Cactus Flowers

Cactus flower anatomy has received little attention. Flowers are large and showy in many genera, with dozens of petals and stamens rather than just the 5 or 10 of each as we might expect of a dicot. All cacti, other than a few species in Pereskia, have an inferior ovary, and that is the point of this portion of the website. The ovaries of cacti are not merely inferior, they are spectacularly inferior.


Carnegiea gigantea (saguaro); notice the hundreds of stamens. Click on any photo for a larger image.	Weingartia hedingiana flowers; notice the scales on the exterior of the flower tube.	Trichocereus lamprochlorus.

Although axillary buds produce flowers, they always produce a bit of stem tissue first: the very first appendage primordia are leaf primordia, not sepal primordia. In most cacti, the axillary bud apical meristem actually produces 20 or 30 or more leaf primordia before it actually switches over to producing floral appendage primordia. The noteworthy thing is that these leaf primordia typically develop much more than do other leaf primordia on the same plant, and they mature into large scales that may be a centimeter or more long and have a thin, flat lamina-like region. People typically do not realize that these cactus leaves are true leaves but they are indeed. They are usually called “scales” or “bracts” but that does not matter: they are leaves. The problem is that the ovary is so deeply inferior that these leaves appear to be part of a flower rather than part of a vegetative stem.


Flowers of Soehrensia bruchii: upper red structures are petals and are parts of the true, flower; lower green tube-like parts are stems which bear leaves (usually called scales). The stem will develop into the outer part of the fruit (but because the outer part develops from something other than a carpel, it is more properly called a false fruit). Click on any photo for a larger image.	Shoot leaves on the flower tube of Rauhocereus riograndensis.	Large leaves on the shoot tissue surrounding the flower of Browningia candelaris.

The confusing nature of the stem and flower is due to the following modified development. The axillary bud apical meristem produces its series of leaf primordia, then a typical series of sepal, petal, stamen and carpel primordia, each above the other on the sides of the shoot apex as happens in any other group of plants. But as these primordia begin to enlarge and differentiate, peripheral shoot tissue below the leaf primordia elongates more than the inner shoot tissue below the carpel primordia. Whereas in ordinary stems all parts of the stem from pith to epidermis elongate exactly the same amount, here the pith lags behind while the epidermis and outer cortex extend, so the flower and subtending shoot turn themselves inside out. Imagine partially inflating a cylindrical toy balloon, then painting on dots to represent the leaf, sepal, petal, stamen and carpel primordia. Once all dots have been painted on with carpel dots at the closed end of the balloon, imagine placing your finger on the end of the balloon then blowing it up further: the balloon will become a hollow cylinder with the stamen dots on the inside next to your finger, the petal and sepal dots at the end and the leaf dots on the outside.

In a cactus flower, the hollow tube has leaves arranged in ordinary phyllotactic spirals on the outer surface almost until the rim of the tube is reached, then there are sepals and petals at the uppermost portions of the outside of the tube as well as at the rim, then down the inside of the tube are the numerous stamens with the first-formed ones at the top near the rim and the last-formed ones at the bottom of the tube’s interior near the carpels. Consequently, when you look at a cactus flower bud, you are not seeing any flower tissue at all except for the sepals and petals at the very tip; all the rest is vegetative stem and leaf, and the true flower parts are located inside. It is easier to comprehend that this really is vegetative tissue by realizing that these leaves typically bear clusters of spines in their axils. Need more proof? In Pereskia sacharosa and in Opuntia fulgida, the axillary buds of the leaves can themselves produce flowers so there are flowers borne on flowers. Opuntia fulgida can repeat this many so many times that as the flowers each mature into fruits, there are fruits hanging from fruits hanging from fruits…… the common name is “chain fruit cholla.” Because the outer part of a cactus “flower” is really vegetative tissue, once this all ripens into a fruit, the outer parts develop into a false fruit and the inner parts – those derived from the carpels – constitute the true fruit (remember that in apples, most of us only eat the false fruit and we throw away the true fruit, the core).


Flower of Neocardenasia herzogiana showing that stamens are attached on the inside of the flower tube. Click on this photo for two larger images and more explanation. Click any photo for a larger image.	Fruit of Cereus forbesii; the innermost red region is composed of funiculi (the stalks of the seeds and ovules), the white layer around that is the true fruit (developed from the carpels) and the greenish-red outermost region is false fruit (developed from the stem tissue that surrounds the ovary).	Fruit of Monvillea kroenleinii; the black seeds are surrounded by white funiculi; outermost green part of "fruit" is false fruit (stem tissue), the thin white layer between the false fruit and the seeds is the true fruit.

I have added this part to the website because the presence of leaves on the flower buds brings up an important point: most cacti still have the genes necessary to make large, flat, photosynthetic leaves. However, those genes are strictly repressed in the ordinary succulent stems of cacti, even though many live in areas that are moist enough that it would appear to be advantageous to have leaves at least during the rainy season. It could be – and this is a wholly unexplored hypothesis – that cacti are indeed making ephemeral leaves at exactly the time that they need them: when ovules are maturing into seeds and carpels are developing into fruits. The presence of leaves on the flowers may truly boost photosynthesis exactly when and where an extra boost is needed. On the other hand, in most cacti the extra amount of photosynthetic surface area produced by the flower’s leaves seems inconsequential compared to the total green surface area of the shoot, and their photosynthetic output seems trivial compared to all the starch that could be stored in the stem throughout the year as the evergreen stem photosynthesizes month after month.

A confounding aspect is that in some genera, there are no leaves or scales on the flowers, and these have been interpreted as having the same organization as other cactus flowers, but with the vegetative part consisting of just a single internode rather than many.


Fruit (only the false fruit portion is visible without dissection) of Acanthorhipsalis monacantha is smooth and mostly without leaves or axillary buds/spines (the white dot near the top is one axillary bud), because it is composed of just one or two internodes rather than many. Click on any photo for a larger image.	This fruit (false fruit visible) of Haageocereus turbidus has numerous axillary buds (about 12 are visible), each with just some hairs but no spines and the subtending leaves are microscopic. This false fruit is composed of numerous internodes.	This immature fruit of this Harrisia divaricata shows its stem-nature very well -- note the numerous leaves and the swollen leaf bases (in a vegetative stem of cactus these would be called tubercles). The true fruit (the carpel wall) is located interior to this green false fruit.

Another aspect of extremely inferior ovaries in cacti: in Calymmanthium, the vegetative tissues extend to the very rim of the tube as expected of cacti, but the rim itself does not expand much during flower development. The rim remains a microscopically narrow hole that is the same diameter as when it was formed by the floral apical meristem whereas in all other cacti the rim grows along with the rest of the flower, forming a wide rim that bears many petals (look at the flowers on the top of this page). Because the rim does not widen during development, a Calymmanthium flower develops inside a pouch of vegetative tissue -- when the flower is ready to open, it actually has to tear the vegetative tissues apart before the petals can emerge.

Flower of Calymmanthium fertile. Whereas the petals of most cactus flowers begin to be inserted at the top of the flower tube, the rim continues to be stem tissue in Calymmanthium. Because all flower parts begin as microscopically small primordia, when the tip of this tube was formed, it too was microscopically small -- and it stayed that way. It did not grow into the large opening like that on the flowers on the top of this page. Consequently, a Calymmanthium flower grows very well protected inside a vegetative pouch, but when it needs to open, it has to fight its way out -- the flower must actually rip the pouch apart. Calymmanthiums are big, rambling bush-like cacti; we have several at the University of Texas, but so far they have not yet become large enough to bloom here. This specimen was provided by Dr. Jean-Marie Solichon, Director of the the Jardin Exotique in Monaco (the Jardin Exotique is a superb collection of cacti and succulents and is well worth a visit).

Click the photo for a larger image.

As mentioned, the biological consequences of cacti having leaves and vegetative tissues completely surrounding their flowers is an area that has not been explored at all and would make a good project for someone.

I haven't done any research on flowers yet, so for further reading in this area, I recommend my two books again, Botany and A Cactus Odyssey.