Jump to content
Toggle menu
Toggle preferences menu
Toggle personal menu
Not logged in
Your IP address will be publicly visible if you make any edits.
ℹ️
This is a project information page for Landrace.Wiki. It describes the project's standards, processes, or structure. It is not an encyclopaedic article.

This page explains how to photograph and measure the seeds of an accession, so that seeds photographed by different people in different places can be compared.

What botanists call the seed of cannabis is an achene: a small dry fruit holding one seed, sometimes still wrapped in a thin, papery perianth from the flower. This page says "seed", as growers do. The measurements and observations go in the Seeds fields of the accession form and show in the accession's infobox under Botanical Characteristics. The words used for colour and marbling are those of the UPOV hemp and cannabis test guideline (TG/276/2); the perianth and base words follow published descriptions of cannabis seeds.

The seed fields describe the seed only. The wiki does not decide from them whether a population is wild, cultivated or mixed.

What you need

  • Dry, ripe seed of the accession: hard seeds from a finished harvest, stored dry. Seed that is still green or soft gives the wrong size and colour.
  • A ruler marked in millimetres, or a printed scale bar checked against such a ruler.
  • A plain, matte background: white or mid-grey card, never patterned, glossy or coloured. Use the same background for every accession you photograph.
  • Even, soft light: daylight from a window, or two lamps from opposite sides. No direct sun and no flash, which leave hard shadows and shine.
  • A camera or phone that can focus close, on a stand or held level and square above the seeds.
  • Tweezers, to turn the seeds over.

Which seeds, and how many

  • Take the seeds from the accession's own seed. If more than one seed lot exists, say in the photographs' descriptions which lot they came from.
  • Take the seeds at random. Pour some out and use those, rather than choosing the biggest, darkest or best-looking.
  • 50 seeds for the measurements and the group photograph.
  • 10 of those 50, again at random, for the close-up photographs.
  • Leave out broken or empty seeds, and say how many there were.

The photographs

Put the scale bar in every photograph, lying on the background beside the seeds, at the same height as the seeds, with the millimetre marks sharp. A photograph without a scale cannot be measured.

  1. Group photograph. Lay the 50 seeds flat in rows, not touching, each with its base (the end where it was attached to the plant) pointing up. Photograph them from straight above, filling the frame with the seeds and the scale.
  2. Close-ups, both faces. A cannabis seed is flattened, with two broad faces and a ridge around the edge. Photograph each of the 10 seeds lying on one face, base up, then turn it over with the tweezers and photograph the other face in the same position. One seed per photograph, or the 10 in a row, each one large enough to show the perianth, the marbling and the shape of the base.
  3. Optional: the edge. One photograph of a few seeds standing on their edge shows how flat or plump they are.

Keep the camera at the same height for all the photographs of one accession, and keep the light the same. Do not correct the colours afterwards beyond a plain white balance, so seed colours can be compared.

Measuring

Measure from the group photograph (free software such as ImageJ or Fiji can do this against the scale bar), or with calipers.

  • Length: from the base to the tip, along the longest line.
  • Width: across the widest point, at right angles to the length, in the same view.
  • Length to width: for each seed, its length divided by its width. Enter the average of these.

Enter the mean length and the mean width in millimetres, to one decimal place, the mean length to width to two decimal places, and the number of seeds measured.

The weight of 1,000 seeds belongs to the seed lot's inventory, not to these fields: weigh a counted number of seeds (at least 100), scale the weight up to 1,000 seeds, and enter it as the inventory's 1,000-seed weight.

Colour, marbling, perianth and base

Look at the seeds dry, in daylight, with the close-ups beside you. If the seeds of one accession differ, enter every state you see.

Field States What to look at
Seed colour Light grey, Medium grey, Grey brown, Yellowish brown, Brown The colour of the seed where no perianth covers it.
Seed marbling Weak, Medium, Strong The dark mottled pattern on the seed, and how strong it is.
Perianth Persistent, Partly persistent, Absent Persistent: the thin papery layer covers the seed. Partly persistent: it stays in patches or stripes. Absent: the seed is bare.
Seed base Elongated, Blunt Elongated: the base is drawn out into a short stub. Blunt: no stub.

Naming the files

Name each file after the accession, then the view, then the year:

  • <accession ID>-seeds-group-2026.jpg for the group photograph;
  • <accession ID>-seed-03-face-a-2026.jpg and <accession ID>-seed-03-face-b-2026.jpg for the two faces of seed 3;
  • <accession ID>-seeds-edge-2026.jpg for the edge view.

If one accession's seeds are photographed again later, the year keeps the two sets apart.

Before uploading

Remove the location and camera data from every file first, as Help:Media and safety explains. Seed photographs carry no landscape, but a phone often writes the place where the photograph was taken into the file. Upload through Special:Upload, saying in the summary who took the photographs, when, and from which seed lot the seeds came, then add them to the accession's page.

See also