Environmental influences on phenotype

We have heard that genes and their products do not have isolated action in forming the phenotype. The phenotype develops within the genotype’s limits, depending on the environment. For this interesting matter, I would like to show some amazing examples here. According to Pierce (2008), most discussed characteristics have slight influence from environment. However, many phenotypes have important effect from the environment.

                A curious example is found in the influence of temperature in some gene expressions. In rabbits, the himalayan allele produces dark fur at the relative cool extremities of the body. When rabbits are reared at 20°C or less, this allele expresses the dark pigments in those extremities. When they are reared above 30°C, no dark portions develop. The enzyme responsible for dark fur is not active in higher temperatures. Another example with temperature is observed in some plant’s albinisms. In barley, a plant homozygous for the albino allele has the chlorophyll production inhibited when grown below 7°C. Above 18°C, the same allele expresses green plant with normal chlorophyll. Drosophila melanogaster has also temperature dependent mutation for vestigial wings (Pierce, 2008).

Influence of temperature on allele himalayan expression (Pierce, 2009)

                In humans, we have heard a lot about some environmental effects on health. Obviously, most people who have a fat diet are more likely to get obesity, mainly those who have genetic trend for obesity. Also, more body exposition for U.V. from sunlight influence the emergence of skin cancer. Silverstone and Searle (1970) showed a highly susceptible population from the state of Queensland, in Australia, to get solar keratosis and skin cancer, as the region provides an advantageous environment for these illnesses due to its proximity to the hottest tropic. Weaver (2009) demonstrated the early environment influences on brain development, whose resultant phenotype persists through life. He stated evidences that maternal care in the early postnatal life determine certain gene expressions in the brain through epigenetic modification, shaping the offspring’s neuroendocrine and behavioural stress response throughout life.

REFERENCE LIST

Pierce BA (2008). Genetic Maternal Effect. In: Genetics: A Conceptual Approach. (third edition) W. H. Freeman and Company, New York. pp 123.

Silverston H & Searle JHA (1970). The Epidemiology of Skin Cancer in Queensland: The Influence of Phenotype and Environment. British Journal of Cancer, 24(2): 235–252.

Weaver ICG (2009). Shaping Adult Phenotypes Through Early Life Environments. Birth Defects Research (Part C) 87:314–326.

Gene interaction for a single phenotype

            We are all very familiar with the Mendelian studies, which show the independent gene assortment for each trait analysed in the peas. However, genes frequently do not act independently in their phenotypic expression. Gene interaction is the term used to call the interaction between expressions of genes at different loci. The effects of genes at one locus depend on the expression of genes at other loci (Pierce, 2008). Thus, the phenotype is not predictable from the perspective of a single locus effects alone, as the products of genes at different loci blend to produce the phenotype.

Figure 1: Gene interaction in fruit colour of Capsicum annum (Pierce, 2008) 

                The fruit colour in the pepper Capsicum annuum is a simple example of gene interaction between two loci producing a single trait. The Y locus and the C locus interact to form the colours. When two heterozygous cross with each other (YyCc x YyCc) the four possible colours of pepper fruit are present in the offspring in a proportion of 9:3:3:1, which are red, peach, orange and cream (Pierce, 2008). You can see on the picture the genotype for each colour.

                

Figure 2: Some coat colours in mice
 (Griffiths, 2000)

         
More complex gene interaction is observed in the genetics of coat colour of mammals. In dogs, many genes participate, and several loci interact with each other. The expression of a particular gene is modified by the effects of other genes. There are four most recognized loci producing many of the noticeable differences in colour and pattern among breed of dogs. However, other poorly known loci may modify the effects of these loci (Pierce, 2008).  In mice, five main genes interact to determine the coat colours (Griffiths et. al. 2000).

                In humans a set of genes are involved in the production of melanin, which provides colour for skin, hair and eyes. Mutations in any of these genes disturb the production of melanin, which reduces pigmentation in the skin, hair, and eyes and causes albinism. Mutations in six genes have been related to be responsible for different types of albinism (Oetting & Kin, 1999).

Figure 3: Albinism in humans

REFERENCE LIST

Griffiths AJF, Miller JH, Suzuki DT, Lewontin RC, Gelbart WM (2000). Gene interaction in coat color of mammals. In An Introduction to Genetic Analysis. 7th edition. New York: W. H. Freeman.

Oetting, WS and King, RA (1999), Molecular basis of albinism: Mutations and polymorphisms of pigmentation genes associated with albinism. Hum. Mutat., 13: 99–115

Pierce BA (2008). Genetic Maternal Effect. In: Genetics: A Conceptual Approach. (third edition) W. H. Freeman and Company, New York. pp 105-114

Figure 3: http://foter.com/search/instant/?q=albinism  

Genomic Imprinting: A different way of gene expression

             Well, I have discussed about sex influences on heredity on the last posts. I could not forget this amazing topic in my literature reviews here. Genomic imprinting is a phenomenon whereby the expression of the genotype is different when inherited from the male or female parent. Some genes have their expression significantly differentiated according to their parental origin (Pierce, 2008). The mechanisms of control of this expression are still not totally clear, but the methylation of DNA is undoubtedly fundamental to the process, being a way of epigenetic phenomenon.

Fig. 1: Genomic imprinting in fetal growth of mice. (Pierce, 2008)

               The gene Igf2 of mice and humans express under genomic imprinting process, encoding the protein insulin-like growth factor II, which determines the size of the fetus and placenta. Both alleles of the gene are inherited from the father and the mother, but only the paternal one is expressed, while maternal copy is completely silent. Demonstrations of deleted paternal copy of Igf2 in mice show small placenta and low-birth-weight in the offspring as result. Like in the placenta in mammals, plants have also exhibited differential expression of maternal and paternal genes in the endosperm, which provides nutrients for the growth of the embryo (Pierce, 2008).

            

Fig. 2: Genomic imprinting in deletion of region in chromosome
15 causing Prader-Willi and Angelman Syndrome.

Genomic imprinting has also received importance in human syndromes. It has been observed this event in the Prader-Willi and Angelman syndromes. Many individuals with the Prader-Willi syndrome do not have a small region on the long arm of chromosome 15. This deletion is always inherited from the father. When the same deletion is maternal, the expression results in a totally different set of symptoms, originating another syndrome: the Angelman syndrome (Pierce, 2008). Children with Prader-Willi syndrome present small hands and feet, mental retardation, poor sexual development and short stature. Symptoms of Angelman syndrome include frequent uncontrolled muscle movement, a large mouth and unusual seizures. A normal development requires the presence of both paternal and maternal regions of the chromosome 15(Pierce, 2008).

REFERENCE:

Pierce B. A. (2008). Genetic Maternal Effect. In: Genetics: A Conceptual Approach. (third edition) W. H. Freeman and Company, New York. p 119-120.

Figure 2: http://atlasgeneticsoncology.org/Deep/GenomImprintID20032.html (accessed 10/05/2014)

Genetic Maternal Effect on Phenotype

                As I have informed about the inheritance of maternal mitochondria, I remembered another sexual influence on phenotype, the genetic maternal effect on some treats. However, it is important to clarify that cytoplasmic inheritance is different from genetic maternal effect. In mithocondrial inheritance, the genes for the characteristic are inherited from only the mother. In genetic maternal effect, the genes are inherited from both parents, but the offspring’s treat is determined not by its own genotype but by its mother’s genotype. The maternal effect usually arises when substances in the egg’s cytoplasm, which are encoded by the mother’s nuclear genes, influence the offspring’s phenotype in its early development (Pierce, 2008). According to William et al., these substances are gene products of regulation, which active or repress the expression of the zygote genome.

Genetic maternal effect in shell coiling of Limnaea peregra (Pierce, 2008)

                The best didactic example of maternal effect on phenotype is seen on the species Limnaea pelegra. The shell of its snails usually coils to the right (dextral coiling) but some snails are left-coiling shell (sinistral coiling). The allele for dextral (s+) is dominant over the allele for sinistral (s). The direction of the coiling is influenced by the way in which the cytoplasm divides soon after the fertilization. This event is determined by a substance produced by the mother and passed to the offspring through the egg’s cytoplasm (Pierce, 2008). As the progeny’s phenotype is determined by the mother’s genotype, not her phenotype, the phenotype of the mother is not necessarily the same as the offspring.

                As you can see on the picture, dextral males s+s+ crossed with sinistral females ss produce a heterozygous offspring (s+s) that have all sinistral shell, as its mother’s genotype encodes sinistral coiling. As result of the F1 self-fertilized, the genotype of F2 is 1s+s+:2s+s:1ss, all with dextral coiling (as F1 genotype produces dextral phenotype). Males can influence phenotypes here just by contributing to the genotype of their daughters, affecting the phenotypes of the daughters’ offspring (Pierce, 2008).

REFERENCES:

Klug W. S.; Cummings M. R.; Spencer C.; Palladino M. A.(2009). Concepts of Genetics. (Ninth edition) Benjamin Cummings, San Francisco. p 489.

Pierce B. A. (2008). Genetic Maternal Effect. In: Genetics: A Conceptual Approach. (third edition) W. H. Freeman and Company, New York. p 119-120.

Mitochondrial DNA

The human mitochondrial genome. (Schon et al. 1997) 

As we know, mitochondria are cytoplasmic organelles exclusive for eukaryotes. They have the central role of oxidative energy metabolism to the viability of the cell. Mitochondria contain a small specialized genome and a complete process of gene expression for their particular activities, distinct from that of the cell nucleus and cytosol. In humans, the mitochondrial genome is constituted by a circle of double-stranded DNA with 16, 569 base pairs. This circle DNA is highly compact and contains only 37 genes: 2 genes encode ribosomal RNAs, 22 encode transfer RNAs, and 13 encode polypeptides. The 13 polypeptides are components of the respiratory chain, the oxidative phosphorylation system (Schon et al. 1997).

           The female egg contains a cytoplasm with many mitochondria and other organelles. Besides the nuclear DNA, it has the entire mitochondrial DNA, which has the same genetic information for all mitochondria. The sperm has mitochondria placed where they can most efficiently control the flagellum, providing energy for their movement. In the fertilization, the only genetic material that the embryo inherits from the male is from the nuclear genetic material contained in the head’s sperm. Therefore, the individual inherits all the genetic mitochondrial material from the female, contained in the egg. In resume, we have the same genetic mitochondrial material and mitochondrial phenotype as our mother.

               Studies in evolutionary biology analyze mitochondrial inheritance to identify the degree of proximity between populations and species. Allen and de Paula (2013) also suggest that the contributions of male/female for the mitochondria arose with the evolutionary origin of separate sexes.  This knowledge is also useful in cases whereby a child mother is not known, and then tests with mitochondrial DNA may be done to prove if a woman is the birth mother. Another importance is observed in studies about human disorders. Pathogenic mutations in the mitochondrial genome can have devastating consequences; one of these is respiratory chain deficiency. 

REFERENCES

Allen J. F. & de Paula W. B. M. (2013). Mitochondrial genome function and maternal inheritance. Biochemical Society Transactions 41(5) 1298-1304.

Schon, E. A.; Bonilla, E; DiMauro S. (1997). Mitochondrial DNA Mutations and Pathogenesis. Journal of Bioenergetics and Biomembranes,  29(2) 131-149.

Chromosomal alterations and human disorders

                On the last post I talked about the relevance of chromosomal alterations for evolution. Therefore, I am going to explain a little more about the effects of these events on the phenotype of humans. As I said before, modifications on chromosome structure, duplication or deletion of entire or parts of chromosomes, can alter the genome, changing the genetic sequence or/and the normal way of gene expression. Chromosomal abnormalities in humans usually lead to patterns known as syndromes, combinations of signs and symptoms (Haydon, 2008).

            As I have mentioned, there are numerical and structural chromosomal alterations. The numerical ones are known as aneuploidy in syndromes. They occur as a result from meiosis error that originates gametes without any chromosome of a chromosome pair and gametes with the both 2 chromosomes. When these gametes fertilize they provide individuals with monosomy (1 chromosome) or trisomy (3 chromosomes) for that pair (Haydon, 2008).

            The most recognized autosomal aneuploidy is the Down syndrome, or trisomy 21 (Some symptoms at the figure above). Despite individuals with this syndrome have developmental delay and heart defect, they usually live well into adult life (Haydon, 2008). Klinefelter syndrome (47, XXY) is the most common sexual trisomy, whereby men have tall stature, small testes, scant body hair and infertility (Haydon, 2008). The only monosomy compatible to life is sexual: the Turner syndrome (45, X). The symptoms are short stature and infertility. Despite, women 45, X usually have a normal life, I believe this is due to the natural activation of only one X in women, which I have explained in previous posts.

            Structural chromosomal alterations are result of breakage and subsequent reunion of chromosome regions. We can have deletions of regions, losing material, or duplication by gaining a copy of a segment at the original location on the chromosome (Theisen and Shaffer, 2010). Also, it is possible to have rearrangements such as inversions of segments after two-break events; translocations by exchange of segments between chromosomes; insertions by translocation and insertion of segments into a new region of the same or other chromosome. These events are pathogenic when they disrupt important genes (Theisen and Shaffer, 2010). An example of deletion is found at the cri du chat syndrome, in chromosome 5. Symptoms include microcephaly, low-set ears, round face, high-pitched cat-like cry, mental retardation and some health problems(Theisen and Shaffer, 2010).

I would like to talk about more syndromes but there are so many for a single post.

Thanks for reading!

REFERENCE LIST:

Theisen, A & Shaffer L G (2010). Disorders caused by chromosome abnormalities. The Application of Clinical Genetics. 2010: 3; 159–174

Haydon, J (2008). Chromosome disorders. In: Genetics in Practice: A Clinical Approach for Healthcare Practitioners (ed. Wiley). pp. 85-100. Hoboken, NJ, USA.

Image source (19/04/2014): http://www.rayur.com/chromosomal-disorder-down-syndrome-trisomy-21.html

Chromosomes Alterations and Evolution

As we know, chromosomes alterations in human being have usually negative effects, causing diseases, sterile individuals or even death.  However, many species not only survive with these alterations but they can evolve from them. These alterations are a type of mutation, which is source of changes in the genome, leading to genome evolution (Reece et al., 2011). I going to explain how some types of chromosome alterations contribute to the evolution: Polyploidy (widely common in plants and rare in animals) and alterations of chromosome structure.

             Duplication of entire chromosome sets is result of accidents in meiosis that produce diploid gametes. According to Freeman and Herron (2004), if an individual self-fertilize its produced male and female diploid gametes, it will originate a polyploid organism. If this organism can self-fertilize or mate with another poplyploid, it is established a tetraploid population.  This event is important in evolution because if the tetraploid population mates with their parental diploid population, they will originate triploid organisms, which are sterile due to the odd number of chromosome set. Therefore, this leads to branching off of a new species. Also, the extra sets can accumulate mutations that promote new functions of the extra genes, maintained by natural selection (Freeman and Herron, 2004). These events change the individual phenotype and the new population diverges further from their parental population.

          Another contribution to evolution is observed in alterations of chromosome structure. Researchers have noticed chromosome rearrangements comparing a species to another. For example, ancestral chimp chromosomes 12 and 13 seem to have fused end to end, forming the chromosome 2 of a human ancestor. Comparing some species, scientists find many duplications and inversions of large portions of chromosomes, as result of DNA breaks and incorrect rejoining during meiotic recombination. These events might have accelerated about 100 million years ago, contributing to the generation of new species (Reece et al., 2011).

REFERENCE LIST:

Freeman, S. & Herron, J.C. (2004) Evolutionary Analysis. 3th ed. Pearson Prentice Hall. United States of America. p 124.

Reece, J. B.; Urry, L. A.; Cain, M. L.; Wasserman, S. A.; Minorsky, P. V.; Jackson, R. N.(2011) Campbell Biology. 9^th ed. Pearson Australia Group. p 446.

Advantageous Female Mosaicism: X-Chromosome Inactivation

As I talked last week about epigenetics, I found the example of X-Chromosome Inactivation extremely interesting. So, I resolved to explain more about.

            As we know, one of the female X-Chromosome in each cell is inactivated by methylation (epigenetic process) as a mechanism of dose compensation related to the male quantity of gene expression.  Half of the female’s cells has a working X chromosome from her mother while the other half has active X chromosome from her father. In this way, a cellular mosaicism is result from the X inactivation. The choice of the X-Chromosome to be inactivated is random. In the fetal development each cell chooses which X they will inactivate: the one from the father or the one from the mother.  

          The gene responsible for inactivating one X-Chromosome is called X inactive specific transcript (XIST). This X-linked gene is transcribed in the chromosome that will be inactivated. Its transcription causes methylation and modification of chromosome’s chromatin, which makes the chromosome inactive. (Migeon, 2007)

             The female mosaicism provides to women a biological advantage. This is not difficult to understand: a single X chromosome in males make them vulnerable. Any mutation that affects a gene on their maternal X chromosome will be expressed in 100% of their cells. In contrast, women have two X chromosome in each cell: when there is a mutation in a gene on one of their X chromosomes, they have a normal copy of that gene on another one. Because of X inactivation, the mutation will be expressed in only half the cells (Migeon, 2007). Therefore, X-linked diseases are more severe and lethal in men than in women.

        According to Migeon (2007), in the course of evolution, mammalians began with identical sex chromosomes of reptilian origin. Then, the Y chromosome has been progressively reduced in size and one of its genes became a sex-determining gene. As result of gene’s losses, this chromosome has now less than 100 genes while X-chromosome has about 1100 genes. Thus, mammals inactivate one of the two X chromosomes for dosage compensation. Other organisms also make dose compensation but curiously in a different way. Flies, for example, increase the expression of the single X chromosome in the male to achieve the same production as that from the two X chromosomes in the female.

An example of female mosaicism caused by X-Chromosome Inactivation: A mutation in a gene for sweat glands from one X-Chromosome affects some parts of the body while tissues that express the normal X-Chromosome are not affected.

REFERENCE LIST:

Migeon, B. R. (2007). Why Females Are Mosaics, X-Chromosome Inactivation, and Sex Differences in Disease. Gender Medicine, vol. 4, no. 2. 

Epigenetics

For each tissue or cell groups in multicellular organisms, cells express specific activity and phenotype, different from other tissues’ cells; despite all of them have the same genetic material. This differentiation happens due to epigenetic mechanism: a regulation of genes that controls which genes will be expressed and which will be ignored. Epigenetics enables identical twins, with identical genes, to have different traits and personalities (Williams, 2013).

                Epigenetics is composed by chemical flags and markers on genes that are copied with genes during the DNA replication. It can alter gene expression without altering the DNA sequence by blocking or activating a gene expression. Evidences show that the environment may alter the epigenetic regulation. Also, epigenetic modifications can confer phenotypic plasticity and in some cases heritability to offispring (Arinmondo, 2012).

                DNA methylation is one type of epigenetic regulation. The linking of methyl groups to certain nitrogenous bases impedes the gene expression. Another way to regulates genes is by histone modifications. Histones are proteins whereby DNA is enrolled around to compact (part of chromatin structure). The turning out of histones enables the DNA transcription (activating gene expression). Paola Arinmondo (2012) affirmed that other epigenetic modifications have been discovered and she cited nucleosome positioning and non-coding RNAs.

              Epigenetics is related to certain important phenomena of controlling gene expression. Imprinting and X-chromosome inactivation are among these phenomena. Imprinting occurs when one allele of certain gene is activated while another one is methylated in cases of gene expression by one single allele. X-chromosome inactivation is one interesting mechanism whereby women balance the quantity of gene expression in their genome by methylation of one of their X-chromosomes. As men have little gene expression in their Y-chromosome, women would have excessive gene expression if both their X-chromosomes were active. 

Biologists are interested in epigenetics researches due to the non-evident genetic causes of certain diseases. Abnormal epigenetic patterns are found in many diseases, including cancer. For example, the decrease in methylation could cause excessive or improper expression of genes such as those that express cell growth. Also, epigenetic errors could methylate tumor suppressor genes. According to Arinmondo (2012), there are chemical modifiers that can reverse epigenetic alterations and it could be a promise to therapeutic use. Some studies have focused in developing chromatin modifying agents as anticancer therapy. However, Williams (2013) argues that we do not have appropriate technology to read a person’s epigenome and it has still a long way to go.

REFERENCE LIST

Williams, S. C. P. February 26, 2013, ‘Epigenetics’, Proceedings of the National Academy of Sciences, vol. 110, no. 9, p. 3209.

Arimondo, P. B. November 2012, ‘Epigenetics’, Biochimie, vol. 94, issue 11, pp. 2191-2192

Phenotypic Plasticity

             To study evolutionary process, it is important to consider the effects of the environment on the process of development. Phenotypic plasticity is the capability of genotypes to response to environmental alterations. Thus, they can modify their phenotypic expression with morphological or physiological adjustments. Some responses are unavoidable, for example, the decrease in growth under adverse conditions for the development. According to Scheiner (1993), variances in allelic expression across environments and changes in interactions among loci are likely the causes of plasticity in a genetic level.

             As it can be seen in the figure 1 (Scheiner, 1993), the environment influences evolutionary processes in two ways. Firstly, it determinates the fitness of an individual’s phenotype by natural selection. In the second way, it establishes the phenotype interacting with the developmental process. Thus, the environment may change the phenotypic expression of a genotype (phenotypic plasticity). Additionally, the figure shows three sources of phenotypic variation: the genome, the environment and random accidents of development. These random processes are due to internal events that causes change; it is distinct from the environmental interaction. Therefore, Scheiner supposes that different phenotypes may come from genetically identical individuals grown in identical environments.

               

Figure 1 (Schiner, 1993) 

                It is important to clarify that plasticity is specific to a trait instead of an entire genotype. Some traits may be plastic responding to certain environment, while another may not. There are labile and fixed traits. The labile ones get change as fast as the environment while some traits are fixed during development such as size following metamorphosis in holometabolous insects. (Scheiner, 1993)

                 Scientists try different measures to quantify the phenotypic plasticity. To know if the plasticity of a trait is adaptive, it is measured the relationship between the trait plasticity and fitness averaged across environments. Despite, the study of phenotypic responses across more than two environments may be complex. The interests in these analyses are to compare populations and to predict responses to selection. The probable processes of adaptation to environmental stress resulting from plasticity could provide predictions about species responses to the climate change. Therefore, these studies could be the key to ensure the stability of populations before global changes.  

                Given this information, do you think it is possible to predict species responses to selection? Could we predict the behaviour of populations before global change? What do you think about the importance of plasticity?

REFERENCE LIST

Scheiner, SM 1993, ‘Genetics and evolution of phenotype plasticity’, The Annual Review of Ecology, Evolution, and Systematics, no. 24, pp. 35-68.

An Introduction: Genetics and Evolution

As basic concept of genetics, phenotype is all morphological, physiological and even behavioural characteristics of organisms. According to Weatherall (2009), phenotype is all the observable characteristics. Examples of phenotype are flower color, eye color, blood type and the amino acid sequence of a protein. The genotype is the genetic constitution of the individual, the genes it inherited from its parents. The genotype with an environmental interaction establishes the phenotype.
          One significant question over genetics is, due to complex interactions of thousands of genes between themselves and the environment, how to relate gene organization and function to a phenotype. This is an issue which many researchers are focused on. For the moment, evolutionary biologists know that mutations can alter the structure of genes products, giving a rise to diversity of phenotypes. Each genetic makeup can provide a phenotype with adaptive capabilities to certain individuals and for a given environment at a given time. This means natural selection does not act on genes or genetic characteristics singly, but on individuals with all their genetic load.

Therefore, genetics is a determinant factor in the evolutionary processes. Indeed, heritable fitness variation within a population is a key factor to evolution by natural selection (Zhang, 2010). Although Darwin understanding that a factor of heritage and a source of variability are required in his theory, he did not know those mechanisms.   

Biologists strove to join Darwinian evolution with Mendelian genetics creating the synthetic theory of evolution, or Neo-Darwinism. They concluded that evolution is a gradual process including mutation, selection and drift which is explained by population genetic (Zhang, 2010). In this way, it was discovered that the primary source of variability are mutations and genetic recombination. Mutations introduce genetic newness to genotypes, while genetic recombination by sexual reproduction creates variability due to the arising of multiple different gene combinations.

The greater the diversity, the greater the probability of a population to adapt to changes occurring in this environment. Homogeneous population could be eliminated from their habitat if change occurs. Over time, certain genes and consequently certain characteristics are eventually eliminated from populations to the detriment of others that are implanted, thereby causing evolution.

From these basic concepts and knowledge, I will post here related matters objects of scientific study.

REFERENCE LIST

Zhang, J. (2010) Evolutionary genetics: progress and challenges. In: Evolution Since Darwin: The First 150 Years. M. A. Bell, D. J. Futuyma, W.F. Eanes, J. S. Levinton, eds., Sinauer, Sunderland, Mass. Pp. 87-118

Weatherall, David J(Mar 2009) Genotype–Phenotype Relationships. In: eLS. John Wiley & Sons Ltd, 
Chichester. http://www.els.net [doi: 10.1002/9780470015902.a0003403.pub2]