(1) Chromosome numbers of 18 forms (35 varieties) of Phlox have been determined. The basic chromosome number is 7. Thirty-two varieties scattered among 15 forms contain 14 chromosomes in the body cells (or 7 in the germ cells). One of these forms (a derivative of a species cross) indicated a partial aneuploid condition-15 chromosomes being seen in some somatic cells. Of the other 3 forms examined 2 were tetraploid, with 28 somatic chromosomes, and the third one was apparently a triploid since it had 20+ somatic chromosomes. Gilia capitata, from a related genus, has 18 somatic chromosomes. (2) Some morphological characters of the chromosomes are described. (3) The known hybridizations among Phlox species are summarized. (4) Three species hybrids were among the 18 forms worked with. Cytological irregularities in 2 of these (and especially in P. procumbens) are followed through the two meiotic divisions and are shown to result in abnormal microspore and pollen production. (5) Mention is made of the types of sterility observed in several forms. (6) A discussion is given which touches on some taxonomic and pbylocytological problems.
1. In Datura stramonium a recessive gene (bd) which is located in the 21·22 chromosome, is responsible for a complete lack of pairing of homologous chromosomes during the first meiotic division in both microsporocytes (pollen-mother-cells) and megasporocytes. The 24 univalents may be scattered along the axis of the bipolar spindle. This results in irregular distribution of chromosomes or even failure to form more than one group (non-reduction). 2. In shape, size and behavior, these univalent chromosomes are similar to the univalents of haploids. 3. The second meiotic division is normal and equational. 4. The tetrad stage is characterized by a varying number of microspores and microcytes. 5. Dyads are the result of non-reduction which varies from zero to eight percent. 6. PMC's showed a limited number of kinds of irregular chromosome distribution; a few occurred more frequently than others. 7. During anaphase I there was a tendency for the chromosomes to separate into two groups, along with a lesser tendency toward non-reduction. There was considerable lagging of chromosomes. 8. Chromosome groups which contain less than a genom produce microcytes only. A formula is given for the probability of a genom in groups of 12 or more chromosomes. 9. Pollen abortion varies from 92 to 98 per cent. Good pollen consists of varying amounts of large grains which have the size characteristic of the pollen of tetraploids and of smaller grains which presumably contain a genom with none to several extra chromosomes. 10. The amount of good pollen found is in fair agreement with the amount expected from application of the formula for probability of a genom to chromosome counts at metaphase II. 11. Trisomic (2n+21·22) plants which are triplex for this gene also show lack of metaphase pairing, 2n+21·22 plants which are duplex for it show normal pairing. 12. Two tetraploids were obtained when bd2 pollen was used on a 4n female.
1. Evidence is presented which indicates that a haploid of coral N. tabacum shows a higher degree of association at pachytene and I-M than is found in normal haploids. 2. At the pachytene stage these associations are mostly nonhomologous in nature except for the duplicate coral segment. T-shaped configurations and foldbacks are also found especially in the coral haploid and the associated sections are much longer than in the normal haploid. 3. As a result of these pachytene associations a variable number of rod-bivalents are found at I-M, the number being higher in the coral haploid. These bivalents have only a single chiasma which may be either terminal, subterminal or interstitial. 4. Fragments are found in both haploids but their frequency is much higher in the coral type, presumably because of the greater amount of non-homologous association. These fragments indicate crossing over has occurred in the non-homologously paired regions or in the foldbacks. 5. It is concluded that observation of bivalents at diakinesis and I-M in haploids and hybrids with variable pairing is not of itself proof of dupliccte segments.
1. The comparison of chromosomes was made between 7- and 8-chromosome rye regarding their, size, shape and behaviour in PMC-s and pollen grains. 2. In common rye, the chromosome number was counted to be 7 at metaphase of primary nuclear division in pollen grains. These chromosomes were classified into three groups according to their size and shape. 3. During the meiosis in 8-chromosome rye, 8 bivalents showed rather regular behaviour, and 8 chromosomes were counted in the majority of their pollen grains. 4. Chromosome set of 8-chromosome rye consists of 7 chromosomes quite similar to those of 7-chromosome rye and an extra small chromosome. 5. The extra chromosome is not the same in its size and shape as the k-chromosome in GOTOH's 8-chromosome rye. 6. The position and shape of nucleus in pollen grains is usually characterized by the stages of their development, and we can distinguish by these the vegetative and the generative nuclei. 7. The extra chromosome shows rather irregular behaviour in primary nuclear division of pollen grain. In most cases, two halves of it were included in the generative nucleus or remained lagging, so that the vegetative nucleus receives 7 ordinary chromosomes. 8. From the irregular distribution of the extra chromosome, the plants having 14, 15 and 16 chromosomes in diploid may be expected in the offspring of 8-chromosome rye. The writer wishes to express his gratitude to Prof. Dr. KAGAWA for his valuable advice, and is also indebted to Prof. Dr. TABATA for the permission to use the material under his management.