![[Home Page]](/images/icons/gramene_logo180new.jpg "[Home Page]"){kind=icon}
1) National Institute of Genetics, Mishima, 411 Japan
2) Institute of Botany, Academia Sinica, Nankang, Taipei, Taiwan 11529, ROC
In rice, all the 12 chromosomes are identifiable morphologically in the
pachytene and somatic prometaphase stages. But it has been agreed to assign
numbers to rice chromosomes according to their pachytene length (RGN 2, p. 16).
The association of respective chromosomes with linkage groups has been worked
out by Khush et al. (1984) as well as by Iwata and Omura (1984) and Iwata et
al. (1984) using different series of trisomics. However, the numbering systems
of chromosomes adopted by the two schools were partly inconsistent. To
establish a unified numbering system, Drs. N. Kurata and R.JU. Singh were
invited by the IRRI to work together with the Triplo series of IR36 in June
1986. They agreed as to the numbers allotted to six Triplo lines, 1,2,5,7,9 and
10, but did not agree as to the remaining six. The second author of this paper
has then investigated the remaining six lines with coworkers. In a RGC
committee meeting held at 4 and 8 to be 3 and 8, respectively (RGN 4,p.1-2).
It is not an easy task to carry out observations of pachytene chromosomes of a trisomic plant to identify its extra chromosome. As the basis of this work, data for normal diploid plants that can be used as the control will be wanted. For this purpose, we have made statistical comparisons of the data on pachytene and somatic prometaphase chromosomes so far made available to us, which are enumerated as follows:
A-Shastry et al. (1960). Norin 6 (Japonica), pachytene; fixation in 1:3 acetic
alcohol and squashing in aceto-carmin. All 12 bivalents were identified and
measured in 10 cells, but standard deviations of the measurements were not
given.
B-Kurata et al. (1981). Nipponbare (Japonica), pachytene; pretreatment in 0.75 M KC1, maceration with pectinase and cellulase, and squashing in 1:3 acetic alcohol. Then, flame-dried and stained in Giemsa.
C-Kurata's report submitted to IRRI (1986). IR36 (Indica), 6 Triplo lines, pachytene; the same techniques as above (B). Means and standard deviations were computed by the authors from the data presented.
D-Chung and Wu (1987); the same data also in Wu et al. (1985). IR36 (Indica), pachytene; fixation in 1:3 acetic-alcohol with 0.5% ferric chloride, squashing in aceto-carmine with ferric hydroxide, and destaining with 45% acetic acid. Then, flame-warmed and pressed. A digitizer was used for measuring length on photographs.
E-Chen et al. (1982). Chianung 242 (Japonica), pachytene; similar techniques as above (D). Standard deviations were computed by the authors from the standard errors of means for 25 cells.
F-Kurata and Omura (1978). Sekitori (Japonica), somatic prometaphase; pretreatment in 0.002 M 8-oxyquinoline, and enzymic maceration, smashing in acetic-alcohol, and Giemsa staining similarly as for pachytene chromosomes (B). Standard deviations of the measurements were not given.
G-Chung and Wu (1987); the same data also in Wu et al. (1985). IR36 (Indica), somatic prometaphase; similar techniques as above (F).
H-Fukui et al. (1988). Nipponbare (Japonica), somatic prometaphase; pretreatment in 0.1% colchicine in phosphate buffer, and the same techniques as used by Kurata and Omura (F). Measurements were taken by a newly devised method of image analysis with the aid of a computer. Standard deviations were obtained through personal communication (cf. Fukui et al., this issue of RGN).
Adjustment of numbering system: Chromosomes were numbered by respective authors similarly in the descending order of relative length, but the numbering systems are not consistent with one another in two respects: First, chromosome 2 in data A,B,C and F has a greater arm ratio than chromosome 3, but that in data D,E,G and H has a smaller arm ratio than the latter. In data D and H, chromosome 2 with smaller arm ratio is significantly longer than chromosome 3 (t-test), but the differences between chromosomes 2 and 3 in length in other data sets are not significant statistically. Therefore, we had better replace chromosomes 2 and 3 with each other in data A,B,D and F so as to make them comparable to other data. The thus replaced chromosomes are shown by R in Tables 1,2 and 5.
Second, IR36 has two nucleolar chromosomes which are numbered 9 and 10 in data C, but 8 and 10 in data D and G. Chromosome 9 in data A,B,F and H has a greater arm ratio than chromosome 8, but that in data D,E and G has a smaller arm ratio than the latter. The 8-9 differences in relative length are not significant statistically except in data H. Therefore, chromosomes 8 and 9 in data D,E and G are replaced with each other so to be comparable to those in other data, as shown by R in Tables 1,2 and 5.
Data H obtained by computer-aided image analysis were not subjected to such readjustment as above. All statistical treatments of the data reported in this note were conducted after these replacements which made all the data comparable to one another.
Most of Japonica varieties have one nucleolar chromosome which is numbered 10. It may be questioned why Shastry et al. (data A) have described chromosomes 2 (originally 3) and 4 to be nucleolus-attached. Probably, some misjudgment is involved.
Comparison of relative-length data given by different authors: The relative lengths of respective chromosomes given by these authors in percent of total length are summarized in Tables 1 (pachytene) and 2 (prometaphase). There are differences among the data sets. How to evaluate the differences must be considered. The t-test for individual chromosomes is limited to the data with standard deviations. We have made the analysis of variance and multiple comparisons for individual chromosomes. As the original data were not available, the mean squares between and within data sets were estimated from means, standard deviations and number of cells observed. Data without standard deviations, A and F, were taken as single measurements.
The results of analysis of variance for pachytene, prometaphase, and for both stages pooled are given in Table 3. The error variances tended to be large in chromosomes 1,2 and 3 having large mean values. Differences among authors were significant in chromosomes 1,6,8,9,10,11 and 12 at pachytene, and in 1,3,8 and 12 in prometaphase. This suggests that biases due to authors are localized in certain chromosomes. It remains unknown whether or not the differences among authors reflect differences due to the varieties used. In the pooled analysis of variance for pachytene and prometaphase, significant differences between the two stages were found in chromosomes 2,5 and 6. The mean relative lengths obtained from the analysis of variance are given in Table 2.
Based on the error variances, multiple comparisons (Steel and Torrie 1980, p. 172-181) were conducted with respect to chromosomes showing significant differences according to authors (Table 4). In pachytene relative length, data A (Shastry et al. 1960) showed greatest differences from other data sets. The same trend was also observed in the comparison between pachytene and prometaphase data. Therefore, the mean for different authors' data was computed excluding data A (Table 1).
Correlations among the data sets were also computed to evaluate agreement between authors (Table 6, above diagonal). The correlation coefficients varied from 0.997 to 0.915, but these high correlations do not imply good agreement among authors because in each data set the 12 chromosomes are ranged in principle in the descending order of relative length, so that their rank correlations are nearly complete. There may be no established method of testing significance of such correlations. A computer experiment was conducted which showed that when 15 sets of 12 random digits so chosen that in each set the largest was less than three times as large as the smallest were ranged in descending order, the 105 correlations obtained among them varied from 0.81 to 0.992 and those exceeding 0.984 were four. Assuming that the probability of a correlation coefficient higher than 0.984 to arise between such sets of random digits is 0.05 or less, we have tentatively taken r>=0.985 as indicating agreement between two data sets. Then, data G gave best agreement with others (5 of 7 comparisons), followed by data D (4/7).
Comparison of arm-ratio data: The long arm/short arm ratios obtained in respective chromosomes by different authors are given in Table 5. Standard deviations have been given in data B,C,D,E,G and H, but they are omitted from the table for brevity. In view of the trend of arm ratios to be more variable than relative length and that they are not ranged in order, analysis of variance and multiple comparisons were not conducted and the agreement between data sets was evaluated by correlation. In computing correlations, the ratios were converted into the percentage of long arm to chromosome length. The results are given in Table 6, below diagonal.
The correlation matrix reveals that data G best agreed with others, followed by data B and C. Data A was totally uncorrelated with others. Data E was correlated only with G. Data A and E were excluded in computing the mean for different authors' data (Table 5).
It is stated in the report of Interim Committee on Chromosome Numbering (RGN 2, p. 16) that the numbering should be done "according to their pachytene length based on studies by Dr. S.V.S. Shastry et al. (1960) and confirmed by Dr. N. Kurata et al. (1981)". However, the data given by Shastry et al. are least concordant with other data. The reliability of a data set cannot be evaluated by agreement with others. Yet there is no tangible evidence showing that the differences in chromosomal measurements come from the use of particular genotypes or cytological techniques. It may be taken for granted that the accumulation of data by different workers has increased the accuracy of measurements. In examining extra chromosomes of the Triplo plants, as the control may be used either the general means as given in this note or a data set showing good agreemement with others.
The data dealt with in this note are confined to relative length and arm ratio. Beside these, pachytene chromosomes are characterized by the distributions of heterochromatin and chromomeres. But these are not suited for numerical determination, and whether or not the observations by two workers are consistent cannot be evaluated objectively. It is also experienced that the distribution of chromomeres changes with time during the pachytene stage and is subject to intentional recognition by an observer. Accordingly, these were not included in this study.
Statistical analysis of chromosomal measurements has not been conducted much in the past. We are indebted to Professor K. Saio of University of Tokyo for his critical discussion from the viewpoint of biometry.
Table 1. Relative lengths of pachytene chromosomes measured by different
authors (mean and standard deviation, in %)
===============================================================================
A B C D E
Chr. Shastry Kurata Kurata's Chung Chen et Overall
no. et al. et al. report and Wu al. 1982 mean*
1960 1981 IR38 1987 1982 exc. A
Norin 6 Nipponbare IR36 Chianung 242
n: 10 4 4-7 25 25
mean mean std mean std mean std mean std mean std
===============================================================================
1 19.75 12.9 0.5 13.62 0.53 13.58 0.87 13.95 1.35 13.51 0.44
2 11.75 12.0R 1.7 11.60R 0.46 12.15 0.74 12.47 1.10 12.06 0.36
3 11.88R 12.0R 0.8 11.68R 0.15 10.82 0.78 10.95 1.10 11.36 0.57
4 9.36n 9.6 1.8 8.93 0.42 8.80 0.54 8.98 0.75 9.07 0.36
5 7.63 8.4 0.3 8.30 0.34 8.06 0.37 8.30 0.60 8.27 0.14
6 6.88 8.3 1.1 7.73 0.16 7.59 0.35 7.86 0.60 7.87 0.31
7 6.63 7.0 0.3 7.23 0.49 7.19 0.31 7.35 0.50 7.19 0.15
8 5.75 6.3 0.4 6.65 0.59 6.70R 0.38 6.31R 0.55 6.49 0.21
9 5.25 6.8 0.8 6.30n 0.22 6.82R 0.37 6.86R 0.40 6.69 0.26
10 5.25 5.8n 0.7 5.70n 0.08 6.39n 0.37 6.11n 0.55 6.00 0.31
11 5.13 5.8 0.8 6.07 0.40 6.21 0.37 5.69 0.55 5.94 0.24
12 4.50 5.2 0.5 6.20 0.41 5.68 0.39 5.18 0.60 5.56 0.48
===============================================================================
A. Recomputed from Table 1 of Shastry et al. (1960) in which the total is 400.
C. Computed from data given in the report submitted to IRRI in 1986.
D. The same data are also given by Wu et al. (1985).
E. Standard deviations were computed from standard errors given by the
authors.Means were adjusted so as to make the total 100.
R. Chromosomes 2 and 3 in data A,B and C, and 8 and 9 in data D and E were
replaced with each other so as to be concordant with other data, as mentioned
in text.
n. Nucleolar chromosome designated by respective authors.
n. No. of cells observed
* Mean for data B,C,D and E. Standard deviations represent the variation among
data sets.
===============================================================================
===============================================================================
F G H
Chr. Kurata and Chung and Fukui et al. Mean derived from ANOVA
no. Omura 1978 Wu 1987 1988 ===============================
Sekitori IR36 Nipponbare Pachytene Prometa- Pooled
n: 10 35 10 phase
Mean Mean s.d. Mean s.d.
===============================================================================
1 15.2 13.81 0.97 12.64 0.60 13.82 13.59 13.72
2 11.2R 11.70 0.63 11.20 0.95 12.24 11.58 11.59
3 12.2R 11.12 0.75 10.08 0.84 11.06 10.92 11.00
4 9.6 9.16 0.45 8.98 0.90 8.97 9.13 9.04
5 7.6 8.71 0.50 8.52 0.63 8.21 8.64 8.40
6 7.6 7.91 0.41 8.28 0.82 7.76 7.98 7.86
7 7.0 7.26 0.60 7.78 0.57 7.25 7.73 7.30
8 6.5 6.51R 0.51 7.20 0.34 6.51 6.66 6.57
9 6.2 6.69R 0.32 6.68 0.25 6.79 6.67 6.74
10 6.2n 6.02n 0.47 6.38 0.57 6.18 6.10 6.14
11 5.5 5.79 0.49 6.12 0.21 5.95 5.86 5.91
12 5.1 5.33 0.42 6.12 0.54 5.48 5.50 5.49
===============================================================================
R. Chromosomes 2 and 3 in data F, and 8 and 9 in data G were replaced with each
other so as to be concordant with other data, as mentioned in text.
n. Nucleolar chromosome.
n. No. of cells observed.
===============================================================================
===============================================================================
Pachytene Prometaphase Both Stages Pooled
Chr. ================= ==================== ===========================
no. Author Error Author Error Stage Author Error
df: 4 54-57 2 43 1 6 97-100
===============================================================================
1 10.27** 1.16 6.66** 0.82 1.43 9.07** 1.01
2 1.31 0.93 1.05 0.87 11.24** 1.22 0.90
3 1.92 0.83 5.05** 0.59 0.55 2.96** 0.73
4 0.70 0.56 0.24 0.33 0.68 0.55 0.46
5 0.38 0.23 0.70 0.33 4.87** 0.48 0.27
6 1.29** 0.27 0.61 0.27 1.29* 0.72* 0.27
7 0.29 0.17 1.12 0.35 0.35 0.56 0.25
8 0.62* 0.23 1.87** 0.23 0.62 1.04** 0.23
9 0.90** 0.17 0.12 0.09 0.31 0.64** 0.14
10 0.88** 0.22 0.51 0.24 0.15 0.75** 0.23
11 0.97** 0.24 0.49 0.20 0.25 0.81** 0.22
12 1.84** 0.25 2.51** 0.20 0.01 2.06** 0.23
===============================================================================
*, ** Significant at 5% and 1% levels, respectively. In ANOVA, data A and F
without standard deviations were taken as single measurements.
===============================================================================
=============================================================================== a) Pachytene =============================================================================== Chr. A-B A-C A-D A-E B-C B-D B-E C-D C-E D-E no. =============================================================================== 1 6.9** 6.2** 6.2** 5.7** -0.7 -0.7 -1.1* 0.04 -0.4 -0.4 6 -1.4** -0.9** -0.7** -1.0** 0.6 0.7** 0.4 0.1 -0.2 -0.3* 8 -0.6 -0.9** -1.0** -0.6** -0.4 -0.4 -0.04 -0.1 0.3 0.4* 9 -1.5** -1.1** -1.6** -1.6** 0.5 -0.02 -0.1 -0.5* -0.6* -0.1 10 -0.6* -0.5 -1.1** -0.9** 0.1 -0.6* -0.3 -0.7** -0.4* 0.3* 11 -0.7* -0.9** -1.1** -0.6** -0.3 -0.4 -0.8 -0.1 0.3 0.5** 12 -0.7* -1.7** -1.2** -0.7** -1.0** -0.5* 0.0 0.5** 1.0** 0.5** (sum) d2/7 7.64 6.28 6.61 5.44 0.33 0.27 0.30 0.14 0.27 0.14 ===============================================================================
=============================================================================== Chr. F-G F-H G-H no. =============================================================================== 1 1.3** 2.6** 1.2** 3 1.1** 2.1** 1.0** 8 -0.01 -0.7** -0.7** 12 -0.2 -1.1** -0.8** (sum) d2/4 0.79 3.20 0.86 =============================================================================== *,** Significant at 5% and 1% levels, respectively, (sum)d2 Sum of squares of differences. ===============================================================================
===============================================================================
Pachytene Prometaphase
====================================== ======================
A B C D E F G H Mean
Chr. Shastry Kurata Kurata's Chung Chen Kurata Chung Fukui exc.
no. et al. et al. report &Wu et al. & Omura &Wu et al. A & E
1960 1981 1986 1987 1982 10 1987 1988
n: 10 4 4-7 19-22 25 10 16-22 10 s.d.*
===============================================================================
1 1.72 1.77 1.34 1.79 1.36 2.00 1.81 1.71 1.74 0.22
2 1.23nR 1.18R 1.13R 1.17 1.16 1.37R 1.28 1.25 1.23 0.09
3 2.17R 1.69R 1.58R 1.93 1.89 1.67R 1.88 1.60 1.73 0.15
4 2.08n 4.71 3.72 2.93 2.05 3.45 3.72 2.91 3.57 0.66
5 2.04 1.76 1.84 1.57 1.20 2.22 1.78 1.46 1.77 0.26
6 4.00 1.16 1.06 1.35 1.23 1.27 1.21 1.12 1.20 0.11
7 1.06 1.87 1.33 2.12 1.38 1.22 1.95 1.59 1.68 0.36
8 1.69 1.16 1.11 1.32R 1.20R 1.37 1.39R 1.16 1.25 0.12
9 3.23 1.97 _n 2.44nR 3.81R 1.64 3.05nR 1.24 2.07 0.70
10 5.88 3.49n _n 1.69n 1.79n 2.63n 2.88n 3.85 2.91 0.83
11 1.56 1.31 1.56 1.31 1.78 1.25 1.79 1.28 1.42 0.21
12 3.03 2.89 1.20 1.21 2.55 2.00 1.38 1.93 1.77 0.65
===============================================================================
A. Recomputed from their Table 1 in which short arm/long arm ratio is given.
C. Report submitted to IRRI; arm ratio is not given for nucleolar chromosomes.
R. Two chromosomes designated by R were replaced with each other, as mentioned
in text.
n. Nucleolar chromosome designated by respective authors.
n. No. of cells observed.
* Standard deviations represent variation among data sets.
===============================================================================
References
Chen, J.T., H.C. Lai, Y.H. Hwang, M.C.Chung and H.K. Wu, 1982. Identification of rice reciprocal translocation and the location of lazy gene. Bot. Bull. Acad. Sinica 23: 71-87. (Chinese/English)
Chung, M.C. and H.K. Wu, 1987. Karyotype analysis of IR36 and two trisomic lines of rice. Bot. Bull. Acad. Sinica 28: 289-304. (Chinese/English)
Fukui, K., K., Iijima and K. Kakeda, 1988. Analysis and utility of chromosome information, 21. Karyotype of rice chromosomes. Jpn. J. Breed. 38, Suppl. 1:474-475. (in Japanese)
Iwata, N. and T. Omura, 1984. Studies on the trisomics in rice plants, VI. An accomplishment of a trisomic series in Japonica rice plants. Jpn. J. Genet 59(3):199-204.
Iwata, N., H. Satoh and T. Omura, 1984. Studies on trisomics in rice plants, V. Relationship between the twelve chromosomes and the linkage groups. Jpn. J. Breed. 34(3):314-321.
Khush, G.S., R.J. Singh, S.C. Sur and A.L. Librojo, 1984. Primary trisomics of rice: Origin, morphology, cytology and use in linkage mapping. Genetic 107: 141-163.
Kurata, N. and T. Omura, 1978. Karyotype analysis in rice, 1. A new method for identifying all chromosome pairs. Jpn. J. Genet. 53: 251-255.
Kurata, N., T. Omura and N. Iwata, 1981. Studies on centromere, chromomere and nucleolus in pachytene nuclei of rice, Oryza sativa, microsporocytes. Cytologia 46: 791-800.
Shastry, S.V.S., D.R. Ranga Rao and R.N. Misra, 1960. Pachytene analysis in Oryza, I. Chromosome morphology in Oryza sativa. Ind. J. Genet. Plant Breed. 20: 15-21.
Steel, R.G.D. and J.H. Torrie, 1980. Principles and Procedures of Statistics (2nd ed.). McGraw-Hill, New York, 633pp.
Wu, H.K., 1967. Note on preparing of pachytene chromosomes by double mordants. Scientific Agriculture (Taipei) 15:40-44. (Chinese/English)
Wu, H.K., M.C. Chung and M.H. Chen, 1985. Karyotype analysis of cultivar IR36. RGN 2:54-57.
