collection management

We have developed the following guidelines for preparing and documenting shipments of lichen and bryophyte specimens to The New York Botanical Garden for digitization for the Lichen-Bryophyte TCN Project.  The goal is to make the transfer of specimens for this project as accurate and efficient as possible, and to minimize the amount of time that the specimens are away from their home institutions.

Preparation of specimens for shipment:

Barcoding of specimens

As soon as the project barcodes have been received at New York, we will distribute these to each institution. 

All specimens sent to New York for digitization should be barcoded in advance by the home institution.  The barcodes should be positioned so they are visible without opening the packet or any  packet flaps, and make sure that there is a white border of at least ¼” inch around the barcode lines – generally the barcode labels have at least this much of a border  on the label itself.  The barcodes should be applied in numerical sequence within a species, preferably within a genus.  Some institutions have collections both mounted on sheets and as individual specimens – these should be barcoded in series within the type of preparation, but not across preparation types.  In other words, barcode all the specimens of Bryum argenteum mounted on sheets in sequence, and barcode all specimens of Bryum argenteum on packets in sequence, but the sequence doesn’t have to  be continuous across mounted and non-mounted specimens.

Spreadsheet inventory of shipments

A spreadsheet should accompany the shipment of specimens that includes the following information: barcode number, preparation type (packet or sheets), genus, species, subspecific  taxon.  If you use Excel or comparable spreadsheet software, you can use functions provided with the software to advance the barcode number automatically and copy data from the preceding entry.   

Below is a mock up of the spreadsheet columns with made- up sample data:

Barcode number

Prep. Type (mounted or packeted, if herbarium has both)

“Filed as” Genus

“Filed as” species

Subspecific qualifier  (if present)

Subspecific name (if present)

NY00111222

p

Bryum

argenteum

var.

vulgaris

NY00111223

p

Bryum

argenteum

var.

vulgaris

 

 

 

 

 

 

 

The names used in the genus and species columns should be those under which the specimens are filed in your herbarium.   The spreadsheet will be used not only to create the skeletal data records for your specimens, but will also serve as inventory control, which will facilitate the checking in and out of specimens when they are sent out and received. The time you take to make the list will be largely offset by the time you will save counting and re-counting specimens pre-shipment and and post return , or, heaven forbid, figuring out where the discrepancy is, if these counts don’t match!

Specimens mounted on herbarium sheets should be placed in thin paper folders labeled with the genus and species name – including the barcode range on the folder is helpful, though not required.  For maximum protection of the specimens, folders should be grouped in 12—16 inch bundles that are sandwiched between corrugates and tied with two evenly-spaced strings.  Bundles should be placed in 12—20 inch high new boxes with a bursting strength of 275 lbs. (approx.).  We will use these same materials to return the specimens to their home institutions.

Specimens not mounted on sheets (that is, loose packets) can be prepared in one of two ways:

Line up in species and barcode order in cardboard trays that fit comfortably inside your shipping boxes.  Use some kind of marker indicating the “filed as” name for each species.  This marker is ideally a differently-colored piece of stiff paper or cardboard cut to the size of the packets that has the species name on it and marks the beginning of the sequence specimens with that “filed as” name.  The trays can be stacked inside each box, but should be separated by corrugates to prevent damage to packets.

Bundled using paper such as unprinted newsprint in groups of 5 – 10 specimens each, with each bundle labeled with the “filed as” name.  This method is a bit more time-consuming, but will contain the specimen contents, and reduce the potential of   contamination, should specimen fragments or dirt fall out of the packets during transfer.

Transaction Management:  We have decided to create a new transaction category for the shipments we receive of specimens for digitization.  We will call these “Incoming Loans for Digitization.”  Unlike other incoming loans to NYBG, these will not be assigned to a particular researcher.  All loans for digitization should be addressed to:

 

Dr. Barbara Thiers

Director, William and Lynda Steere Herbarium

The New York Botanical Garden

2900 Southern Blvd

Bronx, NY 10458-5126

 

LOAN OF [Bryophytes] [Lichens] FOR DIGITIZATION

 

 

The barcode will be used as the primary identifier linking together specimen images, the online specimen records, and the records within the herbarium’s central database. Due to the important role of this identifier, regularly using a barcode reader will avoid common transcription errors that occur when keying numbers by hand. For most of the participating collections, barcodes will be purchased by the TCN. Institutions already purchasing barcodes through their preferred provider will be given funds to continue.

Barcode Requirements

  • Unique Identifiers: The barcode must uniquely identify the specimen within the collection. It is important that methods are in place that ensures that no two specimens receive the same barcode identifier, which is particularly important for collections that make their own barcodes using a barcode printer.
  • Must be stable: Ideally the barcode should never be modified. Therefore, ensure that barcode is truly unique at the time of assignment so that it doesn’t have to be reassigned.
  • Where to Place: The final location for the barcode should be a location that is easy to scan without the need of opening a packet or disturbing the specimen any more than is needed. Remember, that the most important role for barcodes is that they will supply an easy and reliable method for identifying a specimen when preforming curatorial management tasks. If one is processing a group of specimens for a loan, one should be able to quickly go through a stack and scan each specimen without much trouble. Note that OCR returns of text immediately to the right or left of the barcode can be problematic. In order to reduce OCR "noise" that a barcode can create, it is preferable if the barcode is above or below the label with no adjacent text on a horizontal plane.

Barcode Recommendations and Comments

  • Global Unique Identifiers (GUID): Ideally, the barcode identifier would uniquely identify that specimen relative to all other specimens found worldwide. The current TDWG recommendations for creating unique herbarium identifiers are to use: <institution code>:<collection code>:12345678. For more information on the Darwin Core recommendations: http://rs.tdwg.org/dwc/terms/index.htm#occurrenceID
  • Format: Barcodes are often alphanumeric. The most common barcode standard used for herbarium specimens are Code 39. It’s a good idea to avoid using special characters (!@#$%&) and spaces when possible. The size of the barcode label will depend on the space available on the specimen. Smaller sized lichen and bryophyte specimens may make a barcode of the full GUID (ca 18 digits) impractical. In this case, the barcode might only represent the numeric portion of the identifier or have a collection code of only one to two digits.
  • Set Number of Digits: Barcodes with a uniform number of digits aids in catching and avoiding errors within the database. For the numeric portion of the barcode, collections typically use 7-8 digits with left padded zeros. For example, ABC herbarium with 275,429 lichen specimens might have a barcode sequence from ABC:L:0000001 to ABC:L:0275429. If the collection chose to match barcode and accession number, a specimen with an accession number of 1234 would be something like ABC:L:0001234.  
  • Readable Identifier: Include the human readable digits with the barcode so that one can read the identifier without the need of a barcode reader.
  • Ordering Barcodes: When ordering preprinted barcodes, it’s a good idea to order enough extra barcodes to cover incoming specimens for the next 10 years or more.
  • Pre-printed -vs- Barcode Printing: This TCN project recommends using pre-printed barcodes.
    • Pre-ordered barcodes bought in quantity are typically the cheaper way to go in the long run. This option avoids the need to buy and maintain printer, ink, blank barcodes, etc.
    • Purchasing pre-printed barcodes as a batch ensures that each barcode is unique. If one prints their own, ensure that your database application restricts the entry of duplicate barcodes.
    • Affixing pre-printing barcodes to specimen is generally fast since one doesn’t have to wait for printer to pop-out barcode.
    • Barcode printers make it easier to print 3 barcodes of the same number for specimens with 3 sheets. However, a regular printer can be used for printing an occasional barcode.
    • Using a barcode printer may be easier if a collection wishes to match barcodes with accession number, yet one needs to be very careful with typing the accession number in correctly. Errors like this can lead to more than one specimen having the same barcode.
  • Sequential –vs– Matching Accession Number
    • Matching barcodes is more work, time consuming, and typically the more expensive option. This is particularly true if barcodes are preprinted.
    • If one matches barcodes with accession numbers, a method is needed to ensure that no two specimens receive the same barcode identifier. Multiple specimens accidentally being given the same accession number is a typical problem within herbaria (e.g. stamp failed to advance).
    • Not matching produces one more identification number that can be assigned to a specimen. Institutions that decide not to match barcodes with accession numbers, often decide to do away with the old accession number in place of the new barcodes.
  • Multiple Sheet Specimens: There is disagreement on how to handle specimens that consist of multiple sheets. Some prefer that each specimen gets its own barcode while others assign the same barcode to each sheet. From the database perspective, using a single barcode identifier for all sheets of a specimen is preferred. Ideally, a single specimen should be represented by a single record within the database, whether the specimen consists of one or ten sheets. When general users query a database, they typically want the return count to correctly identify the true number of unique specimens rather than the number of sheets. When they look at the details of specimen record, they typically prefer to view images of all the sheets at once rather than having to click on separate records representing a different sheet. Finally, in the event of an annotation, the data managers should not have to enter the same annotation for multiple records within the database. Multiple records representing a single specimen not only increase data maintenance workload, but it also creates an increased possibility of ambiguity if each records states something different because each record was updated differently.

Resources:

This material is based upon work supported by the National Science Foundation grant ADBC#1115116. Any opinions, findings, conclusions, or
recommendations expressed in the material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.