Seeds of Rorippa gigantea

Figure 1. Rorippa gigantea (Brassicaceae) seeds. Collections can often appear healthy and viable, but are they?

There is little value in storing poor quality or empty seeds in a seed bank. The aim of the seed cleaning process, followed by the viability and germination tests, is to ensure that each collection held in long-term storage is of a very high quality given the time and resources available.

Initial viability testing; quality control

If a seed is viable (i.e. living) then it is worth banking for future use. To check that seed collections are viable, a small proportion of each collection is tested for viability before seeds are banked.

At the TSCC, seed viability is assessed via a ‘cut test’, however other seed banks use techniques such as x-ray analysis and/or the ‘tetrazolium chloride (TZ) staining test’.

The cut test

Cut testing pyrenes of Coprosma hirtella. All collections stored by the seed bank have a quality assessment made by dissecting a randomly chosen sample of seeds.

Cut testing pyrenes of Coprosma hirtella. All collections stored by the seed bank have a quality assessment made by dissecting a randomly chosen sample of seeds.

The cut test is simple and easily executed. Viability is determined by cutting the seed open and examining the contents. In particular, the seed embryo must appear intact and healthy if the seed

is to go on and germinate. In addition, the endosperm must appear clean and firm1.

The cut test would appear quite straightforward, however even this technique can be challenging when working with the seed of wild plant species. Seeds vary quite considerably in length from <1mm to several centimetres. Dissection of tiny seeds requires delicate execution with precision tools and examination requires a good microscope. To further complicate matters the internal morphology of seeds is highly variable (fig. 3), and colour and texture of storage tissue (cotyledons and endosperm) is influenced by the storage compounds used by the seed. So an appearance that indicates high viability in one species can indicate low viability in another. Learning to differentiate ultimately comes with experience.

Seed Embryo Variation

Figure 3. Illustration showing a small sample of the variation observed in internal seed morphology. Seed embryo are yellow. Shaded white is endosperm.

  1. Tiny Undifferentiated embryo (e.g. Ranunculaceae)
  2. Small differentiated embryo (e.g. Apiaceae)
  3. Linear embryo (e.g. Oleaceae)
  4. Bent embryo (e.g. Brassicaceae)
  5. Large investing embryo (e.g. Mimosaceae)

Other techniques

X-ray analysis

For seed workers with access to an x-ray machine, x-ray analysis can be a very efficient and non-destructive method of assessing seed quality1. A good x-ray image will reveal details of seed fill, insect infestation and also size or absence of the embryo. To a lesser degree it may also reveal variation in storage tissue quality. However, this method is not as revealing as the simple cut-test.

The TZ test

Seed viability can also be determined over three consecutive days with the tetrazolium chloride (TZ) staining technique2. Hydrated seeds are perforated and placed overnight into a solution of tetrazolium chloride. This compound is colourless until in the presence of living tissue when it is converted to a red dye. This causes living tissue to stain red. Seed staining patterns are examined the following day to determine percentage seed viability.

Usually only uniformly stained red / dark pink embryos are considered ‘viable’. However, staining patterns must be interpreted carefully and, depending on the species and the level of viability, the TZ test can be highly subjective.


  1. Terry J, Probert RJ and Linington SH. 2003. Processing and maintenance of the Millennium Seed Bank collections. In: Seed Conservation. Turning Science into Practice. RD Smith, JB Dickie, SH Linington, HW Pritchard and RJ Probert, eds. Royal Botanic Gardens Kew, 307-325.
  2. International Seed Testing Association (ISTA). 2003. ISTA Working Sheets on Tetrazolium Testing. Vols. I and II. ISTA, Bassersdorf, Switzerland.