Malcolm Buleigh
Cactus and Succulent Journal (USA) 2005 VOLUME 77 NUMBER I

The time is coming when all of the plants in the wild will have been found. In all likelihood, that time is drawing near. This presents a dilemma, since most people who raise cacti seem to thrive on novelty. Collectors scramble to purchase new plants: Geohintonia, Aztekium, Turbinicarpus alonsoi, Mammillaria luethyi, while voracious pilfering of these rare and endangered species is taking place.

As a result of my early years haunting the Paramount nursery, I became familiar with the Paramount Echinopsis hybrids. When I moved to Minnesota, I had a colleague who raised irises, and he took me around to visit many people who bred plants. They asked me what I raised, and I told them I collected cacti. Their next question was whether I bred cacti. At that time my answer was no. I told them that I didn't breed cacti in my collection because the time difference from flower to flower was too long with cacti. They laughed and told me that the time line for iris was about five years and for orchids even longer! I began to think that it might be possible in my lifetime to produce some desirable cacti—plants that have never seen the desert sky. In this article, I describe how varieties of Turbinicarpus schmiedickeanus and Gymnocactus viereckii (L1159 & SB 1570) have been crossed and back-crossed through several generations to produce plants similar to the original T. schmiedickeanus varieties except that these new plants have pink flowers instead of the normal white. Additionally, an alternative soil culture is proposed which focuses on a more natural combination of soil components.

Choosing characteristics

I wanted a genus that was fairly rare but contained a sufficient diversity of characteristics that could bring different forms to light. Concentrating on North American genera, I picked out several target genera with characteristics of interest: Ariocarpus, Thelocactus, Pediocactus, Sclerocactus/Echinomastus, Echinocereus, Escobaria, and Mammillaria. In addition to spination, body, flowers and hardiness, there are other important characteristics which are harder to circumscribe, such as how well and how frequently the plants flower, plant size, robustness, resistance to disease and the overall ability to be grown in a collection.

At the time, I was having trouble flowering my meager Ariocarpus collection. Thelocactus and Echinocereus plants were a bit too large to house in my small 26 m2 greenhouse. Escobaria, Pediocactus and Sclerocactus species proved a bit difficult for me to raise at the time. I was forced to consider easier genera, so my initial endeavors were with Mammillaria. However, in the early 1970s, I began to see offerings for strange plants that I had never seen before—Turbinicarpus. Some looked like Loph but with small weak spines. Others looked like dead grass and animal droppings. The descriptions of Glass and Foster¹ revealed that here was a very small genus with a huge amount of variation among the species. This could be a perfect group for my work. It was small enough to manage all of the species and varied enough to produce many novel forms (Fig 1). The plants tend to stay small, and they flower when quite young. In addition, there are several flower colors that might be readily manipulated. Although not cold hardy, its other properties made it a good prospect, and since I have limited space, the genus Turbinicarpus was made to order.

The process of hybridizing

I use small watercolor brushes to transfer pollen. The best brushes for pollination are made with animal fibers, since natural bristles have a scaly structure on the shaft of the hair that help to remove, hold, and transfer the pollen grains.

Synthetic fiber brushes, though cheaper and longer lasting, simply do not work as well.

Cactus flowers are particularly easy to pollinate. Most have an open bisexual flower with numerous pollen-laden stamens, and a branched central pistil that is easy to spot. I whisk the brush across the stamens of the father plant, and then again across the pistil of the mother. Black bristles make it easy to see the pollen on the brush, and the sticky stigma of the female plant will capture pollen readily. Since most cacti are self-infertile, or would at least prefer to receive pollen from another plant, the reverse cross can be made with the same brush, moving back to the first plant. Both plants get to be mother and father, and considerable differences between the progeny from each parent can occur. When attempting the next cross, a new brush should be used, or the brush should be cleaned. 1 wash the brushes with water frequently and quickly dry them on my sleeve. Rubbing alcohol is another good cleaning agent and ensures that the pollen from the previous transfer is killed, reducing the likelihood of contamination and confusion.

When a flower "takes," the ovary will swell. This can be observed by carefully removing the flower tops. If the flower top has a very narrow base, you do not have a take. It' the base is about 1-2 mm wide then pollination has been successful. In about 6-S weeks the seeds will mature. Cooler weather will cause a slower maturation of the seeds. The fruits of Turbinicarpus and Gymnocactus are small, greenish, berry-like pods about 2—5 mm in diameter. They occur at the top of the plant often hidden amongst the wooly hairs. The entire genus appears to be self-fertile.

There are some problems with hybridization. With Turbinicarpus, it is very difficult to mark the flowers that you have pollinated. The ovaries are generally set within the cottony tuft from which the (lowers emerged. It is easy to lose track of the different seed pods as they mature. Then there are insects that want to do the pollinating for you. Despite keeping records of crosses, there will always be seeds that form from unintended pollination.

Aberrant forms (cresting, probably related to grafting rather than hybridization) and odd colors among the progeny are also a problem. Variegation and general loss of chlorophyll in the hybrid plants are the most serious challenges I have faced. Luckily, variegation shows up quickly. Both regular variegation and a gradual bleaching of the body of the plant (a common problem I call the "yellows" that shows up as plants mature) reduce plant vigor, and must be culled from the hybrid program, no matter how spectacular the plants are, as they are deleterious to both the individual plants and their progeny. Figure 2 shows this effect with two plants of Mammillaria glassii x Mammillaria erythrosperma. Note that the plant on the right has the "yellows." A cross of T. pseudopectinatus x G. beguinii on Myrtillocactus (Fig 3) has been very disappointing in this regard, but since the cross must be made in our winter, I have kept it. However, it will always have the yellows.

Saving seed

During the year, care must be taken to protect the seeds from squirrels, mice and seed-eating insects such as box elder bugs. When the seed pod of a Turbinicarpus plant gets ripe, it splits open (dehisces) in a vertical line, spilling the seed out at the base or within the cottony tops of the plant. It is generally best to wait until the seed pod opens to collect the seed, since this guarantees mature seed. I have been disappointed when I have collected immature seeds that were not viable. In order collect the seeds, you must reach between the spines with forceps to pick them out around the opened fruit, but this frequently allows seeds to drop to the base of the plant when you have not securely grabbed them. The best way to circumvent this is to spray the seeds with a little dish detergent in water. This will wet the top of the plant and the seeds will tend to stick together and onto the surface of the plant. This method also works for Thelocactus, Sclerocactus, Pediocactus and other plants that have dehiscent fruit.

The blooming year

Each Turbinicarpus and Gymnocactus species has a preferred flowering period during the year, and the overlap time for the different species largely determines what crosses can be made. There are two basic blooming periods: winter and summer. The winter bloomers don't seem to be terribly similar in their outward appearance, and similarly the summer bloomers also show the entire run of Turbinicarpus/Gymnocactus form. The early bloomers are Gymnocactus beguinii, T. schmiedickeanus var schmiedickeanus, T. pseudopectinatus and T. valdezianus. I have only been able to cross these amongst themselves, since they prefer to bloom in February and March. In the spring, the earliest bloomer is T. pseudomae-rochele, with the rest following in close succession. The last to flower (late July to early August) is Gymnocactus gielsdorfianus. A table of flowering times for the species is listed below. Not all of the taxa are listed; these either have not flowered for me or are not in my collection.

Turbinicarpus/Gymnocactus blooming times

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

T. schmiedickeanus var schmiedickeanus

T. pseudopectinata

G. beguinii (light flower)

T. valdezianus

G. knuthianus

T. pseudomacrochele

T. krainzianus

T. schmiedickeanus var polaskii

T. schmiedickeanus var schwartzii

T. schmiedickeanus var macrochele

T. schmiedickeanus var klinkerianus

G. viereckii var viereckii

T. loph

T. flaviflorus

T. gracilis var dickisoniae

T. gracilis

T. lauii

G. viereckii var major

G. gielsdorfianus

G. horripilus

G. beguinii (dark red)

G. beguinii var senilis

G. subterraneus var zaragosae

Selecting from the generations

When you produce a cross between two different species, the first generation (called Fl) generally has characters intermediate between both parents. Each plant that is produced from the initial cross between these different plants will all look the same. Even-thing seems to be mixed, like tossing blue dye and yellow dye into a glass of water and producing green. This can be seen in the two hybrid plants of Gymnocactus viereckii (L1159) and Turbinicarpus schmiedickeanus var klinkerianus (Fig 4). If you stop there, you might think that nothing else could come out of this activity but a mixed plant of intermediate characteristics.

However, the second (F2) and subsequent generations will reveal much more diversity, and this is where you can begin to make some real choices. It is not like mixing two dyes into a glass of water. It is more like throwing dice. You may pick and choose for whatever characteristics you want as long as the plants that you are working with are still fertile. Thus, it may be possible to select for a red flower and weak spines or a white flower and strong spines. The various characteristics of the original parents are expressed in a random manner. The F2 and later crosses are generally termed hybrid swarms. Figure 5 is a picture of a 10 cm pot with the out-crossed F2 progeny of Gymnocactus viereckii (L 1159) crossed with T. schmiedickeanus var klinkerianus, schwarzii or polaskii. These types of hybrid swarms may also occur in nature among related plants. It is from these plants that one can pick and choose desirable traits.

When my hybrids flower, I give the ones that I want to keep a list number. I keep a chart with the desirability (rating from 1 to 5) with comments about the plant such as flower color and size. Eventually, I give the best ones more fanciful names.

My first goal was to produce something with large flowers, crossing the large flowered T.loph with the smaller flowered T. schmiedickeanus var polaskii, T. s. var klinkerianus, T. s. var schwarzii and T. s. var macrochele. Some of the progeny of these plants were interesting. Several plants had much darker bodies than their parents. These dark, shiny plant bodies arc quite striking. However, I did not keep any of them—another direction intervened that caught my fancy, which I have been pursuing since the mid 1980s.

Glass and Foster¹ suggested that Gymnocactus might eventually be included in Turbinicarpus. I decided that I should include this genus into my turbinicarpus equation. Here was the possibility of having a pink flowered plant resembling the polaskii/schwarzii/klinkerianus group rather than the normal white flower.

Raising difficult cacti

I was having too much difficulty raising just about all of my target genera. Mealy bug infestations, mites and rot seemed to consume too many of my plants. I resorted to grafting onto Pereskiopsis. This succeeded in accomplishing several things: decreasing the time from seed to flower, producing more flowers and generating more offsets for propagation. In retrospect, however, most of the problems that I had with all of these genera were related to my use of peat moss, which is difficult to wet thoroughly, doesn't drain or dry acceptably, and prevents adequate treatment of root pests. Rotting, poor growth and mealy bug infestations used to be an all too common occurrence in my collection.

Peters² suggested a much different soil medium composed mostly of non-organic ingredients, in 1997, a friend and 1 took a trip to the Big Bend National Park where I noted that Ariocarpus, Escobaria and Neolloydia plants were growing in pure limestone. When we got out of the limestone strata, the plants were no longer there, a phenomenon I had observed on other trips.

Luckily for me, St Paul sits on a huge bed of Pennsylvanian limestone. Dolomite is a variety of limestone that cannot be made into good cement. The local stone yards sells '3/4"-minus' dolomite which I grade, sift and mix in varying proportions to use as a potting and seedling mix. The largest stones (1.5 cm to 1 cm) are used for the pot base. The (5 mm to 1 cm) gravel is for top dressing. The <5 mm portion is mixed with about 50% perlite and used for soil. For all soils I make up a substantial amount and then test the mix in a pot, adding Perlite as needed to improve drainage.

After making the switch to this new mixture, Turbinicarpus, Ariocarpus. Escobaria and many of my Mammillaria plants improved. Now I can water and fertilize my plants without fear of rot. and I can successfully treat for mealy bugs. Additionally, small seedlings from escaped seed grow in the pots, a sure sign of adequate culture. Turbinicarpus loph. (Fig 11) once on Pereskiopsis stock by necessity, is now easy on its own roots. Plants once considered difficult are now run of the mill, and flowering has vastly increased even for difficult-to-flower species.

I had grown Gymnocactus viereckii (L1159) from seed, finally getting it to bloom in 1985. This plant has a small, dark pink flower and was comparatively easy to bloom. I first crossed it with T. schmiedickeanus var klinkerianus. I got the seeds from this cross with the pollen parent as T. schmiedickeanus var klinkerianus. The next year (1986) I started the seeds and got several to sprout. A seeding was grafted onto Pereskiopsis and quickly grew into something I had never seen before. It was obviously a hybrid (#14, Fig 6). The spines of the plant have the typical black tips of Gymnocactus viereckii, but they have a softer feel to them, reminiscent of T. s. var klinkerianus. The body of the plant seems to have an intermediate tubercle shape and spination (Fig 7)—between that of the spiny gym-nocactus with tubercles that are more acute and the flatter tubercles and weaker spines of the T. schmiedickeanus varieties. The next year (1988) T. #14 bloomed, and the flower proved to be as big as T. s. var klinkerianus and almost as magenta as Gymnocactus viereckii. G. viereckii has a smaller flower; the denser spination of the plant partially prevents the flower from opening. In addition to that, it was fertile! I now have several of these plants (#14, 55 and 83, Fig 8) and they all are similar. These latter are Fl crosses between G. viereckii (L1159) and T. s. var schwarzii. Figure 1 illustrates some of the variations in my hybrid collection.

I kept crossing the Fl hybrids together, and finally I picked three plants to propagate. One was an F2 plant (#117, "Marissa's Dream", Fig 9) with a lot of T. schmiedickeanus var schwarzii characteristics. This plant has very nice pink flowers of a similar size to T. s. var schwarzii and the very flat tubercles of T. s. var schwarzii. The spination is also sparse and the plant is spheroid. I have been trying to propagate this from Pereskiopsis stock, but it is shy to offset. It seems not to be very robust, but it is almost exactly what I had been looking for: a plant with the spherical shape of the T. schmiedickeanus varieties, weak papery spines and the larger pink flowers of that group.

Another was a plant that was back-crossed with T. s. var schwarzii, #120 "Little Livi" (Fig 10) These are named after my granddaughters. This plant is much more likely to offset, making it easier to propagate. The body of the plant is similar to T. s. var schwarzii. Figures 7 and 9 show T. s. var klinkerianus and some of the other hybrids. The flower, though, has a nice pale pink cast to it and the plant also seems to show a certain amount of hybrid vigor. The last is T. #115 (Fig 1) which is a plant of unknown parentage, but I suspect that is has some T. lauii in it. It was chosen for its superb vigor. I use it as a secondary grafting stock on Pereskiopsis for Turbinicarpus and Ariocarpus. The work is now ongoing as I approach my goals year by year. What I have to do now is to keep crossing these plants and raise many seeds every year, while looking for the right ones and discarding the ones that are not right. It is a fascinating and rewarding practice anyone can try. Let's bring some new creatures into our collections!

References

1. GLASS C, FOSTER R. 1977. A Revision of the Genus Turbinicarpus. Cact Succ J (US) 52 (4): 161-176.
2. PETERS R. 1998. Growing Plants in Low-organic Soil-based Mixes. Cact Succ J (US) 70 (1): 10-11.