Morphometric methods based on artificial vision algorithms provide measurements for magnitudes descriptive of seed images (i.e., the length, width, area, and surface circularity index). Nevertheless, their results frequently omit the resemblance of the images to geometric figures that may be used as models. A complementary method based on the comparison of seed images with geometric models is applied to seeds of Vitis spp. The J index gives the percentage of similarity between a seed image and the model. Seven new geometric models are described based on the heart-shaped and piriform curves. Seeds of different species, subspecies and cultivars of Vitis adjust to different models. Models 1 and 3, the heart curve and the water drop, adjust better to seeds of V. amurensis, V. labrusca and V. rupestris than to V. vinifera. Model 6, the Fibonacci’s pear, adjusts well to seeds of V. vinifera, in general, and better to V. vinifera ssp. vinifera than to V. vinifera ssp. sylvestris. Seed morphology in species of Cissus and Parthenocissus, two relatives of Vitis in the Vitaceae, is also analysed. Geometric models are a tool for the description and identification of species and lower taxonomic levels complementing the results of morphometric analysis.

Fruit and seed shape are important characteristics in taxonomy providing information on ecological, nutritional, and developmental aspects, but their application requires quantification. We propose a method for seed shape quantification based on the comparison of the bi-dimensional images of the seeds with geometric figures. J index is the percent of similarity of a seed image with a figure taken as a model. Models in shape quantification include geometrical figures (circle, ellipse, oval…) and their derivatives, as well as other figures obtained as geometric representations of algebraic equations. The analysis is based on three sources: Published work, images available on the Internet, and seeds collected or stored in our collections. Some of the models here described are applied for the first time in seed morphology, like the superellipses, a group of bidimensional figures that represent well seed shape in species of the Calamoideae and Phoenix canariensis Hort. ex Chabaud. Oval models are proposed for Chamaedorea pauciflora Mart. and cardioid-based models for Trachycarpus fortunei (Hook.) H. Wendl. Diversity of seed shape in the Arecaceae makes this family a good model system to study the application of geometric models in morphology.

The Vitaceae Juss., in the basal lineages of Rosids, contains sixteen genera and 950 species, mainly of tropical lianas. The family has been divided in five tribes: Ampelopsideae, Cisseae, Cayratieae, Parthenocisseae and Viteae. Seed shape is variable in this family. Based on new models derived from equations representing heart and water drop curves, we describe seed shape in species of the Vitaceae. According to their similarity to geometric models, the seeds of the Vitaceae have been classified in ten groups. Three of them correspond to models before described and shared with the Arecaceae (lenses, superellipses and elongated water drops), while in the seven groups remaining, four correspond to general models (waterdrops, heart curves, elongated heart curves and other elongated models) and three adjust to the silhouettes of seeds in particular genera (heart curves of Cayratia and Pseudocayratia, heart curves of the Squared Heart Curve (SqHC) type of Ampelocissus and Ampelopsis and Elongated Superellipse-Heart Curves (ESHCs), frequent in Tetrastigma species and observed also in Cissus species and Rhoicissus rhomboidea). The utilities of the application of geometric models for seed description and shape quantification in this family are discussed.

Seed shape in the Malvaceae and other families of the order Malvales was investigated. Seed shape was quantified by comparison with the cardioid. The J index is the percent similarity between both images, the seed and the cardioid, and similarity is considered in cases where the J index is over 90. Seed shape was analysed in 73 genera, and seeds resembling the cardioid were found in 10 genera, eight in the Malvaceae and two in the Bixaceae and Cistaceae. Seed shape was quantified by comparison with the cardioid in 105 species. A correlation was found between the values of the J index and plant form, with higher values of the J index in the seeds of herbs, intermediate – in bushes, and lower values in trees. The results suggest a relationship between seed shape and plant form, where seeds resembling the cardioid are associated with plants having small size.

Datasets containing information on seed size have been published and are currently available. Nevertheless, there is a lack in the literature of a dataset dedicated to seed shape. We present a preliminary version for a dataset on seed morphology based on a comparison of seed shape with geometric figures. Similarity of the outline of seed images with geometric models is considered as a basis to classify seeds according to the geometric figures they resemble (e.g., ellipse, oval, cardioid). This allows, first, the classification of plant species according to their geometric type of seed, and second, seed shape quantification. For each seed image, the percent of similarity of their outline with a geometric figure can be calculated as a J index. Similarity in absolute terms is considered only when the J index >90. This criterion is important to avoid ambiguity and increase discrimination. The dataset opens the possibility of studying the relationship between seed shape and other variables such as seed size, genome complexity, life form or adaptive responses.