Pavel Melnikov, Saint-Petersburg, Russia
Translated from Russian by Irina Ten

In my presentation at the First scientific-practical conference "Biological diversity and introduction of succulents in Russia" I made a very brief announcement of my hypothesis and I think few understood it. In this article the hypothesis is expounded.

More than five years ago my Leuchtenbergia principis  grew two tubercles that had green fleshy tips above the areoles (Photo 1) as usual for Ariocarpus agavioides. After that such areoles have never appeared again. The picture of the plant was taken. The discovery of the new Astrophytum  species called at first Digitostigma caput-medusae  makes me to recall those pictures and compare these two species' tubercles which are the longest and the thinnest among all cacti. Besides the similarity there are many differences in anatomy of these species. But the similarity of a tubercle of caput-medusae  and the abnormal tubercle of Leuchtenbergia principis  is impressive.

This year my attention was drawn to the structure of Echinocereus 's flowers. These primitive flowers have spines on its calyx tube resembling spines on the ribs. Echinocereus knippelianus  has specific spines on its ribs that I compared with spines on the calyx tube of its flower. During the summer I was watching with interest flowers of other species of Echinocereus , and then on the calyx tube of flower of Echinocereus pectinatus v. rigidissimus  I found the scales (Photo 2) identical to the structure of abnormal tubercle of Leuchtenbergia principis.

That brings us to the theory of evolutionary process from the bud with long thin scales to the stem with similar tubercles. There are many occasional transformations of an offshoot into a bud and otherwise. Some examples of such metamorphoses are described below.

During the last five years I have had Sulcorebutia rauschii  that had formed floral buds transforming offshoots into them and then the buds developed into self-pollinating flowers. A fruit ripe in the small hollow on the top of the offshoot, i.e. it's deepened into the offshoot's body. The seeds have a good germination rate. Let's take a closer look at the joint point between offshoot and the calyx tube (Photo 3). We can see that areole with a spine is gradually transforming into a calyx tube's scale with a spine and then into a normal scale without spine. This transformation is usual for species of Pterocactus , Wilcoxia  and some others.

I have seen a plant similar to Astrophytum myriostigma  on the Internet under the name Astrophytum ubuchi  and it's likely to be a cultivar (Photo 4). The plant is proliferous and it produces numerous buds in summer which sometimes continue to grow as offshoots. On the pictures we can see how a well-formed calyx tube typical for Astrophytum myriostigma  becomes an offshoot. The series of pictures show the process of transformation in progress (Photo 5 & 6). One of the buds starts to grow as an offshoot increasing in size and forming ribs and areoles typical for Astrophytum myriostigma. And finally the bristles of the new buds appear even in the areoles. This is a typical example of proliferation. It's interesting that while calyx tube matures, it gains red color. The offshoots with former calyx tube as its base fall down after "ripening" of "flowers" that were not pollinated and ovary's fading.

If we take a closer look at the bud-offshoot we notice that in comparison with Sulcorebutia rauschii  the transformation is reverse. The scales of a calyx tube transform into areoles of an offshoot. On the cross-section of the offshoot (Photo 7) we can see how the thin-walled calyx tube fills up with cells, the vascular system starts to develop and as a result the normal vegetative offshoot is formed. The food is delivered through the walls of the calyx tube. After few months the calyx tube fades and the offshoot will fall down like a not pollinated ovary.

Obviously cacti don't have a clear line between vegetative and generative shoots; some ambivalence is present. For many species of Opuntia  proliferation is typical; a generative shoot performs the function of vegetative one. In genera Wilcoxia  and Pterocactus  an offshoot transforms into a bud. One can state that such ambivalence is a common thing for Cactaceae  family. Also one can suggest a possibility of appearance and existence of solitary plants or even a population of the plants with vegetative shoots which have a morphologic anatomy of generative shoots which are more competitive in some conditions. In less arid environment the less succulent anatomy of a bud with sepals and petals provide bigger surface for breathing and photosynthesis and can give some advantages for surviving. As a result of selection a new form becomes dominant, i.e. vegetative shoots obtain some features of generative ones. In the other words, some stages of calyx tube's development are fixated by cacti's stems. This hypothesis explains the fast evolution of globular cacti (Astrophytum, Ferocactus ) and cacti with long tubercles (Leuchtenbergia, Astrophytum caput-medusae ).

Hence stems of cacti that belong to one genus or even species can vary noticeably in structure: from prominent ribbed one to covered with long tubercles as in genera Astrophytum  and Echinocactus. Probably the method how vegetative shoots adopt features of generative ones is absolutely different but the fact of such adoption is evident. I presume that such metamorphoses had happened repeatedly during the process of evolution of some genera. What genera have the shoots resembling buds? Obviously proliferous Opuntia  is a good example. In my opinion the other examples can be Leuchtenbergia, Astrophytum caput-medusae, Ariocarpus, and probably Mammillaria  (scales on its calyx tube have an areole at the bottom, an axilla actually), i.e. cacti with secondary tubercles and cacti which lost their ribs recently as result of evolution. Contemporary botany supposes that main parts of a flower - stamens and carpels - originated from the similar organs of sporophytes not from vegetative shoots. But the hypothesis about metamorphosis of calyx tube into vegetative shoot without stamens and carpels doesn't contradict to this fact. Besides the resemblance of constitution of cacti's shoots and buds the presence of similar process in other families could be a point in favor of this hypothesis.

The accuracy of the hypothesis can be proven if the new cacti are found in nature or in cultivation that differ strongly from other specimen of the genus but resemble buds, or if there is a possibility of selection of cultivars with appearance of bud. Let's go back to the picture of Astrophytum ubuchi. The plant has buds and shoots with features of the buds. From this plant we could produce a cultivar with hairy stem (like the buds) and without ribs although the plant already has these features to some extent. The fuzziness of Astrophytum, especially cv. 'Superkabuto'  could be taken from the buds.

The stated hypothesis allows us to make a whole series of assumptions: the existence of specimens of Gymnocalycium , Eriosyce  and other genera having cacti with long tubercles, and the existence of ribbed specimen without tubercles from genera Ariocarpus  and Mammillaria.