Genetic control of domestication traits in pearl millet (Pennisetum glaucum L., Poaceae)

Quantitative trait loci (QTLs) controlling the morphological differences between pearl millet (Pennisetum glaucum ssp. glaucum) and its wild ancestor (Pennisetum glaucum ssp. monodii, form mollissimum) were investigated in a cultivated/wild F₂ population by means of RFLP markers. The most critical adaptive changes resulting from the domestication process involved the spikelet structure: non-shedding seeds with reduced bracts and bristles and long involucral pedicel. Major differences also concerned characters describing the plant architecture, phenology and spike sizes. Many morphological differences could be attributed to the effect of a small number of loci with relatively large effects. These loci are mainly concentrated on four linkage groups (2, 5, 6 and 7). The loss of shedding ability, due to the absence of a functional abscission layer, is controlled by a single locus on linkage group 6 (al6). Genetic control of the other spikelet traits involved factors with large effects which are located in the region of linkage group 6 close to al6 and to an esterase gene, Esterase-E. Moreover, QTLs with large effects on plant and spike morphology traits such as plant height, number of spikes and weight of the spike were also mapped on linkage groups 6 and 7. This strong linkage of factors in the domestication syndrome may be involved in the maintenance of the phenotypic identity of wild and cultivated populations in sympatry. This result also brings new arguments in the understanding of the domestication process of this allogamous crop. Received: 8 March 1999 / Accepted: 29 April 1999     

Genetic variation in grain-filling ability in dwarf pearl millet [Pennisetum glaucum (L.) R. Br.] restorer lines

The d2 dwarfing gene in pearl millet [Pennisetum glaucum (L.) R. Br.] carries a yield penalty due to an associated reduction in individual grain mass. This reduction, however, varies with genetic background, indicating that it may be possible to select against poor grain filling in d2 dwarfbackgrounds, given an effective measure of grain filling. This study was conducted to assess genetic variability forgrain-filling ability (in contrast to simply grain size),and its relationship to grain yield,indwarf pearl milletrestorer (R) lines. The grain-filling ability (GFA) of an individual R line was defined as the least squares estimate of its effect on individual grain mass in the analysis of variance, following a linear covariance adjustment for grain number. The study was based on 93dwarf hybrids involving31 d2 dwarfR-lines, evaluated over 3 years. Half of the variation in individual grain mass in the 93 hybrids was related to variation in grain number. Covariance adjustment in individual grain mass for grain number resulted in highly significant differences among hybrids and R lines in GFA. The R-line combining ability for GFA accounted for 26% of the variation in the R-line combining ability for yield, compared to 46% for the combining ability for grain number, and just 8% for the combining ability of individual grain mass. The combining ability for GFA was independent of the combining ability for various pre-flowering effects, including grain number, but was related to the combining ability for individual grain mass and harvest index. Improvement in individual grain mass achieved through selection for GFA should translate directly into yield improvement, whereasimprovement by direct selectionfor individual grain mass is less-likely to do so. Received: 9 April 2000 / Accepted: 16 May 2000     

Water saving traits co-map with a major terminal drought tolerance quantitative trait locus in pearl millet [Pennisetum glaucum (L.) R. Br.]

Low transpiration rates in pearl millet under fully irrigated conditions decrease plant water use at vegetative stage and then increase the water availability during grain filling and finally the terminal drought tolerance. Hundred and thirteen recombinant inbred lines developed from a cross between H77/833-2 and PRLT2/89-33 (terminal drought-sensitive?×?tolerant genotype) were evaluated to map transpiration rate (Tr, a proxy for canopy conductance), organ weights, leaf area and thickness and to study their interactions. Transpiration rate was increased by two H77/833-2 and two PRLT2/89-33 alleles on linkage group (LG) 2, whose importance depended on the vapor pressure deficit. The two H77/833-2 and one PRLT2/89-33 alleles co-mapped to a previously identified major terminal drought tolerance quantitative trait locus (QTL), although in a much smaller genetic interval. The other Tr allele from H77/833-2 also enhanced biomass dry weight and co-located with a formerly identified stover and tillering QTL. Leaf characteristics were linked to two loci on LG7. Plant water use was increased and decreased by different loci combinations for Tr, tillering and leaf characteristics, whose respective importance depended on the environmental conditions. Therefore, different alleles influence plant water use and have close interactions with one another and with the environment, so that different ideotypes for plant water use exist or could be designed from specific allele combinations conferring particular physiological characteristics for specific adaptation to a range of terminal drought conditions.     

Genotype identification and assessment of genetic relationships in pearl millet [Pennisetum glaucum (L.) R. Br] using microsatellites and RAPDs

 The potential of DNA markers such as microsatellites, minisatellites and RAPDs was investigated in pearl millet [Pennisetum glaucum (L.) R. Br] with respect to their abundance and variability. Southern analysis, using 22 different di-, tri-, tetra- and penta-oligonucleotide probes and five minisatellite probes, identified (GATA)₄ as the most useful probe for the detection of multiple polymorphic fragments among pearl millet cultivars and landraces from India. The clustering patterns of pearl millet cultivars and landraces based on (GATA)₄ and RAPD (randomly amplified polymorphic DNA) markers differed. The landraces, representing eight states in India, could not be grouped based on their geographical distribution with the DNA markers. RAPD analysis revealed a high degree of genetic diversity among the cultivars and landraces employed in this study. The probability of an identical match by chance for any two genotypes using (GATA)₄ and RAPDs was 3.02×10^-20 for cultivars and 5.2×10^-9 for landraces. The microsatellite (GATA)₄ and RAPDs provide useful tools for genotype identification and for the assessment of genetic relationships in pearl millet. Received: 19 October 1997 / Accepted: 9 December 1997     

Quantitative analysis of ultrastructural changes during zygotic and somatic embryogenesis in pearl millet (Pennisetum glaucum [L.] R. Br.)

Ultrastructural changes during zygotic and somatic embryogenesis in pearl millet (Pennisetum glaucum [L.] R. Br.) were quantified using morphometric techniques. The total area per cell profile and the cell volume percentage of the whole cell, endoplasmic reticulum (ER), Golgi bodies, mitochondria, nuclei, lipids, plastids, starch grains and vacuoles were measured and comparisons made between three zygotic and three somatic embryo developmental stages. All measurements were taken from scutellar or scutellar-derived cells. Zygotic embryogenesis was characterized by increases in cell size, lipids, plastids, starch, Golgi bodies, mitochondria and ER. Somatic embryogenesis was characterized by two phases of cell development: (1) the dedifferentiation of scutellar cells involving a reduction in cell and vacuole size and an increase in cell activity during somatic proembryoid formation and (2) the development of somatic embryos in which most cell organelle quantities returned to values found in late coleoptile or mature predesiccation zygotic stages. In summary, although their developmental pathways differed, the scutella of somatic embryos displayed cellular variations which were within the ranges observed for later stages of zygotic embryogenesis.     

In vitro morphogenesis of pearl millet [Pennisetum glaucum (L.) R.Br.]: efficient production of multiple shoots and inflorescences from shoot apices

 This report presents a procedure for high-frequency multiple shoot production from cultured shoot apical meristems of pearl millet [Pennisetum glaucum (L.) R. Br.]. Shoot apices from 1-week-old aseptically germinated seedlings were cultured in vitro on MS medium containing various concentrations of 2,4-dichlorophenoxyacetic acid (2,4-D) and benzyladenine (BA) with biweekly subculture. A low concentration of 2,4-D coupled with four different concentrations of BA induced the production of adventitious shoots from the enlarged shoot apical meristems. Somatic embryogenesis was also observed at higher concentrations of BA. The use of higher levels of 2,4-D resulted in callusing of shoot apical meristems, while the shoot tips produced many leaves and in vitro flowering in 2,4-D-free media containing BA. All four pearl millet genotypes produced similar results. Fertile pearl millet plants were produced from in vitro-produced multiple shoots. Received: 1 April 1999 / Revision received: 8 July 1999 / Accepted: 17 August 1999     

Phylogeny and origin of pearl millet (Pennisetum glaucum [L.] R. Br) as revealed by microsatellite loci

During the last 12,000 years, different cultures around the world have domesticated cereal crops. Several studies investigated the evolutionary history and domestication of cereals such as wheat in the Middle East, rice in Asia or maize in America. The domestication process in Africa has led to the emergence of important cereal crops like pearl millet in Sahelian Africa. In this study, we used 27 microsatellite loci to analyze 84 wild accessions and 355 cultivated accessions originating from the whole pearl millet distribution area in Africa and Asia. We found significantly higher diversity in the wild pearl millet group. The cultivated pearl millet sample possessed 81% of the alleles and 83% of the genetic diversity of the wild pearl millet sample. Using Bayesian approaches, we identified intermediate genotypes between the cultivated and wild groups. We then analyzed the phylogenetic relationship among accessions not showing introgression and found that a monophyletic origin of cultivated pearl millet in West Africa is the most likely scenario supported by our data set.     

Quantitative trait loci affecting morphology traits in gilthead seabream (Sparus aurata L.)

We report a quantitative trait loci (QTL) mapping study on 18 morphometric characters in gilthead seabream based on a total of 74 informative microsatellite markers genotyped in 409 offspring coming from 10 paternal half‐sib families. Statistical analysis was carried out using a linear regression approach, and various suggestive and significant morphology QTL were detected in three (9, 21 and 25) of nine linkage groups examined. Fitting body weight as a covariate reduced the significance of some QTL but revealed three new QTL in other linkage groups (LG6 and LG10). Current results combined with those obtained from previous studies underline highly significant loci affecting overall growth and morphology in S. aurata.     

Agrobacterium tumefaciens-mediated genetic transformation and production of stable transgenic pearl millet (Pennisetum glaucum [L.] R. Br.)

A synthetic gene encoding the antimicrobial peptide magainin has been designed, cloned, and engineered for regulation by the cauliflower mosaic virus (CaMV) 35S promoter and the nopaline synthase (nos) terminator. The plant expression cassette was introduced into the vector pSB11-bar (with the glyphosate [Basta®] resistance gene, bar), and the recombinant plasmid was mobilized into Agrobacterium tumefaciens strain LBA4404 for the generation of a super-binary vector pSB111-bar-mag. Magainins, positively charged amphipathic antimicrobial peptides of 21–26 amino acid residues, are potential candidates for the development of disease resistant transgenic plants. Six-wk-old pearl millet (Pennisetum glaucum [L.] R. Br.) calli and A. tumefaciens harboring pSB111-bar-mag were cocultivated in a medium supplemented with 400 μM acetosyringone and 3.3 mM l-cysteine. Out of 3,000 infected calli subjected to selection on phosphinothricin medium, 82 calli showed sectors of healthy growth, resulting in a transformation frequency of 2.73%. Among 13 Basta-tolerant putative transformed plants, eight were fertile and their transgenic nature and expression of the transgene was characterized by Southern and Northern blot analyses, respectively. Subsequent T₁ progenies co-segregated for bar and magainin genes in a 3:1 ratio. Bioassays that challenged the eight transgenic T₁ plant progenies against three highly virulent strains of Sclerospora graminicola, viz., Sg 384, Sg 445, and Sg 492 failed to show resistance. The failure of synthetic magainin gene to confer resistance against downy mildew in pearl millet may be attributed to the complexity of the cell wall and cell membrane of the pathogen.     

Host plant resistance to grain mould in germplasm accessions of pearl millet (Pennisetum glaucum [L.] R. Br.)

Abstract

The paucity of information on the moulds in Indian pearl millet (Pennisetum glaucum) led to the studies that were conducted at ICRISAT, India to evaluate (a) 447 germplasm accessions of 32 countries for mould reaction in rainy season, (b) threshed grain mould rating (TGMS) and mycoflora on grains of each accession, and (c) mould scores in field and in vitro. Post physiological maturity evaluation showed that 16% of the accessions secured a mould rating of 2. In TGMS, 18% were mould free and 57% secured a rating of 2 on a 1 – 9 scale. Assessment of twenty representative accessions in vitro against individual and mixed conidial suspensions (1 × 10⁽⁶⁾ conidia ml^(?1)) of Fusarium moniliforme, F. pallidoroseum and Curvularia pennisetti indicated significant correlation (r = 0.97) between the overall field and in vitro scores of mixed spores inoculations. The mycoflora for TGMS in blotter test revealed that Fusarium moniliforme, F. pallidoroseum, Curvularia pennisetti, Helminthosporium spp., Alternaria spp. and Colletotrichum spp. to be the major fungi affecting pearl millet grain. It is advisable to harvest panicles at the physiological maturity stage to obtain better quality grains. A strong negative correlation between TGMS and % GS (r = 0.4601) and positive correlation between TGMS and % UGS (r = 0.4654) indicated that, the lesser the threshed grain mould rating higher the % seed germination.     

Development of transgenic pearl millet (Pennisetum glaucum (L.) R. Br.) plants resistant to downy mildew

Transgenic pearl millet lines expressing pin gene—exhibiting high resistance to downy mildew pathogen, Sclerospora graminicola—were produced using particle-inflow-gun (PIG) method. Shoot-tip-derived embryogenic calli were co-bombarded with plasmids containing pin and bar genes driven by CaMV 35S promoter. Bombarded calli were cultured on MS medium with phosphinothricin as a selection agent. Primary transformants 1T₀, 2T₀, and 3T₀ showed the presence of both bar and pin coding sequences as evidenced by PCR and Southern blot analysis, respectively. T₁ progenies of three primary transformants, when evaluated for downy mildew resistance, segregated into resistant and susceptible phenotypes. T₁ plants resistant to downy mildew invariably exhibited tolerance to Basta suggesting co-segregation of pin and bar genes. Further, the downy mildew resistant 1T₁ plants were found positive for pin gene in Southern and Northern analyses thereby confirming stable integration, expression, and transmission of pin gene. 1T₂ progenies of 1T₀ conformed to dihybrid segregation of 15 resistant:1 susceptible plants.     

Evolutionary history of pearl millet (Pennisetum glaucum [L.] R. Br.) and selection on flowering genes since its domestication

The plant domestication process is associated with considerable modifications of plant phenotype. The identification of the genetic basis of this adaptation is of great interest for evolutionary biology. One of the methods used to identify such genes is the detection of signatures of selection. However, domestication is generally associated with major demographic effects. It is therefore crucial to disentangle the effects of demography and selection on diversity. In this study, we investigated selection in a flowering time pathway during domestication of pearl millet. We first used a random set of 20 genes to model pearl millet domestication using approximate Bayesian computation. This analysis showed that a model with exponential growth and wild-cultivated gene flow was well supported by our data set. Under this model, the domestication date of pearl millet is estimated at around 4,800 years ago. We assessed selection in 15 pearl millet DNA sequences homologous to flowering time genes and showed that these genes underwent selection more frequently than expected. We highlighted significant signatures of selection in six pearl millet flowering time genes associated with domestication or improvement of pearl millet. Moreover, higher deviations from neutrality were found for circadian clock-associated genes. Our study provides new insights into the domestication process of pearl millet and shows that a category of genes of the flowering pathway were preferentially selected during pearl millet domestication.     

Diversity of wild and cultivated pearl millet accessions (Pennisetum glaucum [L.] R. Br.) in Niger assessed by microsatellite markers

Genetic diversity of crop species in sub-Sahelian Africa is still poorly documented. Among such crops, pearl millet is one of the most important staple species. In Niger, pearl millet covers more than 65% of the total cultivated area. Analyzing pearl millet genetic diversity, its origin and its dynamics is important for in situ and ex situ germplasm conservation and to increase knowledge useful for breeding programs. We developed new genetic markers and a high-throughput technique for the genetic analysis of pearl millet. Using 25 microsatellite markers, we analyzed genetic diversity in 46 wild and 421 cultivated accessions of pearl millet in Niger. We showed a significantly lower number of alleles and lower gene diversity in cultivated pearl millet accessions than in wild accessions. This result contrasts with a previous study using iso-enzyme markers showing similar genetic diversity between cultivated and wild pearl millet populations. We found a strong differentiation between the cultivated and wild groups in Niger. Analyses of introgressions between cultivated and wild accessions showed modest but statistically supported evidence of introgressions. Wild accessions in the central region of Niger showed introgressions of cultivated alleles. Accessions of cultivated pearl millet showed introgressions of wild alleles in the western, central, and eastern parts of Niger.Electronic supplementary material Supplementary material is available in the online version of this article at and is accessible for authorized users.Cedric Mariac and Viviane Luong have contributed equally to this work.     

Efficient in vitro plant regeneration from immature zygotic embryos of pearl millet [Pennisetum glaucum (L.) R. Br.] and Sorghum bicolor (L.) Moench

We report here an in vitro culture system that provides reliable, highly efficient regeneration from immature embryos of pearl millet [Pennisetum glaucum (L.) R. Br.] and sorghum [Sorghum bicolor (L.) Moench]. Immature embryos were isolated 10-20 days after pollination and cultured on various L3 media. The influence of different parameters during the callus induction phase was examined with respect to the regeneration rate: (1) the concentrations of 2,4-dichlorophenoxyacetic acid (2,4-D) and various cytokinins; (2) the addition of AgNO₃; (3) the use of maltose or sucrose as a carbon source. Modifications in the phytohormones alone resulted in the regeneration of fertile sorghum plants at high efficiency. Significant increases in the regeneration rates of pearl millet genotypes were achieved by the combination of sucrose as a carbon source and silver nitrate as a potential ethylene inhibitor.     

Telomere Length Dynamics in Pearl Millet (Pennisetum glaucum [L.] R. Br.) as Observed at Different Developmental Stages

Abstract: In plants, the developmental dynamics of telomere length have only been studied in a few species to date. Contrasting results have been reported. To search for the pattern(s) operating in plants, a study of the telomere length was made in pearl millet. Telomere length in cells representing different developmental stages: a) embryo, b) leaves of 1-week-old seedlings, 1-month-old plants and boot leaf and c) germ cells (pollen) were compared. The presence of the consensus plant telomere repeat sequence (5'-TTTAGGG-3') in pearl millet telomeres was first ascertained; the sensitivity of the sequences hybridizing with (TTTAGGG)₄ oligonucleotide to time course Bal 31 exonuclease digestions were studied on dot blots and gel blots. The exonuclease digestion kinetics revealed the presence of the consensus telomere repeat sequence in pearl millet telomeres. The average telomere length (ATL) was measured from autoradiograms of Hae III digested DNA, hybridized with labelled (5'-TTTAGGG-3')₄ oligonucleotide using "UVI band" software. No significant difference in the average telomere length was observed between the embryo, leaves of 1-week-old seedlings, boot leaf and pollen. The ATL in leaves of 1-month-old plants was slightly higher. The results of the present investigation and analysis of the reports in the other plants suggest that there is an occasional increase in telomere length in some telomeres but no significant decrease due to loss during DNA replication.     

Osmotic adjustment to water stress in pearl millet (Pennisetum americanum [L.] Leeke) under field conditions

Abstract. Osmotic adjustment, a mechanism whereby plants maintain positive turgor despite low water potential (ψ), was investigated in pearl millet ( Pennisetum americanum [L.] Leeke) in three types of field experiment at Hyderabad, India:

• (1)

  Osmotic adjustment during the growing season was evaluated by comparing solute potential (ψ_s) of leaves taken at midday from irrigated and droughted plots and allowed to rehydrate in the laboratory. The degree of seasonal adjustment was also estimated by comparing observed values of ψ_s in the field with those expected if ψ_s decreased solely in proportion to water loss. Both types of assessment indicated the maximum seasonal adjustment to be about 0.2 MPa. The cultivars BJ 104 and Serere 39 differed in their capacity to adjust osmotically over the season; Serere 39 was least able to osmoregulate.
• (2)

  Measurements of diurnal variations in ψ and ψ_s in BJ 104 revealed osmotic adjustment during the afternoon hours. At a given value of ψ, turgor (ψ_p) was about 0.1 MPa higher in irrigated, and over 0.2 MPa higher in droughted plants, in the afternoon, than in the morning.
• (3)

  Osmotic adjustment of different leaves within the canopy was investigated. Upper leaves had lower ψ than basal leaves. Differences in ψ were matched by gradients in ψ_s, so that turgor was similar for all leaf layers.

     

The expression and perpetuation of inherent somatic variation in regenerants from embryogenic cultures of Pennisetum glaucum (L.) R. Br. (pearl millet)

Summary Genetic analysis was conducted on the qualitative and quantitative traits of sexual progeny derived from embryogenic cultures of two inbred lines of Pennisetum glaucum (L.) R. Br. (pearl millet). These lines included a genetically stable inbred of Tift 23 BE and a genetic marker line, derived from Tift 23BE, which bore qualitative genetic markers for a dominant purple plant trait (P) and two recessive traits, early flowering (e₁) and yellow stripe (ys). Tissue culture regenerant populations (R₀) and progeny populations (R₁) produced from these plants by selfing showed no qualitative genetic variation when derived from the genetically stable inbred Tift 23BE. In contrast, stably inherited qualitative variation for a number of genetic markers was observed in R₀, R₁, and R₂ progeny of the genetic marker line. In a population of 1,911 plants regenerated over a 12-month period, 0.02% of the population lost or showed reduced expression of the purple plant trait and 92% of plants were chlorophyll deficient. Plants showing reduction or loss of anthocyanin synthesis also flowered later. None of the purple plants showed any significant variation in flowering time. The incidence of chlorophyll deficiency increased with time in culture, 51 % of the progeny regenerated after 1 month were chlorophyll deficient, while 100% of the plants regnerated after 12 months were chlorophyll deficient. Qualitative variation was also observed in control populations of the genetic marker line where 1 plant in a total of 1,010 lacked purple pigmentation and a total of 6% showed chlorophyll variation in the first generation (S₀). The presence of qualitative variation in controls suggests that the inherent variation present in the original explant was expressed and perpetuated in vitro. Quantitative variation was observed for a number of traits in the first sexual cycle (R₁) of the marker line but did not occur in a subsequent generation, suggesting that this variation was epigenetic.     

Transgenic and herbicide resistant pearl millet (Pennisetum glaucum L.) R.Br. via microprojectile bombardment of scutellar tissue

Four different pearl millet breeding lines were transformed and led to the regeneration of fertile transgenic plants. Scutellar tissue was bombarded with two plasmids containing the bar selectable marker and the -glucuronidase reporter gene (gus or uidA) under control of the constitutive CaMV 35S promoter or the maize Ubiquitin1 promoter (the CaMV 35S is not a maize promoter). For the delivery of the DNA-coated microprojectiles, either the particle gun PDS 1000/He or the particle inflow gun was used. The calli and regenerants were selected for their resistance to the herbicide Basta (glufosinate ammonium) mediated by the bar gene. Putative transformants were screened for enzyme activity by painting selected leaves or spraying whole plants with an aqueous solution of the herbicide Basta and by the histochemical GUS assay using cut leaf segments. PCR and Southern blot analysis of genomic DNA indicated the presence of introduced foreign genes in the genomic DNA of the transformants. Five regenerated plants represent independent transformation events and have been grown to maturity and set seed. The integration of the bar selectable and the gus reporter gene was confirmed by genomic Southern blot analysis in all five plants. All five plants had multiple integrations of both marker genes. To date, the T₁ progeny of three out of four lines generated by the PDS particle gun shows co-segregating marker genes, indicating an integration of the bar and the gus gene at the same locus in the genome.     

Genetic architecture of purple pigmentation and tagging of some loci to SSR markers in pearl millet, Pennisetum glaucum (L.) R. Br

This report describes the construction of integrated genetic maps in pearl millet involving certain purple phenotype and simple sequence repeat (SSR) markers. These maps provide a direct means of implementing DNA marker-assisted selection and of facilitating "map-based cloning" for engineering novel traits. The purple pigmentation of leaf sheath, midrib and leaf margin was inherited together 'en bloc' under the control of a single dominant locus (the 'midrib complex') and was inseparably associated with the locus governing the purple coloration of the internode. The purple panicle was caused by a single dominant locus. Each of the three characters (purple lamina, purple stigma and purple seed) was governed by two complementary loci. One of the two loci governing purple seed was associated with the SSR locus Xpsmp2090 in linkage group 1, with a linkage value of 22 cM, while the other locus was associated with the SSR locus Xpsmp2270 in linkage group 6, with a linkage value of 23 cM. The locus for purple pigmentation of the midrib complex was either responsible for pigmentation of the panicle in a pleiotropic manner or was linked to it very closely and associated with the SSR locus Xpsmp2086 in linkage group 4, with a suggestive linkage value of 21 cM. A dominant allele at this locus seems to be a prerequisite for the development of purple pigmentation in the lamina, stigma and seed. These findings suggest that the locus for pigmentation of the midrib complex might regulate the basic steps in anthocyanin pigment development by acting as a structural gene while other loci regulate the formation of color in specific plant parts.     

Correlation between germination and adult stage under water stress in some Tunisian autochthonous pearl millet ecotypes (Pennisetum glaucum (L.) R. Br.)

This study compared the behaviour of six Tunisian autochthonous pearl millet ecotypes collected through the Tunisian territory under drought, from germination to maturity. Moderate water stress was found to improve germination due to better root elongation; it did not have any notable consequence on the yield components of pearl millet. When drought is severe, germination is compromised, radicle and epicotyle lengths decrease, the size of the plant is reduced as well as the surface of its flag sheet, and all the components of yield are penalized. This effect varies according to the ecotype's morphology and its geographical origin. In the same way, we could establish important correlations that show that radicle and epicotyle lengths are related to the final size of the pearl millet plant. All this shows that the study of germination is a scientific tool that can have agronomic repercussions, practical and exploitable.