Search

The kinetic activity of chromosomes is either distributed evenly along the length of the chromosome or localized in a single centromere. Within the group of organisms with evenly distributed (holokinetic) activity great variation occurs, which has its consequences for chromatid segregation. Three major types, with some overlap, can be distinguished. They form a gradient of increasing specialization. Arguments are presented for the hypothesis that the monokinetic chromosomes are nothing but the end point in the same gradient and do not form a fundamentally different category. These arguments include: concentration of kinetic activity at meiosis in some types of holokinetic chromosomes, and ‘atavistic’ traits in monokinetic chromosomes pointing to a holokinetic origin. Possible reasons for the wide distribution of monokinetic chromosomes in the more evolved taxa are given.

In pearl millet, chain trivalents composed of two telocentric and one metacentric chromosomes, showed an excess of linear orientation over the 1/3 expected with random centromere activation and inactivity of a central centromere stretched between the two outer centromeres. Chain trivalents composed of two metacentrics and one telo or of three metacentrics behaved as predicted. The difference was explained by assuming precocious activation of completely terminal centromeres as opposed to median centromeres. This early activity was reflected in precocious separation at late metaphase. In rye, all trivalents composed of two telos and one metacentric showed alternate orientation and anaphase separation did not precede that of metacentric chromosomes. It is concluded that in rye terminal centromeres are not precocious and that the spindle at meiosis is not long enough to permit stretching of the central centromere, which consequently always has the opportunity to orient and to induce the other centromeres to choose the opposite pole either directly or after reorientation, accumulating the most stable (alternate) orientation type.

Meiotic configurations and segregation were studied in tetraploid hybrids (4n=28) between Secale cereale (2n=14) carrying a reciprocal translocation between chromosomes II and VI and a derivative of Secale montanum (2n=14), naturally differing from S.cereale in a translocation involving chromosomes I, III and V. Multivalent frequencies of the translocation complexes were reduced compared to expected but of still appreciable magnitude, which was ascribed to incomplete preferential pairing. Marker segregation (one of the S. cereale translocation chromosomes) also suggested partial preferential pairing.

A number of recessive characters in rye were studied, seven of which in detail: anthocyanin-less (a); waxless (w); brittle (b); canary (c); yellow-green (yg) and two dwarfs (d1 end2). All are inherited monofactorially against the normal dominant character.

Summary No plants with an abnormally high chromosome number were found among 308 plants studied from within old colonies ofColchicum autumnale L.

Standard mapping functions permit the calculation of map length from the frequency by which a chromosome segment has at least one chiasma. In interchange heterozygotes, approximately exact estimates of such frequencies can be obtained from the relative frequencies of specific groups of meiotic metaphase I configurations. Examples are given for two interchanges inSecale cereale.

The estimation of the “crossing-over potentials” of the two arms of a specific chromosome that can not be recognized in the diploid, was earlier found to be inefficient with the use of the primary trisomic. With the telocentric trisomics two groups of two different configurations each can be recognized that permit a reasonably exact estimation of the two parameters. Each telocentric trisomic yields estimates for both arms. The trisomic arm is underestimated as a result of partner-exchange and/or interference by the nucleolus in a nucleolus bearing arm. The other arm is estimated more correctly. Thus the two telocentrics together give a complete picture of the chromosome. After a correction for differences in overall chiasma frequencies the ratio of the “crossing-over potentials” of the two arms of the satellite chromosome ofSecale cereale was found to be approximately 2. This is large for a submedian chromosome in comparison with the ratio for the genome as a whole and it is attributed tentatively to the nucleolus interfering with chiasma formation in the short arm. It is suggested that the three homologous arms, especially the long arms, differ in respect to the tendency to pair and in chiasma frequency.

Information obtained previously and presently on chromosome pairing and chiasma formation in trisomics and in interchange heterozygotes has been applied in newly constructed models for calculating expected MI configuration frequencies in interchange trisomics. Good fit betwen calculated and observed frequencies in some and poor fit in other cases confirmed the expectation of genetic variation in the crossing-over potentials of some or all chromosome regions. If conclusions in respect of chromosome pairing pattern are to be based on relative frequencies of MI configurations, valid values for crossing-over potentials are required. These can only be obtained from genetically comparable material. A few more disturbing factors are recognised. Environmental effects are one of these factors but may have a relatively simple character. Good agreement between expected and observed frequencies of configurations was taken to indicate the validity of the assumption that homologous chromosome end segments have equal probability of being involved in pairing, irrespective of the length of the segment. This conclusion was confirmed by the segregation of chromosomal types in the progenies of interchange trisomics: the excess chromosome was combined as frequently with the interchange set and with the normal set respectively, as expected on basis of the same models, assuming 60–80% viability of trisomes compared to diploids.

On the basis of indirect evidence from meiotic chromosome associations and genetic experiments, a working hypothesis on the initial stages of chromosome pairing is proposed and examples of application are given. It involves the assumption of functional units of chromosome pairing initiation that are provisionally named “zygomeres”. Possible analogies with centromere and nucleolar organized are discussed.

The quadrivalent of an interchange heterozygote can be divided into six regions, each with its own probability of being bound by at least one chiasma. This gives rise to 26 or 64 different possible combinations of bound and not bound regions and as many M I configurations, several of which are isomorph. Sixteen different types of configuration occur, 10 of which involve more than two chromosomes and can thus be distinguished from the non-interchange bivalents. As each combination of bound and not bound regions has a certain probability of occurring depending on the probability of each region to be bound or not, a frequency expectation value can be estimated for each configuration. Also, when the relative frequencies of the different configurations are known, the probabilities of being bound can be estimated for each of the six regions of a quadrivalent, and from these the crossing-over potentials. One translocation heterozygote was studied, involving the long arm of the satellite chromosome of rye (Secale cereale). Crossing-over was reduced, perhaps mainly by partner exchange. The terminal segments retained a relatively high chiasma frequency. The interstitial regions had low chiasma frequencies even when relatively long.

In interchange heterozygotes and interchange trisomics of rye (Secale cereale) correlations between multivalents and bivalents in respect of number of chromosome arm pairs bound at MI, were studied. No significant positive correlations were observed. Significant negative correlations, indicating interchromosome effects, were found only in cell populations that approached anaphase. The effects seem to be generated by non-random loss of chiasmata from chromosome ends. The absence of demonstrable real interchromosome effects on chiasma formation in the material studied may be explained by positive correlations (due to between-cells variation) counteracting possible negative correlations. Another explanation may be found in the fact that intra-chromosome interference, that might be one of the ways through which interchromosome effects act, does not come to expression with the methods of observations used.

The relative rate of development of aneuploid and eudiploid microspores was studied in anthers of autotetraploidTradescantia virginiana (4x=24), Lycopersicum esculentum (4x=48) andSecale cereale (4x=28). Although the species showed slight differences, in all three the first pollen grain mitosis took place considerably later in aneuploid than in euploid cells. Although hypoploid cells were somewhat more retarded than hyperploid cells, the difference was not great, indicating a decidedly lower degree of developmental autonomy of microspores than of embryos, which only infrequently tolerate hypoploidy at the diploid level.