Androecium homologies in eight-staminate maples: a developmental study

preprint OA: closed
Full text JSON View at publisher
Full text 183,728 characters · extracted from preprint-html · click to expand
Androecium homologies in eight-staminate maples: a developmental study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Androecium homologies in eight-staminate maples: a developmental study Alexander Zavialov, Margarita V. Remizowa This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4754778/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 25 Apr, 2025 Read the published version in Journal of Plant Research → Version 1 posted 4 You are reading this latest preprint version Abstract The genus Acer belongs to the family Sapindaceae, whose representatives are characterized by a pentamerous perianth but typically possess only eight stamens. Such an androecium is believed to have evolved through the loss of two stamens. However, there is still no consensus on the origin of eight-staminate androecium including the positions of the two lost stamens and the pathway of their reduction compared to other Sapindaceae. We examined the early stages of flower development in five maple species belonging to different sections – four species with eight stamens and one species with ten stamens – using scanning electron microscopy. Measurements were performed to analyze the relative positions of stamen primordia, their size, and the floral meristem surface area. In addition, the perianth and androecium vasculature was studied to reveal petal-stamen complexes. We found that in three of four 8-staminate species, three stamens are initiated from common petal-stamen primordia, and five arise from single primordia. In A. tegmentosum Maxim., four stamens appear from common primordia with petals, and four from single primordia. Despite developmental differences, stamen distribution within the flower and the angles between adjacent stamens indicate a similar androecium construction in all species. In most species with eight stamens, the differences between two andoecial whorls are vanished. In contrast, A. nikoense (Miq.) Maxim., with ten stamens, possesses two distinct stamen whorls, the antipetalous stamens are initiated from common primordia. In the 8-staminate androecia of the genus Acer , the same two stamens have been lost as in other Sapindaceae. Within genus Acer , there is a certain decrease in the relative size of the floral meristem, accompanied by an increase in the number of common petal-stamen primordia and increased heterogeneity of the androecium (in A. tegmentosum ) or reduction of some floral organs. common petal-stamen primordium stamen loss dynamic morphology Acer Sapindaceae Sapindales Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 INTRODUCTION Acer is one of the most morphologically diverse genera (De Jong 1976 ) in the family Sapindaceae (Acevedo-Rodríguez et al. 2010 ; Buerki et al. 2021 ). Now, 15 sections are recognized within the genus (Areces-Berazain et al. 2021 ; De Jong 2004 ; Li et al. 2019 ). Most species of Acer (sections Acer , Glabra , Ginnala , Platanoidea , Palmata , Parviflora , Pentaphylla , Spicata , Macrantha , Lithocarpa ) share the same flower ground plan (De Jong 1976 ). Typical maple flowers are morphologically bisexual but functionally unisexual, with five usually free sepals, five free petals, eight stamens, and two carpels (Acevedo-Rodríguez et al. 2010 ; De Jong 1976 ). But in other sections ( Rubra, Indivisa, Arguta, Negundo, Macrophylla, Pubescentia , partially Pentaphylla ) there is an increase or decrease in some floral organs or the flower merism (De Jong 1976 ). In particular, in section Pentaphylla , series Grisea there are ten stamens, which probably correspond to the typical for eudicots pentamerous flower with two-whorled androecium (De Jong 1976 ; Ronse De Craene 2022). Non-isomerous perianth and androecium have been provoking numerous hypotheses on the homologies and the origin of eight-staminate androecium in the genus. Payer ( 1857 ) studied flower development in Acer tataricum L. (section Ginnala ) and came to the conclusion that stamens are initiated in two series, though in the mature flower they seemingly form a single whorl. After Payer, sepals are initiated following a 2/5 spiral sequence. After the sepals, the petals are initiated simultaneously. After the petal inception, five stamens appear simultaneously opposite the sepals. Slightly later, the last three stamens arise side by side with already existing primordia. This stamen dédoublement (Ronse De Craene and Smets, 1993 ) is rather unusual because duplication of stamens in certain positions occurs sequentially, whereas typical dédoublement implies the simultaneous appearance of two primordia in the position that was previously occupied by a single organ. According to Payer’s general theory, whorled organs that arise sequentially belong to different whorls. Therefore, five of the eight stamens belong to the first whorl, and three belong to the second whorl. Buchenau ( 1861 ) studied flower development in A. pseudoplatanus L. (section Acer ). His description of sepal and petal initiation is similar to A. tataricum. However, the description of stamen development is different. According to Buchenau ( 1861 ), all eight stamens are initiated simultaneously, but their subsequent growth is unequal. These differences in developmental rates give a false impression of sequential initiation. Buchenau did not provide any interpretation of stamen positions. Eichler ( 1878 ) examined organ arrangement in mature flowers in many maple species and concluded that the androecium consists of two whorls (four stamens in each whorl). He suggested that ancestrally, there were ten stamens, and two stamens from different whorls located opposite the carpels were lost. Yet another hypothesis was proposed by Hall ( 1951 ). He studied the floral vasculature in different species of Acer . Most of the data obtained were not informative for androecium interpretation except Acer pensylvanicum L. (section Macrantha ) whose floral vasculature differs from other species. Vascular bundles of five stamens formed common traces with those of petals, and bundles of three stamens formed common bundles with sepals. On this basis, Hall ( 1951 ) suggested that a typical maple flower has five antipetalous and three antisepalous stamens. All the interpretations mentioned above are based either on the mature flowers or on the floral ontogeny of a single species. Despite this diversity of opinions, Eichler’s interpretation is the most widespread and commonly accepted (Ronse De Craene, 2022). The placement of Aceraceae into the family Sapindaceae (APG IV 2016 ; Buerki et al. 2010 , 2021 Harrington et al.; 2005 ) made the story even more puzzling. Now, eight stamens are considered an important feature and probably a synapomorphy of the family Sapindaceae (Acevedo-Rodríguez et al., 2010 ). The flower development in Dipteronia , the closest relative of the genus Acer , does not fully support any hypothesis proposed for Acer (Zhang et al. 2022 ). The androecium of Dipteronia is organized in the same way as in other Sapindaceae (Ronse De Craene et al. 2000), it consists of three antipetalous and five antisepalous stamens, but the positions of the two lost stamens are different in Dipteronia compared to Acer (Zhang et al. 2022 ). If the androecium of maples is of another nature, this could be an argument in favor of the independent origin of eight stamens in different groups of Sapindaceae. Thus, we decided to conduct a study of androecium construction in several 8-staminate species of Acer in comparison with 10-staminate species to clarify androecium homologies in the genus using both flower development and anatomy. MATERIALS AND METHODS This study focuses on the early flower development of five species: A. tataricum subsp. ginnala (Maxim.) Wesm. (section Ginnala ), A. tegmentosum (Maxim.) Maxim. (section Macrantha ), A. platanoides L. (section Platanoidea ), A. spicatum Lam. (section Spicata ), and A. nikoense (Miq.) Maxim. (section Trifoliata ). The androecium of all species except A. nikoense consists of eight stamens. The plant material was collected in several parks of the city of Moscow, in the Botanical Garden of Moscow State University, in the Tsitsin Main Botanical Garden of the Russian Academy of Sciences, and in the Saint Petersburg Botanical Garden during three years (2021–2023). The vouchers are deposited in the Herbarium of Moscow State University (MW) or in spirit collection (MW). Information about vouchers is presented in Supplementary Data. For scanning electron microscopy (SEM), generative buds at different developmental stages were dissected under an Olympus SZX7 stereomicroscope. Flowers were dehydrated through absolute acetone, critical-point dried using a Hitachi HCP2 critical-point drier, and then coated with gold and palladium using an Eiko IB-3 ion-coater. Observations were made using a CAMSCAN S2 SEM, JEOL JSM-6380LA, and Quattro S at Moscow University. For light microscopy, floral buds of three species ( A. tataricum subsp. ginnala , A. tegmentosum , and A. nikoense ) were sectioned using standard methods of paraffin embedding and serial sectioning at 10 µm thickness (Barykina et al. 2004 ) using a Thermo Scientific Microm HM 355s rotary microtome. Sections were stained with Safranin and Alcian Blue and mounted in Biomaunt mounting medium. Digital photomicrographs were taken using an Olympus BX53 (Olympus) microscope fitted with an Olympus SC50 digital camera. Drawings of serial cross sections showing floral vasculature were performed in Krita 5.1.5 software (free raster graphics editor, software developers are Krita Foundation, KDE). A. tataricum subsp. ginnala was selected as a species with the typical vasculature described by Hall ( 1951 ). The vasculature of A. tegmentosum and A. nikoense were examined for the first time. Using top view SEM images, we analyzed the stamen positions and the area occupied by stamen primordia in three species: A. tataricum subsp. ginnala , A. tegmentosum , A. platanoides . Additionally, we studied the area of the floral meristem in three species: A. negundo L., A. barbinerve Maxim., A. saccharinum L. These species were selected because they possess a relatively flat receptacle at the stage of stamen initiation. This makes measurements more accurate and avoids systematic errors caused by floral meristem convexity. The stamens were color coded in the sepal spiral coordinate system (Fig. 1 ), starting with the stamen opposite the fifth (last initiated) sepal and then moving around the flower towards the third sepal. Each stamen was assigned a different color. This labeling was adopted because the stamens are always located in the same specific way relative to the order of sepal initiation but not their positions relative to the flower-subtending bract. Next, for ten flowers of each species, we calculated the angles between adjacent stamens and the angles between the stamens and the centers of the sepals and petals closest to them. Based on the latter measurements, the “antipetality index” was calculated for each stamen: α p /(α p + α s ), where α p is the angle between the petal and the stamen, α s is the angle between the sepal and the stamen. If the index is 0, the stamen is strictly antipetalous. If the index is 1, the stamen is strictly antisepalous. To calculate the area occupied by stamen primordium on the floral meristem, the stamen primordia were colored black on the SEM photos, then the size of the black areas was calculated in pixels in ImageJ software and converted to square micrometers according to scale bars. The results were recalculated into relative values. The area of each primordium was divided by the area of the largest primordium in this flower. Therefore, the relative size of one of the eight stamens is 1. The average antipetality index and the relative primordial area of each stamen are shown in the scatterplots. These diagrams were made in the Past 4.03. Two qualitative characters were also analyzed: the stamen initiation via petal-stamen primordia and the presence of common vascular bundles with petals or sepals. The presence of a common petal-stamen primordium was scored as 0, the absence as 1. The presence of a common vascular bundle with the petal was scored as 0, with the sepal as 1. If the vascular bundles of the stamens were not associated with the perianth vasculature, this was scored as 0.5. For each stamen, the final parameter was calculated based on the sum of four characters: the presence of a common primordium, the presence of a common vascular bundle, the average relative area of the stamen primordium, and the average antipetality index. This parameter shows the degree of belonging of the stamen to the external or internal androecial whorl, as well as the degree of heterogeneity of the androecium. This parameter varies from 0 to 4, its gradations on the diagrams were displayed in color. The surface area of the floral meristem at the time of corolla and androecium initiation was measured in four species ( A. tataricum subsp. ginalla , A. platanoides , A. tegmentosum , A. nikoense ) using the ImageJ software according to the method described above. We did not measure the meristem area of A. spicatum using this method because the flower meristem of this species is too convex, which increases the measurement error. Also, for comparison with species with different stamen numbers we measured the area of the floral meristem in A. negundo (developmental data published in Zavialov and Remizowa, 2024 ), A. barbinerve and A. saccharinum . A. negundo usually has four stamens and sepals, no petals, A. barbinerve has four stamens, petals, and sepals, A. saccharinum has five or six stamens and sepals, no petals. None of these species have common petal-stamen primordia (Zavialov and Remizowa 2024 ; Zavialov 2024 ). An analysis of variance was performed to compare the areas of the floral meristem in the RStudio program and a histogram was constructed. All illustrations were prepared in Krita 5.1.5. Row data for all measurements are presented in the Supplement. We want to discuss some of the terms used in this paper. We use the terms median (abaxial and adaxial) and transversal to describe the position of sepals and petals. In a typical pentamerous flower, one of the sepals or petals occupies a strictly median position (Fig. 1 ). Four other sepals (petals) occupy intermediate positions in pairs closer to the adaxial or abaxial side. Therefore, the terms “transversal-adaxial” and “transversal-abaxial” are used to describe such positions (Fig. 1 ). We use the same terms to describe the position of the stamens or indicate their approximate position relative to the spiral of sepal initiation. We avoid the widely used terms “antipetalous” (opposite the petal median) and “antisepalous” (opposite the sepal median), because the actual position of the stamens in maples often does not fit such strictly defined concepts. Instead, we divided stamens into those arising from a single primordium and those arising from a common petal-stamen primordium. Only for A. nikoense we used the terms antipetalous and antisepalous, because the positions of stamens are more consistent with these concepts. In SEM photographs, we considered two cases to be a common petal-stamen primordium: an elongate smooth primordium from a single mass of meristem, a primordium with two apices and a common basal part raised above the surface of the receptacle. RESULTS Sequence of organ initiation A. platanoides Sepals are initiated following a 2/5 spiral sequence. The first sepal is formed in a transversal-abaxial or transversal-adaxial position (Fig. 2 a). The median sepal can be in either an abaxial or adaxial position. The sepal primordia are of different sizes, but they are quite wide to occupy the entire flower circumference. The initiation of petals and stamens begins even before the formation of the last fifth sepal (Fig. 2 b). Five stamens appear earlier: four are initiated by a single primordia opposite the first, second, third, and fourth sepals and one is initiated by a common petal-stamen primordium between the first and fourth sepals (Fig. 2 b, c). Simultaneously with the first series of stamens, two petals appear between the first and third sepals and between the second and fourth sepals (Fig. 2 b, c). Later, the three remaining stamens appear: two are initiated by a common petal-stamen primordia on the sides of the fifth sepal, and one is initiated by a separate primordium opposite the fifth sepal (Fig. 2 d). Next, the common petal-stamen primordia divide into the petal and stamen parts (Fig. 2 e, f). The common petal-stamen primordium located between the second and fifth sepals or between the third and fifth sepals is delayed relative to the two others (Fig. 2 f). The stamens emerging from common primordia are slightly smaller at the beginning, but subsequently, the differences in size quickly disappear (Fig. 3 a). The petals do not differ in size or developmental rates. During the development of other floral organs, petals flatten and grow slowly (Fig. 3 b – d). Petal enlargement occurs in the later stages of development, shortly before flowering. A. tataricum subsp. ginnala The sepals are initiated in a 2/5 spiral. The first sepal is always initiated in a transversal-abaxial position, and the second sepal is always initiated in the median adaxial position (Fig. 4 a, b). The bar-shaped sepal primordia are of different size, as in A. platanoides . The first petal appears between the first and third sepals, then the petal between the second and fourth sepals (Fig. 4 b, c). Next, three petals simultaneously arise from common petal-stamen primordia. At the same time, five stamens are initiated from a single primordial (Fig. 4 c). Soon, the common primordia are successively (similar to A. platanoides ) divided into staminate and petal parts (Fig. 4 d, e). At this stage, the stamens located on the radii of the carpels are smaller, than the others (Fig. 4 f). However, differences in stamen size soon disappear. Petal growth is retarded, but by the time of anther formation, the stamens are already covered by petals (Fig. 4 g). A. spicatum The sepals are initiated in a 2/5 spiral as small primordia (Fig. 5 a). As a result, it can be difficult to trace the spiral of sepal initiation at later stages. The first sepal is usually formed in a transverse-abaxial position (Fig. 5 a). At the time of petal initiation, the receptacle remains clearly convex (Fig. 5 b). Otherwise, the pattern of petal and stamen initiation is similar to that of A. tataricum subsp. ginnala . First, two petals appear sequentially between the first and third sepals and between the second and fourth sepals (Fig. 5 b). Then the three remaining petals and eight stamens appear: five stamens are initiated from single primordia, three are initiated from common primordia with petals (Fig. 5 c – f, Fig. 6 a). Immediately after their inception, the stamens are identical in size or their sizes slightly differ, although no regular patterns can be seen in the distribution of stamens with different sizes (Fig. 6 b, c). The petals remain small after initiation (Fig. 6 d). Their main growth occurs in the later stages of flower development. A. tegmentosum The sepals are initiated in a 2/5 spiral by small primordia, almost indistinguishable in size (Fig. 7 a, b). The first sepal is usually in a transversal-abaxial position, and the second is in a median adaxial position (Fig. 7 a, b). Next, a petal between the first and third sepals appears, and a stamen located opposite the third (or less often the first sepal) is initiated by a single primordium (Fig. 7 b). Next, the four stamens are initiated by single primordia in a spiral that continues in the direction of the spiral of sepal initiation. After the initiation of the first four stamens by single primordia, four common petal-stamen primordia appear almost simultaneously, between the third and fifth, second and fifth, first and fourth, second and fourth sepals (Fig. 7 b). The division of common primordia appears sequentially, starting from the primordium between the second and fourth sepals (Fig. 7 c, d). During further development, the stamens differ (Fig. 7 d – f). The stamens, which are initiated via common primordia, remain small. The stamens, which arose from single primordia, have different sizes, corresponding to the order of their initiation (Fig. 7 d – f). The largest stamen is the first, and so on. At later stages of development, the difference in size between the stamens gradually disappears. Corolla development is similar to A. platanoides . A. nikoense This species, unlike all those described above, has an unusual (for maples) floral groundplan. There are three flowers in the inflorescence: two lateral and a terminal one. Lateral flowers usually consist of five sepals, one of which occupies the median abaxial position, five petals, ten stamens (five opposite sepals, five opposite petals), and two carpels located in the transversal plane (Fig. 8 c, d). The terminal flower is usually hexamerous and consists of six sepals, six petals, twelve stamens, and two carpels. Below we provide a description of lateral pentamerous flowers. Two large transversal sepals are first to appear (Fig. 8 a). Next, the median and transverse-adaxial sepals are formed (Fig. 8 b, c). The initiation of one of the transverse-adaxial sepals is sometimes delayed (Fig. 8 c). There is a size difference between the sepals in the early stages: the transversal sepals are larger, the median abaxial sepal is of medium size, and the transverse-adaxial sepals are the smallest (Fig. 8 b). Even before the formation of the last sepal, petals and stamens appear (Fig. 8 b). Single primordia of stamens opposite the sepals and three upper common petal-stamen primordia are formed almost simultaneously (Fig. 8 b). Transversal-abaxial common primordia arise later (Fig. 8 c). Antisepalous stamens are larger than antipetalous ones and are located closer to the center of the flower, which indicates obdiplostemony (Fig. 8 d). At this stage, two whorls of the obdiplostemonous androecium are clearly distinguishable. In some cases, there are two stamens located opposite one perianth element (Fig. 8 e, f). Moreover, in cases where these are antipetalous stamens, apparently, they both are initiated from the same common petal-stamen primordium (Fig. 8 e). Sometimes, there is an increase in the flower: the lateral flowers are hexamerous (Fig. 8 e) and the terminal flowers are heptamerous. The further development of stamens and petals is similar to that of other species. The petals flatten after their appearance and enlarge shortly before anthesis. In the mature flower, antipetalous stamens possess short filaments and are located closer to the flower periphery (Fig. 9 c). The petal bases envelope the bases of the stamen filaments. Antisepalous stamens have long filaments and are located closer to the center of the flower (Fig. 9 c). Floral vasculature In A. tataricum subsp. ginnala and A. nikoense stamen bundles are not associated with the perianth bundles (Fig. 9 a, c). Bundles of sepals and petals are connected. Branches from adjacent common bundles can merge into a bundle of one petal. Bundles of stamens separate from the stele independently. A. tegmentosum demonstrates the same vascular pattern as in A. pensylvanicum (Hall 1951 ). There are independent bundles of two sepals (first and second in the initiation spiral), five common bundles of petals and stamens, and three common bundles of sepals and stamens (Fig. 9 b). Angles between adjacent stamens All three species examined ( A. tataricum subsp. ginnala , A. tegmentosum , A. platanoides ) are similar in the distribution of larger and smaller angles between adjacent stamens (Fig. 10 a, c, e), but the exact angle sizes differ between the species. In A. platanoides , the confidence intervals of all angles overlap (Fig. 10 a). In the other two species, the confidence intervals for the smallest and largest angles do not overlap (Fig. 10 c, e). In all species, the greatest angles are between the stamens near the first and third sepals and between the stamens near the second and fourth sepals. Stamens arising from common petal-stamen primordia (orange, light blue, and black in Fig. 1 and Fig. 10 ) form relatively small angles with adjacent stamens. Size and position of stamen primordia We also analyzed the size of the stamen primordia and their position relative to the sepals and petals (for more details on the measurement technique, see Materials and Methods). The results are presented in scatterplots (Fig. 10 b, d, f). A. platanoides and A. tataricum subsp. ginnala show a similar picture. In these species, three groups of stamens can be roughly distinguished. The three stamens (orange, light blue, and black) are relatively small and located closer to the petals than to the sepals. The other two groups of stamens differ only in their positions. Three stamens (yellow, green, and purple) occupy an intermediate position between the sepals and petals, and the remaining two stamens (red and blue) are located closer to the sepals. In A. platanoides , the stamens belonging to the same group are very similar in size (Fig. 10 b), whereas in A. tataricum subsp. ginnala , black and green stamens stand out from their groups in size (Fig. 10 d). These stamens are located opposite the carpels, which probably affects their size. In A. tegmentosum , only one group of stamens can be distinguished (orange, blue, and black), and these stamens are smaller and closer to the petals compared to the same stamens in the other two species (Fig. 10 f). The five remaining stamens differ considerably from each other in size and/or position. Based on a combination of four characters (the presence of a common primordium, the presence of a common vascular bundle, the average relative area of the stamen primordium, the average antipetality index), each stamen can be characterized as belonging to an external or internal whorl, and the androecium as more or less homogeneous (Fig. 11 ). The results show that A. platanoides and A. tataricum subsp. ginnala are characterized by a relatively homogeneous androecium, but it is still possible to distinguish two groups of stamens from different whorls (Fig. 11 a). One group can be interpreted as stamens of the antipetalous whorl, and the second group as stamens of the antisepalous whorl. Although this interpretation is probably more a matter of agreement. A. tegmentosum exhibits a remarkably heterogeneous androecium (Fig. 11 b). Three stamens distinctly belong to the antipetalous whorl, three distinctly belong to the antisepalous whorl, and two stamens occupy an intermediate position between them. As a result, two whorls are clearly distinguished, but it is impossible to draw a clear boundary between them. Floral meristem size For seven species, we measured the surface area of the floral meristem at the time of petal and stamen initiation. Species can be divided into two groups (Fig. 12 ). The first is represented by A. nikoense , A. platanoides , and A. tataricum subsp. ginnala . The area of their meristem ranges from approximately 2500 to 3500 µm 2 . The second group includes A. tegmentosum, A. saccharinum and A. barbinerve . The area of their floral meristem before the initiation of petals and stamens does not exceed 2500 µm 2 . The smallest floral meristem is observed in A. negundo . This species is statistically significantly different from all species except A. barbinerve . P – values of the analysis of variance are presented in the supplement. DISCUSSION Comparison with other Sapindaceae Flower development has now been studied in a number of species of Sapindaceae (for example, Cao et al. 2018 ; Cao and Xia 2009 ; Payer 1857 ; Ronse De Craene et al. 2000; Zhang et al. 2022 ; Zhou et al. 2019 ). However, a lot of data are disputable. For all species studied, a sequential initiation of sepals in a 2/5 spiral has been described ( Cao et al. 2006 , 2017 , 2018 ; Payer 1857 ; Ronse De Craene et al. 2000; Zhang et al. 2022 ; Zhou et al. 2019 ). After calyx initiation, the floral meristem becomes pentagonal in outline, and petals appear. Petals are retarded in their growth and enlarge shortly before flowering. In some species ( Handeliodendron bodinieri (H.Lév.) Rehder, Delavaya toxocarpa Franch., Koelreuteria bipinnata Franch., Xanthoceras sorbifolium Bunge), petal initiation has been described as simultaneous (Cao et al. 2006 , 2018 ; Cao and Xia 2009 ; Zhou et al. 2019 ); in others ( Koelreuteria paniculata Laxm., Eurycorymbus cavaleriei (H.Lév.) Rehder & Hand.-Mazz., Dipteronia sinensis Oliv.), petals are initiated sequentially (Cao et al. 2017 ; Ronse De Craene et al. 2000; Zhang et al. 2022 ). In D. sinensis and E. cavaleriei , the appearance of petals on each side of the fifth sepal overlaps with the initiation of stamens, giving the false impression of common primordia (Cao et al. 2017 ; Zhang et al. 2022 ). In species of the genus Koelreuteria and D. toxocarpa , one petal (between the second and fifth sepals or between the third and fifth sepals) arises from a common primordium with the stamen (Cao et al. 2018 ; Cao and Xia 2009 ; Ronse De Craene et al. 2000). The time of initiation of stamens and petals overlaps. Stamens are formed in rapid succession, so it is often not possible to determine the exact order of their appearance. In Koelreuteria bipinnata var. integrifolia (Merr.) T.C. Chen and Cardiospermum halicacabum L., however, the order of stamen initiation is unidirectional, starting from the stamen located opposite the fourth sepal and ending on the stamen between the third and fifth sepals (Cao et al. 2018 ; Payer 1857 ). There are two aspects of flower development in the family with great evolutionary and taxonomic importance: the presence of common primordia and the changes in flower symmetry during development (Zhang et al., 2022 ). In most of the studied species of Sapindaceae, flowers at the middle developmental stages are obliquely monosymmetrical. This monosymmetry is established by a delay in the development of some petals or their initiation from common primordia. Zhang et al. ( 2022 ) clearly imply that flower morphogenesis in this case follows evolution, as postulated by the biogenetic law (Haeckel 1866 ). Although these ideas are now rejected by zoologists, they are often used by botanists (Ronse De Craene 2018). On this basis, Zhang et al. ( 2022 ) believe that the ancestral flower in Sapindaceae was most likely monosymmetrical with an oblique plane of symmetry. Changes in flower symmetry during development can occur many times and are observed in many angiosperms (Endress 1999 , 2001 , 2012 ). Developmental changes in symmetry are often associated with the pressure of surrounding structures and flower parts on each other (Bull–Hereñu et al. 2022 ; Endress 1999 , 2012 ; Remizowa et al. 2013 , 2023 ). Therefore, we doubt the correctness of using flower symmetry at early stages of development to reconstruct the symmetry of the ancestral flower. Zhang et al. ( 2022 ) apparently do not distinguish between the concepts of zygomorphy and monosymmetry. Therefore, they probably also imply that the ancestral flower was zygomorphic. However, we will differentiate these concepts to avoid confusion. Monosymmetric flowers are flowers through which only one plane of symmetry can be drawn, so all the organs of the flower will be mirror symmetrical relative to a single plane (Endress, 1999 ). A zygomorphic flower is a flower in which the organs are unevenly distributed in space, so that there are two sides (usually the upper and lower) that differ in some external parameters important for attracting or interacting with the pollinator (Bukhari et al. 2017 ; Nuraliev et al. 2019 ; Remizowa et al. 2023 ). If we assume that eight stamens are arranged into two whorls (one whorl with five, the other with three stamens), then the flower of Sapindaceae is automatically monosymmetrical. According to Zhang et al. ( 2022 ), the flowers of Dipteronia and Acer are disymmetrical. The planes of symmetry are set by a syncarpous gynoecium consisting of two carpels. However, if we take into account the position of both the gynoecium and the androecium, then the flowers of both Dipteronia and maples with eight stamens should be considered asymmetrical because the symmetry planes of different whorls do not coincide. The exception are some flowers of A. spicatum , which can be considered monosymmetrical (Fig. 6 c). Before carpel initiation, the stamens determine the plane of floral symmetry. Monosymmetry at this developmental stage depends on the order of stamen initiation. In some species, stamens are formed almost simultaneously. Therefore, the plane of (mono)symmetry can be drawn only based on the initiation of stamens and petals from common or single primordia. In A. platonoides , stamen initiation is unidirectional, so that the plane of monosymmetry is clearly visible (Fig. 2 b – e). In A. tegmentosum , the first four stamens arise in a spiral, the remaining four appear almost simultaneously from common stamen-petal primordia (Fig. 7 b, c), and therefore this pattern of development is immediately asymmetrical. Flower asymmetry in this species is probably also due to non-integer merism (see below). As soon as the gynoecium appears, the flower becomes asymmetrical. The monosymmetry of the maple flower in its early development is not necessarily evidence of ancestral zygomorphy, but it can be considered a natural pattern associated with the structural features of the androecium. The flower of the common ancestor of maples and Dipteronia was indeed monosymmetric and zygomorphic, at least other members of the subfamily Hippocastanoideae are characterized by zygomorphic monosymmetrical flowers (Acevedo-Rodríguez et al. 2010 ; Buerki et al. 2010 ). However, we believe this is a coincidence. The second important feature of flower morphogenesis in many Sapindaceae is the presence of common primordia of petals and stamens. The formation of common primordia in this particular group can be regarded as a manifestation of petal reduction by delay in their development (Cao et al. 2018 ; Ronse De Craene et al. 1993). The development of three and even four stamens from common primordia is a unique feature of the genus Acer and was previously unknown in Sapindaceae. The most plausible explanation for such uniqueness is still an insufficient knowledge of the diversity of floral organogenesis in this family. However, a certain trait can be traced. Most of the previously studied Sapindaceae have a trimerous gynoecium and, therefore, most likely, a larger size of the floral meristem compared to maples, which typically possess two carpels (Ronse De Craene 2016). Use of developmental data for androecium homologies Developmental data are considered one of the most important sources for inferring homology (Hufford 2003 ; Ronse De Craene 2018; Rutishauser and Moline 2005 ; Sattler and Rutishauser 1997 ). The stamens of initially different whorls may have some peculiarities of development, for example, order of initiation or presence or absence of common petal-stamen primordia. The criterion of initiation sequence to differentiate stamens of the two whorls, as suggested by Payer ( 1857 ), is poorly applicable due to the number of variations in the order of stamen initiation in maples. Moreover, we did not identify the pattern described by Payer ( 1857 ) in any of the species examined here. In A. tataricum subsp. ginnala and A. spicatum , all stamens appear almost simultaneously, in A. platonoides the initiation of stamens is unidirectional, and in A. tegmentosum , the first four stamens are initiated in a spiral, and the remaining stamens appear later simultaneously. All the described variants for the order of stamen initiation do not allow for the separation of stamens into whorls in the sense of Payer ( 1857 ). Using data on the presence of common primordia seems more promising. It is believed that a common primordium is formed by organs located in the same radius (Ronse De Craene et al. 1993). Therefore, we can definitely consider stamens, which arise from common primordia, as antipetalous. In most eight-staminate maples, three stamens are formed by common primordia, so these stamens are probably antipetalous, and the remaining five are antisepalous. This interpretation of maple flowers is consistent with the hypothesis previously proposed for Sapindaceae (Ronse De Craene et al. 2000). But in A. tegmentosum four stamens are formed via common primordia. This fact can be interpreted in different ways: as evidence of an independent origin of the eight-staminate androecium or as a secondary change in the pattern of stamen development. An important problem is the difficulty of determining whether common primordia are present or absent. Sometimes the reason why a particular case is described as a common primordium, or vice versa is not described, is not clear. This problem is caused by several reasons: the presence of transitional forms, the short longevity of common primordia, and the dependence of interpretation on the viewing angle (Remizowa et al. 2006 ; Ronse De Craene et al. 1993; Sattler 1996 ). In rare cases, the presence of common primordia varies between flowers of the same species, for example, in some species of the genus Tofieldia (Remizowa et al. 2006 ). In the family Sapindaceae, common primordia have been described only in the most unambiguous cases. Other developmental patterns have been interpreted as a simultaneous initiation of stamens and petals, giving a false impression of the presence of common primordia (Cao et al. 2017 ; Zhang et al. 2022 ). Is it necessary to clearly distinguish between these cases? They demonstrate the same pattern: synchronization of initiation and spatial shift. “False” common primordia may represent a transitional form between common and single primordia. Such ideas bring us to the question of the reasons for the appearance of common primordia and their possible functions. The reasons for the appearance of common primordia are not well understood. Endress ( 1995 ) noted that in monocots, common primordia are a manifestation of the synorganization of floral organs and that they are formed only on a flat flower apex. Ronse De Craene (2018), in relation to eudicots, puts forward slightly different factors for the formation of common primordia: a decrease in the height of the floral meristem and a reduction of petals, expressed via a certain delay in their development. In his opinion, the presence of common primordia indicates a gradual loss of petals in an evolutionary and phylogenetic context (Ronse De Craene 2018, 2024). We believe that in maples, common primordia are not generally associated with petal reduction. Thus, in A. negundo , the petals were presumably lost in evolution much later than the stamens formed from common primordia, which were lost even before the divergence of the sections Arguta and Negundo (Zavialov and Remizowa 2024 ). Loss of the corolla is quite rare in such a morphologically diverse genus (de Jong 1976 ). The idea of the importance of a certain shape and height of the floral meristem for the formation of common primordia is questionable, because, in some monocots, common primordia occur on a convex or even elongated cylindrical receptacle, as in representatives of the genus Eriocaulon (De Lima Silva et al. 2021 ; Sokoloff et al. 2020 ). The criterion of small height of the floral meristem (Ronse De Craene 2018) is not fully consistent with our data. Among maples, common primordia are formed in both species with a flat receptacle (for example, A. platanoides ) and with a convex receptacle ( A. spicatum ). In our opinion, the only satisfactory explanation for the formation of common primordia is a general decrease in the size of the floral meristem, not necessarily associated with the decrease in meristem height. We tested this hypothesis by estimating the surface area of the floral meristem at the time of petal and stamen initiation. The formation of common primordia is associated with a decrease in space on the floral meristem at the time of the formation of stamens (Fig. 12 ). A decrease in the floral meristem can be expressed as a decrease in its area or height and should be revealed by comparison with closely related species without common primordia or with a smaller number of them. Maple species with eight stamens and three common primordia are similar in floral meristem area (about 30,000 µm 2 ), while species with eight stamens and four common primordia ( A. tegmentosum ) and species with fewer stamens and without common primordia ( A. saccharinum, A. barbinerve and A. negundo; for details, see Zavialov and Remizowa 2024 ; Zavialov 2024 ) have meristem areas of about 20,000 µm 2 or less. A decrease in the floral meristem size is a factor of reduction. Common primordia in this case act as a compensatory mechanism, changing morphogenesis, but preserving the definitive structure as in A. tegmentosum . However, reduction of flower organs still occurs when the floral meristem decreases beyond a critical level, as in A. saccharinum, A. barbinerve and A. negundo . In most cases, the loss of common primordia occurs through the reduction of one organ from a pair or both organs arising from a common primordium. On the other hand, the number of common primordia may increase not due to a decrease in the floral meristem but due to an increase in the number of floral organs, as in A. nikoense . Use of positional criteria for androecium homologies The ancestor of Sapindaceae had ten stamens in two whorls, and two stamens have been lost. In theory, if two stamens were lost, the remaining eight should be evenly distributed in space due to shifts (Ronse De Craene et al. 2000). The position of the stamens can be considered both relative to each other and relative to the perianth. If we consider the position of the stamens relative to the sepals and petals, this allows to evaluate the antisepality and antipetality of the stamens. The stamens adjacent to the disappeared ones should change their position more significantly. These stamens should occupy an intermediate position between the petal and sepal. In reality, there are three stamens, the position of which corresponds to this theory. Two of them are adjacent (yellow and green in Fig. 1 opposite the first and third sepals), between them there was one of the lost stamens. It is difficult to identify the position of the second lost stamen in this way. In general, the size of the stamen primordia and their position relative to the perianth allows to divide the stamens into three groups. The first group of stamens is closer to the antipetalous position (stamens on the sides of the fifth sepal and the stamen between the first and fourth sepals: black, orange, and light blue in Fig. 1 ), the second group is closer to the antisepalous position (approximately in the radii of the fourth and fifth sepals: red and blue in Fig. 1 ), the third group is in an intermediate position (stamens opposite the first, second, and third sepals: yellow, green, purple in the Fig. 1 ). In fact, all stamens are shifted to some degree relative to the median of the sepals and petals. In this regard, the use of the terms “antisepalous” and “antipetalous” to describe the androecium of maples and probably Sapindaceae in general is not entirely correct, although it is widely used. The description of the stamen position is often based not on the actual observed pattern, but rather on the desire to fit a theoretical model. The existing hypothesis about the origin of the androecium in Sapindaceae (Ronse De Craene et al. 2000) is based on similar approximations, which are not entirely consistent with the observed flower construction, although this theory gives a plausible explanation of the androecium evolution in Sapindaceae. Another way to understand the position of the stamens is to compare their positions relative to each other. Presumably, if two stamens are lost, the remaining ones are redistributed more or less equally, so the angles between adjacent stamens should be the same. In reality, there are some larger angles, that could mark the positions of the lost stamens. These angles should be measured only before the gynoecium initiation because carpel growth causes secondary shifts of the stamens, which were initially located in carpel radii. This is clearly visible in A. tataricum subsp. ginnala , because this species is characterized by the relatively early initiation of the gynoecium. However, the interpretation of the androecium of A. tegmentosum is not so obvious. The position of the stamens relative to the perianth indicates the presence of distinctly antipetalous and antisepalous stamens, which is not typical for other studied species with eight stamens. The remaining stamens (green and purple in Fig. 1 ) form a positional gradient between the two extreme states. In other species, these stamens formed their own group with an intermediate position. The position of the stamens relative to each other shows the fundamental similarity of the androecium of this and other species. The stamen positions can also be characterized using vasculature. In most species, the vascular bundles of all eight stamens depart from the receptacle at the same level and are not associated with the perianth, so the vasculature does not add information about the homology of stamens. However, in A. tegmentosum , and apparently in other species of the section Macrantha (Hall 1951 ), the vascular bundles of five stamens are associated with bundles of sepals, and the bundles of three stamens with bundles of petals. Such vasculature is not consistent with either developmental or positional data. Androecium homologies in Acer tegmentosum Different criteria provide different results for Acer tegmentosum . The problem of homology in cases of conflicting interpretations can be considered from different points of view. Using approaches of classical morphology, we believe that the stamens of different whorls have some discrete identity (Arber 1946 ; Timonin 2002 ). The degree of the differences between the stamens of two whorls may change, the whorls may be combined into one, but the stamen identity as belonging to a certain whorl remains. With this approach, it is essential to select the criterion of homology that would show the stamen identity in the most accurate way. If we follow this way of thinking, then it is logical to assume that the most suitable criterion is the position of the stamens relative to each other. This criterion allows not to separate A. tegmentosum from other maples. The assumption that the androecium of A. tegmentosum has the same origin as in all other Sapindaceae is also consistent with the generally accepted principle of Occam's razor (Walsh 1979 ; McFadden 2023 ). Also, the observed position of the stamens is probably non-optimal due to being unevenly spaced, and therefore can be regarded as a special feature most likely inherited from an ancestor (Timonin 1993 ). The developmental and vascular features of this species can be explained by a decrease in the floral meristem while maintaining the definitive floral structure. The same problem can be resolved using Sattler’s dynamic morphology (Sattler 1992 , 1996 ). This approach avoids discrete homology, and considers each case as a combination of structural features and patterns of morphogenesis. According to such ideas, the classical plant organs are nothing more than the most common combinations of characters, between which there is a continuum of other rare intermediate variants (Sattler 1996 ; Sattler and Rutishauser 2022 ). In terms of dynamic morphology, each stamen in a flower could hypothetically possess a unique set of characters. In most cases, this uniqueness is not observed, and all stamens of one whorl are uniform or discretely unequal, with the formation of several types of stamens, for example, in Melastomataceae (Melo et al. 2022 ) or in Cassiinae (Fabaceae) (Tucker 1996 ). The stamens of different whorls usually differ discretely from each other, at least by the level of their insertion. In A. tegmentosum , different stamens have different combinations of characters, usually distinguishing two whorls. Separation of whorls in the androecium is possible in this case, but the boundary between the two whorls is blurred. In other species, the difference between the whorls is even less pronounced. Followers of dynamic morphology often present their concept as a morphogenetic space with more and less stable variants (Jeune and Sattler 1992 ). In relation to the androecium of Acer , stable variants would be haplostemony and diplostemony, in both cases with a whorl non-isomerous to the perianth. The androecium of maples lies between these two extremes. In most species studied here, it is closer to a single-whorled variant, whereas in A. tegmentosum , it is closer to a two-whorled one. Based on the observed features, we propose to use the terms “homogeneous” and “heterogeneous” androecium to characterize the degree of separation of stamens into whorls. The androecium of A. tegmentosum can be regarded as having non-integer merism. Non-integer merism is uncommon and has been described for the perianth or androecium (Choob and Yurtseva 2007 ; Rudall et al. 2005 ;Vislobokov et al. 2014 ). These types of flower merism are considered transitional between typical cases (Choob 2010 ). In this sense, the androecium of all maples with eight stamens can be regarded as a transitional condition evolving towards haplostemony. But only A. tegmentosum has stamens, which cannot be confidently attributed to any whorl. Heterogeneity of the androecium in A. tegmentosum is probably associated with a decrease in the area of the floral meristem while maintaining the number of flower organs. Concluding remarks on androecium homologies in 8-staminate species To summarize, we can conclude that maples have lost two antipetalous stamens: one was located opposite the petal between the first and third sepals, and the second was opposite the petal between the second and fourth sepals. The same stamens are also absent in other Sapindaceae (Ronse de Craene et al. 2000). It is important to note that the positions of the lost stamens are not opposite or near the carpel backs, but rather opposite the septum of the ovary. Therefore, it should be assumed that the gynoecium did not directly influence the reduction of these stamens in maples. Ideas by A.W. Eichler ( 1878 ) turned out to be wrong. Why have stamens been lost in these particular positions? A common answer to this question is certain mechanical forces during flower development (i.e. pressure of the perianth (Brockington et al. 2013 ; Ronse De Craene 2024)) and the influence of gynoecium prepatterning (Choob and Penin 2004 ). Therefore, organs located simultaneously opposite the perianth elements, which are initiated earlier and opposite the carpels are the candidates to be lost (Brockington et al. 2013 ; Ronse De Craene and Wei 2019; Ronse De Craene 2024). Zhang et al. ( 2022 ) suggested that the order of petal initiation inversely correlates with the degree of stamen reduction in Sapindaceae. There are no stamens opposite those petals that usually appear first. Petals that are initiated later than the others tend to emerge from common primordia. According to the ideas of Ronse De Craene (2018), such developmental behavior is an indicator of petal reduction. This point of view is in accordance with the flower development of A. tataricum subsp. ginnala and A. spicatum , but the development of A. platanoides contradicts such ideas. In A. platanoides , the petal, which is formed from a common petal-stamen primordium (between the first and fourth sepals), is one of the first to appear. This common primordium divides into stamen and petal parts even before the initiation of all petals and stamens. Obdiplostemony in A. nikoense In A. nikoense , the androecium is two-whorled, obdiplostemonous, with five stamens in each whorl. Antisepalous stamens appear first in development, antipetalous stamens arise later from common primordia. The stamen initiation is influenced by the flower-subtending bract; therefore, the adaxial organs are formed earlier than the abaxial ones. During further development, the antipetalous stamens are pushed more and more towards the periphery of the flower. This pattern of androecium development indicates secondary obdiplostemony type 1, according to the classification of Ronse De Craene and Bull-Hereñu (2016). Among rosids and, in particular, in the order Sapindales, this type of obdiplostemony is quite common (Ronse De Craene and Bull-Hereñu 2016; Ronse De Craene and Smets 1995). Obdiplostemony is considered a manifestation of certain trends in flower evolution. On the basis of obdiplostemony, synorganization without organ fusion can occur, for example, in Geranium robertianum L. (Endress 2010 ). However, in most cases, it is considered a consequence of the reduction of antipetalous stamens. According to such ideas, secondary obdiplostemony is a transitional state from diplostemony to haplostemony (Ronse De Craene and Bull-Hereñu 2016; Ronse De Craene and Smets 1995). The case of A. nikoense is interesting because the opposite direction of structural transformations can be proposed here. It is even more intriguing because two-whorled androecium is isomerous to the perianth. The androecium of the typical structure for most maples is rather single whorled, at least superficially in the mature flowers. In A. nikoense , two distinct whorls are formed, but in a configuration indicating a lack of space on the floral meristem. This spatial constraint is also evidenced by the initiation of all antipetalous stamens from common primordia. A. nikoense may also be important in understanding the origin of the typical flower of Sapindaceae. The case of A. nikoense can be seen as a re-gain of isomery or as a reversion to an ancestral state for the entire family. This is especially interesting because the androecium of the Biebersteiniaceae, the closest relatives of the Sapindaeaceae, is presumably obdiplostemonous (Joyce et al. 2023 ; Yamamoto et al. 2014 ). CONCLUSIONS The structure and development of typical maple flowers do not show strong differences from the other Sapindaceae. The sepals are initiated in a spiral, and the petals are delayed in their development. Species of the genus Acer demonstrate different patterns of stamen initiation, even those with eight stamens. The most important aspect of morphogenetic evolution in the genus Acer is the change in the number of common petal-stamen primordia compared to other Sapindaceae. These structures affect floral morphogenesis, but not the construction of the mature flower. The topological identity of individual stamens and the entire androecium can vary significantly in different evolutionary lineages of the genus. Our study raises many questions about the criteria for delineating whorls of the androecium, the reasons for the appearance and functions of common primordia, and the limitations of developmental data as a criterion of homology. The hypotheses (related to the questions listed above) we propose in this article should be tested in the future on a wider sampling of representatives of the family Sapindaceae in particular and angiosperms in general. Declarations The authors have no competing interests to declare that are relevant to the content of this article. FUNDING The research was supported by a budgetary subsidy to the Lomonosov Moscow State University (No. 121032500084-6). AUTHOR CONTRIBUTIONS Conceptualization: A.E. Zavialov, M.V. Remizowa; Material preparation, data collection and analysis: A.E. Zavialov, M.V. Remizowa; Writing - original draft preparation: A.E. Zavialov; Writing - review and editing: M.V. Remizowa; Supervision: M.V. Remizowa. ACKNOWLEDGEMENTS We are grateful to to G.A. Boyko for providing specimens of A. nikoense from the Botanical Garden of Moscow State University, to G.A. Firsov for providing specimens of A. spicatum and A. tegmentosum from the Saint Petersburg Botanical Garden, to T.E. Kramina, A.D. Lisitsina, A.O. Astashkin for help in collecting material, to A.C. Timonin for helpful discussions and suggestions. SEM studies were carried out at the Shared Research Facility “Electron microscopy in life sciences” at Moscow State University (Unique Equipment “Three-dimensional electron microscopy and spectroscopy”). References Acevedo-Rodríguez P, Van Welzen PC, Adema F, Van Der Ham RWJM (2010) Sapindaceae. In: Kubitzki K (ed) Flowering plants. Eudicots: Sapindales, cucurbitales, myrtaceae. Springer, Berlin, pp 357–407. APG IV (2016) An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG IV. Bot J Linn Soc 181:1–20. Arber A (1946) Goethe’s botany. Chronica botanica 10:67–124. Areces-Berazain F, Hinsinger DD, Strijk JS (2021) Genome-wide supermatrix analyses of maples (Acer, Sapindaceae) reveal recurring inter-continental migration, mass extinction, and rapid lineage divergence. Genomics 113:681–692. Barykina RP, Veselova TD, Deviatov AG, Djalilova HH, Iljina GM, Chubatova NV (2004) Handbook of the botanical microtechniques. Moscow University Press, Moscow. Brockington S, Dos Santos P, Glover B, De Craene LR (2013) Androecial evolution in Caryophyllales in light of a paraphyletic Molluginaceae. American J of Botany 100:1757–1778. Buchenau F (1861) Morphologische Bemerkungen über einige Acerineen. Botanische Zeitung 19:265–269. Buerki S, Callmander MW, Acevedo‐Rodriguez P, et al. (2021) An updated infra‐familial classification of Sapindaceae based on targeted enrichment data. American J of Botany 108:1234–1251. Buerki S, Lowry Ii PP, Alvarez N, Razafimandimbison SG, Küpfer P, Callmander MW (2010) Phylogeny and circumscription of Sapindaceae revisited: molecular sequence data, morphology and biogeography support recognition of a new family, Xanthoceraceae. Plecevo 143:148–159. Bukhari G, Zhang J, Stevens PF, Zhang W (2017) Evolution of the process underlying floral zygomorphy development in pentapetalous angiosperms. American J of Botany 104:1846–1856. Bull–Hereñu K, Dos Santos P, Toni JFG, et al. (2022) Mechanical Forces in Floral Development. Plants 11:661. Cao L, Liu J, Lin Q, Ronse De Craene LP (2018) The floral organogenesis of Koelreuteria bipinnata and its variety K. bipinnata var. integrifolia (Sapindaceae): evidence of floral constraints on the evolution of monosymmetry. Plant Syst Evol 304:923–935. Cao LM, Ronse De Craene LP, Wang Z-X, Wang Y-H (2017) The floral organogenesis of Eurycorymbus cavaleriei (Sapindaceae) and its systematic implications. Phytotaxa 297:234. Cao LM, Xia N (2009) Floral organogenesis of Delavaya toxocarpa (Sapindaceae; Sapindales). J of Sytematics Evolution 47:237–244. Cao LM, Xia N-H, Deng Y-F (2006) Floral organogenesis of Handeliodendron bodinieri (Sapindaceae) and its systematic implications. Acta Phytotaxonomica Sinica 44:393. Choob VV (2010) The role of positional information in regulation of development of floral organs and leaf series of shoots. Binom, Moscow. Choob VV, Penin AA (2004) Structure of flower in Arabidopsis thaliana: spatial pattern formation. Russian Journal of Developmental Biology 35:224–227. Choob VV, Yurtseva OV (2007) Mathematical modeling of flower structure in the family Polygonaceae. Bot Zhurn 92:114–134. De Jong PC (1976) Flowering and sex expression in Acer L.: a biosystematic study. Wageningen: H. Veenman & Zonen B.V. De Jong PC (2004) World maple diversity. Westonbirt Arboretum and Royal Agricultural College in Gloucestershire, England, pp 2–11. De Lima Silva A, Trovó M, Stützel T, Rudall PJ, Sajo MDG, Coan AI (2021) Floral development and vasculature in Eriocaulon (Eriocaulaceae) provide insights into the evolution of Poales. Annals of Botany 128:605–626. Eichler AW (1878) Blüthendiagramme. Verlag von Wilhelm Engelmann, Leipzig. Endress PK (1995) Major evolutionary traits of monocot flowers. Monocotyledons: systematics and evolution. Royal Botanic Gardens, London, pp 43–79. Endress PK (1999) Symmetry in Flowers: Diversity and Evolution. International Journal of Plant Sciences 160:S3–S23. Endress PK (2001) Evolution of floral symmetry. Current Opinion in Plant Biology 4:86–91. Endress PK (2010) Synorganisation without organ fusion in the flowers of Geranium robertianum (Geraniaceae) and its not so trivial obdiplostemony. Annals of Botany 106:687–695. Endress PK (2012) The Immense Diversity of Floral Monosymmetry and Asymmetry Across Angiosperms. Bot Rev 78:345–397. Haeckel EA (1866) Generelle Morphologie der Organismen: Allgemeine Grundzüge der Organischen Formen-Wissenschaft, Mechanisch Begründet Durch die von Charles Darwin Reformirte Descendenz-Theorie. Druck und Verlag von Georg Reimer, Berlin. Hall BA (1951) Tht floral anatomy of the genus Acer. American J of Botany 38:793–799. Harrington MG, Edwards KJ, Johnson SA, Chase MW, Gadek PA (2005) Phylogenetic Inference in Sapindaceae sensu lato Using Plastid matK and rbcL DNA Sequences. Systematic Botany 30:366–382. Hufford L (2003) Homology and Developmental Transformation: Models for the Origins of the Staminodes of Loasaceae Subfamily Loasoideae. International Journal of Plant Sciences 164:S409–S439. Jeune B, Sattler R (1992) Multivariate analysis in process morphology of plants. Journal of Theoretical Biology 156:147–167. Joyce EM, Appelhans MS, Buerki S, et al. (2023) Phylogenomic analyses of Sapindales support new family relationships, rapid Mid-Cretaceous Hothouse diversification, and heterogeneous histories of gene duplication. Front Plant Sci 14:1063174. Li J, Stukel M, Bussies P, et al. (2019) Maple phylogeny and biogeography inferred from phylogenomic data. J of Sytematics Evolution 57:594–606. McFadden J (2023) Razor sharp: The role of Occam’s razor in science. Annals of the New York Academy of Sciences 1530:8–17. Melo LRF, Vasconcelos TNC, Caetano APS, Brito VLG de (2022) Stamen Diversity in Melastomataceae. Morphology, Color, and Function. Systematics, Evolution, and Ecology of Melastomataceae. Springer International Publishing, Cham, 609–628. Nuraliev MS, Sokoloff DD, Karpunina PV, Oskolski AA (2019) Patterns of Diversity of Floral Symmetry in Angiosperms: A Case Study of the Order Apiales. Symmetry 11:473. Payer J-B (1857) Traité de organogénie comparée de la fleur. V. Masson, Paris. Remizowa MV, Rudall PJ, Choob VV, Sokoloff DD (2013) Racemose inflorescences of monocots: structural and morphogenetic interaction at the flower/inflorescence level. Annals of Botany 112:1553–1566. Remizowa MV, Shipunov AB, Sokoloff DD (2023) When asymmetry mimics zygomorphy: flower development in Chamaelirium japonicum (Melanthiaceae, Liliales). Botanica Pacifica 12:3–14. Remizowa M, Sokoloff D, Rudall P (2006) Patterns of Floral Structure and Orientation in Japonolirion , Narthecium, and Tofieldia. aliso 22:159–171. Ronse De Craene L (2016) Meristic changes in flowering plants: How flowers play with numbers. Flora 221:22–37. Ronse De Craene L (2018) Understanding the role of floral development in the evolution of angiosperm flowers: clarifications from a historical and physico-dynamic perspective. J Plant Res 131:367–393. Ronse De Craene L P (2022) Floral diagrams: an aid to understanding flower morphology and evolution Second Edition. Cambridge University Press, New York. Ronse De Craene LP (2024) The interaction between heterochrony and mechanical forces as main driver of floral evolution. J Plant Res. doi:10.1007/s10265-024-01526-3. Ronse De Craene L, Bull-Hereñu K (2016) Obdiplostemony: the occurrence of a transitional stage linking robust flower configurations. Ann Bot 117:709–724. Ronse De Craene L, Clinckemaillie D, Smets E (1993) Stamen-Petal Complexes in Magnoliatae. Bulletin du Jardin botanique national de Belgique / Bulletin van de National Plantentuin van België 62:97–112. Ronse De Craene LP, Smets EF (1993) Dédoublement revisited: towards a renewed interpretation of the androecium of the Magnoliophytina. Botanical Journal of the Linnean Society 113:103–124. Ronse De Craene LP, Smets EF (1995) The distribution and systematic relevance of the androecial character oligomery. Botanical Journal of the Linnean Society 118:193–247. Ronse De Craene L, Smets E, Clinckemaillie D (2000) Floral ontogeny and anatomy in Koelreuteria with special emphasis on monosymmetry and septal cavities. Pl Syst Evol 223:91–107. Ronse De Craene LP, Wei L (2019) Floral development and anatomy of Macarthuria australis (Macarthuriaceae): key to understanding the unusual initiation sequence of Caryophyllales. Aust Syst Bot 32:49–60. Rudall PJ, Cunniff J, Wilkin P, Caddick LR (2005) Evolution of dimery, pentamery and the monocarpellary condition in the monocot family Stemonaceae (Pandanales). TAXON 54:701–711. Rutishauser R, Moline P (2005) Evo-devo and the search for homology (“sameness”) in biological systems. Theory Biosci 124:213–241. Sattler R (1992) Process morphology: structural dynamics in development and evolution. Can J Bot 70:708–714. Sattler R (1996) Classical Morphology and Continuum Morphology: Opposition and Continuum. Annals of Botany 78:577–581. Sattler R, Rutishauser R (1997) The Fundamental Relevance of Morphology and Morphogenesis to Plant Research. Annals of Botany 80:571–582. Sattler R, Rutishauser R (2022) Fundamentals of Plant Morphology and Plant Evo-Devo (Evolutionary Developmental Morphology). Plants 12:118. Sokoloff DD, Yadav SR, Chandore AN, Remizowa MV (2020) Stability Despite Reduction: Flower Structure, Patterns of Receptacle Elongation and Organ Fusion in Eriocaulon (Eriocaulaceae: Poales). Plants 9:14–24. Timonin AC (1993) Why we are inclined to reveal mainly reductions in evolutionary morphological reconstructions. Zhurnal obshchei biologii 54:104–114. Timonin AC (2002) Sattler’s dynamic morphology: an acme or a reverie. Wulfenia 9:9–18. Tucker SC (1996) Trends in evolution of floral ontogeny in Cassia sensu stricto, Senna , and Chamaecrista (Leguminosae: Caesalpinioideae: Cassieae: Cassiinae); a study in convergence. American J of Botany 83:687–711. Vislobokov N, Sokoloff D, Degtjareva G, Valiejo-Roman C, Kuznetsov A, Nuraliev MS (2014) Aspidistra paucitepala (Asparagaceae), a new species with occurrence of the lowest tepal number in flowers of Asparagales. Phytotaxa 161:270–282. Walsh D (1979) Occam’s Razor: A Principle of Intellectual Elegance. American Philosophical Quarterly 16:241–244. Yamamoto T, Vassiliades DD, Tobe H (2014) Embryology of Biebersteinia (Biebersteiniaceae, Sapindales): characteristics and comparisons with related families. J Plant Res 127:599–615. Zavialov AE (2024) Patterns of flower structure and development in some species of the genus Acer (Sapindaceae). Bachelor dissertation, Moscow State University. Zavialov AE, Remizowa MV (2024) Unstable patterns of flower structure and development in the almost dioecious species Acer negundo (Sapindaceae, Sapindales). Botanical Journal of the Linnean Society 204:267–281. Zhang Y, Cao L, Ronse De Craene LP (2022) Inflorescence and floral development of Dipteronia sinensis (Sapindaceae): Demonstration of developmental constraints in the shaping of the flower. Flora 291:152069. Zhou Q, Cai Q, Zheng Y, Wu Z, Mao J (2019) Floral development and the formation of functionally unisexual flowers in Xanthoceras sorbifolium (Sapindaceae), a morphologically andromonoecious tree endemic to northern China. Trees 33:1571–1582. Supplementary Files supplementaryacerandroecium.xlsx Cite Share Download PDF Status: Published Journal Publication published 25 Apr, 2025 Read the published version in Journal of Plant Research → Version 1 posted Reviewers agreed at journal 22 Jul, 2024 Reviewers invited by journal 20 Jul, 2024 Editor assigned by journal 18 Jul, 2024 First submitted to journal 17 Jul, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4754778","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":329464295,"identity":"1c572bee-3cbe-4b9a-841c-da6624407a29","order_by":0,"name":"Alexander Zavialov","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+UlEQVRIiWNgGAWjYDACdmRGQgWQYGZuwK+FGZnx4QyIwUiCFsaZbSAWAS38zczHJH/UHE7sBzKkeefVRvO3A7X8qNiGU4vEYbZkA4ljhxNnHGZLk+bddjx3xmHGBsaeM7dxW3OYx/CBAVtaYsNhHmNj3m3HchuAWpgZ23BrkT/M/+FAwr+0xPlgLXOO5c4npMXgMA/jg4NtNokbgNY9nNlQk7uBkBbDw2zGho19NsYbD7MlPvhw7EDuRqCWg/j8Ine8+Znkj28SsvOONx84kFBTlzvv/OGDD35U4PE+FDg2QEMDTB4gqB4I7KF0HTGKR8EoGAWjYIQBAFMGXNPo4Z8KAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0007-2280-4324","institution":"Moscow State University: Moskovskij gosudarstvennyj universitet imeni M V Lomonosova","correspondingAuthor":true,"prefix":"","firstName":"Alexander","middleName":"","lastName":"Zavialov","suffix":""},{"id":329464296,"identity":"761385e8-f4eb-4a00-b8b9-a5ddf9cc12db","order_by":1,"name":"Margarita V. Remizowa","email":"","orcid":"","institution":"Moscow State University: Moskovskij gosudarstvennyj universitet imeni M V Lomonosova","correspondingAuthor":false,"prefix":"","firstName":"Margarita","middleName":"V.","lastName":"Remizowa","suffix":""}],"badges":[],"createdAt":"2024-07-17 08:39:46","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4754778/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4754778/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10265-025-01641-9","type":"published","date":"2025-04-25T15:58:10+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":62438664,"identity":"ce90fe37-e089-4b49-9bb8-606c17534a1b","added_by":"auto","created_at":"2024-08-14 08:21:11","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":149659,"visible":true,"origin":"","legend":"\u003cp\u003eFloral diagram with some topological terms and color/numeric code of stamens\u003c/p\u003e","description":"","filename":"floatimage1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/6e05eb5bdfb19500d30ea088.jpg"},{"id":62438663,"identity":"c6aaf710-eed6-44be-b3d2-e7533af8015f","added_by":"auto","created_at":"2024-08-14 08:21:10","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1398083,"visible":true,"origin":"","legend":"\u003cp\u003eEarly stages of flower development in \u003cem\u003eA. platanoides\u003c/em\u003e. a – sepal initiation. b-d – initiation of corolla and androecium. e -f – sequential division of common petal-stamen primordia. Abbreviations: p – petals, I – stamens from individual primordia, II – stamens from common primordia, * – petal-stamen common primordia, s – sepals (numbers indicate the order of initiation). The white dotted line indicates the androecium symmetry plane, and the black dotted line indicates the gynoecium symmetry plane. Scale bars: 50 μm (b – e), 100 μm (a, f)\u003c/p\u003e","description":"","filename":"floatimage2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/3d4b6d117e68453ebda4716c.jpg"},{"id":62438138,"identity":"f29bb843-a032-400d-bcfc-a0d7289113a2","added_by":"auto","created_at":"2024-08-14 08:13:11","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":726159,"visible":true,"origin":"","legend":"\u003cp\u003eLate stages of flower development in \u003cem\u003eA. platanoides\u003c/em\u003e. a – appearance of androecium and corolla in the terminal flower. b – enlargement of stamen primordia. c – thecae formation. d – stamens with fully formed anthers, petals are delayed in development. Abbreviations: p – petals, I – stamens from individual primordia, II – stamens from common primordia, * – petal-stamen common primordia, s – sepals (numbers indicate the order of initiation). Scale bars: 50 μm (a), 100 μm (b – d)\u003c/p\u003e","description":"","filename":"floatimage3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/fe200212af70f8fa40dceca9.jpg"},{"id":62438132,"identity":"5d7273b4-1a65-4a63-bec2-5d5465fdc476","added_by":"auto","created_at":"2024-08-14 08:13:10","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":999466,"visible":true,"origin":"","legend":"\u003cp\u003eFlower development in \u003cem\u003eA. tataricum \u003c/em\u003esubsp. \u003cem\u003eginnala. \u003c/em\u003ea – sepal initiation. b – initiation of two first petals. c – petal and stamen initiation. d, e – sequential division of common petal-stamen primordia. f – appearance of the gynoecium, stamen primordia, located opposite carpel backs, are delayed in development. g – petals and stamens are fully formed. Abbreviations: br – flower-subtending bract, bl – bracteoles, p – petals, I – stamens from individual primordia, II – stamens from common primordia, * – petal-stamen common primordia, s – sepals (numbers indicate the order of initiation). The white dotted line indicates the androecium symmetry plane , and the black dotted line indicates the gynoecium symmetry plane . Scale bars: 50 μm (a – c), 100 μm (d – g)\u003c/p\u003e","description":"","filename":"floatimage4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/fe2f689d3000b8032fb13df0.jpg"},{"id":62439275,"identity":"23dd5125-e8c3-4321-a1c6-a567ab7308d1","added_by":"auto","created_at":"2024-08-14 08:29:11","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1133464,"visible":true,"origin":"","legend":"\u003cp\u003eEarly stages of flower development in \u003cem\u003eA. spicatum\u003c/em\u003e. a – sepal initiation. b – emergence of the first petal. c, d – stamen initiation and division of one common primordium, two views of the same flower. e, f – slightly later stage, division of two common primordia, two views of the same flower. Abbreviations: p – petals, I – stamens from individual primordia, II – stamens from common primordia, * – petal-stamen common primordia, s – sepals (numbers indicate the order of initiation). The white dotted line indicates the androecium plane of symmetry. Scale bars: 30 μm (a – f)\u003c/p\u003e","description":"","filename":"floatimage5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/a78123a8cbb560185ead74d6.jpg"},{"id":62438666,"identity":"b8010b91-655f-437a-b3f1-e5772e52e77e","added_by":"auto","created_at":"2024-08-14 08:21:11","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":787628,"visible":true,"origin":"","legend":"\u003cp\u003eLate stages of flower development in \u003cem\u003eA. spicatum\u003c/em\u003e. a – division of common primordia, side view. b, c – gynoecium initiation, stamen primordia are unequal in size. d – thecae formation. Abbreviations: p – petals, I – stamens from individual primordia, II – stamens from common primordia, * – petal-stamen common primordia, s – sepals (numbers indicate the order of initiation). The white dotted line indicates the symmetry plane of androecium and gynoecium. Scale bars: 50 μm (a), 100 μm (b – d)\u003c/p\u003e","description":"","filename":"floatimage6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/9be89022874baa1584814411.jpg"},{"id":62438134,"identity":"4fc07314-8f17-42f3-b54f-b51e2ffaf212","added_by":"auto","created_at":"2024-08-14 08:13:10","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":1257209,"visible":true,"origin":"","legend":"\u003cp\u003eFlower development of \u003cem\u003eA. tegmentosum.\u003c/em\u003e a – sepal initiation. b – initiation of stamens and petals. c, d – division of common primordia. e – growth of stamens, the difference in size remains. f – beginning of thecae formation. Abbreviations: br – flower-subtending bract, p – petals, I – stamens from individual primordia, II – stamens from common primordia, * – petal-stamen common primordia, s – sepals (numbers indicate the order of initiation). Scale bars: 50 μm (a, d, e), 100 μm (b, c), 200 μm (f)\u003c/p\u003e","description":"","filename":"floatimage7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/83fffae8cd07d16e707126d0.jpg"},{"id":62438667,"identity":"3a07071f-c486-4303-96b6-b9ee10ffa6fd","added_by":"auto","created_at":"2024-08-14 08:21:11","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":1019678,"visible":true,"origin":"","legend":"\u003cp\u003eFlower development of \u003cem\u003eA. nikoense\u003c/em\u003e. a – initiation of transversal sepals. b – initiation of stamens and petals. c – division of adaxial common primordia. d – growth of stamens, obdiplostemony\u003cstrong\u003e \u003c/strong\u003ebecomes visible. e – hexamerous flower with additional stamens, two stamens are located opposite a petal. f – two stamens are located opposite a petal and two stamens are located opposite a sepal. Abbreviations: br – flower-subtending bract, p – petals, I – stamens from single primordia, II – stamens from common primordia, * – petal-stamen common primordia, ts – transversal sepals, ms –transversal-abaxial and transversal-adaxial sepals. Scale bars: 30 μm (c), 50 μm (a, b, e, f), 100 μm (d)\u003c/p\u003e","description":"","filename":"floatimage8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/6a05a363f37f9cd591bd17d1.jpg"},{"id":62438144,"identity":"2f29b0b6-caba-4a14-acdd-cbddf57b14ee","added_by":"auto","created_at":"2024-08-14 08:13:11","extension":"jpg","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":931542,"visible":true,"origin":"","legend":"\u003cp\u003eVasculature of perianth and androecium. a – \u003cem\u003eA. tataricum \u003c/em\u003esubsp\u003cem\u003e. ginnala\u003c/em\u003e (first vertical row). b – \u003cem\u003eA. tegmentosum\u003c/em\u003e (second vertical row). c – \u003cem\u003eA. nikoense\u003c/em\u003e (third vertical row) In \u003cem\u003eA. tataricum \u003c/em\u003esubsp\u003cem\u003e. ginnala \u003c/em\u003eand \u003cem\u003eA. nikoense \u003c/em\u003ethere are common bundles of sepals and petals, bundles of stamens depart separately from the stele. In \u003cem\u003eA. tegmentosum\u003c/em\u003e,\u003cem\u003e \u003c/em\u003etwo bundles of sepals depart separately from the stele, bundles of the remaining perianth elements are common with the bundles of stamens\u003c/p\u003e","description":"","filename":"floatimage9.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/38f63c142036d05ec4dbb37c.jpg"},{"id":62438145,"identity":"262f7fbf-c831-4c0d-b3c9-dc5cf651f0cd","added_by":"auto","created_at":"2024-08-14 08:13:11","extension":"jpg","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":395971,"visible":true,"origin":"","legend":"\u003cp\u003eMeasurements of stamen primordia. a, c, e – the angles between adjacent stamens. b, d, f – graphs of the average antipetality index and the average relative area of the stamen primordium. a, b –\u003cem\u003e A. platanoides\u003c/em\u003e. c, d –\u003cem\u003e A. tataricum \u003c/em\u003esubsp\u003cem\u003e. ginnala\u003c/em\u003e. e, f –\u003cem\u003e A. tegmentosum\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage10.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/1e9967eacad0572c829c0eeb.jpg"},{"id":62439280,"identity":"4b0fb52b-279f-47de-90f1-bce43abdbfbf","added_by":"auto","created_at":"2024-08-14 08:29:11","extension":"jpg","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":228003,"visible":true,"origin":"","legend":"\u003cp\u003eThe degree of division of the androecium into whorls. a – \u003cem\u003eA. platanoides,\u003c/em\u003e \u003cem\u003eA. tataricum \u003c/em\u003esubsp\u003cem\u003e. ginnala\u003c/em\u003e. b – \u003cem\u003eA. tegmentosum. \u003c/em\u003ea – relatively homogenous androecium, three stamens (2, 5, 8) can be attributed to the whorl of antipetalous stamens, and five stamens (1, 3, 4, 6, 7) can be attributed to the whorl of antisepalous stamens. b – relatively heterogeneous androecium, three stamens (2, 5, 8) can be attributed to the whorl of antipetalous stamens, two stamens (3, 6) can be attributed to the whorl of antisepalous stamens, remaining three stamens (1, 4, 7) form a gradient between the two whorls\u003c/p\u003e","description":"","filename":"floatimage11.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/6571917daafacb11a66a391c.jpg"},{"id":62438139,"identity":"082dac71-3de5-4068-b5f6-3616873a1208","added_by":"auto","created_at":"2024-08-14 08:13:11","extension":"jpg","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":145650,"visible":true,"origin":"","legend":"\u003cp\u003eThe area of the floral meristem at the moment of stamen initiation. \u003cem\u003eA. platanoides,\u003c/em\u003e \u003cem\u003eA. tataricum \u003c/em\u003esubsp\u003cem\u003e. ginnala\u003c/em\u003e and \u003cem\u003eA. nikoense\u003c/em\u003eare not statistically different, \u003cem\u003eA. tegmentosum, A. barbinerve\u003c/em\u003e and \u003cem\u003eA. saccharinum\u003c/em\u003e are statistically different from the species of the first group but do not differ from each other, \u003cem\u003eA. negundo\u003c/em\u003e is different from all except \u003cem\u003eA. barbinerve\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage12.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/2018bbcb0bd556ea4a822cef.jpg"},{"id":81569731,"identity":"95df95aa-8650-4b34-a4ed-9e3e64758434","added_by":"auto","created_at":"2025-04-28 16:10:53","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":10052237,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/d88b0857-7947-490f-b2c6-cb55f617e790.pdf"},{"id":62438142,"identity":"b426f7ca-1590-4806-9453-4cd115a96d20","added_by":"auto","created_at":"2024-08-14 08:13:11","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":40118,"visible":true,"origin":"","legend":"","description":"","filename":"supplementaryacerandroecium.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4754778/v1/a8655767c61eaccd84a91f30.xlsx"}],"financialInterests":"","formattedTitle":"Androecium homologies in eight-staminate maples: a developmental study","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003e \u003cem\u003eAcer\u003c/em\u003e is one of the most morphologically diverse genera (De Jong \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1976\u003c/span\u003e) in the family Sapindaceae (Acevedo-Rodr\u0026iacute;guez et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Buerki et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Now, 15 sections are recognized within the genus (Areces-Berazain et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; De Jong \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Li et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Most species of \u003cem\u003eAcer\u003c/em\u003e (sections \u003cem\u003eAcer\u003c/em\u003e, \u003cem\u003eGlabra\u003c/em\u003e, \u003cem\u003eGinnala\u003c/em\u003e, \u003cem\u003ePlatanoidea\u003c/em\u003e, \u003cem\u003ePalmata\u003c/em\u003e, \u003cem\u003eParviflora\u003c/em\u003e, \u003cem\u003ePentaphylla\u003c/em\u003e, \u003cem\u003eSpicata\u003c/em\u003e, \u003cem\u003eMacrantha\u003c/em\u003e, \u003cem\u003eLithocarpa\u003c/em\u003e) share the same flower ground plan (De Jong \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1976\u003c/span\u003e). Typical maple flowers are morphologically bisexual but functionally unisexual, with five usually free sepals, five free petals, eight stamens, and two carpels (Acevedo-Rodr\u0026iacute;guez et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; De Jong \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1976\u003c/span\u003e). But in other sections (\u003cem\u003eRubra, Indivisa, Arguta, Negundo, Macrophylla, Pubescentia\u003c/em\u003e, partially \u003cem\u003ePentaphylla\u003c/em\u003e) there is an increase or decrease in some floral organs or the flower merism (De Jong \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1976\u003c/span\u003e). In particular, in section \u003cem\u003ePentaphylla\u003c/em\u003e, series \u003cem\u003eGrisea\u003c/em\u003e there are ten stamens, which probably correspond to the typical for eudicots pentamerous flower with two-whorled androecium (De Jong \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1976\u003c/span\u003e; Ronse De Craene 2022).\u003c/p\u003e \u003cp\u003eNon-isomerous perianth and androecium have been provoking numerous hypotheses on the homologies and the origin of eight-staminate androecium in the genus. Payer (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1857\u003c/span\u003e) studied flower development in \u003cem\u003eAcer tataricum\u003c/em\u003e L. (section \u003cem\u003eGinnala\u003c/em\u003e) and came to the conclusion that stamens are initiated in two series, though in the mature flower they seemingly form a single whorl. After Payer, sepals are initiated following a 2/5 spiral sequence. After the sepals, the petals are initiated simultaneously. After the petal inception, five stamens appear simultaneously opposite the sepals. Slightly later, the last three stamens arise side by side with already existing primordia. This stamen \u003cem\u003ed\u0026eacute;doublement\u003c/em\u003e (Ronse De Craene and Smets, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e1993\u003c/span\u003e) is rather unusual because duplication of stamens in certain positions occurs sequentially, whereas typical d\u0026eacute;doublement implies the simultaneous appearance of two primordia in the position that was previously occupied by a single organ. According to Payer\u0026rsquo;s general theory, whorled organs that arise sequentially belong to different whorls. Therefore, five of the eight stamens belong to the first whorl, and three belong to the second whorl.\u003c/p\u003e \u003cp\u003eBuchenau (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e1861\u003c/span\u003e) studied flower development in \u003cem\u003eA. pseudoplatanus\u003c/em\u003e L. (section \u003cem\u003eAcer\u003c/em\u003e). His description of sepal and petal initiation is similar to \u003cem\u003eA. tataricum.\u003c/em\u003e However, the description of stamen development is different. According to Buchenau (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e1861\u003c/span\u003e), all eight stamens are initiated simultaneously, but their subsequent growth is unequal. These differences in developmental rates give a false impression of sequential initiation. Buchenau did not provide any interpretation of stamen positions.\u003c/p\u003e \u003cp\u003eEichler (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e1878\u003c/span\u003e) examined organ arrangement in mature flowers in many maple species and concluded that the androecium consists of two whorls (four stamens in each whorl). He suggested that ancestrally, there were ten stamens, and two stamens from different whorls located opposite the carpels were lost.\u003c/p\u003e \u003cp\u003eYet another hypothesis was proposed by Hall (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1951\u003c/span\u003e). He studied the floral vasculature in different species of \u003cem\u003eAcer\u003c/em\u003e. Most of the data obtained were not informative for androecium interpretation except \u003cem\u003eAcer pensylvanicum\u003c/em\u003e L. (section \u003cem\u003eMacrantha\u003c/em\u003e) whose floral vasculature differs from other species. Vascular bundles of five stamens formed common traces with those of petals, and bundles of three stamens formed common bundles with sepals. On this basis, Hall (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1951\u003c/span\u003e) suggested that a typical maple flower has five antipetalous and three antisepalous stamens.\u003c/p\u003e \u003cp\u003eAll the interpretations mentioned above are based either on the mature flowers or on the floral ontogeny of a single species. Despite this diversity of opinions, Eichler\u0026rsquo;s interpretation is the most widespread and commonly accepted (Ronse De Craene, 2022).\u003c/p\u003e \u003cp\u003eThe placement of Aceraceae into the family Sapindaceae (APG IV \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Buerki et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2021\u003c/span\u003e Harrington et al.; \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2005\u003c/span\u003e) made the story even more puzzling. Now, eight stamens are considered an important feature and probably a synapomorphy of the family Sapindaceae (Acevedo-Rodr\u0026iacute;guez et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). The flower development in \u003cem\u003eDipteronia\u003c/em\u003e, the closest relative of the genus \u003cem\u003eAcer\u003c/em\u003e, does not fully support any hypothesis proposed for \u003cem\u003eAcer\u003c/em\u003e (Zhang et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The androecium of \u003cem\u003eDipteronia\u003c/em\u003e is organized in the same way as in other Sapindaceae (Ronse De Craene et al. 2000), it consists of three antipetalous and five antisepalous stamens, but the positions of the two lost stamens are different in \u003cem\u003eDipteronia\u003c/em\u003e compared to \u003cem\u003eAcer\u003c/em\u003e (Zhang et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). If the androecium of maples is of another nature, this could be an argument in favor of the independent origin of eight stamens in different groups of Sapindaceae. Thus, we decided to conduct a study of androecium construction in several 8-staminate species of \u003cem\u003eAcer\u003c/em\u003e in comparison with 10-staminate species to clarify androecium homologies in the genus using both flower development and anatomy.\u003c/p\u003e"},{"header":"MATERIALS AND METHODS","content":"\u003cp\u003eThis study focuses on the early flower development of five species: \u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e (Maxim.) Wesm. (section \u003cem\u003eGinnala\u003c/em\u003e), \u003cem\u003eA. tegmentosum\u003c/em\u003e (Maxim.) Maxim. (section \u003cem\u003eMacrantha\u003c/em\u003e), \u003cem\u003eA. platanoides\u003c/em\u003e L. (section \u003cem\u003ePlatanoidea\u003c/em\u003e), \u003cem\u003eA. spicatum\u003c/em\u003e Lam. (section \u003cem\u003eSpicata\u003c/em\u003e), and \u003cem\u003eA. nikoense\u003c/em\u003e (Miq.) Maxim. (section \u003cem\u003eTrifoliata\u003c/em\u003e). The androecium of all species except \u003cem\u003eA. nikoense\u003c/em\u003e consists of eight stamens.\u003c/p\u003e \u003cp\u003eThe plant material was collected in several parks of the city of Moscow, in the Botanical Garden of Moscow State University, in the Tsitsin Main Botanical Garden of the Russian Academy of Sciences, and in the Saint Petersburg Botanical Garden during three years (2021\u0026ndash;2023). The vouchers are deposited in the Herbarium of Moscow State University (MW) or in spirit collection (MW). Information about vouchers is presented in Supplementary Data.\u003c/p\u003e \u003cp\u003eFor scanning electron microscopy (SEM), generative buds at different developmental stages were dissected under an Olympus SZX7 stereomicroscope. Flowers were dehydrated through absolute acetone, critical-point dried using a Hitachi HCP2 critical-point drier, and then coated with gold and palladium using an Eiko IB-3 ion-coater. Observations were made using a CAMSCAN S2 SEM, JEOL JSM-6380LA, and Quattro S at Moscow University.\u003c/p\u003e \u003cp\u003eFor light microscopy, floral buds of three species (\u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e, \u003cem\u003eA. tegmentosum\u003c/em\u003e, and \u003cem\u003eA. nikoense\u003c/em\u003e) were sectioned using standard methods of paraffin embedding and serial sectioning at 10 \u0026micro;m thickness (Barykina et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2004\u003c/span\u003e) using a Thermo Scientific Microm HM 355s rotary microtome. Sections were stained with Safranin and Alcian Blue and mounted in Biomaunt mounting medium. Digital photomicrographs were taken using an Olympus BX53 (Olympus) microscope fitted with an Olympus SC50 digital camera. Drawings of serial cross sections showing floral vasculature were performed in Krita 5.1.5 software (free raster graphics editor, software developers are Krita Foundation, KDE). \u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e was selected as a species with the typical vasculature described by Hall (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1951\u003c/span\u003e). The vasculature of \u003cem\u003eA. tegmentosum\u003c/em\u003e and \u003cem\u003eA. nikoense\u003c/em\u003e were examined for the first time.\u003c/p\u003e \u003cp\u003eUsing top view SEM images, we analyzed the stamen positions and the area occupied by stamen primordia in three species: \u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e, \u003cem\u003eA. tegmentosum\u003c/em\u003e, \u003cem\u003eA. platanoides\u003c/em\u003e. Additionally, we studied the area of the floral meristem in three species: \u003cem\u003eA. negundo\u003c/em\u003e L., \u003cem\u003eA. barbinerve\u003c/em\u003e Maxim., \u003cem\u003eA. saccharinum\u003c/em\u003e L. These species were selected because they possess a relatively flat receptacle at the stage of stamen initiation. This makes measurements more accurate and avoids systematic errors caused by floral meristem convexity. The stamens were color coded in the sepal spiral coordinate system (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), starting with the stamen opposite the fifth (last initiated) sepal and then moving around the flower towards the third sepal. Each stamen was assigned a different color. This labeling was adopted because the stamens are always located in the same specific way relative to the order of sepal initiation but not their positions relative to the flower-subtending bract. Next, for ten flowers of each species, we calculated the angles between adjacent stamens and the angles between the stamens and the centers of the sepals and petals closest to them. Based on the latter measurements, the \u0026ldquo;antipetality index\u0026rdquo; was calculated for each stamen:\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eα\u003csub\u003ep\u003c/sub\u003e/(α\u003csub\u003ep\u003c/sub\u003e\u0026thinsp;+\u0026thinsp;α\u003csub\u003es\u003c/sub\u003e), where α\u003csub\u003ep\u003c/sub\u003e is the angle between the petal and the stamen, α\u003csub\u003es\u003c/sub\u003e is the angle between the sepal and the stamen. If the index is 0, the stamen is strictly antipetalous. If the index is 1, the stamen is strictly antisepalous.\u003c/p\u003e \u003cp\u003eTo calculate the area occupied by stamen primordium on the floral meristem, the stamen primordia were colored black on the SEM photos, then the size of the black areas was calculated in pixels in ImageJ software and converted to square micrometers according to scale bars. The results were recalculated into relative values. The area of each primordium was divided by the area of the largest primordium in this flower. Therefore, the relative size of one of the eight stamens is 1. The average antipetality index and the relative primordial area of each stamen are shown in the scatterplots. These diagrams were made in the Past 4.03.\u003c/p\u003e \u003cp\u003eTwo qualitative characters were also analyzed: the stamen initiation via petal-stamen primordia and the presence of common vascular bundles with petals or sepals. The presence of a common petal-stamen primordium was scored as 0, the absence as 1. The presence of a common vascular bundle with the petal was scored as 0, with the sepal as 1. If the vascular bundles of the stamens were not associated with the perianth vasculature, this was scored as 0.5.\u003c/p\u003e \u003cp\u003eFor each stamen, the final parameter was calculated based on the sum of four characters: the presence of a common primordium, the presence of a common vascular bundle, the average relative area of the stamen primordium, and the average antipetality index. This parameter shows the degree of belonging of the stamen to the external or internal androecial whorl, as well as the degree of heterogeneity of the androecium. This parameter varies from 0 to 4, its gradations on the diagrams were displayed in color.\u003c/p\u003e \u003cp\u003eThe surface area of the floral meristem at the time of corolla and androecium initiation was measured in four species (\u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginalla\u003c/em\u003e, \u003cem\u003eA. platanoides\u003c/em\u003e, \u003cem\u003eA. tegmentosum\u003c/em\u003e, \u003cem\u003eA. nikoense\u003c/em\u003e) using the ImageJ software according to the method described above. We did not measure the meristem area of \u003cem\u003eA. spicatum\u003c/em\u003e using this method because the flower meristem of this species is too convex, which increases the measurement error. Also, for comparison with species with different stamen numbers we measured the area of the floral meristem in \u003cem\u003eA. negundo\u003c/em\u003e (developmental data published in Zavialov and Remizowa, \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), \u003cem\u003eA. barbinerve\u003c/em\u003e and \u003cem\u003eA. saccharinum\u003c/em\u003e. \u003cem\u003eA. negundo\u003c/em\u003e usually has four stamens and sepals, no petals, \u003cem\u003eA. barbinerve\u003c/em\u003e has four stamens, petals, and sepals, \u003cem\u003eA. saccharinum\u003c/em\u003e has five or six stamens and sepals, no petals. None of these species have common petal-stamen primordia (Zavialov and Remizowa \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zavialov \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). An analysis of variance was performed to compare the areas of the floral meristem in the RStudio program and a histogram was constructed.\u003c/p\u003e \u003cp\u003eAll illustrations were prepared in Krita 5.1.5. Row data for all measurements are presented in the Supplement.\u003c/p\u003e \u003cp\u003eWe want to discuss some of the terms used in this paper. We use the terms median (abaxial and adaxial) and transversal to describe the position of sepals and petals. In a typical pentamerous flower, one of the sepals or petals occupies a strictly median position (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Four other sepals (petals) occupy intermediate positions in pairs closer to the adaxial or abaxial side. Therefore, the terms \u0026ldquo;transversal-adaxial\u0026rdquo; and \u0026ldquo;transversal-abaxial\u0026rdquo; are used to describe such positions (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWe use the same terms to describe the position of the stamens or indicate their approximate position relative to the spiral of sepal initiation. We avoid the widely used terms \u0026ldquo;antipetalous\u0026rdquo; (opposite the petal median) and \u0026ldquo;antisepalous\u0026rdquo; (opposite the sepal median), because the actual position of the stamens in maples often does not fit such strictly defined concepts. Instead, we divided stamens into those arising from a single primordium and those arising from a common petal-stamen primordium. Only for \u003cem\u003eA. nikoense\u003c/em\u003e we used the terms antipetalous and antisepalous, because the positions of stamens are more consistent with these concepts.\u003c/p\u003e \u003cp\u003eIn SEM photographs, we considered two cases to be a common petal-stamen primordium: an elongate smooth primordium from a single mass of meristem, a primordium with two apices and a common basal part raised above the surface of the receptacle.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003e \u003cb\u003eSequence of organ initiation\u003c/b\u003e \u003c/p\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eA. platanoides\u003c/h2\u003e \u003cp\u003eSepals are initiated following a 2/5 spiral sequence. The first sepal is formed in a transversal-abaxial or transversal-adaxial position (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). The median sepal can be in either an abaxial or adaxial position. The sepal primordia are of different sizes, but they are quite wide to occupy the entire flower circumference. The initiation of petals and stamens begins even before the formation of the last fifth sepal (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). Five stamens appear earlier: four are initiated by a single primordia opposite the first, second, third, and fourth sepals and one is initiated by a common petal-stamen primordium between the first and fourth sepals (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb, c). Simultaneously with the first series of stamens, two petals appear between the first and third sepals and between the second and fourth sepals (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb, c). Later, the three remaining stamens appear: two are initiated by a common petal-stamen primordia on the sides of the fifth sepal, and one is initiated by a separate primordium opposite the fifth sepal (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eNext, the common petal-stamen primordia divide into the petal and stamen parts (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee, f). The common petal-stamen primordium located between the second and fifth sepals or between the third and fifth sepals is delayed relative to the two others (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef). The stamens emerging from common primordia are slightly smaller at the beginning, but subsequently, the differences in size quickly disappear (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea). The petals do not differ in size or developmental rates. During the development of other floral organs, petals flatten and grow slowly (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb \u0026ndash; d). Petal enlargement occurs in the later stages of development, shortly before flowering.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eA. tataricum\u003c/b\u003e \u003cb\u003esubsp.\u003c/b\u003e \u003cb\u003eginnala\u003c/b\u003e\u003c/p\u003e \u003cp\u003eThe sepals are initiated in a 2/5 spiral. The first sepal is always initiated in a transversal-abaxial position, and the second sepal is always initiated in the median adaxial position (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea, b). The bar-shaped sepal primordia are of different size, as in \u003cem\u003eA. platanoides\u003c/em\u003e. The first petal appears between the first and third sepals, then the petal between the second and fourth sepals (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb, c). Next, three petals simultaneously arise from common petal-stamen primordia. At the same time, five stamens are initiated from a single primordial (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec). Soon, the common primordia are successively (similar to \u003cem\u003eA. platanoides\u003c/em\u003e) divided into staminate and petal parts (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed, e). At this stage, the stamens located on the radii of the carpels are smaller, than the others (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ef). However, differences in stamen size soon disappear. Petal growth is retarded, but by the time of anther formation, the stamens are already covered by petals (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eg).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eA. spicatum\u003c/h2\u003e \u003cp\u003eThe sepals are initiated in a 2/5 spiral as small primordia (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea). As a result, it can be difficult to trace the spiral of sepal initiation at later stages. The first sepal is usually formed in a transverse-abaxial position (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea). At the time of petal initiation, the receptacle remains clearly convex (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb). Otherwise, the pattern of petal and stamen initiation is similar to that of \u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e. First, two petals appear sequentially between the first and third sepals and between the second and fourth sepals (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb). Then the three remaining petals and eight stamens appear: five stamens are initiated from single primordia, three are initiated from common primordia with petals (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ec \u0026ndash; f, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea). Immediately after their inception, the stamens are identical in size or their sizes slightly differ, although no regular patterns can be seen in the distribution of stamens with different sizes (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb, c). The petals remain small after initiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ed). Their main growth occurs in the later stages of flower development.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eA. tegmentosum\u003c/h2\u003e \u003cp\u003eThe sepals are initiated in a 2/5 spiral by small primordia, almost indistinguishable in size (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea, b). The first sepal is usually in a transversal-abaxial position, and the second is in a median adaxial position (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea, b). Next, a petal between the first and third sepals appears, and a stamen located opposite the third (or less often the first sepal) is initiated by a single primordium (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb). Next, the four stamens are initiated by single primordia in a spiral that continues in the direction of the spiral of sepal initiation. After the initiation of the first four stamens by single primordia, four common petal-stamen primordia appear almost simultaneously, between the third and fifth, second and fifth, first and fourth, second and fourth sepals (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb). The division of common primordia appears sequentially, starting from the primordium between the second and fourth sepals (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ec, d). During further development, the stamens differ (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ed \u0026ndash; f). The stamens, which are initiated via common primordia, remain small. The stamens, which arose from single primordia, have different sizes, corresponding to the order of their initiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ed \u0026ndash; f). The largest stamen is the first, and so on. At later stages of development, the difference in size between the stamens gradually disappears. Corolla development is similar to \u003cem\u003eA. platanoides\u003c/em\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eA. nikoense\u003c/h2\u003e \u003cp\u003eThis species, unlike all those described above, has an unusual (for maples) floral groundplan. There are three flowers in the inflorescence: two lateral and a terminal one. Lateral flowers usually consist of five sepals, one of which occupies the median abaxial position, five petals, ten stamens (five opposite sepals, five opposite petals), and two carpels located in the transversal plane (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ec, d). The terminal flower is usually hexamerous and consists of six sepals, six petals, twelve stamens, and two carpels. Below we provide a description of lateral pentamerous flowers.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTwo large transversal sepals are first to appear (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ea). Next, the median and transverse-adaxial sepals are formed (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eb, c). The initiation of one of the transverse-adaxial sepals is sometimes delayed (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ec). There is a size difference between the sepals in the early stages: the transversal sepals are larger, the median abaxial sepal is of medium size, and the transverse-adaxial sepals are the smallest (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eb). Even before the formation of the last sepal, petals and stamens appear (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eb). Single primordia of stamens opposite the sepals and three upper common petal-stamen primordia are formed almost simultaneously (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eb). Transversal-abaxial common primordia arise later (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ec). Antisepalous stamens are larger than antipetalous ones and are located closer to the center of the flower, which indicates obdiplostemony (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ed). At this stage, two whorls of the obdiplostemonous androecium are clearly distinguishable. In some cases, there are two stamens located opposite one perianth element (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ee, f). Moreover, in cases where these are antipetalous stamens, apparently, they both are initiated from the same common petal-stamen primordium (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ee). Sometimes, there is an increase in the flower: the lateral flowers are hexamerous (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ee) and the terminal flowers are heptamerous. The further development of stamens and petals is similar to that of other species. The petals flatten after their appearance and enlarge shortly before anthesis. In the mature flower, antipetalous stamens possess short filaments and are located closer to the flower periphery (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003ec). The petal bases envelope the bases of the stamen filaments. Antisepalous stamens have long filaments and are located closer to the center of the flower (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003ec).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eFloral vasculature\u003c/h2\u003e \u003cp\u003eIn \u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e and \u003cem\u003eA. nikoense\u003c/em\u003e stamen bundles are not associated with the perianth bundles (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003ea, c). Bundles of sepals and petals are connected. Branches from adjacent common bundles can merge into a bundle of one petal. Bundles of stamens separate from the stele independently.\u003c/p\u003e \u003cp\u003e \u003cem\u003eA. tegmentosum\u003c/em\u003e demonstrates the same vascular pattern as in \u003cem\u003eA. pensylvanicum\u003c/em\u003e (Hall \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1951\u003c/span\u003e). There are independent bundles of two sepals (first and second in the initiation spiral), five common bundles of petals and stamens, and three common bundles of sepals and stamens (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eb).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eAngles between adjacent stamens\u003c/h2\u003e \u003cp\u003eAll three species examined (\u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e, \u003cem\u003eA. tegmentosum\u003c/em\u003e, \u003cem\u003eA. platanoides\u003c/em\u003e) are similar in the distribution of larger and smaller angles between adjacent stamens (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003ea, c, e), but the exact angle sizes differ between the species. In \u003cem\u003eA. platanoides\u003c/em\u003e, the confidence intervals of all angles overlap (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003ea). In the other two species, the confidence intervals for the smallest and largest angles do not overlap (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003ec, e). In all species, the greatest angles are between the stamens near the first and third sepals and between the stamens near the second and fourth sepals. Stamens arising from common petal-stamen primordia (orange, light blue, and black in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003e) form relatively small angles with adjacent stamens.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eSize and position of stamen primordia\u003c/h2\u003e \u003cp\u003eWe also analyzed the size of the stamen primordia and their position relative to the sepals and petals (for more details on the measurement technique, see Materials and Methods). The results are presented in scatterplots (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eb, d, f). \u003cem\u003eA. platanoides\u003c/em\u003e and \u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e show a similar picture. In these species, three groups of stamens can be roughly distinguished. The three stamens (orange, light blue, and black) are relatively small and located closer to the petals than to the sepals. The other two groups of stamens differ only in their positions. Three stamens (yellow, green, and purple) occupy an intermediate position between the sepals and petals, and the remaining two stamens (red and blue) are located closer to the sepals. In \u003cem\u003eA. platanoides\u003c/em\u003e, the stamens belonging to the same group are very similar in size (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eb), whereas in \u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e, black and green stamens stand out from their groups in size (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003ed). These stamens are located opposite the carpels, which probably affects their size.\u003c/p\u003e \u003cp\u003eIn \u003cem\u003eA. tegmentosum\u003c/em\u003e, only one group of stamens can be distinguished (orange, blue, and black), and these stamens are smaller and closer to the petals compared to the same stamens in the other two species (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003ef). The five remaining stamens differ considerably from each other in size and/or position.\u003c/p\u003e \u003cp\u003eBased on a combination of four characters (the presence of a common primordium, the presence of a common vascular bundle, the average relative area of the stamen primordium, the average antipetality index), each stamen can be characterized as belonging to an external or internal whorl, and the androecium as more or less homogeneous (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e). The results show that \u003cem\u003eA. platanoides\u003c/em\u003e and \u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e are characterized by a relatively homogeneous androecium, but it is still possible to distinguish two groups of stamens from different whorls (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003ea). One group can be interpreted as stamens of the antipetalous whorl, and the second group as stamens of the antisepalous whorl. Although this interpretation is probably more a matter of agreement. \u003cem\u003eA. tegmentosum\u003c/em\u003e exhibits a remarkably heterogeneous androecium (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003eb). Three stamens distinctly belong to the antipetalous whorl, three distinctly belong to the antisepalous whorl, and two stamens occupy an intermediate position between them. As a result, two whorls are clearly distinguished, but it is impossible to draw a clear boundary between them.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eFloral meristem size\u003c/h2\u003e \u003cp\u003eFor seven species, we measured the surface area of the floral meristem at the time of petal and stamen initiation. Species can be divided into two groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003e). The first is represented by \u003cem\u003eA. nikoense\u003c/em\u003e, \u003cem\u003eA. platanoides\u003c/em\u003e, and \u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e. The area of their meristem ranges from approximately 2500 to 3500 \u0026micro;m\u003csup\u003e2\u003c/sup\u003e. The second group includes \u003cem\u003eA. tegmentosum, A. saccharinum\u003c/em\u003e and \u003cem\u003eA. barbinerve\u003c/em\u003e. The area of their floral meristem before the initiation of petals and stamens does not exceed 2500 \u0026micro;m\u003csup\u003e2\u003c/sup\u003e. The smallest floral meristem is observed in \u003cem\u003eA. negundo\u003c/em\u003e. This species is statistically significantly different from all species except \u003cem\u003eA. barbinerve\u003c/em\u003e. P \u0026ndash; values of the analysis of variance are presented in the supplement.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"DISCUSSION","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eComparison with other Sapindaceae\u003c/h2\u003e \u003cp\u003eFlower development has now been studied in a number of species of Sapindaceae (for example, Cao et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Cao and Xia \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Payer \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1857\u003c/span\u003e; Ronse De Craene et al. 2000; Zhang et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Zhou et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). However, a lot of data are disputable.\u003c/p\u003e \u003cp\u003eFor all species studied, a sequential initiation of sepals in a 2/5 spiral has been described ( Cao et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2006\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2017\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Payer \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1857\u003c/span\u003e; Ronse De Craene et al. 2000; Zhang et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Zhou et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). After calyx initiation, the floral meristem becomes pentagonal in outline, and petals appear. Petals are retarded in their growth and enlarge shortly before flowering. In some species (\u003cem\u003eHandeliodendron bodinieri\u003c/em\u003e (H.L\u0026eacute;v.) Rehder, \u003cem\u003eDelavaya toxocarpa\u003c/em\u003e Franch., \u003cem\u003eKoelreuteria bipinnata\u003c/em\u003e Franch., \u003cem\u003eXanthoceras sorbifolium\u003c/em\u003e Bunge), petal initiation has been described as simultaneous (Cao et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2006\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Cao and Xia \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Zhou et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2019\u003c/span\u003e); in others (\u003cem\u003eKoelreuteria paniculata\u003c/em\u003e Laxm., \u003cem\u003eEurycorymbus cavaleriei\u003c/em\u003e (H.L\u0026eacute;v.) Rehder \u0026amp; Hand.-Mazz., \u003cem\u003eDipteronia sinensis\u003c/em\u003e Oliv.), petals are initiated sequentially (Cao et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Ronse De Craene et al. 2000; Zhang et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In \u003cem\u003eD. sinensis\u003c/em\u003e and \u003cem\u003eE. cavaleriei\u003c/em\u003e, the appearance of petals on each side of the fifth sepal overlaps with the initiation of stamens, giving the false impression of common primordia (Cao et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Zhang et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In species of the genus \u003cem\u003eKoelreuteria\u003c/em\u003e and \u003cem\u003eD. toxocarpa\u003c/em\u003e, one petal (between the second and fifth sepals or between the third and fifth sepals) arises from a common primordium with the stamen (Cao et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Cao and Xia \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Ronse De Craene et al. 2000).\u003c/p\u003e \u003cp\u003eThe time of initiation of stamens and petals overlaps. Stamens are formed in rapid succession, so it is often not possible to determine the exact order of their appearance. In \u003cem\u003eKoelreuteria bipinnata\u003c/em\u003e var. \u003cem\u003eintegrifolia\u003c/em\u003e (Merr.) T.C. Chen and \u003cem\u003eCardiospermum halicacabum\u003c/em\u003e L., however, the order of stamen initiation is unidirectional, starting from the stamen located opposite the fourth sepal and ending on the stamen between the third and fifth sepals (Cao et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Payer \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1857\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThere are two aspects of flower development in the family with great evolutionary and taxonomic importance: the presence of common primordia and the changes in flower symmetry during development (Zhang et al., \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In most of the studied species of Sapindaceae, flowers at the middle developmental stages are obliquely monosymmetrical. This monosymmetry is established by a delay in the development of some petals or their initiation from common primordia. Zhang et al. (\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) clearly imply that flower morphogenesis in this case follows evolution, as postulated by the biogenetic law (Haeckel \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1866\u003c/span\u003e). Although these ideas are now rejected by zoologists, they are often used by botanists (Ronse De Craene 2018). On this basis, Zhang et al. (\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) believe that the ancestral flower in Sapindaceae was most likely monosymmetrical with an oblique plane of symmetry. Changes in flower symmetry during development can occur many times and are observed in many angiosperms (Endress \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1999\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2001\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Developmental changes in symmetry are often associated with the pressure of surrounding structures and flower parts on each other (Bull\u0026ndash;Here\u0026ntilde;u et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Endress \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1999\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Remizowa et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2013\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Therefore, we doubt the correctness of using flower symmetry at early stages of development to reconstruct the symmetry of the ancestral flower.\u003c/p\u003e \u003cp\u003eZhang et al. (\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) apparently do not distinguish between the concepts of zygomorphy and monosymmetry. Therefore, they probably also imply that the ancestral flower was zygomorphic. However, we will differentiate these concepts to avoid confusion. Monosymmetric flowers are flowers through which only one plane of symmetry can be drawn, so all the organs of the flower will be mirror symmetrical relative to a single plane (Endress, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). A zygomorphic flower is a flower in which the organs are unevenly distributed in space, so that there are two sides (usually the upper and lower) that differ in some external parameters important for attracting or interacting with the pollinator (Bukhari et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Nuraliev et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Remizowa et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIf we assume that eight stamens are arranged into two whorls (one whorl with five, the other with three stamens), then the flower of Sapindaceae is automatically monosymmetrical.\u003c/p\u003e \u003cp\u003eAccording to Zhang et al. (\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), the flowers of \u003cem\u003eDipteronia\u003c/em\u003e and \u003cem\u003eAcer\u003c/em\u003e are disymmetrical. The planes of symmetry are set by a syncarpous gynoecium consisting of two carpels. However, if we take into account the position of both the gynoecium and the androecium, then the flowers of both \u003cem\u003eDipteronia\u003c/em\u003e and maples with eight stamens should be considered asymmetrical because the symmetry planes of different whorls do not coincide. The exception are some flowers of \u003cem\u003eA. spicatum\u003c/em\u003e, which can be considered monosymmetrical (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec). Before carpel initiation, the stamens determine the plane of floral symmetry. Monosymmetry at this developmental stage depends on the order of stamen initiation. In some species, stamens are formed almost simultaneously. Therefore, the plane of (mono)symmetry can be drawn only based on the initiation of stamens and petals from common or single primordia. In \u003cem\u003eA. platonoides\u003c/em\u003e, stamen initiation is unidirectional, so that the plane of monosymmetry is clearly visible (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb \u0026ndash; e). In \u003cem\u003eA. tegmentosum\u003c/em\u003e, the first four stamens arise in a spiral, the remaining four appear almost simultaneously from common stamen-petal primordia (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb, c), and therefore this pattern of development is immediately asymmetrical. Flower asymmetry in this species is probably also due to non-integer merism (see below).\u003c/p\u003e \u003cp\u003eAs soon as the gynoecium appears, the flower becomes asymmetrical. The monosymmetry of the maple flower in its early development is not necessarily evidence of ancestral zygomorphy, but it can be considered a natural pattern associated with the structural features of the androecium. The flower of the common ancestor of maples and \u003cem\u003eDipteronia\u003c/em\u003e was indeed monosymmetric and zygomorphic, at least other members of the subfamily Hippocastanoideae are characterized by zygomorphic monosymmetrical flowers (Acevedo-Rodr\u0026iacute;guez et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Buerki et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). However, we believe this is a coincidence.\u003c/p\u003e \u003cp\u003eThe second important feature of flower morphogenesis in many Sapindaceae is the presence of common primordia of petals and stamens. The formation of common primordia in this particular group can be regarded as a manifestation of petal reduction by delay in their development (Cao et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Ronse De Craene et al. 1993).\u003c/p\u003e \u003cp\u003eThe development of three and even four stamens from common primordia is a unique feature of the genus \u003cem\u003eAcer\u003c/em\u003e and was previously unknown in Sapindaceae. The most plausible explanation for such uniqueness is still an insufficient knowledge of the diversity of floral organogenesis in this family. However, a certain trait can be traced. Most of the previously studied Sapindaceae have a trimerous gynoecium and, therefore, most likely, a larger size of the floral meristem compared to maples, which typically possess two carpels (Ronse De Craene 2016).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eUse of developmental data for androecium homologies\u003c/h2\u003e \u003cp\u003eDevelopmental data are considered one of the most important sources for inferring homology (Hufford \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Ronse De Craene 2018; Rutishauser and Moline \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Sattler and Rutishauser \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e1997\u003c/span\u003e). The stamens of initially different whorls may have some peculiarities of development, for example, order of initiation or presence or absence of common petal-stamen primordia.\u003c/p\u003e \u003cp\u003eThe criterion of initiation sequence to differentiate stamens of the two whorls, as suggested by Payer (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1857\u003c/span\u003e), is poorly applicable due to the number of variations in the order of stamen initiation in maples. Moreover, we did not identify the pattern described by Payer (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1857\u003c/span\u003e) in any of the species examined here. In \u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e and \u003cem\u003eA. spicatum\u003c/em\u003e, all stamens appear almost simultaneously, in \u003cem\u003eA. platonoides\u003c/em\u003e the initiation of stamens is unidirectional, and in \u003cem\u003eA. tegmentosum\u003c/em\u003e, the first four stamens are initiated in a spiral, and the remaining stamens appear later simultaneously. All the described variants for the order of stamen initiation do not allow for the separation of stamens into whorls in the sense of Payer (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1857\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eUsing data on the presence of common primordia seems more promising. It is believed that a common primordium is formed by organs located in the same radius (Ronse De Craene et al. 1993). Therefore, we can definitely consider stamens, which arise from common primordia, as antipetalous. In most eight-staminate maples, three stamens are formed by common primordia, so these stamens are probably antipetalous, and the remaining five are antisepalous. This interpretation of maple flowers is consistent with the hypothesis previously proposed for Sapindaceae (Ronse De Craene et al. 2000). But in \u003cem\u003eA. tegmentosum\u003c/em\u003e four stamens are formed via common primordia. This fact can be interpreted in different ways: as evidence of an independent origin of the eight-staminate androecium or as a secondary change in the pattern of stamen development.\u003c/p\u003e \u003cp\u003eAn important problem is the difficulty of determining whether common primordia are present or absent. Sometimes the reason why a particular case is described as a common primordium, or vice versa is not described, is not clear. This problem is caused by several reasons: the presence of transitional forms, the short longevity of common primordia, and the dependence of interpretation on the viewing angle (Remizowa et al. \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Ronse De Craene et al. 1993; Sattler \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). In rare cases, the presence of common primordia varies between flowers of the same species, for example, in some species of the genus \u003cem\u003eTofieldia\u003c/em\u003e (Remizowa et al. \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2006\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn the family Sapindaceae, common primordia have been described only in the most unambiguous cases. Other developmental patterns have been interpreted as a simultaneous initiation of stamens and petals, giving a false impression of the presence of common primordia (Cao et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Zhang et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Is it necessary to clearly distinguish between these cases? They demonstrate the same pattern: synchronization of initiation and spatial shift. \u0026ldquo;False\u0026rdquo; common primordia may represent a transitional form between common and single primordia. Such ideas bring us to the question of the reasons for the appearance of common primordia and their possible functions.\u003c/p\u003e \u003cp\u003eThe reasons for the appearance of common primordia are not well understood. Endress (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e1995\u003c/span\u003e) noted that in monocots, common primordia are a manifestation of the synorganization of floral organs and that they are formed only on a flat flower apex.\u003c/p\u003e \u003cp\u003eRonse De Craene (2018), in relation to eudicots, puts forward slightly different factors for the formation of common primordia: a decrease in the height of the floral meristem and a reduction of petals, expressed via a certain delay in their development. In his opinion, the presence of common primordia indicates a gradual loss of petals in an evolutionary and phylogenetic context (Ronse De Craene 2018, 2024). We believe that in maples, common primordia are not generally associated with petal reduction. Thus, in \u003cem\u003eA. negundo\u003c/em\u003e, the petals were presumably lost in evolution much later than the stamens formed from common primordia, which were lost even before the divergence of the sections \u003cem\u003eArguta\u003c/em\u003e and \u003cem\u003eNegundo\u003c/em\u003e (Zavialov and Remizowa \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Loss of the corolla is quite rare in such a morphologically diverse genus (de Jong \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1976\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe idea of the importance of a certain shape and height of the floral meristem for the formation of common primordia is questionable, because, in some monocots, common primordia occur on a convex or even elongated cylindrical receptacle, as in representatives of the genus \u003cem\u003eEriocaulon\u003c/em\u003e (De Lima Silva et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Sokoloff et al. \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The criterion of small height of the floral meristem (Ronse De Craene 2018) is not fully consistent with our data. Among maples, common primordia are formed in both species with a flat receptacle (for example, \u003cem\u003eA. platanoides\u003c/em\u003e) and with a convex receptacle (\u003cem\u003eA. spicatum\u003c/em\u003e).\u003c/p\u003e \u003cp\u003eIn our opinion, the only satisfactory explanation for the formation of common primordia is a general decrease in the size of the floral meristem, not necessarily associated with the decrease in meristem height. We tested this hypothesis by estimating the surface area of the floral meristem at the time of petal and stamen initiation.\u003c/p\u003e \u003cp\u003eThe formation of common primordia is associated with a decrease in space on the floral meristem at the time of the formation of stamens (Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e12\u003c/span\u003e). A decrease in the floral meristem can be expressed as a decrease in its area or height and should be revealed by comparison with closely related species without common primordia or with a smaller number of them.\u003c/p\u003e \u003cp\u003eMaple species with eight stamens and three common primordia are similar in floral meristem area (about 30,000 \u0026micro;m\u003csup\u003e2\u003c/sup\u003e), while species with eight stamens and four common primordia (\u003cem\u003eA. tegmentosum\u003c/em\u003e) and species with fewer stamens and without common primordia (\u003cem\u003eA. saccharinum, A. barbinerve\u003c/em\u003e and \u003cem\u003eA. negundo;\u003c/em\u003e for details, see Zavialov and Remizowa \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zavialov \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) have meristem areas of about 20,000 \u0026micro;m\u003csup\u003e2\u003c/sup\u003e or less. A decrease in the floral meristem size is a factor of reduction. Common primordia in this case act as a compensatory mechanism, changing morphogenesis, but preserving the definitive structure as in \u003cem\u003eA. tegmentosum\u003c/em\u003e. However, reduction of flower organs still occurs when the floral meristem decreases beyond a critical level, as in \u003cem\u003eA. saccharinum, A. barbinerve\u003c/em\u003e and \u003cem\u003eA. negundo\u003c/em\u003e. In most cases, the loss of common primordia occurs through the reduction of one organ from a pair or both organs arising from a common primordium. On the other hand, the number of common primordia may increase not due to a decrease in the floral meristem but due to an increase in the number of floral organs, as in \u003cem\u003eA. nikoense\u003c/em\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eUse of positional criteria for androecium homologies\u003c/h2\u003e \u003cp\u003eThe ancestor of Sapindaceae had ten stamens in two whorls, and two stamens have been lost. In theory, if two stamens were lost, the remaining eight should be evenly distributed in space due to shifts (Ronse De Craene et al. 2000).\u003c/p\u003e \u003cp\u003eThe position of the stamens can be considered both relative to each other and relative to the perianth. If we consider the position of the stamens relative to the sepals and petals, this allows to evaluate the antisepality and antipetality of the stamens. The stamens adjacent to the disappeared ones should change their position more significantly. These stamens should occupy an intermediate position between the petal and sepal. In reality, there are three stamens, the position of which corresponds to this theory. Two of them are adjacent (yellow and green in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e opposite the first and third sepals), between them there was one of the lost stamens. It is difficult to identify the position of the second lost stamen in this way.\u003c/p\u003e \u003cp\u003eIn general, the size of the stamen primordia and their position relative to the perianth allows to divide the stamens into three groups. The first group of stamens is closer to the antipetalous position (stamens on the sides of the fifth sepal and the stamen between the first and fourth sepals: black, orange, and light blue in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), the second group is closer to the antisepalous position (approximately in the radii of the fourth and fifth sepals: red and blue in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), the third group is in an intermediate position (stamens opposite the first, second, and third sepals: yellow, green, purple in the Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn fact, all stamens are shifted to some degree relative to the median of the sepals and petals. In this regard, the use of the terms \u0026ldquo;antisepalous\u0026rdquo; and \u0026ldquo;antipetalous\u0026rdquo; to describe the androecium of maples and probably Sapindaceae in general is not entirely correct, although it is widely used. The description of the stamen position is often based not on the actual observed pattern, but rather on the desire to fit a theoretical model. The existing hypothesis about the origin of the androecium in Sapindaceae (Ronse De Craene et al. 2000) is based on similar approximations, which are not entirely consistent with the observed flower construction, although this theory gives a plausible explanation of the androecium evolution in Sapindaceae.\u003c/p\u003e \u003cp\u003eAnother way to understand the position of the stamens is to compare their positions relative to each other. Presumably, if two stamens are lost, the remaining ones are redistributed more or less equally, so the angles between adjacent stamens should be the same. In reality, there are some larger angles, that could mark the positions of the lost stamens. These angles should be measured only before the gynoecium initiation because carpel growth causes secondary shifts of the stamens, which were initially located in carpel radii. This is clearly visible in \u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e, because this species is characterized by the relatively early initiation of the gynoecium.\u003c/p\u003e \u003cp\u003eHowever, the interpretation of the androecium of \u003cem\u003eA. tegmentosum\u003c/em\u003e is not so obvious. The position of the stamens relative to the perianth indicates the presence of distinctly antipetalous and antisepalous stamens, which is not typical for other studied species with eight stamens. The remaining stamens (green and purple in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) form a positional gradient between the two extreme states. In other species, these stamens formed their own group with an intermediate position. The position of the stamens relative to each other shows the fundamental similarity of the androecium of this and other species.\u003c/p\u003e \u003cp\u003eThe stamen positions can also be characterized using vasculature. In most species, the vascular bundles of all eight stamens depart from the receptacle at the same level and are not associated with the perianth, so the vasculature does not add information about the homology of stamens. However, in \u003cem\u003eA. tegmentosum\u003c/em\u003e, and apparently in other species of the section \u003cem\u003eMacrantha\u003c/em\u003e (Hall \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1951\u003c/span\u003e), the vascular bundles of five stamens are associated with bundles of sepals, and the bundles of three stamens with bundles of petals. Such vasculature is not consistent with either developmental or positional data.\u003c/p\u003e \u003cp\u003e \u003cb\u003eAndroecium homologies in\u003c/b\u003e \u003cb\u003eAcer tegmentosum\u003c/b\u003e\u003c/p\u003e \u003cp\u003eDifferent criteria provide different results for \u003cem\u003eAcer tegmentosum\u003c/em\u003e. The problem of homology in cases of conflicting interpretations can be considered from different points of view.\u003c/p\u003e \u003cp\u003eUsing approaches of classical morphology, we believe that the stamens of different whorls have some discrete identity (Arber \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e1946\u003c/span\u003e; Timonin \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). The degree of the differences between the stamens of two whorls may change, the whorls may be combined into one, but the stamen identity as belonging to a certain whorl remains. With this approach, it is essential to select the criterion of homology that would show the stamen identity in the most accurate way. If we follow this way of thinking, then it is logical to assume that the most suitable criterion is the position of the stamens relative to each other. This criterion allows not to separate \u003cem\u003eA. tegmentosum\u003c/em\u003e from other maples. The assumption that the androecium of \u003cem\u003eA. tegmentosum\u003c/em\u003e has the same origin as in all other Sapindaceae is also consistent with the generally accepted principle of Occam's razor (Walsh \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e1979\u003c/span\u003e; McFadden \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Also, the observed position of the stamens is probably non-optimal due to being unevenly spaced, and therefore can be regarded as a special feature most likely inherited from an ancestor (Timonin \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e1993\u003c/span\u003e). The developmental and vascular features of this species can be explained by a decrease in the floral meristem while maintaining the definitive floral structure.\u003c/p\u003e \u003cp\u003eThe same problem can be resolved using Sattler\u0026rsquo;s dynamic morphology (Sattler \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e1992\u003c/span\u003e, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). This approach avoids discrete homology, and considers each case as a combination of structural features and patterns of morphogenesis. According to such ideas, the classical plant organs are nothing more than the most common combinations of characters, between which there is a continuum of other rare intermediate variants (Sattler \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e1996\u003c/span\u003e; Sattler and Rutishauser \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn terms of dynamic morphology, each stamen in a flower could hypothetically possess a unique set of characters. In most cases, this uniqueness is not observed, and all stamens of one whorl are uniform or discretely unequal, with the formation of several types of stamens, for example, in Melastomataceae (Melo et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) or in Cassiinae (Fabaceae) (Tucker \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). The stamens of different whorls usually differ discretely from each other, at least by the level of their insertion. In \u003cem\u003eA. tegmentosum\u003c/em\u003e, different stamens have different combinations of characters, usually distinguishing two whorls. Separation of whorls in the androecium is possible in this case, but the boundary between the two whorls is blurred. In other species, the difference between the whorls is even less pronounced. Followers of dynamic morphology often present their concept as a morphogenetic space with more and less stable variants (Jeune and Sattler \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e1992\u003c/span\u003e). In relation to the androecium of \u003cem\u003eAcer\u003c/em\u003e, stable variants would be haplostemony and diplostemony, in both cases with a whorl non-isomerous to the perianth. The androecium of maples lies between these two extremes. In most species studied here, it is closer to a single-whorled variant, whereas in \u003cem\u003eA. tegmentosum\u003c/em\u003e, it is closer to a two-whorled one. Based on the observed features, we propose to use the terms \u0026ldquo;homogeneous\u0026rdquo; and \u0026ldquo;heterogeneous\u0026rdquo; androecium to characterize the degree of separation of stamens into whorls. The androecium of \u003cem\u003eA. tegmentosum\u003c/em\u003e can be regarded as having non-integer merism. Non-integer merism is uncommon and has been described for the perianth or androecium (Choob and Yurtseva \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Rudall et al. \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2005\u003c/span\u003e;Vislobokov et al. \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). These types of flower merism are considered transitional between typical cases (Choob \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). In this sense, the androecium of all maples with eight stamens can be regarded as a transitional condition evolving towards haplostemony. But only \u003cem\u003eA. tegmentosum\u003c/em\u003e has stamens, which cannot be confidently attributed to any whorl.\u003c/p\u003e \u003cp\u003eHeterogeneity of the androecium in \u003cem\u003eA. tegmentosum\u003c/em\u003e is probably associated with a decrease in the area of the floral meristem while maintaining the number of flower organs.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eConcluding remarks on androecium homologies in 8-staminate species\u003c/h2\u003e \u003cp\u003eTo summarize, we can conclude that maples have lost two antipetalous stamens: one was located opposite the petal between the first and third sepals, and the second was opposite the petal between the second and fourth sepals. The same stamens are also absent in other Sapindaceae (Ronse de Craene et al. 2000). It is important to note that the positions of the lost stamens are not opposite or near the carpel backs, but rather opposite the septum of the ovary. Therefore, it should be assumed that the gynoecium did not directly influence the reduction of these stamens in maples. Ideas by A.W. Eichler (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e1878\u003c/span\u003e) turned out to be wrong.\u003c/p\u003e \u003cp\u003eWhy have stamens been lost in these particular positions? A common answer to this question is certain mechanical forces during flower development (i.e. pressure of the perianth (Brockington et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Ronse De Craene 2024)) and the influence of gynoecium prepatterning (Choob and Penin \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). Therefore, organs located simultaneously opposite the perianth elements, which are initiated earlier and opposite the carpels are the candidates to be lost (Brockington et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Ronse De Craene and Wei 2019; Ronse De Craene 2024).\u003c/p\u003e \u003cp\u003eZhang et al. (\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) suggested that the order of petal initiation inversely correlates with the degree of stamen reduction in Sapindaceae. There are no stamens opposite those petals that usually appear first. Petals that are initiated later than the others tend to emerge from common primordia. According to the ideas of Ronse De Craene (2018), such developmental behavior is an indicator of petal reduction. This point of view is in accordance with the flower development of \u003cem\u003eA. tataricum\u003c/em\u003e subsp. \u003cem\u003eginnala\u003c/em\u003e and \u003cem\u003eA. spicatum\u003c/em\u003e, but the development of \u003cem\u003eA. platanoides\u003c/em\u003e contradicts such ideas. In \u003cem\u003eA. platanoides\u003c/em\u003e, the petal, which is formed from a common petal-stamen primordium (between the first and fourth sepals), is one of the first to appear. This common primordium divides into stamen and petal parts even before the initiation of all petals and stamens.\u003c/p\u003e \u003cp\u003e \u003cb\u003eObdiplostemony in\u003c/b\u003e \u003cb\u003eA. nikoense\u003c/b\u003e\u003c/p\u003e \u003cp\u003eIn \u003cem\u003eA. nikoense\u003c/em\u003e, the androecium is two-whorled, obdiplostemonous, with five stamens in each whorl. Antisepalous stamens appear first in development, antipetalous stamens arise later from common primordia. The stamen initiation is influenced by the flower-subtending bract; therefore, the adaxial organs are formed earlier than the abaxial ones. During further development, the antipetalous stamens are pushed more and more towards the periphery of the flower. This pattern of androecium development indicates secondary obdiplostemony type 1, according to the classification of Ronse De Craene and Bull-Here\u0026ntilde;u (2016). Among rosids and, in particular, in the order Sapindales, this type of obdiplostemony is quite common (Ronse De Craene and Bull-Here\u0026ntilde;u 2016; Ronse De Craene and Smets 1995). Obdiplostemony is considered a manifestation of certain trends in flower evolution. On the basis of obdiplostemony, synorganization without organ fusion can occur, for example, in \u003cem\u003eGeranium robertianum\u003c/em\u003e L. (Endress \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). However, in most cases, it is considered a consequence of the reduction of antipetalous stamens. According to such ideas, secondary obdiplostemony is a transitional state from diplostemony to haplostemony (Ronse De Craene and Bull-Here\u0026ntilde;u 2016; Ronse De Craene and Smets 1995). The case of \u003cem\u003eA. nikoense\u003c/em\u003e is interesting because the opposite direction of structural transformations can be proposed here. It is even more intriguing because two-whorled androecium is isomerous to the perianth. The androecium of the typical structure for most maples is rather single whorled, at least superficially in the mature flowers. In \u003cem\u003eA. nikoense\u003c/em\u003e, two distinct whorls are formed, but in a configuration indicating a lack of space on the floral meristem. This spatial constraint is also evidenced by the initiation of all antipetalous stamens from common primordia. \u003cem\u003eA. nikoense\u003c/em\u003e may also be important in understanding the origin of the typical flower of Sapindaceae. The case of \u003cem\u003eA. nikoense\u003c/em\u003e can be seen as a re-gain of isomery or as a reversion to an ancestral state for the entire family. This is especially interesting because the androecium of the Biebersteiniaceae, the closest relatives of the Sapindaeaceae, is presumably obdiplostemonous (Joyce et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Yamamoto et al. \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"CONCLUSIONS","content":"\u003cp\u003eThe structure and development of typical maple flowers do not show strong differences from the other Sapindaceae. The sepals are initiated in a spiral, and the petals are delayed in their development. Species of the genus \u003cem\u003eAcer\u003c/em\u003e demonstrate different patterns of stamen initiation, even those with eight stamens. The most important aspect of morphogenetic evolution in the genus \u003cem\u003eAcer\u003c/em\u003e is the change in the number of common petal-stamen primordia compared to other Sapindaceae. These structures affect floral morphogenesis, but not the construction of the mature flower. The topological identity of individual stamens and the entire androecium can vary significantly in different evolutionary lineages of the genus. Our study raises many questions about the criteria for delineating whorls of the androecium, the reasons for the appearance and functions of common primordia, and the limitations of developmental data as a criterion of homology. The hypotheses (related to the questions listed above) we propose in this article should be tested in the future on a wider sampling of representatives of the family Sapindaceae in particular and angiosperms in general.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eThe authors have no competing interests to declare that are relevant to the content of this article.\u003c/p\u003e\u003ch2\u003eFUNDING\u003c/h2\u003e \u003cp\u003eThe research was supported by a budgetary subsidy to the Lomonosov Moscow State University (No. 121032500084-6).\u003c/p\u003e\u003ch2\u003eAUTHOR CONTRIBUTIONS\u003c/h2\u003e \u003cp\u003eConceptualization: A.E. Zavialov, M.V. Remizowa; Material preparation, data collection and analysis: A.E. Zavialov, M.V. Remizowa; Writing - original draft preparation: A.E. Zavialov; Writing - review and editing: M.V. Remizowa; Supervision: M.V. Remizowa.\u003c/p\u003e\u003ch2\u003eACKNOWLEDGEMENTS\u003c/h2\u003e \u003cp\u003eWe are grateful to to G.A. Boyko for providing specimens of \u003cem\u003eA. nikoense\u003c/em\u003e from the Botanical Garden of Moscow State University, to G.A. Firsov for providing specimens of \u003cem\u003eA. spicatum\u003c/em\u003e and \u003cem\u003eA. tegmentosum\u003c/em\u003e from the Saint Petersburg Botanical Garden, to T.E. Kramina, A.D. Lisitsina, A.O. Astashkin for help in collecting material, to A.C. Timonin for helpful discussions and suggestions. SEM studies were carried out at the Shared Research Facility \u0026ldquo;Electron microscopy in life sciences\u0026rdquo; at Moscow State University (Unique Equipment \u0026ldquo;Three-dimensional electron microscopy and spectroscopy\u0026rdquo;).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAcevedo-Rodr\u0026iacute;guez P, Van Welzen PC, Adema F, Van Der Ham RWJM (2010) Sapindaceae. In: Kubitzki K (ed) Flowering plants. Eudicots: Sapindales, cucurbitales, myrtaceae. Springer, Berlin, pp 357\u0026ndash;407.\u003c/li\u003e\n \u003cli\u003eAPG IV (2016) An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG IV. Bot J Linn Soc 181:1\u0026ndash;20.\u003c/li\u003e\n \u003cli\u003eArber A (1946) Goethe\u0026rsquo;s botany. Chronica botanica 10:67\u0026ndash;124.\u003c/li\u003e\n \u003cli\u003eAreces-Berazain F, Hinsinger DD, Strijk JS (2021) Genome-wide supermatrix analyses of maples (Acer, Sapindaceae) reveal recurring inter-continental migration, mass extinction, and rapid lineage divergence. Genomics 113:681\u0026ndash;692.\u003c/li\u003e\n \u003cli\u003eBarykina RP, Veselova TD, Deviatov AG, Djalilova HH, Iljina GM, Chubatova NV (2004) Handbook of the botanical microtechniques. Moscow University Press, Moscow.\u003c/li\u003e\n \u003cli\u003eBrockington S, Dos Santos P, Glover B, De Craene LR (2013) Androecial evolution in Caryophyllales in light of a paraphyletic Molluginaceae. American J of Botany 100:1757\u0026ndash;1778.\u003c/li\u003e\n \u003cli\u003eBuchenau F (1861) Morphologische Bemerkungen \u0026uuml;ber einige Acerineen. Botanische Zeitung 19:265\u0026ndash;269.\u003c/li\u003e\n \u003cli\u003eBuerki S, Callmander MW, Acevedo‐Rodriguez P, et al. (2021) An updated infra‐familial classification of Sapindaceae based on targeted enrichment data. American J of Botany 108:1234\u0026ndash;1251.\u003c/li\u003e\n \u003cli\u003eBuerki S, Lowry Ii PP, Alvarez N, Razafimandimbison SG, K\u0026uuml;pfer P, Callmander MW (2010) Phylogeny and circumscription of Sapindaceae revisited: molecular sequence data, morphology and biogeography support recognition of a new family, Xanthoceraceae. Plecevo 143:148\u0026ndash;159.\u003c/li\u003e\n \u003cli\u003eBukhari G, Zhang J, Stevens PF, Zhang W (2017) Evolution of the process underlying floral zygomorphy development in pentapetalous angiosperms. American J of Botany 104:1846\u0026ndash;1856.\u003c/li\u003e\n \u003cli\u003eBull\u0026ndash;Here\u0026ntilde;u K, Dos Santos P, Toni JFG, et al. (2022) Mechanical Forces in Floral Development. Plants 11:661.\u003c/li\u003e\n \u003cli\u003eCao L, Liu J, Lin Q, Ronse De Craene LP (2018) The floral organogenesis of Koelreuteria bipinnata and its variety K. bipinnata var. integrifolia (Sapindaceae): evidence of floral constraints on the evolution of monosymmetry. Plant Syst Evol 304:923\u0026ndash;935.\u003c/li\u003e\n \u003cli\u003eCao LM, Ronse De Craene LP, Wang Z-X, Wang Y-H (2017) The floral organogenesis of Eurycorymbus cavaleriei (Sapindaceae) and its systematic implications. Phytotaxa 297:234.\u003c/li\u003e\n \u003cli\u003eCao LM, Xia N (2009) Floral organogenesis of Delavaya toxocarpa (Sapindaceae; Sapindales). J of Sytematics Evolution 47:237\u0026ndash;244.\u003c/li\u003e\n \u003cli\u003eCao LM, Xia N-H, Deng Y-F (2006) Floral organogenesis of Handeliodendron bodinieri (Sapindaceae) and its systematic implications. Acta Phytotaxonomica Sinica 44:393.\u003c/li\u003e\n \u003cli\u003eChoob VV (2010) The role of positional information in regulation of development of floral organs and leaf series of shoots. Binom, Moscow.\u003c/li\u003e\n \u003cli\u003eChoob VV, Penin AA (2004) Structure of flower in Arabidopsis thaliana: spatial pattern formation. Russian Journal of Developmental Biology 35:224\u0026ndash;227.\u003c/li\u003e\n \u003cli\u003eChoob VV, Yurtseva OV (2007) Mathematical modeling of flower structure in the family Polygonaceae. Bot Zhurn 92:114\u0026ndash;134.\u003c/li\u003e\n \u003cli\u003eDe Jong PC (1976) Flowering and sex expression in Acer L.: a biosystematic study. Wageningen: H. Veenman \u0026amp; Zonen B.V.\u003c/li\u003e\n \u003cli\u003eDe Jong PC (2004) World maple diversity. Westonbirt Arboretum and Royal Agricultural College in Gloucestershire, England, pp 2\u0026ndash;11.\u003c/li\u003e\n \u003cli\u003eDe Lima Silva A, Trov\u0026oacute; M, St\u0026uuml;tzel T, Rudall PJ, Sajo MDG, Coan AI (2021) Floral development and vasculature in Eriocaulon (Eriocaulaceae) provide insights into the evolution of Poales. Annals of Botany 128:605\u0026ndash;626.\u003c/li\u003e\n \u003cli\u003eEichler AW (1878) Bl\u0026uuml;thendiagramme. Verlag von Wilhelm Engelmann, Leipzig.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eEndress PK (1995) Major evolutionary traits of monocot flowers. Monocotyledons: systematics and evolution. Royal Botanic Gardens, London, pp 43\u0026ndash;79.\u003c/li\u003e\n \u003cli\u003eEndress PK (1999) Symmetry in Flowers: Diversity and Evolution. International Journal of Plant Sciences 160:S3\u0026ndash;S23.\u003c/li\u003e\n \u003cli\u003eEndress PK (2001) Evolution of floral symmetry. Current Opinion in Plant Biology 4:86\u0026ndash;91.\u003c/li\u003e\n \u003cli\u003eEndress PK (2010) Synorganisation without organ fusion in the flowers of Geranium robertianum (Geraniaceae) and its not so trivial obdiplostemony. Annals of Botany 106:687\u0026ndash;695.\u003c/li\u003e\n \u003cli\u003eEndress PK (2012) The Immense Diversity of Floral Monosymmetry and Asymmetry Across Angiosperms. Bot Rev 78:345\u0026ndash;397.\u003c/li\u003e\n \u003cli\u003eHaeckel EA (1866) Generelle Morphologie der Organismen: Allgemeine Grundz\u0026uuml;ge der Organischen Formen-Wissenschaft, Mechanisch Begr\u0026uuml;ndet Durch die von Charles Darwin Reformirte Descendenz-Theorie. Druck und Verlag von Georg Reimer, Berlin.\u003c/li\u003e\n \u003cli\u003eHall BA (1951) Tht floral anatomy of the genus Acer. American J of Botany 38:793\u0026ndash;799.\u003c/li\u003e\n \u003cli\u003eHarrington MG, Edwards KJ, Johnson SA, Chase MW, Gadek PA (2005) Phylogenetic Inference in Sapindaceae sensu lato Using Plastid matK and rbcL DNA Sequences.\u0026nbsp;Systematic Botany 30:366\u0026ndash;382.\u003c/li\u003e\n \u003cli\u003eHufford L (2003) Homology and Developmental Transformation: Models for the Origins of the Staminodes of Loasaceae Subfamily Loasoideae. International Journal of Plant Sciences 164:S409\u0026ndash;S439.\u003c/li\u003e\n \u003cli\u003eJeune B, Sattler R (1992) Multivariate analysis in process morphology of plants. Journal of Theoretical Biology 156:147\u0026ndash;167.\u003c/li\u003e\n \u003cli\u003eJoyce EM, Appelhans MS, Buerki S, et al. (2023) Phylogenomic analyses of Sapindales support new family relationships, rapid Mid-Cretaceous Hothouse diversification, and heterogeneous histories of gene duplication. Front Plant Sci 14:1063174.\u003c/li\u003e\n \u003cli\u003eLi J, Stukel M, Bussies P, et al. (2019) Maple phylogeny and biogeography inferred from phylogenomic data. J of Sytematics Evolution 57:594\u0026ndash;606.\u003c/li\u003e\n \u003cli\u003eMcFadden J (2023) Razor sharp: The role of Occam\u0026rsquo;s razor in science. Annals of the New York Academy of Sciences 1530:8\u0026ndash;17.\u003c/li\u003e\n \u003cli\u003eMelo LRF, Vasconcelos TNC, Caetano APS, Brito VLG de (2022) Stamen Diversity in Melastomataceae. Morphology, Color, and Function. Systematics, Evolution, and Ecology of Melastomataceae. Springer International Publishing, Cham, 609\u0026ndash;628.\u003c/li\u003e\n \u003cli\u003eNuraliev MS, Sokoloff DD, Karpunina PV, Oskolski AA (2019) Patterns of Diversity of Floral Symmetry in Angiosperms: A Case Study of the Order Apiales. Symmetry 11:473.\u003c/li\u003e\n \u003cli\u003ePayer J-B (1857) Trait\u0026eacute; de organog\u0026eacute;nie compar\u0026eacute;e de la fleur. V. Masson, Paris.\u003c/li\u003e\n \u003cli\u003eRemizowa MV, Rudall PJ, Choob VV, Sokoloff DD (2013) Racemose inflorescences of monocots: structural and morphogenetic interaction at the flower/inflorescence level. Annals of Botany 112:1553\u0026ndash;1566.\u003c/li\u003e\n \u003cli\u003eRemizowa MV, Shipunov AB, Sokoloff DD (2023) When asymmetry mimics zygomorphy: flower development in Chamaelirium japonicum (Melanthiaceae, Liliales). Botanica Pacifica 12:3\u0026ndash;14.\u003c/li\u003e\n \u003cli\u003eRemizowa M, Sokoloff D, Rudall P (2006) Patterns of Floral Structure and Orientation in Japonolirion , Narthecium, and Tofieldia. aliso 22:159\u0026ndash;171.\u003c/li\u003e\n \u003cli\u003eRonse De Craene L (2016) Meristic changes in flowering plants: How flowers play with numbers. Flora 221:22\u0026ndash;37.\u003c/li\u003e\n \u003cli\u003eRonse De Craene L (2018) Understanding the role of floral development in the evolution of angiosperm flowers: clarifications from a historical and physico-dynamic perspective. J Plant Res 131:367\u0026ndash;393.\u003c/li\u003e\n \u003cli\u003eRonse De Craene L P (2022) Floral diagrams: an aid to understanding flower morphology and evolution Second Edition. Cambridge University Press, New York.\u003c/li\u003e\n \u003cli\u003eRonse De Craene LP (2024) The interaction between heterochrony and mechanical forces as main driver of floral evolution. J Plant Res. doi:10.1007/s10265-024-01526-3.\u003c/li\u003e\n \u003cli\u003eRonse De Craene L, Bull-Here\u0026ntilde;u K (2016) Obdiplostemony: the occurrence of a transitional stage linking robust flower configurations. Ann Bot 117:709\u0026ndash;724.\u003c/li\u003e\n \u003cli\u003eRonse De Craene L, Clinckemaillie D, Smets E (1993) Stamen-Petal Complexes in Magnoliatae. Bulletin du Jardin botanique national de Belgique / Bulletin van de National Plantentuin van Belgi\u0026euml; 62:97\u0026ndash;112.\u003c/li\u003e\n \u003cli\u003eRonse De Craene LP, Smets EF (1993) D\u0026eacute;doublement revisited: towards a renewed interpretation of the androecium of the Magnoliophytina. Botanical Journal of the Linnean Society 113:103\u0026ndash;124.\u003c/li\u003e\n \u003cli\u003eRonse De Craene LP, Smets EF (1995) The distribution and systematic relevance of the androecial character oligomery. Botanical Journal of the Linnean Society 118:193\u0026ndash;247.\u003c/li\u003e\n \u003cli\u003eRonse De Craene L, Smets E, Clinckemaillie D (2000) Floral ontogeny and anatomy in Koelreuteria with special emphasis on monosymmetry and septal cavities. Pl Syst Evol 223:91\u0026ndash;107.\u003c/li\u003e\n \u003cli\u003eRonse De Craene LP, Wei L (2019) Floral development and anatomy of Macarthuria australis (Macarthuriaceae): key to understanding the unusual initiation sequence of Caryophyllales. Aust Syst Bot 32:49\u0026ndash;60.\u003c/li\u003e\n \u003cli\u003eRudall PJ, Cunniff J, Wilkin P, Caddick LR (2005) Evolution of dimery, pentamery and the monocarpellary condition in the monocot family Stemonaceae (Pandanales). TAXON 54:701\u0026ndash;711.\u003c/li\u003e\n \u003cli\u003eRutishauser R, Moline P (2005) Evo-devo and the search for homology (\u0026ldquo;sameness\u0026rdquo;) in biological systems. Theory Biosci 124:213\u0026ndash;241.\u003c/li\u003e\n \u003cli\u003eSattler R (1992) Process morphology: structural dynamics in development and evolution. Can J Bot 70:708\u0026ndash;714.\u003c/li\u003e\n \u003cli\u003eSattler R (1996) Classical Morphology and Continuum Morphology: Opposition and Continuum. Annals of Botany 78:577\u0026ndash;581.\u003c/li\u003e\n \u003cli\u003eSattler R, Rutishauser R (1997) The Fundamental Relevance of Morphology and Morphogenesis to Plant Research. Annals of Botany 80:571\u0026ndash;582.\u003c/li\u003e\n \u003cli\u003eSattler R, Rutishauser R (2022) Fundamentals of Plant Morphology and Plant Evo-Devo (Evolutionary Developmental Morphology). Plants 12:118.\u003c/li\u003e\n \u003cli\u003eSokoloff DD, Yadav SR, Chandore AN, Remizowa MV (2020) Stability Despite Reduction: Flower Structure, Patterns of Receptacle Elongation and Organ Fusion in Eriocaulon (Eriocaulaceae: Poales). Plants 9:14\u0026ndash;24.\u003c/li\u003e\n \u003cli\u003eTimonin AC (1993) Why we are inclined to reveal mainly reductions in evolutionary morphological reconstructions. Zhurnal obshchei biologii 54:104\u0026ndash;114.\u003c/li\u003e\n \u003cli\u003eTimonin AC (2002) Sattler\u0026rsquo;s dynamic morphology: an acme or a reverie. Wulfenia 9:9\u0026ndash;18.\u003c/li\u003e\n \u003cli\u003eTucker SC (1996) Trends in evolution of floral ontogeny in Cassia sensu stricto, Senna , and Chamaecrista (Leguminosae: Caesalpinioideae: Cassieae: Cassiinae); a study in convergence. American J of Botany 83:687\u0026ndash;711.\u003c/li\u003e\n \u003cli\u003eVislobokov N, Sokoloff D, Degtjareva G, Valiejo-Roman C, Kuznetsov A, Nuraliev MS (2014) Aspidistra paucitepala (Asparagaceae), a new species with occurrence of the lowest tepal number in flowers of Asparagales. Phytotaxa 161:270\u0026ndash;282.\u003c/li\u003e\n \u003cli\u003eWalsh D (1979) Occam\u0026rsquo;s Razor: A Principle of Intellectual Elegance. American Philosophical Quarterly 16:241\u0026ndash;244.\u003c/li\u003e\n \u003cli\u003eYamamoto T, Vassiliades DD, Tobe H (2014) Embryology of Biebersteinia (Biebersteiniaceae, Sapindales): characteristics and comparisons with related families. J Plant Res 127:599\u0026ndash;615.\u003c/li\u003e\n \u003cli\u003eZavialov AE (2024) Patterns of flower structure and development in some species of the genus Acer (Sapindaceae). Bachelor dissertation, Moscow State University.\u003c/li\u003e\n \u003cli\u003eZavialov AE, Remizowa MV (2024) Unstable patterns of flower structure and development in the almost dioecious species Acer negundo (Sapindaceae, Sapindales). Botanical Journal of the Linnean Society 204:267\u0026ndash;281.\u003c/li\u003e\n \u003cli\u003eZhang Y, Cao L, Ronse De Craene LP (2022) Inflorescence and floral development of Dipteronia sinensis (Sapindaceae): Demonstration of developmental constraints in the shaping of the flower. Flora 291:152069.\u003c/li\u003e\n \u003cli\u003eZhou Q, Cai Q, Zheng Y, Wu Z, Mao J (2019) Floral development and the formation of functionally unisexual flowers in Xanthoceras sorbifolium (Sapindaceae), a morphologically andromonoecious tree endemic to northern China. Trees 33:1571\u0026ndash;1582.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"journal-of-plant-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jpre","sideBox":"Learn more about [Journal of Plant Research](http://link.springer.com/journal/10265)","snPcode":"10265","submissionUrl":"https://www.editorialmanager.com/jpre/default2.aspx","title":"Journal of Plant Research","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"common petal-stamen primordium, stamen loss, dynamic morphology, Acer, Sapindaceae, Sapindales","lastPublishedDoi":"10.21203/rs.3.rs-4754778/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4754778/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe genus \u003cem\u003eAcer\u003c/em\u003e belongs to the family Sapindaceae, whose representatives are characterized by a pentamerous perianth but typically possess only eight stamens. Such an androecium is believed to have evolved through the loss of two stamens. However, there is still no consensus on the origin of eight-staminate androecium including the positions of the two lost stamens and the pathway of their reduction compared to other Sapindaceae. We examined the early stages of flower development in five maple species belonging to different sections \u0026ndash; four species with eight stamens and one species with ten stamens \u0026ndash; using scanning electron microscopy. Measurements were performed to analyze the relative positions of stamen primordia, their size, and the floral meristem surface area. In addition, the perianth and androecium vasculature was studied to reveal petal-stamen complexes. We found that in three of four 8-staminate species, three stamens are initiated from common petal-stamen primordia, and five arise from single primordia. In \u003cem\u003eA. tegmentosum\u003c/em\u003e Maxim., four stamens appear from common primordia with petals, and four from single primordia. Despite developmental differences, stamen distribution within the flower and the angles between adjacent stamens indicate a similar androecium construction in all species. In most species with eight stamens, the differences between two andoecial whorls are vanished. In contrast, \u003cem\u003eA. nikoense\u003c/em\u003e (Miq.) Maxim., with ten stamens, possesses two distinct stamen whorls, the antipetalous stamens are initiated from common primordia. In the 8-staminate androecia of the genus \u003cem\u003eAcer\u003c/em\u003e, the same two stamens have been lost as in other Sapindaceae. Within genus \u003cem\u003eAcer\u003c/em\u003e, there is a certain decrease in the relative size of the floral meristem, accompanied by an increase in the number of common petal-stamen primordia and increased heterogeneity of the androecium (in \u003cem\u003eA. tegmentosum\u003c/em\u003e) or reduction of some floral organs.\u003c/p\u003e","manuscriptTitle":"Androecium homologies in eight-staminate maples: a developmental study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-14 08:13:06","doi":"10.21203/rs.3.rs-4754778/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2024-07-22T04:04:56+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-07-20T10:09:46+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-07-18T11:21:33+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Plant Research","date":"2024-07-17T04:01:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-plant-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jpre","sideBox":"Learn more about [Journal of Plant Research](http://link.springer.com/journal/10265)","snPcode":"10265","submissionUrl":"https://www.editorialmanager.com/jpre/default2.aspx","title":"Journal of Plant Research","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"037c3772-07c4-47ac-bf1d-59a24085a24c","owner":[],"postedDate":"August 14th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-04-28T16:03:39+00:00","versionOfRecord":{"articleIdentity":"rs-4754778","link":"https://doi.org/10.1007/s10265-025-01641-9","journal":{"identity":"journal-of-plant-research","isVorOnly":false,"title":"Journal of Plant Research"},"publishedOn":"2025-04-25 15:58:10","publishedOnDateReadable":"April 25th, 2025"},"versionCreatedAt":"2024-08-14 08:13:06","video":"","vorDoi":"10.1007/s10265-025-01641-9","vorDoiUrl":"https://doi.org/10.1007/s10265-025-01641-9","workflowStages":[]},"version":"v1","identity":"rs-4754778","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4754778","identity":"rs-4754778","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

Ask this paper AI returns verbatim quotes from the full text · source: preprint-html

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2024) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.

Source provenance

europepmc
last seen: 2026-05-20T01:45:00.602351+00:00