The genus Crinum is an attractive group of tropical and subtropical lilies found about Africa, Asia, Australia, and often in frost free areas of North and South America. Illustrations or color photographs appear in botanical publications from time to time, but an organized collection has not been published previously. As a consequence my many collecting friends have contributed photos and comments such that some of the better known members of the genus can be presented as an organized group. At the same time some cultural and breeding notes are included. They are a tempting group to hybridize, but tricky.
Back in 1820-40 Dean William Herbert became interested in the group after flowering quite a striking hybrid. At that time the existing ecclesiastical opinions were that most plant and animal species had evolved entirely by cross-breeding after Noah's flood. So Herbert initiated quite a breeding program and ended up 15 years later with only 24 named hybrids. Out of the 24 only two had produced seed. The old hybrid theory silently fell flat on its face!
The same breeding problems still exist today. The more colorful crinum species present a natural incentive for one to try various hybridization experiments, and hybrids are commonly obtained. But, many are usually sterile mules, as Herbert gradually realized and so stated. Often the hybrid pollens are viable, but most hybrid combinations are disinclined to accept any pollens, unless parental, and even then quite reluctantly. Occasionally hybrid pollens will back-cross onto parental or allied species. But success is often limited due to genetic conflicts. In 50 years I have only developed two or three interspecific hybrids which actually yield seed freely. On the other hand intraspecific breeding between related variants often results in fertile hybrids, as well as some nonfertile. So intraspecific breeding is the more successful, though seldom attempted, except with C. scabrum and its many variants.
Back in 1888 J. G. Baker, author of the Handbook of the Amaryllidaceae, divided the genus Crinum into three subdivisions or subgenera, with these based upon the shape and arrangement of the petals: (1) linear petals in an actinomorphic arrangement for subgenus Stenaster, (2) lorate petals, also actinomorphically arranged, for subgenus Platyaster and (3), for the subgenus Codonocrinum he cited those forms with broad elliptical petals in curved trumpet-shaped blossoms having bilateral symmetry. Genetically, and on the evolutionary basis, we find another basic division noted by Ker-Gawler in 1787 where the respective blossoms either have their seed capsules sessile or subsessile to the umbels, or in contrast (2) have their seed capsules and blossoms mounted on pedicels several inches in length. This sessile-pedicellated division gives us six distinct morphological groups which aid greatly in any preliminary classification, as well as suggesting breeding combinations. So we have obviously arranged our illustrations of species in accordance to these specific groups.
Then, according to G. Ledyard Stebbins, a well respected plant geneticist at U. C. Davis, the primitive Crinum were marsh plants with rush-like leaves, which bore simple sessile blossoms with narrow linear petals, (or more specifically, 'tepals' which botanists prefer to use in lieu of petals). The point is, during the last few million years with the many climatic changes, these aquatic plants, particularly coastal, spread readily on ocean currents and suffered less form the long dry cycles, and made a few adaptive changes. So our more primitive Crinum are apparently the sessile-flowered, linear-petaled actinomorphic forms like C. defixum found mostly about Asia and Indonesia, or the many variants of C. americanum which range from northern Florida and the Louisiana Gulf down to northern Argentina. (Some American forms are found with linear tepals, not lorate, so logically they are among the most primitive).
Concurrently, we find some linear tepaled forms having elongated pedicels, normally with numerous blossoms like the C. asiaticum which are basically aquatic. Several of these early forms still exist in Africa, but have been subjected to more modifications there than Asia. With prolonged droughts the inland African plants have experienced greater climatic stress with resulting modifications. Here a group succeeded in the development of the zygomorphic blossoms with the broad tepals and curved tepaltubes, primarily to protect the stigma and allied parts from direct sunlight. From these, both sessile and pedicellated Codonocrinum type-forms evolved, and in all probability the latter gave rise to Nerine, Brunsvigia, Amaryllis belladonna, Cybistetes, and Crossyne as all possess the same bilateral zygomorphic features. Amongst the sessile-zygomorphic Crinum forms (all tropical) we have the C. jagus series (ex C. giganteum) and the Crinum series Ornatae which contains C. scabrum, C. abyssinicum, and even C. zeylanicum from India. In turn, the long-pedicellated zygomorphic group contains forms like C. bulbispermum, C. macowanii, and C. moorei, all semitropicals or subtropicals.
Some ten years back the International Botanical Congress agreed that a first described species for a genus should also be identified as the subgenus for that genus, thus Crinum americanum is both the representative for the genus Crinum and now the subgenus Crinum. This new subgenus now includes all Crinum with actinomorphic blossoms (the original members of Baker's subgenera Stenaster and Platyaster) as members of subgenera Crinum. We now find this lumping inconvenient since Baker's original subdivisions are normally quite obvious, but the basic morphological distinctions between those Crinum having umbels with sessile blossoms which flower in groups and the pedicellated forms which flower sequentially have been ignored since 1821, although this is a basic evolutionary split, and as such, this becomes particularly evident when one attempts interspecific breeding of the two umbel forms. Thus, as an aid in Crinum identification we recognize these two umbel types as well as Baker's original subgeneric division to be the best practical means of subgenus Crinum species recognition. Likewise, the subgenus Codonocrinum with its zygomorphic trumpet shaped blossoms can be subdivided into distinctive Sessile and Pedicellated groups with the sessile being essentially tropical species.
So once we know these six distinctive morphological groups and their most obvious features and habits, their identifications are simplified, as are their possibilities for intercrossing.