olivaceus olivaceus Schomburgk, 1848Cebus olivaceus castaneus I. Geoffroy Saint-Hilaire, 1851Cebus unicolor Spix, 1823Cebus versicolor Pucheran, 1845Cebus yuracus Hershkovitz, 1949

Chiropotes Lesson, 1840

Chiropotes albinasus (I. Geoffroy Saint-Hilaire et Deville, 1848)

Chiropotes chiropotes (Humboldt, 1812)Chiropotes sagulatus (Traill, 1821)Chiropotes satanas (Hoffmannsegg, 1807)Chiropotes utahickae Hershkovitz, 1985

Lagothrix É. Geoffroy Saint-Hilaire, 1812

Lagothrix (cana) cana (É. Geoffroy Saint-Hilaire, 1812)Lagothrix (cana) tschudii Pucheran, 1857Lagothrix (lagotricha) lagotricha (Humboldt, 1812)

Lagothrix (lagotricha) lugens Elliot, 1907Lagothrix poeppigii Schinz, 1844

Leontopithecus Lesson, 1840

Leontopithecus caissara Lorini et Persson, 1990Leontopithecus chrysomelas (Kuhl, 1820)Leontopithecus chrysopygus (Mikan, 1823)Leontopithecus rosalia (Linnaeus, 1766)

Mico Lesson, 1840

Mico acariensis (van Roosmalen M.G.M., van Roosmalen T., Mittermeier et Rylands, 2000)

Mico argentatus (Linnaeus, 1771)Mico chrysoleucos (Wagner, 1842)Mico emiliae (Thomas, 1920)Mico humeralifer (É. Geoffroy Saint-Hilaire, 1812)Mico intermedius (Hershkovitz, 1977)Mico leucippe Thomas, 1922Mico manicorensis (van Roosmalen M.G.M., van

Roosmalen T., Mittermeier et Rylands, 2000)Mico marcai (Alperin, 1993)Mico mauesi (Mittermeier, Schwarz et Ayres, 1992)Mico melanurus (É. Geoffroy Saint-Hilaire, 1812)Mico nigriceps (Ferrari et Lopes, 1992)Mico rondoni Ferrari, Sena, Schneider et Silva, 2010Mico saterei (Silva et Noronha, 1998)

Oreonax Thomas, 1927

Oreonax flavicauda (Humboldt, 1812)

Pithecia Desmarest, 1804

Pithecia aequatorialis Hershkovitz, 1987Pithecia albicans Gray, 1860Pithecia (irrorata) hirsuta Spix, 1823Pithecia (irrorata) irrorata Gray, 1842Pithecia (irrorata) vanzolinii Hershkovitz, 1987Pithecia (monachus) milleri Allen, 1914Pithecia (monachus) monachus (É. Geoffroy

Saint-Hilaire, 1812)Pithecia (monachus) napensis Lönnberg, 1938Pithecia (pithecia) pithecia (Linnaeus, 1766)Pithecia (pithecia) chrysocephala I. Geoffroy

Saint-Hilaire, 1850Pithecia (pithecia) lotichiusi Mertens, 1925

Saguinus Hoffmannsegg, 1807

Saguinus bicolor (Spix, 1823)Saguinus fuscicollis fuscicollis (Spix, 1823)Saguinus fuscicollis avilapiresi Hershkovitz, 1966Saguinus fuscicollis cruzlimai Hershkovitz, 1966

191On the origin of allopatric primate species


less intelligent, sensitive, and sociable than primatesin general are - being forced to live as outcastswould equal a sure death. But, if it were healthymale individuals deviant from the socially selectedskin and/or hair color pattern that are discriminatedagainst merely for being slightly depilated orhaving its coat somewhat bleached somewhere,such young males pushed out of the group’s corearea by high-ranking males will ally for the sake ofsurvival alone. Their shared forced-upon marginal-ity could well drive them into looking beyond thehorizon and together leaving the pack in search ofa living space wherever it could be found. Once thatliving condition is fulfilled, they can start a newsocial group incorporating some females that theywere able to attract from other resident groups ontheir way out. This phenomenon is known to com-monly take place in hierarchically structuredprimate societies that are ruled and defended by do-minant (alpha)-males (e.g., Alouatta). It guaranteesa certain primate to reach optimal densities in un-

disturbed populations. Furthermore, it selects formales that are capable to lead and defend a socialgroup. All-male parties of slightly eumelanin and/orpheomelanin bleached, or somewhat depilatedmales that are pushed out of their parental group’sliving space and that follow the ‘trend to allopatry’,will range further and further away from the coreof a taxon’s distribution. If suitable habitat to settledown is not encountered, the animals eventuallywill weaken, suffer from diseases, starve to death,or get predated upon. Very rarely, they happen toventure into for that species marginal or unsuitablehabitat, being forced to adapt to an alien habitat ora different feeding niche. In extremely rare cases,such founder-groups or -colonies may diverge alongthis path into a different subspecies (whatever thatmay be) and eventually into a different species(whatever that may be) or ecospecies. This kind ofsympatric speciation may have taken place in suchcases as the cream-white, near-albinotic fair woollymonkey living year-round in the várzeas between


Saguinus fuscicollis mura Röhe, Silva Jr., Sampaio et Rylands, 2009

Saguinus fuscicollis primitivus Hershkovitz, 1977Saguinus fuscus (Lesson, 1840)Saguinus geoffroyi (Pucheran, 1845)Saguinus illigeri (Pucheran, 1845)Saguinus (imperator) imperator (Goeldi, (1907)Saguinus (imperator) subgrisescens (Lönnberg, 1940)Saguinus inustus (Schwarz, 1951)Saguinus labiatus labiatus (É. Geoffroy Saint-Hilaire,

1812)Saguinus labiatus rufiventer (Gray, 1843)Saguinus labiatus thomasi (Goeldi, 1907)Saguinus lagonotus (Jiménez de la Espada, 1870)Saguinus leucogenys (Gray, 1865)Saguinus leucopus (Günther, 1876) Saguinus martinsi martinsi (Thomas, 1912)Saguinus martinsi ochraceus Hershkovitz, 1966Saguinus midas (Linnaeus, 1758)Saguinus mystax mystax (Spix, 1823)Saguinus mystax pileatus (I. Geoffroy Saint-Hilaire et Deville, 1848)

Saguinus mystax pluto (Lönnberg, 1926)Saguinus niger (É. Geoffroy Saint-Hilaire, 1803)Saguinus nigrifrons (I. Geoffroy Saint-Hilaire, 1850)Saguinus nigricollis nigricollis (Spix, 1823)Saguinus nigricollis graellsi (Jiménez de la Espada, 1870)Saguinus nigricollis hernandezi Hershkovitz, 1982Saguinus oedipus (Linnaeus, 1758)Saguinus tripartitus (Milne-Edwards, 1878)Saguinus weddelli weddelli (Deville, 1849)

Saguinus weddelli crandalli Hershkovitz, 1966Saguinus weddelli melanoleucus (Miranda Ribeiro, 1912)

Saimiri Voigt, 1831

Saimiri boliviensis boliviensis (I. Geoffroy Saint-Hilaire et de Blainville, 1834)

Saimiri boliviensis peruviensis Hershkovitz, 1984Saimiri (cassiquiarensis) cassiquiarensis (Lesson, 1840)Saimiri (cassiquiarensis) albigena (von Pusch, 1942)Saimiri macrodon Elliot, 1907Saimiri oerstedii oerstedii (Reinhardt, 1872)Saimiri oerstedii citrinellus (Thomas, 1904)Saimiri (sciureus) sciureus (Linnaeus, 1758)Saimiri (sciureus) collinsi Osgood, 1916Saimiri ustus (I. Geoffroy Saint-Hilaire, 1843)Saimiri vanzoliniiAyres, 1985

Sapajus Kerr, 1792

Sapajus apella apella (Linnaeus, 1758)Sapajus apella margaritae (Hollister, 1914)Sapajus cay (Illiger, 1815)Sapajus flavius (Schreber, 1774)Sapajus libidinosus (Spix, 1823)Sapajus macrocephalus (Spix, 1823)Sapajus nigritus (Goldfuss, 1809)Sapajus robustus (Kuhl, 1820)Sapajus xanthosternos (Wied-Neuwied, 1826)


the lower Rio Javarí and the right bank of the upperRio Solimões. Or, the pheomelanin bleached, over-all orange-colored woolly monkey from the head-waters of the Rio Jutaí. Somewhat metachromicbleached founder-colonies of woollies driven by the‘trend to allopatry’ once must have diverged fromarchetypic agouti-colored or saturated eumelaninancestral La. poeppigiiwhile adapting to a differentecological niche that was new to woolly monkeys -in this case that of a frugivorous, canopy-dwelling,brachiating inhabitant of white-water inundatedfloodplain forest (várzea). During our systematicsurveys of primate distribution and diversity carriedout in the matrix terra firme rain forest that stretchesout behind the floodplain of some white-waterrivers (e.g., Javarí, Juruá, Purús, Madeira), we werenot able to detect any differences in phenotypebetween individual monkeys of a given taxon thatwe observed along the entire course (from sourceto headwaters) of each of these far-apart rivers.Contrary to what is the common presumptionamong primatologists, this would mean that interritorial monkeys such as pygmy marmosets orsaddle-back tamarins that occupy large distributionsdelineated by some of the largest tributaries of theAmazon, phenotypic characters of skin and pelagecoloration, and/or local hair growth or depilation,seem to have stabilized across their entire distribu-tions. In other words, within the distribution of agiven Amazonian monkey there does not existsomething like a gradient of slightly different phen-otypes, color forms, morphs, or races. These obser-vations from the larger field have led us attributingfull-species status to primate taxa like Cebuellapygmaea and C. niveiventris that are phenotypicallystable throughout their (sometimes huge) distribu-tions. Consequently, we here introduced the conceptof eco-species. This concept is firmly corroboratedby the here proposed theory on the origin of allo-patric primate species. An ecospecies may be bestdefined as: “A genetically isolated population orgroup of populations of a kind that does notundergo gene flow from adjacent populations of oneor more closely related kinds; and that shows astabilized phenotype across the entire range inwhich it occupies a well-defined ecological niche,which it defends against any outside competitor,even beyond generic level.” This eco-speciesconcept (ESC) avoids the often confusing arbitrarydistinction between species, subspecies, race,

morph, or form, for it adds sociobiological restric-tions to environmental (geographic, geomorpholo-gical and phytosociological) ones that use to act onspeciation and radiation in sociable territorialprimates. Defined as such, the ESC may apply alsoto similarly socially structured mammals like coatis,peccaries, and some canids. In accordance to thisdefinition, an enclave population of Callibellahumilis that lives year-round in igapó forest fringingthe Rio Atininga - genetically isolated from themain population that lives at least one hundred kmto the north in primary terra firme rain forest -should be assigned a different species name in itsown right. Or, in case the ranges of two Saddle-backTamarins of the S. fuscicollis Clade, hitherto beingtreated as subspecies, are only separated by a nar-row contact zone - where its aggressive territorialdefense effectively impedes any gene flow throughcross-breeding or hybridization - each populationshould be given valid species status. But, wherevera former distributional boundary between two suchecospecies has been disrupted, removed by a vi-cariance, or overtaken by the more aggressive oropportunistic of two ecospecies, the latter willexpand its distribution to the cost of the other. Then,a process of replacement is set in motion along asteadily moving frontline, which inevitably willlead to the extinction of the less aggressive, morevulnerable, or more sensitive of the two ecospecies.According to our doctrine of allopatric primatespeciation this will always be the ecospecies that isthe more advanced metachromic bleached one.Here we have mentioned at least four cases acrossthe Amazon where such process of replacement(through physical extermination) of one primate byanother is ongoing or about to be terminated: 1) thearchetypic agouti, gray, and dark red-brown coatedLake Baptista Titi Monkey Callicebus baptistaextending its range along the southbank of the RioAmazonas to the cost of the advanced bleached,yellow- and-gray coated Hoffmanns’s Titi MonkeyCallicebus hoffmannsi and the near-albinotic eco-species from the right bank of the Rio Mamurú; 2)the archetypic saturated eumelanin Midas TamarinSaguinus midas versus the progressively bleached,halfway to fully albinotic Pied Two-colored Tam-arin S. bicolor (including S. ochraceus) and Mar-tins’s Bare-face Tamarin S. martinsi, the latter threeecospecies being currently at the verge of extinctioncaused by a rapid southern expansion of midas


(Fig. 10); 3) the saturated eumelanin Weddell’sSaddle-back Tamarin S. (fuscicollis) weddelli ex-panding its range to the cost of the near-albinoticRondon’s Marmoset Mico rondoni, pushing thefrontline eastward into the interfluve delineated bythe Rios Guaporé and Jí- Paraná after havingtraversed the upper Rio Madeira in the recent past;4) Gray’s Saki Pithecia hirsuta (or P. mittermeieri)extending its range northwards to the cost of thenear-albinotic Buffy Saki taxon P. albicans. In casesof replacement it is always the more advancedmetachromic bleached to albinotic ecospecies thatis loosing the battle and eventually will go extinct.Though only documented by us in semi-captive andfree-ranging, but artificially composed multi- peciespopulations, during social conflicts it was invari-ably the more advanced metachromic bleached in-dividual monkey or group of monkeys that sufferedfrom dominant-male discriminatory behavior, beingbullied, repeatedly physically attacked or violentlyassaulted, and eventually forced out of the core(compound) area, where we provided additionalfood on feeding platforms constructed up in thecanopy. If not moving voluntarily to the periphery,so turning into outcasts, these monkeys could bebitten to death by the invariably less bleached, moreaggressive, conspecific leading male(s). In retro-spect, we recall that all neotonic, advanced meta-chromic bleached and near-albinotic individualmonkeys kept free-ranging in our respectivehalfway-houses by comparison were invariablymore soft-hearted, more sensitive, cooperative,adaptable, and (not surprisingly?) smarter than themale congeners by whom they were discriminated,pushed into the periphery, or banned from the corearea. Applying these observations to the wild, thetrend to allopatry boosted by seemingly non-ad-aptive social selection - leading males that discrim-inate upon phenotypically deviant mutant youngmales - in evolutionary sense could well turn out tobe truly adaptive. To cite Charles Darwin (1859):“In the long history of humankind -and animalkind,too- those who learned to collaborate and impro-vise most effectively have prevailed”. And: “It is notthe strongest of the species that survives, or themost intelligent that survives. It is the one that isthe most adaptable to change.”Applying the doctrine to the evolution of hom-

inins, in particular Homo sapiens, one may ponderand speculate about questions like the following:

“Why, and driven by what force about six millionyears ago somewhere in Tropical Africa an ape-likelineage of primates -our hominid ancestors- left therain-forest canopy and ventured into an arid open-savanna scrub landscape?”The common ancestors of the Great Apes and

the human line of hominins (Homo) were arborealprimates that had adopted brachiation (suspendedarm-over-arm-swinging underneath the twig/branchsubstrate) as a special locomotor pattern. Brachi-ation allows large-bodied arboreal primates toquickly move through the canopy and get to thefleshy fruits that are, as is the rule in any tropicalrain-forest environment, distributed in the far peri-phery (small- branch/twig micro-habitat) of canopy-and emergent-tree tops. Brachiation is a primarilyarboreal type of locomotion that evolved exclus-ively in some Neotropical Monkeys (i.e., spider,woolly and woolly spider monkeys) as well asin the Old-World Apes (i.e., gibbons, siamangs,bonobos/pygmy chimps, chimpanzees, orang-utans,and gorillas). It may never have evolved in Prosimi-ans, which are the more primitive among all theworld’s primates. It followed an independent evol-utionary path, a convergent or parallel evolution, ina physiognomically similar natural environment -the tropical forests of Southeast Asia, CentralAfrica, and South America (the larger AmazonBasin). A major intercontinental difference is thatsome monkeys in the Neotropics developed a pre-hensile tail as extra support in suspensory loco-motion, therefore called “semi-brachiation”,whereas apes during the evolutionary processtoward brachiation lost a functional tail. Brachiationwithout use of a fifth limb is called “true brachi-ation”. Most plausibly, our early ape-like hominidancestors that about 6 MYA descended from the can-opy of Central-African rain forest much resembledextant Spider Monkeys in their general locomotorpattern and diet. Brachiation is associated with adietary preference for ripe, pulpy, nutritious fruitsthat contain a single to few large seeds. The uprightposition of the trunk associated with an arboreallife-style involving much brachiation happened tobe a crucial pre-adaptation for later bipedal (two-legged) upright walking on the ground. It enabledour early ancestors to leave the trees in the sameway as gorillas once did, but different in thatthe Great Apes adopted ‘knuckle-walking’ as theprincipal locomotor pattern to walk on the ground.


Similarities between Spider Monkeys and Chim-panzees are striking as we consider that at leasttwenty-five million years of evolution on differentcontinents do separate these primates from one an-other. Cognitive features that both brachiatingprimates share are the mental capacity to visualize,pre-plan, and map out in time and space complexeconomic foraging routes to be followed that veryday, tomorrow, the day after tomorrow, and perhapseven over several days ahead. Moreover, theseprimates are able to lay out these foraging routesacross a landscape that is covered with dense trop-ical rain forest containing only few seasonal, widelydispersed food sources at any given time (VanRoosmalen, 1985a; 2013a). Consequently, bothspider monkeys and pygmy chimpanzees (bonobos)may well depict a marked period or stage in theevolution of our early ancestors that may havespecialized first in feeding upon ripe, juicy, lipid-and protein-rich, large-seeded fruits. Perhaps, thatfeeding niche may have been the condition thatpredestined our ancestors, both locomotorily andmentally, to leave the trees and become two-leggedground-dwelling foragers with an advanced use ofthe hands (e.g., dexterity, precision grips, tool-fashioning). And at the same time growing bigbabies and three to four times bigger brains (Lynch& Granger, 2008). In physical, anatomical, physiolo-gical, and mental respect, therefore, descendingfrom the trees and adapting locomotorily to bipedalwalking and running over the ground was not the‘near-impossible’ step that it may seem to be. If weput it in Darwinian evolutionary perspective, how-ever, to let it happen, until now an intraspecificsocial driver was missing that must have acted onthe undoubtedly territorially and hierarchicallyorganized communities of these ape-like ancestorswith the brain size of contemporary chimpanzees(400 cc). Forthcoming our thirty-five years livingin the Amazon and conducting long-term researchon captive, feral, as well as wild monkeys - thelatter mostly representing pristine populations thatwere never in any way disturbed by humans - wehere suggest the ‘trend to allopatry’ among slightlydepilated and/or metachromic bleached male indi-viduals (mutants) in primate populations being theprincipal force that has driven founder-colonies ofour early ancestors - for the mere sake of survival -out of their preferred habitat -canopy trees- into (tothem) new, with respect to natural enemies risky

and hostile landscapes. As sociable and intelligentmammals suffering from intraspecific populationpressures and discriminatory social constraints,outcast males must have taken on the challenge totraverse whatever barrier on their way out. So, theyventured into the arid, in many aspects hostilenatural environment of savanna scrub and openwoodlands. In a similar way as a small populationof Gracile Capuchins on the slopes of tepuís likePico da Neblina successfully adapted to a predom-inantly ground-dwelling life-style; the MountainGorilla successfully adapted to a fully terrestriallife-style in the cloud forests of the Virunga vulca-noes in Central Africa; the Western Chimpanzee ofthe ‘subspecies’ verus once adapted to a predom-inantly terrestrial life-style in an arid, for specialistfrugivores inappropriate or marginal natural envir-onment - the open savanna scrub of West Africa(Patterson et al., 2006); the near-albinotic Rio JavaríFair Woolly Monkey with an overall cream-whitecolored coat, and the Rio Jutaí Orange WoollyMonkey with an overall orange colored coat, ad-apted to várzea floodplain forest along the upperAmazon and lower Javarí Rivers, and the upperJutaí River, respectively; the Peruvian Yellow-tailedWoolly Monkey in complete isolation adapted tohigh-altitude cloud forest in the NE PeruvianAndes; the advanced pheomelanin to near-albinoticBald-headed Uakaris adapted to seasonally inund-ated white-water floodplain forest (várzea) alongthe Amazon River and some of its southern tributa-ries that drain the southeastern flanks of the Andes;among others. Looking at the distribution ofCentral-American spider monkeys of the Atelesgeoffroyi Clade, we could speculate about animaginary evolutionary path that could have beenfollowed by an advanced metachromic bleached,near-albinotic founder-colony of the Central Amer-ican Yucatán Spider Monkey Ateles (geoffroyi)yucatanensis from the tropical forest of YucatánPeninsula in SE Mexico. By the ‘trend to allopatry’forced out of the canopy of a semi-deciduous rainforest somewhere on the Yucatán Peninsula - thetaxon’s current deadend distribution - some founder-colony may venture into the savanna and desertscrub of SE Mexico and from there further into theMidwest of the US. To survive in such (for spidermonkeys) alien landscape it would quickly have toloose a functional tail and adopt bipedal uprightwalking as its main locomotor pattern. It is tempting


to imagine a similar scenario for progressivelymetachromic bleached, depilated, red- or white-skinned near-albinotic early hominids 6 MYAradiating away from their archetypic, saturated-eumelanin congeners they had in common withancestral chimpanzees. Driven by the trend to allo-patry, in a similar way founder-colonies may haveleft the semi-deciduous rain forests of C Africa andventured first into the savannas, plains and desertscrub of N Africa and, thereafter, into the tem-perate-clime dominated landscape of S+C Europe,the Middle East and SE Asia. Recent evidence frommolecular biology suggests that it took several hun-dreds of thousands years for our early ancestors toevolve in two distinct animals: the open savannaexplorers leading toward proto-humans, and thoseremaining arboreal resulting in chimpanzees (Pat-terson et al., 2006). In accordance with recentphylogenetic research, the modern ChimpanzeePan troglodytes diverged from the proto- or ar-chetypic, saturated eumelanin, overall blackish-brown colored Bonobo (Pygmy Chimpanzee) Panpaniscus. The common Chimpanzee is an oppor-tunist having an omnivorous diet, whereas theBonobo holds a predominantly specialist frugivor-ous diet. In comparison to common Chimpanzees,Bonobos are egalitarian, peaceable, non-violentcreatures that live in loosely organized, matriarchalsocial groups in which the males may defend theirterritories, but rather adopt a “Make Love No War”philosophy of life. Bonobos have never been repor-ted to involve in raids on neighboring group males,whereas common chimpanzees have been seenperforming a kind of troop-hunting culture in whichbeta-males led by one alpha-male sometimes doattack neighboring males or small mixed parties,killing and eating some of them. Bonobos live inthe dense tropical rain forests of Central Congo.Their distribution is thought to represent the cradleof chimpanzee evolution or the center of chimpan-zee (genus Pan) dispersion. Applying Hershkovitz’hypothesis of metachromism, Chimpanzees maywell have derived from (proto)-Bonobos. Nowadays,the two species are allopatric. The trend to allopatrymay have forced ancestral chimpanzees to swimacross or circumvent the Congo River that does actas a geographic barrier in present-day distributions.The farther in any but southern direction fromthe center of Pan troglodytes dispersion, locatedjust north of the Congo River, the more arid the

landscape becomes, the more often chimps dodescend from the trees and ‘knuckle-walk’ on theground, and the more chimps have adapted to whatbonobos would consider inappropriate or marginalhabitat - unsuitable to highly specialized maturefruit-eaters that bonobos are. At the same time, wesee chimpanzees becoming more pheomelaninto euchromic bleached, their skin getting lightercolored (less pigmented), their coat thinner andlocally depilated or almost hairless, and elderlyindividuals becoming gray with age.Another question to ponder about with the doc-

trine in mind: “Why, and driven by what force someof our Homo ancestors between 100,000 and50,000 years ago left the origin and center of hom-inid dispersion -Central and North Africa beingconsidered the cradle of human evolution- to ven-ture into the clime- and habitat-wise new, but un-suitable or (at least) marginal landscape of Europe,the Middle East and Asia?”After Homo erectus having grown much bigger

brains on the plains, some millions of years later thetrend to allopatry may have been again the principaldriving force for some founder-colonies of Homosapiens to move ‘Out of Africa’. The pioneers thatventured into new landscapes to the north could doso only by occupying an ecological feeding nichethat was new to former small-game hunter-fisher-gatherers, that of big-game hunter-gatherers. Hereby,the invention to first carrying along fire, soon fol-lowed by the skill to kindle it, was essential in theadaptation process to a new feeding niche, as their(our) digestive system is not apt to decompose rawmeat. It has to be cooked or barbecued. Apparently,in very low densities - recent estimates place thepopulation of Europe 30,000 years ago at about5,000 people - these humans following herds ofprehistoric megafauna (e.g., mammoth) and drivingthem to extinction in the Holocene, have spreadrapidly across the whole of Europe and SoutheastAsia, one route taking them as far as Australia andTasmania, the other to the far northeastern cornerof Siberia. From these places they eventually couldreach and inhabit some Pacific Islands, and mostamazingly also the continent of South America, firstabout 30–40,000 years ago by bordering the Antarc-tic during one of the glacials, and a second time,about 15,000 years ago, via Beringia and NorthAmerica (Van Roosmalen, 2013c). We could askourselves if these all could have been advanced



metachromic bleached, euchromic to albinoticfounder-colonies or colonizing parties that werepushed out from dead-end distributions in Africaand Asia following the male-territorial primate-borntrend to allopatry?

ACKNOWLEDGMENTS

We would like to dedicate the theory laid downin this paper to the memory of Alfred Russel Wal-lace (1823–1913), who inspired us to follow hisearly footsteps into the Amazonian realm. We arealso immensely indebted to the late Philip Her-shkovitz of the Chicago Field Museum of NaturalHistory, a true pioneer in the fields of primatologyand metachromism. Many thanks are expressed toRussell A. Mittermeier of Conservation Interna-tional, who supported our fieldwork in the Amazonever since we started our careers in the pristineancient forests around the Voltzberg Mt in Suri-name, also financially through the Margot MarshBiodiversity Foundation in the period 1998–2002.We would like to thank Noel Rowe and DougBrandon-Jones for reviewing an early draft of thepaper. Special thanks go to Stephen Nash of Con-servation International who kindly provided us withhis very fine standardized color illustrations of allknown Neotropical monkey taxa. It enabled us tobuild our theory on the equally colorful speciation,radiation and diversification in New World prim-ates.

REFERENCES

Allen J.A., 1914. New South American monkeys. Bul-letin of the American Museum of Natural History,Vol. 23, Art. 18: 647–655.

Boubli J.P., Da Silva M.N. F., Amado M.V., HrbekT., Pontual F.B. & Farias I.P., 2008. A taxonomicreassessment of Cacajao melanocephalus, Humboldt(1811), with the description of two new species.International Journal of Primatology, 29: 723–741.

Chiou K.L., Pozzi L., Lynch Alfaro J.W. & Di FioreA., 2011. Pleistocene diversification of living squirrelmonkeys (Saimiri spp.) inferred from completemitochondrial genome sequences. Molecular Phylo-genetics and Evolution, 59: 736–745.

Da Cruz Lima E., 1945. Mammals of Amazônia. 1. Gen-eral introduction and primates. Belém do Pará: Con-tribuições do Museu Paraense Emílio Goeldi deHistória Natural e Etnografia.

Darwin C., 1859. On the Origin of Species by Means ofNatural Selection, or the Preservation of FavouredRaces in the Struggle for Life. J. Murray, London.(Reprint edition: Avenel Books, New York, 1979).

Grabert H., 1991. Der Amazonas: Geschichte undProbleme eines Stromgebietes zwischen Pazifik undAtlantik. Springer-Verlag, Heidelberg, 235 pp.

Groves C.P., 2001a. Primate taxonomy. Washington, DC:Smithsonian Institution Press.

Groves C.P., 2001b. Why taxonomic stability is a badidea, or why are there so few species of primates (orare there?). Evolutionary Anthropology, 10: 192–198.

Groves C.P., 2004. The what, why and how of primatetaxonomy. International Journal of Primatology, 25:1105–1126.

Groves C.P., 2005. Order Primates. In: Mammal speciesof the world: a taxonomic and geographic reference,(3rd edition) 1, Wilson D.E. & Reeder D.M. (Eds).,Johns Hopkins University Press, MD, 111–184.

Hershkovitz P., 1968. Metachromism or the principleof evolutionary change in mammalian tegumentarycolors. Evolution, 22: 556–575.

Hershkovitz P., 1977. Living New World Monkeys(Platyrrhini) with an Introduction to Primates. Vol. I.University of Chicago Press, Chicago, 1117 pp.

Hershkovitz P. 1984. Taxonomy of squirrel monkeys,genus Saimiri (Cebidae, Platyrrhini): a preliminaryreport with description of a hitherto unnamed form.American Journal of Primatology, 7: 155–210.

Hershkovitz P., 1985. A preliminary taxonomic reviewof the South American bearded saki monkeys genusChiropotes (Cebidae, Platyrrhini), with the descrip-tion of a new subspecies. Fieldiana Zoology, 27: 1–46.

Hershkovitz P., 1987a. Uacaries, New World monkeys ofthe genus Cacajao (Cebidae, Platyrrhini): a prelim-inary taxonomic review with the description of anew subspecies. American Journal of Primatology, 12:1–53.

Hershkovitz P., 1987b. The taxonomy of South Americansakis, genus Pithecia (Cebidae, Platyrrhini): a pre-liminary report and critical review with the descrip-tion of a new species and new subspecies. AmericanJournal of Primatology, 12: 387–468.

Hershkovitz P., 1988. Origin, speciation, dispersal ofSouth American titi monkeys, genus Callicebus(family Cebidae, Platyrrhini). Proceedings of theAcademy of Natural Sciences of Philadelphia, 140:240–272.

Hershkovitz P., 1990. Titis, New World monkeys of thegenus Callicebus (Cebidae, Platyrrhini): a prelim-inary taxonomic review. Fieldiana Zoology, 55: 1–109.

Jacobs S.C., Larson A. & Cheverud J.M., 1995. Phylo-genetic relationships of coat color among tamar-

198

ins (genus Saguinus). Systematic Biology, 44: 515–532.

Kellogg R. & Goldman E.A., 1944. Review of the spidermonkeys. Proceedings of the United States NationalMuseum, 96: 1–45.

Lynch G. & Granger R., 2008. Big brain: the origins andfuture of human intelligence. New York, PalgraveMacmillan, 259 pp.

Mittermeier R.A., Rylands A.B., & Wilson D.E. (Eds.),2013. Handbook of the Mammals of the World. Vol.3. Primates. Lynx Edicions, Barcelona, 951 pp.

Patterson N., Richter D.J., Gnerre S., Lander E.S.& Reich D., 2006. Genetic evidence for complexspeciation of humans and chimpanzees. Nature,441: 1103–1108.

Silva Jr. J.S., 2001. Especiação nos macacos-prego ecaiararas, gênero Cebus Erxleben, 1777 (Primates,Cebidae). Doctoral thesis, Universidade Federal doRio de Janeiro, Brazil.

Silva Jr. J.S., 2002. Sistemática dos macacos-prego ecaiararas, gênero Cebus Erxleben, 1777 (Primates,Cebidae). In: Livro de Resumos, X CongressoBrasileiro de Primatologia: Amazônia - A ÚltimaFronteira, 10–15 November 2002, 35 (abstract).Sociedade Brasileira de Primatologia. Belém,Pará.

Van Roosmalen M.G.M., 1985a. Habitat preferences,diet, feeding strategy and social organization of theblack spider monkey (Ateles paniscus paniscusLinnaeus 1758) in Suriname. Acta Amazonica, 15(Supplement), Sep/Dec 1985.

Van Roosmalen M.G.M., 1985b. Fruits of the GuiananFlora. Institute of Systematic Botany, Utrecht, theNetherlands, 483 pp.

Van Roosmalen M.G.M. & Van Roosmalen T., 2003. Thedescription of a new marmoset genus, Callibella(Callitrichinae, Primates), including its molecularphylogenetic status. Neotropical Primates, 11: 1–10.

Van Roosmalen M.G.M., Van Roosmalen T., MittermeierR.A. & Fonseca G.A.B.Da., 1998. A new and dis-

tinctive species of marmoset (Callitrichidae, Prim-ates) from the lower Rio Aripuanã, state of Amazonas,central Brazilian Amazonia. Goeldiana Zoologia, 22:1–27.

Van Roosmalen M.G.M., Van Roosmalen T., Mitter-meier R.A. & Rylands A.B., 2000. Two new speciesof marmoset, genus Callithrix Erxleben, 1777 (Cal-litrichidae, Primates), from the Tapajós/Madeirainterfluvium, south central Amazonia, Brazil.Neotropical Primates, 8: 1–18.

Van Roosmalen M.G.M., Van Roosmalen T. & Mitter-meier R.A., 2002. A taxonomic review of the titimonkeys, genus Callicebus Thomas, 1903, with thedescription of two new species, Callicebus bernhardiand Callicebus stephennashi. Neotropical Primates,10 (Supplement): 1–52.

Van Roosmalen M.G.M., 2013a. Barefoot through theAmazon - On the Path of Evolution. http://www.amazon.com Print version & Kindle e-book version,500 pp.

Van Roosmalen M.G.M., 2013b. Distributions andPhylogeography of Neotropical Primates - A PictorialGuide to All Known New-World Monkeys.http://www.amazon.com Print version & Kindlee-book version: 1–71.

Van Roosmalen M.G.M., 2013c. A Shaman’s Apprentice- Bio-History in South America and the Amazon.http://www.amazon.com Print version & Kindlee-book version: 1–161.

Van Roosmalen M.G.M., 2015. Live from the Amazon.http://www.amazon.com - Print version & Kindlee-book version, 442 pp.

Wallace A.R., 1852. On the monkeys of the Amazon.Proceedings of the Zoological Society of London,20: 107–110.

Wallace A.R., 1876. The geographic distribution ofanimals, with a study of the relations of living andextinct faunas as elucidating the past changes of theearth’s surface. Boston University. New York: Harper& Brothers Publishers, 668 pp.


on the origin of allopatric primate species · structure in most primate societies, in particular...

Documents