ArticlePDF Available

Book review. Guide des Serpents d’Afrique Occidentale, Centrale et d’Afrique du Nord.

Authors:

Abstract

We provide a review of Jean-François Trape’s opus “Guide des Serpents d’Afrique Occidentale, Centrale et d’Afrique du Nord” covering the whole snake fauna of 32 countries in West, Central, and North Africa (Bioko is not included, and only the part of Angola which is north of 13°S is included). The guide is divided into a foreword, a geographical introduction, chapters on snake bites, classification, identification keys followed by synoptic tables allowing for a direct comparison of diagnostic characters, species accounts, literature cited, and an alphabetical index of scientific names. We report some mistakes in the identification keys for several genera (Hydraethiops, Leptotyphlops, Letheobia, Naja, Philothamnus, Polemon, Prosymna, Psammophis, Tricheilostoma).
Herpetological Review 54(4), 2023
700700 BOOK REVIEWS
better. The lack of some icon or text pointing to venomous
and potentially deadly species is regrettable, as adding memes
could have helped readers avoid dangerous species and allow
harmless snakes to avoid maltreatment. This would have
helped to educate people not to harm most species, which the
authors mention is a problem in their discussion, and to avoid
envenomation. The guide appears aimed at English speaking
locals and perhaps people from Lebanon, southeastern Turkey,
Iraq, and northern Jordan—and for those (like me) fascinated by
precise distribution data. It is probably not intended for those
interested more in aspects of natural history or in a less restricted
geographic extent.
I think that this little book forms an excellent platform to
educate local enthusiasts about how to identify Syrian snakes,
and at least gives international colleagues an idea of the diversity
of species inhabiting or potentially inhabiting the region.
Hopefully in the future, it will lead to the production of an
expanded edition when the political situation within Syria, and
its relations with its neighbors, allows for easier, safer, and more
in-depth study of the fascinating snakes of the region.
liteRatuRe cited
al-sheiKhly, o. F., W. bohMe, n. a. Fazaa, h. K. abdulzahRa, M. K. haba,
R. h. al-assady, and s. b. Mousavi. 2021. Extraordinary colour
morphs in the lebetine (Levantine) viper Macrovipera lebetinus
(Linnaeus, 1758) from the Mesopotamian marshes of southern
Iraq and southwestern Iran. Sauria 43:38–50.
baR, a., g. haiMovitch, and s. MeiRi. 2021. Field Guide to the Amphib-
ians and Reptiles of Israel. Edition Chimaira, Frankfurt Am Main,
Germany. 511 pp.
beRgeR-dellMouR, v. h. 1986. Zur Herpetofauna des Golan. Annalen
des Naturhistorischen Museum in Wien 87B:59–67.
boettgeR, o. 1878. Reptilien und Amphibien aus Syrien. Bericht
der Senckenbergischen Naturforschenden Gesellschaft 1878–
1879:57–84.
disi, a. M., d. ModRý, P. necas, and l. RiFai. 2001. Amphibians and
Reptiles of the Hashemite Kingdom of Jordan: An Atlas and Field
Guide. Edition Chimaira, Frankfurt am Main, Germany. 408 pp.
esteRbaueR, h. 1992. Die herptofauna des ostlichen Golan- und Her-
mongebietes funde und bemerkungen zur systematik und okolo-
gie. Zool. Middle East 7:21–54.
hooFien, J. h. 1968. Contributions to the herpetofauna of Mount Her-
mon. II. On some lacertids and colubrids. Israel J. Zool. 17:199–
204.
JablonsKi, d., M. a. RibeiRo-JúnioR, e. siMonov, K. soltys, and s. MeiRi.
2023. A new, rare, small-ranged, and endangered mountain snake
of the genus Elaphe from the Southern Levant. Sci. Rep. 13:4839.
sivan, n. and y. l. WeRneR. 1992. Survey of the reptiles of the Golan
plateau and Mt. Hermon, Israel. Israel J. Zool. 37:193–211.
uetz, P. 2023. The Reptile Database. Available at http://reptile-data-
base.reptarium.cz. Accessed 28 Aug 2023.
WeRneR, y. l. and e. avital. 1980. The herpetofauna of Mt. Hermon
and its altitudinal distribution. Israel J. Zool. 29:192–193.
zinneR, h. 1972. Contributions to the herpetofauna of Mount Hermon
No. iii Elaphe quatuorlineata (Ophidia: Colubridae). Israel J. Zool.
21:123–127.
Herpetological Review, 2023, 54(4), 700–702.
© 2023 by Society for the Study of Amphibians and Reptiles
Guide des Serpents d’Afrique Occidentale,
Centrale et d’Afrique du Nord
Jean-François Trape. 2023. IRD Editions, Marseille, France (www.
editions.ird.fr). 896 pp. Softcover. €65.00 (ca. US $69.00). ISBN: 978-2-
7099-2974-5.
OLIVIER S. G. PAUWELS
Royal Belgian Institute of Natural Sciences,
Rue Vautier 29, 1000 Brussels, Belgium;
e-mail: opauwels@naturalsciences.be
JONATHAN BRECKO
Royal Belgian Institute of Natural Sciences, Rue
Vautier 29, 1000 Brussels, Belgium;
Royal Museum for Central Africa,
Leuvensesteenweg 13, 3080 Tervuren, Belgium
This massive and solidly bound field
guide covers all snake species found in
32 countries in West, Central, and North
Africa (Bioko is not included, and only
the part of Angola which is north of 13°S
is included). No one, except the French herpetologist Laurent
Chirio, approaches the level of knowledge of field herpetology
in western and central Africa acquired by Jean-François Trape
during decades of expeditions in all geographical areas covered
by this book. Taxa treated here and not encountered by Trape
in the field are few. Of the 375 recognized snake taxa inhabiting
the area covered, more than 30 were (co-) described by Trape,
belonging to various taxonomic groups, from worm snakes to
cobras and vipers. Besides a personal collection of more than
25,000 West and Central African snakes that he has gathered
since 1980 and deposited in several institutions, Trape visited
many museums in Europe and Africa in order to examine
preserved specimens of species that he had not or only rarely
collected himself.
The guide is divided into a foreword, a geographical
introduction, chapters on snake bites, classification,
identification keys followed by synoptic tables allowing for
a direct comparison of diagnostic characters (pp. 65–172),
species accounts (pp. 174–837), literature cited (pp. 839–890),
and an alphabetical index of scientific names. The introduction
to regional geography is abundantly illustrated with various
biotopes photographed by Trape in not less than 22 countries,
and each photograph is accompanied by geographical
coordinates. Some of these photographs are truly spectacular,
and show rarely illustrated sites.
The chapter on snake bites gives general information on the
occurrence of snake bites in the region, followed by an overview
of the main taxa accounting for snake bites arranged in four
main groups: vipers, elapids, colubrids, and other snakes. For
each of these groups, the effects and general treatment of the
bites are mentioned.
An introduction to the diagnostic morphological characters
of snakes and their nomenclature, with several scalation
illustrations, precedes the identification keys. The keys are
abundantly illustrated with clear drawings and photographs
and are very practical. The photographs presented in the keys
often illustrate different individuals than those shown in the
species accounts, thus allowing for better documentation of the
Herpetological Review 54(4), 2023
701BOOK REVIEWS 701
variation within these species. We successfully used the keys with
preserved specimens of Grayia spp., Philothamnus spp., and
Psammophis spp. We noted a number of discrepancies between
the scale variation provided in the keys and in the species
accounts, but they generally have no or little consequence for
correctly identifying the species, so we will not list them here.
Some mistakes in the keys could misdirect users, however;
we therefore mention several of the most important ones as
follows. In the key to families on p. 65, the first couplet leads to
couplets 2 and ‘’32;’’ the latter should be corrected to ‘‘3.’’ In the
key to Leptotyphlopidae on p. 67, the 2nd couplet mentions that
Tricheilostoma broadleyi has a single anterior supralabial between
the nasal and the ocular, although in the accompanying scalation
drawing and the account for this species there are two. The first
part of the 9th couplet of the same key leading to Leptotyphlops
emini and L. monticolus should read that the posterior supralabial
does not reach the level of the eye. The first part of the 13th couplet
proposes ‘‘less than 45 subcaudals’’ (versus 49–55 subcaudals),
but this leads to Myriopholis adleri, which shows a variation
of 41 to 50 subcaudals. In couplet 9 of the key to Typhlopidae,
‘‘Less than 600 longitudinal scales; 24 to 27 transversal scale
rows’’ (versus > 628 longitudinal scales and 22 transversal scale
rows) leads to species with up to 660 longitudinal scales or 22
transversal scale rows. In part of couplet 13 leading to Letheobia
weidholzi, ‘‘frontal slightly narrower than frontal’’ should be
corrected to ‘‘frontal slightly narrower than rostral.’’ In the key
to Viperidae, the second part of the 3rd couplet mentions ‘‘more
than 40 subcaudals’’ (versus less than 40 subcaudals), but this
leads to species showing as little as 35 subcaudals. The second
part of the 21st couplet in the key to Elapidae mentions “17 to 35
dorsal scale rows at midbody” (versus 15), but this leads to Naja
anchietae, which has 15–19 dorsal scale rows. The first part of the
10th couplet of the key to Colubridae, “Single anal scale” (versus
divided) leads to Hydraethiops melanogaster, which has a divided
anal, as correctly stated in the species account. In the first part
of the 18th couplet of the same key, “51 to 79 ventrals” should be
corrected to “51 to 79 subcaudals,” and “About 40 ventrals” to
About 40 subcaudals.” In the 33rd couplet, “11 to 15 dorsal scale
rows” (versus 17–35) leads to Prosymna ambigua, which has 15
or 17 dorsal scale rows. In couplet 49, “Subcaudals unkeeled”
(versus keeled), lead to P. dorsalis, which has keeled subcaudals, as
stated in the species account. In the 100th couplet, “8 supralabials,
of which the 4th and 5th contact the eye” (versus 9, of which the
5th and 6th contact the eye) leads to Psammophis elegans, which
has 9 supralabials, of which the 5th and 6th contact the eye. In the
key to the snakes of North Africa and Sahara, there is no access
to couplet 40, but two accesses to couplet 39: this should be
corrected in couplet 32, which then leads to couplets 33 and 40,
not 33 and 39. There are also numerous discrepancies between
the tables showing the character variations on pp. 164–168 and
the respective species accounts, including contradictions about
the state of the anal scale (single versus divided) in several species
(for example, for Polemon ater and P. christyi). For the tables that
span several pages, the headings of the columns appear only on
the first page, which renders consultation of the characters on the
following pages difficult.
Due to their morphological similarities, the species accounts
for the Colubridae and Lamprophiidae and some incertae sedis
taxa are grouped into a single section. Here, they appear in
alphabetical order regardless of family or subfamily they belong
to, although their subfamily and family affiliations, when known,
are always indicated near the species name.
Each species account spans two pages: the left one for the
text and the map; the right one for two photos. The text page
includes the scientific name, the French and English common
names, a short diagnosis, and sections on size, distribution
and habitat, morphological description, and natural history,
respectively. The latter section unfortunately does not include
data on reproduction, with an exception for three species,
Duberria lutrix, Pseudaspis cana and Indotyphlops braminus.
Trape indicates that the diet of Hypoptophis wilsonii is unknown,
but Das et al. (2022) recorded an amphisbaenian (Zygaspis
quadrifrons) in an individual from Upemba National Park in
southeastern Democratic Republic of the Congo. The text page
includes a gridded map (square degrees longitude/latitude) and
generally two drawings illustrating head scalation in different
views.
Without exception, the photographs are of good to excellent
quality. They were most often taken by Trape himself, in all areas
covered by the book, often in remote and herpetologically poorly
explored locations. The fact that photographs are accompanied
by the localities where they were taken should be commended,
as this important information is rarely provided in guides. Only
a few photographs are lacking a locality beyond the country in
which they were taken. The locality for a photograph taken by
Wolfgang Wüster of a live Naja multifasciata on p. 96 is given
as “Congo-Kinshasa.” In fact, this individual was a long-term
captive maintained in the Liverpool School of Tropical Medicine,
and its country of origin is actually unknown (W. Wüster, pers.
comm.). The photograph of an adult Causus lichtensteini on p.
287 is said to have been taken in “Ituri (Congo-Kinshasa)”; it
was in fact taken in “Bombatia (3°00’17.7”N 25°09’32.2”E), Bas-
Uele” (E. Greenbaum, pers. comm.). On p. 501, the caption of the
photograph of a Limaformosa capensis mentions “Botswana”; it
was more precisely taken in Moeding College, Otse (25°01’48.0”S
25°42’36.0”E), a small village in South-East District, Botswana (S.
Spawls, pers. comm.).
Head scale drawings are unfortunately not accompanied by
the collection number or geographical origin of the specimens
upon which they are based. In some cases, only dead or
preserved individuals could be illustrated, and this is not always
indicated in the caption of the photographs (for example,
for the Kladirostratus togoensis shown on the top of p. 499, or
the Gabonese Polemon fulvicollis on p. 633). The Bothrolycus
ater from Bamenda Highlands shown on the top of p. 391 has
a strikingly large eye and a well visible banded pattern; this is
probably due to the fact that it was a tiny newborn individual (V.
Gvoždík, pers. comm.).
The maps owe much to the field and lab work of Trape: “Nearly
all squares on the gridded maps for Senegal, Mali, Guinea, Togo,
Congo-Brazzaville, Niger, and Chad, plus three quarters of those
for Mauritania, Benin, and Ivory Coast, correspond to specimens
I collected.” And Trape’s contribution to the knowledge of the
geographic distribution of African snakes is not limited to these
countries.
There are occasionally discrepancies between the scale
numbers provided in the diagnoses and the main text in a number
of species accounts, but these have little consequence for species
identification (for example, in Psammophis afroccidentalis,
the diagnosis gives a variation of 159–183 ventrals and 97–120
subcaudals, while the main text gives a variation of 156–185 and
96–121, respectively). In addition, the main text of the species
account of Philothamnus ruandae on p. 610 erroneously says
that its subcaudals are keeled, whereas the diagnosis correctly
Herpetological Review 54(4), 2023
702702 BOOK REVIEWS
says they are smooth, as can be verified on the photograph on
p. 611. The diagnosis of Polemon ater and P. christyi on p. 624
erroneously says that their cloacal scale is single, whereas the
main text of the species account correctly says it is divided. The
diagnosis and text for Polemon barthi says that its subcaudals are
divided, whereas they are actually single (de Witte and Laurent
1947), as correctly stated in the key and the table.
The literature cited, although not exhaustive, includes no
fewer than 1202 references, including all of the most pertinent
and useful sources; it is a real goldmine for taxonomists needing
to assemble a library on the snakes of the region. Publications
up to the year 2022 are included in the references, and some of
Trape’s papers in press for 2023 are mentioned.
For West Africa, this new opus comes as a perfect complement
to a comparable book on all reptiles except snakes published by
Trape et al. (2012; see review by Pauwels and Kok 2013). The new
book is more accurate and covers many more species than the
latest guide to snakes of a comparable geographical area, co-
authored by Jean-Philippe Chippaux and Kate Jackson, which
was partly outdated by the time of its release in 2019 due to a
four-year long publication delay; it also did not cover North
Africa and did not provide species accounts for the Scolecophidia
(Pauwels and Brecko 2020). Given the pace at which new snake
species are being described in this part of the World (as noted by
Trape, 53 have been named since 2000), it is important to have
an updated reference. Trape provides many important remarks
in the species accounts about possible undescribed species and
taxa in dire need of revision, and further includes opinions on
the validity of some taxa, including a suggestion that Myriopholis
boueti (Chabanaud, 1917) is a senior synonym of M. rouxestevae
(Trape and Mané 2004).
A new species is even described within the book, Naja
obscura Saleh and Trape in Trape, 2023. In order to fit within the
same format as the other species accounts, the description is
brief. The account states that two specimens were used for the
description, but a collection number is provided only for the
holotype. The number of ventrals is given as 200 to 210, but one
does not know to which of the two specimens they respectively
refer, and the number of subcaudals is given as environ une
soixantaine” (more precisely stated as being about 65 in a table
on p. 170). In fact, the paratype has been deposited in the reptile
collections of the Institut de Recherche pour le Développement
under the number IRD TR.4654, and it is the paratype that has
200 ventral scales (Trape, pers. comm.). A field guide is not the
ideal context for describing new species because such a brief
description calls for a follow-up with more comprehensive
analysis, thus providing missing details, as had to be done for
gecko species described in the book on non-snake West African
reptiles by Trape et al. (2012; see Ineich et al. 2015).
Referring to as yet undescribed species by name in a field guide
also can lead to nomenclatural issues of priority. For example,
one species account is dedicated to ‘Dipsadoboa riparia Trape,
Mediannikov et Baldé, 2023,’ whereas the description of this
snake was not yet published when the book was made available.
The book was distributed from 14 July 2023, and OSGP received
his copy in late July. The species’ description (Trape et al. 2023)
was published on 1 September 2023. Concerning Philothamnus
carinatus, Trape indicates that it includes two genetically distinct
populations; shortly after the book was released, the review of
this species was published, and the population living south and
east of the Congo River was named Philothamnus chifunderai
(Greenbaum et al. 2023). On p. 589, the photograph above
shows a typical Philothamnus carinatus from Gabon, and the
one below a P. chifunderai from Rwanda. The genus Micrelaps,
included by Trape in the Lamprophiidae, was demonstrated to
belong, along with the genus Brachyophis, to a distinct family,
the Micrelapidae (Das et al. 2023).
The text is written in a flawless, pleasant to read, French.
We noted only a few mistypings, mostly in scientific names and
English common names. Thanks to his taxonomic work over the
last decade and this new, brilliant opus, Trape has demonstrated
that this region of the World is much richer herpetologically than
was formerly thought. This book, compiling the encyclopedic
knowledge of Trape on African snakes, will greatly facilitate
herpetological research in the region, and is definitely a must in
any herpetological or natural history library.
Acknowledgments.—We are grateful to Eli Greenbaum, Václav
Gvoždík, Stephen Spawls, Jean-François Trape, and Wolfgang Wüster
for providing useful information.
liteRatuRe cited
chiPPauX, J.-P. and K. JacKson. 2019. Snakes of Central and Western Af-
rica. Johns Hopkins University Press, Baltimore, Maryland. x + 429
pp.
das, s., J. bRecKo, o. s. g. PauWels, and J. MeRilä. 2022. Cranial osteolo-
gy of Hypoptophis (Aparallactinae: Atractaspididae: Caenophidia),
with a discussion on the evolution of its fossorial adaptations. J.
Morph. 283:510–538.
———, e. gReenbauM, s. MeiRi, a. M. baueR, F. t. buRbRinK, c. J. RaX-
WoRthy, J. l. Weinell, R. M. bRoWn, J. bRecKo, o. s. g. PauWels, n.
Rabibisoa, a. P. RaseliManana, and J. MeRilä. 2023. Ultraconserved
elements-based phylogenomic systematics of the snake super-
family Elapoidea, with the description of a new Afro-Asian family.
Mol. Phylogen. Evol. 180:107700
de Witte, g.-F. and R. lauRent. 1947. Revision d’un groupe de Colubri-
dae africains. Genres Calamelaps, Miodon, Aparallactus et formes
affines. Mém. Mus. Roy. Hist. Natur. Belg. 29:1–134.
gReenbauM, e., o. s. g. PauWels, v. gvoždíK, e. R. vaughan, t. chaney,
M. buonteMPo, M. a. MWenebatu, W. M. Muninga, and h. M. engel-
bRecht. 2023. Systematics of the thirteen-scaled green snake Philo-
thamnus carinatus (Squamata: Colubridae), with the description
of a cryptic new species from Central and East Africa. Afr. J. Her-
petol. 72:119–144.
ineich, i., J.-F. tRaPe, l. chiRio, and J. couRtois. 2015. Rectifications sur
les séries-types et nouvelles données concernant les geckos afric-
ains Hemidactylus kundaensis Chirio & Trape, 2012 et Tarentola
pastoria Trape, Baldé & Ineich, 2012 déposés au Muséum national
d’Histoire naturelle. Bull. Soc. Herpétol. France 153:49–52.
PauWels, o. s. g. and J. bRecKo. 2020. Book review. Snakes of Central
and Western Africa. Herpetol. Rev. 51:161–164.
——— and P. J. R. KoK. 2013. Book review. Lézards, crocodiles et Tor-
tues d’Afrique Occidentale et du Sahara by Jean-François Trape,
Sébastien Trape and Laurent Chirio. Herpetol. Rev. 44:165–167.
tRaPe, J.-F., o. MedianniKov, and M. c. baldé. 2023. Une nouvelle es-
pèce de couleuvre arboricole du genre Dipsadoboa Günther, 1858
(Squamata: Colubridae: Colubrinae) d’Afrique de l’Ouest. Bull.
Soc. Herpétol. France 183:1–13.
———, s. tRaPe, and l. chiRio. 2012. Lézards, Crocodiles et Tortues
d’Afrique Occidentale et du Sahara. IRD Editions, Marseille,
France. 503 pp.
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Recent molecular phylogenies of African green snakes suggested the geographically widespread species Philothamnus carinatus includes at least two distinct lineages. We utilised an integrative taxonomic approach with morphological and genetic data to reconcile the taxonomic status of these cryptic lineages, including the recently described taxon P. brunneus from West Africa. We sequenced three mitochondrial (16S, cyt b and ND4) and two nuclear (c-mos and RAG1) genes from several Central African populations of P. carinatus and combined our data with other closely related species to infer a maximum likelihood phylogenetic tree with IQ-TREE. Our results are consistent with previous studies that showed P. cf. carinatus populations from Democratic Republic of the Congo (DRC) represent a cryptic lineage that is distinct from P. carinatus sensu stricto in Cameroon, Central African Republic, Equatorial Guinea (including Bioko Island), Gabon, eastern Nigeria, Republic of the Congo, and extreme western DRC. In our preferred tree, P. brunneus (limited to 16S molecular data) was recovered as a relatively long branch in a moderately supported clade with P. carinatus sensu stricto, whereas P. cf. carinatus populations from northern Angola, most of DRC, and East Africa (Burundi, Kenya, Rwanda, Tanzania and Uganda) were described as a new species, Philothamnus chifunderai n.sp. A possible hybrid population between south-eastern Cameroon and north-western DRC is consistent with an increasing body of evidence suggesting the Ubangi River might represent a hybrid zone area.
Article
Full-text available
Fossoriality evolved early in snakes, and has left its signature on the cranial morphology of many extinct Mesozoic and early Caenozoic forms. Knowledge of the cranial osteology of extant snakes is indispensable for associating the crania of extinct lineages with a particular mode of life; this applies to fossorial taxa as well. In the present work, we provide a detailed description of the cranium of Hypoptophis wilsonii, a member of the subfamily Aparallactinae, using micro‐computed tomography (CT). This is also the first thorough micro‐CT‐based description of any snake assigned to this African subfamily of predominantly mildly venomous, fossorial and elusive snakes. The cranium of Hypoptophis is adapted for a fossorial lifestyle, with increased consolidation of skull bones. Aparallactines show a tendency towards reduction of maxillary length by bringing the rear fangs forward. This development attains its pinnacle in the sister subfamily Atractaspidinae, in which the rear fang has become the ‘front fang’ by a loss of the part of the maxilla lying ahead of the fang. These dentitional changes likely reflect adaptation to subdue prey in snug burrows. An endocast of the inner ear of Hypoptophis shows that this genus has the inner ear typical of fossorial snakes, with a large, globular sacculus. A phylogenetic analysis based on morphology recovers Hypoptophis as a sister taxon to Aparallactus. We also discuss the implications of our observations on the burrowing origin hypothesis of snakes. This article is protected by copyright. All rights reserved.
  • I Ineich
  • J J.-F. Trape
  • Courtois
ineich, i., J.-F. tRaPe, l. chiRio, and J. couRtois. 2015. Rectifications sur les séries-types et nouvelles données concernant les geckos africains Hemidactylus kundaensis Chirio & Trape, 2012 et Tarentola pastoria Trape, Baldé & Ineich, 2012 déposés au Muséum national d'Histoire naturelle. Bull. Soc. Herpétol. France 153:49-52.
Une nouvelle espèce de couleuvre arboricole du genre Dipsadoboa Günther, 1858 (Squamata: Colubridae: Colubrinae) d'Afrique de l'Ouest
  • J.-F Trape
  • M C Mediannikov
tRaPe, J.-F., o. MedianniKov, and M. c. baldé. 2023. Une nouvelle espèce de couleuvre arboricole du genre Dipsadoboa Günther, 1858 (Squamata: Colubridae: Colubrinae) d'Afrique de l'Ouest. Bull. Soc. Herpétol. France 183:1-13. ---, s. tRaPe, and l. chiRio. 2012. Lézards, Crocodiles et Tortues d'Afrique Occidentale et du Sahara. IRD Editions, Marseille, France. 503 pp.