龙胆科蔓龙胆属及其相关类群的系统学研究

本文回顾了龙胆科蔓龙胆属、双蝴蝶属、龙胆属狭蕊组和翼萼蔓属的研究历史，对这些类群的外部形态、花部解剖结构、染色体、花粉形态及胚胎学进行了研究，并结合已有的资料，探讨了这些类群的分类学问题。主要结论如下： 1．外部形态 蔓龙胆属、双蝴蝶属、龙胆属狭蕊组和龙胆属其他组之 间以及蔓龙胆属和翼萼蔓属之间在外部形态上均存在差异。 2．花部解剖结构 蔓龙胆属、双蝴蝶属、龙胆属狭蕊组和龙胆属其他组的花部解剖结构存在较多地相似之处，同时蔓龙胆属和双蝴蝶属的花被维管束和腺体比龙胆属狭蕊组和龙胆属其他组的特化。而翼萼蔓属的花部解剖结构与蔓龙胆属的差异较大，与扁蕾属的花部解剖结构较相似。 3．染色体 本文对蔓龙胆属、双蝴蝶属、龙胆属狭蕊组和翼萼蔓属的7种植物的染色体数目、基数和核型对称性作了研究，其中6种植物的染色体为首次报道，它们是：披针叶蔓龙胆C．delavayi、双蝴蝶T.chnense、峨眉双蝴蝶T．cordatum、锯齿龙胆 G．serra、毛脉龙胆G．souliei、和翼萼蔓P．volubilis。蔓龙胆属、双蝴蝶属和龙胆属狭蕊组的核型不对称性稍强于龙胆属其他组，但在染色体组成上十分相似。翼萼蔓属和蔓龙胆属的染色体差异较大，面与扁蕾属的染色体较相似。 4.花粉形态蔓龙胆属、双蝴蝶属、龙胆属狭蕊组和龙胆属其他组均具有球形或长球形、3孔沟、沟长而宽、外壁纹饰为条纹一无穿孔和条纹一穿孔的花粉类型。而翼萼蔓属的花粉为球形、纹饰为网状纹馋与蔓龙胆属不同，与扁蕾属的花粉较相似。 5．胚胎学 本文对披针叶蔓龙胆C，delavayi、双蝴蝶T．chinense、峨眉双蝴蝶T．cordatum、条纹龙胆G. striata和翼萼蔓P．volubilis作了较详细的胚胎学研究，这些类群在花药壁的分化和形成、胚珠和胎座类型、卵器中的助细胞和反足细胞、种皮以及胚发育等特征上存在差异，并且在胚胎学特征上各自拥有原始和进化的性状。蔓龙胆属、双蝴蝶属和龙胆属之间的胚胎学性状差异较大，龙胆属狭蕊组与双蝴蝶属的胚胎学性状较相似，而与蔓龙胆属和龙胆属其他组的差吴较大。翼萼蔓属的胚胎学性状与扁蕾属的较相似，而与蔓龙胆属的差异较大。 6．综合分析外部形态、花部解剖结构、染色体、花粉形态及胚胎学性状，结果表明，蔓龙胆属、双蝴蝶属、龙胆属狭蕊组和龙胆属其他组是亲缘关系十分相近的类群，在某些特征上已存在明显地演化极向，但在大多数特征上演化极向不明。蔓龙胆属和双蝴蝶属分别作为独立属处理较为合适，而且两属的亲缘关系很近。狭蕊组仍应放在龙胆属，蔓龙胆属比双蝴蝶属与龙胆属的亲缘关系更近一些。翼萼蔓属也应作为一独立属处理，而且该属可能与扁蕾属的亲缘关系较近，而与蔓龙胆属亲缘关系较远。 7．本文还讨论了蔓龙胆属、双蝴蝶属、龙胆属狭蕊组和翼萼蔓属的地理分布。

条纹龙胆的胚胎学研究(英文)

首次报道了条纹龙胆(Gentiana striata Maxim.) 的胚胎学特征, 研究结果用以讨论龙胆属狭蕊组(Gentiana Sect.Stenogyne) 的系统演化关系。主要研究结果如下: 花药四室; 药壁发育为双子叶型; 绒毡层细胞仅来源于初生壁细胞, 故绒毡层起源属单型起源, 细胞具单核,原位退化, 属腺质型绒毡层, 药隔处的绒毡层细胞经多次平周分裂形成2 层至多层的绒毡层细胞, 其余部位的绒毡层细胞仍为1 层细胞; 中层细胞1 层；在花药成熟时, 花药的表皮细胞和药室内壁均部分纤维状加厚且柱状伸长。小孢子母细胞减数分裂为同时型, 四分体的排列主要为四面体形；成熟花粉为32细胞型。子房为2心皮, 1室, 侧膜胎座。胚珠4列。薄珠心,单珠被, 珠心基部产生珠被原基, 进而形成珠被, 条纹龙胆仅有1 层珠被。珠被沿珠心向上生长并将珠心包围, 于胚珠顶部形成珠孔。胚珠在发育过程中, 整个胚珠的本体倒转, 而且珠柄继续生长并弯曲, 使珠孔与合点端的连线与珠柄垂直, 形成Hypertropous 胚珠。大孢子母细胞减数分裂形成的4 个大孢子呈直列式排列, 合点端的大孢子具功能。胚囊发育为蓼型。极核在受精前融合为次生核, 反足细胞3 个、多宿存。雄蕊先熟。珠孔受精。胚乳发育为核型。胚胎发育为茄型酸浆Ⅱ变型。通过比较龙胆属狭蕊组与龙胆属其它组和双蝴蝶属的胚胎学特征表明, 龙胆属狭蕊组在一些重要的胚胎学特征上与双蝴蝶属较相似, 而与龙胆属其它组存在较大差异, 故建议应将龙胆属狭蕊组从龙胆属中移出。

Metagentiana, a new genus of Gentianaceae

The evidence from cross morphology, floral anatomy, chromosomes, palynology, and embryology all indicates that sect. Stenogyne is discordant within the genus Gentiana and is as distinct from the other sections of Gentiana as are other genera, such as Tripterospermum and Crawfurdia. In light of these characters, sect. Stenogyne is removed from Gentiana and given generic rank as the new genus Metagentiana. It is more related to Tripterospermum and Crawfurdia than to Gentiana, though it is more primitive than the first two genera. Together with Tripterospermum and Crawfurdia the new genus forms a monophyletic group, which is the sister group to the genus Gentiana. Fourteen new combinations required at specific rank are proposed.

Taxonomy, phylogeny and reproductive ecology of Gentiana lutea L.

This research focuses on taxonomy, phylogeny and reproductive ecology of Gentiana lutea. L.. Taxonomic analysis is a critical step in botanical studies, as it is necessary to recognize taxonomical unit. Herbarium specimens were observed to assess the reliability of several subspecies-diagnostic characters. The analysis of G. lutea genetic variability and the comparison with that of the other species of sect. Gentiana were performed to elucidate phylogenetic relationships among G. lutea subspecies and to propose a phylogenetic hypothesis for the evolution and the colonization dynamics of the section. Appropriate scientific information is critical for the assessment of species conservation status and for effective management plans. I carried out field work on five natural populations and performed laboratory analyses on specific critical aspects, with special regard to G. lutea breeding system and type and efficiency of plant-pollinator system. Bracts length is a reliable character to identify subsp. vardjanii, however it is not exclusive, hence to clearly identify subsp. vardjanii, other traits have to be considered. The phylogenetic hypotheses obtained from nuclear and chloroplast data are not congruent. Nuclear markers show a monophyly of sect. Gentiana, a strongly species identity of G. lutea and clear genetic identity of subsp. vardjanii. The little information emerging from plastid markers indicate a weak signal of hybridization and incomplete sorting of ancestral lineages. G. lutea shows a striking variation in intra-floral dichogamy probably evolved to reduce pollen-stigma interference. Although the species is partially self-compatible, pollen vectors are necessary for a successful reproduction, and moreover it shows a strong inbreeding depression. G. lutea is a generalist species: within its spectrum of visitors is possible to recognize "nectar thieves" and pollinators with sedentary or dynamic behaviour. Pollen limitation is frequent and it could be mainly explained by poor pollen quality.

Germinating the seeds of three species of Pimelea sect. Epallage (Thymelaeaceae)1

Pimelea trichostachya Lindl., P. simplex F.Muell. and P. elongata Threlfall frequently cause pimelea poisoning of cattle. Fresh seeds of these species, belonging to sect. Epallage (Endl.) Benth. of Pimelea Gaertn. (Thymelaeaceae) are strongly dormant for years when in laboratory storage. Common methods of stimulating germination, such as scarification, dry heat and cold stratification, did not remove much of the dormancy. ‘Smoke water’ stimulated some germination but its effect was unpredictable and many seedlings then grew aberrantly. Exposure of imbibed seeds to gibberellic acid greatly and reliably improved the germination of all three species. However, the manner of application and the concentration of gibberellic acid used had to be appropriate or many young seedlings grew abnormally or died suddenly, limiting successful plant establishment rates. The dormancy type involved is non-deep Type 2 physiological. Ten days of good moisture, in addition to gibberellic acid exposure, is required before appreciable laboratory germination occurs at optimal temperatures. Thus, the mechanism by which gibberellic acid stimulates good germination does not appear to be the same as that which primes seeds for the rapid and prolific germination often seen under natural conditions in arid Australia. Seeds of P. simplex subsp. continua (J.M.Black) Threlfall proved most difficult to germinate and those of P. elongata the easiest.