Technical protocol for the ISTA Experiment on Sampling and Testing ...

<strong>Technical</strong> <strong>protocol</strong> <strong>for</strong> <strong>the</strong> <strong>ISTA</strong> <strong>Experiment</strong> on Sampling and 

Testing Seed Mixtures 

Background 

The present <strong>ISTA</strong> Rules only cover <strong>the</strong> issuance of <strong>ISTA</strong> Certificates <strong>for</strong> seed lots of 

one variety or species. The following experiment shall provide data to establish <strong>the</strong> 

necessary in<strong>for</strong>mation that may allow a Rules change proposal <strong>for</strong> <strong>the</strong> testing of seed 

mixtures. 

1. Scope of <strong>the</strong> experiment 

This experiment is directed at checking <strong>the</strong> sampling and testing of seed mixtures. 

The experiment does not include ei<strong>the</strong>r <strong>the</strong> certification of <strong>the</strong> mixture lot or <strong>the</strong> 

issuance of <strong>ISTA</strong> Certificates <strong>for</strong> <strong>the</strong> lot. 

All accredited <strong>ISTA</strong> member laboratories can participate in this experiment. 

2. Course of <strong>the</strong> experiment 

1. <strong>ISTA</strong> laboratories should send a short in<strong>for</strong>mal notice about <strong>the</strong>ir 

willingness to participate in this experiment to <strong>the</strong> <strong>ISTA</strong> Secretariat. 

2. Participants are expected to sample and test mixture seed lots during <strong>the</strong> 

course of <strong>the</strong> experiment. However, <strong>the</strong>re is no minimum number of seed 

lots that have to be sampled and tested by participants. 

3. For sampling and testing of seed mixture lots, <strong>the</strong> methods used shall be 

as described under 3. of this <strong>protocol</strong>. 

4. Participants are expected to report <strong>the</strong> test results regarding purity, 

composition and thousand seed mass as well as <strong>the</strong> declared composition 

of <strong>the</strong> mixture lot toge<strong>the</strong>r with some additional in<strong>for</strong>mation on <strong>the</strong> attached 

<strong>for</strong>m to <strong>the</strong> <strong>ISTA</strong> Secretariat. 

5. Participants cannot get a financial refund from <strong>ISTA</strong> <strong>for</strong> sampling and 

testing expenses. 

6. The <strong>ISTA</strong> Secretariat will <strong>for</strong>ward submitted <strong>for</strong>ms to Michael Kruse, 

member of <strong>the</strong> Bulking and Sampling Committee, <strong>for</strong> statistical analysis. 

The goal of this analysis is to quantify <strong>the</strong> distribution of deviations between 

<strong>the</strong> declared composition and <strong>the</strong> test results obtained by <strong>the</strong> proposed 

method. This distribution will be <strong>the</strong> basis used to establish tolerance tables 

as part of a Rules Change Proposal. 

7. The <strong>Experiment</strong> will run up to <strong>the</strong> end of 2009. The results of <strong>the</strong> 

experiment shall be reported during <strong>the</strong> <strong>ISTA</strong> Congress 2010. The identity 

of <strong>the</strong> mixtures, companies, countries and <strong>ISTA</strong> laboratories will not be 

evaluated and published.

3. Methods <strong>for</strong> sampling and testing seed mixtures 

Sampling 

The <strong>ISTA</strong> Rules Chapter 2 shall be followed <strong>for</strong> sampling mixture seed lots and <strong>for</strong> 

obtaining working samples <strong>for</strong> testing by taking into account <strong>the</strong> following additional 

regulations, restrictions and changes. 

Definitions 

(The <strong>ISTA</strong> Rules 2.2) 

2.2. Seed Mixture Lot 

A seed mixture lot is a specified quantity of seed that is physically and uniquely 

identifiable and includes by declaration seed of two or more species. 

2.2 Mixture Components 

A mixture component is seed of a species or variety that is given in <strong>the</strong> declaration of 

<strong>the</strong> seed mixture lot. Inert Matter, e.g. additives to improve sowability, are also 

regarded as mixture components if given as such in <strong>the</strong> declaration. 

The Seed Mixture Lot 

2.5.4.1 Size of <strong>the</strong> lot 

(iii) For mixtures where all <strong>the</strong> components have <strong>the</strong> same maximum seed lot size, 

<strong>the</strong> mixture will have this maximum seed lot size. Where <strong>the</strong> mixture contains 

components that have different maximum seed lot sizes <strong>the</strong> components are grouped 

according to <strong>the</strong> maximum seed lot size and <strong>the</strong> maximum seed lot size of <strong>the</strong> 

mixture is equal to that of <strong>the</strong> most abundant group in terms of percentage by weight 

of <strong>the</strong> lot. If a seed mixture lot includes coated seeds, <strong>the</strong> mass of <strong>the</strong> coated seeds 

is used <strong>for</strong> <strong>the</strong> determination of <strong>the</strong> lot size and not <strong>the</strong> mass of <strong>the</strong> seed without 

coating material. Inert additives, if given in <strong>the</strong> declaration, are included in <strong>the</strong> lot 

size. 

Appendix B of <strong>the</strong> <strong>ISTA</strong> Rules as well as <strong>the</strong> “<strong>ISTA</strong>/ISF Herbage Seed Lot Size 

<strong>Experiment</strong>” are not extended to seed mixtures. 

2.5.4.1.1 Declaration of <strong>the</strong> seed mixture lot 

The applicant must provide in<strong>for</strong>mation about <strong>the</strong> 

- identity of <strong>the</strong> mixture components, i.e. <strong>the</strong> genus and, if applicable, <strong>the</strong> species 

within <strong>the</strong> genus and/or <strong>the</strong> varieties within <strong>the</strong> species (or <strong>the</strong> identity of <strong>the</strong> 

additives) and 

- <strong>the</strong> mass proportions of <strong>the</strong> mixture components. 

2.5.1.2 Sampling Intensity 

Sampling intensity <strong>for</strong> seed mixtures is as described in The <strong>ISTA</strong> Rules, Tables 2.1 

and 2.2, however it is strongly recommended to take more samples than <strong>the</strong> 

minimum number and when sampling containers to take care that all positions in <strong>the</strong> 

containers (i.e. center, top, bottom, corners, edges) are represented when primary 

samples are taken.

2.5.1.3 Taking primary samples 

Seed mixtures. Containers or positions in <strong>the</strong> seed mixture lots to be sampled should 

be selected according to a systematic plan. If <strong>the</strong> sequence of <strong>the</strong> containers is not 

detectable, <strong>the</strong> containers to be sampled should be selected with equal distances 

from each o<strong>the</strong>r. 

For taking primary samples from seed mixture lots in containers, methods listed in 

2.5.1.3 must be used. It is also possible to obtain <strong>the</strong> primary sample from <strong>the</strong> 

container by one of <strong>the</strong> sample reduction method listed below. 

For taking primary samples from <strong>the</strong> stream of seeds, automatic or hand sampling 

methods may be used according to <strong>the</strong> <strong>ISTA</strong> Rules. 

2.5.4.4 Submitted sample 

The size of <strong>the</strong> submitted sample shall be equal to <strong>the</strong> sample size of <strong>the</strong> component 

with <strong>the</strong> greatest sample size according to <strong>the</strong> <strong>ISTA</strong>-Rules Table 2A. 

2.5.4.5 Sample reduction 

For reduction of samples taken from seed mixture lots, <strong>the</strong> methods listed in 2.5.2.2 

of <strong>the</strong> <strong>ISTA</strong> Rules may be used. 

The variable divider is not recommended <strong>for</strong> seed mixtures containing species with 

multiple seed units since <strong>the</strong>y may cause cross-contaminations. 

For seed mixtures, <strong>the</strong> hand halving method or <strong>the</strong> spoon method may be used when 

hand halving or spoon method is allowed <strong>for</strong> at least one of <strong>the</strong> components in <strong>the</strong> 

mixture seed lot. 

Chapter 3. Purity 

3.2.4 Component species 

In a seed mixture, <strong>the</strong> pure seed of each species stated by <strong>the</strong> applicant to be part of 

that mixture shall constitute a component species. The pure seed assessment shall 

follow <strong>the</strong> appropriate PSD of each species. 

3.5.1 Working sample 

The purity analysis shall be made on a working sample taken from <strong>the</strong> submitted 

sample in accordance with 2.5.2, <strong>the</strong> submitted sample having been received in 

accordance with 2.5.1. Except <strong>for</strong> species of Poaceae (Gramineae) <strong>for</strong> which <strong>the</strong> 

uni<strong>for</strong>m blowing method is to be used, <strong>the</strong> size of <strong>the</strong> working sample shall be: 

ei<strong>the</strong>r a weight estimated to contain at least 2500 seed units, 

or not less than <strong>the</strong> weight indicated in column 4 of Table 2A. 

For seed mixtures, a calculation is required to define <strong>the</strong> right sample size on <strong>the</strong> 

basis of <strong>the</strong> weights of table 2.A assumed as weights of 2500 seeds of each species. 

The following <strong>for</strong>mula can be used to obtain a mixture sample of 2500 seeds: 

Working sample weight <strong>for</strong> mixture = 100 / (%C 1 /PWC 1 + %C 2 /PWC 2 + %C 3 /PWC 3 + 

%C 4 /PWC 4 + …) 

Where: 

%C i = percentage of <strong>the</strong> component “i” in <strong>the</strong> mixture 

PWC i = purity sample weight of component “i” (table 2.A)

To illustrate <strong>the</strong> use of this equation, see <strong>the</strong> following example: 

Species in <strong>the</strong> mixture 

Component 

percentage 

Weight of purity 

working sample 

Table 2.A 

Weight of 

each species 

in <strong>the</strong> sample 

of calculated 

weight 

Number of 

each species 

in <strong>the</strong> sample 

of calculated 

weight 

C1 Festuca arundinacea 30 5 0,97 484 

C2 Festuca rubra 30 3 0,97 806 

C3 Lolium perenne 30 6 0,97 403 

C4 Poa pratensis 10 1 0,32 806 

Total 3,23 2500 

Working sample weight <strong>for</strong> mixture 

= 100 / (%C 1 /PWC 1 + %C 2 /PWC 2 + %C 3 /PWC 3 + %C 4 /PWC 4 ). 

= 100/ (30/5 + 30/3 + 30/6 + 10/1) 

= 100/ (6+10+5+10) 

= 100/31 

= 3.23 

3.2.1 Pure seed 

The pure seed shall refer to <strong>the</strong> species stated by applicant, or found to predominate 

in <strong>the</strong> test, and shall include all botanical varieties (varietas) and cultivars of that 

species (or more than one species, in <strong>the</strong> case of seed mixtures) including: 

3.5.3 Seed Mixtures 

The pure seed shall be separated into <strong>the</strong> component species and each component 

species shall be weighed in grams to <strong>the</strong> minimum number of decimal places 

necessary to calculate <strong>the</strong> percentage to one decimal place (3.5.1.A). 

3.5.2.A Separation 

3.5.2.A.5 Uni<strong>for</strong>m blowing method 

This method is obligatory <strong>for</strong> Poa pratensis, Poa trivialis and Dactylis glomerata, 

except when <strong>the</strong>y <strong>for</strong>m components of a seed mixture. The uni<strong>for</strong>m blowing method 

shall not be used <strong>for</strong> species components of seed mixtures. 

Chapter 4. O<strong>the</strong>r seeds by number 


a) The size of <strong>the</strong> working sample shall be ei<strong>the</strong>r a weight estimated to contain at 

least 25000 seed units or not less than <strong>the</strong> weight indicated in Table 2A, Part 1, 

column 5. For seed mixtures <strong>the</strong> working sample size shall be ten times <strong>the</strong> purity 

sample working size determined according to <strong>the</strong> calculation in 3.5.1.

17.5.A Reporting results 

17.5.A.23.1 Purity 

The percentage by weight of pure seed, inert matter, ant o<strong>the</strong>r seeds shall be given 

to one decimal place. In <strong>the</strong> case of a seed mixture, <strong>the</strong> pure seed percentage shall 

be calculated using <strong>the</strong> total weight of <strong>the</strong> pure seed of all component species. The 

percentage by weight of <strong>the</strong> pure seed of each component species shall be included 

under „O<strong>the</strong>r determinations‟. If <strong>the</strong> percentage by weight is below 0.05, it shall be 

stated as „TR‟ (<strong>for</strong> „Trace‟). If no „Inert matter‟ or „O<strong>the</strong>r seeds‟ are found, this should 

be entered as „0.0'. The scientific name shall be used <strong>for</strong> every species of „O<strong>the</strong>r 

species‟ found in <strong>the</strong> purity test. 

Chapter 5. Germination 


Four hundred seeds in replicates of 100 are taken at random from <strong>the</strong> pure seed 

(5.3) and spaced uni<strong>for</strong>mly and adequately apart on <strong>the</strong> moist substrate. Replicates 

may be divided into sub-replicates of 50 or 25 seeds depending on <strong>the</strong> size of seeds 

and <strong>the</strong> amount of space needed between <strong>the</strong>m. For seed mixtures, four hundred 

seeds are planted from <strong>the</strong> pure seed of each component species, except that <strong>for</strong> 

species representing less than 15% of <strong>the</strong> mixture, 2 x 100 seeds are planted. For 

species representing less than 5% of <strong>the</strong> mixture germination testing is not carried 

out except on <strong>the</strong> specific request of <strong>the</strong> customer. When such a request is made <strong>the</strong> 

germination test is carried out on 2 x 100 seed replicates of seed. If sufficient seed 

are not available from <strong>the</strong> pure seed fraction <strong>the</strong> test will be carried out on <strong>the</strong> all of 

<strong>the</strong> pure seed of <strong>the</strong> species in <strong>the</strong> pure seed fraction. 

5.9 Reporting results 

The following items shall be entered in <strong>the</strong> report when reporting <strong>the</strong> result of a 

germination test: 

duration of test. 

percentage of normal, abnormal seedlings, hard seeds, fresh seeds and dead 

seeds. If <strong>the</strong> result <strong>for</strong> any of <strong>the</strong>se categories is found to be nil, it shall be 

entered as „0'. 

In <strong>the</strong> case of a seed mixture, <strong>the</strong> results of <strong>the</strong> germination test are reported to <strong>the</strong> 

nearest whole number under <strong>the</strong> categories normal seedlings, hard seed, fresh seed, 

abnormal seedlings, and dead seeds. No germination results are reported in <strong>the</strong> 

„Germination‟ section of a report. The germination results <strong>for</strong> each component 

species shall be included under „O<strong>the</strong>r determinations‟ under <strong>the</strong> categories normal 

seedlings, hard seed, fresh seed, abnormal seedlings, and dead seeds. When 400 or 

200 seeds have been tested <strong>the</strong> results are reported as a percentage and <strong>the</strong> 

number of seeds tested is also reported. Where less than 200 seeds have been 

tested <strong>the</strong> actual number of seeds in each category are reported toge<strong>the</strong>r with <strong>the</strong> 

total number of seeds tested.

Chapter 10: Weight Determination 

10.5.4 Seed Mixtures 

For seed mixtures <strong>the</strong> 1000 seed weight is determined by counting <strong>the</strong> number and 

weight of each seed component in <strong>the</strong> purity sample ra<strong>the</strong>r than counting replicates 

as in 10.5.3. 

10 .6 Calculation and expression of results 

If counting <strong>the</strong> number of seed of each component of a mixture in a purity sample <strong>the</strong> 

1000 seed weight is calculated according to <strong>the</strong> weight and number of seeds of each 

component 

e.g. 

Species in <strong>the</strong> mixture 

Number of 

seed in purity 

sample 

Weight of 

seed 

Thousand Seed Weight 

Festuca arundinacea 484 0.9677 = (0.9677/484)*1000 = 2.00 

Festuca rubra 806 0.9677 1.20 

Lolium perenne 403 0.9677 2.40 

Poa pratensis 806 0.3226 0.40 

Chapter 17. Certificates 

As <strong>ISTA</strong> certificates are not allowed to be issued to a <strong>for</strong> a seed lot of a 

mixture, chapter 17 should be referred to as ‘Not applicable’.

2.5% TOLERANCE RANGES TO BE USED TO COMPARE REPLICATE RESULTS OF 

GERMINATION TESTS 1 

Average Number of Category * in Sample 

Seed Tested: 

2x100 Tolerance 

Seed Tested: 

4x100 Tolerance 

A B C D 

99 2 4 5 

98 3 5 6 

97 4 6 7 

96 5 6 8 

95 6 7 9 

94 7 8 10 

93 8 8 10 

92 9 9 11 

91 10 9 11 

90 11 9 12 

89 12 10 12 

88 13 10 13 

87 14 11 13 

86 15 11 14 

85 16 11 14 

84 17 11 14 

83 18 12 15 

82 19 12 15 

81 20 12 15 

80 21 13 16 

79 22 13 16 

78 23 13 16 

77 24 13 17 

76 25 13 17 

75 26 14 17 

74 27 14 17 

73 28 14 17 

72 29 14 18 

71 30 14 18 

70 31 14 18 

69 32 14 18 

68 33 15 18 

67 34 15 18 

66 35 15 19 

65 36 15 19 

64 37 15 19 

63 38 15 19 

62 39 15 19 

61 40 15 19 

60 41 15 19 

59 42 15 19 

58 43 15 19 

57 44 15 19 

56 45 15 19 

55 46 15 20 

54 47 16 20 

53 48 16 20 

52 49 16 20 

51 50 16 20 

1 Derived from 

Miles S. R. (1963) Handbook of tolerances and measures of precision <strong>for</strong> seed testing Proc.Int. Seed Test. Ass. 28 pg 644 

* Category can be number of: normal or abnormal seedlings; hard, fresh, dormant or dead seed; or seed with a particular 

disease

Technical protocol for the ISTA Experiment on Sampling and Testing ...

Create successful ePaper yourself

Delete template?

Save as template?