From fruit to young cactus. How Rhipsalis colonises tree crowns

A fruit above the forest floor

Small fleshy fruits ripen along the hanging stems of Rhipsalis. A cactus growing on a tree uses the trunk and branches for support. Its seeds must reach a place where a young plant can remain attached and obtain water. Distance from the parent is therefore one of several dimensions of effective dispersal. The surface on which a seed is deposited, its illumination and its pattern of drying also matter. These connected requirements provide a way to relate observations of birds to germination experiments [Guaraldo et al. 2013] [Gonçalo et al. 2026].

Guaraldo and colleagues connected fruit biology in five species with observations in the Brazilian Atlantic Forest. The fruits were several millimetres across, with watery pulp containing little fat or protein. Small seeds were embedded in a mucilaginous material. White, pink and orange fruits provided food for several kinds of birds. For the plant, the consequential part of this interaction was the subsequent route of its seeds, including passage through the digestive tract and the way in which droppings were deposited [Guaraldo et al. 2013].

Fruit retention can extend the period during which food remains available. In a field experiment, Rhipsalis teres fruits protected from consumption remained on plants for many months, occasionally for as long as a year. The authors linked this persistence to the opportunity to await infrequent disperser visits. Protection changed the fruits’ availability to animals, however, so the result describes potential persistence under the experimental conditions. A persistent fruit still depends on a suitable consumer and on the destination of its seeds [Guaraldo et al. 2013].

Feeding behaviour and the destination of seeds

Species of Euphonia played a prominent role among birds visiting the fruits. They accounted for a substantial share of visits in observations combined with published records. Captive feeding trials revealed behaviour particularly relevant to an epiphyte. These birds rubbed the cloacal region against a perch, leaving seeds embedded in sticky material on its surface. Mucilage enabled the seeds to adhere. Under the same conditions, birds belonging to other tested genera more often dropped their faeces onto the cage floor. The difference concerned the destination of seeds as well as fruit consumption itself [Guaraldo et al. 2013].

Gut passage also altered germination. In Rhipsalis baccifera, almost all tested seeds germinated after passage through birds of the genera Euphonia and Tangara. The average was substantially lower in the control, from which pulp had been removed by hand. The response depended on the particular cactus and bird combination. Seeds of Rhipsalis pachyptera recovered from Euphonia droppings germinated better than controls, whereas those passed by the tested Turdus thrushes performed worse. Fruit consumption thus combines transport with an effect on the seed itself [Guaraldo et al. 2013].

In Rhipsalis teres, final germination following Euphonia gut passage remained similar to the control. This separates the effectiveness of seed transport from digestive effects on seed physiology. A bird can deliver a seed to a branch even when it does not increase the number germinating in a dish. Conversely, a high laboratory percentage leaves the question of survival on a tree unresolved. Captive observations support a mechanism directing seeds onto supports, while the contribution of individual bird species to wild recruitment requires following the subsequent fate of seedlings [Guaraldo et al. 2013].

Fruit scent and the limits of a hypothesis

Rhipsalis juengeri broadens the picture of interactions between fruits and animals. Schlumpberger and colleagues described a conspicuous scent resembling blackcurrants. They analysed volatile substances released by fruits of one clone cultivated at the Bonn botanical garden. Ketones dominated the mixture, and chromatography combined with sensory assessment identified compounds contributing to its aroma. This was a measurement of a fruit property. Comparison with other cultivated representatives of the group highlighted the unusually strong scent of the material examined [Schlumpberger et al. 2006].

Fruit colour, position on hanging stems and strong scent led the authors to propose dispersal by bats. That hypothesis remains distinct from the chemical result. The study did not observe bats eating the fruits or seeds transported by these mammals. Confirmation would require evidence from natural populations. This case explains why a combination of fruit traits can suggest a likely animal partner, while the actual dispersal process is established by observing animals and tracking the subsequent fate of seeds [Schlumpberger et al. 2006].

Light triggers germination

Once released from a fruit, a seed responds to its microsite. Lone and colleagues compared germination in Rhipsalis floccosa, Rhipsalis pilocarpa and Rhipsalis teres. In a separate trial, seeds were kept in darkness at 20 °C for 30 days. None of the three species germinated under those conditions, whereas germination occurred in illuminated treatments. This response is called positive photoblastism, meaning that light promotes germination. The result applies to the seed batches and observation period used in the experiment [Lone et al. 2016].

Light as a germination signal must be distinguished from a seedling’s tolerance of intense exposure. Comparing illumination with darkness does not establish a safe radiation level on an exposed branch. It does indicate that deep burial may remove a necessary signal. For sowing these three species, this supports allowing light to reach the seed surface. Moisture is needed at the same time, because illumination cannot replace the water absorbed as the embryo begins to grow. A useful sowing environment must satisfy both requirements together [Lone et al. 2016] [Gonçalo et al. 2026].

Temperature changes both the amount and timing of germination

In the study by Lone and colleagues, Rhipsalis floccosa achieved its highest germination at 20 °C. Temperatures of 15 and 20 °C were favourable for Rhipsalis pilocarpa and Rhipsalis teres. Constant temperatures of 30 and 35 °C reduced performance, and Rhipsalis teres did not germinate at either temperature. The researchers also measured the timing of root emergence. Temperature could therefore alter both the final number germinating and the pace of the process. A high percentage and an early start describe different properties of a seed batch [Lone et al. 2016].

Gonçalo and colleagues extended the comparison to eight species. In the article assigned to the 2026 volume, most seed batches performed best at moderate temperatures of 20–25 °C. Observations continued for 60 days, after which seeds without an emerged root were transferred to conditions favourable for germination. In some species, growth resumed when stress was removed. Inhibition can therefore leave a living seed available for a later favourable period, although the ability to recover germination differed among the species tested [Gonçalo et al. 2026].

Differences between the 2016 and 2026 experiments help place the numbers in context. The researchers used different seed batches, observation periods and partially different sets of species. In the newer study, Rhipsalis floccosa was represented by its nominotypical subspecies. Performance at a particular temperature remains tied to that sample. For cultivation, the shared implication is the value of moderate warmth when sowing the species examined. Germination temperatures cannot be converted into winter minimum temperatures for adult plants, which these experiments did not investigate [Lone et al. 2016] [Gonçalo et al. 2026].

A seed can survive conditions that threaten a seedling

In the newer study, water availability was reduced with PEG 6000 solutions. These lowered osmotic potential, making water uptake more difficult for seeds. This series was conducted at 25 °C, separately from the temperature experiments. Germination generally declined as the restriction increased. At the strongest restrictions used, none of the species began to germinate. Some seeds germinated after transfer to water, revealing a distinction between temporary inhibition and loss of viability. The recovery phase was therefore essential for interpreting an initially empty germination dish [Gonçalo et al. 2026].

Root emergence begins a stage with different requirements. The researchers counted living and dead seedlings separately, allowing germination to be distinguished from survival. At −0.4 MPa, seeds of Rhipsalis dissimilis germinated, but the seedlings produced in that treatment did not survive. A similar outcome occurred in Rhipsalis floccosa subsp. floccosa. Seeds that remained inhibited could subsequently produce living seedlings after conditions improved. Starting growth early is therefore advantageous only when the environment can sustain the next developmental stage [Gonçalo et al. 2026].

The osmotic experiment separates physiological responses, but it does not reproduce every feature of a tree crown. In nature, moisture changes after rain, branches warm unevenly, and a thin layer of organic material dries differently from filter paper in a dish. Experimental potential values are not a ready-made prescription for moisture in a pot. They provide a strong basis for identifying a vulnerable stage, while its importance in a forest must be assessed alongside microsite measurements and observations of actual seedlings [Gonçalo et al. 2026].

A tree contains many different places to live

The distribution of adult plants retains the effects of many earlier events. Guaraldo and colleagues found clustered distributions of the investigated Rhipsalis and more frequent occupation of larger trees. In one forest, Rhipsalis teres occurred more often on smooth-barked supports than their availability would predict. This observation illustrates the importance of local context. Assigning one ideal surface to all epiphytes overlooks bird behaviour, tree characteristics and plant developmental history. The associations identify useful questions for further investigation rather than a complete mechanism of host selection [Guaraldo et al. 2013].

After germination begins, seedling survival and further development require separate observations. Changes in the young plant’s structure are also discussed in the article on the development of epiphytic cacti. Dispersal determines where this history begins, while germination conditions restrict the number of plants that can enter it. Counting fruits, seeds, newly emerged roots and surviving seedlings therefore addresses successive questions. Bringing these stages together is what makes it possible to identify where a species encounters its strongest limitation to recruitment [Guaraldo et al. 2013] [Gonçalo et al. 2026].

Conservation includes the entire recruitment process

Species identification affects how a plant’s situation in the forest is assessed. The 2025 revision of Rhipsalis in Minas Gerais corrected identifications of some herbarium specimens and documented additional occurrence of Rhipsalis neves-armondii. The authors also identified habitat-loss pressures and species poorly represented within protected areas. Such records establish the geographical context for recruitment research. An incorrect specimen name can lead to germination results or dispersal information being assigned to populations belonging to a different species [Tortorelli et al. 2025].

Described in 2026, Rhipsalis boanova was known to its authors from a single plant discovered in Bahia. Its fruits contained seeds, but a description of fruiting does not establish how often the species recruits in nature. With such limited material, documenting provenance, maintaining living resources and understanding seedling requirements become particularly consequential. Protecting trees preserves growing sites, protecting dispersers maintains opportunities for transport, and retaining suitable microclimates gives young plants a chance to survive after germination. Each element corresponds to a different stage of the same process [Taylor et al. 2026a] [Guaraldo et al. 2013] [Gonçalo et al. 2026].

Related plant cards