Nectar outside the flower. Ants on chollas

A gland among the spines

Small extrafloral nectaries occur on young growth of Cylindropuntia imbricata, developing as modified spines. The gland surface consists of epidermis, beneath which lie nectary and subnectary parenchyma. Vascular bundles occur in this deeper zone. The gland's surroundings include hairs and tissues containing mucilage and crystals. The areole thus combines defensive armature with secretion of a resource available to insects. [Ávila-Argáez et al. 2019]

The nectary's supply of water and organic compounds depends on the whole plant's condition. Anatomy indicates possible transport routes, while the ecological response is measured separately through visit numbers, visitor composition and insect behaviour. The anatomical study of C. imbricata was combined with a field water experiment. [Ávila-Argáez et al. 2019]

Water changes visitation

In San Luis Potosí, 20 adult plants were selected, half of which were watered once a week for 20 weeks. Ant inspections were conducted on young shoots. More ants visited watered cacti, especially the dominant Liometopum apiculatum. Crematogaster opuntiae was also found, and Pseudomyrmex pallidus occasionally. The authors could not quantify the amount of nectar secreted, so the principal result is a change in insect activity with greater water availability. [Ávila-Argáez et al. 2019]

A separate experiment on 40 plants covered a broader insect community. A total of 6286 individuals from 18 taxa was recorded, with 3137 at watered plants and 3149 in the control. Higher diversity in the watered treatment resulted from the distribution of taxon proportions at a similar total abundance. Visitor composition depended strongly on time of day. Observing presence on a stem measures spatial association. An insect's function, such as pollination or feeding, is established through additional observations. [Ramírez-Hernández et al. 2021]

Which ant protects the plant

In the Granite Mountains in the Mojave, C. echinocarpa was studied. In two seasons, 60 and 80 plants were selected, half provided with additional honey solution. Certain ant species responded particularly strongly, and community composition changed. The bugs Chelinidea vittiger and seeds in ripe fruits were counted in parallel. [Braun and Lortie 2025]

The presence of Crematogaster depilis was associated with fewer herbivores and more seeds per fruit. Solenopsis xyloni, which responded strongly to extra food, showed a different pattern. Supplementary feeding alone did not significantly increase seeds per fruit. Plant protection thus depends on partner identity, behaviour and the stage at which benefit is measured. [Braun and Lortie 2025]

Three levels of the result

Ecological comparisons become more precise when partner attraction, herbivore suppression and cactus reproduction are recorded separately. Extra food can change the first stage more strongly than the last. The experiment used a solution supplying both sugars and water, so their relative contributions to the response remain unresolved. [Braun and Lortie 2025]

The surroundings change the partnership

In Sevilleta, long-term observations showed that greater density of C. imbricata was accompanied by greater dominance of L. apiculatum in occupying plants. Traps in the surroundings allowed the broader extent of this effect to be tested. The diversity and basic composition of the entire ant community did not show an analogous restructuring. The result links the scale of an individual cactus with the community scale. [Donald and Miller 2022]

In Albuquerque and non-urban areas, tests used Drosophila melanogaster larvae placed beside cacti. Prey detection, attack and removal were measured, including after providing sucrose solution. Ants outside the city showed stronger defensive responses. These behavioural tests complement field observations, while long-term offspring production by plants remains a separate question. [Lynch and Savage 2023]