Cytogenetic analysis of chromosomal intermediates from a hybrid zone between two chromosome races of the Sceloporus grammicus complex (Sauria, Phrynosomatidae).
Reduced fecundity of heterozygotes, through the production of chromosomally unbalanced gametes in small isolated demes, is an assumption common to most models of chromosomal speciation including the cascade model (Sites and Moritz 1987; Hall 1983). As stated by Hall (1983, p. 671), "The most critical assumption in the chromosomal speciation model is that the chromosomal rearrangements fixed between species were negatively heterotic when they originally occurred as new mutations. Presumably, where they are involved in narrow hybrid zones which appear to function as sinks, such as found in the S. grammicus complex, they should still be negatively heterotic." Meiotic disruption has not been demonstrated in hybrid individuals heterozygous for the centric fission mutations that delineate most S. grammicus hybrid zones.
The hybrid zone between the F5 (2n = 33[male]; 34[female]) and FM2 (2n = 43-45[male]; 44-46[female]) cytotypes, located west of Tulancingo in the Mexican state of Hidalgo, offered a unique opportunity to assess the effects of heterozygosity at multiple chromosomes on meiotic processes and test the critical assumption of Hall. In addition to the multiple-fission rearrangement of chromosome 2 (Reed et al. 1992c), the F5 and FM2 cytotypes differ by centric fissions at chromosomes 1, 3, and 6 ([ILLUSTRATION FOR FIGURE 2 OMITTED] in Sites et al. 1995), and each cytotype is heteromorphic at chromosome 4 (pericentric inversion, F5; fission, FM2). Therefore, [F.sub.1] hybrids between the cytotypes should be heterozygous for rearrangements of up to five chromosomes (Sites et al. 1993). A statistical analysis of clinal changes in chromosome markers by Sites et al. (1995) shows that although hybridization occurs along a broad front, chromosomal clines show very narrow transitions between individuals homozygous at three unambiguous chromosomal markers (1, 2, and 6). Additionally, this zone is structured into a series of adjoining local patches characterized by a deficit of heterozygotes and strong linkage disequilibria. Heterozygote deficits may result from positive assortative mating, habitat selection (or association), or selection against heterozygotes.
According to Hall's phylogenetic hypothesis, the F5 and FM2 cytotypes represent nearly opposite ends of the cascade sequence and are separated by several transitional chromosome races. The F5 cytotype differs from the hypothesized ancestral S (standard, 2n = 31[male]; 32[female]) cytotype by a single fission rearrangement. The FM2 cytotype is considered the most chromosomally derived cytotype of the S. grammicus complex. Recent molecular data have shown the pattern of chromosomal evolution is more complex than the linear series proposed by Hall (1973, 1980) but confirmed that the F5 and FM2 cytotypes are not sister taxa (Arevalo et al. 1994). Because of the number of chromosomal differences, and the nonsister relationship of the F5 and FM2 cytotypes, this zone is not equivalent to the primary zones envisioned by Hall (1983). However, if the cumulative effects of multiple rearrangements (see also Walsh 1982) fail to function as substantial barriers to gene flow between hybridizing forms such as the F5 and FM2 cytotypes, then speciation mechanisms such as proposed by the cascade model are seriously compromised. Through analysis of chromosomal pairing and segregation, this study was designed to assess the level of chromosomally based underdominance associated with chromosome rearrangements in the F5 x FM2 hybrid zone (see also Reed et al. 1995).
MATERIALS AND METHODS
The lizards examined in this study were collected during June and July of 1989 and represent a subset of those mapped onto quadrants for the clinal study by Sites et al. (1995). To eliminate the analytical complexity involved with chromosome 2 (see Reed et al. 1995), only individuals homozygous for parental morphologies of this chromosome were included. Eighteen individuals collected from throughout the hybrid zone and heterozygous for up to three fission rearrangements (table 1), were chosen for meiotic analyses. These chromosomally intermediate individuals were classified as either "F5-like" or "FM2-like" based on the morphology of chromosome 2 and their remaining chromosomal complement [ILLUSTRATION FOR FIGURE 1 OMITTED]. Voucher specimens for all karyotyped individuals are cataloged into research collections either at Brigham Young University (BYU) or the Instituto de Biologia, Laboratory de Herpetologia, Universidad Nacional Autonoma de Mexico (IBH).
Surface-spread nuclei for synaptonemal complex (SC) analysis were prepared and substaged as described by Reed et al. (1992a). Diakinetic nuclei were examined for the frequency of chiasmata (bivalent/trivalent formation) and for frequency and incidence of crossing over within the inverted region of pericentric inversion heterozygotes. To estimate the relative frequency of nondisjunction and malsegregation, secondary spermatocytes (metaphase II cells, MII) were scored for the number of metacentric and acrocentric macrochro-mosomes and presence of either the Y or two X chromosomes. Cells in which the number of observed elements differed from the expecteds by more elements than could be accounted for by nondisjunction of any one chromosome were excluded from the analysis. This scoring method provides a conservative estimate of aneuploidy, which is minimally biased by preparation artifact.
Synaptonemal Complex Analyses
A total of 1066 surface-spread primary spermatocytes from 17 males (table 1) was analyzed from electron photomicro-graphs. Axial element and synaptomenal complex (SC) formation was consistent with that observed from individuals from parental populations adjacent to the hybrid zone (Reed et al. 1992a,b,c). Synapsis of the lateral elements of the trivalents proceeded unidirectionally from both sets of telo-meres of the metacentric element towards the centromere [ILLUSTRATION FOR FIGURE 2A OMITTED]. In all individuals examined, the number of bivalents and trivalents at pachynema corresponded to the number expected based on the somatic chromosomal data (table 1).
The degree of synapsis of the two acrocentric elements with the metacentric element of the trivalents was variable within individual pachytene nuclei [ILLUSTRATION FOR FIGURE 2A OMITTED]. This difference could be attributed to the particular behavior of one of the trivalent elements [ILLUSTRATION FOR FIGURE 2B OMITTED] or to hindrance caused by non-homologous telomeric associations [ILLUSTRATION FOR FIGURE 2C, D OMITTED]. Associations between autosomal trivalents and other elements were observed in 36.7% of the pachytene nuclei. The frequency of nuclei with nonhomologous associations involving trivalents increased with the number of heterozygous chromosomes, occurring in 30.3%, 44.5%, and 54.0% of the pachytene nuclei from individuals heterozygous for 1, 2, and 3 chromosomes, respectively. In individuals heterozygous for single fission rearrangements, the frequency of telomeric associations involving the elements of the trivalents (30.3%) was comparable to that observed in chromosome-4 heterozygotes from a nonhybrid FM2 population (31.6%; Reed et al. 1992c). The frequency of nonhomologous associations involving the autosomal trivalents did not appreciably vary with the predominant chromosomal background (F5 or FM2). In individuals heterozygous for a single rearrangement, the percent of associations averaged higher for the F5 intermediates (34.01%) than for the FM2 intermediates (28.00%). However, in individuals heterozygous for two rearrangements, the percent of associations averaged higher for the FM2 intermediates (46.96%) than for the F5 intermediates (42.50%).
Telomeric associations involving the autosomal bivalents were observed in all individuals examined. In the heterozygous condition, the proximal telomeres (i.e., those closest to the centromere) of the acrocentric fission products lack homologous telomeres with which to associate. This fact coupled with the tendency for the telomeres to become polarized on the nuclear membrane, increases the likelihood for associations. Alternatively, the associations may be "homologous" if the chromosomes of Sceloporus grammicus share a common telomeric sequence (cf., Blackburn and Szostack 1984). No data exist to indicate that telomeric associations disrupt the meiotic process in S. grammicus.
Synapsis of the autosomal trivalents was generally delayed relative to the bivalents. In the absence of evident telomeric associations, the timing of complete trivalent formation was generally dictated by the size of the element. In those individuals heterozygous for more than one rearrangement, the smaller of the trivalents was usually first to complete synapsis. In those individuals heterozygous only for fission of chromosome 6 (the smallest Robertsonian chromosome), the elements of the trivalent often completed synapsis before the larger autosomal bivalents (i.e., during zygonema). For example, in one such individual (BYU 39983), 15 of the 30 nuclei assigned to zygonema contained a completely synapsed chromosome-6 trivalent.
Meiotic behavior of the inversion bivalents in the fission/inv(4) heterozygotes was identical to that observed in other inv(4) heterozygotes (Reed et al. 1992b). Analyses of the fission/inv(4) heterozygote (IBH 7190) revealed heterosynapsis of the inverted region without prior reverse-loop formation. Several studies in vertebrates have documented similar phenomenon in inversion heterozygotes (Thorneycroft [TABULAR DATA FOR TABLE 1 OMITTED] 1975; Shields 1976; Ashley et al. 1981; Sites 1983; Kaelbling and Fechheimer 1985; Hale 1986; Chandley et al. 1987; and Gabriel-Robez et al. 1988). In (IBH 7190), heterosynapsis of the inverted region of chromosome 4 proceeded prior to the complete synapsis of the chromosome-3 trivalent. Pachytene nuclei with complete chromosome-3 trivalents and partially synapsed inv(4) bivalents were not observed.
Meiotic Chromosomal Analyses
Chromosomal configurations observed at diakinesis and metaphase II [ILLUSTRATION FOR FIGURES 3, 4 OMITTED] were consistent with the number of elements predicted from the SC analyses. Examination of 1810 diakinetic nuclei from the 18 heterozygotes revealed uniformly low levels of abnormalities (mean = 0.4%, range = 0.0% - 4.0%, table 1). In most cases, the abnormalities were the result of the apparent lack of chiasma formation between a metacentric and one acrocentric element of one or more trivalent. Three of the aberrant nuclei each contained two univalents. Grouping of individuals by the number of heterozygous chromosomes indicated that the number of abnormal diakinetic nuclei increased only slightly with the increase in the number of heterozygous chromosomes (table 2A). Abnormalities were not observed in any nuclei examined from the F5 intermediates. The percentage of abnormalities in FM2 intermediates increased by 2.2% between the single and double heterozygotes (table 2B).
Analysis of secondary spermatocytes (MII configurations) was complicated by the possible number and morphologies of the elements. For example, in the individual heterozygous at three chromosomes (BYU 39909), balanced disjunction of the three trivalents would result in eight different but normal (i.e., genetically balanced) haploid MII combinations. The number of metacentric and/or acrocentric elements for the three heteromorphic chromosomes can range from three metacentric and zero acrocentric elements to zero metacentric and six acrocentric elements per nucleus [ILLUSTRATION FOR FIGURE 3B-E OMITTED]. Because of the difficulty in determining which elements were present as metacentric and acrocentric elements in individuals heterozygous for fissions of two or three chromosomes, MII cells were grouped by the total number of elements for analysis.
Examination of 1224 secondary spermatocytes suggested completely balanced disjunction in eight of the individuals (table 1). For the remaining individuals, the percentage of aneuploid nuclei at MII ranged from 1.0% to 6.7%. Whereas most of the aneuploidy could be attributed to trivalent non-disjunction, three nuclei contained complements that could not be attributed to this phenomenon [ILLUSTRATION FOR FIGURE 3F OMITTED]. The frequency of aneuploidy did not appreciably increase with the number of heterozygous chromosomes (table 2A). When individuals were grouped on the basis of chromosomal background, the frequency of aneuploidy was higher in the FM2 intermediates than in F5 intermediates heterozygous at the same number of chromosomes (table 2B), although the frequency remained low for all groups.
[[Chi].sup.2] analyses indicated that the frequency of spermatocytes containing either the acrocentric (fission products) or the metacentric chromosomes was significantly different from random expectations in 3 of the 16 chromosomally intermediate individuals (table 3; inversion heterozygotes excluded). These three individuals had primarily F5 chromosomal backgrounds and were heterozygous for the fission rearrangement of chromosome 1. It is unclear as to the cause of the unbalanced segregation of chromosome 1. In these individuals, the number of secondary spermatocytes with the fission products (acrocentric elements) was significantly greater than those with the metacentric element. The segregational irregularities may be indicative of selection for the fission morphology resulting perhaps from cytomechanical constraints (White 1969, 1975). Interestingly, the two larger elements of the fissioned chromosome 1 found in the FM2 cytotype are not truly telocentric but possess minute short arms (subacrocentric). This morphology is similar to a chromosome-2 inversion variant seen in another FM cytotype (FM1, Reed et al. 1992c). It is possible that chromosome 1 harbors an undetected pericentric inversion. If so, this might account for the additional segregation irregularities.
Because of the potential bias in the scoring of the sex chromosomes (see discussion in Reed et al. 1992a), direct comparisons of sex-chromosomal segregation were not made. However, contingency [[Chi].sup.2] tests for independence, which are unprejudiced by the potential scoring bias, were performed on euploid MII counts from nine individuals (table 4). In six individuals, the occurrence of the Y chromosome with the metacentric element of the heterozygous macrochromosome and the two X chromosomes with the two acrocentric elements was greater than the reciprocal combinations (table 4A-C). This difference was significant at the [Alpha] = 0.01 level in one of the lizards examined.
The greater association of the metacentric elements with the Y chromosome and the acrocentric elements with the two X chromosomes is similar to that observed for chromosome-4 heterozygotes (Reed et al. 1992c). This may result from associations between the trivalent elements and sex-chromosomal elements during prophase I. Association of the acrocentric elements may cause the trivalents to be oriented such that the metacentric and acrocentric elements of both trivalents are preferentially drawn to opposite poles at anaphase I.
A total of 300 diakinetic nuclei was scored from the two fission/inv(4) heterozygotes. Of the nuclei examined, 96.7% contained a single bivalent (corresponding in size to chromosome 4) with chiasmata in only one arm [ILLUSTRATION FOR FIGURE 4A-B OMITTED]. In the remaining nuclei, this bivalent displayed an association between the telomeres of the inverted region similar to that reported for other inv(4) heterozygotes (Reed et al. 1992b). A total of 197 secondary spermatocytes from the fission/inv(4) heterozygotes was scored for the presence of either the metacentric (noninverted) or acrocentric (inverted) chromosome 4 and the metacentric or two acrocentric elements of the fission trivalent [ILLUSTRATION FOR FIGURE 4C-F OMITTED]. Individual counts and the results of the [[Chi].sup.2] analyses are summarized in table 5. Significant deviation from the expected equal ratio of inverted to noninverted and fissioned to nonfissioned chromosomes was detected in one of the fission/inv(4) heterozygotes (BYU 39833). In this individual, MII cells were biased towards the presence of the inverted chromosome 4 with the fission products of chromosome 1. The segregational differences in BYU 39833, as evident in the [[Chi].sup.2] tests, could be caused by the presence of the chromosome-1 fission heteromorphism. Because of the delayed synapsis of the chromosome-1 trivalent, as observed in SC analyses of other chromosome-1 heterozygotes, the segregational difference could be due to associations between the chromosome-1 trivalent and the inv(4) bivalent during meiotic prophase I. Synaptonemal complex data were not collected from this animal, and thus, no definitive statement can be made regarding the early meiotic behavior of this chromosomal combination. Tests of independence between the inverted, fissioned, and sex chromosomes (table 5B-D) revealed no significant deviation from random expectations.
TABLE 2. Results of the meiotic chromosomal analysis. A. Individuals pooled by number of heterozygous chromosomes. B. Individuals pooled by number of heterozygous pairs and grouped by chromosomal background. Abbreviations are as follows: number of heterozygous macrochromosomes (Het); total number of nuclei scored (N), number and percent abnormal nuclei, and number and percent aneuploid nuclei.
Diakinesis Metaphase II
Abnor- Aneu- Abnor- mal Aneu- ploid Het Ind N mal (%) N ploid (%)
1 11 1010 2 0.20 689 10 1.45 2 6 700 5 0.71 364 9 2.47 3 1 100 1 1.00 171 3 1.75
Abnor- Aneu- Abnor- mal Aneu- ploid Het Cytotype N mal (%) N ploid (%)
1 F5 410 0 0.00 313 2 0.60 FM2 600 2 0.33 376 8 2.13
2 F5 500 0 0.00 255 4 1.57 FM2 200 5 2.50 109 5 4.58
Chromosomal Pairing Behavior and Segregation
Meiotic irregularities observed in this study were on the same order of magnitude as those found in other S. grammicus individuals heterozygous for within-population heteromorphisms (Porter and Sites 1985, 1987; Reed et al. 1992b,c). Trivalent formation was similar to that described for other simple Robertsonian rearrangements (Grao et al. 1989; Wallace and Searle 1990; Reed et al. 1992c; Wallace et al. 1992). The retention of the metacentric pairing pattern (biterminal synaptic initiation) is presumably instrumental in the proper formation of fission trivalents. The presence of at least one chiasma per chromosomal arm appears to facilitate proper orientation and disjunction in simple Robertsonian heterozygotes. Trivalent configurations at diakinesis, and their nearly balanced disjunction, as evident from the MII counts, support this hypothesis.
Results of the analysis of diakinetic nuclei from the fission/inv(4) heterozygotes were consistent with the observed suppression of chiasmata reported for the inv(4) rearrangement in S. grammicus (Reed et al. 1992b). Estimates of the frequency of nondisjunction in the fission/inv(4) heterozygotes (BYU 39833, 2.0% and IBH 7190, 4.0%) were within the range of that estimated for the simple fission heterozygotes. This low level of abnormal secondary spermatocytes further supports the interpretation of chiasma suppression within the inverted region. Estimates of nondisjunction or aneuploid frequency for fission/fusion heterozygotes in other vertebrates range from 0% to 34% (Ford and Evans 1972; Cattanach and Moseley 1973; Gropp and Winking 1981; Searle 1986; Porter and Sites 1987; Stewart-Scott and Bruere 1987; Wallace et al. 1992). If the frequency of nondisjunction is assumed to be approximately equal to the frequency of aneuploidy at MII (i.e., each nondisjunction event producing one hypo- and one hyperhaploid nucleus, although see Cattanach and Moseley 1973; Ford and Evans 1972; Logue and Harvey 1978; Guichaoua et al. 1986), nondisjunction in the fission heterozygotes examined in this study ranged from 0% to 6.7%. This low level of nondisjunction is consistent with previous examinations of within-population heteromorphisms in the S. grammicus complex (Porter and Sites 1987; Reed et al. 1992b,c). If similar levels of nondisjunction occur in females, the effect of the rearrangements examined in this study on individual fitness would be inconsequential (although see Gropp and Winking 1981).
Genetic Background Influences on Meiosis
Several studies have demonstrated increased meiotic disruption for similar chromosomal rearrangements when present on divergent genetic backgrounds (Coates and Shaw 1984; Gropp and Winking 1981; John et al. 1983). Chromosomal background did not appear to be an important factor in the dynamics of male meiosis in the F5 x FM2 hybrid zone. If, as hypothesized by Hall (1973, 1983), the derived FM2 cytotype has a genetic background that mitigates the potential meiotic problems associated with structural chromosomal heterozygosity, lizards with a primarily FM2 chromosomal background should show lower levels of meiotic irregularities than those with a primarily F5 chromosomal complement. Excepting the segregational differences in chromosome 1, FM2 intermediates generally showed increased (but low) levels of both diakinetic abnormalities and MII aneuploidy. However, because the FM2 intermediates contain an appreciable amount of the F5 genome, these findings may not be an adequate test of Hall's hypothesis. One phenomenon, however, may be indicative of differences in either chromosomal or genic background. Individuals heterozygous for the inversion of chromosome 4 were found only on the F5 (east) side of the hybrid zone, and no individuals heterozygous for both the fission and inversion phenotypes of this chromosome were obtained. These findings suggest that the inv(4) rearrangement is not compatible with the FM2 genome or that the fission/inversion combination of chromosome 4 is not viable. This lack of inversions of chromosome 4 has been noted in other populations of S. grammicus heteromorphic for fission of chromosome 4 (see Reed et al. 1992b).
Implications for the Cascade Model
According to Hall's cascade model, the speciation "cascade" is initiated by the fixation of chromosomal rearrangements [TABULAR DATA FOR TABLE 3 OMITTED] in small isolated populations. The origin and fixation of rearrangements, assumed to be negatively heterotic, are enhanced by genetic backgrounds that mitigate meiotic problems associated with structural heterozygosity. If the loci that make up a favorable background are themselves polymorphic, some demes will by chance have genetic backgrounds more favorable to chromosomal differentiation. Those populations (species) in which the new rearrangement survives will tend [TABULAR DATA FOR TABLE 4 OMITTED] to perpetuate a favorable genetic background (the positive feedback mechanism) and be more likely to undergo further chromosomal rearrangements than the ancestral species. This process of rearrangement fixation coupled with positive genetic feedback would result in an increasingly rapid and primarily linear chain of chromosomal rearrangements (Hall 1973, 1983).
In the cascade model, group selection acts in concert with [TABULAR DATA FOR TABLE 5 OMITTED] positive genetic feedback. However, the effects of selection are seen as an ecological consequence of chromosomal speciation and not a causal force. Selection acts according to the competitive exclusion principle with the major selective force being the tension zone formed between the newly differentiated founder populations and the ancestral stock. Founder populations, once established, expand their range by displacing the ancestral stock or by invading a new niche. Termination of the speciation cascade occurs when: (1) the chromosomal substrate for a particular rearrangement is exhausted (for example in S. grammicus when all chromosomes are fixed for fission); (2) selection produces a genetic system in which chromosomal heterozygotes no longer display the negative heterosis required for successful speciation; or (3) niche saturation. The first two types of termination would produce terminal taxa with primarily fissioned chromosomal complements or which are heteromorphic for the final rearrangements of the series, and in which the heteromorphisms are meiotically neutral or nearly so. Within the framework of Hall's chromosomally based phylogeny of the S. grammicus complex, this is predicted to be exclusively or primarily the FM2 cytotype.
Recent studies have examined the meiotic effects of chromosomal heterozygosity in the S. grammicus complex. Heterozygosity for fission (Porter and Sites 1985, 1987; Reed et al. 1992c) and pericentric-inversion rearrangements (Reed et al. 1992b) was found to be of little meiotic consequence when present within populations (i.e., of nonhybrid origin). These findings are inconsistent with the cascade model because the meiotically neutral heteromorphisms are not confined to FM2.
The cascade model depends on the same major assumption as the stasipatric model of White (1968); both require the rearrangements to be underdominant, at least in hybrid zones. Underdominance is likely strong enough in chromosome 2 to play a significant role in the dynamics of the F5 x FM2 hybrid zone (see Reed and Sites 1995; Reed et al. 1995), but this study provides no direct evidence for underdominance in the Tulancingo hybrid zone associated with the centric fissions of chromosomes 1, 3, 4, and 6 or the pericentric inversion of chromosome 4. We conclude that these rearrangements have not played the primary role in the origin of the S. grammicus cytotypes that has been attributed to them by some proponents of chromosomal speciation (White 1978a; Hall 1983; King 1993).
We thank E. Arevalo, D. Hutchison, F. Mendoza, M. Mancilla, and J. and H. Sites for assistance with the field collections and processing. Use of EM facilities was made possible by the Texas A & M Electron Microscopy Center. Critical review of various drafts of the manuscript was provided by J. Bickham, S. Davis, D. Hale, and several anonymous reviewers. This research was supported by a grant from the United States National Science Foundation (BSR-8822751).
Arevalo, E., S. K. Davis, G. Casas, G. Lara, and J. W. Sites, Jr. 1993. Parapatric hybridization between chromosome races of the Sceloporus grammicus complex (Phrynosomatidae): structure of the Ajusco transect. Copeia 1993:320-340.
Arevalo, E., S. K. Davis, and J. W. Sites, Jr. 1994. Mitochondrial DNA sequence divergence and phylogenetic relationships among eight chromosome races of the Sceloporus grammicus complex (Phrynosomatidae) in central Mexico. Systematic Biology 43:387-418.
Ashley, T., M. J. Moses, and A. J. Solari. 1981. Fine structure and behaviour of a pericentric inversion in the sand rat, Psammomys obesus. Journal of Cell Science 50:105-119.
Blackburn, E. H., and J. W. Szostack. 1984. The molecular structure of centromeres and telomeres. Annual Review of Biochemistry 53:163-194.
Cattanach, B. M., and H. Moseley. 1973. Nondisjunction and reduced fertility caused by the tobacco mouse metacentric chromosomes. Cytogenetics and Cell Genetics 12:264-287.
Chandley, A. C., S. McBeath, R. M. Speed, L. Yorston, and T. B. Hargreave. 1987. Pericentric inversion in human chromosome 1 and the risk for male sterility. Journal of Medical Genetics 24: 325-334.
Coates, D. J., and D. D. Shaw. 1984. The chromosomal component of reproductive isolation in the grasshopper Caledia captiva. III. Chiasma distribution patterns in a new chromosomal taxon. Heredity 53:85-100.
Ford, C. E., and E. P. Evans. 1972. Robertsonian translocations in mice: segregation irregularities in male heterozygotes and zygotic unbalance. Chromosomes Today 4:387-397.
Gabriel-Robez, O., C. Ratomponirina, M. Croquette, J. Couturier, and Y. Rumpler. 1988. Synaptonemal complexes in a subfertile man with a pericentric inversion in chromosome 21. Heterosynapsis without previous homosynapsis. Cytogenetics and Cell Genetics 48:84-87.
Grao, P., M.D. Coll, M. Ponsa, and J. Egozcue. 1989. Trivalent behavior during prophase I in male mice heterozygous for three Robertsonian translocations: An electron-microscopic study. Cytogenetics and Cell Genetics 52:105-110.
Gropp, A., and H. Winking. 1981. Robertsonian translocations: Cytology, meiosis, segregation patterns, and biological consequences of heterozygosity. Symposia of the Zoological Society of London 47:141-181.
Guichaoua, M. R., S. Ayme, and J. M. Luciani. 1986. Direct estimation of the non-disjunction rate at first meiotic division in the human male. Preliminary results. Human Genetics 72:174-176.
Hale, D. W. 1986. Heterosynapsis and suppression of chiasmata within heterozygous pericentric inversions of the Sitka deer mouse. Chromosoma 64:425-432.
Hall, W. P. 1973. Comparative population cytogenetics, speciation, and evolution in the iguanid lizard genus Sceloporus. Ph.D. diss. Harvard University, Cambridge, Mass.
-----. 1980. Chromosomes, speciation, and evolution of Mexican iguanid lizards. National Geographic Society Research 12:309-329.
-----. 1983. Modes of speciation and evolution in the sceloporine iguanid lizards. I. Epistemology of the comparative approach and introduction to the problem. Pp. 643-679 in A. G. J. Rhodin and K. Miyata, eds. Advances in herpetology and evolutionary biology. Museum of Comparative Zoology, Cambridge, Mass.
John, B., D. C. Lightfoot, and D. B. Weissman. 1983. The meiotic behavior of natural F1 hybrids between Trimerotropis suffusa Scudder and T. cyaneipennis Bruner (Orthoptera: Ocdipodinae). Canadian Journal of Genetics and Cytology 25:467-477.
Kaelbling, M., and N. S. Fechheimer. 1985. Synaptonemal complex analysis of a pericentric inversion in chromosome 2 of domestic fowl, Gallus domesticus. Cytogenetics and Cell Genetics 39:82-86.
King, M. 1993. Species evolution: the role of chromosome change. Cambridge University Press, New York.
Logue, D. N., and M. J. A. Harvey. 1978. Meiosis and spermatogenesis in bulls heterozygous for a presumptive 1/29 Robertsonian translocation. Journal of Reproductive Fertility 54: 159-165.
Porter, C. A., and J. W. Sites, Jr. 1985. Normal disjunction in Robertsonian heterozygotes from a highly polymorphic lizard population. Cytogenetics and Cell Genetics 39:250-257.
-----. 1987. Evolution of Sceloporus grammicus complex (Sauria: Iguanidae) in central Mexico. II. Studies on rates of non-disjunction and the occurrence of spontaneous chromosomal mutations. Genetica 75:131-144.
Reed, K. M., and J. W. Sites, Jr. 1995. Female fecundity in a hybrid zone between two chromosome races of the Sceloporus grammicus complex (Sauria, Phrynosomatidae). Evolution 49:61-69.
Reed, K. M., J. W. Sites, Jr., and I. F. Greenbaum. 1992a. Chromosomal synapsis and the meiotic process in male mesquite lizards, Sceloporus grammicus complex. Genome 35:398-408.
-----. 1992b. Synapsis, recombination, and meiotic segregation in the mesquite lizard, Sceloporus grammicus, complex. I. Pericentric inversion heteromorphism of the F5 cytotype. Cytogenetics and Cell Genetics 61:40-45.
-----. 1992c. Synapsis, recombination, and meiotic segregation in the mesquite lizard, Sceloporus grammicus complex. II. Fission heteromorphism of the FM2 cytotype and evolution of chromosome 2. Cytogenetics and Cell Genetics 61:46-54.
Reed, K. M., I. F. Greenbaum, and J. W. Sites, Jr. 1995. Dynamics of a novel chromosomal polymorphism within a hybrid zone between two chromosome races of the Sceloporus grammicus complex (Sauria, Phrynosomatidae). Evolution 49:48-60.
Searle, J. B. 1986. Meiotic studies of Robertsonian heterozygotes from natural populations of the common shrew, Sorex araneus L. Cytogenetics and Cell Genetics 41:154-162.
Shields, G. F. 1976. Meiotic evidence for pericentric inversion polymorphism in Junco (Aves). Canadian Journal of Genetics and Cytology 18:747-751.
Sites, J. W., Jr. 1983. Chromosome evolution in the iguanid lizard Sceloporus grammicus. I. Chromosome polymorphisms. Evolution 37:38-53.
Sites, J. W., Jr., and C. Moritz. 1987. Chromosome evolution and speciation revisited. Systematic Zoology 36:153-174.
Sites, J. W., Jr., S. K. Davis, D. W. Hutchison, B. A. Maurer, and G. Lara. 1993. Parapatric hybridization between chromosome races of the Sceloporus grammicus complex (Phrynosomatidae): structure of the Tulancingo transect. Copeia 1993:373-398.
Sites, J. W., Jr., N.H. Barton, and K. M. Reed. 1995. The genetic structure of a hybrid zone between two chromosome races of the Sceloporus grammicus complex (Sauria, Phrynosomatidae) in central Mexico. Evolution 49:9-36.
Stewart-Scott, I. A., and A. N. Bruere. 1987. Distribution of heterozygous translocations and aneuploid spermatocyte frequency in domestic sheep. Journal of Heredity 78:37-40. 2 Thorneycroft, H. B. 1975. A cytogenetic study of the white-throated sparrow, Zonotrichia albicollis (Gmelin). Evolution 29:611-621.
Wallace, B. M. N., and J. B. Searle. 1990. Synaptonemal complex studies of the common shrew (Sorex araneus). Comparison of Robertsonian heterozygotes and homozygotes by light microscopy. Heredity 65:359-367.
Wallace, B. M. N., J. B. Searle, and C. A. Everett. 1992. Male meiosis and gametogenesis in wild house mice (Mus musculus domesticus) from a chromosomal hybrid zone; a comparison between "simple" Robertsonian heterozygotes and homozygotes. Cytogenetics and Cell Genetics 61:211-220.
Walsh, J. B. 1982. Rate of accumulation of reproductive isolation by chromosome rearrangements. American Naturalist 120:510-532.
White, M. J. D. 1968. Models of speciation. Science 159:1065-1070.
-----. 1969. Chromosomal rearrangements and speciation in animals. Annual Review of Genetics. 3:75-98.
-----. 1975. Chromosomal repatterning - regularities and restrictions. Genetics 79:63-72.
-----. 1978a. Modes of speciation. W. H. Freeman, San Francisco.
-----. 1978b. Chain processes in chromosomal speciation. Systematic Zoology 27:285-298.