(3) Further detailed deep investigation on the distant hybridization plants.
During a period of years from the middle of 1980's to the middle 1990's, studies were mainly made concentrating on the following subjects, and the results will be summarized as below:
(a) Cross compatibility of A. tsukushiense, A.ciliare and A. humidorum with the rest of genera in Triticeae.
(b) Chromosome number instability and production of polyhaploid plants.
(c) Production of amphiploids.
(d) Analysis of the chromosome pairing configurations using Agropyron species.
(e) Production of addition monosomics.
(f) Studies on agronomically useful characteristics in relation to plant genetics and breeding.
(g) Anomalous cell divisions.
(a) Compatibility
Extensive studies were carried out on cross compatibility using A. tsukushiense and A. ciliare as female, pollinated by species and strains of the genera involving Triticum, Aegilops, Elymus, Leymus, Hystrix and Hordeum*. In all cross combinations, it had been tested that seeds obtained are abortive and never germinate. The hybrid plants are only possible to obtain by means of embryo rescue (Kaneta et al. 1984, Muramatsu et al. 1983, Muramatsu et al. 1988, Muramatsu et al. 1989). Therefore, the term "cross-compatibility" in this paper will be used in a meaning of ability that a zygote to develop into adult phase, with/ without embryo rescue. It was expected that the rate to produce hybrid plant may vary among cross combinations.

Wide and extensive experiments were carried out.With the total number of pollinations included more than 13,000 florets. The results of the paper submitted to the Proceedings of the 1st International wnTriticeae Symposium, Sweden are in Tables 5a-5e (Muramatsu et al. 1992). The totaled results represented experiments conducted during years from 1988-1990. Separate results will be described in the next section. Both A. tsukushiense and A. ciliare have shown cross-compatibilitywith Triticum, Aegilops and Hordeum by embryo rescue (Kudo et al. 1990; Muramatsu et al. 1990; Uno et al. 1990; Muramatsu 1991; Muramatsu et al. 1991). With various species of Triticum, A. ciliare produced a total of 45 hybrid plants out of 1,982 florets pollinated, while in A. tsukushiense two plants out of 1,756 pollinated. Similarly, the difference was clear when A. ciliare as female was pollinated by wheat and yielded 18.9 % of seed set, while in A. tsukushiense only 0.7% Also, higher tendencies were obtained with pollen of Aegilops species and Secale. However, it is interesting that when barley pollen, H. vulgare and H, bulbosum, were used, the tendencies indicated that the number of resulting plants increased in the crosses using A. tsukushiense. Out of 2,407 florets pollinated 544 (22.6%) of seeds were obtained and 368 embryos cultured gave 217(59.0%) plants. With A. ciliare, out of 2,151 florets 41.3% of seeds were obtained and from 354 embryos cultured 7 plants (2.0%) were recovered. Results of studies on the Agropyron - barley hybrids will be described in a later section.

Crosses with rye was obtained in combinations, A. tsukushiense x rye 2x (Kaneta et al. 1984), and in A. ciliare x rye (Kudo et al. 1990). The hybridization was also possible with A. ciliare and A. yezoense as female to Hystrix longe-aristata (Muramatsu 2001). With further experiment using A. humidorum as an indigenous Agropyron species, similarly we produced hybrid plants. Cytogenetic results obtained in the hybrids resembled those involved A. tsukushiense (Muramatsu et al. 1992; Nakatsuji et al. 1992; Muramatsu et al, 1993,)

Results show the wide cross-compatibility throughout the genera of the tribe Triticeae. However, there are tendencies that the hybrids show cytological abnormalities in certain cross combination, presumably due to the interaction or complementary effects of the genotypes brought together in the hybrid from the parents. In Fig. 7 is shown the spike of the F₁ hybrid between A. humidorum x T. aestivum and the parents.