Ectodermal Dysplasia 1, Hypohidrotic, X-Linked

A number sign (#) is used with this entry because X-linked hypohidrotic ectodermal dysplasia-1 (ECTD1) is caused by mutation in the gene encoding ectodysplasin-A (EDA; 300451) on chromosome Xq13.

Description

Some ectodermal dysplasias are here classified as congenital disorders characterized by abnormal development in 2 or more ectodermal structures (hair, nails, teeth, and sweat glands) without other systemic findings.

Hypohidrotic, or anhidrotic, ectodermal dysplasia (HED/EDA) is characterized by a triad of signs comprising sparse hair (hypotrichosis), abnormal or missing teeth (anodontia or hypodontia), and inability to sweat (anhidrosis or hypohidrosis). Typical clinical manifestations also include dryness of the skin, eyes, airways, and mucous membranes presumably due to the defective development of several exocrine glands. Hypohidrotic ectodermal dysplasia can be associated with dysmorphic features (forehead bumps, rings under the eyes, everted nose, and prominent lips) and occasionally with absent nipples. Ectodermal dysplasia-1, due to mutation in the EDA gene, is the most frequent form of hypohidrotic ectodermal dysplasia (summary by Cluzeau et al., 2011).

Clinical Features

Pinheiro and Freire-Maia (1979) reported a large Brazilian kindred with multiple affected individuals over 6 generations. Thirteen males were affected and 27 females were variably affected. Males had a characteristic facies, with frontal bossing, maxillary hypoplasia, 'saddle' nose, prominent lips, and linear wrinkles around the eyes. Teeth were often missing or misshapen. Hair was fine, dry, brittle, and sparse, and skin was thin, glossy, smooth, and dry with hypohidrosis. Females had sparse, thin scalp hair and mosaic patchy distribution of body hair. Many females had abnormal teeth and mild hypohidrosis. Pinheiro and Freire-Maia (1979) suggested that the syndrome had 2 forms: a 'major' form in males, and a 'minor' form in females. They estimated that mothers of affected males showed 66% penetrance, daughters of carriers showed 81% penetrance, and daughters of affected males showed 49% penetrance. Pinheiro et al. (1981) revised these numbers to 72%, 77%, and 67%, respectively.

Nakata et al. (1980) reported the clinical findings in 23 affected males from 15 families as well as 21 mothers. All affected males had the characteristic clinical findings of HED, including very sparse hair, small, misshapen, and missing teeth, diminished sweating with a history of frequent hospitalization for high fevers during infancy, and symptoms of atrophic rhinitis and decreased salivary secretions. All affected males had normal nails. Seven other male family members were reported to be affected, including 6 who died before 1 year of age with high fevers. Two affected males had total anodontia, several had missing teeth bilaterally, and most had significantly smaller upper incisors, upper and lower first molars, and lower second molars than controls. Seventy-three percent of obligate heterozygous females had 1 or more congenitally missing teeth, and most had smaller teeth, One daughter had severe thinning of the hair and several mothers were known to wear wigs. The findings suggested that HED is a highly penetrant X-linked trait with intermediate expression in the heterozygous female.

Autopsy in 1 patient (Reed et al., 1970) showed absence of mucous glands in the pharynx, larynx, trachea, and large and small bronchi. The finding was thought to be the basis for the observed increase in susceptibility to respiratory infections. Mucous glands were also absent in the upper esophagus and hypoplastic in the colon.

In 13 HED families with 16 affected males, Saksena and Bixler (1990) described detailed facial characteristics, including prominent forehead, narrow and short maxillary regions, small palatal depth, small cranial length, and depressed nasal root and bridge. Heterozygous carriers showed variable manifestations of these facial characteristics. Crawford et al. (1991) reported on the dental changes in 34 British families. Abinun et al. (1996) reported immunodeficiency characterized by persistence of failure of antibody production on exposure to polysaccharide antigen in a 4-year-old patient with anhidrotic ectodermal dysplasia. Halperin and Curtis (1942) reported an association with mental retardation, but this is usually not a feature of the syndrome.

Affected Females

In the family reported by Roberts (1929), skin involvement in heterozygous females was patchy. Singh et al. (1962) described a severe case in a 27-year-old Sikh woman in India. Two brothers had died of the disease. Whether this was a homozygous affected or a heterozygous manifesting female is uncertain, especially since no information was provided on whether the father was affected. Consanguineous matings of the types that are expected to result in homozygous affected females are frequent in some Indian groups.

Richards and Kaplan (1969) described a female infant with neonatal pyrexia due to anhidrotic ectodermal dysplasia. The mother had 'somewhat sparse hair and wrinkled appearance of the eyelids.' Two of the sisters and 4 of the brothers of the mother, as well as her mother and the son of a maternal uncle, had absence of upper canine teeth. The authors suggested autosomal dominant inheritance. Earlier, Kerr et al. (1966) had expressed the view that dominant inheritance had not adequately been documented. The family of Richards and Kaplan (1969) is consistent with X-linked inheritance with partial expression in heterozygous females.

Nakata et al. (1980) found small teeth and congenital missing teeth as rather consistent findings in carriers. Happle and Frosch (1985) demonstrated that heterozygotes show a pattern of lyonization that corresponds, over the back, for example, to lines of Blaschko. They reproduced one of Blaschko's original drawings (Blaschko, 1901) and showed a photograph of the iodine-starch test of the back of a patient showing the same lines with a typical V-shape over the spine. (Happle (1991) presented diagrams of the lines of Blaschko with a demonstration of the findings on the scalp where spiral streaks converge on the vertex.) Carrier detection is often possible by dental examination; whole-back sweat tests are useful to supplement the dental examination when necessary (Harper, 1986).

Clark et al. (1990) found abnormal skin temperature patterns consistent with altered peripheral vascular perfusion in heterozygotes for XHED. Clarke and Burn (1991) found positive sweat tests as indicated by mosaic hypohidrosis following the lines of Blaschko in 35 of 36 obligate female carriers. Furthermore, female carriers identified in HED families on the basis of unequivocal dental signs were found to give positive results in 44 of 47 cases. The authors noted that diagnosis of females carriers is important to optimize neonatal and pediatric care for affected male infants who may be at substantial risk of death in infancy.

At the first Human Gene Mapping Workshop, in New Haven in 1973, Gerald and Brown (1974) noted that a girl with severe manifestations of HED and an X;9 translocation had been described by Cohen et al. (1972). MacDermot and Hulten (1990) reported follow-up on the girl with the X;9 translocation and confirmed the diagnosis of hypohidrotic ectodermal dysplasia with moderately severe mental retardation. The patient had been born of a nonconsanguineous couple with maternal age 43 and paternal age 42. During her first year, she had severe feeding problems necessitating gastrostomy and episodes of unexplained hyperpyrexia. She was noted to have fine and sparse hair but with no family history. The diagnosis of HED was not made until she was 1 year old.

Zankl et al. (2001) described female monozygotic twins with X-linked hypohidrotic ectodermal dysplasia due to a de novo t(X;9) translocation disrupting the EDA gene and nonrandom inactivation of the normal X chromosome. Like the patient studied by MacDermot and Hulten (1990), these patients were severely affected. One of the girls died unexpectedly at 2.5 years of age of severe acute respiratory distress. Autopsy revealed complete absence of sebaceous, submucous, and eccrine sweat glands as well as severe hypoplasia of hair follicles. The left main stem bronchus was obstructed by mucous debris. Lack of normal tracheobronchial secretions leading to complete tracheal obstruction by mucous debris was considered the likely cause of death.

Lexner et al. (2008) reported 2 unrelated Danish carrier females of pathogenic EDA gene mutations with clinical features of HED, including agenesis of multiple teeth and reduced salivary flow. Both affected women had activation of the mutated allele in 82% and 84% of cells, respectively. None of 41 additional females with sparse clinical features had skewed X-chromosome inactivation.

Diagnosis

Prenatal Diagnosis

Gilgenkrantz et al. (1989) described a 3-generation family and reported that in 2 cases prenatal diagnosis was performed by skin biopsy and by fetoscopy. Specimens showed a complete lack of pilosebaceous units.

Zonana et al. (1990) diagnosed HED at 9 weeks' gestation by linkage analysis involving chorionic villus biopsy material. On the basis of advice, the pregnancy was terminated. The diagnosis could not be confirmed by histologic analysis because of the early developmental stage of the fetus.

Mapping

Chautard-Freire-Maia et al. (1981) presented evidence against linkage with Xg.

At the first Human Gene Mapping Workshop, in New Haven in 1973, Gerald and Brown (1974) noted that a girl with an X;9 translocation had been described by Cohen et al. (1972). In a comment at the first Human Gene Mapping Workshop, the late Dr. P. J. L. Cook suggested that the disease gene might be located in Xq12, based on the girl with HED and the X;9 translocation. This statement was not recorded in the workshop report, but was communicated by Dr. Cook to Albert de la Chapelle in 1981 (de la Chapelle (1982, 1990)). This was the first instance of deducing the location of a disease gene on the X chromosome from the location of the breakpoint in the X/autosome translocation in a female with the disease. Since the break in the X chromosome was in Xq12, the possibility that the EDA locus is situated there was raised (de la Chapelle, 1982). Zonana et al. (1988) restudied the fibroblast cell line from this patient and demonstrated a karyotype of 46,X,t(X;9)(q13.1;p24) with an Xq breakpoint distal to the one previously reported. Turleau et al. (1989) reported a second case of HED in a female with an X/autosome translocation; the breakpoint appeared to be at the same site, Xq13.1.

From comparative mapping studies of the X chromosomes of mouse and man, including mapping of 'Tabby' (Ta), the presumed mouse homolog of EDA, Buckle et al. (1985) concluded that the assignment of EDA to Xq12 is consistent. MacDermot et al. (1986) found a lod score of 2.66 at a recombination fraction of 0.06 for linkage between EDA and DXYS1. Kolvraa et al. (1986) found no recombination in 9 informative meiotic events (7 of which were phase known) for EDA and a RFLP located in the region Xp11-q12. This gave a maximum lod score of 2.41 at a recombination fraction of 0.00.

Clarke et al. (1987) conducted a linkage study of 24 families with hypohidrotic ectodermal dysplasia. They confirmed the previously suggested linkage to DXYS1 and established linkage to probes DXS14 and DXS3. They concluded that the HED locus lies in the centromeric region between DXYS1 on the long arm and DXS14 on the short arm, probably on proximal Xq. MacDermot et al. (1987) analyzed linkage with 5 polymorphic DNA markers in 30 families. Hanauer et al. (1987) added linkage data to those previously reported to arrive at a total lod score of 12.07 at theta = 0.05 for linkage to DXYS1. They concluded that the EDA locus is proximal to DXYS1; earlier reports placed EDA distal to DXYS1. Clarke et al. (1987) found that EDA is closely linked to PGK1 (311800), with a lod score of 13.30 at theta = 0.02. A full review was provided by Clarke (1987). Zonana et al. (1988) extended their previous study of linkage by analyzing 36 families by means of 10 DNA probes at 9 marker loci. They concluded that the disorder is located in the region Xq11-q21.1, probably Xq12-q13. Hanauer et al. (1988) concluded that DXS159, located in Xq11-q12, is the closest marker to EDA. Physical mapping placed DXS159 proximal to the Xq12 breakpoint of an X/autosome translocation found in a female with EDA. On the basis of studies in 24 families using 7 DNA markers from the centromeric region of the X chromosome, Kruse et al. (1989) concluded that the most likely location of the EDA locus is 10 cM distal to DXS1 on proximal Xq between DXS1 and DXYS1. Zonana et al. (1992) found no recombination between EDA and the DXS732 locus. This locus was defined by a conserved mouse probe pcos169E/4, otherwise known as the DXCrc169 locus, which cosegregates with the mouse Ta locus. The absence of recombination lends support to the hypothesis that the DXCrc169 locus in the mouse and the DXS732 locus in humans contain candidate sequences for the Ta and EDA genes, respectively. Limon et al. (1991) described a de novo translocation (X;1)(q13.1;p36.33) in a 2-year-old girl with typical clinical features of X-linked anhidrotic ectodermal dysplasia. The breakpoint at Xq13.1 was approximately the same as that described in 2 earlier translocations.

Molecular Genetics

In a male patient with the classic EDA phenotype, Zonana et al. (1993) identified a partial deletion at the DXS732 locus within the Xq12-q13 region, with a unique junctional fragment identified in the proband and in 3 of his maternal relatives. This was the first determination of carrier status for EDA in females by direct mutation analysis. Since the DXS732 locus contains a highly conserved sequence, Zonana et al. (1993) concluded that it may be a candidate locus for the EDA gene.

Kere et al. (1996) found that the EDA gene was disrupted in 6 patients with X/autosome translocations or submicroscopic deletions; 9 patients had point mutations (see, e.g., 300451.0001). The authors noted that mutations were detected in only one-tenth of patients studied.

In 17 of 18 families with X-linked hypohidrotic ectodermal dysplasia, Monreal et al. (1998) identified mutations in the EDA gene, including 12 missense, 1 nonsense, and 4 deletion mutations (see, e.g., 300451.0005-300451.0010 and 300451.0023).

In 2 Han Chinese brothers with X-linked hypohidrotic ectodermal dysplasia and their unaffected mother, Huang et al. (2006) identified a 1-bp insertion in the EDA gene (300451.0016). The mutation was not found in the maternal grandparents or in 200 controls. The authors stated that this was the first de novo insertion identified in the EDA gene.

Van der Hout et al. (2008) identified mutations in the ED1 gene in 24 (57%) of 42 unrelated European probands with hypohidrotic ectodermal dysplasia.

Lexner et al. (2008) identified 16 different mutations in the EDA gene in 19 Dutch families with X-linked HED. Nine of the mutations were novel. In addition, multiplex ligation-dependent probe amplification (MLPA) analysis detected a deletion in exon 1. There were no genotype/phenotype correlations.

Van Steensel et al. (2008) described a 19-year-old Pakistani man, born of consanguineous parents, who presented with severe generalized hyperkeratosis. The patient reported occasional blistering as a child, inability to sweat, although he could tolerate heat, slow-growing hair, and congenital absence of teeth, except for 2 conical teeth in the upper jaw. He attended business school and was otherwise healthy. Physical examination revealed a massive, hystrix-like dark brown malodorous hyperkeratosis covering large skin areas, which diminished toward the extremities and was absent on the palms and soles. Scalp hair was thinly implanted, dry and wiry, and pubic and axillary hair were absent. Histologic examination of a skin biopsy showed acanthosis and hyperkeratosis, as well as vacuolar degeneration of keratinocytes, in the upper spinous layer. There were prominent keratohyaline granules and large lamellar bodies in the granular layer. Electron microscopy showed numerous vacuoles and lipid droplets in and around corneocytes and many giant keratohyaline granules. Van Steensel et al. (2008) noted that the findings were not consistent with acanthosis nigricans and suggested that this disorder could be classified as a hystrix-like ichthyosis. Despite the initial diagnosis of Lelis syndrome (608290) in this patient, the facial features, hypotrichosis, and anodontia were reminiscent of HED, leading van Steensel and van der Hout (2009) to analyze the EDA gene (300451), which revealed a known causative missense mutation, R156H (300451.0007). Van Steensel and van der Hout (2009) suggested that Lelis syndrome may be a manifestation of X-linked HED.

Associations Pending Confirmation

In a 13-year-old male with mild symptoms of hypohidrotic ectodermal dysplasia in whom mutation was excluded in known causative genes, as well as in the TRAF6 gene (602355), Wisniewski and Trzeciak (2012) identified a hemizygous frameshift mutation in exon 3 of the XEDAR gene (252delG; 300276.0001). Physical examination of the patient revealed hyperthermia, dry skin, hyperpigmentation around the eyes, sparse hair, eyebrows, and eyelashes, and hypodontia with irregularly shaped teeth. An iodine test revealed deficiency of sweat glands. The mutation was predicted to encode a nonfunctional, truncated receptor that lacks both the transmembrane domain and an intracellular domain that interacts with the TRAF6 gene. The mutation was thought to occur de novo because it was not found in the patient's mother or sister. The mutation was also not found in 46 other HED patients, in 150 healthy control individuals, or in the 1000 Genomes Project database.

Inheritance

Ferrier et al. (2009) studied a family in which X-linked hypohidrotic ectodermal dysplasia (XHED) was transmitted from father to son by paternal sex chromosome heterodisomy. The proband was first seen at 5 years of age and had sparse eyebrows and sparse, fine, slow-growing scalp hair. Hypodontia was evident; none of the teeth were peg-shaped. He had normal nail growth and adequate eyelashes, and perspiration appeared adequate with visible sweat pores on dermatoglyphs. His father had a similar history and manifestations. Family history revealed an affected paternal great-great uncle, a male first cousin once removed, and 2 male second cousins. The first cousin once removed was examined and found to have a mild form of ectodermal dysplasia, with a paucity of sweat pores, fine scalp hair with early loss, minimal eyebrows, short and sparse eyelashes, and hypodontia. Teeth were not peg-shaped, androgenic hair was normally distributed, and nails were not dystrophic. Analysis of the EDA gene in the 2 second cousins revealed hemizygosity for a missense mutation (R276C; 300451.0019) that was inherited from their carrier mother. Assay of polymorphic X chromosome markers in the proband and both of his parents was consistent with paternal inheritance of the X chromosome, and subsequent analysis of EDA confirmed that the proband was hemizygous for the R276C mutation segregating in his paternal family. Ferrier et al. (2009) stated that this was only the second reported case of father-to-son transmission of an X-linked condition due to sex chromosome heterodisomy (see 306700 and Vidaud et al., 1989).

Pathogenesis

Patients with HED and 'Tabby' mice lack sweat glands. Exogenous epidermal growth factor (EGF) reverses the Ta phenotype (Kapalanga and Blecher, 1990; Blecher et al., 1990). Vargas et al. (1996) studied the EGF signaling pathway in HED/Ta. Fibroblasts from HED patients had a 2- to 8-fold decrease in binding capacity for radioiodinated EGF, a decreased expression of the immunoreactive 170-kD EGF receptor (EGFR; 131550) protein, and a corresponding reduction in EGFR mRNA. Reduced expression of EGFR also was observed in Ta fibroblasts and liver membranes. Other aspects of the EGF signaling pathway, including EGF concentration in urine and plasma, were normal in both HED patients and Ta mice. Vargas et al. (1996) proposed that a decreased expression of EGFR plays a causal role in the HED/Ta phenotype.

Clinical Management

Schneider et al. (2018) administered the recombinant fusion protein Fc-EDA, consisting of the receptor-binding domain of EDA and the Fc domain of human IgG1, intraamniotically at 26 and 31 weeks' gestation to a pair of monochorionic, diamniotic twin male fetuses diagnosed prenatally with XLHED based on lack of tooth germs and because they had an affected older brother who was hemizygous for a missense mutation in the EDA gene (Y304C). The treated infants had normal sweat-duct density on their soles and produced amounts of sweat similar to those of healthy infants, whereas their older brother did not sweat at all. The twins had no hyperthermic episodes or respiratory-related hospitalizations over 22 months of follow-up, including 2 summers. Transillumination revealed 3 to 5 and 6 to 7 meibomian-gland ducts per lower eyelid in the twins, respectively, compared to only 1 gland duct in their untreated brother. Postnatal MRI and x-rays showed 10 and 8 tooth germs in the twins, compared to 3 teeth and 1 tooth germ in their untreated brother at age 5 years. An unrelated affected male fetus, who was hemizygous for a 1-bp splice site duplication in the EDA gene (c.924+1dupG), was treated with a single dose of Fc-EDA at gestational week 26 due to the limited supply of Fc-EDA. At birth he had slightly fewer sweat pores per square centimeter on his soles compared to healthy controls, and pilocarpine-induced sweat production at age 6 months was lower than in the twins, suggesting slower maturation of sweat-gland function. He had a near-normal number of meibomian glands, with 11 and 15 gland ducts per lower eyelid, and he had 9 tooth germs, compared to only 2 tooth buds in his affected 2-year-old brother. Schneider et al. (2018) concluded that prenatal treatment with Fc-EDA can restore sustained sweating ability in patients with EDA mutations that abrogate perspiration, but noted that it was unknown whether these therapeutic effects would be permanent.

Nomenclature

The acronym HED has been used in the literature to designate both hypohidrotic ectodermal dysplasia and hidrotic ectodermal dysplasia (see, e.g., 129500). In OMIM, HED is used to designate hypohidrotic ectodermal dysplasia.

Freire-Maia and Pinheiro (1980) insisted that 'anhidrotic ectodermal dysplasia' is a poor designation because the condition is, in fact, hypohidrotic. They considered 'hypohidrotic X-linked ectodermal dysplasia' misleading because there are 2 X-linked ectodermal dysplasias, this disorder and Lenz dysplasia. (Actually ectodermal dysplasia does not seem to be a conspicuous feature of the latter condition.) The designation they proposed, Christ-Siemens-Touraine (CST) syndrome, runs the risk of confusion with the CRST syndrome (calcinosis-Raynaud-sclerodactyly-telangiectasia; see 181750), which has phenotypic similarities to the Osler-Rendu-Weber syndrome (187300).

Animal Model

Mouse Model

The 'Tabby' mouse represents the murine equivalent of anhidrotic ectodermal dysplasia. Srivastava et al. (1997) cloned the mouse Ta gene, the homolog of the human EDA gene, and confirmed that it was mutated in 2 independent 'Tabby' mouse strains.

Gaide and Schneider (2003) showed that treatment of pregnant Tabby mice with a recombinant form of EDA, engineered to cross the placental barrier, permanently rescued the Tabby phenotype in the offspring. Notably, sweat glands can also be induced by EDA after birth. This was said to be the first example of a developmental genetic defect that can be permanently corrected by short-term treatment with the recombinant protein.

By administration of a recombinant fusion protein, Fc-EDA, consisting of the receptor-binding domain of EDA and the Fc domain of human IgG1, into the amniotic fluid of Eda Y/- male mice, or Eda -/- mice of either sex, Schneider et al. (2018) prevented the development of XLHED. Mice without the neonatal Fc receptor did not have their condition corrected, whereas all fetuses from the same litter with expression of the neonatal Fc receptor were corrected and exhibited only a few remnants of the disease phenotype. Administration at embryonic day 12.5, 13.5, or 14.5 showed similar results; the authors observed that structures formed close to the time of injection were rescued more efficiently than those formed at later stages of development. Schneider et al. (2018) concluded that Fc-EDA provided in amniotic fluid must first enter the organism in a manner that is dependent on the neonatal Fc receptor, presumably through the gut, before it can act on developing EDA-dependent structures.

Bovine Model

A probably homologous X-linked condition occurs in cattle (Ohno, 1973). In black-and-white German Holstein cattle with X-linked anhidrotic ectodermal dysplasia, Drogemuller et al. (2001) identified partial deletion of the bovine ED1 gene. In red-and-white German Holstein cattle with X-linked anhidrotic ectodermal dysplasia, Drogemuller et al. (2002) identified a splice site mutation in the bovine ED1 gene.

Canine Model

Casal et al. (1997) described X-linked ectodermal dysplasia in a male German shepherd pup with symmetrical areas of hairlessness as well as missing and misshapen teeth.

Casal et al. (2007) used the canine model of X-linked hypohidrotic ectodermal dysplasia (XLHED) to study the developmental impact of ectodysplasin A (EDA) on secondary dentition, since dogs have dentition similar to that in humans with respect to both development and morphology of teeth. Also, clinical signs in humans and dogs with XLHED are virtually identical, whereas several are missing in the murine equivalent. Casal et al. (2007) found that in the dog model the genetically missing EDA was compensated for by postnatal intravenous administration of soluble recombinant EDA. Untreated XLHED dogs have an incomplete set of conically shaped teeth similar to those seen in human patients with XLHED. After treatment with EDA, significant normalization of adult teeth was achieved in 4 of 5 XLHED dogs. Moreover, treatment restored normal lacrimation and resistance to eye and airway infections and improved sweating ability.

History

Thurnam (1848) reported 2 male first cousins and described a carrier, their maternal grandmother, with a hereditary syndrome associated with sparse hair, missing teeth, and dry skin. Hypohidrotic ectodermal dysplasia was the condition affecting the 'toothless men of Sind,' members of a Hindu kindred which resides in the vicinity of Hyderabad and was described by Darwin (1875) and by Thadani (1934). Darwin (1875) wrote as follows: 'I may give an analogous case, communicated to me by Mr. W. Wedderburn, of a Hindoo family in Scinde, in which ten men, in the course of four generations, were furnished, in both jaws taken together, with only four small and weak incisor teeth and with eight posterior molars. The men thus affected have very little hair on the body, and become bald early in life. They also suffer much during hot weather from excessive dryness of the skin. It is remarkable that no instance has occurred of a daughter being affected...though the daughters in the above family are never affected, they transmit the tendency to their sons: and no case has occurred of a son transmitting it to his sons. The affection thus appears only in alternate generations, or after long intervals.' Hutt (1935) called attention to Darwin's description.

Graves (1963) wrote a charming, highly literate account of the large southern Mississippi group afflicted with this disorder. The group was also described in the WPA guide on Mississippi (WPA, Federal Writers' Project, 1938) where they were referred to as 'Whitaker Negroes.' They were said to 'have sub-normal sweat glands; consequently, in warm weather, they have to be near a pool or creek in which they can immerse themselves. Frequently the Negroes take buckets of water to the field with them, turning the water over their heads to soak their clothing...they have few teeth, perhaps 2 or 3 at the top and a few below, and these are fine and pointed....Their hair is fine and silky but thin and short...their peculiarities seem to be inherited only by the male children, the females being normal.'