Natural history collections are critically valuable for many areas of research requiring reliable biological occurrence data. There are no other data sources quite so reliable as museum data since specimens are stored indefinitely and can be examined to verify specimen identification. They are becoming more and more relevant as the need for historic historical data to research climate change and other human impacts on ecosystems grows (Pyke and Ehrlich, 2009). This need is growing while at the same time funding for natural history collections is decreasing, resulting in various problems including lack of storage space, overcrowded storage units, lack of trained staff, and damaged facilities (Denis, 2011) leading to degradation and loss of valuable specimens and their associated data.

In our own experience (as part of the Fishes of Texas Project) working with and visiting fish collections across the United States, we are frustrated to see this loss. Often, we act as archaeologists, attempting to piece together museum records from fragmented labels and specimens that lack critical diagnostic anatomy due to poor preservation. Attempting to salvage information from these specimens and their labels is critical to our mission. Problems we've noticed include: lack of regular maintenance (particularly maintaining jar fluid); lids prone to cracking which results in alcohol loss; poor housing with unstable temperatures and exposure to UV light; fire risk; lack of digital databases and in rare cases no database; lack of ancillary documentation such as field notes and photographs; lack of space making it difficult to find specimens; and in some cases, stacked jars at risk of falling. In our work visiting nearly a dozen fish collections, we find these problems widespread and even institutions that are better-funded have specimens that are uncatalogeduncatalogued, unsorted, or otherwise in need of preservation improvements (Figure 1).

Figure 1. Examples of specimen and data

...

degradation.

Most museums have a specimen backlog, often out of sight and not included with the cataloged collection, representing collection events from the past that have not been curated. These collections are often unsorted, hidden, and housed in worse conditions than the cataloged collection. These sometimes-forgotten backlogs represent occurrences that are often unknown to science and recovering them is the only way to add "new" specimen data to the historical record.

The TNHC has reduced it's its backlog considerably over recent years as part of the Fishes of Texas project and in the process recovered and cataloged into the TNHC's database numerous valuable specimens. At the same time, the TNHC has acquired several museum collections as donations including collections from: Lamar University, Texas A&M University in Kingsville, University of Texas' Marine Science Institute, and Texas Tech University. The TNHC is currently seeking specimens from more donors and funding to continue this work. The examples below document some of the work we have done to recover data from those collections that would otherwise likely be lost.

...

Examples

The Texas Natural History Collections (TNHC) recently received the fish collection from a small Texas University. This collection was in a condition typical of many of the smaller museums which we have visited as described above. It was donated to the TNHC and we were able to salvage most of it and include it in our collection and databases, making these data useful to researchers around the world for the first time. Specimens were often desiccated and/or moldy as a result of using various jars and lid types (including Bakelite lids) with cracks (Figure 2). Note the use of Parafilm to maintain seals as a temporary solution to drying -- this often makes jars difficult to open and decreases the usefulness of the collection. Most jars were in decent condition and only needed a new lid and fresh alcohol -- some lots were in need in of much more work.

...

The following series of photos (Figure 3) document the condition of one of the jars included in the Lamar University donation. This jar is almost dry and mold has begun to grow in the jar making the jar label almost entirely indecipherable and disintegrated. The specimen has been degraded to the point that it cannot be handled and must be discarded. Disposing of a specimen is unusual and we typically keep specimens even if in very poor condition. We were able to examine enough of the specimen, while it floated in the jar to have a reasonable level of confidence in the labeled ID of Ammocrypta vivax. The specimen tag remains intact and other critical data can be retrieved:

...

Thus this jar, which is an extremely poorly preserved example, can be salvaged as a reasonably reliable occurrence record. Although not included as part of the TNHC database, since the specimen voucher no longer exists, this record is now databased as a photo vouchered record (with moderate confidence in the ID) in our Fishes of Texas Project databases and the occurrence can be mapped for use in various research projects.

Figure 3. Example of a data salvage project.

Example 2

This jar (Figure 4) is damaged by mold and desiccated making the jar label extremely fragile and difficult to read. It must be handled with extreme care using forceps and probes to remove it from the jar intact. The specimen has been degraded and dried, but salvageable as a verifiable occurrence record. Specimen can be verified as Ammocrypta vivax. The label can be read with the aid of a dissecting scope. Data are salvaged as:

...

The contents of this jar (Figure 5) are damaged by black mold making the label difficult to read. The specimens are salvageable as a verifiable occurrence record. Specimen can be verified as Micropterus salmoides. The label is faded in some areas but can be read with the aid of a dissecting scope. Data are salvaged as:

...

The contents of this jar (Figure 6) are in decent condition except for the jar label which has entirely disintegrated forming a white "broth". A back-up label preserved the basic data. Specimen is determined to be Lepomis microlophus not L. humilis as written on label. Data are salvaged as:

...

The contents of this jar (Figure 7) include a large white fungal mat which was easily removed. Specimen is in decent condition and can be verified as Cyprinus carpio. The specimen label is easily read. Data are salvaged as:

Cyprinus carpio; Culvert 50 yd E. of Walden Rd on I10; Jeff Co. TEX; 7-17-72; Leg. Ichth. Class; Det. Simpson

...