| Abstract: | The karyotypes of five species of the germs Consolida from SE Europe, Turkey and Iran (see Appendix Ⅰ for the detail information concerning the vouchers) were studied with 0.05% colchicin pretreatment followed by Carney Ⅰ fixation and Fenlgen squashing. The result shows that C. regalis ssp. regalis has a karyotype of 2n=16= 1(L)m- (SAT)+ 3(L)m + 6(S)st + 5(S)t+1(S)t(SAT)(Fig. 1A and Plate Ⅰ, G) and ssp. paniculata has a karyotype of 2n = 16 = 2(L)m (SAT) + 2(L)m + 6(S)st + 6(S)t (Plate Ⅱ,A) and 8 bivalents in meiosis (Piate Ⅱ, E, F); C. persica 2n = 14 = 2 (L)m (SAT) + 2(L)m + 3(S) st + 7(S) t(Plate Ⅲ, B; Fig. 1,B); C. stenocarpa 2n =16 = 2(L)m(SAT) + 2(L)m + 1(S)st + 11(S)t (Plate Ⅰ, C, Fig. 1, C); C, teheranica (see Appendix Ⅰ for the nomenclature) 2n = 16 = 4(L)m(SAT) + 2(S) st + 10(S) t (Plate Ⅲ, D; Fig. 1, D); C. scleroclada var. rigida 2n = 18 = 2(L)m (SAT) + 2(S)st + 14(S)t(Plate Ⅱ H, Ⅰ; Fig. 1, E) All the karyotypes here descr- ibed are highly asymmetrical and bimodal, and belong to the type 3C in the karyotype classification system established by Stebbins14,15]. 2n=18 for the last mentioned taxon has been confirmed by the ,standard microtome sectioning method. Its meiosis was also examined with acetoorcein staining, and 9 bivalents were always found at MI, no meiotic aberrations being Observed. x=7 and x=9 are two new basic numbers even for the whole tribe of Delphineae. It is considered that the karyotype of 2n=18 is derived from that of 2n=16 by centic fission (Robertsonian exchange), while the karyotype of 2n=14 is derived from that of 2n=16 probably by successive unequal interchanges. As shown to Fig. 1 and 2. the complement of C. sclerocleda var. rigida (2n=18) has only one pair of large and metacentric chromosomes instead of 2 pairs of such chromosomes in the other species, but it has 2 extra pairs of small and terminal chromosomes as compared with the species with 2n=16. The complement in the taxon has, therefore, exactly the same fundamental number of chromosome arms as that of the other species with 2n=16 (for example, of C. stenocarpon), but it has two more centromeres. There are at least 2 pairs of chro- mosomes in the complement (3 and 5) which may be telocentric, i.e. T chromosomes in the sense of Levan et al8] The small dots at the ends of the chromosomes may be the chromomeres in centric regions rather than short arms (Jones6]). As the plants constantly show 9 bivalents .at the first meiotic division and have very high pollen fertility (98%) as well as good seed-set, the karyotype seems to be a stable one. Therefore, Consolida scleroclada var. rigida may have provided another example of spontaneous centric fission which has resulted in homozygous and stable telocentrics. John and Freemanm have argued for the mechanism of chromosomal structural variation based on the observed facts both in animals and plants. The cytogenetic model for the variation was formulated by Lima-de-Faria as early as in 1956 and revised by Jones6], and the mo- lecular model for the mechanism recently by Holmquist et al.4] As in the genus Delphinium, most species of Consolida are pollinated by long-tongued bumble-bees. In C. regalis (incl. both subspecies), C. stenocarpa and C. scleroclada var. rigida the isolated flowers (3–15 for each species) gave no any seeds. The flowers first emasculated and isolated and then pollinated with pollen collected from the same individuals in these three species (10–25 flowers for each species), however, all gave full seed-set. The experiment clearly shows that these three species, though self-com- patible, are obligately out-pollinated. It was Observed that the three species are pro- tandrous. When stigmata become 2-lobed and show their receptive surface, all the stamens have recurved down or laterally, forming a semi-circle, but the styles remain erect. Therefore the receptive stigmata are over 3 mm (C. regalis) or 5 mm (C. stenocarpa and C. scloroclada var. rigida) away from and above the anthers of the same flower Plate I, B, D and F). Self-pollination is thus prevented. Just-opened flowers, however, have always some stamens erect and with their dehiscing anthers correspondent in position to 2-lobed and receptive stigmata of other flowers (compare A with B, C with D, E with F in Plate II). Pollen grains are therefore easily taken by a bumble-bee from dehiscing anthers onto receptive stigmata. Here we see a perfect mechanism which prevents self-pollination and secure out-pollination. It was observed during the experiment that a bumble-bee, Bombus agrorum F., only visited the straight-spurred species, C. regalis (both subspecies), but never visited the curved-spurred species, C. scleroclada var. rigida and C. stenocarpa. Another bumble-bee, B. hortorum L., however, visited both the straight-spurrod and curved-spurred species, but when it visited C. stenocarpa it sometimes kept the body upside down. Consolida teheranica (Boiss.) Hong, on the contrary, is an inbreeder. Its stigmata and anthers become mature at the same time, and its styles and stamens always remain erect with the dehiscing anthers right over the receptive stigmata. It was also found that its corollae are not fully opened (Fig. 3). As expected, two isolated flowers gave 9 good seeds. The results of crossing experiment axe shown in Fig. 4. Only the cross between two subspecies of C. regalis resulted in an interfertile hybrid, which was vigorous, showed normal meiosis, had 94% pollen fertility, and gave good seed-set. The cross combination C. stenocarpa×C. scleroclada var. rigida gave some 50 % seed set, but the seeds yielded from the cross did not germinate though looked good. The other crosses gave no any seeds. |
