Journalof TropicalEcology(2005) 21:493-500. Copyright? 2005 CambridgeUniversity Press Printed in the United Kingdom doi:10.1017/S0266467405002506

ofHyperiodrilus africanus Population dynamics Eudrilidae) (Oligochaeta, inIvory Coast andPatrick Tondoh*l Lavellet Jer6meEbagnerin * UFRdesSciencesdelaNature/Centre deRecherche en

23 BP4727Abidjan 23,C6ted'lvoire Ecologie, d'Abobo-Adjame, Universite desSolsTropicaux, Paris avenue 6, 32, Henri 93143 Cedex, t Laboratoire France d'Ecologie Bondy Varagnat, IRD/Universite I February (Accepted 2005)

Abstract:Thepopulationdynamicsof the exoticearthwormHyperiodrilus was investigatedin a secondary africanus forestof the NaturalReserveof Lamto(IvoryCoast)overa periodof 19 mo. The objectiveswereto assessseasonal abundancepatternsand to determinethe adaptivestrategiesof this species.Eachmonth, 10 soil samplesof 100 x 100 x 40 cm and 20 monolithsamplesof 25 x 25 x 30 cm were randomlyexcavatedin a plot of 50 x 95 m and earthwormswereextractedbybothhand-sortingandwash-sievingmethods.Theresultsshowsignificantinter-annual and seasonalfluctuationsin populationsize. Threefactorsare likelyto controlpopulationdynamics:(1) rainfall, (2) soil watercontentand (3) seasonality.The dry season appearsto be the most importantenvironmentalfactor thatregulatespopulationabundancewhenpredation,density-dependent regulationandcompetitionphenomenaare ignored.Hyperiodrilus africanusexhibitedan r strategy,suggestinga high abilityto recoverpopulationsaffectedby drought. KeyWords:demographicprofile,IvoryCoast,populationdynamics,tropicalearthworms,WestAfrica

INTRODUCTION Soil ecosystem engineers (predominantly termites, ants and earthworms) and roots are involved in the regulation of soil processes since they produce structures including aggregates and pores that influence the structure of soil (Blanchart 1990, Blanchart et al. 1999, Lavelle 2002, Rossi 1998). It is well established that earthworms significantly influence soil fertility via nutrient cycling processes (Blair et al. 1995, Lee 1985, Villenave et al. 1999). However most agricultural practices have detrimental effects on the diversity and/or abundance of earthworm communities (Kouassi 1999, Tondoh 1993). Adequate management of croppingsystems may optimize conditions for the persistence of earthworm communities and thereforeenhance their activity and impact on the soil system (Decdienset al. 1994). The design of such practices requires knowledge of species ecology, especially species that have a large tolerance to perturbations. The understanding of earthworm behaviour in adverse conditions is of great relevance to the design of new

'Corresponding

author. Email:[email protected]

low-input agricultural systems in which the introduction and subsequent maintenance of specificearthworms can improve soil fertility (Lavelleet al. 1987, Swift 1984). Hyperiodrilusafricanus(Beddard 1891) is one of the species best suited for such management options in humid tropical Africa (Lavelle et al. 1999). Populations of this species are found throughout West (IvoryCoast, Nigeria) and CentralAfrica(Congo,DemocraticRepublicof Congo, Angola) both in natural and disturbed areas derived from humid savannas and forests (Madge 1969, Omodeo 1954, Lavelle unpubl. data, Tondoh unpubl. data). This exotic worm is known to be active in the burial and the decomposition of litter in various agro-ecosystems (Hauser 1993, Hauser et al. 1998, Tian 1995). The knowledge of population ecology of this earthworm is limited to one field study undertaken 30 y ago in Nigeria (Madge 1969). More recently, general demographic parameters such as growth rate, generation time and fecundity have been studied under laboratory conditions (Tondoh 1998a, Tondoh & Lavelle 1997). This study led to two main hypotheses to explain the widespread distribution of H. africanus:(1) significant capacity for movement at the soil surface and of colonization of new areas, (2) high demographic profile that allows rapid

This content downloaded on Sun, 10 Mar 2013 18:01:44 PM All use subject to JSTOR Terms and Conditions

494

EBAGNERIN TONDOHAND PATRICKLAVELLE JEROME

growth of a population when environmental conditions are favourable. The aim of this study was to collect field data on demographic parameters (survival rate, natality, fecundity) and to determine key factors affecting population dynamics and the life cycle of H. africanus.

STUDY SITE

Table1. Average density (ind m-2) of Hyperiodrilusafricanusindividuals for differentbiomass classes obtained through hand-sorting and washsieving over a 19-mo period (June 1994 to December 1995). Biomass class (mg) 0-50 50-100

98.8

274.4

36.0

2.8

50.9 28.9 23.7

69.9 80.5 91.9

1.6

100-150

79.7 35.9 25.8

150-200

This investigation was conducted at the Station d'Ecologie de Lamto (5002'W, 6013'N) in the forest-savanna boundary region of Ivory Coast. The research station is surrounded by a 2500-ha natural reserve characterized by a mosaic of grass and shrub savannas and gallery forests. The mean annual temperature over 10 y (19861996) was 28.4? C and the average monthly rainfall ranged from 8.4 mm in January to 189.7 mm in June with an annual total of 1138.1 mm. The climate is characterized by two dry seasons: from November to March and from July to August. The study site is located in a portion of shrub savanna protected from fire for 32 y which has turned into a forest succession characterized by a significant growth of the tropical shrub Chromolaenaodoratato which H. africanus is uniquely associated (P. Lavelle unpubl. data). The soil is classifiedas a sandy ferralsolwith 75%sand, 13.8% silt and 11.2% clay (FAO-UNESCO 1989).

METHODS Samplingprocedure: samplingpointsandsamplingarea One hundred and ninety quadrats (5 x 5 m), regularly arranged on a grid of 50 x 95 m within the protected savanna area were sampled fromJune 1994 to December 1995. Each month, ten replicate quadrats were selected randomly from the available array. The sampling unit was a monolith of 100 x 100 x 40 cm excavated from each quadrat. Earthworms were separately collected from each 10-cm layer by hand-sorting and fixed in 4% formaldehyde solution for species identification. In order to correct for errors due to undersampling of small individuals, earthworms were also collected from each 10-cm layer of two additional small monoliths (25 x 25 x 30cm) adjacent to the large 1-m2 samples using a wash-sieving techniques (Lavelle 1978). Individuals of H. africanuswere separated into four age classes (cocoons, juveniles, immatures and adults) on the basis of cocoon identification, clitellum development and weight (juveniles: 2-48 mg, Tondoh 1998a). Cocoons and individuals were counted and weighed separately. Ten soil samples were taken from each 1 x 1-m monolith at 0-10 cm and 10-20 cm depth to estimate

Hand-sorting Wash-sieving WS/HS Correctionindex (HS) (WS) (%) (N = WS/HS)

1.2 1.1

the soil water content expressed as per cent dry weight at 105 oC. The temperature of the plot was measured using thermometerslocated at the four corners in the top 10 cm of soil.

Dataprocessing Population density. The correctionindex N (Table1) applied to the density of individuals in each weight class was estimated using the equation: N = WS/HS (Lavelle 1978) where: WS is the density of populations of a given biomass class collected by wash-sieving; HS is the density of populations of the same size class hand-sorted. There was a significant correlation (r=0.99, P= 0.0196, n=4) between the efficiency of hand-sorting (WS/HS) and the size of the worms. Hand-sorting efficiency for juveniles (0-50 mg) was the lowest (36%) while the recovery efficiency for individuals weighing 200-250 mg or more was beyond 100%. Cocoondensity. To adjust for errors in the estimation of numbers of cocoons regardless of their weight because of insignificant variation, the monthly hand-sorted number was multiplied by the mean of the correction indices (NC= 4.3) calculated over 19 mo.

Estimating monthlyadultfecundity.Asuming that the cocoons collected (C)were laid by the hand-sorted adults (A), fecundity (F) was estimated as the ratio between cocoons and adults at a given date, C/A.

the fixedbiomassof individuals.One hundred Correcting H. africanus collected from the field and fixed in 4% formaldehydesolution for several days lost 12.9% of their initial biomass. The average correcting index for each individual fixed without weighing is then estimated at 1.15. This index was applied to the fixed biomass to estimate the fresh one.

This content downloaded on Sun, 10 Mar 2013 18:01:44 PM All use subject to JSTOR Terms and Conditions

Populationdynamicsof Hyperiodrilusafricanus (Oligochaeta,Eudrilidae)in IvoryCoast Table2. Survival rate (%mo-') of different developmental stages of a Hyperiodrilusafricanuspopulation. Cocoon

Juvenile

Immature

December 1994

31.1

50.0

37.7 66.8

49.7 28.8

33.2

January 1995 December 1995

39.8

46.6

59.9

37.0 0.0

Adult

the survivalrateof individuals.In earthworm Estimating populations, survival rate of individuals is difficultto estimate without any specific information on reproduction in the field. A common assumption considers the dry season characterized by quiescence as associated with mortality in earthworm populations (Lavelle 1978, Tondoh 1998b). Therefore, assuming that drought was the main mortality factor in H. africanuspopulations, the survival rate was estimated for each weight class as the ratio between population density before and during the dry season. The decrease in population size during the dry season enabled us to estimate the survival rate of individuals. Once the peak before the long dry season is identified, the survival rate was calculated from the number of individuals in the previous month. Survival rateofcocoons(so). Ifwe assume that there were no cocoons laid during the quiescence period and predation has no consistent effect, a reasonable estimate of survival can be obtained by writing: * December1994 Number in November 1994: 51.2 cocoons m-2 Number in December 1994: 15.9 cocoons m-2 so = 15.9 x 100/51.2 so = 31.1% * January1995 Number in December 1994: 15.9 cocoons m-2 Number in January 1995: 6 cocoons m-2 so= 6 x 100/15.9 so = 37.7% * December1995 Number in November 1995: 36.1 cocoons m-2 Number in December 1995: 24.1 cocoons m-2 so = 24.1 x 100/36.1 so = 66.8% The survival rate of other developmental stages was estimated in December 1994, January 1995 and December 1995 (Table 2) using the same procedure.

495

stage within this time. The short development time (59 d) between cocoon and adult stages of this worm in laboratory conditions (mean temperature= 28 ?C) does not permit the recognition and tracking of cohorts on a monthly sampling basis and therefore the age-classified structure (Caswell1989, Tondoh 1998a) was considered a better way to study the demography of this worm. Fourclasses of individualswere distinguished in a given population of H. africanus: cocoons: oval shaped weighing 15-58 mg; juveniles: newly hatched individuals weighing 2-49 mg; immatures: individuals larger than 50 mg with no clitellum; adults: clitellate individuals.

profile Demographic Given the difficulty of estimating the intrinsic rate of increase r of the population in field conditions, we found the demographicindex proposedby Lavelle(19 79, 1983) useful for synthesizing the demographic profile (D). This index makes it possible to rank earthworm species along an r-K gradient according to the values of their D index, showing their ability to increase in numbers. This ability itself depends on maximum body weight and the average depth of soil layer in which they are living (Lavelle1983). The demographic profile of earthworm populations may be described by three parameters: adult fecundity (F: number of cocoons produced per adult per year), generation time (T: time of growth until the first cocoon is produced) and the life expectancy of newly hatched worms (Ev)(Lavelle 1979). These parameters were combined in the demographic index D = (F x 103)/(T x Ev).

Statistical analysis The relationships between demographic parameters (fecundity, developed-stages, population activity: rate of active or quiescent individuals) and environmental factors (rainfall, soil moisture, soil temperature) were assessed by performinglinear and non-linear regression, using the software Statistica, 1999. RESULTS

structure population Age-classified

Populationcharacterization

The cohort methodology used by Lavelle (19 78) to study the demography of earthworm populations was not efficient for H. africanus.The sampling interval of 1 mo used in the field investigation was too long and as a result, new individuals could have reached the immature

Densityandbiomass. The population density ofH. africanus exhibited marked annual and seasonal variations. Maximum values of density in 1994 were 90.5 ind m-2 at the end of the long rainy season (July-August) and 130.9 ind m-2 during the short rainy season in October

This content downloaded on Sun, 10 Mar 2013 18:01:44 PM All use subject to JSTOR Terms and Conditions

496

EBAGNERIN TONDOHAND PATRICKLAVELLE JEROME 1994

1995

40

monthly rainfall level (r = 0.84, P = 0.0092, n = 19). Below a monthly rainfall of 25 mm, 75-100% of individuals were inactive.

160 - biomass

---.

30

140

density

120 E

E

100 '.

20-

80

(,

-

ccS.60

10

Verticaldistribution. Most individuals (57-99%; mean monthly percentage) were located in the top 10 cm of the soil in the rainy season while up to 36.3% occurred at 10-20 cm during the dry season. More than 50% of cocoons were laid in the first 10-cm layer except in the dry season when up to 6 7%were found in the 10-20-cm layer from January to February 1995.

"40 20

structure Population JJASONDJFMAM J J ASOND Samplingperiod(month) Figure1. Seasonal fluctuation in mean monthly density (ind m-2 - SE, n = 10) and biomass (g fresh wt m-2 ? SE, n = 10) of Hyperiodrilus africanus.

(Figure 1); the lowest value (39.5 ind m-2) was recorded in December. In 1995, the highest values ranged from 51.9 to 74. 9 ind m-2 during the rainy season (JuneAugust). Biomass ranged from 4.9 g m-2 (December)to 25.3 g m-2 (July)in 1994 and from0.7 g m-2 (December) to 18.9 g m-2 (June)in 1995 (Figure 1). In general, populations were characterized by high abundance and large biomass towards the end of the rainy seasons and by a decrease of these parameters during the dry seasons. This trend was revealed by a significant linear regression (r = 0.60, P = 0.0062, n = 19) between population biomass and rainfall. Most individuals (60-100%) were inactive during the dry season (Figure2). There was a significant logarithmic correlation between the inactivity of individuals and the 1994

250 .4

1995 ,

200-

----

The inter-annual fluctuation of mean population abundance of H. africanuswas markedly characterized by a decrease in the overall abundance from 1994 (115.8 ind m-2) to 1995 (43.1 ind m-2) (Figure 3). The inter-annual fluctuation of cocoons (246 to 176) and juveniles (315 to 178) was marked while the immatures (181 to 161) and adults (69 to 53) fluctuated much less. Cocoon production was characterized by a peak at the end of the major rainy season (June-July) and at the beginning of the dry one (November) (Figure 4a). The number of cocoons laid during the rainy season decreased as the dry season approached. Mean cocoon density ranged from 0 to 66 ind m-2 in 1994 and from 0 to 36 in 1995. Cocoons represented a large proportion of the population in the dry season i.e. 71% in November and 790/, in December 1995. The peaks in juvenile numbers followed those of cocoons in 1994 where juvenile densities fluctuated between 22 and 83 ind m-2 (Figure 4b). In 1995, the 1994

1995 200

0-"

Srainfall

g200

rainfall

-population size

200 -

activeworm

150

150

150--

CY

E 150 4 0

a,

-100 eS 1 10000

100

,--,-•

-

100+ .100

50"

'eS

50 50

50

0

r

.

C 2

0

J JASONDJ FMAMJ J ASSOND Samplingperiod(month) Figure2. Proportion of active Hyperiodrilusafricanusand rainfall during the sampling period.

JJ

ASONmDJF

MAMJJ

ASOND

Sampling period (month)

Figure 3. Seasonal variation in mean monthly population size (ind m-2 ? SE) of Hyperiodrilusafricanus.(Data presented are the sum of earthworm and cocoon number.)

This content downloaded on Sun, 10 Mar 2013 18:01:44 PM All use subject to JSTOR Terms and Conditions

PopulationdynamicsofHyperiodrilusafricanus (Oligochaeta,Eudrilidae)in Ivory Coast 120

E

0

(a) 80-

Cocoons

497

was likely to be related to the accumulation of cocoons that had survived the drought. The annual fecundity rate estimated was more than 28.2 cocoons per adult in 1994 (because the sampling has begun in June) and 29.9 cocoons per adult in 1995.

JJASONDJFMAMJJASOND 120 ,S E

(b)

80

S40Juveniles c_ O o

J J ASONDJ

FMAMJ

J ASOND

60 E

S

M (c) 40

200-

C

Immatures

I

J JASONDJFMAM

J J ASOND

60 E

(d)

40 20 0

Adults.

JJASONDJFMAMJJASOND 1994

1995

Samplingperiod(month)

Factorsaffectingpopulation dynamics Soil water content was the major abiotic factor affecting population dynamics because it was significantly related n=19) to individual mean (r=0.78, P=0.0002, biomass. The seasonal pattern of cocoon production, juvenile emergence, adult occurrence and the global activity of the population showed that the life cycle and population dynamics of H. africanusare clearly under the control of the seasonal cycle. During the long rainy season characterized by higher soil water content, populations were dominated by adult individuals as a consequence of a rapid growth of the juveniles. The short wet and dry seasons were characterized by a large number of cocoons hatching and a high proportion of immatures and juveniles in populations. At the beginning of the following long dry season, the fecundity was high and cocoons accounted for 31-75% of the population.

Figure4. Dynamicsof Hyperiodrilus africanuspopulationstructure (ind m-2 ? SE)during the sampling period.

peak of juveniles occurred in the long rainy season and ranged from 3-75 ind m-2. Fluctuations were largely due to the high sensitivity to drought of this age class. The variation in number of immatures was less marked. Their densities ranged from 13-43 indm-2 in 1994 and fluctuated from 3-24 ind m-2 in the following year (Figure 4c). Clitellate individuals displayed the lowest density despite their abundance during the rainy season (June, October, July): 2.7-23indm-2 in 1994; 0.712 ind m-2 in 1995 (Figure 4d). They represented 221% of the overall population throughout the sampling period.

Reproduction Mating periods were identified by following fluctuations in adult and cocoon densities. Fecundity was high at the end of the long rainy season and at the beginning of the long dry season. The values ranged from 0 (September) to 9.5 cocoons per adult (November) in 1994; and from 0 (September)to 14.7 cocoons per adult (November) in 1995. In most cases, the peaks of fecundity corresponded to the period of high density of juveniles. The peak in cocoon production in February (13.5 cocoons per adult)

profile Demographic The demographic index D = (F x 103)/(C x Ev) was calculated, using the following data: F: annual fecundity per adult in 1995 is 29.9 cocoons per adult y-1 T (month):the generation time in the fieldis 2 mo (Tondoh 1998a) Ev(month): life expectancy at hatching. Lifeexpectancy was estimated by following the dynamics of cocoons, juveniles and adults, taking into account the length of the growth period. Assuming that AugustSeptember is a hatching period because of the peak in juvenile densities, the mean life expectancy of an individual was estimated at 5 mo, in other words, from August to December. The demographic index was calculated as follow: D=(29.9 x 103)/(2 x 5)=2990. DISCUSSION Populationcharacterization The fluctuation of H. africanus natural populations is influenced by the seasonal changes of soil water

This content downloaded on Sun, 10 Mar 2013 18:01:44 PM All use subject to JSTOR Terms and Conditions

498

EBAGNERIN TONDOHAND PATRICKLAVELLE JEROME

availability. Two contrasting periods can be recognized, i.e. a rapid population growth during the rainy season (from January to June-July) followed by a decrease during the dry season (from October to December). The annual changes in population dynamics may be due to fluctuations in rainfall patterns between 1994 and 1995 or biotic factors such as predation, interspecific competition since up to nine earthworm species were identifiedin this study. Biotic factors are known to play a crucial role in natural population dynamics although they are difficultto estimate under field conditions (Hellriegel

When the soil is wet, large individuals (1100 mg) deposit small numbers of large cocoons. The beginning of the dry season is considered as the main factor determining the occurrence of adult reproduction. It causes a large deposition of cocoons that is thought to maximize the chance of survival of populations after the dry period. Cocoons, juveniles, immatures and adults that survive during the drought period may continue their active life in March during the early rainy season. Demographic patterns of H. africanusin Ivory Coast are, to some extent, similar to the observations made in Nigeria (Madge 1969). There are two generations per year: the firstpeaks in June-July,followedby the second in September-October.Consequently, one can assume that the longevity of H. africanusis about 5 mo. Compared with other Lamto earthworm species (1.5-5 y, Lavelle 19 78) and Lumbricusterrestris(6-7 y) (Lakhani&Satchell 19 70), H. africanusis a short-lived species. Juveniles and adults are likely to be the resistant stages of the population ensuring the survival of H. africanuspopulations.

2000). This simple model of earthworm population dynamics may be modifiedby some seasonal perturbations,leading to the appearance of one or several peaks. Other earthworm populations in the Lamto savannas are also markedly under the influence of soil moisture content (Lavelle 1978). Similar patterns of population dynamics were recorded for Indian earthworms (Lampitomauritii, Drawida willsii and Octochaetonasurensis) commonly found in pastures, crop-fieldsand in compost pits (Dash & Senapati 1980). Bennour & Nair (1997) found the same significant relationship between soil water content and the abundance of the endogeic earthworm Apporrectodea caliginosain north-east Libya. In the same way, highest values of biomass in the tropical earthworm Martiodrilus carimaguensisappear at the beginning of the wet season when all the population is supposed to be active after several months of diapause (Jimenez et al. 1998). Both vertical distributionand the dependence on coarse particulate organic matter indicate that H. africanusis an epi-endogeic earthworm species. However, individuals can migrate down to 10-20 cm depth in soil in the early dry season to lay cocoons. This strategy is common in other earthworms like M. anomala at Lamto (Lavelle 1971), A. caliginosain Libya (Bennour & Nair 1997), Lampitomauritii,Drawidawillsii and Octochaetonasurensis in India (Dash & Senapati 1980).

Agestructure The age-classifiedstructure of the H. africanuspopulation is highly sensitive to both annual and seasonal variations. The high proportionof young individuals was followed by that of cocoons and adults. Similarpatterns of population structure have been reported for Lumbricus rubellus and Dendrobaenarubida (Rbmbke 1987), A. caliginosa (Bennour & Nair 1997, Fraser et al. 1996) and Lamto earthworm species (Lavelle 1978). Despite the lack of cocoons in September,H. africanusreproduction is likely to be continuous when compared to Millsonialamtoiana that lays cocoon from April to June (Lavelle 1978). Cocoon production is observed in the phases of rapid growth and appearsto be induced by the onset of drought.

Factorsaffectingpopulation dynamics Our study showed that under field conditions, rainfall and soil water content significantly influence population dynamics of H. africanus. A similar conclusion was reached by Whalen et al. (1998) concerning the seasonal fluctuations in earthworm density and biomass in corn agrosystems in the USA which received organic or inorganic fertilizeramendments. However, the weakness of our study results in the ignorance of biotic factors such as predation by driver ants (Tondoh unpubl. data) or density-dependent regulation in earthworm populations recently highlighted by Kammenga et al. (2003). Density-dependent effects in H. africanusgrowth and cocoon production have been reported under laboratory conditions (Tondoh 1998a). Given the results obtained in the temperate opportunistic earthworm Eisenia fetida pointing out the significant influence of density dependence (Kammenga et al. 2003) on population dynamics, one can assume that H. africanus may be subject to a density-dependent regulation.

Demographicprofileand adaptivestrategies The high value of the demographic index is probably due to a high ability of H. africanusto colonize newly available habitat. This demographic index is far higher than that of native species of Lamto (Lavelle 1979): Dichogaster agilis (200), Chuniodriluszielae (200), M. (50), M. lamtoiana(17), Dichogasterterrae-nigrae anomnala (4.5), Agastrodrilusopisthogynus(4.9), M. ghanensis(2.9).

This content downloaded on Sun, 10 Mar 2013 18:01:44 PM All use subject to JSTOR Terms and Conditions

PopulationdynamicsofHyperiodrilusafricanus (Oligochaeta,Eudrilidae)in IvoryCoast The ability of H. africanusto colonize a wide range of humid tropical soils is probablydue to three main factors: (1) relatively small size (2-15 cm) associated with a high reproductive potential; (2) rich diet of organic residues, (3) high fecundity. The adaptation of H. africanusto the environment is accomplished through different strategies: demographic (deposition of cocoons almost throughout the year) and high population expansion ability: D=2990; behavioural (migration to 10-20 cm layer of soil during the drought) and physiological (quiescence: 60-100% of individuals were inactive during the dry season). Quiescence is an efficient strategy since it stops the life cycle when the soil is dry and the whole life cycle is resumed as soon as suitable growth conditions arise. The main ecological constraint for individuals is drought. On the basis of previous observations, H. africanus exhibits an r strategy. This worm is sensitive to environmental fluctuations and displays a high ability for expansion. This suggests a high ability of renewal or reconstitution of populations affectedby drought.

ACKNOWLEDGEMENTS This work was supported by a research grant from the European Community through the Macrofauna project (STD3).The authors would like to thank Raphael Zouzou Bi Danko and Yao N'goran for their field assistance. They are grateful to Dr David Bignell for kindly reviewing the manuscript.

LITERATURE CITED

499

DECAENS, T., LAVELLE,P., JIMENEZ,J. J., ESCOBAR, G. & RIPPSTEIN.G. 1994. Impact of land management on soil macrofauna in the Oriental Llanos of Colombia. EuropeanJournalof Soil Biology 30:157-168. FAO-UNESCO 1989. Cartemondialedes sols. Legende revis6e. 125 pp. P. M., WILLIAMS,P. H. & HAYNES,R. J. 1996. Earthworm FRASER, species, population size and biomass under differentcroppingsystems across the CanterburyPlains, New Zealand.AppliedSoilEcology3:4957. HAUSER. S. 1993. Distribution and activity of earthworms and contribution to nutrient recycling in alley cropping. Biology and Fertilityof Soils 15:16-20. HAUSER,S., ASAWALAM,D. O. & VANLAUWE,B. 1998. Spatial and temporal gradient of earthworm casting activity in alley cropping systems. AgroforestrySystems41:12 7-137. HELLRIEGEL. B. 2000. Single or multistage regulation in complex life cycles: does it make a difference?Oi'kos88:883-892. JIMENEZ.J. J.. MORENO, A. G., LAVELLE,P. & DECAENS, T. 1998. Population dynamics and adaptive strategies of Martiodrilus carimaguensis(Oligochaeta, Glossoscolecidae),a native species from the well-drained savannas of Colombia.AppliedSoil Ecology9:153160. KAMMENGA,E. J., SPURGEON,J. D., SVENDSEN,C. & WEEKS,J. M. 2003. Explaining density-dependent regulation in earthworm populations using life-historyanalysis. Oikos100:89-9 5. KOUASSI,K. P. 1999. Structureet dynamiquedes groupetrophiquesde la macrofaunedu sol d'ecosystemesnaturelset transformesde C6tedIvoire. These d'Etat,Universit6d'Abidjan.206 pp. LAKHANI,K. H. & SATCHELL, J. E. 1970. Production by Lumbricus terrestris.Journalof AnimalEcology39:4 73-492. P. 1971. Etudeddmographique et dynamiquedespopulationsde LAVELLE, Millsoniaanomala(Acanthodrilidae, Oligochetes)dansla savanedeLamto de Universite Paris VI. 88 pp. Th&se 3e cycle, (C6tedi'hoire).

BENNOUR,S. A. & NAIR, G. A. 1997. Density, biomass and vertical distribution of Aporrectodea calignosa(Savigny, 1929) (Oligochaeta, Lumbricidae)in Benghazi, Libya.BiologyandFertilityofSoils24:102105. R. W. & LAVELLE, P. 1995. Influences of BLAIR,J. M., PARMELEE, earthworm on biogeochemistry. Pp. 127-158 in Hendrix. P. F. (ed.). Earthwormecology and biogeographyin North America.Lewis Publishers, Ann Arbor. BLANCHART,E. 1990. Rdle des vers de terre dans la formation et la conservationde la structuredes sols de savanede Lamto(C6ted'Ioire). Th&sede Doctorat, Universite de Rennes I. 263 pp. BLANCHART, E., ALBRECHT,A., ALEGRE,J., DUBOISSET,A., GILOT,C., PASHANASI,B., LAVELLE, P. & BRUSSAARD,L. 1999. Effects of earthworms on soil structure and physical properties. Pp. 149-172 in Lavelle, P., Brussard, L. & Hendrix, P. F. (eds). Earthwormmanagementin tropicalagroecosystemns. OxfordUniversity Press, Oxford. CASWELL,H. 1989. Matrix populationmodels. Sinauer, Sunderland, Massachusetts. 328 pp. DASH, C. M. & SENAPATI,K. B. 1980. Cocoon morphology, hatching and emergence pattern in tropicalearthworms. Pedobiologia20:316324.

LAVELLE,P. 1978. Les vers de terre de Ia savane de Lamto (C6te d'hoire):Peuplements,populationsetfonctionsdansl'cosystime. These de Doctorat d'Etat,Universit6Paris VI. Publication du Laboratoirede Zoologie ENS 12. 301 pp. LAVELLE,P. 1979. Relations entre types 6cologiques et profils d6mographiques chez les vers de terre de la savane de Lamto (Cbte d'Ivoire).Revued' Ecologieet deBiologiedu Sol 16:8 5-101. P. 1981. Strat6giesde reproductionchez les vers de terre.Acta LAVELLE, Oecologica(Oecologiageneralis)2:117-133. LAVELLE,P. 1983. The soil fauna of tropical savannas. II. The earthworms. Pp. 485-504 in Bourliere, F. (eds). Tropicalsavannas. ElsevierPublishing Company, Amsterdam. LAVELLE,P. 2002. Functional domains in soils. EcologicalResearch 17:441-450. LAVELLE,P., BAROIS,I., CRUZ,I., FRAGOSO,C., HERNANDEZ,A., PINEDA,A. &RANGEL,P. 1987. Adaptative strategies of Pontoscolex corethrurus(Glossoscolecidae,Oligochaeta), a peregrine geophagous earthworm of the humid tropics. BiologyandFertilityof Soils 5:188194. LAVELLE,P., BRUSSARD,L. & HENDRIX,P. F. 1999. Earthworm management in tropical agroecosystems. Oxford University Press, Oxford.300 pp.

This content downloaded on Sun, 10 Mar 2013 18:01:44 PM All use subject to JSTOR Terms and Conditions

500

EBAGNERIN TONDOHAND PATRICKLAVELLE JE?ROME

LEE,K. 1985. Earthworms:their ecologyand relationshipswith soil and landuse. Academic Press, London. 411 pp.

TONDOH. E. J. 1998a. Effect of coffee residues on growth and reproduction of Hyperiodrilusafricanus(Oligochaeta, Eudrilidae)in

MADGE,D. S. 1969. Fieldand laboratorystudies on the activities of two species of tropical earthworms. Pedobiologia9:188-214. OMODEO,P. 1954. Eudrilinae e octochaetina delle Costa d'Ovorio

Ivory Coast. Biologyand Fertilityof Soils 26:336-340. TONDOH.E. J. 1998b. Dmnographieet fonctionnementdes populations du ,er de terreHyperiodrilus africanus (Eudrilidae)dans une savane

(Oligochaeta). Momoriedel Museo Civicodi StoriaNaturaledi Verona 4:162-166.

protdgdedujeu en mioyenneCdted'Ivoire.These de Doctorat, Universit6 Paris VI. 209 pp. TONDOH. E. J. & LAVELLE,P. 1997. Effect of two organic residues on growth and cocoon production of the earthworm Hyperiodrilusafricanus(Eudrilidae).EuropeanJournalof Soil Biology 33:13-18.

ROMBKE, J. 1987. Population dynamics of earthworms in a moder soil beech forest. Pp. 199-214 in Bonvicini-Pagliai, A. M. &Omodeo, P. (eds). Onearthworms.Mucchi, Modena. ROSSI,J.-P. 1998. Rle fonctionneldela distributionspatialedesversdeterre dans une savanehumidede Cdted'IxToire. These de Doctorat, Universite Paris VI. 240 pp. SWIFT,M. J. 1984. Soil biological processes and tropical soil fertility: a proposal for a collaborative programme of research. Biology

VILLENAVE,C., CHARPENTIER,F., LAVELLE,P., FELLER,C., BRUSSARD,L., PASHANASI. B., BAROIS, I., ALBRECHT,A. & PATRON,J. C. 1999. Effectsof earthworms on soil organic matter and nutrient dynamics following earthworm Inoculation in field

International5 [Specialissue]:1-38. experimental situations. Pp. 173-197 in Lavelle, P., Brussard, L. & TIAN, G. 1995. Breakdownof plant residues with contrasting chemical Hendrix.P. F. (eds). Earthwormmanagementin tropicalagroecosystems. OxfordUniversityPress, Oxford. compositions under humid tropical conditions: effectsof earthworms and millipedes. Soil BiologyandBiochemistry2 7:2 77-280. WHALEN. K. J., PARMELEE,W. R. & EDWARDS, C. A. 1998. en culturesur les peuplemnents TONDOH,E. J. 1993. Influencede la Population dynamics of earthworm communities in corn agrormise de la macrofaunedu sol en moyenneC6te d'Ivoire.M6moire de DEA, ecosystems receiving organic or inorganic fertilizer amendments. Biologyand Fertilityof Soils 2 7:400-407. Universit6d'Abidjan. 78 pp.

This content downloaded on Sun, 10 Mar 2013 18:01:44 PM All use subject to JSTOR Terms and Conditions