
PLANT
PROPAGATION
Magnoliaceae
Magnoliaceae
This highly decorative but primitive Family, from the early period of Angiosperm evolution, contains over two hundred species of deciduous and evergreen woody plants. It is characterised by having the floral parts occurring in a spiral (rather than rings) on a cone shaped receptacle and with no specific differentiation into petals and sepals. The seeds carry differentiated but very small, immature embryos.
The Family has a disjunct distribution - on the eastern side of North America down into subtropical and tropical Central and South America and then in Eastern and South Eastern Asia to the Himalayas, India and Sri Lanka - again with a distribution which encompasses temperate to tropical. Most species are found in warm temperate and subtropical climates and about two thirds of the species are in Asia. There are, however, many which are suitable for cultivation in a temperate climate – such as the UK.
In recent years DNA analysis and the resultant assessment of the phylogeny of the Family has toppled the previously recognised categorisation of genera. It has separated the Family into two sub-families – the monogeneric Liriodendroideae (Liriodendron) and the remaining group of genera – the Magnolioideae. This latter grouping has proved difficult for taxonomists to differentiate with any great exactitude or agreement and it is variously regarded as one diverse portmanteau genus (Magnolia) or some seven genera which include the familiar Magnolia, Manglietia, Michelia and Talauma and it is these four genera which contain the species suitable for cultivation in temperate climates. References, in this text, to generic and specific names will be debatable and reference will need to be made to up-to-date literature and thinking in order to be current.
Magnolia (sensu stricto)
The seeds of Magnolias are produced in tough, leathery to woody, cone-like fruits. These are made up of clusters of sessile, coalescent, coarse, fleshy follicles, each one of which may contain one or two seeds or may be empty. These fruits ripen in the late summer or autumn and, as they dry, each follicle gradually splits open (along a longitudinal median suture) to liberate its seeds. The final stage in the maturation of the fruit - which leads to the dispersal of the seed is readily recognised – for as the follicles split they expose the immediately obvious, brightly coloured seeds, the outer seedcoats of which are usually orange, scarlet or red.
Depending on the weather conditions it may take as long as three to four weeks for the fruit case to open sufficiently for the seeds to be released. Often the seeds are temporarily retained by the fruit as they hang down on a mucilaginous thread or ‘suspensor’ for a day or two before detaching and falling to the ground. In a few species the fruits will absciss and split open, to expose the seeds, on the ground. Dispersal of the seeds is effected by birds from the tree and birds and small mammals from the ground. The outer seed coat of the seed represents a desirable and nutritious source of fats and oils for birds and small mammals stocking up on their body fat for the winter.
As the fruits mature and the seeds develop and enlarge, each follicle, containing a fertilised seed, will swell. If the pattern of pollination has been irregular and incomplete the development of the cone will consequently be twisted and contorted. Hence the shape of the cone at maturity will reflect the suitability of the season for pollination, at the time of flowering and the availability of suitable agents for pollination (normally beetles), which thus determines the crop of seeds produced.
Although the seeds will have developed to ‘maturity’ before the dispersal stage begins, it is tedious and difficult to extract the seeds from the fruits until the follicles split. It is usually prudent to leave collection of the fruits, until this opening stage is reached on the tree - as harvesting beforehand and drying artificially does not necessarily cause the follicles to split and separate along the suture. If they fail to open, extraction from these tough fruits is time consuming. Thus the fruits should be harvested for easy extraction when the follicles have split open but before the seeds disperse – usually this phase lasts for a couple of weeks – so that it is not critical. Extraction is by hand - picking the seeds from the follicles if they do not fall out readily.
The seed of Magnolia has a seed coat that consists of three layers:-
-
the outer layer which is relatively thick and consists of edible fats and oils and is retained by a brightly coloured skin. This material fulfils two functions the first is as the ‘bait’ which causes the vectors to eat the seed and so disperse it and secondly while awaiting this activity it protects the seed from drying;
-
the inner layer is hard and stony and generally the colour of this material is black but in M. virginiana it is beige. This layer is not a barrier to water uptake (and hence germination) but protects the seed in its passage through the digestive tract of the vector. It does not exert any physical constraint or impede moisture flow in either direction and
-
the third, innermost layer is a thin, water permeable membrane which has no other function than to retain the endosperm and structure of the seed itself.
Internally the seed consists of a firm fleshy endosperm which virtually fills the seed. The greater proportion of this food reserve is oils and lipids. Embedded in this endosperm at the basal, pointed end of the seed and along the longitudinal axis is a small but only partially developed embryo (ie an immature, as opposed to a rudimentary, embryo) with its radicle towards the base. At this stage the embryo length - in most species of Magnolia - is unlikely to exceed one sixth of the overall internal length of the seed and is usually much smaller than this.
At the dispersal stage no further development of this immature embryo occurs until the embryo has imbibed and then has experienced a period of chilling (at least for species from temperate provenances).
Storage of Magnolia seeds is not straight forward but, in the short to medium term, they can be cold stored in the ‘fresh’ condition with their outer seed coat intact (this oily fleshy material should not be dried or permitted to go rancid). A suitable regime would be to place them in a polythene bag (to prevent water loss) and refrigerate them at 2-4ºC. This biological material, if kept en masse, will have the potential to ‘heat up’ and go rancid: thus it is prudent to store the material in relatively small packets. This should keep the seeds in viable, usable condition for several weeks – until treatments to allow germination to begin are initiated.
When the process leading to germination is to be initiated the first activity is to extract the seed from the outer layer. The seeds are pulped or macerated and left in some warm water for a couple of days - until the seed can be rubbed clean – the detritus will float and can be decanted off. The stony seeds will fall to the bottom and can be washed clean – the final washing and rinsing should include a small amount of detergent in order to ensure that the seed is free of any surface lipid as this may create a waterproofing effect and so affect the movement of moisture into the seed.
Once the seed has been prepared and cleaned it is possible to fully imbibe the seed and proceed with its treatment to achieve germination. To ensure that the seed (both embryo and food reserve) is fully imbibed the sample should be soaked in warm water for a few hours. It is then (without delay) mixed with a relevant stratifying medium (ie one with adequate water retention and a sufficient ‘air filled porosity’) and placed in cold temperature storage at 3ºC (+/-1ºC). This treatment is needed to eliminate the inhibition to further embryo development (practically this equates to exposure to a winter’s cold). Potentially germination will then occur when it is subjected to a warm environment. The period of cold treatment needed to achieve a situation where the sample can be moved into warmth, is generally – for temperate climate species - given at 56 days but reasonably accurate observation has demonstrated that 42 days is sufficient for the deciduous American species and most of the Asiatic species tested – this period is sufficient to achieve eventual seedling emergence in a uniform and synchronised fashion. The recommendations for a longer period might well be due to incomplete cleaning of the seed after extraction and prior to treatment (ie the seed was still ‘waterproofed’ with surface lipid and this had to degrade before full and rapid imbibition occurred).
Following the chilling period the sample is transferred to a warm environment for the embryo to develop to a stage at which it will emerge through the seed coat (ie germination). At a constant temperature of 20ºC it will require 21 days to such emergence.
If it is possible, it will be more accurate and efficient to transfer the sample directly from refrigeration to an incubator and not sow the seeds until after 20 days of this latter treatment have been completed. The seeds can then be sown on a standard seed tray, covered with grit and germinated on a bottom heat of 20ºC when emergence should be virtually immediate (ie one to two days).
For ease of practical operation it is useful if a known number of seeds are designated in each sample – say 100/120 seeds in a sample - which also represents a reasonable number to sow on a standard seed tray.
Once the accurate assessment of these treatments (cleaning, chilling and warm period) has been confirmed it allows the development of a blueprint for production in order to achieve a rapid and synchronised germination. Thus the beginning of the treatment process will be determined by working backwards from the day on which germination is scheduled.
The seeds can be germinated on any well drained medium and after a few days of protected growth can be pricked off into liner containers (before any major radicle extension growth and branching occurs). It eases the practical operation of separating the seedlings for potting on if the seeds are sown in a medium with a high proportion of grit so that the seedlings can be readily separated from the medium and extracted without damage to the root system.
The propagation of Magnolias from seed is chiefly used in the production of rootstocks for grafting particular hybrids and cultivars. It is prudent to select a closely related species for use as the rootstock in order to provide a relevant vigour. Most species of Magnolia are intercompatible and in the past the traditional rootstocks used (for deciduous species) were M. kobus for Asiatic species (but it is hardly vigorous enough for many of the modern tree hybrids and selections) and M. acuminata for American species. M. grandiflora has usually been used for evergreen species
On a small scale seed propagation may be used for the production of rarer species and during breeding programmes.
The following species appear to fruit fairly satisfactorily in the UK when mature trees are available - however care should be taken to make an assessment of potential hybridity in Parks and Gardens where different species and hybrids are growing in reasonable proximity. The level of seed production will reflect the suitability of climatic conditions at flowering - these can vary dramatically from year to year.
Spring flowering, deciduous - American – M. acuminata, M. cordata, M. fraseri, M. ashei, M. macrophylla, M. pyramidata and M. tripetala. Himalayan - M. campbellii and M. mollicomata. Chinese - M. cylindrica, M. dawsoniana, M. denudata, M. liliflora, M. officinalis, M. rostrata, M. sargentii and M. sprengeri. Japanese - M. kobus, M. stellata, M. obovata and M. salicifolia.
Summer flowering, evergreen - M. delavayi (Asiatic) and M. grandiflora and M. virginiana (North American).
Summer flowering, deciduous (all Asiatic) - M. globosa, M. sieboldii, M. sinensis and M. wilsonii.
Newly available species - specimens of these species are generally not yet seeding prolifically - M. biondii and M. zenii.
As most deciduous Magnolias hybridise freely the offspring from hybrid sources may be extremely variable and exhibit a broad spectrum of characteristics. In order to make use of such material it can be recognised that Magnolias are hardy trees and can be used as windbreaks. They withstood the 1990 gales with some ease – inevitably the leaves to the windward will be damaged but that is a feature of any deciduous windbreak.
Manglietia, Michelia and Talauma
As indicated above these three ‘genera’ present a dilemma taxonomically and are increasingly regarded as part of Magnolia. All the species within these three groups are more or less native to warm temperate to sub-tropical climates and only very few are suitable for cultivation in the UK outside the mildest niches.
The fruits are similar to Magnolia in structure and the same comments apply as to harvesting and extraction. The seeds have the same aril type construction and the same seed processing and cleaning regime will be required.
Despite their warmer provenance the seeds of many of the species which fringe into warm temperate climate areas also require a period of chilling (often because of their high altitude provenance) in order to overcome physiological dormancy. Although there is little concrete information on what period(s) are effective – probably the requirement for most would be relatively marginal (ie some three weeks at below 7˚C). However in order to ensure success the logical programme to be adopted would be the same as for Magnolia unless other information is available; the seeds are unlikely to be distressed using this protocol.
Up until the recent past the numbers of species from these groups which have been grown, with any success in the British Isles, has been very limited – usually only Manglietia insignis, Michelia compressa, M. doltsopa and M. figo and sometimes M. champaca; nowadays there are many other species being grown and assessed.
Liriodendron
This genus of tall trees, which can reach thirty metres in height, are variously known by several vernacular names - the most prominent being Tulip tree and Yellow Poplar. There are two species – one in Eastern North America (L. tulipifera) and one in China and Vietnam (L. chinense).
The fruit is an elongated, narrow cone consisting of tightly adpressed woody samaras – 80 to 100 per cone. This cone dries and breaks up - chiefly in the autumn but sometimes continuously over the winter period. The seeds are wind dispersed. Thus harvesting is most readily achieved just before the cone disintegrates.
In normal circumstances the great majority of the seedcases are void – the number of filled seed cases can be as low as 10%. Separating the void seeds from the filled cases is not an exact science and usually depends on attempting to select the heavier samaras.
It has proved difficult to germinate the seeds with any exactness and consistency. However a reasonable protocol would start with the processing of fresh seed. Despite the hard case of the samara, there is not any constraint on water penetration to the seed. It is therefore sensible to start with a twenty four hour, warm water soak in order to ensure an imbibed embryo and then chilling the seed for 63 days at 3˚C. Germination at 20+˚C is then successful.