"Trillium Germination and Seedling Growth"
John F. Gyer, November 2005
There have been frequent discussions on Trillium-L of pollination, seed germination and seedling growth in the genus trillium. The following discussion and supporting photographs are a comprehensive description of the earliest stages of trillium seed germination and seedling growth.
In one way or another most of the trillium seed discussions I have read or heard involve the concept of DOUBLE DORMANCY and the steps used to break dormancy. Double dormancy in trillium was described by Barton in 1944 (Cont. Boyce Thompson Institute, 13: 259-271). The double dormancy concept implies that two cold periods are needed for seedling development. Germination is implicitly defined as the appearance above ground of a photosynthetic cotyledon. Double dormancy treats the seed like a "black box", a diagrammatic device used in engineering when the inputs and outputs of a system are known or measurable, but the inner workings of the system are not known. For trillium seeds, the commonly known or measurable inputs are planting time and the number of cold periods prior to cotyledon emergence. The output is the development above ground of the photosynthetic cotyledon. The biologic processes of germination and growth between planting and cotyledon appearance are ignored. The following research presents a picture of this "black box" period in the growth of trillium seedlings.
The easiest way to visualize the process of trillium germination and seedling growth is to refer to pictures of the process. I used Trillium grandiflorum because the garden department at Winterthur Museum, Garden and Library allowed me to collect seed from the extensive and beautifully naturalized planting in the Azalea Woods section of the garden. The skotomorphogenic germination and seedling growth seen in T. Grandiflorum is repeated, with minor modifications, in all the trillium species I have examined.
Picture 1: "Trillium Grandiflorum
Photograph: John F Gyer
Trillium seed forms in a berry like that in the first picture, "Trillium Grandiflorum Berry Section". Seed is fertilized by the two sperm in a pollen tube that grows through the style and placenta, and into the micropore of the ovule. The embryo forms just inside the micropore from the fusion of one sperm with the egg to form a diploid cell. The second sperm of the same pollen tube joins with a diploid cell to form a triploid cell that will become endosperm, the energy storage tissue of the seed. The endosperm is fed by photosynthate; sugars, lipids, etc. produced in the leaves. Photosynthate enters the seed through the Chalazal Cap. The vascular system that delivers photosynthate runs from the placenta into the side of the aril and through the raphe to the chalazal cap.
Picture 2: "Trillium Grandiflorum
Seed (stained for starch)"
Photograph: John F Gyer
The second picture, "Trillium Grandiflorum Seed (stained for starch)", shows the surface structure of a single seed. Starch, stained black by Iodine, is concentrated in the aril. Although not shown, starch also is present in the placenta. This is significant for, as the berry matures, the starch begins to ferment and releases volatile compounds that attract ants, hornets, etc. These carry the seed about in the local environment to expand and maintain healthy populations.
Picture 3: "Trillium Grandiflorum
Seed & Aril Longitudinal Section".
Photograph: John F Gyer
The third picture is a section of a T. Grandiflorum seed ,"Trillium Grandiflorum Seed & Aril Longitudinal Section". The most striking feature of the section is the thin layer of starchy cells on the exterior of the aril. The bulk of the aril contains cells with lipids, some of which are similar to the lipids that exist in the skin of some insect larva. These larva are the food for species of predatory ants that are known to carry trillium seed about as they feed on the aril. The vascular strand is just discernable in this illustration, but it is lignified and traceable up to the Chalazal Cap where its identity is lost.
The Chalazal Cap is important for it is through the cap that the seed "communicates" with its environment. It is permeable to small molecules and pathogenic bacteria and fungus. In tissue culture experiments the chalazal area of fresh seeds was very difficult to sterilize. However, when it was cut away, the rest of the seed remained uninfected in culture. In nature and in germination experiments, seed rot generally begins at the chalazal area and often spreads down the wall of the endosperm resulting in seed coat separation from endosperm tissue.
Under "normal" conditions the seed is discharged into a relatively dry ( high humidity at the soil surface, but no prolonged soaking) environment where ants or other creatures feed on the aril, but leave the seed intact. Aril removal makes this tissue unavailable to pathogenic fungus and bacteria while the relative dryness allows the chalazal cap to seal as compounds (probably phenolic) that leak out through it are oxidized. David Mellard has reported on Trillium-L that mildly drying fresh seed before storage or shipment greatly reduces rot. This rot reduction is at least partially due to oxidative sealing of the chalazal cap during the mild drying he gives the seed. Continued soaking of fresh seed in a wet environment will retard chalazal cap sealing and the normal exudates will be food for pathogens that thrive in films of water.
About half the weight of trillium seed is water. Some can be lost without damage to the seed, but often fully dried seed - particularly seed of the pedicillate species such as T.grandiflorum - is damaged and may not germinate. Because of this trillium seed is usually stored or shipped slightly moist, a condition that favors seed rot organisms. Various seed surface sterilants are used to reduce the tendency to rot. Some can damage the seed, if not used carefully.
A 10% Chlorox solution is often used as a seed surface sterilant. The solution is alkaline and dissolves the waxy seed coat that repels water and resists fungal attack. The wax loss can leave the seed subject to pathogen attack during germination. After a prolonged soak, Clorox can diffuse through the chalazal cap and disrupt endosperm cells. The disrupted cells will leak their contents through the chalazal cap during the long time needed for germination. The leakage will attract pathogens and may undo the good the initial sterilization accomplished.
The permeability of the chalazal cap is the main reason that the surface sterilant, hydrogen peroxide, must be used with caution. Hydrogen peroxide will not dissolve seed coat wax, but it can diffuse beneath the cap where it decomposes into gas that separates the cap tissue from the endosperm. This promotes leakage of endosperm cell contents into the environment The leakage attracts pathogens that can ultimately rot the seed. However, in the case of T. grandiflorum, aril tissue also decomposes hydrogen peroxide into gas bubbles that rupture aril cells. Leakage from ruptured aril cells can be washed away. This reduces the importance of the aril as a food source for fungus and bacteria during germination and storage.
The endosperm occupies the bulk of the seed volume. This is energy storage tissue, the stored sunlight that the growing embryo feeds upon. The endosperm/embryo volume/volume ratio of T. Grandiflorum is in the 700 to 1,000 range. It is less for the smaller seeds of species in the erectum group (500 to 800) and much higher for the large seeds of sessile species (1,000 to 1,800). All the species I have examined have very similar embryo volumes (see the 4th picture for typical dimensions) but the endosperm volume varies with seed size and seed size varies with species. The endosperm/embryo volume ratios roughly correlate with typical environments for the species - cooler/wetter for the erectum group, more mesic for grandiflorum, and warmer/dryer for the sessiles.
Although they are not well defined in the picture, there are three distinguishable regions in the endosperm; a core that generally has smaller cells, a cortex with larger cells that sometimes may have a little starch accumulation near the cell wall, and a thin wall area of starch free cells adjacent to the seed coat. The embryo grows through the core of the seed. If the cortex contains cells with starch, they are distributed toward the micropore half of the seed and form a band around the core.
The embryo lies about 0.5mm above the micropore. It seems surrounded by a thin gel-like layer. At the magnification I use, this layer does not appear cellular. There is no vascular connection between the embryo and endosperm. The embryo is positioned by a thin line of cells (not seen in the picture) called the suspensor apparatus. These cells contain a little starch and extend from the micropore wall to the embryo.
The micropore is firmly plugged by what appears to be the same sort of brown
polymeric material that seals the chalazal cap.
Picture 4: "Trillium Grandiflorum
Embryo in culture".
Photograph: John F Gyer. Photo taken at University of Delaware.
The embryo, shown in the 4th picture, "Trillium Grandiflorum Embryo in culture", is the new trillium plant. At this stage it is minute - 0.2 mm by 0.25 mm. It consists of a rounded dome that will become the cotyledon and hypocotyl and a more pointed base that will become the rhizome. When sectioned and stained for starch, the rhizome pole has a cortex of starchy cells surrounding a core of starch free cells that will become the vascular system as the embryo grows. When the embryo is dissected from the seed, a section of the suspensor apparatus usually is attached, as shown in the picture.
Culture experiments have shown that most embryos from mature seeds are dormant,
i.e. they do not grow over a period of 60 days at about 70 degrees F. Dormant
embryos can be induced to grow when cultured in a medium with as little as 1ppm
GA-3. However, not all seeds in an individual mature berry will have a dormant
embryo. Some will produce an above ground cotyledon the spring after they are
planted, i.e. after one cold period. These embryos grow in culture without added
hormones. The reasons for this will be the subject of another discussion.
Picture 5: "Germinated Trillium
Photograph: John F Gyer. Photo taken at University of Delaware.
The 5th picture, "Germinated Trillium Seed", shows an embryo after about 60 days of growth. This seed is germinated, i.e. the rhizome is emerging from the seed. Further growth is the skotomorphogenitic development of a seedling.
When the germination process begins the cotyledonary pole (see picture 4) begins to expand. Based on the appearance of seeds dissected in the early stages of germination, it is probable that the cotyledonary pole of the embryo excretes enzymes and/or hormones that begin to digest the core endosperm immediately around it. The digested endosperm is absorbed by the growing cotyledon and is translocated to the hypocotyl and rhizome where much of it is stored as starch. During about 60 days of this growth the embryo lengthens to ten times its initial length or to about 2.5 mm. About mid way in this growth starch granules appear in the endosperm near the micropore around the tip of the rhizome. These endosperm cells become enlarged and water filled - the hyaline cells in picture 5. The hyaline cells are weaker than regular endosperm and provide an easy exit for the rhizome as it is pushed out by hypocotyl and cotyledon elongation.
The hyaline cells at the micropore are the result of gibberellin that is probably released from the tip of the rhizome. In experiments when exogenous gibberellin is used as a seed soak the same sort of hyaline cells can appear randomly on the seed surface wherever the seed coat is abraded or damaged and the endosperm is exposed. These hyaline cell clusters are a target for pathogens that rot the seed.
At germination a vascular system is evident in the embryo. In picture 5 parts
of it appear as a clear area in the cotyledon and another clear area in the
rhizome that is a meristematic bud. Although it will not be a three-leafed whorl,
this bud will be the first true leaf two years from germination.
Picture 6: "Trillium Grandiflorum
early seedling growth".
Photograph: John F Gyer. Photo taken at University of Delaware.
Picture 6, "Trillium Grandiflorum early seedling growth", shows the skotomorphogenic seedling about 2 weeks after germination. When the rhizome emerges from the seed, it immediately begins growth of the primary root which is quickly covered in root hairs. These absorptive surfaces may begin to absorb mineral nutrients that aid the development of the seedling. Although the endosperm is rich in energy reserves, it may not have large stores of Nitrogen, Phosphorus, Potash, Calcium and Magnesium. The primary root could absorb these from the environment and make them available for seedling growth. In addition it can absorb water which provides the growing seedling with increased ability to survive temporary dry spells.
After primary root growth has started, the hypocotyl begins to elongate and
forces the rhizome clear of the seed. At this point all the energy for growth
must be absorbed from digested endosperm by the cotyledon.
Picture 7: "Advanced Seedling
Photograph: John F Gyer
Picture 7, "Advanced Seedling - TrilliumGrandiflorum", shows the seedling after about 5 weeks of skotomorhpogenic growth. The primary root continues to elongate and may reach 2 to 3 inches. A secondary root buds from the rhizome. About the time this root appears, cotyledonary tissue can be seen just emerging from the micropore area.
Picture 8: "Trillium Grandiflorum
Seedling - November 2005".
Photograph: John F Gyer. Photo taken at Rowan University.
Picture 8, "Trillium Grandiflorum Seedling - November 2005", shows a fully developed seedling. Some cotyledonary tissue remains in the seed husk to digest and absorb the last traces of endosperm, but over half is outside the seed. This tissue will turn green and photosynthetic when it is exposed to light. The hypocotyl looks solid and relatively thick for it has stored starch to provide energy for rapid spring elongation and growth that will bring the cotyledon to the surface. The bud meristem is obvious and well developed. The rhizome is starch filled as is the upper section of the primary root.. The seedling is now an independent plant that will undergo the same growth cycles as the adult it will soon become. This means that some cooling is needed to initiate spring elongational growth - just as cooling of the developed buds on a mature rhizome is needed for normal growth.
Picture 9: ""Trillium Maculatum
mature rhizome & seedling".
Photograph: John F Gyer
Picture 9, "Trillium Maculatum mature rhizome & seedling", clearly shows the amount of hypocotyl elongation needed to bring the cotyledon above the leaf litter layer. In this case the hypocotyl has stretched about 2.5 inches (about 6 cm). The energy for this spring growth was mobilized from the starch reserves stored in the hypocotyl and rhizome during the skotomorphogenic growth phase of the seedling.
Picture 9 also compares the size of a seedling rhizome as its first photosynthetic year begins with a fully mature, flowering size rhizome that is at least 6 years old. Both plants were photographed in situ in a habitat that is not subject to significant frost.
Picture 10: "Contractile Root
in Soil ( Trillium grandiflorum)".
Photograph: John F Gyer
Picture 10, "Contractile Root in Soil ( Trillium grandiflorum)", shows a rhizome that is flowering for the first time about 6 years after germination. The primary root is still visible. Subsequent growth cycles have enlarged the rhizome and contractile roots have pulled the rhizome deeper into the soil each year, an action that is probably responsible for the "J" shape of some rhizomes.
The idea that seeds of Trillium grandiflorum have "double dormancy" dates from the work of L. V. Barton in 1944 (Some Seeds Showing Special Dormancy, Cont. Boyce Thompson Institute, 13: 259-271). In Barton's work seed was considered germinated when the cotyledon appeared above ground. Since in her work the cotyledons of most Trillium grandiflorum seeds appeared after two winters, the seed was considered doubly dormant, i.e. two cold periods were needed for germination. Barton did not report on the stages of embryo growth in the seed prior to cotyledon emergence
. Barton's work is cited by Baskin & Baskin ( "SEEDS, Ecology, Biogeography,
and Evolution of Dormancy and Germination"; Academic Press, 1998 ; pg 95
and Table 10.18, pg 372) as an example of morphophysiological dormancy. In their
analysis of Barton's work Baskin & Baskin state that the first cold period
is required to break "radicle dormancy" and the second is needed to
break "epicotyl dormancy". After "epicotyl dormancy" is
broken the "shoot bud" will grow. Thomas Patrick in his MS thesis
("Observations on the Life History of Trillium grandiflorum (Michaux) Salisbury"
Cornell University; May 1973) observed and illustrated the growth stages of
T.grandiflorum seedlings. His work and the photos in this report clearly show
that after the first cold period the cotyledon itself expands and the "shoot
bud" continues to develop at the apex of the rhizome. Typically this "shoot
bud" will grow into a true leaf after the third cold period in the seedling's
Trillium seed and seedling structures, as I interpret them, are defined in
the preceding pictures. These definitions reinterpret some of Patrick's observations,
correct the terminology that Baskin & Baskin applied to trillium, and shift
the descriptive emphasis of trillium seed germination from a series of dormancies
to a single dormancy followed by germination and skotomorphogenic plant growth.
The reasoning behind my definitions follow.
The nomenclature used in literature that discusses the external appendage of
a trillium seed is inconsistent and can be confusing. This confusion is illustrated
in an article by Solt (Solt, Stephanie; Propagation Protocol for Trillium L.
(Liliaceae); Native Plants Journal; Spring 2002; pg. 18) that describes the
appendage as "either a strophiole or an aril or an eliasome".
An eliasome refers to any seed structure that attracts ants or other insects that then carry the seed about in its environment. It does not refer to a specific physiological structure. The term has relevance to discussions of ecology, but is not relevant to this discussion of trillium seed structures and seed germination.
STROPHIOLE is defined as:
"An appendage at the hilum of certain seeds" in both Gray's Manual, 8th edition and Gleason & Cronquest, 2nd edition.
Patrick's 1973 thesis refers to the external seed appendage as the "strophiole" and the scar at the chalazal end of the seed as the "hilum". Although Patrick's pairing of strophiole and hilum is consistent with the broad glossary definition, current usage, as revealed by a GOOGLE Scholar search on strophiole, identifies a strophiole as part of the seed coat that regulates the seed coat's permeability to water before and during imbibition. In this literature strophiole is a term used primarily in discussion of seeds of the Fabaceae (Leguminosae).
Hilum is defined as:
"The scar or point of attachment of the seed" (Gray);
"The scar of a seed at its point of attachment." (Gleason & Cronquist);
"a scar on a seed, marking the place where it was attached to the seed stalk" (Webster's New World Dictionary, 2nd College edition, 1970).
The "seed stalk" is the funiculus generally defined as the free stalk
of an ovule. In trillium the funiculus is poorly differentiated from the placenta.
In Picture 1 above it is best seen as the thick stalk below the unfertilized
ovule. The above definitions place the hilum of trillium at the base of the
appendage where it attaches to the funiculus/placenta tissue and not at the
chalazal end of the seed.
Although not commonly seen in discussions of trillium seed, the term raphe as defined in glossaries describes the underlying structure of the trillium seed appendage:
Raphe is defined as:
"The ridge or adnate funicle which in an anatropous ovule connects the two ends." (Gray's Manual);
"The part of the funiculus that is permanently adnate to the integument of the ovule, commonly visible as a line or ridge on the mature seed coat." (Gleason & Cronquist);
"That part of the funiculus or ovule stalk attached permanently to the ovule or seed." (Radford, Ahles, & Bell in The Manual if the Vascular Flora of the Carolinas, 1968);
"a ridge of tissue along the side of an ovule, indicating the position of the vascular bundle which supplies the developing seed" (Webster's Dictonary).
The raphe can lie below or internal to the integument as in the seed of Helleborus
thibetanus or it can lie external to the integument as in Stylophorum diphyllum,
Sanguinaria canadensis and all the trillium species I have examined. For seeds
with an internal raphe, the hilum is at the base of the seed where the raphe
and the funiculus/placenta structures join. For seeds with an external raphe
the hilum is similarly defined. The junction of the raphe and the funiculus/placenta
is at or near the joining of the raphe and the integument. For seeds with an
internal raphe, the vascular trace ends at the chalazal area that is hidden
within the integument. By analogy for an external raphe the vascular trace ends
at the visible chalazal end of the seed. Because of these structural considerations,
I propose the term CHALAZAL CAP for the scar at the chalazal end of the
trillium seed and the term EXTERNAL RAPHE to describe the underlying
structure of the trillium seed appendage. In trillium the abscission zone of
the raphe-funiculus/placenta structure is obscure and does not form a distinct
scar in the soft tissue of the raphe. Therefore the term hilum is not applicable
to trillium. Instead of a distinct separation of the seed from the placenta
Trillium grandiflorum seed separates because the tissues involved decompose
- probably by a fermentation process.
The base of the external raphe of Trillium grandiflorum is significantly enlarged. This enlargement, clearly seen in the first 3 pictures above, fits the definition of an ARIL.
"ARIL" is defined as:
"An appendage growing at or about the hilum of a seed" (Gray's Manual);
"A specialized, usually fleshy outgrowth from the funiculus that covers or is attached to the mature seed; more loosely, any appendage or thickening on the seed coat." (Gleason & Cronquist);
"Enlarged raphe, an appendage on the seed." (; Radford, Ahles & Bell);
"an additional covering that forms on certain seeds after fertilization, developing from the stalk of the ovule" (Webster's dictionary).
Fermentation or decomposition of the starch in the outer cells of the aril
likely produces volatiles attractive to ants or insects that are the dispersal
agents for this trillium in its environment. Thus the ecological function of
Trillium grandiflorum's external raphe is as an eliasome.
As a result of these considerations I propose that the appendage of a trillium seed consists of an aril that covers the lower part of an underlying raphe that ends at the chalaza of the seed. When the appendage is removed by insects or rot, the chalazal area of the seed is covered and closed by the chalazal cap.
"HYPOCOTYL" is defined as:
"the part of the axis, or stem, below the cotyledons in the embryo of a plant" (Webster's dictionary).
This is a clear definition for dicots, but is less clear for Trillium, a monocot.
Patrick applies the term to the storage organ that emerges from the seed coat
at germination (seed coat rupture) and discusses the "hypocotyl-young rhizome"
for the storage organ seen after the photosynthesizing cotyledon has senesced.
He mentions that the "hypocotyl" can still be found as long as 12
years after germination. On pg. 88 he defines the trillium hypocotyl
"- that region between the primary root and the cotyledonary node. ---- this means that roughly half the structure is actually part of the cotyledonary petiole and sheath, epicotyl, or young plumule.".
He clearly sees a difference between the storage organ - his hypocotyl- and
the tissues at the base of the cotyledon.
At germination, Picture 5 above, the embryo has a clearly differentiated storage (starchy) organ that supports a meristem or bud at its apex. The cotyledon has little starch, but the part of the cotyledon stem just above the bud is thickened and starchy. The primary root grows from the base of the storage organ very soon after the organ emerges from the seed husk. During the skotomorphogenic growth of the trillium seedling, the storage organ enlarges until it comprises about 40% of the dry mass of the seedling (J.F.Gyer, unpublished data). The starchy tissue at the base of the cotyledon stem persists and expands proportionally. At about the time that cotyledonary tissue appears outside the seed husk, the storage organ develops a secondary root from an area just below the shoot meristem. These morphologic growth phases are seen in the pictures above and follow the pattern seen in the mature rhizome:
Patrick's thesis presents a good description of the vascular system of the skotomorphogenic trillium seedling rhizome (his hypocotyl) as the system transitions from that of the primary root to that of the developed rhizome structure. The structural changes involved fit well with Easau's description of the "transition zone" between root and stem (Esau, K.; PLANT ANATOMY; John Wiley & Sons, Inc, 1953, pg. 521). The concept of a transition zone in this structure is consistent with the position of the seedling rhizome in the plant architecture, the obvious definition of "root" and the definition of "RHIZOME" as:
"Any prostrate or subterrranean stem, usually rooting at the nodes and becoming upcurved at the apex."(Gray's Manual";
"A creeping underground stem", (Gleason & Cronquist)
"Usually elongate, horizontal underground or subsurface stem, usually rooting at the nodes.", (Radford, Ahles & Bell)
"a creeping stem lying, usually horizontally, at or under the surface of the soil and differing from a root in having scale leaves, bearing leaves or aerial shoots near its tips, and producing roots from its undersurface", (Webster's dictionary).
Based on these definitions, an interesting speculation is that the visible
photosynthetic leaves and above ground stem of trillium are actually petioles
and flowering leaves that develop from the stem-like rhizome.
The thickened, starchy cotyledon stem area just above the rhizome's apical meristem fits Webster's definition of "hypocotyl". That term is applied to this area because of position (between the cotyledon and the rhizome) and its function. The starch reserves are available to power the elevation of the cotyledon to the surface where it becomes photosynthetic. The thinner, relatively starch free stem above the starchy hypocotyl becomes a cotyledonary petiole.
"EPICOTYL" is defined as:
"that part of the stem of a seedling or embryo just above the cotyledons", (Webseter's dictionary)
"Epicotyl" is a useful and descriptive term for dicots and hypogeal
(cotyledons remain below ground and do not become photosynthetic) monocots like
Smilacina racemosa. Trillium however has an epigeal cotyledon and the term "epicotyl"
results in confusion. If "epicotyl" is applied to the bud meristem
area of the seedling rhizome, then three, not two, vernalizations are required
to activate it - 1, breaks embryo dormancy; 2, induces cotyledon expansion;
3, induces development of the bud meristem (epicotyl). Baskin & Baskin on
page 95 rely on the work of Barton to support trillium "double dormancy".
Baskin & Baskin describe the process as follows:
"After radicle dormancy [= embryo dormancy, JFG] is broken [cold period 1, JFG] , seeds should be exposed to temperatures of 20-30EC in a greenhouse for 3 months (to simulate spring and summer) to permit emergence of the radicle [= rhizome, JFG], production of a root system, and formation and growth of the shoot bud. A 4-month period of cold stratification at 5EC [cold period 2, JFG] will break dormancy of the epicotyl (shoot bud)"
This confuses the development of the photosynthetic epigeal cotyledon, where
no "shoot bud " is involved, with growth of the still dormant or at
least quiescent shoot bud meristem. Application of the concept of skotomorphogenic
seedling growth, as outlined with the pictures above, avoids this confusion
and accurately represents the growth of trillium seedlings.