BULLETIN OF THE ALLYN MUSEUM
3621 Bay Shore Rd.
Sarasota, Florida 34234
Florida Museum of Natural History
University of Florida
Gainesville, Florida 32611
Number 143 19 September 2003
AFROTROPICAL SKIPPERS (LEPIDOPTERA:
HESPERIOIDEA) AND THE EMERGENCE OF
THE COMBINED REFUGIUM THEORY
John F Douglass
3347 Airport Hwy., Toledo, Ohio 43609
Lee D. Miller
Allyn Museum of Entomology, Florida Museum of Natural History,
3621 Bay Shore Rd., Sarasota, Florida 34234
ABSTRACT. Eighty-nine (89) species of skippers are reported from a collection made in
NE Zaire (esp. the Ituri Forest) during June-July 1991. Eight of these species are reported for
the first time from Zaire (now DR Congo); in five cases these represent sizable eastward
range extensions. A discussion of various faunistic, biogeographic, and systematic relation-
ships is provided, particularly with reference to the past fragmentation of Africa's equatorial
forest (glacial refugia).
KEY WORDS: Afrotropical biogeography, Hesperiidae, Zaire, Democratic Republic of the
Congo, Ituri Forest, refuge theory, forest history, paleoecology, faunistics, vicariance,
The Ituri Forest has intrigued outsiders since H. M. Stanley first made his way through this
vast tropical wilderness ('Darkest Africa') in 1887. Inhabited by the world's most diminutive
people (Mbuti Pygmies) and an unusual endemic fauna, it has eluded intensive biological
study until recently: such striking endemic animals as the okapi or 'rainforest giraffe' (Okapia
johnstoni (Sclater)) (Figure 1) and Congo peacock (Afropavo congensis Chapin) were
unknown to Western science until 1900 and 1936, respectively. Despite the compelling
interest the Ituri Forest holds in these respects, little intensive work has been undertaken on
its lepidopteran fauna (e.g., Talbot 1921, 1922, Barns 1922, Ducarme 1993, 1999).
In 1991, with the support of the Allyn Museum (Florida Museum of Natural History), JFD
organized an expedition to Zaire and spent 40 days collecting Lepidoptera at Epulu, in the
heart of the Ituri Forest, aided by Pygmy and Bantu assistants.
Butterflies and skippers were collected during the summer of 1991 at three disjunct local-
ities in Haut-Zaire: Nyankunde (WSW of Bunia) (19 June), Mambasa (20 June), and the
general vicinity of Epulu (incl. Lenda and Afarama research camps) (20 June-29 July). The
latter two of the three localities lie within the Ituri Forest; Nyankunde represents a different
(i.e., elevated, open) environment (site descriptions below).
In addition to aerial nets, bait traps were used: nylon lines and pulleys placed in trees
allowed these to be lowered and checked frequently. Acraea skippers (genus Fresna) were
generally collected while perched on bird droppings on the upper surfaces of mongongo
(phrynium) leaves (Marantaceae), 1-2 m above the forest floor (and on moist sand in one
Determinations included here are as made by LDM and verified by JFD and others.
Nomenclature and the sequence in which taxa are arranged follows Ackery et al. (1995).
Specimens are deposited at the Allyn Museum of Entomology (AME), Sarasota, Florida, USA.
Throughout this paper, we use Zaire as the name of the country under consideration (in
lieu of the recently-restored Democratic Republic of the Congo [DR Congo]) because: 1) Zaire
was the country's name when the present study was carried out; 2) the cardinal reference we
cite (Ackery et al. 1995) in developing taxonomic and distributional contexts for information
in the present paper uses this name.
THE ITURI FOREST
Location The Ituri Forest (0o-3 N, 27o-30 E; Figure 1) is a 65,000-kmn expanse of closed
evergreen and semi-evergreen forest centered on the upper watershed of the Ituri River, its
elevation ranges from 600 to 1200 m. The village of Epulu (1025' N, 28035' E, elev. 750 m) in
the central Ituri Forest lies on the single E-W-running dirt road traversing the forest (variously
called the trans-Ituri road, Kisangani-to-Bunia road, Nia Nia-Mambasa road, and trans-Africa
It is regrettable that the scientific study of Africa's tropical forests lags behind that of the
Neotropical realm, both in terms of funding and the number of Western and indigenous
researchers who are involved (pers. obs. 1991, Weber and Vedder 2001). The village of Epulu
has a rich legacy of traditions in both anthropology and mammalogy (e.g., Turnbull 1965, J. A.
Hart and T. B. Hart 1986-2001, Nat. Geogr. Soc. 1992, 1996). The okapi capture station (Station
d'Epulu) represents a superb base for research, but as yet it is only rarely used by
entomological investigators. Year-round collecting and rearing work carried out there would
be well rewarded; lepidopterological studies which suggest themselves include comparative
surveys of mixed and mbau forest, vast tracts of which are easily accessible from Epulu. A
promising recent development in the area is the establishment in 1992 of a huge sanctuary
surrounding Epulu, the 13,726-klm Reserve de Faune A Okapis (RFO) (Okapi Wildlife
Reserve) (Hart and Hall 1996, Stephenson and Newby 1997, Curran and Tshombe 2001).
Seasonality Annual rainfall in the Ituri region is 1700-1800 mm, with a dry season of less
than 40 days (Bultot 1971). Mean annual rainfall at Epulu during the period 1987-1993 was
1700 mm (range, 1307-2084 mm) (T. B. Hart 1995).
The central Ituri experiences a single, well-defined dry season ('long dry'), generally
extending from mid-December to the end of February (T. B. Hart 1985, Hart and Hart 1989,
J. A. Hart 2001). The driest month is January (Bultot 1971). Total rainfall during the three-
month period January-March varies widely (76-359 mm) (T. B. Hart 1995). In addition, the
Epulu region experiences a variable 'short dry' season from ca. mid-June to mid-August (pers.
obs. 1991, S. Shurter pers. conun.).
The rainy season begins in March; the wettest periods are March-May and September-
November (Hart and Hart 1989). New leaf-growth in the understory occurs in every month
but peaks in February-April at the onset of rains (Hart and Hart 1988).
During the period of collecting work in 1991 (i.e., JFD based at Epulu, 20 June-29 July),
excellent conditions for butterfly-collecting prevailed. Most days were sunny. Mornings
generally began overcast (misty, gray), clearing to sunshine (often with fair-weather cumulus
clouds) by mid- to late morning. Rain (drizzle, thunderstorms) occurred at the site on the
average only every three days, and when heavy downpours did occur they tended to happen
in the pre-dawn hours of darkness.
Because this is the first in what we hope will be a series of reports (and continuing field
work at the sites described), Ituri Forest habitats are described here in some detail. Further
useful descriptions of the structure and composition of Central Africa's forests may be found
in Chapin (1932), Moreau (1966), White (1983), Sayer et al. (1992), Richards (1996), Morley
(2000), and White (2001).
Prior to the immigration of agriculturalists (ca. 1,000-4,000 ybp), the Ituri was comprised of
closed primary forest, interrupted only sparingly by small seral communities at treefalls and
lightning strikes (Hart and Hart 1986, Wilkie and Finn 1990).
At the time of the present study (1991), human disturbance of the forest remained minimal.
There are few permanent settlements, and significant stands of mature forest remain intact.
Based on satellite imagery, D. S. Wilkie (unpubl. 1989 report cited in J. A. Hart 2001 and
Curran and Tshombe 2001) has determined that ca. 90% of the Ituri Forest (RFO area)
remains intact, i.e., covered with closed-canopy forest.
Present-day habitats in the central Ituri region consist of the distinct types below. In the
present study, skippers were collected in all of these environments.
I. Primary forests: A. Mature upland forests
(1, monodominant mbau; 2, mixed)
B. Riparian/Swamp forests
II. Disturbed areas: A. Open areas
B. Secondary forests
I. Primary forests
A. Mature upland forests The Ituri Forest is characterized by two distinct rainforest
types, mbau (monodominant) forest ("single-dominant moist evergreen" forest of White 1983)
and mixed forest ("mixed moist semi-evergreen" forest of White 1983); the central Ituri region
is notable for the close co-occurrence of these (T. B. Hart 1985, 1995, Hart et al. 1989). Both
vegetation types are mature upland forest formations dominated by Caesalpiniaceae. The
habitat descriptions below are based on the pioneering synecological work of T. B. Hart
(1985, 1995, 2001), Hart et al. (1989), and Makana et al. (1998). Further details of habitat
description are provided by J. A. Hart (2001; eight habitat types are distinguished in the RFO).
1) Monodominant mbau forest This evergreen forest type is particularly distinctive.
It occurs in large, continuous stands. It is floristically less rich than many other tropical
evergreen forests, and than adjoining mixed forests in the Ituri. A single tree species, mbau
(Gilbertiodendron dewevrei (De Wild.) Leonard (Caesalpiniaceae)), predominates in all
size classes (sapling, pole, tree, large tree); in local studies (T. B. Hart 1985, 1995, Hart et
al. 1989) it was found to comprise more than 90% of canopy trees and to account for 88% of
total basal area
With the exception of the near-absence of G. dewevrei in mixed forest, mbau and mixed
forests share the same tree species (Hart et al. 1989). Interestingly, however, the transition
from mbau to adjacent mixed forest is floristically abrupt, typically with no zone of
intergradation between the communities (Hart et al. 1989).
In both mbau and mixed forests, the canopy attains heights of 30-40 m (Hart et al. 1989).
In mbau forest, however, the canopy is more homogeneous (denser, deeper, less broken),
the subcanopy is more open, and the understory is more uniformly shady than in mixed
forest (T. B. Hart 1985, 1995, Hart et al. 1989). The understory is rather open in both forest
types and allows easy foot passage and visibility to 10 m or more (T. B. Hart 1985, pers. obs.
Mbau forests are steadily encroaching on adjacent mixed forests, and have apparently
undergone a marked expansion in the central Ituri region over the last 4,000 yrs. (Hart et al.
1989, Hart et al. 1996). G. dewevrei has a low rate of dispersal (White 1983, Hart et al. 1989),
and the establishment of monodominant stands (mbau forest) is apparently dependent upon
the long-term absence of large-scale disturbances (e.g., fire) or climatic change (Hart et al.
2) Mixed forest This semi-evergreen forest type is characteristic of large expanses of the
central Ituri region. It is considerably more diverse (more species-rich) than mbau forest (T.
B. Hart 1985, Hart et al. 1989), and appears to have undergone large-scale disturbance more
recently. The most abundant tree species, Julbernardia seretii (De Wild.) Troupin
(Caesalpiniaceae), comprises up to 40% of canopy trees and accounts for 32% of total basal
area (T. B. Hart 1985, 1995, Hart et al. 1989). Species with the greatest mean basal area after
J. seretii are Cynometra alexandri C. H. Wright (Caesalpiniaceae), Cleistanthus
michelsonii J. Ieonard (Euphorbiaceae), and Klainedoxa gabonensis Pierre (Irvingiaceae)
(Hart et al. 1989).
The canopy in mixed forest has a more broken aspect than that in mbau forest and is more
heterogeneous. Crowns of the dominant and emergent species tend to be broad and shallow
and are often not contiguous (T. B. Hart 1985, Hart et al. 1989). Emergent individuals (to
40-50 m) and treefall gaps are more common in mixed than in mbau forest (Hart et al. 1989),
and a well-developed subcanopy layer (10-25 m) is dense in irregular patches (T. B. Hart
1985). Lianas are much more abundant in mixed than in mbau forest (Makana et al. 1998).
Scaphopetalum dewewrei De Wild. and Th. Dur. (Sterculiaceae) treelets dominate in the
understory, and dense stands of herbaceous Marantaceae, Commelinaceae, and
Zingiberaceae occur in treefall gaps, blowdowns, and swampy areas (Makana et al. 1998, J. A.
B. Riparian/Swamp forests These formations (i.e., swamp forest and narrow bands of
periodically-inundated forest along streams) comprise a small percentage of the total area in
both primary forest types (Hart and Hart 1986, T. B. Hart 1995, Hart et al. 1996, Stephenson
and Newby 1997). Swampy sites may include dense stands of herbaceous Marantaceae;
swamp forest vegetation in the central Ituri is further described by Hart and Hart (1988, 1989).
II. Disturbed areas
A. Open areas These consist of roadsides, village settlements, and agricultural clearings
(i.e., active gardens, village fields).
B. Secondary forests The only large areas of secondary forest in the Ituri are of agricul-
tural origin (Hart and Hart 1986). Cultivation of crops in the forest interior apparently began
when agriculturalists (i.e., Bantu- and Sudanic-speaking shifting cultivators) advanced into
the forest ca. 1,000-4,000 ybp (Hart and Hart 1986, Wilkie and Finn 1990, Hart et al. 1996).
Introduction of agriculture (i.e., shifting (swidden) horticulture) into the forest permitted
semi-nomadic hunter-gatherers (i.e., Mbuti and Efe Pygmies) to occupy what had been closed
forest on a permanent basis (Hart and Hart 1986); long-standing trade relations between
Pygmies and forest agriculturalists continue to the present day (Hart and Hart 1986).
More recently, roadside recurrent cultivation has been practiced, and the past 50 years
have seen an expansion of slash-and-burn agriculture along roads in the area (Wilkie and Finn
1990, Stephenson and Newby 1997); today, most young secondary forest is concentrated
along roads traversing the forest (Hart and Hart 1986).
Patterns of post-cultivation secondary forest succession in the Ituri Forest (incl.
clearing/cultivation/forest fallow/reclearing cycles) have been described by T. B. Hart (1985),
Hart and Hart (1986), and Wilkie and Finn (1988, 1990). It is characterized by these distinct
1) active field (shamba) (2-yr. cultivation cycle);
2) years 1-3 post-cultivation, abandoned gardens, "farm-bush" secondary communities:
open sere colonized by herbaceous and shrub species (incl. feral cultigens and
pioneer plant species);
3) years 4-20, young regrowth forest: Musanga cecropioides (Moraceae) and other
emergent tree species become dominant, canopy height 10-20 m, dense understory
and middle layers (large, contiguous zones of M. cecropioides-dominated secondary
forest now border roads in the region);
4) years 21-50, old seral forest: replacement of senescing trees by slow-growing, late-
seral species which persist in canopy until sere is no longer distinguishable from
climax forest (T. B. Hart 1985);
5) years 50-100, regenerated climax forest.
Introduction of agriculture into the Ituri has essentially extended the forest-savanna eco-
tone into the forest interior (Hart and Hart 1986). Relative to Epulu (at the center of the Ituri
Forest), the closest forest/savanna border is more than 150 km distant to the NE (Makana et
al. 1998, T. B. Hart 2001). In this connection it is interesting to note the capture in the present
study of several open-country and transition-zone skipper species (Kielland 1990, Larsen
1991) in the vicinity of Epulu: Eretis lugens (Rogenhofer), Sarangesa maculata (Mabille),
Fresna netopha (Hewitson), Fresna nyassae (Hewitson), and Borbo holtzi (Pltz). Also,
Carcasson (1964) indicated that the presence within the forest of such open-country forms as
Coeliades forestan (Stoll), Tagiades flesus (Fabricius), Borbo borbonica (Boisduval), and
Gegenes niso (Linnaeus) tends to reflect these adaptable species' ability to colonize man-
Collecting was carried out at seven localities in Haut-Zaire: each of these has been assigned
a three-letter code name, indicated below. Lenda (LDA) and Afarama (AFM) are research
reserves administered by WCS-IZCN (Wildlife Conservation Society [formerly the New York
Zoological Society] and the Institut Zairois pour la Conservation de la Nature [now the
Institute Congolais pour la Conservation de la Nature (ICCN)]). Skippers were collected at all
of these localities except sites MMB and LDA; the great majority of the specimens are from
the general vicinity of Epulu (EVC).
1) NYK: village of Nyankunde, 45 km WSW of Bunia, esp. grounds of the Centre Medical
2) MMB: village of Mambasa (disturbed ground).
3) EVC: vicinity of Epulu, elev. ca. 750 m; mosaic of habitats (mbau forest, mixed forest,
secondary forest, active gardens/cleared agricultural fields) (specimens were generally
collected within a one-hour walk of Epulu; few or none were collected in the village itself).
4) TLD: trail (footpath) to Lenda, commencing 6 km E of Epulu on the E-W Bunia-to-
Kisangani road, running 8 km S to site LDA; northerly 2.5-3 km of trail traverses second-
growth forest with small settlements and plantations, remainder of trail traverses mosaic of
mbau and mixed forest.
5) LDA: Lenda Study Area (LSA), 15-km2 research camp established by Drs. J. A. Hart and
T. B. Hart at S terminus of TLD; 12 km SE of Epulu, elev. ca. 700 m; mosaic of monodominant
mbau forest, including stands of up to 10 km' in which mbau comprises more than 70% of the
large trees (Hart et al. 1996).
6) TAF: trail (footpath) to Afarama (AFM), commencing 10 km W of Epulu on the E-W
Bunia-to-Kisangani road, running 19 km N to site AFM; mixed forest.
7) AFM: Edoro Study Area (ESA) ("Afarama"), 45-km' research camp established by Drs.
J. A. Hart and T. B. Hart at N terminus of TAF, at confluence of Edoro and Afarama rivers;
22 km NNW of Epulu, elev. ca. 850 m; broken-canopy mixed forest (habitats at ESA are
further described by Hart and Hart 1988, 1989).
RESULTS AND DISCUSSION
FAUNISTIC AND BIOGEOGRAPHIC RELATIONSHIPS
I. Composition of fauna
A. Numbers of individuals collected Lindsey and Miller (1965) commented on the
paucity of Hesperiidae (i.e., 1-3% of all lepidopteran specimens) in various collections made
in the forests of West and Central Africa, and suggested that collections in which skippers
comprise 7-10% probably provide a truer picture of the fauna in these areas. Skipper
specimens (n = 501) comprise 13% of the total collection of Rhopalocera (n = ca. 3,933)
resulting from the collecting work in 1991. (It should be noted that a concerted effort was
made by JFD and assistants to collect Hesperioidea, along with other relatively neglected
groups (e.g., Satyridae, Lycaenidae).) Within the Hesperiidae, specimens collected in 1991
were distributed as follow: Coeliadinae, n = 25 (5% of total); Pyrginae, 96 (19%); Hesperiinae,
B. Species diversity Skipper species (n = 89) comprise 21% of the ca. 431 species of
Rhopalocera collected in 1991. An analysis of species diversity by subfamily within the
Hesperiidae appears in Table 1. Eighteen percent (18%) of the known Afrotropical skipper
fauna is represented in the 1991 collection; the proportions in which species in each of the
three hesperiid subfamilies were collected closely matches the proportions these groups
comprise in the Afrotropical fauna overall. Further collecting in the Epulu area, especially
during other parts of the year, will no doubt augment the list of species now known from the
site (cf. Ducarme 1999).
Table 1. Hesperiidae resulting from the collecting work in Haut-Zaire ("1991" column)
compared to total numbers of species listed for the Afrotropical Region by Ackery et al.
(1995) ("Afrotrop." column). In each case, the number of species is followed in parentheses
by the same expressed as a percent of total.
1991 AfrotroD. 1991 as % of Afrotrop.
Coeliadinae: 5 (6) 18 (3) 28
Pyrginae: 21 (23) 145 (29) 14
Hesperiinae: 63 (71) 340 (68) 19
TOTAL: 89 (100) 503 (100) 18
C. Sex ratios Most species of skippers collected in this study show a strong apparent
bias toward (6 in terms of numbers of individuals captured. Interestingly, however, in two
species, the observed sex ratio did not differ significantly from 50:50 (i.e.: Tagiades flesus,
X2 = 0.04, P >> 0.05; Semalea pulvina, x2 = 0.64, P >> 0.05).
Two taxa known to be endemic to the Ituri region are represented among the species
collected: Celaenorrhinus b. beni Bethune-Baker and C. boadicea howarthi Berger
(howarthi is also known from Uganda).
III. Geographic range extensions
Some of the Ituri specimens appear to represent considerable range extensions of the
species involved. In some cases, indicated with asterisks (*), these appear to be the first
known records of particular species from Zaire.
A. Eastward extensions (disjuncts?) The most striking biogeographic patterns
evident within the Ituri Forest collection are sizable eastward extensions of the known ranges
of some species, hitherto known only from western Africa. Whether these represent truly
disjunct populations in the Ituri region or are an artifact (i.e., poor sampling in intervening
areas, Louette 1984, Mayr and O'Hara 1986, Hamilton 1992) is not clear. The Ituri species
*Coeliades bixana Evans (also collected in E Zaire by Ducarme 1999);
*Xanthodisca rega (Mabille);
*Meza mabillei (Holland);
*Caenides maracanda (Hewitson);
*Melphina tarace (Mabille) (also collected in E Zaire by Ducarme 1999).
Taxa previously known or suspected to show this disjunct (E-W) distribution pattern are
also represented in the Ituri collection, i.e.: Celaenorrhinus beni Bethune-Baker,
Celaenorrhinus bitjena Evans, Eagris subalbida aurivillii (Neustetter), Sarangesa thecla
(Pltz), Acleros sparsum Druce, Semalea atrio (Mabille), *Fresna carlo Evans, and *Fresna
This distributional pattern is of interest in that it may reflect the locations of prehistoric
refugia in Africa's central and western forest blocks (PALEOECOLOGY, below).
B. Other range extensions The Haut-Zaire records of each of three species reported
herein may represent eastward extensions of their respective known ranges within Zaire:
Pardaleodes edipus (Stoll), Andronymus helles Evans, and Borbo f fanta (Evans). In
addition, specimens in the Ituri collection appear to extend the known ranges of *Sarangesa
thecla mabira Evans (west- & southwestward) and Meza area (Neave) (northward).
PALEOECOLOGY OF THE ITURI REGION ('OKAPILAND')
Forest dynamics Contrary to first impressions, perhaps, the Congo Basin's vast forest
belt has not always been continuous and green. Large areas of it are underlain by wind-blown
Kalahari sands, giving evidence that at arid times in the past desert conditions have stretched
north to join with the southern Sahara (Moreau 1966, Kingdon 1971, 1989, Maley 1996, Larsen
1997, Livingstone 2001, White 2001).
Tropical rainforest areas themselves have not all been the havens of long-term stability
they were once thought to represent. It is now clear that the Pleistocene was a time of great
climatic instability in tropical Africa: throughout the Quaternary, wide fluctuations in climate
have periodically altered environments across the continent.
Periods of global climatic cooling were apparently associated with tropical aridity and
forest retreat, alternating with warmer, wetter periods which allowed the forest to advance.
During glacial intervals (cool, dry) in tropical Africa, lowland rainforests were apparently
drastically reduced in area, contracting to relatively stable refuge sites (ecological islands)
(Figure 1) around the periphery of the present forest belt where relatively high rainfall per-
sisted. These purported sanctuaries are variously called refugia, core areas, centers of ende-
mism, and (more recently) biodiversity "hotspots". With the subsequent return of favorable
conditions during interglacial intervals (warm, wet), the forest apparently expanded from
these refuge sites to colonize intervening areas from which it had disappeared. Rainforest
organisms were then able to emanate into the reforested areas, and did so at varying rates.
Repeated fragmentation and coalescence of the forest in this manner during glacial/inter-
glacial cycles is thought to have functioned as one mechanism of, or as one 'engine' driving,
speciation ('species pump' of Morley 2001, part of 'evolutionary whirlpool' of Kingdon 1989).
Recognition of the idea that much of the tropical rainforest is not very old geologically and
that its periodic fragmentation has influenced speciation was pioneered by work on
Amazonia by Haffer (1969) and on African environments by Chapin (1932), Carcasson (1964),
Moreau (1966, but see Diamond and Hamilton 1980 and Hamilton 1982), Kingdon (1971), and
It appears that at least 21 glacial/interglacial cycles have occurred in the past 2.3 million
yrs. (Hamilton 1988, 1992, Livingstone 2001, White 2001), each of them presumably with some
degree of concomitant retreat and subsequent advance of organisms from forest refugia. The
last of these (the most recent world glacial maximum, ca. 18,000 ybp) appears to have been
of extreme severity (Hamilton 1988, Larsen 1997); it appears that rainforest has spread out
fairly recently from particular refugia, beginning ca. 12,000 ybpfide Hamilton (1976, 1988).
Support for the validity of the 'Pleistocene Refuge Theory' in tropical Africa as outlined
above has come from the work of many authors, e.g., Hamilton (1976, 1982, 1988, 1992),
Kingdon (1980, 1989), Prance (1982a), Maley (1987, 1996, 2001), Richards (1996), van der
Maesen et al. (1996), Morley (2000), and Hamilton et al. (2001).
In addition, studies of particular groups of African forest organisms have provided support
for the theory and rather wide agreement about the locations of postulated refugia: plants
(van der Maesen et al. 1996, Hamilton et al. 2001); butterflies (Carcasson 1964, Larsen 1997,
Hamilton et al. 2001); reptiles and amphibians (Laurent 1973, Lawson and Klemens 2001);
birds (Diamond and Hamilton 1980, Crowe and Crowe 1982, Mayr and O'Hara 1986, Prigogine
1988, Dowsett-Lemaire and Dowsett 2001, Hamilton et al. 2001); mammals (Booth 1958,
Bigalke 1968, Kingdon 1971, 1980, 1989, Grubb 1978, 1982, Rodgers et al. 1982, Hamilton 1976,
1988, Hamilton et al. 2001, Ray 2001, Tutin and Vedder 2001); other organisms (Prance
1982a; summaries by Hamilton 1976, 1982, and by Rodgers et al. 1982).
The Ituri Forest Africa's two most species-rich forest blocks, Cameroon/Gabon and E
Zaire, lie on either side of the great, forested, Congo Basin (Carcasson 1964, Diamond and
Hamilton 1980, Hamilton 1982, 1992, Hamilton et al. 2001). The Ituri Forest adjoins closely
and overlaps the postulated location of the E Zaire refugium (Figure 1). The Ituri region
illustrates well the three characteristics typical of forest refugia outlined by Hamilton (1976,
1982, 1988, 1992), Diamond and Hamilton (1980), Mayr and O'Hara (1986), Kingdon (1989),
and others: 1) high species richness; 2) high levels of endemism; 3) the presence of isolated
populations of disjunctly-distributed taxa. The spatially coincident occurrence of these sorts
of centers gives support to the view that the Ituri region has served as a refugium in which
lowland forest has persisted during climatic vicissitudes of the Quaternary.
1) Species richness The E Zaire refugium was the largest, most ecologically varied, and
most centrally located of the three great African forest core areas (Upper Guinea,
Cameroon/Gabon, E Zaire), and it is presently faunally the most species-rich of the three
(Kingdon 1971, Hamilton 1976, Wilklde and Finn 1988; forest mammals, Bigalke 1968, Rodgers
et al. 1982; forest passerine birds, Diamond and Hamilton 1980). The Albertine Rift region of
Central Africa (where Zaire, Uganda, Rwanda, and Burundi meet), in fact, leads the continent
in overall biodiversity (Pomeroy 1993).
The Ituri region is wonderfully rich in species, representing a wide range of taxonomic
groups. Gradients of biotic diversity can be seen to fall away in every direction from the Ituri
core area (Figure 1). About 1,000 species of forest butterflies (Hesperioidea and
Papilionoidea) are estimated to occur in the E Zaire (Ituri) region; within Africa this number
is exceeded only by that in the Cameroon/Gabon core area, where more than 1,400 species
occur (Carcasson 1964, Larsen 1997).
2) Endemism Among the species endemic to the Ituri Forest are some of the continent's
most unusual and intriguing forms (e.g., Kingdon 1971, 1989). About 15% of forest mammal
species in the Ituri region are endemic to NE Zaire/DR Congo (Bigalke 1968, Hart and Hart
1992). The okapi (Figure 1), the 'flagship species' of the Ituri ecosystem, has become an icon
in Zaire and a symbol of Zairian national identity (pers. obs. 1991). The importance of E Zaire
as a refugium and center of endemism has been discussed as it concerns forest mammals by
Bigalke (1968), Rodgers et al. (1982), and Grubb (2001), and concerning forest birds by
Diamond and Hamilton (1980) and Prigogine (1988, Albertine Rift Refugium). Maley (2001)
and others have argued that the numerous endemic taxa typical of particular precincts within
the forest could have resulted from repeated forest fragmentation in times past, each time
with refuges in approximately the same locations.
3) Disjunct populations Some of the most compelling evidence for the existence and
locations of former refugia in the African forest belt comes from the present-day distributions
of particular organisms: core areas tend to be foci of isolated populations of species showing
disjunct (apparently relictual) distributions (e.g., Hamilton 1988). Numerous taxa in Africa
are represented by widely-separated, isolated populations at the western and eastern
periphery of the Congo Basin; the most common pattern of disjunction involves populations
centered on Cameroon/Gabon and on E Zaire (e.g.: butterflies and skippers, Carcasson 1964,
present study; forest birds, Diamond and Hamilton 1980, Prigogine 1988). The distribution of
populations of the gorilla (G. gorilla (Savage and Wyman)) is a well-known example of this
pattern: western and eastern subspecies are separated by a gap of ca. 750 km of unoccupied
rainforest (Hamilton 1982, 1992, Tutin and Vedder 2001). The most likely explanation for
these disjunct distributions in many cases is that a once continuous range was split during a
previous interval or intervals of forest contraction, and with subsequent forest expansion the
respective isolated populations have been slow to colonize the relatively recently-established
intervening areas of suitable habitat (i.e., they have not yet come into secondary contact)
(Diamond and Hamilton 1980, Mayr and O'Hara 1986, Hamilton 1992).
Although additional field work in the central Congo Basin (between the sites of postulated
refugia) might fill in some of the distributional gaps (Louette 1984, Dowsett-Lemaire and
Dowsett 2001), there is little doubt that the patterns of disjunction are real, and that the area
of intervening Congolese forest is faunally depauperate (Moreau 1966, Kingdon 1971, Crowe
and Crowe 1982, Hamilton 1982, 1992, Louette 1984, Mayr and O'Hara 1986, Grubb 2001).
The validity of the refuge theory as outlined above has been challenged (e.g., Connor 1986,
Endler 1982 [though refuted by Mayr and O'Hara 1986], White 1993), and alternative
explanations for the observed patterns of biodiversity have been put forth (e.g., Endler 1982
(incl. butterflies, birds), Colyn et al. 1991 and Grubb 2001 (mammals)). However, a
preponderance of scientific evidence supports the view that isolation in glacial refugia has
been a prime factor in the speciation of rainforest organisms, and the theory has gained wide
acceptance (e.g., Diamond and Hamilton 1980, van der Maesen et al. 1996, Larsen 1997, We-
ber and Vedder 2001).
The refuge model outlined here does not exclude the validity of other models (Prance
1982b, Maley 2001), and factors other than vicariance (isolation in refugia) have no doubt
been important in speciation in many taxa; an excellent summary of the possible role of oth-
er factors contributing to speciation is provided by Carcasson (1964). A more robust, com-
posite theory ('combined model') is now emerging, one in which the idea of isolation in larger
refugia is retained but which recognizes: 1) that peripheral regions of gallery forests and
small patches of lowland forest probably remained within the Congo Basin during arid inter-
vals (Grubb 2001, White 2001); 2) that additional 'forcing mechanisms' were at work (e.g.,
orogeny, rivercourses, various ecological barriers) inhibiting gene flow and dispersion (e.g.,
Morley 2000, Maley 2001); 3) that multiple immigrations into colonizing forest (rather than
persistence in ancient forest) may have occurred (Colyn et al. 1991 (primates)); and 4) that a
major lowland or fluvial refuge existed in the S/SW portion of the Congo Basin (shown as site
B, Figure 1).
It also seems that speciation of rainforest organisms (including butterflies) began and has
taken place over much longer time periods than was previously thought, apparently antedat-
ing events of the late Pleistocene (e.g., Hamilton 1976, Larsen 1997, Morley 2000). The
suggestion made by Larsen (1997), that disjunct populations have been slow to differentiate,
is borne out by our own observations, e.g., that the genitalia of d specimens (Coeliades
bixana Evans, Meza mabillei (Holland)) from disjunct populations in West Africa (Ghana)
and the Ituri Forest show no important differences.
SPECIES ACCOUNTS: HESPERIIDAE
Five hundred and one (501) skippers representing 89 species were collected, as follow.
Coeliades bixana Evans. 1S; site EVC; 28.vi.1991.
The range of this lowland forest species is described in Ackery et al. (1995) as "west Africa
from Guinea to Angola". Although the S collected may represent a stray, the possibility of
disjunct populations exists, in NE Zaire and possibly in Shaba (5 specimens in the AME
collection are labeled "Katanga: Kafakumba"). Ducarme (1999) reports having collected this
species at Beni and Mangina (Nord-Kivu) and at Teturi (Haut-Zaire). The Zairian records
reported here represent large eastward extensions of this species' known range.
Comparisons of the genitalia of the Ituri d with AME specimens from Ghana (Likpe [TL is
"Gold Coast"]) and Katanga ("Kafakumba") show no important differences.
Coeliades c. chalybe (Westwood). 3 86, 19; site EVC; 3-11.vii.1991.
Coeliadesf. forestan (Stoll). 29 ?; site EVC; 27.vi.-23.vii.1991.
Coeliades hanno (Plotz). 19; site EVC; 18.vii.1991.
Coeliades libeon (Druce). 12< 6, 59 9; sites EVC, AFM; 2-28.vii.1991.
Celaenorrhinus b. beni Bethune-Baker. 16; site AFM; 20.vii.1991.
Subsp. beni is an Ituri District endemic; Miller (1971) observed that it appears to be rather
rare and local.
Celaenorrhinus bitjena Evans. 16; site EVC; 23.vii.1991.
Ducarme (1999), too, reports having collected bitjena at Epulu. This species (figured in
Evans 1937) is known only from Cameroon, Gabon, and the Ituri Forest (Ackery et al. 1995).
Celaenorrhinus boadicea howarthi Berger. 16; site EVC; 25.vi.1991.
Subsp. howarthi is restricted to the Ituri region and Uganda (Ackery et al. 1995).
Celaenorrhinus galenus (Fabricius). 4 6, 22 9; sites EVC, TAF; 26.vi.-22.vii.1991.
Celaenorrhinus homeyeri (PlItz). 26 6; site EVC; 28.vi.-10.vii.1991.
Celaenorrhinus n. nigropunctata Bethune-Baker. 16, 19; site EVC; 28.vi.-25.vii.1991.
Celaenorrhinus ovalis Evans. 233, 29 9; site EVC; 24.vi.-17.vii.1991.
Celaenorrhinus proxima (Mabille). 46 d, 2 2 ; site EVC; 28.vi.-26.vii.1991.
Tagiadesflesus (Fabricius). 126 6, 139 9; site EVC; 23.vi.-28.vii.1991.
Eagris lucetia (Hewitson). 26 ; site EVC; 26-30.vi.1991.
Eagris subalbida aurivillii (Neustetter). 96 6, 4 9; site EVC; 26.vi.-28.vii.1991.
An apparently disjunct population of subsp. aurivillii is known in E Cameroon (Ackery et
Eagris t. tigris Evans. 16; site EVC; 16.vii.1991.
Calleagris lacteus (Mabille). 4d ; site EVC; 29.vi.-23.vii.1991.
Lindsey and Miller (1965) questioned whether it is appropriate to recognize any subspecies
within this species.
Eretis lugens (Rogenhofer). 563; site EVC; 29.vi.-28.vii.1991.
Sarangesa bouvieri (Mabille). 16; site EVC; 1.vii.1991.
Sarangesa b. brigida (Pl6tz). 66 3, 19; site EVC; 13-28.vii.1991.
Sarangesa maculata (Mabille). 1d; site EVC; 23.vii.1991.
Sarangesa tertullianus (Fabricius). Id; site EVC; 24.vi.1991.
Sarangesa thecla mabira Evans. 66 6; site EVC; 25.vi.-28.vii.1991.
Ducarme (1999) reports having collected mabira at Beni (Nord-Kivu) and at Mt. Hoyo
(Haut-Zaire). His records, the Ituri specimens reported here, and 266 from Katanga
("Kafakumba" and "Kapanga") in the collection at AME are apparently the first records of
thecla for Zaire and represent west- and southwestward extensions of the known range of
subsp. mabira (Ackery et al. 1995).
Abantis contigua Evans. 1d; site EVC; 11.vii.1991.
Spialia p. ploetzi (Aurivillius). 463, 29 9; site EVC; 26.vi.-18.vii.1991.
This is the only species of Spialia confined to evergreen forest (Kielland 1990, Larsen 1991);
de Jong and Congdon (1993) cite it as an example of a species which has successfully made
the adaptive shift from open grassland to forest.
Gorgyra bina Evans. 1d; site EVC; 16.vii.1991.
Teniorhinus ignita (Mabille). 216 19; sites EVC, AFM; 23.vi.-28.vii.1991.
Teniorhinus watsoni Holland. 66 6; sites EVC, AFM; 3-25.vii.1991.
Records of T watsoni exist from as far east as Uganda (Lindsey and Miller 1965).
Teniorhinus niger (Druce). 16; site AFM; 20.vii.1991.
T watsoni and T niger are sympatric in Cameroon, where they behave as distinct species
(Lindsey and Miller 1965; T. B. Larsen in litt., 5 Dec. 1996). The capture of T watsoni and T
niger in flight at the same Ituri microlocality on the same day (AFM, 20.vii.) lends support to
the view that they are distinct entities.
Ceratrichia aurea Druce. 53 6, 2 9; sites EVC, AFM; 28.vi.-21.vii.1991.
Ceratrichia h. hollandi Bethune-Baker. 86 39 9; sites EVC, AFM; 30.vi.-28.vii.1991.
Ceratrichia semilutea Mabille. d1; site AFM; 21.vii.1991.
Ceratrichia wollastoni Heron. 36 ; sites EVC, AFM; 25.vi.-21.vii.1991.
Pardaleodes bule Holland. 766, 39 9; site EVC; 29.vi.-22.vii.1991.
One pair was captured while in copula, 29.vi.
Pardaleodes edipus (Stoll). 46 6, 69 9; site EVC; 27.vi.-28.vii.1991.
Pardaleodes incerta (Snellen). 25dd, 79 9; site EVC; 24.vi.-27.vii.1991.
Pardaleodes sator (Westwood). 29 9; site EVC; 4-8.vii.1991.
Pardaleodes t. tibullus (Fabricius). 2dd, 19; site EVC; 25.vi.-18.vii.1991.
S 500 km
forest refugia (18,000 yr BP)
- forest limit (present)
FI savannas within present forest area
.......annual rainfall isohyets in mm (presi
gradients of decreasing biotic divers
Figure 1. The Ituri Forest of NE Zaire in relation to postulated lowland forest refugia of West and Central Africa during the
last world glacial maximum (ca. 18,000 ybp), at which time cool, dry conditions prevailed across equatorial Africa. The indicated
refugia appear to have been important as moist-forest sanctuaries during dry periods in past millennia; from them, newly-
differentiated species have apparently emanated during favorable periods when wetter conditions prevailed across the region.
The Ituri region is an important component of the E Zaire center of endemism. (Map modified after Maley (1987, 1996, 2001) and
Richards (1996); diversity-gradient arrows within the forest belt follow Carcasson (1964) and Hamilton (1982, 1988, 1992).)
(1), Upper Guinea core area (la, Sierra Leone/Liberia; Ib, E Ivory Coast/W Ghana);
(2), Cameroon/Gabon core area;
(3), E Zaire core area (Central Refuge of Kingdon 1971, 1980, 1989; Albertine Rift Refugium (in part) of Prigogine 1988).
(A) (inset), forest profile (Ituri region), adapted from Chapin (1932): a, primary forest; b, secondary forest; c, agricultural
clearing (village garden). (For a further sense of scale, arrowhead to right of "c" points to a relatively minuscule 6-ft-tall villager
standing to the right of a dwelling-place.)
(B), location of S/SW Zaire Basin lowland refuge postulated by Laurent (1973), Kingdon (1980, 1989), Prigogine (1988, Congo
Basin Refugium), and Maley (1996, 2001) (also cf. Major Fluvial Refuge of Colyn et al. 1991, Salongo and Kasai centers within
Fluvial Super-Refuge of Grubb 2001). Not shown: Montane forest refugia to E and W of Zaire Basin (e.g., Kingdon 1989, Maley
Other inset features are adapted as follow: Mbuti arrow indicating north is from Turnbull (1965); okapi image (center) is from
Kenneth W. Fink-Root Resources.
Although Lindsey and Miller (1965) concluded that torensis Bethune-Baker is indistinguish-
able from subsp. tibullus, torensis is now recognized as a distinct subspecies (LDM).
Ducarme (1999) reports having collected torensis at Beni and Mt. Atonza (Nord-Kivu), and at
Katanga [sic] and Mt. Hoyo (Haut-Zaire).
Ankolafan (Holland). 226 4 9 ; site EVC; 24.vi.-27.vii.1991.
Xanthodisca rega (Mabille). 16; site EVC; 10.vii.1991.
The range of this species as described in Ackery et al. (1995) is Senegal to Cameroon and
Gabon; the Ituri record is apparently the first for Zaire and extends the known range of this
species far to the east.
Rhabdomantis galatia (Hewitson). 13; site EVC; 11.vii.1991.
Osmodes adonia Evans. 266; sites EVC, AFM; 1-21.vii.1991.
Osmodes adosus (Mabille). 36 6, 29 ; sites EVC, AFM; 26.vi.-21.vii.1991.
Osmodes distinct Holland. 16, 49 9; sites EVC, TAF, AFM; 9-22.vii.1991.
Osmodes laronia (Hewitson). 26 ; site EVC; 4-14.vii.1991.
Osmodes thora (Pltz). 46 6, 29 ; site EVC; 28.vi.-26.vii.1991.
Paracleros biguttulus (Mabille). 76 6, 49 9; sites EVC, TLD, TAF, AFM; 27.vi.-28.vii.1991.
The genitalic pattern of the Ituri 3 6 closely matches that of AME specimens of P biguttulus
from Kenya we examined, and contrasts with P placidus (Plotz) from Ghana.
Acleros bibundica Strand. 26 19; site EVC; 26.vi.-l1.vii.1991.
Acleros nigrapex Strand. 26 6, 29 9; site EVC; 8-26.vii.1991.
Acleros ploetzi Mabille. 106 6, 3Y 9; site EVC; 26.vi.-28.vii.1991.
Acleros sparsum Druce. 46 3; site EVC; 24.vi.-26.vii.1991.
This species appears to consist of two disjunct populations, one in the Ituri Forest and one
in Cameroon (Ackery et al. 1995).
Semalea area (Mabille). 263 6, 129 9; sites EVC, AFM; 25.vi.-27.vii.1991.
Semalea atrio (Mabille). 16; site AFM; 21.vii.1991.
This species apparently consists of separate West African and E Zairian populations
(Ackery et al. 1995).
Semalea pulvina (Pltz). 22 6 179 9; sites EVC, NYK (19); 19.vi.-28.vii.1991.
Semalea sextilis (Plotz). 3 6; site EVC; 18-26.vii.1991.
This species is apparently rare throughout its range (Larsen 1991).
Hypoleucis ophiusa ophir Evans. 16; site EVC; 14.vii.1991.
Meza cybeutes pallida (Evans). 19; site AFM; 21.vii.1991.
Meza indusiata (Mabille). 19; site EVC; 17.vii.1991.
Meza larea (Neave). 19; site EVC; 28.vii.1991.
It is interesting that Lindsey and Miller (1965) placed 19 Meza sp. from Liberia doubtfully
Meza mabillei (Holland). 4 6, 19; sites EVC, AFM; 30.vi.-23.vii.1991.
This species' range as described in Ackery et al. (1995) is Sierra Leone to Nigeria and Gabon;
the Ituri records are apparently the first for Zaire and extend the known range of mabillei far
to the east.
Comparison of 6 genitalia shows no important differences between the Ituri specimens and
AME specimens of mabillei from Ghana.
Meza meza (Hewitson). 866, 69 9; site EVC; 27.vi.-18.vii.1991.
Andronymus caesar philander (Hopffer). 26 6, 19; site EVC; 25.vi.-28.vii.1991.
Andronymusfenestrella Bethune-Baker. 36 ; sites EVC, AFM; 8-20.vii.1991.
Andronymus helles Evans. 46 6; site EVC; 28.vi.-4.vii.1991.
Ducarme (1999) reports having collected helles at Mt. Hoyo (Haut-Zaire).
Andronymus n. neander (Plotz). 22& 6, 5 9 ; sites EVC, NYK (d1); 19.vi.-27.vii.1991.
Chondrolepis cynthia Evans. 16; site EVC; 14.vii.1991.
Although the range of this species is described in Ackery et al. (1995) as highland forest in
E Zaire and SW Uganda, the specimen reported here is from lowlands, as are two Ugandan
specimens in the AME collection (Entebbe).
Chondrolepis n. niveicornis (Plotz). 1d, 19; site EVC; 27-30.vi.1991.
Gamia shelleyi (Sharpe). 19; site EVC; 24.vi.1991.
Gretna cylinda (Hewitson). 26 ; site EVC; 30.vi.-23.vii.1991.
Caenides dacela (Hewitson). 3 6; site EVC; 23.vi.-17.vii.1991.
Caenides dacena (Hewitson). 3 6 29 9; sites EVC, AFM; 20-26.vii.1991.
Caenides kangvensis Holland. 19; site EVC; 29.vi.1991.
Caenides maracanda (Hewitson). 19; site TLD; 9.vii.1991.
The range of this species, extremely rare in collections, is described in Ackery et al. (1995)
as Nigeria, Cameroon, and Angola. The Ituri specimen, and 19 in the AME collection labeled
"Katanga: Kapanga", are apparently the first records of maracanda from Zaire and extend the
known range of this large skipper far to the east.
Monza alberti (Holland). 1d; site EVC; 15.vii.1991.
Monza cretacea (Snellen). 2 d 12; sites EVC, NYK (16); 19.vi.-26.vii.1991.
Melphinaflavina Lindsey and Miller. d1; site EVC; 8.vii.1991.
Melphina tarace (Mabille). Id; site EVC; 23.vii.1991.
The range of this species as described in Ackery et al. (1995) is Sierra Leone, Ivory Coast,
Nigeria, Cameroon, and the Congo Republic; the Ituri 6, and the collection record from Beni
(Nord-Kivu) reported by Ducarme (1999), are apparently the first records of tarace from Zaire
and extend this species' known range far to the east. In addition, the collection at AME
includes a long series of tarace from Ghana (coll. Fr. Theo. Maessen) and one specimen from
the Central African Republic (coll. S. C. Collins).
Melphina unistriga (Holland). d1, 19; site EVC; 28.vi.-26.vii.1991.
Fresna carlo Evans. 1d; site EVC; 4.vii.1991.
This rare species (figured in Evans 1937) has been known heretofore only from Ivory Coast,
Nigeria, Cameroon, and W Uganda (Ackery et al. 1995).
Fresna cojo (Karsch). 19; site EVC; 23.vii.1991.
The Ituri specimen, and 19 in the AME collection labeled "BELG. CONGO: Kafakumba",
appear to be the first Zairian records of this rare species (Ackery et al. 1995). Its range may
be comprised of two disjunct populations (Senegal to Nigeria; Uganda to W Kenya, and W
Tanzaniafide Kielland 1990).
Fresna netopha (Hewitson). 1d; site EVC; 23.vii.1991.
Fresna nyassae (Hewitson). 236, 19; site EVC; 23-27.vii.1991.
Platylesches galesa (Hewitson). d1; site EVC; 27.vi.1991.
Borbo b. borbonica (Boisduval). 19; site EVC; 17.vii.1991.
Borbo f fanta (Evans). 5 63, 29 9; site EVC; 29.vi.-18.vii.1991.
Borbo ffatueUus (Hopffer). 1d; site EVC; 12.vii.1991.
Borbo holtzi (Plotz) [epithet is often misspelled holtzii]. 19; site EVC; 14.vii.1991.
Gegenes niso brevicornis (Plotz). 43d, 29 9; sites EVC, NYK (ld, 19); 19.vi.-27.vii.1991.
It is a pleasure to thank Dr. Jacqueline Y. Miller (AME) for her invaluable assistance during
the course of this project, at every step of the way. Funding and encouragement was provided
by Drs. L. D. Miller, J. Y. Miller, and Mark J. Simon (Lighthouse Point, FL). Stephen R.
Steinhauser (AME) generously provided help with genitalic determinations. Our participation
in survey work in the Ituri Forest was encouraged by Dr. Julian C. Kerbis and Pamela K.
Austin (Field Museum of Natural History, Chicago) and by Drs. John A. and Terese B. Hart
(WCS): each of these people provided valuable help before, during, and after JFD's field work
in the region. Dr. David S. Wilkie (Tufts University [currently at WCS]) was a source of
important information about local conditions and interactions with the Mbuti Pygmies.
The late Monsieur Karl Ruf and the Gilman Investment Company (GIC [now Gilman
International Conservation]) contributed immeasurably to the success of this project by mak-
ing possible the use of the okapi station at Epulu as a base of operation. M. Rufs hospitality
and generosity were unfailing; his willingness and ability to facilitate investigators' field work
in the remoteness of the Ituri was extraordinary.
Personnel of the Mission Aviation Fellowship (MAF) and Centre Medical Evangelique
(CME, Nyankunde) gave willingly of their services and hospitality, and made possible JFD's
(uneventful) travel into Zaire: in this connection, special thanks are due Douglas M. Kimanthi,
Tom and Pam Tharp and the pilots of MAF, and Luis and Elida Perfetti (CME). Additional
assistance in arranging travel and security was provided by Dr. Theodore W. Walters, S.J.
(Toledo, OH), Sandra and J. E Nicolet (Mambasa), Jean Pierre Kahindo-Kataka (Epulu), and
Francis Voet (Sabena Belgian World airlines, Brussels).
We are grateful for help of various kinds received by JFD during the course of collecting
work at Epulu, especially from: collectors Wamba Mavungu (Siraphin), Ngadjo Ndjaikpa
Lotchu, Mbivo Lotchu, Muzinga-Fataki, and hired Mbuti assistants Abdallah, Yuma Fayizi, and
Banyale; advisers Marie-Louise, Steve Shurter (GIC), Marcel Enckoto, and Katembo Vital; and
Matatu and Tabu, Mbuti Pygmies hired as camp hands.
Several people and institutions have been of great help in facilitating follow-up work since
the 1991 effort in Epulu. Through the generous cooperation of Drs. J. A. and T. B. Hart and
Martha Schwartz (WCS), continuing communication with our collectors in Epulu has been
possible. In 1991, WCS-IZCN established a modern forestry research facility at Epulu, the
Centre de Formation et de Recherche en Conservation Forestiere (CEFRECOF), and in 1992,
IZCN established the Reserve de Faune A Okapis (RFO) (Okapi Wildlife Reserve) surround-
ing Epulu. We have been able to maintain an exchange of entomological information with
these institutions thanks to the Harts, Mme. Caroline Mandango (CEFRECOF Librarian),
Norbert Mushenzi Lusenge (Conservateur of IZCN's Station d'Epulu), and Dr. Mbieme Lokwa
L. (Conservateur previously).
For providing determinations of specimens and other useful correspondence, we thank
Steve C. Collins (African Butterfly Research Institute [ABRI], Nairobi), Dr. Torben B. Larsen
(Butterflies of West Africa project), Robert Ducarme (Beni [Zaire] and Brussels), Jacques
Hecq (Mont-sur-Marchienne, Belgium), and Dr. John E. Rawlins (Carnegie Museum of Natural
History, Pittsburgh). The Florida Museum of Natural History and AME have provided funding
for JFD's study of the Ituri collection in Sarasota.
This work would not have been possible without the use of the wonderful library resourc-
es of the University of Michigan (Ann Arbor): we sincerely thank Dr. Philip Myers and the U.
of M. Library staff, especially Charlene M. Stachnik (Museums Library), for facilitating JFD's
access to reference materials. Gary H. Bendig (Polygraphics, Toledo, OH) and M. P. Crawford
(Maumee, OH) provided very capable graphical help (Figure 1). Ms. Joni Trempe (Toledo,
OH) and T. R. Rosebrock (AME) assisted with typing of the manuscript, and Ms. Pauline Walle
(Rochester, MN) helped by publicizing our efforts to sustain local collectors through Zaire's
disturbances of 1991-1997.
Finally, we thank Charlotte N. Douglass and Dr. Bruce E. Douglass (parents of JFD, Roch-
ester, MN) and the Jesuit Community of St. John's (Toledo, OH) for their steadfast support.
Ackery, P. R., C. R. Smith, and R. I. Vane-Wright, eds. 1995. Carcasson's African Butterflies: an
annotated catalogue of the Papilionoidea and Hesperioidea of the Afrotropical Region.
CSIRO Australia, East Melbourne. xii + 803 pp.
Barns, T. A. 1922. The wonderland of the eastern Congo, the region of the snow-crowned
volcanoes the Pygmies the giant gorilla and the okapi. G. P. Putnam's Sons, London and
New York. xxxv + 288 pp., 108 pls., 1 map.
Bigalke, R. C. 1968. Evolution of mammals on southern continents. HI. The contemporary
mammal fauna of Africa. Quart. Rev. Biol. 43: 265-300.
Booth, A. H. 1958. The Niger, the Volta and the Dahomey Gap as geographic barriers.
Evolution 12: 48-62.
Bultot, F 1971. Atlas climatique du bassin congolais, deuxibme parties. Publications de l'Inst.
nat. pour I'ltude agronomique du Congo (I.N.E.A.C.) [Bruxelles], Rep. dem. Congo.
Unnumbered text pp. + maps, figs., tables.
Carcasson, R. H. 1964. A preliminary survey of the zoogeography of African butterflies. E. Aft.
Wildl. J. 2: 122-157.
Chapin, J. P. 1932. The birds of the Belgian Congo, Part I. Bull. Amer. Mus. Nat. Hist. 65. x +
756 pp., 10 pis., 1 map.
Colyn, M., A. Gautier-Hion, and W. Verheyen. 1991. A re-appraisal of palaeoenvironmental
history in Central Africa: evidence for a major fluvial refuge in the Zaire Basin. J. Biogeogr.
Connor, E. F 1986. The role of Pleistocene forest refugia in the evolution and biogeography
of tropical biotas. Trends in Ecol. & Evol. 1: 165-168.
Crowe, T. M., and A. A. Crowe. 1982. Patterns of distribution, diversity and endemism in
Afrotropical birds. J. Zool. (London) 198: 417-442.
Curran, B. K., and R. K. Tshombe. 2001. Integrating local communities into the management
of protected areas: lessons from DR Congo and Cameroon. Pp. 513-534 in Weber, W., L. J.
T. White, A. Vedder, and L. Naughton-Treves, eds., African rain forest ecology and
conservation: an interdisciplinary perspective, Yale University Press, New Haven and
London, xiii + 588 pp.
de Jong, R., and T. C. E. Congdon. 1993. The montane butterflies of the eastern Afrotropics.
Pp. 133-164 + Appendices 8.1 & 8.2 in Lovett, J. C., and S. K. Wasser, eds., 1993, Biogeogra-
phy and ecology of the rain forests of eastern Africa, Cambridge University Press, ix + 341
Diamond, A. W, and A. C. Hamilton. 1980. The distribution of forest passerine birds and
Quaternary climatic change in tropical Africa. J. Zool. (London) 191: 379402.
Dowsett-Lemaire, F, and R. J. Dowsett. 2001. African forest birds: patterns of endemism and
species richness. Pp. 233-251 in Weber, W., L. J. T. White, A. Vedder, and L. Naughton-
Treves, eds., African rain forest ecology and conservation: an interdisciplinary perspective,
Yale Univ. Press, New Haven and London, xiii + 588 pp.
Ducarme, R. 1993. Liste des lIpidopteres rhopaloceres r6colt6s dans le nord est du Zaire.
Unpubl. MS, ii + 67 pp., 4 tables.
Ducarme, R. 1999. Liste des lepidopteres rhopaloc6res recolt6s dans le nord est du Congo.
Unpubl. MS, ii + 93 pp., 1 map.
Endler, J. A. 1982. Pleistocene forest refuges: fact or fancy? Pp. 641-657 in Prance, G. T, ed.,
Biological diversification in the Tropics, Columbia University Press, New York, xvi + 714
Evans, W. H. 1937. A catalogue of the African Hesperiidae indicating the classification and
nomenclature adopted in the British Museum. Jarrold and Sons Ltd., Norwich. xii + 212 pp.,
Grubb, P 1978. Patterns of speciation in African mammals. Bull. Carnegie Mus. Nat. Hist. No.
Grubb, P 1982. Refuges and dispersal in the speciation of African forest mammals. Pp. 537-
553 in Prance, G. T., ed., Biological diversification in the Tropics, Columbia University
Press, New York, xvi + 714 pp.
Grubb, P. 2001. Endemism in African rain forest mammals. Pp. 88-100 in Weber, W., L. J. T.
White, A. Vedder, and L. Naughton-Treves, eds., African rain forest ecology and conserva-
tion: an interdisciplinary perspective, Yale University Press, New Haven and London, xiii +
Haffer, J. 1969. Speciation in Amazonian forest birds. Science 165: 131-137.
Hamilton, A. 1976. The significance of patterns of distribution shown by forest plants and
animals in tropical Africa for the reconstruction of Upper Pleistocene palaeoenvironments:
a review. Palaeoecology of Africa & of the surrounding islands & Antarctica 9: 63-97. A. A.
Balkema, Cape Town, Rotterdam.
Hamilton, A. C. 1982. Environmental history of East Africa: a study of the Quaternary.
Academic Press, London, New York. xiv + 328 pp.
Hamilton, A. C. 1988. Guenon evolution and forest history. Pp. 13-34 in Gautier-Hion, A., F.
Bourlifre, J.-P Gautier, and J. Kingdon, eds., A primate radiation: evolutionary biology of
the African guenons, Cambridge University Press, Cambridge, New York, Melbourne, viii +
567 pp., 7 pls.
Hamilton, A. C. 1992. History of forests and climate. Pp. 17-25 in Sayer, J. A., C. S. Harcourt,
and N. M. Collins, eds., The conservation atlas of tropical forests: Africa, Simon & Schuster,
New York, London, Toronto, Sydney, Tokyo, Singapore, 288 pp.
Hamilton, A., D. Taylor, and P. Howard. 2001. Hotspots in African forests as Quaternary
refugia. Pp. 57-67 in Weber, W., L. J. T. White, A. Vedder, and L. Naughton-Treves, eds.,
African rain forest ecology and conservation: an interdisciplinary perspective, Yale Univer-
sity Press, New Haven and London, xiii + 588 pp.
Hart, J. A. 2001. Diversity and abundance in an African forest ungulate community and
implications for conservation. Pp. 183-206 in Weber, W., L. J. T. White, A. Vedder, and L.
Naughton-Treves, eds., African rain forest ecology and conservation: an interdisciplinary
perspective, Yale University Press, New Haven and London, xiii + 588 pp.
Hart, J. A., and J. S. Hall. 1996. Status of eastern Zaire's forest parks and reserves. Conserva-
tion Biol. 10: 316-324.
Hart, J. A., and T. B. Hart. 1988. A summary report on the behaviour, ecology and conserva-
tion of the okapi (Okapiajohnstoni) in Zaire. Acta Zool. Pathol. Antverp. 80: 19-28.
Hart, J. A., and T. B. Hart. 1989. Ranging and feeding behaviour of okapi (Okapiajohnstoni)
in the Ituri Forest of Zaire: food limitation in a rain-forest herbivore? Symp. Zool. Soc.
London 61: 31-50.
Hart, J. A., and T. B. Hart. 1992. The Ituri Forest. P. 281 in Sayer, J. A., C. S. Harcourt, and N.
M. Collins, eds., The conservation atlas of tropical forests: Africa, Simon & Schuster, New
York, London, Toronto, Sydney, Tokyo, Singapore, 288 pp.
Hart, T. B. 1985. The ecology of a single-species-dominant forest and of a mixed forest in
Zaire, Africa. Ph.D. dissertation, Michigan State Univ., East Lansing. xi + 215 pp.
Hart, T. B. 1995. Seed, seedling and sub-canopy survival in monodominant and mixed forests
of the Ituri Forest, Africa. J. Tropical Ecol. 11: 443-459.
Hart, T. B. 2001. Forest dynamics in the Ituri Basin (DR Congo): dominance, diversity, and
conservation. Pp. 154-164 in Weber, W., LJ. T. White, A. Vedder, and L. Naughton-Treves,
eds., African rain forest ecology and conservation: an interdisciplinary perspective, Yale
University Press, New Haven and London, xiii + 588 pp.
Hart, T. B., and J. A. Hart. 1986. The ecological basis of hunter-gatherer subsistence in African
rain forests: the Mbuti of eastern Zaire. Human Ecol. 14: 29-55.
Hart, T. B., J. A. Hart, R. Dechamps, M. Fournier, and M. Ataholo. 1996. Changes in forest
composition over the last 4000 years in the Ituri basin, Zaire. Pp. 545-563 in van der
Maesen, L. J. G., X. M. van der Burgt, and J. M. van Medenbach de Rooy, eds., 1996, The
biodiversity of African plants, Proc. XIVth AETFAT Congress, Kluwer Academic
Publishers, Dordrecht, Boston, London, xv + 861 pp.
Hart, T. B., J. A. Hart, and P. G. Murphy. 1989. Monodominant and species-rich forests of the
humid tropics: causes for their co-occurrence. Amer. Nat. 133: 613-633.
Kielland, J. 1990. Butterflies of Tanzania. Hill House, Melbourne & London. 363 pp., 68 pls.,
Kingdon, J. 1971. East African mammals: an atlas of evolution in Africa, Vol. I. Academic
Press, London, New York. x + 446 pp.
Kingdon, J. S. 1980. The role of visual signals and face patterns in African forest monkeys
(guenons) of the genus Cercopithecus. Trans. Zool. Soc. London 35: 425-475.
Kingdon, J. 1989. Island Africa: the evolution of Africa's rare animals and plants. Princeton
University Press, Princeton, New Jersey. 287 pp.
Larsen, T. B. 1991. The butterflies of Kenya and their natural history. Oxford University Press,
New York. xxii + 490 pp., 64 pls., 19 + 4 figs.
Larsen, T. B. 1997. An overview of the distribution patterns of Afrotropical rainforest butter-
flies (Lepidoptera: Rhopalocera). Metamorphosis [Lep. Soc. Afr.], Occas. Suppl. 3: 183-188.
Laurent, R. F 1973. A parallel survey of equatorial amphibians and reptiles in Africa and South
America. Pp. 259-266 in Meggers, B. J., E. S. Ayensu, and W. D. Duckworth, eds., Tropical
forest ecosystems in Africa and South America: a comparative review, Smithsonian
Institution Press, Washington, DC, viii + 350 pp.
Lawson, D. P, and -M. W. Klemens. 2001. Herpetofauna of the African rain forest: overview and
recommendations for conservation. Pp. 291-307 in Weber, W., L. J. T. White, A. Vedder, and
L. Naughton-Treves, eds., African rain forest ecology and conservation: an interdisciplinary
perspective, Yale Univ. Press, New Haven and London, xiii + 588 pp.
Lindsey, A. W., Jr., and L. D. Miller. 1965. Superfamily Hesperioidea, family Hesperiidae. Pp.
47-146 in Fox, R. M., A. W. Lindsey, Jr., H. K. Clench, and L. D. Miller, The butterflies of
Liberia, Mem. Amer. Ent. Soc. No. 19, ii + 438 pp., 1 pl., 233 figs.
Livingstone, D. A. 2001. A geological perspective on the conservation of African forests. Pp.
50-56 in Weber, W., L. J. T. White, A. Vedder, and L. Naughton-Treves, eds., African rain
forest ecology and conservation: an interdisciplinary perspective, Yale University Press,
New Haven and London, xiii + 588 pp.
Louette, M. 1984. Apparent range gaps in African forest birds. Pp. 275-286 in Proc. 5th Pan-
Afr. Ornithol. Congr. [held in 1980], S. Afr. Ornithol. Soc., Johannesburg.
Makana, J.-R., T. B. Hart, and J. A. Hart. 1998. Forest structure and diversity of lianas and
understory treelets in monodominant and mixed stands in the Ituri Forest, Democratic
Republic of the Congo. Pp. 429-446 in Dallmeier, F, and J. A. Comiskey, eds., Forest
biodiversity research, monitoring and modeling: conceptual background and Old World
case studies, Man and the Biosphere Ser. Vol. 20, UNESCO, Paris, and Parthenon Publish-
ing Group, Pearl River, New York, and Carnforth, Lancs., UK, xxiii + 671 pp.
Maley, J. 1987. Fragmentation de la forCt dense humide africaine et extension des biotopes
montagnards au Quaternaire recent: nouvelles donnees polliniques et chronologiques.
Implications paldoclimatiques et biogdographiques. Palaeoecology of Africa and the sur-
rounding islands 18: 307-334. A. A. Balkema, Rotterdam, Brookfield (Vermont).
Maley, J. 1996. Le cadre paleoenvironnemental des refuges forestiers africains: quelques
donn6es et hypotheses. Pp. 519-535 in van der Maesen, L. J. G., X. M. van der Burgt, and J.
M. van Medenbach de Rooy, eds., The biodiversity of African plants, Proc. XIVth AETFAT
Congress, Kluwer Academic Publishers, Dordrecht, Boston, London, xv + 861 pp.
Maley, J. 2001. The impact of arid phases on the African rain forest through geological histo-
ry. Pp. 68-87 in Weber, W., L. J. T. White, A. Vedder, and L. Naughton-Treves, eds., African
rain forest ecology and conservation: an interdisciplinary perspective, Yale University
Press, New Haven and London, xiii + 588 pp.
Mayr, E., and R. J. O'Hara 1986. The biogeographic evidence supporting the Pleistocene
forest refuge hypothesis. Evolution 40: 55-67.
Miller, L. D. 1971. Descriptions of new species and notes on other Hesperiidae of Africa. Bull.
Allyn Mus. No. 2, 17 pp., 75 figs.
Moreau, R. E. 1966. The bird faunas of Africa and its islands. Academic Press, New York,
London. ix + 424 pp.
Morley, R. J. 2000. Origin and evolution of tropical rain forests. John Wiley & Sons, Ltd.,
Chichester and New York [Chichester, New York, Weinheim, Brisbane, Singapore,
Toronto]. xv + 362 pp.
National Geographic Society. 1992. "Heart of Brightness" (television episode). Nat. Geogr.
Explorer, produced by Alan Root, ca. 45 min. [This TV program documents the work of
Drs. J. A. and T. B. Hart in the Ituri Forest.]
National Geographic Society. 1996. "Heart of Africa: Forest Primeval" (videocassette). Nat.
Geogr. Video and Survival Anglia, Ltd., produced by Alan Root and (for Nat. Geogr.
Television) Cathy McConnell, ca. 60 min. [This video program features the forest biome of
Pomeroy, D. 1993. Centers of high biodiversity in Africa. Conservation Biol. 7: 901-907.
Prance, G. T., ed. 1982a. Biological diversification in the Tropics. Columbia University Press,
New York. xvi + 714 pp.
Prance, G. T. 1982b. Introduction. Pp. 3-5 in Prance, G. T, ed., Biological diversification in the
Tropics, Columbia University Press, New York, xvi + 714 pp.
Prigogine, A. 1988. Speciation pattern of birds in the central African forest refugia and their
relationship with other refugia. Proc. Internat. Ornithol. Congr. 19 [Acta XIX Congr.
Internat. Omithol., Vol. II, Univ. of Ottawa Press, Canada]: 2537-2546.
Ray, J. C. 2001. Carnivore biogeography and conservation in the African forest: a community
perspective. Pp. 214-232 in Weber, W., L. J. T. White, A. Vedder, and L. Naughton-Treves,
eds., African rain forest ecology and conservation: an interdisciplinary perspective, Yale
University Press, New Haven and London, xiii + 588 pp.
Richards, P. W. 1996. The tropical rain forest: an ecological study, 2nd ed. Cambridge
University Press, Cambridge, New York, Melbourne. xxiii + 575 pp.
Rodgers, W. A., C. F. Owen, and K. M. Homewood. 1982. Biogeography of East African forest
mammals. J. Biogeogr. 9: 41-54.
Sayer, J. A., C. S. Harcourt, and N. M. Collins, eds. 1992. The conservation atlas of tropical
forests: Africa. Simon & Schuster, New York, London, Toronto, Sydney, Tokyo, Singapore.
Stephenson, P. J., and J. E. Newby. 1997. Conservation of the Okapi Wildlife Reserve, Zaire.
Oryx 31: 49-58.
Talbot, G. 1921. New Lepidoptera collected by Mr. T. A. Barns, in east central Africa. I.-
Introduction. Bull. Hill Mus. 1(1): 40-44, 1 map.
Talbot, G. 1922. Mr. Barns' collections and aspects of butterfly life in Africa. Pp. 267-270 in
Barns, T. A., The wonderland of the eastern Congo, the region of the snow-crowned
volcanoes the Pygmies the giant gorilla and the okapi, G. P. Putnam's Sons, London and
New York, xxxv + 288 pp., 108 pls., 1 map.
Turnbull, C. M. 1965. The Mbuti Pygmies: an ethnographic survey. Anthropol. Pap. Amer. Mus.
Nat. Hist. 50: 137-282, pls. 26-38.
Tutin, C. E. G., and A. Vedder. 2001. Gorilla conservation and research in Central Africa: a
diversity of approaches and problems. Pp. 429-448 in Weber, W, L. J. T. White, A. Vedder,
and L. Naughton-Treves, eds., African rain forest ecology and conservation: an interdisci-
plinary perspective, Yale Univ. Press, New Haven and London, xiii + 588 pp.
van der Maesen, L. J. G., X. M. van der Burgt, and J. M. van Medenbach de Rooy, eds. 1996.
The biodiversity of African plants, Proc. XIVth AETFAT Congress. Kluwer Academic
Publishers, Dordrecht, Boston, London. xv + 861 pp.
Weber, W, and A. Vedder. 2001. Trends and challenges in applied research and conservation:
a practitioner's view from the African rain forest. Pp. 547-556 in Weber, W., L. J. T. White,
A. Vedder, and L. Naughton-Treves, eds., African rain forest ecology and conservation: an
interdisciplinary perspective, Yale University Press, New Haven and London, xiii + 588 pp.
White, E 1983. The vegetation of Africa: a descriptive memoir to accompany the
UNESCO/AETFAT/UNSO vegetation map of Africa. U. N. Educ., Sci. and Cult. Org., Paris.
356 pp. + 1 legend sheet + map (3 sheets dated 1981).
White, F. 1993. Refuge theory, ice-age aridity and the history of tropical biotas: an essay in
plant geography. Fragm. Flor. Geobot. Suppl. 2(2): 385-409.
White, L. J. T. 2001. The African rain forest: climate and vegetation. Pp. 3-29 in Weber, W., L.
J. T. White, A. Vedder, and L. Naughton-Treves, eds., African rain forest ecology and
conservation: an interdisciplinary perspective, Yale University Press, New Haven and
London, xiii + 588 pp.
Wilkie, D. S., and J. T. Finn. 1988. A spatial model of land use and forest regeneration in the
Ituri Forest of northeastern Zaire. Ecol. Modelling 41: 307-323.
Wilkie, D. S., and J. T. Finn. 1990. Slash-burn cultivation and mammal abundance in the Ituri
Forest, Zaire. Biotropica 22: 90-99.
Numbers of the Bulletin of the AUyn Museum are published at irregular intervals. Price: $2.40