The chance discovery of an unusual Ficus specimen near Katherine in the Northern Territory prompted an investigation into hybridisation between two morphologically distinct endemic Australian sandpaper figs, Ficus aculeata A.Cunn. ex Miq. and F. coronulata Miq. In this study, genome-wide scans and morphological measurements were used to investigate the perceived hybridisation by using herbarium and freshly collected samples. Most of the putative hybrids displayed a wide variety of intermediate morphology and some individuals had characteristics consistent with the description of a third species, F. carpentariensis D.J.Dixon. Both genomic and morphometric results provided evidence of naturally occurring hybridisation events within Ficus. Additionally, the findings from this study showed possible taxonomic issues within the Northern Australian sandpaper figs that warrant further investigation.

The creek sandpaper fig of southeastern Australia, Ficus coronata Spin, is culturally significant to Australian traditional owners who made use of the leaves to smooth timber and ate the fruit. The species is thought to have a long history on the continent, with some suggesting a Gondwanan origin. However, distributional patterns and overall ecology suggest a recent expansion across suitable habitats. We used landscape genomic techniques and environmental niche modelling to reconstruct its history and explore whether the species underwent a recent and rapid expansion along the east coast of New South Wales. Genomic analysis of 178 specimens collected from 32 populations throughout the species’ New South Wales distribution revealed a lack of genetic diversity and population structure. Some populations at the species’ southern and western range limits displayed unexpected diversity, which appears to be the result of allele surfing. Field work and genetic evidence suggest a Holocene expansion which may have increased since European colonisation. We also present a novel method for detecting allele surfing—MAHF (minor allele at highest frequency).

A new species of lithophytic fig, Ficus desertorum B.C.Wilde & R.L.Barrett, endemic to arid Central Australia, is described and illustrated. It is distinguished from other species in Ficus section Malvanthera Corner by having stiff lanceolate, dark green, discolorous leaves; many parallel, often obscure lateral veins; petioles that are continuous with the midrib; with minute, usually white hairs and non- or slightly sunken intercostal regions on the lower surface. Previously included under broad concepts of either Ficus platypoda (Miq.) Miq. or Ficus brachypoda (Miq.) Miq., this species has a scattered distribution throughout Central Australia on rocky outcrops, jump-ups (mesas) and around waterholes. This culturally significant plant, colloquially referred to as the desert fig, grows on elevated landscapes in central Australia, including Uluru (Ayers Rock), Kata Tjuta (The Olgas) and Karlu Karlu (Devils Marbles), three of Central Australia’s best-known natural landmarks. Evidence is provided to show these plants are geographically and morphologically distinct from Ficus brachypoda, justifying the recognition of F. desertorum as a new species. Taxonomic issues with F. brachypoda and F. atricha D.J.Dixon are also discussed. Lectotypes are selected for Urostigma platypodum forma glabrior Miq. and Ficus platypoda var. minor Benth.

Premise Continental‐scale leaf trait studies can help explain how plants survive in different environments, but large data sets are costly to assemble at this scale. Automating the measurement of digitized herbarium collections could rapidly expand the data available to such studies. We used machine learning to identify and measure leaves from existing, digitized herbarium specimens. The process was developed, validated, and applied to analyses of relationships between leaf size and climate within and among species for two genera: Syzygium (Myrtaceae) and Ficus (Moraceae). Methods Convolutional neural network (CNN) models were used to detect and measure complete leaves in images. Predictions of a model trained with a set of 35 randomly selected images and a second model trained with 35 user‐selected images were compared using a set of 50 labeled validation images. The validated models were then applied to 1227 Syzygium and 2595 Ficus specimens digitized by the National Herbarium of New South Wales, Australia. Leaf area measurements were made for each genus and used to examine links between leaf size and climate. Results The user‐selected training method for Syzygium found more leaves (9347 vs. 8423) using fewer training masks (218 vs. 225), and found leaves with a greater range of sizes than the random image training method. Within each genus, leaf size was positively associated with temperature and rainfall, consistent with previous observations. However, within species, the associations between leaf size and environmental variables were weaker. Conclusions CNNs detected and measured leaves with levels of accuracy useful for trait extraction and analysis and illustrate the potential for machine learning of herbarium specimens to massively increase global leaf trait data sets. Within‐species relationships were weak, suggesting that population history and gene flow have a strong effect at this level. Herbarium specimens and machine learning could expand sampling of trait data within many species, offering new insights into trait evolution.