Herbertia 1979
At Long Last—Seeds On Lycoris squamigera
Sam Caldwell

In the early 1950s when I started hybridizing lycorises I tried crossing everything available, including, of course L. squamigera, Over several years I spread pollen from L. squamigera to every other Lycoris that was blooming and pollen from other species onto L. squamigera, with very discouraging results—not a single seed. Then a botanist friend told me, "Lycoris squamigera is an infertile triploid," and I read in Traub and Moldenke's Amaryllidaceae: Tribe Amarylleae that "L. squamigera appears to be a sterile natural hybrid, not represented in any fertile wild populations. . . ." Having little botanical knowledge myself, I was impressed by these statements and concluded that in my innocence I had been trying to do something impossible. Also at the time I was getting good results and many seedlings by using L. sprengeri, L. sanguinea and a fertile strain of L. radiata, so I just dropped all efforts with L. squamigera.

Nevertheless, over years that followed I would occasionally hear or read of someone claiming to have seeds on L. squamigera. I carefully investigated every report and rumor that I could and in every case, found it to be in error. The reason seemed to be that to average gardeners L. squamigera (see Fig. 6) does appear to make seeds. After the flowers fade large seed capsules often develop, filled with chaff and bits of aborted seeds which are worthless but to the uninitiated do look like flower seeds.

Speculating that somewhere, someone might really have a fertile strain of L. squamigera, I wrote a letter that was published in the September, 1975 issue of Flower and Garden magazine, which circulates among more than a half million gardeners. I explained the situation surrounding L. squamigera, the Magic Lily, and requested any reader having one making seeds to get in touch with me.

My letter drew some thirty responses—tributes to the warmheartedness, generosity and spirit of helpfulness that prevails among garden-minded people. Unfortunately, these attributes are not always coupled with much horticultural knowledge, for it soon became clear that more than twenty of the writers didn't know what a lycoris was. The common name, Magic Lily, covered a lot of ground for them and they sent me seeds of Hemerocallis, Hymenocallis, Amaryllis, Brunsvigia rosea and even stem-bulbils from stalks of Lilium tigrinum. Then eight correspondents sent dried seed capsules of L. squamigera containing the usual worthless chaff that they thought was seed. Exactly one person—a lady in Dublin, Ohio—sent me three real lycoris seeds, which I planted. However, from information supplied in her letter and from the appearance of the one seedling I now have growing. I feel sure that this is L. sprengeri.

By coincidence I received from Japan at about this time a letter from my good gardener friend, Dr. Shuichi Hirao, who wrote: "I want this letter to report you my 'finding' on lycoris. It is to obtain seed from sterile species. The practice is very simple: just cut the bloom-stalk after pollination and hang it downwards in the shade, or just lay on a shady ground. The stalk will shrivel gradually from the cut end, but the umbel will continue to live and the pod will swell gradually. If lucky you will find one or more perfect seed in the pod after four to five weeks after the pollination. I got a perfect seed out of three umbels of Lycoris squamigera treated above."

So in the flowering season of 1976 I went back to work on L. squamigera, using pollen from L. sanguinea, L. sprengeri, L. chinensis, L. "Sperryi" and from a new unidentified yellow lycoris that looks much like L. squamigera. Reciprocal crosses were made. I cut about 40 scapes, labeled them and hung them in light shade in my greenhouse. For a time they made progress; seed capsules fattened in an encouraging way. However, in September when they were fully ripened it was a disappointing task to shell out the capsules, umbel after umbel, and find no seed. But one scape looked particularly good and, sure enough, when I peeled away the capsule covering, there they were—three large, shiny, hard black seeds, one of them fully 3/8" in diameter. This may sound absurd but plant breeders will understand—it was like finding gold nuggets after a 20-year search!

A label showed that this was the only scape of L. squamigera on which I had applied pollen from L. chinensis. The secret was out. Lycoris squamigera could produce seed in cooperation with the right partner. And as one might guess, L. chinensis is something special. It is the big, beautiful hardy yellow "spiderlily" type lycoris received in 1948 at the U.S.D.A.. Plant Introduction Garden in Glenn Dale, Maryland from the Nanking, China Botanic Garden. It came under the label, "L. aurea," but proved quite distinct from what we regard as true L. aurea, that grows in Ft. Augustine, Florida and other mild-winter areas. Dr. Traub named it L. chinensis in 1958. (See Fig. 7).

There are few bulbs of the species in this country. I have had it since 1958 and still have only two flowering size bulbs. Last year I dug my first acquisition, planted as a single bulb in 1958. There was still the original bulb, 1 1/2" in diameter, one additional bulb slightly smaller, and two offsets about 3/4" in diameter—a very miserly natural increase after 19 years of growth, during which time it has borne a single good flower scape in July of nearly every year. It is very fertile and sets good crops of seed to its own and other pollens. Oddly enough, though I have raised hybrids with L. chinensis as the seed parent, I have never been able to bring up a mature bulb from self-pollinated seed. They germinate fairly well but under exactly the same condition that I routinely grow other lycoris seedlings, the little L. chinensis bulbs just disappear after two or three years. No doubt it is possible to grow them; I just haven't found the way.

Now back to the story of my 1976 crop of the three hybrid seeds, L. squamigera X L. chinensis. These were planted in a 4" pot kept under a bench in my cool greenhouse over winter, as I customarily handle all lycoris seed. In March I carefully dug into the planting medium and was happy to find two small white bulblets emerged from black seed covers. It was a sad day a few weeks later when I again dug in and discovered both bulblets rotting.

Fortunately the next flowering season, 1977, was a good one. There were two scapes of L. chinenis, supplying plenty of pollen for use on dozens of L. squamigera flowers. Some of these I cut and hung in the greenhouse, others were placed in the greenhouse in a jar of water which was changed occasionally, and still others were left outside on the bulbs as an experiment. Those in water all decayed after a few weeks, but from the greenhouse hung scapes and those gathered outside in mid-September I was delighted to get 36 mature, sound seeds. (See Fig. 8). In fact, those pollinated scapes left outside on their bulbs were quite as productive as those brought into the greenhouse, so it would appear that cutting and hanging is unnecessary.

Actually, the ratio of seeds to pollinated scape was low. Many scapes produced no seeds at all. Six was the most from any one scape, while the over-all average was slightly over one seed per scape, although I had generally pollinated every flower.

This time, hoping to avoid decay of small bulblets, I planted the seeds in a peatmoss-vermiculite mix with a fungicide added. The same medium was used for all my other lycoris seeds of the 1977 season, and these germinated and grew well. Not so the squamigera seeds. It is sad to report that about half of them apparently rotted without germinating at all. Others did germinate and develop plump little bulblets but these also decayed within a few months.

Discussing the problem with friends who are experienced plant propagators, I was told that these seeds undoubtedly can be started successfully under aseptic culture in a properly equipped laboratory. In fact, one who is connected with a government agency having adequate facilities offered to start seeds that I might supply. I hoped to take advantage of that offer this year but it turned out to be one of those years when I had no bloom on L. chinensis, nor were there any flowers on the species at the Plant Introduction Garden in Maryland. Thus there was no pollen for fertilizing L. squamigera flowers.

But there'll be another year ... more lycoris blooms ... more pollen spreading ... more of the illusive seeds on L. squamigera ... more years of waiting as seedlings grow to flowering bulbs. What a hybrid might result! How will the traits of orchid-tinted L. squamigera combine with those of the golden ruffled Chinese parent? Answering a question like that provides endless fascination for the avid plant breeder.