Herbarium Curation Standards

Report to CHAH 2000

Background

A statistical sampling of herbarium specimens to assess physical and scientific curation standards was commenced in 1999. The National Herbarium of New South Wales evaluates curatorial standards with this method every three months as part of regular reporting to their Trust. Each State and Territory herbarium has been asked to complete at least one sampling prior to the annual CHAH meeting.

The method has many weaknesses, not the least being the small sample size. NSW now reports to its Trust using a rolling annual average to help smooth out the anomalies in the data. Once we have accumulated a few year’s State and Territory data it will be useful to use rolling averages to report on the national summary as well. In this report we simply document the results of the two years in tabular and graphic form.

Method

A stratified sample of 25 specimens is assessed in each herbarium. The recommended sampling procedure is:

• Divide the herbarium into major units (NSW uses the four levels) and then calculate how many samples need to be made in each unit to ensure even coverage of the collection (i.e. divide 25 by the proportion of boxes or cupboards in each unit).
• Divide the number of rows in each unit by the number of samples to be taken, and then locate the first row of each of these subunits
• Number the boxes to be sampled along the row, then up the columns progressively
• Sample the third specimen in the box or cupboard (if box/cupboard empty, go up to the next one with specimens; if box with less than 3 specimens, sample 2^nd, or if only 1, 1^st specimen)
• For each succeeding evaluation the sample box or cupboard should be progressed by one, so that a representative coverage of the collection will be taken over time.
• As a separate task (the ‘retrievability test’), generate a random list of 25 specimens from the herbarium database and check whether each one is in its correct box/shelf.

For the specimen selected, assess the following criteria as ‘0’ or ‘1’:

1. name on specimen consistent with label on box and folder = 1
2. specimen databased = 1
3. specimen on archival sheet = 1
4. specimens mounted/unmounted/bagged (to acceptable standard for group) = 1
5. specimen undamaged = 1

For the retrievabilty test, score as ‘1’ if the selected specimen is in the correct box or cupboard, and ‘0’ if not.

In addition to the evaluation criteria, three additional fields of information should be gathered that will eventually assist in valuation and recollection. These fields are not currently reported to CHAH.

• number of specimens in box counted
• where specimen was collected (eg NSW, WA, Africa)
• if specimen was an original collection or an exchange specimen

Results of 2000 survey

*Includes specimens databased but unshelved, in botanist’s rooms, and misshelved.

Recommendations

1. Herbaria continue to provide annual data to NSW
2. 3-5 year rolling averages to be used to present data (to overcome small sample size)