2014 AOSA/SCSTMid-West Region

SORGHUM WARM TEST REFEREE

A comparison of the 10 day AOSA warm test to a 7 day warm test.

Purpose

The purpose of this referee was to determine if the AOSA sorghum warm test procedure can be shortened from 10 days at 20°-30°C to 7 days.

Potential Benefits• To reduce the turnaround time by 3 days• To make this test easier to schedule • To make the samples easier to evaluate, better

uniformity within and among labs• To provide more planting and evaluation days per

week

Procedure• Follow the AOSA Rules for Sorghum bicolor, 20°C-30°C

for 10 days. Then use the same method, but substitute 7 days instead of the recommended 10 day test duration.

• Carryover sorghum was used so there would be no need to prechill any of the samples. Twelve samples (6 lots duplicated and randomized) were sent to each participant.

• This larger set of samples provided more data points to determine uniformity within and among participating laboratories.

• Results will presented at the 2014 AOSA/SCST meeting

Survey• Media a. All respondents but 5 used rolled towels. Other media used

included TB, TC and agar*.

b. One respondent using TC had insufficient plumule development at 4 days for germination.

c. One respondent using TB had substantially lower 4 day counts than respondents using RT at 20-30C.

• Temperature

a. Most respondents used 20-30C, four respondents used 25C

Discussion• The 28 participants provided a good cross section of regulatory and commercial

seed testing laboratories to compare the 7 and 10 test durations for the sorghum warm test. (25 participants 2013 referee)

• Test durations proved to be statistically equivalent. • 75% of the labs finalize their tests early as a routine, which is not the intention of

the current recommended procedure (2013 referee survey).• If the procedure was changed to 7 day test duration and a problem exists with a

sample the test can all ways be extended to 10 days.• There seemed to be a slight edge for optimum results for the 7 day test when it

comes to testing lower quality seed where fungal disease may interfere with the final evaluations.

• It was interesting to see how many different sizes of towels were used, especially the larger sizes 12”x16”, 12”x18” and 12”x 24” given the small seed size of sorghum. The majority of the labs used 38# paper.

Within Lab VariabilityTests Comparisons 10 Day OOT 7 Day OOT 10 vs. 7day OOT

336 26 15 35

Within a laboratory, on the same sample, the lab was almost twice as likely to be out of tolerance when testing on a 10 day basis as a 7 day basis.11% of the time a lab was out of tolerance between the 10 day test and the 7 day test.

Paired T Test 10 Day vs.. 7 Day

20100-10-20-30

X_

Ho

Differences

10 Day Vs. 7 Day Paired T Test Boxplot of Differences(with Ho and 95% t-confidence interval for the mean)

ANOVA Results of 2 Test Duration, 6 Samples, 24 Participants and

• Nested ANOVA: Germination Ave versus Days, Sample ID, Referee Participant # • Analysis of Variance for Germination Ave• Source DF SS MS F P• Days 1 79.99 79.99 0.004 0.949 NS• Sample no 10 183332.63 18333.26 127.604 0.000 ***• Paricipant 324 46550.57 143.67 14.946 0.000 ***• Error 336 3229.95 9.61 • Total 671 233193.14

• There was no statistical difference between test duration• The referee was designed for statistical difference between samples• Unfortunately there were statistical differences between respondents

Participants had the most difficult time with very low quality sorghum

Comparison of Overall Results vs. Agar (one participant)

Agar showed some potential, slightly higher germination was observed with agar

Additional Participants may close the gap between testing on paper toweling and creped cellulose blotters and agar

There were insufficient comparisons for valid statistics

5.02.50.0-2.5-5.0-7.5-10.0-12.5

X_

Ho

Differences

Overall 10 Day Vs. Agar Boxplot of Differences(with Ho and 95% t-confidence interval for the mean)

5.02.50.0-2.5-5.0-7.5-10.0-12.5

X_

Ho

Differences

7 Day vs. Agar Boxplot of Differences(with Ho and 95% t-confidence interval for the mean)

Conclusions• The 7 day and 10 day warm test results showed no

statistical differences in both years of a large scale trial

• Currently many labs finalize before the 10 day, 75% of labs finalize early according to previous surveys

• Poor quality samples have their optimum germination prior to the 10 day evaluation for most labs

• Some labs like a first count

RecommendationsBased on the referee data a rule change can be proposed to shorten the current 10 day sorghum warm test to a 7 day duration.

a. This would provide a 3 day quicker turnaround on data

b. Make the tests easier to schedulec. Tests can be planted and evaluated everydayd. Replications should be easier to evaluate

providing more uniform datae. New crop seed would still need to pre-chilled