Facebook Twitter Flickr Google + Blogger

Australasian Plant Conservation

Originally published in Australasian Plant Conservation 21(3) December 2012 - February 2013, p 17-19

Seeds behaving badly: Conservation of rainforest species

Graeme Errington, Peter Cuneo and Catherine A. Offord
Royal Botanic Gardens & Domain Trust, Sydney. Email: graeme.errington@rbgsyd.nsw.gov.au

Determination of seed storage classification e.g. orthodox seeds maintain viability when dried to 5% and stored at -20º C (seed bank condition). Adapted from (Kew, n.d.).

Diagram of 100 seed test for desiccation tolerance (Gold & Hay, 2008). Equilibrium relative humidity (eRH) is a non‑destructive measurement of moisture content, actual moisture content is calculated using an oven-dry method.

The aim of Rainforest Seed Conservation Program of the Royal Botanic Gardens & Domain Trust is to identify species that can be seed banked and those with ‘bad’ seed storage behaviour that require alternative ex situ germplasm conservation techniques. The exploration of the seed storage behaviour of seeds can be complicated and there is much to learn about how seeds from rainforests respond to seed storage treatments and how we should approach alternatives for germplasm conservation.


Rainforests are important repositories of biodiversity that contain many threatened species of flora and fauna. Across the planet, rainforests also have important economic and cultural values. In NSW, rainforests cover less than 1% of the land area with around 30% of the pre-European extent lost. Many of the remaining remnants have been subjected to logging during past management. Some rainforest communities have been dramatically changed, with remnants occurring as small fragmented areas that are infested with invasive species and diseases such as the recently arrived fungus myrtle rust.

Seed banking

Seed banking is an efficient and effective long-term ex situ conservation strategy for species that tolerate drying and storage at low temperature. Many rainforest species have large, fleshy fruits - characteristics that are commonly associated with poor seed storage potential. To date, there is limited understanding of the seed storage behaviour of rainforest species and as a result, these species are under-represented in seed bank conservation programs. Highly successful programs like the Millennium Seed Bank Partnership have focussed on dryland flora, with good reason, as a high proportion of the species from plant communities in these areas are suited to seed banking. By comparison there has been relatively little work on seed storage of rainforest species globally or within Australia.

Seed storage behaviour

Roberts (1973) described two classes of seed storage behaviour: orthodox and recalcitrant. Orthodox seeds have increased storage longevity in response to decreased seed moisture content and decreased storage temperature, with maximum longevity at conditions of low moisture content (2-6%) and at a temperature of -20º C. Internationally many seedbanks have used these parameters to develop seed storage facilities for crop and wild species. In Australia, dominant plant groups such as Acacia and Eucalyptus have orthodox seed and long-term conservation seed banks such as the NSW Seedbank have developed extensive collections of these and other dryland species.

The term recalcitrant is applied to seeds that do not tolerate drying. This type of seed cannot be dried below a relatively high moisture content without a significant drop in viability (this ‘critical moisture content’ varies between species), consequently recalcitrant seeds are not suited to conservation seed banking. Terms such as desiccation tolerance and desiccation sensitivity have been applied to the response of seeds to moisture loss. Seeds that can be dried to low moisture contents, 4-7% are termed desiccation tolerant. Desiccation tolerance and dormancy are two plant traits that allow seeds to develop soil seedbanks and survive until suitable conditions for germination occur. Desiccation tolerant seeds can be dried and stored at -20º C which extends the period of viability, and as indicated previously, these types of seeds are termed orthodox.

There is an important distinction between desiccation tolerant and orthodox. Some desiccation tolerant species, such as some Australian Citrus spp. and Macadamia ternifolia (Hamilton, et al., 2010), have high oil content and this oil has a phase transition at freezing temperature and has the potential to cause a loss of viability as a result of low temperature storage, thus are not classified as orthodox species.

Since the initial classification of seed storage behaviour by Roberts, it has become clear that a number of species cannot be accounted for by the two existing categories of seed storage behaviour. Ellis (1990) proposed a new classification of intermediate seed storage behaviour. This classification is indicative of a spectrum of desiccation tolerance and seed storage behaviour, Figure 1. Seeds of the intermediate category can be dried to around 10-12% moisture content, but they are sensitive to storage at sub-zero temperatures.

The categorisation of some species is complicated by unusual seed behaviour. In paw paw (Carica papaya), drying induces a dormancy that may result in seeds being classified as non -viable or desiccation sensitive in a germination test. This is further complicated by crystallisation of lipids at sub-zero temperatures, which then require exposure to high temperatures before germination can occur. In Azadirachta indica, slow drying can allow seeds to be fully desiccation tolerant and stored at -18ºC or less, but the seeds must imbibe moisture at high temperature (above 25ºC), or they may suffer chilling damage (Kew, n.d.). These types of behaviours make it difficult to make general recommendations about the storage of intermediate species.

Identifying seed storage behaviour

The main purpose of the Rainforest Seed Conservation Project is to identify orthodox species that can be seed-banked and to ensure they are adequately represented in our seed bank. Initially the desiccation tolerance of a species must be determined, and then further viability testing examines the response to storage at -20º C. There are a number of sources of existing information regarding seed storage behaviour, such as the Kew Seed Information Database and some general indicative characteristics that can be used to make assessments about potential seed storage behaviour:

  • dry capsules and dehiscent fruits generally have desiccation tolerant seeds
  • taxonomy – some families such as Dipterocarpaceae are almost exclusively recalcitrant but other families have a proportion of all seed storage behaviour categories e.g. Myrtaceae and Proteaceae
  • wetter environments have a high proportion of recalcitrant species and drier environments have a high proportion of orthodox species (aquatic seeds are generally orthodox)
  • seeds that are released during the wet season have a high proportion of recalcitrant species and seeds released during the dry season have a high proportion of orthodox species
  • larger seeds tend to be recalcitrant
  • seeds with very thin seed coats tend to be recalcitrant (determined by the seed coat ratio).

Our screening process compares seed germination response to three treatments: fresh seed, dried seed and moist seed (stored for the same time period as seed takes to dry) based on the 100 seeds test (Pritchard, et al., 2004). If the germination rate in the dried seed treatment is similar to the fresh seed, then the species is considered to be desiccation tolerant. Germination is tested again after storage at -20º C.

Strategies for desiccation intolerant seeds

The seeds of some recalcitrant species can be stored at low temperature: 0-5º C, for up to 12 months for temperate species and 16º C for 3-6 months for tropical species (Kew, n.d.). The response to these treatments can be variable; some species germinate if the temperature is too high, and other species may suffer chilling damage. Alternative germplasm conservation techniques such as tissue culture and cryostorage may be suitable strategies but protocols can be specific and difficult to develop.

Rainforest seed collection program

A key part of this seed conservation program is the collection of rainforest seed from wild locations. There are a number of advantages and disadvantages of seed collecting in rainforest habitats as opposed to dry-land communities. Asynchronous fruit ripening both within and between species offers a greater window of opportunity for collection across the year. This is an important factor when considering the need to process seed in a timely manner regarding potential viability loss during handling. A particular disadvantage is that some rainforest species exist as individuals spread across the landscape rather than closely clustered populations. This has implications for the effort required in maximising the genetic diversity of collections.

A high priority for the program is the collection of species that are affected by myrtle rust. First detected in 2010, myrtle rust has spread from the central coast of NSW to Victoria and far north Queensland. The rust has had a dramatic effect on some Myrtaceae species, such as Rhodamnia rubescens, which suffer defoliation and reduced flower and seed production. This may have profound effects for the long-term survival of these species. The seed storage behaviour of R. rubescens and many other affected rainforest species is not yet known. Future collection trips will focus on collecting, assessing seed and where possible storing seed from these species.


Ellis, R. H., Hong, T. D., & Roberts, E. H. (1990). An intermediate category of seed storage behaviour? I. Coffee. Journal of Experimental Botany, 41, 1167-1174.

Gold, K., & Hay, F. (2008). Identifying desiccation-sensitive seeds - Technical Information Sheet No. 10, Millenium Seed Bank Project. Kew: Royal Botanic Gardens.

Hamilton, K. N., Offord, C. A., Cuneo, P., Ashmore, S. E., & Deseo, M. (2010). Conserving Australia’s Unique Rainforest Fruits and Wild Relatives. Paper presented at the International Horticultural Congress, Lisbon.

Kew (n.d.). Seed Storage Behaviour - Difficult Seeds Project. Electronic. Retrieved 20 November 2012, from http://www.kew.org/science-research-data/kew-in-depth/difficult-seeds/resources/index.htm

Pritchard, H. W., Wood, C. B., Hodges, S., & Vautier, H. J. (2004). 100-seed test for desiccation tolerance and germination: a case study on eight tropical plam species. Seed Science and Technology, 32.

Roberts, E. H. (1973). Predicting the storage life of seeds. Seed Science and Technology, 1, 499-514.