J Med Genet: first published as 10.1136/jmg.3.3.230 on 1 September 1966. Downloaded from Review Article

7. med. Genet. (I966). 3, 230.

A Proposed Classification of Genetically Determined Mosaicism in Man MARGARET J. COREY and JAMES R. MILLER From the Department of Paediatrics, University of British Columbia, Vancouver, B.C., Canada Terms are often used in medical literature with- Biological Mosaicism out a true awareness of their original meaning and According to Wilson (I928), Roux was the first previous usage. Although the recent advances in to use the term '' in a biological context. In human cytogenetics have focused attention on the his work Uber Mosaikarbeit und neuere Entwick- phenomenon of mosaicism, in fact the term lungshypothesen (I892-I893) he states, 'The develop- 'mosaic' was used in biology and human genetics ment of the frog gastrula and of the embryo formed long before human cytogenetics was a recognized from it is from the second cleavage division onward discipline. a mosaic work, consisting of at least four vertically The term 'mosaic' was borrowed, originally as a independent developing pieces' (from Wilson, purely descriptive term, from the realm of fine arts. I928). The implication here is that all living The term in its original context leaves no room for multicellular organisms in which differentiation confusion or ambiguity. It is defined as 'the process occurs are mosaics. During development differenti-copyright. of producing pictures or patterns by cementing ation consists of a programmed progressive together small pieces of stone or glass, etc. of various mosaicism and all human beings are mosaics. This colours; pictures or patterns thus produced; the form of mosaicism, recognized as an aspect of constructive or decorative materials of these' normal development, will not be discussed further (Oxford Shorter Dictionary). This clarity of its and in view of its universality will not be considered meaning seems to make it a particularly appropriate in our classification. However, it should be descriptive term for certain naturally occurring

acknowledged that there may be some difficulty inhttp://jmg.bmj.com/ patterned organisms (pathological or normal) where distinguishing between normal mosaicism and living tissues show the clear-cut difference of a anomalous or pathological conditions in which the mosaic. When used outside the realm of the mosaicism is at a cellular level and involves a single inanimate it should always be thought of as tissue. 'mosaic-like', and any definition or classification Pathological mosaicism is basically a develop- should encompass all such mosaic-like situations mental problem. Although of great interest to regardless of their origin or level of observation. geneticists, it is not concerned with the classical

Such situations are sufficiently rare to attract wide- Mendelian concept of the transfer of hereditary on September 26, 2021 by guest. Protected spread interest. They have been extensively characteristics from generation to generation but investigated in plants (see Cramer, I954 and rather with the branch of genetics which deals with Neilson-Jones, 1937, for reviews). More recently the transfer and utilization of this hereditary general discussions of their occurrence in the information from the undifferentiated single-celled animal kingdom (Hannah-Alava, I960) and in mice zygote to the millions of specialized cells of the (Russell, I964) have been published. Recent dis- complete organism. In this context, therefore, cussions of the subject (Lancet, I965; Bain and a mosaic will be defined as an organism which displays Scott, I965; Engel, I965) have tended to ignore the abnormal genotypic or phenotypic variation from cell diversity of origin of mosaics, and the purpose of the to cell within the same tissue or genotypic variation present communication is to propose a general between tissues. classification of mosaicism in man. Environmental Factors. Mosaicism may occasion- Received December 3, I965. ally result from the action of non-genetic factors. 230 J Med Genet: first published as 10.1136/jmg.3.3.230 on 1 September 1966. Downloaded from

Proposed Classification of Genetically Determined Mosaicism in Man 23I

Variation in biochemical or other environmental factors examples of demonstrable functional differences in may cause the same gene to be expressed differently the cells, i.e. situations in which given genes are under different circumstances or even in different parts known to function in only some cells. To avoid of the body. This mosaicism may closely mimic the exclusion of phenotypic variants indistinguishable genetically determined form in some instances. An origin, we prefer a more example of this would be the heterochromia iridum from those of known observed in Homer's syndrome (Robinson, Dikrainian, inclusive definition. Causative agents may be and Roseborough, i965), which is similar to the entirely extranuclear, extracellular, or even remote mosaicism of eye colour found in the Waardenburg from the affected areas. syndrome and other genetically-determined hetero- chromias. (i) Caused by Specific Genes-Pattern Loci. Early genetic investigation of mammalian coat An outline of the proposed classification of colour revealed the existence of specific genes for genetically determined mosaicism is presented in mosaics or patterned coats, just as there were genes the Table. It should be pointed out that though for coat colour itself. These can be inherited as some of the categories have not yet been described simple Mendelian factors. The white forelock and heterochromia associated with the Waardenburg TABLE syndrome is the most striking example of this type of human mosaicism (Waardenburg, Franceschetti, A PROPOSED CLASSIFICATION OF GENETICALLY DETERMINED MOSAICISM IN MAN and Klein, I96I). This condition is determined by an autosomal dominant gene with variable expressiv- is no A MOSAIC is an organism which displays abnormal genotypic or ity and incomplete penetrance, and there phenotypic variation from cell to cell within the same tissue or reason to doubt that the mutant gene is present in genotypic variation between tissues. all cells of an affected individual. Other examples of A: Genetically uniform mosaics this phenomenon in man would include the cafe au (1) Caused by specific genes-pattem loci (2) Caused by variation in gene expression lait spots observed in neurofibromatosis and (a) Lyon inactive-X theory effect xeroderma pigmentosa (Waardenburg et al.,- I96I; (b) Variegated-type position copyright. (c) Controlling elements Stern, I960). Russell (I964) has excluded 'spotting (3) Non-genetic inheritance-cytoplasmic structures genes' from her discussion of functional mosaicism B: Chimeras-a is an organism that is not genetically in the mouse on the grounds that they probably uniform throughout (1) Post-zygotic represent a process of histological differentiation. (a) Involving cells derived from a single zygote (i) somatic crossing-over (ii) somatic mutation (2) Caused by Variation in Gene Expression. (iii) somatic segregation and reduction (iv) polyploidy (a) LYON PRINCIPLE OF MAMMALIAN X CHROMO- (b) Involving cells derived from different zygotes certain In order to explain http://jmg.bmj.com/ (i) chorionic vascular anastomosis SOME INACTIVATION. (ii) maternal-foetal exchange observations relating to the distribution of colour (iii) transfusion and transplantation (iv) radiation chimeras pattern in mice heterozygous for two (2) Peri-zygotic chimerism X-linked mutants, Lyon (I96I) proposed that early in embryonic development of the female mammal, one X in each cell becomes genetically inactive and that maternally and paternally derived in man, they are included because it seems reason- X are inactivated randomly. Each able to expect that examples will be found in the female thus becomes a mosaic for the expression of on September 26, 2021 by guest. Protected future. Mosaics can be divided into two basic genes on the X chromosome and two distinct cell those that are genetically uniform and categories: populations will exist in heterozygotes for X-linked those in which the mosaicism results from a genetic observations on difference. This second category is chimerism. genes. Lyon later presented human X-linked conditions which appear to support her theory (I962, i963). Several lines of evidence have tended to substantiate mosaicism for X Genetically Uniform Mosaics chromosome activity in human (Beutler, Included in this class are all organisms with Yeh, and Fairbanks, I962; Davidson, Nitowsky, mosaic-like appearance which are not demonstrably and Childs, i963), and the Lyon hypothesis that the due to genetic differences between the pheno- normal human female is a mosaic of X chromosome typically or functionally different cells. Russell activity can be considered as a general principle. (I964), in her discussion of functional mosaicism in the mouse, includes in this category only Since this phenomenon is a normal aspect of all J Med Genet: first published as 10.1136/jmg.3.3.230 on 1 September 1966. Downloaded from

232 Corey a]rnd Miller human females it might be considered out of place in loci are under the influence of nuclear elements this classification. We include it because of its signifi- which seem to be similar to the heterochromatin of cance in the expression of X-linked mutants in the Drosophila. Both workers have drawn attention to female and the understanding of certain fundamental and have suggested aspects of embryogenesis in mammals as well as its the similarity of the phenomena relation to certain abnormal conditions which will be a similar underlying mechanism. McClintock discussed. (I96I) has drawn a parallel between controlling elements in corn and those in bacteria and bacterio- (b) VARIEGATED-TYPE POSITION EFFECT. Muller phage where precise mechanisms for control of (1930) in classifying X-ray induced changes in gene activity are known to exist. Brink (i964b) has found an unexpected class specifically indicated the relation of differential which he called 'ever-sporting displacement'. This pyknosis and gene control and has suggested a class was characterized by chromosomal rearrange- similarity of mechanism involving controlling ments which resulted in a mottled phenotype. This elements in corn, variegated type position effects mottled phenotype resulted only when a gene that in Drosophila and mice, and in the inactivation of was normally located in a euchromatic section of a the female mammalian X chromosome. chromosome became involved in a translocation and (3) Non-genetic Inheritance-Extranuclear was placed next to heterochromatin. Variegation Structures. The significance of non-genetic or was attributed to and dependent on the abnormal extranuclear structures as a cause of human cellular proximity of the affected loci to a region of hetero- mosaicism has yet to be demonstrated. The chromatin. That no permanent change in the possibility that extranuclear structures can contri- affected loci occurred was shown by a return to bute to tissue mosaicism is well established in normal activity when the locus was returned to its plants where chloroplast variation is a known normal position (Kaufmann, I942). The extent of source of mosaicism (Cramer, 1954; Neilson-Jones, variegation could be influenced by environmental I937). The existence of self-replicating structures agents such as temperature and the presence of a in the cytoplasm (Catcheside, I958) and the sug- different heterochromatin (Gowen and Gay, 1934). gestion that extranuclear structures may be copyright. an The influence of the heterochromatin was not important influence on subsequent structure (Mar- restricted to immediately adjacent loci, but showed kert, I964; Stern, 1957) increase the likelihood of a spreading effect along the transposed segment. the hypothesis that accidents or changes in This phenomenon recently has been observed in non-nuclear structures may lead to subsequent mice (Russell and Bangham, I96I; Cattanach, mosaicism. I963) in which a coat colour gene normally on an autosome was translocated to the X chromosome, Chimeras resulting in a mosaic coat colour in females. The http://jmg.bmj.com/ similarity in phenotype occurring from a trans- The term 'chimera' is a borrowed descriptive location to heterochromatin and to the X chromo- term. The original chimera was a Greek mythologi- some suggested some relation between the mechan- cal being with a lion's head, a goat's body, and a isms involved in the normal X chromosome serpent's tail. The term was first used in a bio- inactivation and the heterochromatic influence on logical context by Winkler (I907) to describe the genes placed abnormally close to heterochromatin. plant he created by grafting together two different (c) CONTROLLING ELEMENTS. Variegations of species. Such an organism is composed of two both leaf and flower colour in plants have been distinctly different cell lines exactly like the species on September 26, 2021 by guest. Protected selected for novelty value and are now a familiar of their origin. Therefore a chimera is defined as an observation. A similar phenomenon in corn in organism which is not genetically uniform throughout. which the kernels have normally pigmented areas Although Winkler demonstrated that such variega- alternating with unpigmented areas has been tion could have its origin in the combination of two extensively investigated genetically. Variegation is different individuals, it was soon recognized that associated with the activity of a specific colour gene not all variegation was attributed to different and colour variegation seems to be due to an genotypes (genetically uniform mosaics) nor did all instability ofthis locus. Rhodes (1941) demonstrated genetic variegation necessarily originate in two that the mutability was associated with the presence different individuals. of a specific genetic element other than that of the The term 'chimera' has been used extensively in colour locus and such an element was termed a botanical literature with reference to plant variega- controlling element. McClintock (I950, 1956) and tion resulting from genotypic differences as Brink (I964a) have demonstrated that these mutable distinguished from variegation due to non-genetic J Med Genet: first published as 10.1136/jmg.3.3.230 on 1 September 1966. Downloaded from

Proposed Classification of Genetically Determined Mosaicism in Man 233 factors such as chloroplasts or viruses or to geneti- pairing observed in Diptera, genetic evidence of cally uniform mosaics. It was used in the zoological mitotic recombination has been demonstrated in field to describe the erythrocyte mosaicism in cattle other organisms (Pontecorvo and Kafer, I958). observed by Owen in 1945. Cytogenetic and Conclusive evidence for somatic crossing-over in cytochemical techniques are currently revealing the man has not been demonstrated. However, both existence and complexity of chimerism in man. At cytological and genetic observations have prompted times the term has been used in its fullest sense to its suggestion as a possibility. German (I964) has include all genetic mosaics, but there has been some observed association of homologous chromosomes suggestion that the term should be restricted to from human blood cell cultures in a configuration those genetic mosaics in which the mosaicism which suggests chromatid interchange. As yet, no involves an entire genome, and exclude those proof of this interchange is available, nor has it been mosaics in which alternations of a basic genome demonstrated in vivo. Goudie (I957) has suggested give rise to a second or further cell lines (Russell, somatic crossing-over to explain the apparently I964; Chu, Thuline, and Norby, I964). We homozygous red blood cells occasionally encountered consider it preferable to include these. in low percentages in heterozygous individuals. Chimerism may be the result of an event (ii) Somatic mutation. All replication of somatic occurring at any point during the life of the material is subject to errors and mutational changes, individual (post-zygotic) or it may be the result of whether of mitotic or meiotic origin. If a meiotic an event before or at the time of fertilization and chromosome or gene is involved, the resulting the formation of the zygote (peri-zygotic). Post- individual is a mutant or is heterozygous for a zygotic chimerism may involve a single zygote and mutation. If chromosomes subsequent to fertiliza- result from an event in a single genome, which tion are involved, only part of the organism will be gives rise to one or two new genotypes which are affected and the individual will be a mosaic. The mutational alterations of the original genotype. It demonstrated frequency of spontaneous mutation may also involve cells derived from different makes it seem likely that all human beings are to zygotes by some accidental or artificial transfer of some extent mosaics. Such mutational events have tissue from one individual to another. Currently, a been suggested as the probable origin of demon- copyright. great deal of attention is being given to the possi- strated mosaicism, particularly involving pigmenta- bility of one or more genomes being involved tion (i.e. heterochromia, spotting). Davis and Shaw through participation of more than two germ cells (I964) reported a human mosaic for skin pigmenta- at the time of fertilization (peri-zygotic) and the tion involving large areas of the body and have different mechanisms which may be involved suggested an early somatic mutation as its origin. (Lancet, I965). Blood cell mosaicism has been explained by somatic and Scheinberg, I958; Atwood mutation (Atwood http://jmg.bmj.com/ Post-zygotic Chimerism. and Pepper, I96I). Neoplasms should perhaps be (a) INVOLVING CELLS FROM A SINGLE ZYGOTE. excluded from this group on the basis that they do Although all mosaics included in this category are not involve areas of uniform differentiation. in general the result of a mutational event, four (iii) Somatic segregation and reduction. We have known mechanisms will be discussed. In all four, included somatic segregation as a separate entity, the extent and nature of the mosaicism depend on though it is really a specific somatic mutational the time of the event and the type of cells deriving event. It involves a mitotic segregational disturb- from it. The detection of this form of mosaicism ance resulting in an abnormal chromosome distribu- on September 26, 2021 by guest. Protected may depend on whether or not germ tissue is tion. Its significance in human disorders is currently involved. being demonstrated cytologically in cases in which (i) Somatic crossing-over. Of the four suggested certain individuals who seem to have only a mechanisms, only somatic crossing-over has not 'partial' syndrome are found to be cytological been demonstrated in man, though its existence as mosaics, and the disorder is due to a genetic a cause of mosaicism has been recognized since imbalance in some of the cells. The widely accepted Stern's classic genetic demonstration in Drosophila mechanism for chromosome mosaicism is non- (1936). Basically, it is an exchange between homolo- disjunction or a failure of two chromatids to move gous chromosomes at (rather than meiosis). to opposite poles at mitotic anaphase resulting in If it involves a heterozygous locus, it will result in two new cell lines (2n + I, and 2n - i) as well two new genotypes, each homozygous for one allele. as an original 2n cell line (except at the first cleavage Although early opinion attributed exchange between division). According to Wilson (I928), non- mitotic chromosomes to the intimate mitotic disjunction was first described in Oenothera and J Med Genet: first published as 10.1136/jmg.3.3.230 on 1 September 1966. Downloaded from

234 Corey and Miller Datura by Gates in I908 and has subsequently been erythrocytes could be antigenically detected until widely demonstrated in both plants and animals, adulthood. Placental fusions have been described though the severity of its effect is much less in plants. in mice (McLaren and Michie, i959) and in man It is now thought to be a frequent cause of prob- (Benirschke, I96I). lems in differentiation resulting in pseudoherma- Vascular anastomoses are apparently rare in phrodism, sexual ambiguity, and intersexuality human dizygotic twins, but blood group studies (Ford, I960; Ferrier, Shepard, Gartler, and Burt, indicate that they can occur. Dunsford, Bowley, I96I; Greenblatt, Dominguez, Mahesh, and Demos, Hutchison, Thompson, Sanger, and Race (I953) I964; Ross, Holland, Kiser, and Douglas, I965). documented the first case ofred blood cell mosaicism Although mitotic pairing of homologous chromo- in twins. Since then several cases have been somes normally occurs in many species of Diptera described (Race and Sanger, I962). The most it is rare in other organisms. However, chromosome recent case, to our knowledge, is that described bv arrangements suggestive of somatic pairing have Chown, Lewis, and Bowman (I963). Although been sporadically reported and somatic reduction vascular anastomosis in cattle results in a freemartin resulting from meiotic-like segregation of paired condition, the fertility of human blood group chromosomes observed in both treated and chimeras does not appear to be diminished. untreated plant material (Huskins, I948) indicates (ii) Maternal-foetal exchange. The regularity another possible mechanism for somatic segregation. with which erythroblastosis occurs under predicted Cells produced by such a division would be circumstances has long since destroyed the concept hemizygous or homozygous for all genes. that the placental barrier between mother and (iv) Polyploidy. A number of recent reports have foetus is complete. It has been further demonstrated stressed observations of endoreduplicated chromo- that violation of the barrier is not restricted to somes in cultured cells from patients with normal soluble substances. Bleeding of the foetus into the chromosome complements (Aspillaga, Neu, and maternal circulation has been recognized as a cause Gardner, I964; Bishun and Morton, I965). of anaemia in the newborn (Chown, I954; Wiener, Polyploidy, or duplication of complete haploid sets 1948; Gunson, I957; Kirkman and Riley, I959). of chromosomes, is a rather common deviation from There is some evidence that the reverse exchange,copyright. diploidy (Wilson, I928). It is usually observed in maternal-foetal transfusion, can also occur, and it old, highly specialized or degenerate cells. For a has been proposed as a cause of plethora (Michael long time it was presumed that these cells probably and Mauer, I96I; Smith, Duhring, Greene, arose by a replication of the chromosomes without Rochlin, and Blakemore, I96I). In the first case, cell division (endoreduplication). Although this the mother, and in the second case the foetus, would process automatically immediately creates a chromo- be, at least temporarily, a blood cell chimera. There somal mosaicism, it appears to have little effect on is as yet no evidence as to whether a persisting variant gene expression. More recently, fusion of mosaicism is established in viable embryos by this http://jmg.bmj.com/ somatic cells has been suggested and it may not be mechanism. Recently, Taylor and Polani (I965) a rare event (Ephrussi and Sorieul, I962). Such have reported an XX/XY aborted foetus in which somatic cell mating has been suggested as the they suggest the chimerism may have resulted from possible origin of a 'hybrid' blood cell type in a colonization in the embryo of maternal cells bovine twin chimera (Stone, Friedman, and Fregin, through breakdown of the placental barrier. I964). Generalized mosaicism for cells of different (iii) Transfusions and transplantations. Winlder's

ploidy will also be discussed below. original chimera resulted from successfully uniting on September 26, 2021 by guest. Protected (b) INVOLVING CELLS DERIVED FROM DIFFERENT parts of two different plants. In the resulting plant, ZYGOTES. cells descended from each of the ancestral plants (i) Chorionic vascular anastomosis. Chorionic maintained their integrity, a fact that can be vascular anastomoses are common in bovine twins, demonstrated by asexual propagation. A similar and were recognized as early as I9I6 by Lillie as a human chimera is created, for a short time, when cause of the freemartin condition of the female twin blood of one individual is transfused into another. of an unlike sex pair. The classic demonstration Such transfusion is successful only if certain geneti- that this type of anastomosis could give rise to cally determined characters are the same or com- mosaicism was made by Owen (1945), who was able patible. A much closer parallel to plant grafting is to demonstrate two populations of serologically found in tissue transplants in which apparently different red cells in such twins. He concluded that animals are much more able to 'recognize' and exchange between the twins must have included reject foreign tissue. Although tissue transplanta- ancestral haemapoietic tissue, since two types of tion is still at an undeveloped and primitive stage in J Med Genet: first published as 10.1136/jmg.3.3.230 on 1 September 1966. Downloaded from

Proposed Classification of Genetically Determined Mosaicism in Man 235 human beings, it is a routine and extremely useful In the recent medical literature, several cases of procedure in certain areas of mouse genetics true hermaphrodism in humans have been reported (Snell, I958). in which the condition is attributed to XX/XY (iv) Radiation chimeras. An exception to the chromosome mosaicism (see Bain and Scott, i965, isogenicity required for successful tissue transplant for review). The possibility that such mosaicism is found following post-irradiation injection of has arisen through a series of mitotic events during bone-marrow. Since Jacobson, Simmons, Marks, early cleavage has been ruled out in most cases by Gaston, Robson, and Eldredge (I95I) and Lorenz, red blood cell antigen mosaicism which indicates Uphoff, Reid, and Shelton (i95I) discovered that that more than two germ cells must have participated mice recovered from an expected lethal dose of in the fertilization process. Since such individuals X rays if irradiation was followed by injection of are mosaics of normal yet cytologically and geneti- bone-marrow from non-irradiated individuals, this cally distinct cells, it seems likely that more than has been the object of extensive investigation both two germ cells have participated in fertilization and from the immunological and radiation treatment in such a manner that the cytological equivalent of viewpoints (Congdon, I962). Implantation and two complete zygotes is involved (termed geminism persistence of grafted cells from genetically distinct by B. Chown, i965, personal communication). donors have been demonstrated immunologically Several mechanisms have been described to account (Lindsley, Odell, and Tausche, I955), cytologically for these mosaic . Several other (Ford, Hamerton, Barnes, and Loutit, i956), and by mechanisms for production through a number of other means (Congdon, i962), and mosaicism in other organisms have been described apparent permanent chimerism is established in in cytogenetic papers. Suggested mechanisms marrow, lymphatic, and blood cells. include dispermic fertilization of the ovum plus a second meiotic product and fusion of two simul- taneous and independently fertilized ova. The latter Peri-zygotic Chimerism. Goldschmidt and has been artificially achieved in mice (Mintz, I964). that in the Katsuki (I927) determined Bombyx mori, copyright. silkworm, gynandromorphism was an inherited Summary condition resulting from the action of a single gene The term 'mosaicism' is now used extensively in which produced somatic mosaics through the human cytogenetics. In fact, the term has had long retention of one polar body nucleus. That is, the usage and specific meanings in human genetics. A eggs had two nuclei and fertilization of both nuclei proposed classification is presented. by different sperm followed by fusion of the two into a single developing mass produced mosaics for We should like to express our appreciation to sexual or somatic characteristics or both, depending Professor F. C. Fraser and Dr. B. B. Mukherjee and http://jmg.bmj.com/ upon the nature of the sperm participating in the their students for a critical reading of the manuscript. fertilization. In 1957 Beatty suggested chromosome mosaicism as an alternative to mutation to explain REFRNCES mammalian colour and sex mosaicism. Three Aspillaga, M. J., Neu, R. L., and Gardner, L. I. (I964). Chromo- possible peri-zygotic mechanisms were suggested: somal endoreduplication in a case of testicular feminisation. a Lancet, I, 937. (x) fertilization of double secondary oocyte; Atwood, K. C., and Pepper, F. J. (I96I). Erythrocyte automosaicism (2) fertilization of a double matured oocyte; and (3) in some persons of known genotype. Science, 134, 2I00. polynuclear ovarian eggs. The first mechanism is -, and Scheinberg, S. L. (I958). Somatic variation in human on September 26, 2021 by guest. Protected erythrocyte antigens. J. cell. comp. Physiol., 52, Suppl. I, 97. identical with the 'immediate cleavage', as observed Bain, A. D., and Scott, J. S. (I965). Mixed gonadal dysgenesis with by Braden (I957) in the mouse. This process XX/XY mosaicism. Lancet, I, I035. Beatty, R. A. (I957). Parthenogenesis and Polyploidy in Mammalian involves the production of two almost equally sized Development. Cambridge University Press, London. cells from the first meiotic division, both of which Benirschke, K. (I96I). Accurate recording of twin placentation. Obstet. and Gynec., I8, 334. may participate in embryo formation. One, neither, Beutler, E., Yeh, M., and Fairbanks, V. F. (I962). The normal or both may extrude a second polar body, and one, human female as a mosaic of X chromosome activity. Proc. nat. neither, or both may be fertilized. This process can Acad. Sci. (Wash.), 48, 9. Bishun, N. P., and Morton, W. R. M. (I965). Endoreduplicated result in various forms of chimerism, In/2n, 2n/2n, chromosomes. Lancet, i, II69. 2n/3n, depending upon the nature of the fertiliza- Book, J. A., and Santesson, B. (I960). Malformation syndrome in tion process. The diploid-triploid mosaics described man associated with triploidy (69 chromosomes). ibid., I, 858. -3, and - (I96I). Nuclear sex in triploid XXY human cells. by Book and Santesson (i960, I96I) and Ferrier, ibid., 2, 3I8. Bamatter, and Klein Braden, A. W. H. (I957). Variation between strains in the incidence Ferrier, Stalder, Biihler, (I964) of various abnormalities of egg maturation and fertilization in the are probably examples of this type of mosaicism. mouse. J. Genet., 55, 476. J Med Genet: first published as 10.1136/jmg.3.3.230 on 1 September 1966. Downloaded from

236 Corey and Miller

Brink, R. A. (I964a). Genetic repression of R action in maize. In (Mus musculus L.). Nature (Lond.), 190, 372. The Role of Chromosomes in Development, ed. M. Locke, p. I83. (1962). Sex chromatin and gene action in the mammalian

Academic Press, New York. X-chromosome. Amer. J. hum. Genet., 14, I35. (I964b). Genetic repression in multicellular organisms. Amer. (i963). Attempts to test the inactive-X theory of dosage Nat., 98, 193. compensation in mammals. Genet. Res., 4, 93. Catcheside, D. G. (I958). A discussion on the cytoplasm in variation McClintock, B. (ig5o). The origin and behavior of mutable loci Proc. roy. and development. Soc. B, 148, 285. in maize. Proc. nat. Acad. Sci. (Wash.), 36, 344. Cattanach, B. M. (I963). The inactive-X hypothesis and position (i956). Controlling elements and the gene. Cold Spr. Harb. effects in the mouse. (Abstract.) Genetics, 48, 884. Chown, B. (1954). Anaemia from bleeding of the fetus into the Symp. quant. Biol., 21, I97. Some between gene systems mother's circulation. Lancet, 1, 12I5. (i96i). parallels control in maize Lewis, M., and Bowman, J. M. (I963). A pair of newborn and bacteria. Amer. Nat., 95, 265. , McLaren, A., and Michie, D. Experimental studies in human blood chimeric twins. Transfusion (Philad.), 3, 494. (i959). Chu, E. H. Y., Thuline, H. C., and Norby, D. E. (I964). Triploid- placental fusion in mice. J. exp. Zool., 141, 47. diploid chimerism in a male tortoiseshell cat. Cytogenetics, 3, I. Markert, C. L. (i964). The role of chromosomes in development. In The Role Congdon, C. C. (I962). Radiation injury: bone marrow trans- of Chromosomes in Development, ed. M. Locke, p. I. plantation. Ann. Rev. Med., 13, 203. Academic Press, New York. A. Cramer, P. J. S. (s954). Chimeras. Bibliogr. genet., I6, 193. Michael, F., Jr., and Mauer, A. M. (i96i). Maternal-foetal Davidson, R. G., Nitowsky, H. M., and Childs, B. (I963). Demon- transfusion as a cause of plethora in the neonatal period. stration of two populations of cells in the human female Pediatrics, 28, 458. heterozygous for glucose-6-phosphate dehydrogenase variants. Mintz, B. (i964). Formation of genetically mosaic mouse embryos Proc. nat. Acad. Sci. (Wash.), 50, 48I. and early development of "lethal (t"1/t12)-normal" mosaics. Davis, D. G., and Shaw, M. W. (I964). An unusual human mosaic 7. exp. Zool., 157, 273. for skin pigmentation. New Engl. 7. Med., 270, 1384. Muller, H. J. (1930). Types of visible variation induced by X-rays Dunsford, I., Bowley, C. C., Hutchison, M., Thompson, J. S., in Drosophila. J. Genet., 22, 299. Sanger, R., and Race, R. R. (I953). A human blood-group Neilson-Jones, W. (I937). Chimeras: a summary and some special chimera. Brit. med. J'., 2, 8I. aspects. Bot. Rev., 3, 546. Engel, E. (I965). Mosaicism. New Engl. J1. Med., 272, 34. Owen, R. D. (I945). Immunogenetic consequences of vascular Ephrussi, B., and Sorieul, S. (I962). Mating of somatic cells in vitro. anastomoses between bovine twins. Science, 102, 400. Univ. Mich. med. Bull., 28, 347. Pontecorvo, G., and Kafer, E. (1958). Genetic analysis based on Ferrier, P., Ferrier, S., Stalder, G., Bishler, E., Bamatter, F., and mitotic recombination. Advanc. Genet., 9, 7I. Klein, D. (I964). Congenital asymmetry associated with diploid- Race, R. R., and Sanger, R. (i962). Blood Groups in Man, 4th ed., triploid mosaicism and large satellites. Lancet, I, 8o. p. 375. Blackwell, Oxford. , Shepard, Th., Gartler, S., and Burt, B. (I96I). Chromatin- Rhodes, M. M. (1940. The genetic control of mutability in maize. positive gonadal dysgenesis and mosaicism. ibid., I, II70. Cold Spr. Harb. Symp. quant. Biol., 9, 138. Ford, C. E. (I960). Human cytogenetics: its present place and Robinson, G. C., Dikrainian, D. A., and Roseborough, G. F. (i965). future possibilities. Amer. hum. Genet., 12, I04. Congenital Horner's syndrome and heterochromia iridum. copyright. -, Hamerton, J. L., Barnes, D. W. H., and Loutit, J. F. (I956). Pediatrics, 35, I03. Cytological identification of radiation-chimaeras. Nature (Lond.), Ross, G. T., Holland, J. M., Kiser, W. S., and Douglas, G. W. 1779 452. (i965). XO/XY chromosomal mosaicism and extragenital stigmata German, J. (I964). Cytological evidence for crossing-over in vitro in of Turner's syndrome in a phenotypic male. clin. Endocr., 25, human lymphoid cells. Science, 144, 298. I4I. Goldschmidt, R., and Katsuki, K. (I927). Erblicher Gyandro- Russell, L. B. (i964). Genetic and functional mosaicism in the morphismus und somatische Mosaikbildung bei Bombyx mori L. mouse. In The Role of Chromosomes in Development, ed. M. Locke, Biol. Zbl., 47, 45. p. I53. Academic Press, New York. Goudie, R. B. (I957). Somatic segregation of "inagglutinable" -, and Bangham, J. W. (i96i). Variegated-type position effects erythrocytes. Lancet, I, 1333. in the mouse. Genetics, 46, 509. Gowen, J. W., and Gay, E. H. (I934). Chromosome constitution Smith, K., Duhring, J. L., Greene, J. W., Jr., Rochlin, D. B., and and behaviour in eversporting and mottling in Drosophila Blakemore, W. S. (i96i). Transfer of maternal erythrocytes http://jmg.bmj.com/ melanogaster. Genetics, I9, I89. across the human placenta. Obstet. and Gynec., I8, 673. Greenblatt, R. B., Dominguez, H., Mahesh, V. B., and Demos, R. Snell, G. D. (1958). Histocompatibility genes of the mouse. II. (1964). Gonadal dysgenesis with XO/XY mosaicism. Production and analysis of isogenic resistant lines. 7. nat. Cancer 7. Amer. med. Ass., I88, 221. Inst., 2I1 843. Gunson, H. H. (1957). Neonatal anaemia due to foetal haemorrhage Stern, C. (1936). Somatic crossing over and segregation in Drosophila into the maternal circulation. Pediatrics, 20, 3. melanogaster. Genetics, 2I, 625. Hannah-Alava, A. (I960). Genetic mosaics. Scient. Amer., 202, 1 I9. (957). Perspectives in chromosome physiology. Proc. genet. Huskins, C. L. (I948). Segregation and reduction in somatic tissues: Soc. Can., 2, 3. I. Initial observations in Allium cepa. J. Hered., 39, 3II. (ig60). Principles of Human Genetics, 2nd ed. W. H. Freeman Jacobson, L. O., Simmons, E. L., Marks, E. K., Gaston, E. O., Co., San Francisco. Robson, M. J., and Eldredge, J. H. (I95I). Further studies on Stone, W. H., Friedman, J., and Fregin, A. (i964). Possible somatic on September 26, 2021 by guest. Protected recovery from radiation injury. J. Lab. clin. Med., 37, 683. cell mating in twin cattle with erythrocyte mosaicism. Proc. nat. Kaufmann, B. P. (I942). Reversion from roughest to wild type in Acad. Sci. (Wash.), 51, I036. Drosophila melanogaster. Genetics, 27, 537. Taylor, A. I., and Polani, P. E. (i965). XX/XY mosaicism in man. Kirkman, H. N., and Riley, H. D., Jr. (I959). Posthaemorrhagic Lancet, I, I226. anaemia and shock in the newborn: a review. Pediatrics, 24, 97. Waardenburg, P. J., Franceschetti, A., and Klein, D. (i96i). Lancet (I965). Chimaeras and mosaics. I, 1052. Genetics and Ophthalmology. Koninklijke van Gorcum, The Lillie, F. R. (I 9 I 6). The theory of the freemartin. Science, 43, 6I I. Netherlands. Lindsley, D. L., Odell, T. T., Jr., and Tausche, F. G. (I955). Wiener, A. S. (1948). Diagnosis and treatment of anaemia of the Implantation of functional erythropoietic elements following newborn caused by occult placental hemorrhage. Amer. 7. Obstet. total-body irradiation. Proc. Soc. exp. Biol. (N. Y.), 90, 5I2. Gynec., 56, 71I7. Lorenz, E., Uphoff, D., Reid, T. R., and Shelton, E. (i95I). Wilson, E. B. (1928). The Cell in Development and Heredity, 3rd ed. Modification of irradiation injury in mice and guinea pigs by Macmillan, New York. bone-marrow injections. J. nat. Cancer Inst., 12, 197. Winkler, H. (I907). Uber Pfropfbastarde and pflanzliche Lyon, M. F. (I96I). Gene action in the X-chromosome of the mouse Chimaeren. Ber. dtsch. bot. Ges., 25, 568.