Construction of microsatellite-based linkage map and mapping of nectarilessness and hairiness genes in Gossypium tomentosum

Gossypium tomentosum, a wild tetraploid cotton species with AD genomes, possesses genes conferring strong fibers and high heat tolerance. To effectively transfer these genes into Gossypium hirsutum, an entire microsatellite (simple sequence repeat, SSR)-based genetic map was constructed using the interspecific cross of G. hirsutum × G. tomentosum (HT). We detected 1800 loci from 1347 pairs of polymorphic primers. Of these, 1204 loci were grouped into 35 linkage groups at LOD?≥?4. The map covers 3320.8 cM, with a mean density of 2.76 cM per locus. We detected 420 common loci (186 in the At subgenome and 234 in Dt) between the HT map and the map of TM-1 (G. hirsutum) and Hai 7124 (G. barbadense; HB map). The linkage groups were assigned chromosome numbers based on location of common loci and the HB map as reference. A comparison of common markers revealed that no significant chromosomal rearrangement exist between G. tomentosum and G. barbadense. Interestingly, however, we detected numerous (33.7%) segregation loci deviating from 3:1 ratio (P?<?0.05) in HT, mostly clustering on eight chromosomes in the Dt subgenome, with some on three chromosomes in At. Two morphological traits, leaf hairiness and leaf nectarilessness were mapped on chromosomes 6 (A6) and 26 (D12), respectively. The SSR-based map constructed in this study will be useful for further genetic studies on cotton breeding, including mapping loci controlling quantitative traits associated with fiber quality, stress tolerance and developing chromosome segment specific introgression lines from G. tomentosum into G. hirsutum using marker-assisted selection.     

Development and Evaluation of Intron and Insertion–Deletion Markers for Gossypium barbadense

A total of 588 Gossypium barbadense coding sequences (CDSs) from nucleotide databases were selected for marker development. After selection, 125 CDSs were used to design 126 markers, including 39 intron polymorphisms (GbIPs) and 87 insertion?Cdeletion polymorphisms (GbIDPs). These markers were evaluated by analyzing the genetic diversity of 66 tetraploid cotton accessions including 56 G. barbadense accessions and 10 Gossypium hirsutum accessions. The amplification efficiencies of the GbIPs and GbIDPs were 0.560 and 0.489 for polymorphism information content, 0.744 and 0.690 for effective multiplex ratio (E), 0.653 and 0.438 for qualitative of nature of data, and 0.272 and 0.148 for effective marker index. Principal coordinate analysis showed profound differences between G. hirsutum and G. barbadense accessions. In addition, most of the G. barbadense accessions of Xinjiang, China were clearly different from foreign and other Chinese G. barbadense accessions. The 126 markers were also evaluated for their ability to enrich genetic maps, and 16 polymorphic loci were mapped on nine chromosomes with six loci on A subgenome and 10 loci on D subgenome. The mapping efficiencies of GbIPs and GbIDPs primers were 15.38% and 11.49%, respectively. This study well proves that GbIPs and GbIDPs can be successfully applied to the analysis of genetic diversity and construction of genetic maps.     

GbEXPATR,a species‐specific expansin,enhances cotton fibre elongation through cell wall restructuring

Cotton provides us the most important natural fibre. High fibre quality is the major goal of cotton breeding, and introducing genes conferring longer, finer and stronger fibre from Gossypium barbadense to Gossypium hirsutum is an important breeding strategy. We previously analysed the G. barbadense fibre development mechanism by gene expression profiling and found two homoeologous fibre‐specific α‐expansins from G. barbadense, GbEXPA2 and GbEXPATR. GbEXPA2 (from the D_T genome) is a classical α‐expansin, while its homoeolog, GbEXPATR (A_T genome), encodes a truncated protein lacking the normal C‐terminal polysaccharide‐binding domain of other α‐expansins and is specifically expressed in G. barbadense. Silencing EXPA in G. hirsutum induced shorter fibres with thicker cell walls. GbEXPA2 overexpression in G. hirsutum had no effect on mature fibre length, but produced fibres with a slightly thicker wall and increased crystalline cellulose content. Interestingly, GbEXPATR overexpression resulted in longer, finer and stronger fibres coupled with significantly thinner cell walls. The longer and thinner fibre was associated with lower expression of a number of secondary wall‐associated genes, especially chitinase‐like genes, and walls with lower cellulose levels but higher noncellulosic polysaccharides which advocated that a delay in the transition to secondary wall synthesis might be responsible for better fibre. In conclusion, we propose that α‐expansins play a critical role in fibre development by loosening the cell wall; furthermore, a truncated form, GbEXPATR, has a more dramatic effect through reorganizing secondary wall synthesis and metabolism and should be a candidate gene for developing G. hirsutum cultivars with superior fibre quality.     

Postinfection Development of Rotylenchulus reniformis on Resistant Gossypium barbadense Accessions

The reniform nematode (Rotylenchulus reniformis) causes significant cotton (Gossypium hirsutum) losses in the southeastern United States. The research objective was to describe the effects of two resistant G. barbadense lines (cultivar TX 110 and accession GB 713) on development and fecundity of reniform nematode. Nematode development and fecundity were evaluated on the resistant lines and susceptible G. hirsutum cultivar Deltapine 16 in three repeated growth chamber experiments. Nematode development on roots early and late in the infection cycle was measured at set intervals from 1 to 25 d after inoculation (DAI) and genotypes were compared based on the number of nematodes in four developmental stages (vermiform, swelling, reniform, and gravid). At 15, 20, and 25 DAI, egg production by individual females parasitizing each genotype was measured. Unique reniform nematode developmental patterns were noted on each of the cotton genotypes. During the early stages of infection, infection and development occurred 1 d faster on susceptible cotton than on the resistant genotypes. Later, progression to the reniform and gravid stages of development occurred first on the susceptible genotype, followed by G. barbadense cultivar TX 110, and finally G. barbadense accession GB 713. Egg production by individual nematodes infecting the three genotypes was similar. This study corroborates delayed development previously reported on G. barbadense cultivar TX 110 and is the first report of delayed infection and development associated with G. barbadense accession GB 713. The different developmental patterns in the resistant genotypes suggest that unique or additional loci may confer resistance in these two lines.     

Bidirectional cytoplasmic and nuclear introgression in the New World cottons Gossypium barbadense and G. Hirsutum (Malvaceae)

Patterns of interspecific cytoplasmic (plastid and mitochondrial) and nuclear introgression are typically asymmetrical: cytoplasmic gene flow or “capture” is frequently observed without evidence of nuclear introgression. In contrast, nuclear introgression without concomitant cytoplasmic introgression has rarely been demonstrated. Gossypium barbadense L. and G. hirsutum L. have large indigenous ranges in the New World semiarid tropics, with an extensive area of sympatry in the Caribbean and Central America. Numerous accessions of both species were surveyed for diagnostic cpDNA restriction sites. These data, in conjunction with previous information on nuclear markers, lead to several conclusions: 1) introgression between G. hirsutum and G. barbadense is bidirectional for both nuclear and cytoplasmic genes; 2) patterns of introgression between the two species are not symmetrical—in G. barbadense, introgression of G. hirsutum alleles is largely restricted to modem cultivars and is uncommon in areas of sympatry; in contrast, introgression of G. barbadense alleles into G. hirsutum is relatively common in areas of sympatry and is rare in modem cultivars; 3) nuclear introgression is geographically more widespread and more frequently detected than cytoplasmic introgression. Several mechanisms may underlie the differential patterns of cytoplasmic and nuclear gene flow observed, including differential fitness of infraspecific and interspecific cytonuclear combinations and selection against female function in interspecific backcrosses. Possible explanations for the observed asymmetrical patterns of introgression include differences in population sizes combined with phenological differences that bias interspecific pollen transfer.     

Wide coverage of the tetraploid cotton genome using newly developed microsatellite markers

Microsatellite [simple-sequence repeat (SSR)] markers were developed and positioned on the genetic map of tetraploid cotton. Three hundred and ninety-two unique microsatellite sequences, all but two containing a (CA/GT) repeat, were isolated, and the deduced primers were used to screen for polymorphism between the Gossypium hirsutum and G. barbadense parents of the mapping population analyzed in our laboratory. The observed rate of polymorphism was 56%. The 204 polymorphic SSRs revealed 261 segregating bands, which ultimately gave rise to 233 mapped loci. The updated status of our genetic map is now of 1,160 loci and 5,519 cM, with an average distance between two loci of 4.8 cM. The presence of a total of 466 microsatellite loci, with an average distance of 12 cM between two SSR loci, now provides wide coverage of the genome of tetraploid cotton and thus represents a powerful means for the production of a consensus map and for the effective tracking of QTLs.Electronic Supplementary Material Supplementary material is available in the online version of this article at http://dx.doi.org/10.1007/s00122-004-1612-1Communicated by C. Möllers     

A BIL Population Derived from G. hirsutum and G. barbadense Provides a Resource for Cotton Genetics and Breeding

To provide a resource for cotton genetics and breeding, an interspecific hybridization between Gossypium hirsutum cv. Emian22 and G. barbadense acc. 3–79 was made. A population of 54 BILs (backcross inbred lines, BC₁F₈) was developed with the aim of transferring G. barbadense genes into G. hirsutum in order to genetically analyze these genes’ function in a G. hirsutum background and create new germplasms for breeding. Preliminary investigation of the morphological traits showed that the BILs had diverse variations in plant architecture, seed size, and fuzz color; the related traits of yield and fiber quality evaluated in 4 environments also showed abundant phenotypic variation. In order to explore the molecular diversity of the BIL population, 446 SSR markers selected at an average genetic distance of 10 cM from our interspecific linkage map were used to genotype the BIL population. A total of 393 polymorphic loci accounting for 84.4% MAF (major allele frequency) > 0.05 and 922 allele loci were detected, and the Shannon diversity index (I) was 0.417 per locus. The average introgression segment length was 16.24 cM, and an average of 29.53 segments were introgressed in each BIL line with an average background recovery of 79.8%. QTL mapping revealed 58 QTL associated with fiber quality and yield traits, and 47 favored alleles derived from the donor parent were discovered. This study demonstrated that the interspecific BIL population was enriched with much phenotypic and molecular variation which could be a resource for cotton genetics and breeding.     

A new interspecific, <Emphasis Type="Italic">Gossypium hirsutum</Emphasis> × <Emphasis Type="Italic">G. barbadense,</Emphasis> RIL population: towards a unified consensus linkage map of tetraploid cotton

We report the development of a new interspecific cotton recombinant inbred line (RIL) population of 140 lines deriving from an interspecific cross between Gossypium hirsutum (Gh) and G. barbadense (Gb), using the same two parents that have served for the construction of a BC₁ map and for the marker-assisted backcross selection program underway at CIRAD. Two marker systems, microsatellites and AFLPs, were used. An important feature of the RIL population was its marked segregation distortion with a genome-wide bias to Gh alleles (parental genome ratio is 71/29). The RIL map displays an excellent colinearity with the BC₁ map, although it is severely contracted in terms of map size. Existence of 255 loci in common (between 6 and 14 per chromosome) allowed the integration of the two data sets. A consensus BC₁–RIL map based upon 215 individuals (75 BC1 + 140 RIL) was built. It consisted of 1,745 loci, spanned 3,637 cM, intermediate between the sizes of the two component maps, and constituted a solid framework to cross align cotton maps using common markers. The new RIL population will be further exploited for fiber property QTL mapping and eQTL mapping. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Mapping quantitative trait loci for lint yield and fiber quality across environments in a Gossypium hirsutum × Gossypium barbadense backcross inbred line population

Identification of stable quantitative trait loci (QTLs) across different environments and mapping populations is a prerequisite for marker-assisted selection (MAS) for cotton yield and fiber quality. To construct a genetic linkage map and to identify QTLs for fiber quality and yield traits, a backcross inbred line (BIL) population of 146 lines was developed from a cross between Upland cotton (Gossypium hirsutum) and Egyptian cotton (Gossypium barbadense) through two generations of backcrossing using Upland cotton as the recurrent parent followed by four generations of self pollination. The BIL population together with its two parents was tested in five environments representing three major cotton production regions in China. The genetic map spanned a total genetic distance of 2,895 cM and contained 392 polymorphic SSR loci with an average genetic distance of 7.4 cM per marker. A total of 67 QTLs including 28 for fiber quality and 39 for yield and its components were detected on 23 chromosomes, each of which explained 6.65–25.27 % of the phenotypic variation. Twenty-nine QTLs were located on the At subgenome originated from a cultivated diploid cotton, while 38 were on the Dt subgenome from an ancestor that does not produce spinnable fibers. Of the eight common QTLs (12 %) detected in more than two environments, two were for fiber quality traits including one for fiber strength and one for uniformity, and six for yield and its components including three for lint yield, one for seedcotton yield, one for lint percentage and one for boll weight. QTL clusters for the same traits or different traits were also identified. This research represents one of the first reports using a permanent advanced backcross inbred population of an interspecific hybrid population to identify QTLs for fiber quality and yield traits in cotton across diverse environments. It provides useful information for transferring desirable genes from G. barbadense to G. hirsutum using MAS.     

Quantitative trait locus analysis of Verticillium wilt resistance in an introgressed recombinant inbred population of Upland cotton

Verticillium wilt (VW) of Upland cotton (Gossypium hirsutum L.) is caused by the soil-borne fungal pathogen Verticillium dahlia Kleb. The availability of VW-resistant cultivars is vital for control of this economically important disease, but there is a paucity of Upland cotton breeding lines and cultivars with a high level of resistance to VW. In general, G. barbadense L. (source of Pima cotton) is more VW-resistant than Upland cotton. However, the transfer of VW resistance from G. barbadense to Upland cotton is challenging because of hybrid breakdown in the F2 and successive generations of interspecific populations. We conducted two replicated greenhouse studies (tests 1 and 2) to assess the heritability of VW resistance to a defoliating V. dahliae isolate and identify genetic markers associated with VW resistance in an Upland cotton recombinant inbred mapping population that has stable introgression from Pima cotton. Disease ratings at the seedling stage on several different days after the first inoculation (DAI) in test 1, as well as the percentages of infected and defoliated leaves at 2 DAI in test 2, were found to be low to moderately heritable, indicating the importance of a replicated progeny test in selection for VW resistance. With a newly constructed linkage map consisting of 882 simple sequence repeat, single nucleotide polymorphism, and resistance gene analog–amplified fragment length polymorphism marker loci, we identified a total of 21 quantitative trait loci (QTLs) on 11 chromosomes and two linkage groups associated with VW resistance at several different DAIs in greenhouse tests 1 and 2. The markers associated with the VW resistance QTLs will facilitate fine mapping and cloning of VW resistance genes and genomics-assisted breeding for VW-resistant cultivars.     

Cotton genome mapping with new microsatellites from Acala ‘Maxxa’ BAC-ends

Fine mapping and positional cloning will eventually improve with the anchoring of additional markers derived from genomic clones such as BACs. From 2,603 new BAC-end genomic sequences from Gossypium hirsutum Acala ‘Maxxa’, 1,316 PCR primer pairs (designated as MUSB) were designed to flank microsatellite or simple sequence repeat motif sequences. Most (1164 or 88%) MUSB primer pairs successfully amplified DNA from three species of cotton with an average of three amplicons per marker and 365 markers (21%) were polymorphic between G. hirsutum and G. barbadense. An interspecific RIL population developed from the above two entries was used to map 433 marker loci and 46 linkage groups with a genetic distance of 2,126.3 cM covering approximately 45% of the cotton genome and an average distance between two loci of 4.9 cM. Based on genome-specific chromosomes identified in G. hirsutum tetraploid (A and D), 56.9% of the coverage was located on the A subgenome while 39.7% was assigned to the D subgenome in the genetic map, suggesting that the A subgenome may be more polymorphic and recombinationally active than originally thought. The linkage groups were assigned to 23 of the 26 chromosomes. This is the first genetic map in which the linkage groups A01 and A02/D03 have been assigned to specific chromosomes. In addition the MUSB-derived markers from BAC-end sequences markers allows fine genetic and QTL mapping of important traits and for the first time provides reconciliation of the genetic and physical maps. Limited QTL analyses suggested that loci on chromosomes 2, 3, 12, 15 and 18 may affect variation in fiber quality traits. The original BAC clones containing the newly mapped MUSB that tag the QTLs provide critical DNA regions for the discovery of gene sequences involved in biological processes such as fiber development and pest resistance in cotton. Electronic Supplementary Material Supplementary material is available for this article at and is accessible for authorized users.     

Identification of candidate genes for key fibre‐related QTLs and derivation of favourable alleles in Gossypium hirsutum recombinant inbred lines with G. barbadense introgressions

Fine mapping QTLs and identifying candidate genes for cotton fibre‐quality and yield traits would be beneficial to cotton breeding. Here, we constructed a high‐density genetic map by specific‐locus amplified fragment sequencing (SLAF‐seq) to identify QTLs associated with fibre‐quality and yield traits using 239 recombinant inbred lines (RILs), which was developed from LMY22 (a high‐yield Gossypium hirsutumL. cultivar) × LY343 (a superior fibre‐quality germplasm with G. barbadenseL. introgressions). The genetic map spanned 3426.57 cM, including 3556 SLAF‐based SNPs and 199 SSR marker loci. A total of 104 QTLs, including 67 QTLs for fibre quality and 37 QTLs for yield traits, were identified with phenotypic data collected from 7 environments. Among these, 66 QTLs were co‐located in 19 QTL clusters on 12 chromosomes, and 24 QTLs were detected in three or more environments and determined to be stable. We also investigated the genomic components of LY343 and their contributions to fibre‐related traits by deep sequencing the whole genome of LY343, and we found that genomic components from G. hirsutum races (which entered LY343 via its G. barbadense parent) contributed more favourable alleles than those from G. barbadense. We further identified six putative candidate genes for stable QTLs, including Gh_A03G1147 (GhPEL6), Gh_D07G1598 (GhCSLC6) and Gh_D13G1921 (GhTBL5) for fibre‐length QTLs and Gh_D03G0919 (GhCOBL4), Gh_D09G1659 (GhMYB4) and Gh_D09G1690 (GhMYB85) for lint‐percentage QTLs. Our results provide comprehensive insight into the genetic basis of the formation of fibre‐related traits and would be helpful for cloning fibre‐development‐related genes as well as for marker‐assisted genetic improvement in cotton.     

Genetic dissection of lint yield and fiber quality traits of <Emphasis Type="Italic">G. hirsutum</Emphasis> in <Emphasis Type="Italic">G. barbadense</Emphasis> background

Gossypium hirsutum L. is a widely cultivated species characterized by its high yield and wide environmental adaptability, while Gossypium barbadense is well known for its superior fiber quality. In the present report, we, for the first time, developed G. hirsutum chromosome segment introgression lines (ILs) in a G. barbadense background (GhILs_Gb) and genetically dissected the inheritance of lint yield and fiber quality of G. hirsutum in G. barbadense background. The GhILs_Gb contains introgressed segments spanning 4121.20 cM, which represents 82.20% of the tetraploid cotton genome, with an average length of 18.65 cM. A total of 39 quantitative trait loci (QTLs) for six traits are identified in this IL population planted in Xinjiang. Four QTL clusters are detected. Of them, however, three clusters have deleterious effects on fiber length and strength and boll weight, and only one cluster on Chr. D9 can be used in marker-assisted selection (MAS) to increase lint percentage and decrease micronaire value in G. barbadense. QTL mapping showed that most of yield-related QTLs detected have positive effects and increase lint yield in G. barbadense, while most of fiber quality-related QTLs have deleterious effects except for micronaire. It suggested that G. hirsutum evolved to have a high lint yield. Several lines improved in lint percentage and boll size in G. barbadense by introgressed one fragment of G. hirsutum have been developed from the GhILs_Gb. The ILs developed, and the analyses presented here will enhance the understanding of the genetics of lint yield and fiber quality in G. hirsutum and facilitate further molecular breeding to improve lint yield in G. barbadense.     

Genetic mapping of new cotton fiber loci using EST-derived microsatellites in an interspecific recombinant inbred line cotton population

There is an immediate need for a high-density genetic map of cotton anchored with fiber genes to facilitate marker-assisted selection (MAS) for improved fiber traits. With this goal in mind, genetic mapping with a new set of microsatellite markers [comprising both simple (SSR) and complex (CSR) sequence repeat markers] was performed on 183 recombinant inbred lines (RILs) developed from the progeny of the interspecific cross Gossypium hirsutum L. cv. TM1 × Gossypium barbadense L. Pima 3-79. Microsatellite markers were developed using 1557 ESTs-containing SSRs (≥10 bp) and 5794 EST-containing CSRs (≥12 bp) obtained from ~14,000 consensus sequences derived from fiber ESTs generated from the cultivated diploid species Gossypium arboreum L. cv AKA8401. From a total of 1232 EST-derived SSR (MUSS) and CSR (MUCS) primer-pairs, 1019 (83%) successfully amplified PCR products from a survey panel of six Gossypium species; 202 (19.8%) were polymorphic between the G. hirsutum L. and G. barbadense L. parents of the interspecific mapping population. Among these polymorphic markers, only 86 (42.6%) showed significant sequence homology to annotated genes with known function. The chromosomal locations of 36 microsatellites were associated with 14 chromosomes and/or 13 chromosome arms of the cotton genome by hypoaneuploid deficiency analysis, enabling us to assign genetic linkage groups (LG) to specific chromosomes. The resulting genetic map consists of 193 loci, including 121 new fiber loci not previously mapped. These fiber loci were mapped to 19 chromosomes and 11 LG spanning 1277 cM, providing approximately 27% genome coverage. Preliminary quantitative trait loci analysis suggested that chromosomes 2, 3, 15, and 18 may harbor genes for traits related to fiber quality. These new PCR-based microsatellite markers derived from cotton fiber ESTs will facilitate the development of a high-resolution integrated genetic map of cotton for structural and functional study of fiber genes and MAS of genes that enhance fiber quality. Electronic Supplementary Material Supplementary material is available for this article at Names are necessary to report factually on available data, however, the USDA neither guarantees nor warrants the standard of products or service, and the use of the name by the USDA implies no approval of the products or service to the exclusion of others that may also be suitable.     

Development of EST-based SNP and InDel markers and their utilization in tetraploid cotton genetic mapping

Background

Availability of molecular markers has proven to be an efficient tool in facilitating progress in plant breeding, which is particularly important in the case of less researched crops such as cotton. Considering the obvious advantages of single nucleotide polymorphisms (SNPs) and insertion-deletion polymorphisms (InDels), expressed sequence tags (ESTs) were analyzed in silico to identify SNPs and InDels in this study, aiming to develop more molecular markers in cotton.

Results

A total of 1,349 EST-based SNP and InDel markers were developed by comparing ESTs between Gossypium hirsutum and G. barbadense, mining G. hirsutum unigenes, and analyzing 3′ untranslated region (3′UTR) sequences. The marker polymorphisms were investigated using the two parents of the mapping population based on the single-strand conformation polymorphism (SSCP) analysis. Of all the markers, 137 (10.16%) were polymorphic, and revealed 142 loci. Linkage analysis using a BC₁ population mapped 133 loci on the 26 chromosomes. Statistical analysis of base variations in SNPs showed that base transitions accounted for 55.78% of the total base variations and gene ontology indicated that cotton genes varied greatly in harboring SNPs ranging from 1.00 to 24.00 SNPs per gene. Sanger sequencing of three randomly selected SNP markers revealed discrepancy between the in silico predicted sequences and the actual sequencing results.

Conclusions

In silico analysis is a double-edged blade to develop EST-SNP/InDel markers. On the one hand, the designed markers can be well used in tetraploid cotton genetic mapping. And it plays a certain role in revealing transition preference and SNP frequency of cotton genes. On the other hand, the developmental efficiency of markers and polymorphism of designed primers are comparatively low.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-1046) contains supplementary material, which is available to authorized users.     

A Comparison of Leaf Anatomy in Field-grown Gossypium hirsutum and G. barbadense

Gossypium hirsutum L. (upland cotton) and G. barbadense L. (Pimacotton) are two of the most important fibre producing cottonspecies in cultivation. When grown side-by-side in the field,G.hirsutum has higher photosynthetic and transpiration rates (Luet al., 1997. Australian Journal of Plant Physiology24: 693–700).The present study was undertaken to determine if the differencesin physiology can be explained by leaf and canopy morphologyand anatomy. Scanning electron microscopy was used to comparethe leaf anatomy of field-grown upland (‘Delta’and ‘Pine Land 50’) and Pima (‘S6’)cotton. Compared to G. hirsutum, mature leaves of G. barbadenseare larger and thinner, with a thinner palisade layer. G. barbadenseleaves show significant cupping or curling which allows fora more even absorption of insolation over the course of theday and much more light penetration into the canopy. AlthoughG. barbadense leaves have a 70–78% higher stomatal densityon both the abaxial and the adaxial surfaces, its stomates areonly one third the size of those of G. hirsutum. This resultsin G. barbadense having only about 60% of the stomatal surfacearea per leaf surface area compared to G. hirsutum. These resultsare indicative of the anatomical and physiological differencesthat may limit the yield potential of G. barbadense in certaingrowing environments. Copyright 2000 Annals of Botany Company Cotton, leaf anatomy, leaf development, photosynthesis, Gossypium hirsutum, Gossypium barbadense, stomatal density     

Some observations on photoperiodism and the development of annual forms of domesticated cottons

Observations made over the last fifteen to twenty years have discovered no primitive forms of G.barbadense L. orG. hirsutum L. (other than G.hirsutum racelatifolium Hutchinson) which are capable of flowering during the long summer days of temperate latitudes. Annual Upland cottons, grown in the southern United States since the mid-eighteenth century, were probably derived from perennial day-neutral forms of Mexican racelatifolium, though not necessarily by direct introduction from Mexico into the United States. The origin of Sea Island cottons (day-neutral forms of G.barbadense), which were formerly grown on the southeastern seaboard of the United States since the mid-eighteenth century, remains obscure. No primitive forms of this species, capable of flowering during the long summer days of temperate latitudes, have been found. An experiment is reported in which it has been possible to synthesize a day-neutral form of G.barbadense through introgression between primitive short-day sensitive forms of G.barbadense and G.hirsutum, both native to the Caribbean region. It is suggested that the ancestors of Sea Island cottons may have originated naturally by a similar mechanism involving the same species.