Monstera

The holes are made by cells dying on schedule inside a leaf that has not yet unrolled — and nobody has demonstrated what they are for.

Petiole, sheath and geniculum

Aroid usage differs from the rest of the monocots: what is called the petiole here takes in the sheath as well, everything from the stem to the base of the blade. The sheathing part protects the next leaf and the terminal bud, and how it behaves is one of the genus's best characters. It may be confined to the lower petiole or run all the way to the blade; it may end bluntly, or in wide auricles, or be drawn out into a ligule — in Monstera tuberculata the ligule is about as long as the petiole itself. After the next leaf expands, the sheath either stays green and persistent, withers in place, or falls away cleanly, and which of the three it does is often the quickest way to separate two similar species. The petiole is swollen at both ends; the upper swelling, at the base of the blade, is the geniculum, and it is usually conspicuously bent.

Leaves are produced in two ranks but seldom stay in them. They move after unfolding, fanning out so that upper leaves shade lower ones less. A cultivated Monstera deliciosa receiving light from one window was turned through 180°; over the next two weeks the leaves rearranged themselves into a less self-shading array, flexing at the pulvini even in leaves more than two years old.

How a hole is made

The mechanism was worked out in 1854 and has survived every re-examination since. Early in the development of the blade, while it is still furled, small patches of cells discolor and die, and the tissue they would have become never forms. The death starts in the mesophyll; if it starts late enough, the mesophyll dies but both epidermes survive, leaving a window rather than a hole. Anatomical sections published in 1969 already showed a discrete population of cells dying simultaneously while its immediate neighbors carried on dividing; a 2005 study on Monstera obliqua named the process programmed cell death and added that DNA cleavage comes early, the cell walls are left intact, and a small disc of dead tissue detaches as the expanding leaf stretches the opening. Exposed cells at the rim then differentiate into epidermis, which is why a mature hole has a finished edge rather than a torn one.

Timing decides the shape of the leaf. If the holes begin early enough they extend to the margin, and the leaf ends up pinnatifid instead of perforate; a second series started later can leave a blade that is both at once. That single variable accounts for most of the range of adult leaf shapes in the genus.

Why, and the honest answer

Four explanations have been offered and none tested in the field. In 1892 the holes were said to let rain fall through to the roots rather than be shed like an umbrella, an idea the 1977 revision called fanciful. That revision proposed instead that a dissected blade breaks up the still air against the leaf, increasing convection and keeping it cooler in strong sun — reasoning from torn banana leaves, which do run cooler, though banana is a full-sun pioneer and Monstera lives in shade. A 2006 suggestion added camouflage, which does not explain why the juvenile leaves of the same plant are entire. A 2013 model proposed that fenestration reduces the variance in a plant's carbon gain where most of the light arrives as brief unpredictable sunflecks: a perforated crown of the same leaf area intercepts flecks more evenly. It was published as a testable hypothesis, and the test has not been done.

Fig · Leaf and holes

Bisexual flowers, no perianth, a spathe that lasts two or three days — and a female phase that runs longer than the male, which is unusual enough to be worth stating.

Arrangement and the spathe

There is no marked flowering season. In any region where a species is reasonably collected, plants in flower turn up in every month of the year. Records for single species can be narrower, though: the 2024 Central American revision lists flowering for Monstera deliciosa in March, April and May, with fruit present all year, and for Monstera dubia in January, April and May.⁴ When a plant does flower it usually makes several inflorescences in succession, each shoot of the sympodium carrying two cataphylls and one inflorescence: two to four spadices in a cluster is normal, six to eight in Monstera obliqua, and solitary in Monstera lechleriana.

Peduncles are erect in every species but Monstera tuberculata, where peduncle and spadix both hang. They are terete or flattened, 5–45 cm long and 0.2–4.0 cm thick, and they keep growing after flowering — up to half again in length and thickness between anthesis and ripe fruit. The developing spathe is green, wrapped tightly around the spadix and longer than it. At anthesis it unfurls and colors: white, or yellow, or rose, depending on the species. It often shows a constriction about a third of the way up, marking where it had closed over the top of the spadix in bud. Fully open it may be almost globose with only a slit, or hemispherical with the spadix in plain view. Then it abscises, two or three days later.

The flowers

The spadix carries numerous spirally arranged bisexual flowers, each with four stamens and no perianth; those at the base are usually sterile. Flowers are protogynous by a day or two and mature from the base upward. Stigmatic drops have finished before pollen is shed, which works against an inflorescence fertilizing itself. The stamens have a broad flat filament 1.0–1.5 mm wide, the anthers stay hidden until just before anthesis and then push up between the pistils, and the pollen is shed through slits in sticky threads rather than as free grains. The ovary is bilocular with a perforation at the base of the septum and two basal ovules per locule. The stylar region is packed with needle-like trichosclereids 2–3 mm long, aligned with the axis of the pistil — a detail that matters later, when the fruit is handled.

Who pollinates it

For most of the twentieth century the answer was stingless bees, on the strength of an unpublished observation of two Ecuadorian species passed on in the 1977 revision and a 1978 report of Trigona bees taking the stigmatic secretion from Monstera deliciosa. Two field studies have since watched a Monstera through a full cycle — Monstera obliqua in French Guiana and Monstera lentii in Costa Rica — and in both the likely pollinator was a nitidulid beetle held for a night or more in the chamber behind the spathe. Where bees appeared, they were robbing pollen. The heat is modest, two to four degrees above the air, against eleven to nineteen in Philodendron. Only those two species of the genus have been studied this way, and they disagree with each other on almost every quantity, including whether the plant can set seed on its own — which is the subject of the reproduction guide. A message to the Aroid-L mailing list from Sarawak, far from where Monstera deliciosa is native, reports that it sets viable seed there regularly and names a melolonthin scarab, tentatively identified as Apogonia pygidialis Ritsema, as the main visitor and possibly the pollinator; the same beetle visits native aroids, notably Homalomena.¹²

Fig · Inflorescence

A fruit that takes anywhere from two months to over a year, and sheds its own armor when it is ready.

Once pollinated, the spathe falls and the spadix turns green — photosynthetic, and inconspicuous against the foliage while the seeds develop. The better protection is mechanical: the trichosclereids packed into the stylar end of every pistil make a barrier of needles around the seeds. Predation of developing fruit appears not to have been recorded.

Ripening takes 2 to 15 months depending on the species. When it is finished the stylar caps abscise — sometimes over the whole spadix, more often in a band that widens as ripening spreads — exposing the seeds in a sweet pulp, usually gray but in some species brightly colored. Ripe infructescences are white, yellow, orange or moss green by species, and in three species of one group the whole thing turns orange.

Each ovary starts with four ovules and usually finishes with one seed; two-seeded berries are uncommon. Endosperm is laid down early and then consumed, so the mature seed has none and stores its reserves in a swollen cotyledon instead. Cut open, a fresh seed is green or bluish inside. The seed coat is soft, unlike the rest of the tribe, and the seeds behave accordingly: they lose viability quickly if they are allowed to dry, but keep for months if kept moist.

Size varies more than almost anything else in the genus — roughly 5 to 22 mm — and it varies geographically. The largest seeds are in Mexico and they get smaller southward and eastward; every South American species has seeds under 10 mm. Size also tracks how the seedling behaves: species whose seedlings run as stolons have distinctly heavier seeds than those whose seedlings simply grow, which fits, since a stolon travels on what the seed packed for it.

A correction worth keeping

Unripe Monstera fruit is notoriously prickly in the mouth, and this is almost universally attributed to calcium oxalate raphides. The genus does contain raphides, abundantly, including in the fruit and seed coat. But it also contains trichosclereids, and the stylar portion of the fruit is full of them — they visibly penetrate the skin when the fruit is handled. That portion is shed and not normally eaten. The 1977 revision put the matter plainly and it has not moved since: it remains unclear whether the prickly character of unripe fruit is due to raphides or to trichosclereids. Ripe Monstera deliciosa fruit is not trouble-free for everyone, either. Writing to the Aroid-L mailing list in 2008, Leland warned that some people are sensitive to the calcium oxalate even in fully ripe fruit,¹³ and in 2009 D. Christopher Rogers wrote that some people who eat ripe fruit from his plant get a strong tingling, pin-pricking feeling on the lips, while others eating the same plant's fruit do not; he suggested the needle-like crystals.¹⁴ The cause is no more settled for ripe fruit than for unripe.

Fig · Fruit and seed