Search

We examined the floral development of Dichocarpum fargesii, Thalictrum fargesii, Thalictrum przewalskii, and Aquilegia yabeana in Thalictroideae, Ranunculaceae, by scanning electron microscope. The sepals are initiated spirally in D. fargesii and A. yabeana, and in two pairs (with four sepals) or spirally (with five sepals) in T. fargesii and T. przewalskii. The petals in D. fargesii and A. yabeana and the stamens and carpels are initiated in a whorled pattern in all three genera. The floral phyllotaxis is whorled in these genera. The primordia of sepals are lunular and truncate, but that of petals and/or stamens are hemispherical, rounded, and much smaller than the sepal primordia. A relatively long plastochron exists between the last sepal and the first petal in D. fargesii and A. yabeana or the first stamen in T. fargesii and T. przewalskii. The similarity between the primordia of petals and stamens may indicate an evolutionary relationship between petals and stamens. The petals develop slower than the stamens in D. fargesii, but faster than stamens in A. yabeana. The early developmental stages of the staminodes in A. yabeana are similar to that of stamens, so they may be phylogenetically homologous organs. The carpel primordia are initiated in a single whorl; are lunular in shape and plicate in A. yabeana and D. fargesii; and are initiated spirally and hemispheric in shape and ascidiate in T. fargesii and T. przewlaskii. The stigma is everted and decurrent with unicellular papillae in T. fargesii and T. przewalskii; the head has unicellular papillae in D. fargesii and is smooth in A. yabeana. The floral development features of Aquilegia are unique in Thalictroideae. [ABSTRACT FROM AUTHOR]

• Petal spurs in Aquilegia kansuensis, A. rockii, A. semicalcarata and A. ecalcarata have different lengths and nectary structures. • Early petal development is similar, and petals undergo delayed development, resulting in different mature petals via developmental remodeling. • There is no obvious one-to-one correspondence between spur length and pollinators in the four species of Aquilegia , and abiotic factors may be an important explanation for spur lengths in our study. The diversification of spur length in Aquilegia is often thought to be driven by pollinators. In this study, the morphology and pollination biology of four sympatric Aquilegia species distributed in China with different petal spur lengths were investigated through scanning electron microscopy, light microscopy, and field observations and measurements. The morphology and spur lengths of the four Aquilegia species differed: Aquilegia kansuensis has long incurved spurs (14.82 ± 0.18 mm), A. rockii has long straight spurs (16.34 ± 0.18 mm), A. semicalcarata has short slender spurs (5.16 ± 0.17 mm), and A. ecalcarata is effectively spurless (1.20 ± 0.03 mm). Early petal development was similar among the species; petals undergo initiation and delayed development, resulting in mature petals with different spur lengths and morphologies via developmental remodeling. The long spurs of A. kansuensis and A. rockii were derived from an increased cell number and anisotropic cell expansion, with well-developed secretory tissue forming at the spur bottom. No secretory tissue formed in the spurs of A. semicalcarata and A. ecalcarata , even though their cell number increased. Although their spur morphology and nectary structure show considerable differences, the pollination systems of the four species are similar. Rhingia campestris is the dominant pollinator of A. kansuensis, A. semicalcarata , and A. ecalcarata , whereas Pipiza familiaris is the dominant pollinator of A. rockii. Both R. campestris and P. familiaris forage for pollen when visiting flowers. Given that the four species studied differ in spur length but are all pollinated by short proboscid pollinators, there is no obvious one-to-one correspondence between spur length and pollinator mouthpart length in A. rockii, A. kansuensis, A. semicalcarata and A. ecalcarata , which is similar to the pattern in the European Aquilegia alpina , and abiotic factors may be an important explanation for spur lengths in our study. [ABSTRACT FROM AUTHOR]