Fig. 1. Silica bodies of rice. a)Schematic depiction and four dimensions measured; b) A silica body of alpha type; c) A silica body of Beta type.
1) National Institute of Genetics, Mishima, 411 Japan
2) Faculty of Agriculture, Miyazaki University, Miyazaki, 889-21 Japan
The plant opal, a plant-derived silica body (SiO\2\) excavated from soil, has been used in archaeological and palaeobotanical studies to estimate the past vegetation and crops by analyzing its species-specific morphology (Smithson 1956; Fujiwara 1976).
We have attempted to distinguish between subspecies indica and japonica of Oryza sativa by certain characteristics of silica bodies. We examined 96 cultivars from south, southeast and east Asian countries and 40 wild strains (0. rufipogon Griff., from India, Burma, Thailand, Malaysia, Philippines and China). For classifying the cultivars into indica and japonica, a discriminant function given by Sato et al. (1986) was used in which values of phenol reaction (P, 0 or 1), KClO\3\ susceptibility of seedlings (K, 0 to 2) and apiculus hair length (H, mm) were combined as Z\1\=P+ 1.31K -0.82H.
Silica bodies were extracted from the leaves burnt to ashes, and four dimensions shown in Fig. 1 were measured with 50 samples of each strain. A preliminary study of variations revealed that the silica bodies were classifiable into the alpha (with smaller a, b, d and b/a values and larger c value) and beta (showing opposite trends; Fig. 1) types. A discriminant formula maximizing the difference between the two types was obtained from the data as: Z\2\=O.049(a+b) -0.019(c) +0.197(d) -4.792(b/a) -2.614 (standardized).
The Z\1\ and Z\2\ scores showed a negative correlation (r= -0.682, P<0.001; Fig. 2), indicating that the indica and japonica cultivars tend to have the alpha and beta type of silica bodies, respectively. The Z2 values for silica bodies showed a continuous variation although the variation in Z\1\ value was more discontinuous. Excluding 24 cultivars having an intermediate Z\2\ value (-O.5 to 0.5; 25% of the total), most of alpha-type carriers (36/39) belonged to indica and most of beta-type carriers (31/33) to japonica.
Fig. 1. Silica bodies of rice. a)Schematic depiction and four dimensions
measured; b) A silica body of alpha type; c) A silica body of Beta type.
Fig. 2. Ninety-six rice cultivars (shown by solid circles) and 40 wild
strains (open circles) scattered according to Z\1\ and Z\2\ scores. Z\1\ is
given by a discriminant function distinguishing between subspecies indica and
japonica by combining values of KClO\3\ susceptibility, apiculus hair length
and phenol reaction. Z\2\ is given by another discriminant function
maximizing the difference between alpha and beta types of silica bodies.
The Z\1\ and Z\2\ scores were also calculated in 40 wild strains. The range of their Z\2\ values largely overlapped that of indica cultivars, although their Z1 values were distributed over the space between indica and japonica groups.
The result of this study may be used for analyzing plant opals obtained from archaeological remains, as they remain unchanged for thousands of years when buried in soil. The time and place of origin of subspecies indica and japonica may be elucidated if plant opals excavated from different sites are gathered. The genetic control of the variation between alpha and beta types is under observation in segregating populations.
References
Fujiwara, H., 1976. Fundamental studies of plant opal analysis. On the silica bodies of motor cell of rice plants and their near relatives, and the method of quantitative analysis. Archaeology & Natural Sciences 9: 15-29. (in Japanese)
Sato, Y. I., S. Chitrakon and H. Morishima, 1986. The Indica.Japonica differentiation of rice cultivars in Thailand and its neighboring countries. In New Frontiers in Breeding Researches, p. 185-191. Kasetsart University, Bangkok.
Smithson, F., 1956. Plant opal in soil. Nature 176: 107.