From forest to plantation? Obscure articles reveal alternative host plants for the coffee berry borer, Hypothenemus hampei (Coleoptera: Curculionidae)

From forest to plantation? Obscure articles revealalternative host plants for the coffee berry borer,Hypothenemus hampei (Coleoptera: Curculionidae)

FERNANDO E. VEGA FLS1*, AARON P. DAVIS FLS2 and JULIANA JARAMILLO3,4

1Sustainable Perennial Crops Laboratory, United States Department of Agriculture, AgriculturalResearch Service, Building 001, BARC-W, Beltsville, MD 20705, USA2Royal Botanic Gardens, Kew, Richmond, Surrey, TW9 3AB, UK3Institute of Plant Diseases and Plant Protection, University of Hannover, Hannover, Germany4International Centre of Insect Physiology and Ecology (ICIPE), Nairobi, Kenya

Received 22 February 2012; revised 28 March 2012; accepted for publication 28 March 2012bij_1912 1..9

The coffee berry borer, Hypothenemus hampei (Coleoptera: Curculionidae: Scolytinae), is the most devastatinginsect pest of coffee throughout the world. The insect is endemic to Africa but can now be found throughout nearlyall coffee-producing countries. One area of basic biology of the insect that remains unresolved is that of itsalternative host plants, i.e. which fruits of plants, other than coffee, can the insect survive and reproduce in. Anin-depth survey of the literature revealed an article by Schedl listing 21 genera in 13 families in which the insectwas collected, mainly in the Democratic Republic of Congo. This overlooked reference, together with informationprovided in other early articles, suggests that H. hampei is polyphagous, and could provide, if confirmed in thefield, critical information on the evolution of this insect’s diet, ecology and host range. © 2012 The Linnean Societyof London, Biological Journal of the Linnean Society, 2012, ••, ••–••.

ADDITIONAL KEYWORDS: Coffea – ecology – evolution – insect host range – speciation.

INTRODUCTION

The coffee berry borer, Hypothenemus hampei(Ferrari, 1867) (Coleoptera: Curculionidae: Scolyti-nae), an insect endemic to Africa (Vega et al., 2009;Gauthier, 2010), was first reported in exported coffee(Coffea L.) seeds (‘beans’) by Ferrari (1867), describedas Cryphalus hampei. It was first collected in the fieldin 1897 in Mount Coffee, Liberia, classified as Stepha-noderes cooki (Hopkins, 1915), which subsequentlybecame a synonym for H. hampei (Schedl, 1959;Wood, 2007). In 1901 it was reported as a pest ofCoffea canephora Pierre ex. A. Froehner in the Repub-lic of Congo (Fleutiaux, 1901), and ever since theinsect has disseminated throughout nearly all coffee-producing countries in the world, and has become themost devastating insect pest of the two commer-cial Coffea species, C. arabica L. and C. canephora,

causing an estimated $500 million in losses yearly(Vega, Franqui & Benavides, 2002).

Hypothenemus hampei has a cryptic life cycle, withan adult female boring a hole in the coffee berry(botanically known as a drupe) and then a tunnel inthe endosperm (i.e. the major component of a coffeeseed), in which eggs are laid. Upon hatching, larvaecommence to feed on the endosperm, thereby produc-ing small galleries throughout the seed, reducing theyield and quality of the marketable product. To com-plicate matters, there is sibling mating inside theberry, and once adult females emerge from the berry,they are already inseminated and ready to oviposit inother berries. The cryptic life cycle inside the berrymakes the insect quite difficult to control by bothchemical and biological methods. The use of biologicalcontrol agents such as parasitoids, predators, andfungal entomopathogens is a pest management alter-native that has been practiced in several countries(Jaramillo, Borgemeister & Baker, 2006; Vega et al.,*Corresponding author. E-mail: [email protected]

Biological Journal of the Linnean Society, 2012, ••, ••–••. With 1 figure

bs_bs_banner

© 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, ••, ••–•• 1

2009), and yet the insect continues to pose a formi-dable challenge to coffee growers throughout theworld.

The natural habitats of C. arabica and C. cane-phora are the humid, evergreen forests of Africa(Davis et al., 2006): the former is a high-altitudespecies (900–2000 m a.s.l.) of south-western Ethiopiaand the surrounding regions; the latter is a predo-minately lowland plant (50–1500 m a.s.l.) foundthroughout much of tropical Africa, west of the RiftValley (Davis et al., 2006). One aspect of paramountimportance in dealing with the basic biology ofH. hampei has been determining its alternative hostplants in Africa, i.e. are there plants other thanC. arabica and C. canephora that, in their naturalhabitat, are suitable for the reproduction and devel-opment of the insect?

Recently, as part of a project aimed at locatingalternative host plants for H. hampei in Africa, wecompiled a list of Hypothenemus species that havebeen reported in Africa together with the host plantsfor these species, when known. We identified 94Hypothenemus species out of 179 described species ashaving been recorded in Africa (F. E. Vega, unpubl.data). This process required keeping track of oldnames for Hypothenemus species, based on Wood &Bright (1987, 1992), Bright & Skidmore (1997, 2002),and Wood (2007), and consulting the literaturereviewed in these works. During this review, and toour great surprise and excitement, we discoveredthree obscure studies, either not previously cited orundervalued by subsequent researchers working inour field. These works have important consequencesfor the study of alternative hosts of H. hampei.

MATERIAL AND METHODS

Three articles have revealed important aspects of thebiology of the coffee berry borer: Beille (1925), Ghes-quière (1933), and Schedl (1960). The most significantwas Schedl’s (1960) article, which was found whiletracking records for Hypothenemus species in Africa(see previous paragraph), and was cited by Wood& Bright (1987). It appears the importance of thisarticle was not recognized because of its title, Insectesnuisibles aux fruits et aux graines (Insect pests offruits and grains), which does not give any indicationabout its coverage of H. hampei.

RESULTS

The first significant finding we came across was aarticle by Beille (1925), in which he writes:

Le début de la maladie a coïncidé avec la destruction de la forêt;il semble que le Stephanoderes, privé des végétaux sur lesquels

il pullulait, ait trouvé dans les nouvelles plantations de Caf-éiers, des conditions favorable à son évolution. [The onset of thedisease coincided with the destruction of the forest; it appearsthat Stephanoderes (i.e. Hypothenemus hampei), lacking theplants that it frequents, has found in the new coffee planta-tions, conditions that are favourable to his evolution.]

This article is significant because, as far as we know,it is the first mention of plants in the forest beingoriginal host plants for the insect, and ascribes forestdestruction as a contributor to the insect moving intocoffee plantations. Before being commercially planted,C. canephora was endemic to humid, evergreenforests in Africa (Davis et al., 2006), and Beille’s state-ment would require the insect to have used C. cane-phora (or other wild Coffea species) as one of its manyhost plants in the forest, in order to make it plausiblefor the insect to have then been able to exploit coffeeplantations.

The second significant finding comes from an articleby Schedl (1960), in which he reports results of a fieldsurvey he conducted in the Belgian Congo (present-day Democratic Republic of Congo, DRC), togetherwith findings from other scientists. Karl E. Schedl(1898–1979) was a world-renowned bark beetle tax-onomist who published 342 articles dealing with barkbeetle taxonomy (Wood & Bright, 1992). What isquite striking about Schedl’s (1960) article is thatH. hampei was recovered from 20 plant genera (otherthan Coffea) in 13 families (Table 1). Some examplesof damaged fruits presented by Schedl (1960) areshown in Figure 1. The valid name for the coffeeberry borer at the time of publication was Stephano-deres hampei, and what Schedl (1960) reports asStephanoderes punctatus was later re-classified asH. hampei by Wood (1972). Concerning H. hampei,Schedl (1960: 13) writes:

Il est intéressant, du point de vue biologique, de noter que,d’après les recherches faites par l’auteur à Yangambi etd’après les données bibliographiques au sujet de Stephano-deres hampei Ferr., il existe dans la forêt ombrophile touteune série d’hôtes naturels du parasite qui lui donnent lapossibilité de se développer indépendamment des plantationsde caféiers. Il est certain que cet insecte est une espèceendémique de la forêt ombrophile; il ne s’établit que secon-dairement dans les plantations de caféiers. [It is interesting,from the biological point of view, to note that after the inves-tigations conducted by the author in Yangambi and after thebibliographical data on the subject of Stephanoderes hampeiFerr., there exists inside the rainforest a series of naturalhosts for the parasite that give it the possibility to developindependently from coffee plantations. It is certain that thisinsect is an endemic species of the rainforest; it becomesestablished only secondarily in coffee plantations.]

Schedl (1960) specifically mentions that the forestplants provide the insect with ‘the possibility to

2 F. E. VEGA ET AL.

© 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, ••, ••–••

Tab

le1.

Pla

nts

inw

hic

hS

ched

l(1

960)

repo

rted

the

coff

eebe

rry

bore

r(H

ypot

hen

emu

sh

ampe

i)as

atta

ckin

gfr

uit

sor

seed

s.S

teph

anod

eres

pun

ctat

us

and

Ste

phan

oder

esh

ampe

iar

esy

non

yms

for

H.h

ampe

i.A

tota

lof

21ge

ner

ain

13fa

mil

ies

are

repo

rted

.Hos

tpl

ant

nam

esin

pare

nth

eses

indi

cate

the

curr

ent

nam

e

Fam

ily

Hos

tpl

ant

(cu

rren

tn

ame)

H.h

ampe

ire

port

edas

:R

epor

ted

loca

tion

Pla

nt

hab

itP

lan

tdi

stri

buti

onP

lan

tE

colo

gyF

ruit

See

dsS

eed

fat

con

ten

t

Ach

aria

ceae

Cal

onco

baaf

f.cr

epin

ian

aS

teph

anod

eres

pun

ctat

us

DR

CTr

ee(r

arel

ya

shru

b),

10(–

25)

mta

ll

Trop

ical

Wes

tan

dC

entr

alA

fric

aH

um

idfo

rest

;85

0–15

00m

.a.s

.l.C

apsu

le,

subg

lobu

lar,

(5–)

7(–8

)¥

5–6(

–8)

cm,

oran

ge;

wit

hn

um

erou

sse

eds

See

dsan

gula

r,0.

7–0.

8¥

0.5–

0.6

mm

Un

know

n.

Oth

ersp

ecie

sin

gen

us

said

tobe

‘oil

rich

’

Apo

cyn

acea

eP

leio

carp

atu

bici

na

Sta

pf(P

leio

carp

apy

cnan

tha

(K.

Sch

um

)S

tapf

)

Ste

phan

oder

esh

ampe

iD

RC

Tree

20(–

30)

mta

ll;

bole

up

to50

cmin

diam

eter

Trop

ical

Wes

tan

dC

entr

alA

fric

a,an

dw

este

rnE

ast

Afr

ica

Hu

mid

fore

st,

prim

ary

and

seco

nda

ry;

up

to23

00m

.a.s

.l.

Com

pose

dof

seve

ral

basa

lly

join

edca

rpel

s,ea

chca

rpel

glob

ose

toel

lips

oid,

1.3–

2.3(

–3)

cmlo

ng,

yell

owto

oran

ge;

each

carp

el2-

seed

ed

See

dsel

lips

oid

toob

lon

g,6.

5–13

.5m

mlo

ng

Un

know

n

Big

non

iace

aeS

path

odea

cam

pan

ula

taP.

Bea

uv.

S.h

ampe

iD

RC

Tree

,25

(–35

)m

tall

;bo

leu

pto

60cm

indi

amet

er

Trop

ical

Wes

tan

dC

entr

alA

fric

a,an

dw

este

rnE

ast

Afr

ica

Hu

mid

fore

stan

dsa

van

na,

prim

ary

and

seco

nda

ry;

up

to20

00m

.a.s

.l.

Woo

dyca

psu

le,

deh

isci

ng

by2

valv

es,

nar

row

lyel

lips

oid,

15–2

7¥

3.5–

7cm

;se

eds

nu

mer

ous

See

dsth

inan

dfl

at,

very

broa

dly

win

ged,

~1.5

¥2

cm

Un

know

n

Cal

oph

ylla

ceae

Mam

mea

afri

can

aS

abin

eS

.ham

pei

DR

CTr

ee,

30–4

5m

tall

;bo

leu

pto

125

cmin

diam

eter

Trop

ical

Afr

ica

Hu

mid

fore

st,

prim

ary

and

seco

nda

ry,

up

to10

00m

.a.s

.l.

Dru

pe,

wit

h1–

4(o

ne

seed

ed)

pyre

nes

,gl

obos

eto

pear

-sh

aped

,7–

18(–

30)

¥10

cm

See

ds1–

4,fl

atte

ned

,~3

cmlo

ng

See

ds:

~10%

oil

(9.6

%oi

l:co

nsi

stin

gof

am

ixtu

reof

olei

cac

id34

.9%

,pa

lmit

icac

id32

.3%

,li

nol

eic

acid

22.9

%)

[pal

mit

icac

id(2

8%),

stea

ric

acid

(27.

5%),

myr

isti

cac

id(1

.5%

),la

uri

cac

id(1

%)]

Clu

siac

eae

All

anbl

acki

afl

orib

un

da

Oli

v.S

.ham

pei

DR

CTr

ee,

~30

mta

ll;

bole

~50

cmor

mor

ein

diam

eter

Trop

ical

Wes

tan

dC

entr

alA

fric

aH

um

idfo

rest

,pr

imar

yan

dse

con

dary

;u

pto

1000

m.a

.s.l.

Ber

ry,

elli

psoi

d,20

–50

¥5–

14cm

(col

our)

;40

–80-

seed

ed,

each

encl

osed

ina

pin

kish

aril

See

dsov

oid,

2.5–

3¥

1.5–

2cm

Ker

nel

(~60

%se

edm

ass)

:~7

2%fa

t(s

tear

icac

id45

–58%

and

olei

cac

id40

–51%

)C

ombr

etac

eae

Term

inal

iasu

perb

aE

ngl

.&

Die

ls.

S.p

un

ctat

us

DR

CTr

ee,

up

to45

(–50

)m

tall

;bo

le12

0(–1

50)

cmin

diam

eter

Trop

ical

Wes

tan

dC

entr

alA

fric

aH

um

idfo

rest

,pr

imar

yan

dse

con

dary

;u

pto

1000

m.a

.s.l.

Win

ged

nu

t(i

nde

his

cen

t),

tran

sver

sely

oblo

ng-

elli

ptic

alin

outl

ine,

1.5–

2.5

¥4–

7cm

;1-

seed

ed

See

d~1

.5¥

7cm

Un

know

n

FROM FOREST TO PLANTATION 3


Tab

le1.

Con

tin

ued

Fam

ily

Hos

tpl

ant

(cu

rren

tn

ame)

H.h

ampe

ire

port

edas

:R

epor

ted

loca

tion

Pla

nt

hab

itP

lan

tdi

stri

buti

onP

lan

tE

colo

gyF

ruit

See

dsS

eed

fat

con

ten

t

Leg

um

inos

aeP

has

eolu

slu

nat

us

L.

S.h

ampe

iU

gan

daC

lim

bin

g,tr

aili

ng

orm

ore

orle

sssh

rubb

y,an

nu

alor

pere

nn

ial;

stem

su

pto

4.5(

–8)

mlo

ng

Cu

ltiv

ated

thro

ugh

out

the

wor

ld,

intr

opic

alan

dte

mpe

rate

regi

ons;

intr

odu

ced

from

Trop

ical

Sou

thA

mer

ica

Cu

ltiv

ated

un

der

vari

ous

con

diti

ons

Pod

,[?

deh

isce

nt]

,ob

lon

g,(4

.5–)

5–10

.5(–

13)

¥1–

2(–3

)cm

;2–

4(–5

)-se

eded

See

dski

dney

-sh

aped

torh

ombo

idor

glob

ose,

8–11

¥6–

7m

m

Edi

ble

port

ion

[per

100

g(i

mm

atu

rebe

an)]

:w

ater

66.3

g,pr

otei

n8.

3g,

carb

ohyd

rate

23.1

g,fi

bre

1.0

g~0

.7%

fat

Leg

um

inos

aeC

ath

orm

ion

alti

ssim

um

(Hoo

kf.

)H

utc

h&

Dan

dy(A

lbiz

zia

alti

ssim

aH

ook

f.)

S.p

un

ctat

us

DR

CTr

ee,

15–3

5m

tall

;bo

leu

pto

80cm

indi

amet

er

Trop

ical

Wes

tan

dC

entr

alA

fric

aH

um

idan

dsu

bhu

mid

fore

st,

prim

ary

and

seco

nda

ryfo

rest

;10

–100

0m

.a.s

.l.

Pod

,br

eaki

ng

up

into

1-se

eded

segm

ents

,n

arro

wly

oblo

ng,

10–2

8¥

1–2

cm,

red-

brow

nto

blac

kish

;u

pto

20-s

eede

d

See

dsob

lon

gto

len

s-sh

aped

,fl

atte

ned

,0.

6–0.

9¥

0.6–

0.7

mm

Fer

men

ted

seed

s:25

.3%

prot

ein

,16

.9%

fat

and

10%

carb

ohyd

rate

Leg

um

inos

aeC

aesa

lpin

iapu

lch

erri

ma

(L.)

Sw

.

S.p

un

ctat

us

DR

CS

hru

bor

smal

ltr

ee,

up

to5

mta

llP

roba

bly

nat

ive

totr

opic

alA

mer

ica

(th

ough

ofte

nal

lege

dto

beA

siat

ic);

wid

ely

cult

ivat

edin

the

trop

ics

Cu

ltiv

ated

un

der

vari

ous

con

diti

ons

intr

opic

alan

dsu

btro

pica

lre

gion

s

Pod

,de

his

cin

gdo

wn

both

side

s,ob

liqu

ely

oblo

ng

toob

lan

ceol

ate,

flat

ten

ed,

6–12

¥1.

4–2.

5cm

;se

vera

lse

eded

See

ds,

obov

ate

orob

tria

ngu

lar,

com

pres

sed,

9–10

¥7–

8¥

3m

m

Un

know

n

Leg

um

inos

aeD

iali

um

laco

urt

ian

um

Ver

moe

sen

(D.

engl

eria

nu

mH

enri

q.)

S.h

ampe

i,S

.pu

nct

atu

sD

RC

Tree

8–23

mta

ll,

orsh

rub

1–3

mta

llTr

opic

alW

est

and

Cen

tral

Afr

ica

Hu

mid

fore

st;

900–

1450

m.a

.s.l.

Pod

,in

deh

isce

nt,

±ob

ovoi

d,2.

2–3.

7¥

1.4–

2.5

cm;

1-or

2-se

eded

,ea

chem

bedd

edin

afl

oury

pith

See

ds±

len

ticu

lar,

8–13

¥7–

8.5

¥3.

5–5

mm

Un

know

n

Leg

um

inos

aeO

xyst

igm

aox

yph

yllu

m(H

arm

s)L

eon

ard

S.h

ampe

iD

RC

Tree

,u

pto

45m

tall

;bo

le13

0(–1

80)

cmin

diam

eter

Trop

ical

Wes

tan

dC

entr

alA

fric

aH

um

idan

dsu

bhu

mid

fore

st;

up

to80

0m

.a.s

.l.

Pod

inde

his

cen

t,sa

mar

a-li

ke(l

eaf-

like

,ve

ryfl

atte

ned

),3.

5–13

¥2.

2–8

cm;

1-se

eded

See

dsfl

atte

nde

d-el

lips

oid,

2–5

¥2–

4cm

Un

know

n.

Oth

ersp

ecie

sin

gen

us

said

tobe

‘oil

rich

’

Mal

vace

aeC

ola

gris

eifl

ora

De

Wil

d.S

.pu

nct

atu

sD

RC

Tree

,u

pto

22(–

35)

mta

ll;

bole

up

to30

cmin

diam

eter

Gab

onan

dD

RC

Hu

mid

and

subh

um

idfo

rest

;u

pto

800

m.a

.s.l.

Fru

itco

nsi

stin

gof

foll

icle

s,ov

oid,

6¥

3.5

cm;

3–7-

seed

ed

See

dspl

ano-

con

vex,

~0.2

¥0.

15cm

Un

know

n

Mal

vace

aeH

ibis

cus

sp.

S.h

ampe

iD

RC

Spe

cifi

cin

form

atio

nn

otav

aila

ble

4 F. E. VEGA ET AL.


Mal

vace

aeT

heo

brom

aca

cao

L.

S.h

ampe

iD

RC

Tree

,4–

20m

tall

Nat

ive

toTr

opic

alS

outh

Am

eric

a;cu

ltiv

ated

thro

ugh

out

Trop

ical

Afr

ica

Cu

ltiv

ated

un

der

vari

ous

con

diti

ons

Ber

ry-l

ike

dru

pe,

com

mon

lyca

lled

apo

d,gl

obos

eto

cyli

ndr

ical

,10

–32

¥6–

12cm

,20

–60-

seed

ed,

seed

sem

bedd

edin

mu

cila

geou

s,w

hit

ish

,su

gary

,ac

idic

pulp

See

dsgl

obos

eto

elli

psoi

d,2–

4¥

1–2

cm

See

ds:

~50%

fat

Mel

iace

aeTr

ich

ilia

gilg

ian

aH

arm

sS

.ham

pei

DR

CTr

ee,

up

to30

mta

ll;

bole

up

to10

0cm

indi

amet

er

Trop

ical

Wes

tan

dC

entr

alA

fric

aH

um

idan

dsu

bhu

mid

fore

st;

up

to95

0m

.a.s

.l.

Cap

sule

,de

his

cen

t,pe

ar-s

hap

ed(o

bovo

id)

tofi

g-sh

aped

,3–

3.5

cmin

diam

eter

;u

pto

6-se

eded

See

ds,

±an

gula

rel

lips

oid,

~2¥

1.2

cm

See

ds:

repo

rted

as‘o

ily’

Myr

isti

cace

aeP

ycn

anth

us

ango

len

sis

(Wel

w).

War

b.

S.h

ampe

iD

RC

Tree

,25

–35(

–40)

mta

ll;

bole

120(

–150

)cm

indi

amet

er

Trop

ical

Wes

tan

dC

entr

alA

fric

aH

um

idan

dsu

bhu

mid

fore

st;

1200

(–14

00)

m.a

.s.l.

Dru

pe,

spli

ttin

glo

ngi

tudi

nal

lyw

ith

2va

lves

,el

lips

oid

toob

lon

gor

glob

ose,

3–4.

5¥

2–4

cm,

1-se

eded

See

del

lips

oid,

1.5–

3¥

1–1.

5cm

See

ds:

45–7

0%fa

t

Rh

izop

hor

acea

eA

nop

yxis

klai

nea

na

(Pie

rre)

Pie

rre

exE

ngl

.

S.p

un

ctat

us

DR

CTr

ee,

~45

mta

ll;

bole

~120

cmin

diam

eter

Trop

ical

Wes

tan

dC

entr

alA

fric

aH

um

idfo

rest

;20

0–50

0m

.a.s

.l.C

apsu

le,

5-va

lved

,ob

ovoi

d,2.

6–3.

2¥

2–2.

4cm

;se

vera

lse

eded

See

ds±

elli

psoi

d,fl

aten

don

one

side

,w

inge

dat

apex

,0.

8–1

¥0.

5cm

Un

know

n

Ros

acea

eR

ubu

ssp

.S

.ham

pei,

S.p

un

ctat

us

Tan

zan

iaS

peci

fic

info

rmat

ion

not

avai

labl

eR

ubi

acea

eC

offe

asp

.,C

.ar

abic

a,C

.ca

nep

hor

a,C

.ex

cels

a(C

.li

beri

cava

r.d

ewev

rei

(De

Wil

d&

T.D

ura

nd)

Leb

run

),C

.li

beri

caB

ull

.ex

Hie

rn

S.h

ampe

i,S

.pu

nct

atu

sD

RC

,U

gan

da,

An

gola

,Ta

nza

nia

,L

iber

ia,

Ivor

yC

oast

Tree

let,

smal

ltr

eeor

bush

(in

cult

ivat

ion

),3–

10m

tall

Trop

ical

Wes

tan

dC

entr

alA

fric

a;cu

ltiv

ated

thro

ugh

out

Trop

ical

Afr

ica

Hu

mid

tosu

bhu

mid

fore

st;

0–22

00m

.a.s

.l.;

cult

ivat

edu

nde

rva

riou

sco

ndi

tion

sin

hu

mid

and

subh

um

idre

gion

s

Dru

pe,

wit

h(1

–)2

pyre

nes

,ea

chpy

ren

e1-

seed

ed,

oblo

ng-

elli

poso

id,

0.9–

2.2

¥(0

.6–)

0.9–

1.6

cm

See

ds,

hem

i-el

lips

oid,

(0.7

–)1.

3–1.

6¥

0.4–

0.7

cm

See

ds(e

ndo

sper

m):

7–17

%fa

t

Ru

biac

eae

Nau

clea

did

erri

chii

(De

Wil

d.)

Mer

r.

S.h

ampe

iD

RC

Tree

,30

–40(

–50)

mta

ll;

bole

90–1

50cm

indi

amet

er

Trop

ical

Wes

tan

dC

entr

alA

fric

aH

um

idto

subh

um

idfo

rest

;0–

500

m.a

.s.l.

Syn

carp

,fl

esh

y,gl

obos

e,1.

8–2.

5cm

indi

amet

er;

man

yse

eded

See

ds±

elli

psoi

d,~1

.2¥

0.8

mm

Un

know

n

Ru

biac

eae

Oxy

anth

us

sp.

S.h

ampe

iU

gan

daS

hru

bsan

dsm

all

tree

s,(1

–)2–

5(–1

2)m

tall

Trop

ical

Afr

ica

Hu

mid

tosu

bhu

mid

fore

st;

0–20

00m

.a.s

.l.

Dru

pe,

sph

eric

alto

ovoi

d,1–

4¥

1–2

cm;

seve

ral

seed

ed

See

ds,

±le

nti

cula

r,5–

11¥

2–6

mm

Un

know

n

DR

C,

Dem

ocra

tic

Rep

ubl

icof

Con

go.



A

B

C

D

Fig

ure

1.S

ome

exam

ples

ofda

mag

edfr

uit

spr

esen

ted

byS

ched

l(1

960)

.A

,da

mag

eca

use

dby

Hyp

oth

enem

us

ham

pei

onC

offe

asp

.B

,da

mag

eca

use

dto

Col

agr

isei

flor

a:th

esm

allh

oles

onth

ele

ftw

ere

mad

eby

H.h

ampe

ian

dth

ela

rge

hol

esw

ere

mad

eby

aC

urc

uli

onid

ae;t

he

fru

iton

righ

tsh

ows

dam

age

bybo

thin

sect

s.C

,Tr

ich

ilia

gilg

ian

ada

mag

edby

H.h

ampe

ian

dP

oeci

lips

sp.,

anot

her

bark

beet

le.

D,

Ple

ioca

rpa

pycn

anth

a:le

ftan

dm

iddl

efr

uit

ssh

owda

mag

eca

use

dby

H.h

ampe

i;ri

ght

fru

itsh

ows

dam

age

cau

sed

bya

Cu

rcu

lion

idae

.

6 F. E. VEGA ET AL.


develop independently from coffee plantations’, thusimplying that the insect could possibly complete itslife cycle in these plants (i.e. they could be alternativehost plants). Furthermore, Schedl (1960: 13) alsostates:

Comme le caféier est également cultivé en dehors de la forêtombrophile typique, il y a lieu de considérer que dans cesautres situations, S. hampei Ferr. possède également deshôtes naturels à partir desquels il infeste les plantations decaféiers, même lorsque la lutte menée contre lui est intense.[As coffee is also cultivated outside of the typical rainforest,it should be considered that in these other situations,S. hampei Ferr. also has natural hosts from which it infestsplantations of coffee, even when the struggle against it isintense.]

This indicates that in areas where there is coffeeoutside the vicinity of a rainforest (more correctlytermed humid, evergreen forest), it is possible thatother plants might serve as hosts for the insects.

The article by Schedl (1960) is also significantbecause it suggested that H. hampei could bepolyphagous (‘generalists that exploit plants in morethan one family’; Price et al., 2011). In his landmarkbook on coffee pests, Le Pelley (1968) discussedH. hampei host plants, and after mentioning severalreports of the insect having been found in plantsother than Coffea, he concludes that ‘In none of theabove cases was breeding found in the plants andthey can be considered as occasional food plants only;some of the records may even be erroneous throughmisidentification of the insect found in them’ (p. 118).Interestingly, even though Le Pelley (1968) citesSchedl (1960), he does not include the host plantslisted in that publication. It should be emphasizedthat Schedl’s expertise in bark beetle taxonomymakes it highly unlikely that he would have beenmistaken in identifying H. hampei, and therefore hisidentifications must be considered reliable based onsubsequent work by Wood & Bright (1987, 1992) andWood (2007). The only caveat is that as mentionedabove, what he classified as S. punctatus was laterre-classified as H. hampei by Wood (1972).

Sixteen of the 21 genera listed in Table 1 werereported by Waller, Bigger & Hillocks (2007), based ona subsequent article by Schedl (1961). Waller et al.(2007), in reference to the list of plants they compiled,wrote: ‘. . . it has been assumed by many authors(e.g. Le Pelley, 1968; Hill, 1975) that these are eitherexploratory attacks by the beetle on plants in which itcannot breed or that the beetle has been confusedwith other, similar species of Scolytid’. This meansthat Schedl’s (1960) article was unknown to Walleret al. (2007), as it is not cited by them, and further-more, as mentioned above, their assumption thatthese beetles had been confused with similar species

is unlikely given that Schedl was an expert on theidentification of H. hampei. It is clear then thatSchedl (1960) reveals important information on pos-sible alternative host plants for H. hampei, for whichfield confirmation is now urgently required. At thispoint, it is important to observe that Schedl (1960: 14)mistakenly reported that Roepke (1919) reported hostplants for H. hampei in Indonesia. Roepke (1919: 21–22) actually reported that none of the beetles boredinto the branches of the plant species tested, and thatall of H. hampei had died within 2–3 days in thevarious seeds tested.

The third significant article reporting alternativehost plants for H. hampei was published by Ghes-quière (1933). In this article, he stated:

Le Stephanoderes hampei Ferr. est certainement, au point devue économique, le plus dangereux, il a toujours été considéré,jusqu’a présent, comme un sténomère spécifique des Coffea.Il y a une vingtaine d’années cependant, notre collègueR. Mayné signalait que cet insecte fréquentait, outre le Coffearobusta son hôte habituel, des Hibiscus et diverses Légu-mineuses. Cette assertion mise en doute par d’autres ento-mologistes, est actuellement confirmée par les résultats de nosrecherches sur Cacaoyer, Caesalpinia pulcherrima, et Dialiumlacourtianum. Les récoltes de larves, nymphes et adultes surce dernier, surtout, ne peuvent laisser subsister aucun doute,cette Légumineuse croissait, par bosquets, dans les savanesboisées du Kasai, loin de toute plantation et de galerieforestière quelconque. L’utilisation du D. lacourtianum commeplante-piège serait sans doute à préconiser, il murit ses fruitstardivement en saison sèche, à une époque différente desCaféiers et fournirait en même temps on ombrage léger néces-saire aux plantations équatoriales. [Stephanoderes hampeiFerr. is certainly, from an economic point of view, the mostdangerous, and it has always been considered, until now, likea specific stenomere (old term used in classification ofColeoptera) of Coffea. It has been 20 years since our colleagueR. Mayné (see Mayné, 1914) noted that this insect frequented,in addition to its habitual host Coffea robusta (valid name:C. canephora), Hibiscus and various Leguminosae. This asser-tion, doubted by other entomologists (Ghesquière, (1933),includes a footnote to specify that these doubts by otherentomologists are based on possible mistakes in taxonomicdetermination), is currently supported by the results of ourresearch on cocoa, Caesalpinia pulcherrima, and Dialiumlacourtianum (valid name: D. englerianum). The collection oflarvae, pupae and adults on the latter, above all, can leave nodoubt, that this Leguminosae growing in groves in the wood-lands of the Kasai, away from all plantations and any galleryforest. (Our interpretation of this apparently incomplete sen-tence is that all stages of the insect were found on this plantaway from coffee plantations and the forest, thus indicatingthat the insect does not need coffee or other forest plants tosurvive.) The use of D. lacourtianum as a trap plant should beadvocated because its fruits ripen late in the dry, at a timedifferent from coffee, while providing at the same time thelight shade needed in equatorial plantations.]



Ghesquière’s (1933) article confirms the presence ofother plants on which H. hampei can reproduce.

Habit, distribution, and ecology of the plants listedin Table 1, as well as information of the fruit andseeds, were obtained from the herbarium at the RoyalBotanic Gardens, Kew, and from various literaturesources (Burkill, 1985, 1994, 1997; PROTA, 2012).

CONCLUSIONS

The articles by Beille (1925), Ghesquière (1933), andparticularly Schedl (1960), raise important issues,which if resolved, would greatly expand our knowl-edge of H. hampei. For example, were wild Coffeaspecies the original host plants for the insect in addi-tion to those plants reported by Schedl (1960), and didforest disturbances push the insect towards areaswhere food resources, in the form of C. canephoraplantations, were readily available, resulting in peststatus, as implied by Beille (1925)? To our knowledge,H. hampei is the only insect pest that completes itslife cycle in Coffea seeds, even though eight additionalHypothenemus species have been collected fromCoffea plants: Hypothenemus areccae (Hornung),Hypothenemus crudiae (Panzer), Hypothenemuseruditus Westwood, Hypothenemus grandis Schedl,Hypothenemus liberiensis (Hopkins), Hypothenemusplumeriae (Nordlinger), Hypothenemus seriatus(Eichoff), and Hypothenemus solitarius (Schedl) (F.E. Vega, unpubl. data). Nevertheless, the use of coffeeseeds as food by H. hampei requires biochemicalmechanisms to break down the alkaloid caffeine,known to be toxic to other insects (see Vega et al.,2003).

Our assessment of the species listed by Schedl(1960) shows that many of the proposed alternativehosts for H. hampei have similar biological charac-teristics. The first of these, and one that immedi-ately appears at odds with the Coffea host, is theheight of the plant species listed in Table 1. Fifteenof the 20 (75%) species (i.e. excluding Coffea) canattain a height greater than 15 m; and 13 (65%) aretrees of more than 25 m in height, with severalof these attaining heights between 30 and 45 m(Table 1). Thus, this list includes many canopy treesand canopy emergents. A canopy existence has notpreviously been suggested for H. hampei. Coffeaspecies rarely attain heights of greater than 10 m.Secondly, a large number of the species listedpossess seeds with a high oil/fat content. Eight of the20 species (40%; excluding Coffea, and including twogenera without specific information) are reported tohave seeds with high oil or fat content. This per-centage could be considerably larger, as the oil/fatcontent of many of the listed species are unknown tous. Significantly, the seeds (endosperm) of Coffea are

also rich in fats, having a fat content between 7 and17% (Speer & Kölling-Speer, 2006). This relationshipis, however, by no means universal for the specieslisted by Schedl (1960), as some species, notablyPhaseolus lunatus, are low in oil/fat; we also have aconsiderable quantity of missing data for seed com-position. Thirdly, large seed size figures prominentlyin the list. Thirteen of the 20 species (65%; excludingCoffea, and including two genera without specificinformation) have seeds with at least one dimension(e.g. length or width) greater than 1 cm. For Coffeaspecies reported as hosts of H. hampei (Table 1), theseeds are hemi-ovoid, and usually between 0.9 and1.4 cm long. In summary, a high proportion of thealternative hosts reported by Schedl (1960) are sub-stantial rainforest trees, and the majority havelarge, nutritious seeds capable of providing a worth-while food source for H. hampei. It is also evidentfrom Table 1 that there are three species listed ascultivated and introduced from South America[Caesalpinia pulcherrima (L.) Sw., Phaseolus luna-tus L., and Theobroma cacao L.]. It is clear that in atleast the latter two species (lima bean and cocoa,respectively), H. hampei is not a significant pest, asit would have been widely reported as such for thesesignificant crop species.

Another interesting issue that would need to beelucidated is whether a field survey would reveal theinsect in Dialium englerianum, and if so, whetherthe plant can be used as a trap, as proposed byGhesquière (1933). Finally, another important impli-cation of finding the alternative host plants forH. hampei is that these plants might reveal previ-ously unreported biological control agents (i.e. preda-tors, parasitoids, pathogens) that might be moreeffective in controlling the insect than thosecurrently being used.

Based on the findings presented here, the pursuitof H. hampei in the field in the alternative hostplants listed in Table 1 is being planned, under theleadership of J. Jaramillo, as part of a project fundedby the German Research Foundation (Deutsche For-schungsgemeinschaft). Once these alternative hostplants are located it will be essential to determine ifH. hampei can be found naturally infecting and com-pleting its life cycle in the fruits produced by theseplants. This will be the only way to ascertainwhether these plants do indeed serve as host plantsfor H. hampei, and not as exploratory or temporaryhosts in which reproduction and development doesnot occur. It will also be important to determinewhether H. hampei insects taken from these alter-native plants can complete their development incoffee berries: this information will provide insightson how the insect might have become adapted tothrive on Coffea fruits.

8 F. E. VEGA ET AL.


ACKNOWLEDGEMENTS

We thank Harry K. Kaya (University of California,Davis) and three anonymous reviewers for theirhelpful comments. We give special thanks to LulúRico-Arce, Legume Specialist at the Royal BotanicGardens, for providing seed information on the Legu-minosae presented in Table 1.

REFERENCES

Beille L. 1925. Les Stephanoderes sur les Caféiers cultivésa la Côte d’Ivoire. Revue de Botanique Appliquée etd’Agriculture Tropicale 5: 387–388.

Bright DE, Skidmore RE. 1997. A catalog of Scolytidaeand Platypodidae (Coleoptera), Supplement 1 (1990–1994).Ottawa: NRC Research Press.

Bright DE, Skidmore RE. 2002. A catalog of Scolytidaeand Platypodidae (Coleoptera), Supplement 2 (1995–1999).Ottawa: NRC Research Press.

Burkill HM. 1985. The Useful Plants of West Tropical Africa– Volume 1. Families A–D. Kew: Royal Botanic Gardens.

Burkill HM. 1994. The Useful Plants of West Tropical Africa– Volume 3. Families J–L. Kew: Royal Botanic Gardens.

Burkill HM. 1997. The Useful Plants of West Tropical Africa– Volume 4. Families M–R. Kew: Royal Botanic Gardens.

Davis AP, Govaerts R, Bridson DM, Stoffelen P. 2006.An annotated taxonomic conspectus of the genus Coffea(Rubiaceae). Botanical Journal of the Linnean Society 152:465–512.

Ferrari JAG. 1867. Die Forst- und BaumzuchtschädlichenBorkenkäfer (Tomicides Lac.) aus der Familie der Holzver-derber (Scolytides Lac.), mit besonderer Berücksichtigungvorzüglich der europäischen Formen, und der Sammlungdes k. k. zoologischen Kabinetes in Wien. Wien: Carl Gerold’sSohn.

Fleutiaux E. 1901. Un ennemi du café du Ronilon (Congo).La Nature – Revue des sciences et de leur application à l’artet à l’industrie 29: 4.

Gauthier N. 2010. Multiple cryptic genetic units inHypothenemus hampei (Coleoptera: Scolytinae): evidencefrom microsatellite and mitochondrial DNA sequence data.Biological Journal of the Linnean Society 101: 113–129.

Ghesquière J. 1933. Rôle des Ipides dans la destructiondes végétaux au Congo belge. Annales de Gembloux 39:24–37.

Hill DS. 1975. Agricultural insect pests of the tropics and theircontrol. London: Cambridge University Press.

Hopkins AD. 1915. Classification of the Cryphalinæ, withdescriptions of new genera and species. United StatesDepartment of Agriculture, Contributions from the Bureau ofEntomology, Report No. 99.

Jaramillo J, Borgemeister C, Baker P. 2006. Coffee berryborer Hypothenemus hampei (Coleoptera: Curculionidae):

searching for sustainable control strategies. Bulletin ofEntomological Research 96: 223–233.

Le Pelley RH. 1968. Pests of coffee. London: Longmans,Green and Co. Ltd.

Mayné R. 1914. Travau de l’Entomologiste de la Colonie.Bulletin Agricole du Congo Belge 5: 577–600.

Plant Resources of Tropical Africa (PROTA). 2012.Available at: http://database.prota.org/search.htm (accessed5 June 2012).

Price PW, Denno RF, Eubanks MD, Finke DL, Kaplan I.2011. Insect ecology: behavior, population and communities.New York: Cambridge University Press.

Roepke W. 1919. Gegevens omtrent de Koffiebessen-boeboek(Stephanoderes hampeï Ferr. = coffeae Hgd.). Mededeelingenvan het Instituut voor Plantenziekten 38: 1–32.

Schedl KE. 1959. A check list of the Scolytidae and Platypo-didae (Coleoptera) of Ceylon with descriptions of newspecies and biological notes. Transactions of the RoyalEntomological Society of London 111: 469–534.

Schedl KE. 1960. Insectes nuisibles aux fruits et aux graines.Publications de l’institut national pour l’étude agronomiquedu Congo Belge, Série Scientifique 82: 1–133.

Schedl KE. 1961. Scolytidae und Platypodidae Afrikas.Band 1. Unterfamilie Hylesinae. Revista de Entomologia deMoçambique 4: 335–742.

Speer K, Kölling-Speer I. 2006. The lipid fraction of the coffeebean. Brazilian Journal of Plant Physiology 18: 201–216.

Vega FE, Blackburn MB, Kurtzman CP, Dowd PF. 2003.Identification of a coffee berry borer-associated yeast: does itbreak down caffeine? Entomologia Experimentalis et Appli-cata 107: 19–24.

Vega FE, Franqui RA, Benavides P. 2002. The presence ofthe coffee berry borer, Hypothenemus hampei, in PuertoRico: fact or fiction? Journal of Insect Science 2: 13.

Vega FE, Infante F, Castillo A, Jaramillo J. 2009.The coffee berry borer, Hypothenemus hampei (Ferrari)(Coleoptera: Curculionidae): a short review, with recentfindings and future research directions. Terrestrial Arthro-pod Reviews 2: 129–147.

Waller JM, Bigger M, Hillocks RJ. 2007. Coffeepests, diseases and their management. Wallingford: CABIPublishing.

Wood SL. 1972. New synonymy in the bark beetle tribeCryphalini (Coleoptera: Scolytidae). Great Basin Naturalist32: 40–54.

Wood SL. 2007. Bark and ambrosia beetles of South America(Coleoptera, Scolytidae). Provo, UT: Brigham Young Univer-sity.

Wood SL, Bright DE. 1987. A catalog of Scolytidae andPlatypodidae (Coleoptera), Part 1: Bibliography. GreatBasin Naturalist Memoirs 11: 1–685.

Wood SL, Bright DE. 1992. A catalog of Scolytidae andPlatypodidae (Coleoptera), Part 2: Taxonomic Index, VolumeB. Great Basin Naturalist Memoirs 13: 835–1553.



Documents

From forest to plantation? Obscure articles reveal alternative host plants for the coffee berry borer, Hypothenemus hampei (Coleoptera: Curculionidae)