315
CHAPTER 19
Sampling Vertebrate Collections for
Molecular Research:Practice and Policies
MARK D. ENGSTROM, ROBERT w: MuRPHY
AND OLIVER HADDRATH
Abstract.-With
the widespread use of protein electrophoresisin verte-
brate systematicsin the 1960sand 70s, severalmuseumsand universities
established extensive collections of frozen tissues.Use of these special
collections expanded exponentially with the development of the
polymerasechain reaction (PCR) and improvements in obtaining nucleotide sequences.Moreover, using PCR, DNA sequencescan now be
routinely obtained directly from traditional voucher collections (skins,
bone, alcohol-preservedspecimens,etc.), accentuating the issue of destructive sampling of this material. Herein we briefly review: the development of special tissue collections for consumptive use; the rationale
for and methods of collecting both voucher specimensand tissue samples amenableto genetic analyses;tissuecollection storageand management; the suitability of voucher collections as direct sourcesof DNA;
and policies on the consumptive sampling of specialcollections and destructive sampling of voucher collections.
Museumshavethe fundamental role of building and maintaining collections of biological specimens for documentation of
biodiversity, studiesof evolutionary pattern and proc~ss,and other
evolutionary research.With the advent of the polymerasechain
reaction (PCR; Mullis and Faloona, 1987) and other analytical
tools for directly assayinggeneticvariation, useof thesecollections
-
316
I
I
MANAGINGTHE MODERN HERBARIUM
has rapidly shifted to include molecular-oriented researchin addition to traditional morphological studies. In vertebrate collections, this change in orientation has been accompanied by the
development of ancillary specialcollections usedmainly for direct
genetic assays.These special collections, such as frozen tissues,
samplesof isolatedDNA, and ethanol-preservedtissues,haveposed
novel challengesfor collection developmentand management,especially as they are ultimately designedfor consumptive use.The
rationale and intent behind these special collections contrasts
sharply with those for traditional voucher collections. Accordingly, institutional policies for consumptivesampling of specialcollections will differ from the necessarilymore restrictive policies
associatedwith destructivesampling of voucher collections.
Extensiveuseof comparativemolecular methods in population
genetics and systematicsof vertebrateshas occurred only in the
past 30 years (see reviews in Dessauer and Hafner, 1984;
Honeycutt and Yates,1994). In particular, the wide adaptation of
protein electrophoresisin the 1960s and 70s (seeRichardson et
al., 1986; Murphy et al., 1996) provided the impetus for assembling frozen tissuecollections.This was the first comparativetechnique that allowed routine, direct assessmentof genetic variation
within and betweenpopulations. Focuson the microevolutionary
level, including the examination of large numbers of individuals
to estimateparameterssuch asheterozygosity,polymorphism, and
population subdivision led to the development of large, comprehensive tissue collections from wild populations (Baverstockand
Moritz, 1996). More recently, these collections have proven invaluablefor direct extraction of nucleic acids(DNA or RNA) used
in a variety of studies, including molecular systematics and
phylogenetics.For example, the number of comparative molecular papersusing thesecollections and published in the Journal of
Mammalogy increasedfrom zero in 1954 to two in 1964, six in
1974, 14 in both 1984 and 1994, and 15 in 1996. Further, there
has been a decided shift in techniquesfrom karyology, immunology, and protein electrophoresisin the early part of this period to
the current emphasison comparative studies of DNA. These collections, together with the voucher specimensfrom which they
were derived, also serveas baselinesamplesin forensic studies by
law enforcement arms of agenciessuch as the US Fish and Wild-
ENGSTROM ET AL.: VERTEBRATE COLLECfIONS AND MOLECULAR RF.5EARCH
317
life Service,responsiblefor monitoring traffic in animals and their
products (Dessauerand Goddard, 1984).
Concomitant with the acceptance,development, and routine
application of molecular techniques, special collections have
grown rapidly. In 1984, there were five North American collections with over 10,000 frozen tissue samplesof vertebrates,all in
the United States(Dessauerand Hafner, 1984). By 1994, there
were 13 North American collections of this size and large collections were establishedboth in Canada(Dessaueret aI., 1996) and in
Mexico (F.Cervantes,UniversidadNacionaIAut6noma de Mexico,
pers. comm.). Severalof thesecollections, such as the amphibian
and reptile collections in the Royal Ontario Museum (ROM),
now havemore than 30,000 individual tissuesamples(Dessaueret
al., 1996). Growth of thesecollections has been facilitated by requirements of granting agencies.For example, the National Science Foundation and the National Geographic Society now
routinely require that tissue samplesbe taken and deposited in
major museumsduring the courseof biological surveywork.
More recently,the issueof destructivesampling of existing traditional collections of vertebratesasdirect sourcesof DNA, where
specialcollections are not available,has reopenedan old problem
for those responsiblefor maintaining the archival quality of those
collections. Destructive sampling of voucher specimensfor anatomical, physiological, medical, and other avenuesof researchhas
been a long-standing issuebut hasbeen brought to the fore by the
sheer volume of requests(occasionallydemands) from the molecular researchcommunity. All researchcollections are designed
to be used, and curators and collection managersare faced with
the conundrum of balancing present research needs with the
charge of preserving the collections for future research (Cato,
1993). Although the prospect of using traditional specimensas
sourcesof DNA hasadded an unforeseenresearchdimension, further justifying their archival value to sometimesscepticaladmini~trations, this use must be carefully regulated. The widespread
availability of special collections designed for such consumptive
use in molecular systematicshas helped to ameliorate, but not
eliminate, thesepressuresfor vertebratecollections.
The aim of this paper is to: 1) briefly review techniques for
field collection, storage, and collection management of special
-
I
318
I
I
I.
MANAGING THE MODERN HERBARIUM
collections for vertebrates (mainly frozen tissue collections); 2)
discussmethodology and policies for destructive sampling of traditional, existing collections of voucher specimens;and 3) discuss
institutional policies for consumptive use of special collections
and destructive sampling of traditional voucher collections. It is
hoped that this review will provide some background into the debate on consumptive sampling as it has evolved in the vertebrate
museum community.
SPECIAL COLLECTIONS
Field Methods
Field preservation, transport, and storageof tissue collections
were recently reviewedby Dessaueret al. (1996), and will be only
briefly summarizedhere. For genetic analyses,tissuesand their included polymers need to be collected and maintained in a biochemically active form. This necessityposesa seriesof challenges
for field collecting, depending on location of the work and the
types of molecular studiesto be pursued. For vertebrates,samples
aremost often quick-frozen in either liquid nitrogen (LN2: -196°C)
or dry ice (-60 °C) in the field and later stored at ultra-low temperatures (near -80°C or colder) upon return to the laboratory.
For certain molecular approaches,such asstudiesof cellular DNA
content (Sharbelet al., 1997) or isolation of whole, intact, mitochondrial DNA, initial freezing in LN2 is required or highly preferable. Other approaches,such as DNA sequencing,pose fewer
obstacles.Macromolecules can be preservedpartially intact in a
variety of solutions (seebelow), or even in dried piecesof skin or
muscle (the latter enabling extraction of small segmentsof DNA
from traditional specimens).To minimize financial and logistic
constraintson field work and collection storageand maintenance,
it is critical to determine at the outset which molecular approaches
are to be supported by the collections. We routinely savetissue
samplesfrom most specimenscollected.
A variety of tissue types can be collected in the field, and that
variety may be affected by the intended molecular approach. For
vertebrates,the array of tissuesmost commonly preservedincludes
blood, heart muscle, skeletal muscle, kidney, and liver, but may
alsoincludebrain, spleen,stomach,eye(in fishes),and testes(in
ENGSTROM ET AL.: VERTEBRATE COU.ECflONS AND MOLECULAR RESEARCH
319
birds and mammals). Tissuesshould be dissectedfrom appropriately euthanized specimensas soon after death as possible. Skin
biopsies,blood and other tissuescan sometimesbe removedwithout killing the animal (e.g.,Amos and Hoelzel, 1991; Seutin et al.,
1991; Whitmer and Barratt, 1996) but we recommend preparation of at leastsomespecimensof eachtaxon asvouchers.For protein electrophoretic studies, the selection of tissues is critical
becausethe enzymatic expressionof individual protein loci is often restricted to specific tissues.Tissue-specificexpressionof proteins varies widely among species(Matson, 1984; Murphy and
Crabtree, 1985; Murphy and Matson, 1986) with the result that
this information is sometimesuseful in reconstructing phylogeny
(Fisher et al., 1980; Buth, 1984; Murphy and Crabtree, 1985). In
this regard, taking a greater diversity of tissuesmakes the collection more useful. For extraction of DNA, tissuetype is lessimportant, although mitochondrial-rich tissue such as liver or spleenis
best for extraction of whole mtDNA molecules. Longmire et al.
(1997) noted that for partially decomposedspecimensof mammals, DNA in brain may be lesssubject to degradation than that
in someother tissuetypes.
Extended field trips in remote regions sometimespreclude the
useof LN 2 and other cryogenic options for tissue collection. Although freezing tissues remains the most reliable and versatile
method of preserving tissue for long-term storage, cryopreservationis not required for somemolecular studies.DNA can
be preserved(although not wholly intact) in either 95% ethanol
or in 35% isopropyl alcohol. It is best to mince tissuesto allow
quick penetration by alcohol, although whole specimenscan also
be preservedin ethanol. Long-term stability of DNA molecules
may also be enhancedby adding ethylenediaminetetraaceticacid
(EDTA) which inhibits nucleases.Given that DNA is preservedin
alcohol, severalinstitutions such asthe Museum of VertebrateZoology, University of California, Berkeley,and the ROM ornithol;ogy collection, now maintain alcohol-fIXed tissue collections.
Voucher collections never preservedin formaldehyde (e.g., Zoological Institute of St. Petersburg,Russia)also have been used as
sourcesof material for molecular research,although their value in
anatomical or histological studies may be compromised. Macerated tissuesinitially preservedin lysis buffer in the field (Seutin et
320
MANAGING THE MODERN HERBARIUM
al., 1991; Longmire et al., 1997), a preliminary step in DNA isolation, can be stored at room temperature, refrigerated, or frozen
for severalyears prior to use. Finely minced tissuescan also be
saved in a saturated salt solution containipg dimethyl sulfoxide
(DMSO) and EDTA (Amos and Hoelzel, 1991; Seutin et al.,
1991; Whitmer and Barratt, 1996) and stored at room temperature for six months or more. Preservationin either lysis buffer or
DMSO salt solution produces larger yields of high molecular
weight fragments of DNA than fIXation in alcohol (Seutin et al.,
1991).
Transportation and regulations for importation of animals and
tissuesto the United Statesare discussedin Dessauerand Hafner
(1984) and Dessaueret al. (1996). In addition to obtaining any
required research,collecting and/or export permits from the country of origin, Canadian regulations require a "Permit to Import
Material of Animal or Microbial Origin Into Canada," obtained
in advancefrom Agriculture and Agri-Food Canada (present fee,
Can$21.00), to import tissuesand specimens.Within Canada,
written authorization is also required before exchangingimported
tissuesamong institutions. If return airline schedulesinclude a
stop and customsclearancein the United States,the material must
be declared there and all United Statesregulations satisfied (see
Dessauerand Hafner, 1984) before proceeding.
Curation
Unlike traditional voucher specimens,it is usually impractical
to re-number tissuevials once they have been returned from the
field. Thus the field number initially written on the tissue container is the number usedto identify the tissueand retrieve it from
storage.Regardlessof the numbering system, the sample should
always be cross-referencedto the museum catalogue number of
the voucher specimenfrom which it was taken.
The problem of field cataloguinghas been handled in a variety
of ways. Some institutions use a separatehard-bound field catalogue for tissues,wherein every tissuesampleis assigneda unique
sequential number, in addition to the field collector's number.
These tissue cataloguepagesare subsequentlyannotated with the
permanent museum cataloguenumber of the voucher, thus crossreferencingthe voucherwith parts derivedfrom it. A secondsystem
ENGSTROM
ETAL.:V'=-=ONS
.
ANOMOlECUlARRESEARCH32 [
usesa single field number for both the tissue sample(s)and the
voucher, and this number is recordeddirectly on the vial. On return to the lab, these samplesare assignedanother frozen tissue
number and cross-referencedto the voucher collection. In both of
thesesystems,three numbers are associatedwith each tissuesample: the collector'sfield number, the tissuecataloguenumber, and
the catalogue number of the voucher specimen (Baker and
Hafner, 1984).
Alternatively, the tissuecataloguecan be eliminated altogether.
In the ROM mammal and herpetology collections, collectors use
pre-printed field cataloguepagesand rolls of unique, sequential
field numbers. These field numbers are usedfor both the voucher
specimensand tissuesamples.On return to the museum, the collection data are immediately entered into a temporary file and
voucher specimensare assigneda permanent cataloguenumber.
Thus, like the voucher specimen,only two numbers are associated
with the tissue samples:the field number (used for retrieval purposesin the frozen tissue collection) and the permanent museum
cataloguenumber of the voucher. Once field identifications are
verified, the temporary databaseis updated and specimen information is sent to two permanent databases:the databasefor the
main specimencollection and the frozen tissue database,each of
which contains some unique fields (Woodward and Hylwka,
1993). The advantageof this systemis that final determinations of
the voucher and its collection cataloguenumber are automatically
updated and cross-referencedto the tissuesample.
Frozen samples are permanently stored in cardboard boxes
with dividers, which in turn are arranged in a system of metal
racks in an ultra-cold freezer.Samplespreservedin ethanol, lysis
buffer or DMSO-salt solutions can be kept in the dark at room
temperature, but DNA is more stable if tissuesare kept cool. In
the ROM ornithology collection, ethanol-preservedsamplesare
stored at -20°C for short intervals and between-70°C and -80o~
for long term storage.
Two systemsfor sorting samplesfor retrieval are often used,either numeric, where samplesare placed in order of field or tissue
cataloguenumbers (the systemusedin the ROM mammal collection), or taxonomic, where samplesare arrangedin the sameorder
as the voucher collection (the
system
by the ROM
'~.
' ~---;- used
---
~
322
MANAGINGTHE MODERN HERBARIUM
herpetology collection). The numeric system is more spaceefficient, as samplesare simply added to the end of the numeric sequence. However, retrieval of specific taxa is more time
consuming. Regardlessof the system used, random gaps can occur in the boxes as samplesare consumed.
Another challenge to the management of tissue collections is
that samplesare often completely consumed.To maintain a running inventory in the ROM mammal collection, we record the
number of tissue vials originally present and update the database
recordsany time a sample is granted to an outside investigator or
used in-house. When the vial number equalszero, the samplehas
been totally consumed and it is effectively deaccessioned.However, records of how the sample was used and by whom are still
maintained. One person (preferablya tissuecollection manager)is
assignedto maintain the tissuedatabase,store and retrievesamples,
and processinquiries and loan requests.Tissue grants are made
only on the approval of the curator responsiblefor that collection.
DESTRUCTIVE SAMPLING
OF VOUCHER COLLECTIONS
Tissuesdo not necessarilyneed to be frozen or preservedin solution if collections are solely for studies of DNA. One of the
great advantagesofPCR is its ability to produce large quantities of
DNA from very small numbers of target nucleic acids.Short fragments can be recoveredfrom skin, bone, feathers,teeth, and other
dried body parts that are hundreds and, in some cases,thousands
of (or even more) years old (Paabo et al., 1988; Ellegren, 1991;
Hagelbergand Clegg, 1991).
Traditional voucher collections are now recognizedby molecular systematistsnot only as the primary repository of information
on morphological charactersand relationships but also as storehousesof short strandsof DNA. Although thesespecimenscan be
used, they present many problems for the recovery of DNA not
experiencedwith frozen tissues.For example, the autolytic processesthat follow cell death degrade nucleic acids. The resulting
low concentration of target DNA is very susceptibleto contamination from extraneousDNAs in the laboratory or from other
sources.For example,Haddrath recentlyamplified a mitochondrial
~
ENGSTROM Kf AL-' VEmURAlE
Co=ONS
AND MolE1JLAR
RESEARCH
323
cytochrome b sequencefrom bird lice, only to find that the sequence recoveredwas actually that of the host (a kiwi) on which
the lice had most recently fed. Inhibitors of the polymerasechain
reaction that are found in traditionally preservedskins and tissues
also tend to co-purify with DNA, often resulting in no or low
yields (Hummel and Herrmann, 1994). Given these problems,
frozen or otherwise preservedtissuesfrom the special collections
are much preferredassourcesof DNA for PCR. However, for rare,
endangered or, especially,extinct species,the original voucher
specimensmay be the best or only practical alternative.These factors should be considered in both the investigator'sinitial selection of sourcematerials,and in the curatorial decision whether or
not to grant a requestto destructively samplespecimens.
Although there are a variety of protocols for removing samples
from museum specimens,severalaspectsare universal.When handling a specimen, latex gloves must be worn and any cutting
should be done with sterile utensils. The smallest and least conspicuous tissue sample possible should be removed and then
placed in a sterile plastic tube to lower the chanceof contamination. It is best to removetwo samplesfrom separateplaceson the
specimen. Both samplescan then be processedindependently to
confirm the authenticity of any resulting DNA sequencesand to
minimize the possibility that any aliquots of DNA returned to the
granting institution are contaminated. Double sampling and amplification is now required by somejournal editors to confirm the
repeatability of experiments involving ancient DNA. In the case
of soft tissues,samplesshould be removed from subsurfaceareas
when possibleto limit the effectsof contamination and action of
preservatives(e.g., formaldehyde). In dried or mummified tissues,
sampling near the extremities of the specimen enhances the
chances of recovering DNA because these regions dehydrate
quickly, limiting degrading autolytic processes.
The amount of material required from voucher specimensfor
PCR varieswith the type of tissuebeing sampled.For soft tissues,
such as dried skins, small pieceslessthan 0.1 g (lor 2 mm2) will
often suffice (Thomas et al., 1990; Scott Woodward, Brigham
Young University, pers. comm.), whereasfor hard tissues,such as
bone, between 0.5 g and 1.0 g usually is needed (Hagelberg and
Clegg, 1991). One advantagein using bone is that longer &agments
,"
324
MANAGINGTHE MODERN HERBARIUM
of DNA canbe amplifiedfrom it, in somecases
up to 1,000base
pairs (bp; Hagelberg et al., 1991). DNA in soft tissuesis often
more degraded and maximum fragment size is usually between
150 bp to 350 bp. In contrast, length of fragments in frozen tissuesis limited only by the maximum size that can be amplified
using PCR. State of preservation,rather than absolute age of the
specimen,is the best indicator of the likelihood of successfullyrecoveringa DNA sample.
We have been able to recoverand amplify DNA fragments between 150-350 bp from skin samplesin the voucher collection
that are up to 100 yearsold, greaterthan 500 bp from bone up to
3,000 years old, and fragments greater than 350 bp from bone
severalthousand yearsold. For example,Haddrath useda 1 g sample of bone to isolate 400 bp fragments of DNA from a 10,000
year old New Zealand moa, and, using overlapping primers, reconstructed a 1,000 bp sequencefrom this extinct bird.
Methods of preservationwhich alter the chemical structure of
vouchers may limit their utility in the recoveryof DNA. For example, for large mammals it is very difficult to recoverDNA from
chemically-altered,tanned skins. In caseswhere large hides are to
be tanned, we clip a piece of skin beforehand and store it along
with the voucher skeleton. Likewise, dried museum skins are prepared without any preserving agents or insecticides, such as arsenic or mercuric chloride, and bones are rinsed only in water
during their final cleaning. Although it is sometimespossible to
recoverDNA from specimensinitially preservedin formalin, this
is usually problematic. We havehad little successamplifying DNA
from well-fixed material, even when specimenshave been subsequently transferredto ethanol for long-term storage.
Policies for Consumptive and Destructive Sampling
An important collections management issue is to determine
under which circumstancesit is appropriate to permit consumptive sampling from specialcollections or destructive sampling of
traditional vouchers. Some balance must be struck between the
need for accessto collections by individual researchersand the
long-term responsibility of the holding institution to maintain the
future value and integrity of its collections. At the ROM, we are
much more strict about destructive sampling of specimensin the~
ENGSTROM ET AL.: VERTEBRATE CoLLEGnONS AND MOLECULAR RESEARCH
325
main voucher collection than in loaning samplesfrom specialcollections for consumptive use. For example,over a recent five year
period, the ROM mammalogy collection has granted only three
requestsfor destructive sampling of vouchers,whereasloans from
the specialfrozen tissuecollection comprised over 20% of all filled
requests.
The issue of both consumptive and destructive sampling
should be addressedin the form of a collections policy statement,
such as that of the Museum of VertebrateZoology (MVZ), University of California, Berkeley(currently postedon the Internet at:
http://www.mip.berkeley.edu/mvz/fcpolicy.html). A number of
authors have discussedpoints that should be formalized in a written policy statement (e.g., Paabo et al., 1992; Hafner, 1994;
Whitfield and Cameron, 1994; American Society of Mammalogists, Systematic Collections Committee, F. Villablanca, in
litt.), and Cato (1993) provided an example.
The following is a brief summary and amplification of points
that we feel merit discussion in any formulation of a policy for
both consumptive and destructive sampling.
1. Any requestmust be made in writing and should include a researchproposal. Student proposalsmust be signed by the supervisor who must accept accountability for the material
loaned.
2. The researchershould have demonstrated competency in the
proposed methods, and positive results using a common, related taxon might be required before a requestis granted.
3. Portions of samplesthat are not usedshould be returned to the
collection at the conclusion of the project. Transferof materials
from the borrower to a third party should never be allowed,
unless expresslyauthorized in writing by the lending institution. Such third party loans can lead to the loss of the connection between the voucher specimen and the tissue or
subsequentDNA extraction. It is very important to avoid ~he
loss of this link becauseit can result in the original lending institution losing both control of, and the rightsto, information derived from its specimens.
4. Some institutions require that aliquots of DNA extractions be
returned, to be stored by the lender (seediscussion between
Whitfield and Cameron, 1994; and Hafner, 1994). Return of
326
MANAGING THE MODERN HERBARIUM
DNA extractions may prevent unnecessaryresampling of the
samespecimensor tissuesamplesin the future sincethe extraction could be used instead. Our own experiencewith returned
extractionsis mixed, however,and we requestthis only for rare
specimens,and then on a case-by-casebasis.We have found
that the long-term stability of DNA extractscan be affectedby
either the choice of reagentsusedto isolate the DNA or by the
storagemethod. For example,while Chelex is very effective at
extracting DNA from a wide range of cell types, the resultant
high pH environment has been observed to affect the longterm stability of the DNA. Repeated freezing and thawing
similarly compromiseslong-term stability. It should be noted
that this can occur when specimensare stored in frost-free
freezersdue to the oscilation in temperaturesof these units.
There is also the possibility of the extractions being contaminated with extraneousDNA, a problem mostly found with extractions of ancient DNA. Unfortunately, these are often
precisely the rare specimenswhich one would prefer not to
resample.
5. For vertebrates,central depositories are being identified that
are willing to accept and store returned extractions when the
original lender doesnot havethe facilities or personnelto do so
(Hafner, 1994). It is unclear, however, how these central repositorieswill maintain the link betweenthe original specimen
and the extraction of PCR product, and under what circumstancestheseproducts might be loaned, given that the original
lender should retain the rights to material derived from their
specimens.Any repository systemshould be regulatedby a disciplinary body (such as the Society for the Preservation of
Natural History Collections or the Association of Systematics
Collections) to insure that rights of the original collections are
not inadvertently violated.
6. Collections differ in their policies regardingthe scopeof materials to be loaned (i.e., whether the lending institution is willing to serve as the major source of tissuesand/or taxa for an
outside study, or whether it is willing only to provide supplementary material to augment existing data). Some institutions,
like the MVZ, will only provide supplementary material
ENGSTROM ET AL.: VERTEBRATE COLLECI10NS AND MOLECULAR RESEARCH
327
whereasothers, like some collections in the ROM, may provide the lion's shareof the material required.
7. For destructivesampling of study specimens,someinstitutions
allow known and qualified researchersto remove samplesof
skin, bone, feathers,etc., from specimenseither under supervision in the collection at the lending institution or at the borrower's laboratory. Other institutions are more restrictive and
insist on removing the samplesthemselvesor in some cases,
even performing the DNA extractions (e.g., zoological collections of the Natural History Museum, London). By choosing
the specimen(s) and removing the tissue clips in house, the
lending institution retains the most control over its material
and insuresthat damageto specimensis minimized. This is the
coursemost often taken at the ROM.
8. As noted by most authors of sequencing studies, DNA sequences should be submitted by the researcher to either
GenBank or an equivalent database,and the accessionnumbers
of these sequencesshould be returned to the lending institution to become a permanent part of the specimen record.
Whether or not other resultsare required to be returned varies
among institutions and types of studies. For example, in protein studies the individual genotypic frequency data might be
depositedwith the lending institution so that thesedata can be
correlatedwith individual vouchers.This is particularly important if the original genotypic data are not published. However,
accumulating, filing, and cross-referencingraw data hasits limits. As pointed out by Hafner (1994), museumscurate specimens, not reamsof unpublished methods and resultsbasedon
them. In the ROM we require, at minimum, the return of
GenBank numbers and reprints of published papers, and ask
for other data only in special circumstances.In all cases,any
publications resulting from a study of borrowed material
should both acknowledgethe lending institution and list the
individualspecimens
that wereused.
-:
9. Unlike loans of study specimens,reimbursement for shipping
costs is usually required, particularly for frozen tissueswhich
are sent by overnight courier. For large collections with relatively liberal grant policies, the cost of shipping soon becomes
328
MANAGINGTHE MODERN HERBARIUM
insupportable unless the trade of material is evenly reciprocated.We occasionallywaive this rule for starving graduatestudents. Some institutions also charge a retrieval fee, ranging
from $10 to $50 per tissue sample, to recovercostsassociated
with assemblingand processingloans. Other institutions require that some tissuesamplesbe receivedin exchangefor any
tissuesgranted.
CONCLUSION
The distinction between consumptivesampling of special collections assembledfor laboratory analysesand destructivesampling
of voucher collections is important and should be consideredby
both molecular systematistsand curators. Voucher collections are
preserved,processed,and stored to ensuretheir long-term value in
the documentation and study of biodiversity and related specimen-oriented systematic research,and their utility in molecular
systematicsis a secondaryconsideration. Frozen and other special
tissuecollections designedfor consumptive use alleviate the need
to extensivelysample traditional collections and remain the primary source of materials for molecular systematics.Institutional
policies for destructive sampling of vouchersare necessarilymore
stringent than those regarding sampling of specialcollections designedfor consumption. Although the advent of PCR hasresulted
in important new usesfor both traditional and specialcollections,
there is a continued requirement to balancethe need for immediate usewith the preservationof their archival value.
ACKNOWLEDGMENTS
This is contribution number 46 from the Centre for Biodiversity and
Conservation Biology, Royal Ontario Museum.
LITERATURE CITED
AMos, W ANDA. R. HoEIZEL. 1991. Long term preservation of whale skin for
DNA analysis.Pp. 99-103 in A. R. Hoelzel, editor. Geneticecologyof whales
and dolphins.Reportof the InternationalWhaling Commission,SpecialIssue13.
BAKER,R. J. ANDM. S. HAFNER.1984. Curation of collections of frozen tissues.
ENGSTROM ET AL.: VERTEBRATE CoU.ECflONS
AND MOLECULAR RESEARCH
329
Curatorial problems unique to frozen tissue collections. Pp. 35-40 in H. C.
Dessauer and M. S. Hafner, editors. Collectiom of Frozen Tissues: Value,
Management, Field and Laboratory Procedures, and Directory of Existing Collectiom. Association of Systematics Collections, Lawrence, Kansas.
BAVERSTOCK,
P. R. AND C. MORITZ. 1996. Project design. Pp. 17-27 in D. M.
Hillis, C. Moritz,
and B. K. Mable, editors. Molecular Systematics. Second
edition. Sinauer Associates, Inc., Sunderland, Mississippi.
Burn,
D. G. 1984. Applicability
of electrophoretic
data in systematic studies.
Annual Review of Ecology and Systematics 15:501-522.
CAm, P. S. 1993. InstitUtion-wide policy for sampling. Collection Forum 9:27-39.
DESSAUER,
H. C. AND K. W GODDARD. 1984. Value of frozen tissue collections
for forensic studies. Pp. 12-13 in H. C. Dessauer and M. S. Hafner, editors.
Collectiom of Froun Tissues:Value, Management, Field and Laboratory Procedures, and Directory of Existing Collectiom. Association of Systematics Collections, Lawrence, Kansas.
DESSAUER,H. C. AND M. S. HAFNER. 1984. Collectiom of Froun Tissues:Value,
Management, Field and Laboratory Procedures, and Directory of Existing Collectiom. Association of Systematics Collections, Lawrence, Kansas.
DESSAUER,
H. C., C. J. COLE, AND M. S. HAFNER. 1996. Collections and storage
of tissues. Pp. 29-47 in D. M. Hillis, C. Moritz,
and B. K. Mable, editors.
Molecular Systematics. Second edition. Sinauer Associates, Sunderland, MA.
EU.EGREN,H. 1991. DNA typing of museum birds. Nature 354:113.
FISHER,S. E., J. B. SHAKLEE,S. D. FARRIs,AND G. S. WHITT. 1980. Evolution of
five multilocus isozyme systems in the chordates. Genetica 52/53:73-85.
HAFNER,M. S. 1994. Reply: Molecular extracts from museum specimens canand should -
be saved. Molecular Phylogenetics and Evolution 3:270-271.
HAGELBERG,E. AND J. B. CLEGG. 1991. Isolation and characterization of DNA
from archaeological bone. Proceedings Royal Society of London, B. 244:45-50.
HAGELBERG,E., L. S. BELL, T. ALLEN, A. BOYDE, S. J. JONES,AND J. B. CLEGG.
1991. Ancient
bone DNA:
techniques
and applications.
Philosophical
Transactions Royal Society London, B. 333:399-407.
HONEYCUTT,R. L. AND T. L. YATES.1994. Molecular systematics. Pp. 288-309
in E. C. Birney and J. R. Choate, editors. Seventy-five Yearsof Mamma~gy
(1919-1994).
Special
Publication
no.
11,
American
Society
'of
Mammalogists.
HUMMEL, S. AND B. HERRMANN. 1994. General aspects of sample preparation.
Pp. 59-68
in B. Herrmann
and S. Hummel,
editors. Ancient
DNA.
Springer-Verlag, New York.
loNGMIRE, J. L., M. MALTBIE,
ANDR. J. BAKER.1997. The useof "lysis buffer" in
,
330
MANAGING THE MODERN HERBARIUM
DNA
isolation.
Occasional Papers, The Museum, Texas Tech University
163:1-4.
MATSON, R. H. 1984. Applications
of electrophoretic data in avian systematics.
Auk 101:717-729.
MULLIS, K B. AND F. FALOONA. 1987. Specific synthesis of DNA
in vitro via a
polymerase catalyzed chain reaction. Methods in Enzymology 155:335-350.
MURPHY, R. w: AND C. B. CRABTREE.1985. Evolutionary aspects of isozyme patterns, number of loci, and tissue-specific gene expression in the prairie rattlesnake, Crotalus viridis viridis. Herpetologica 41 :451-470.
MURPHY, R. w: AND R. H. MATSON. 1986. Gene expression in the tuatara,
Sphenodon punctatus. New Zealand Journal of Zoology 13:573-581.
MURPHY, R.W:, J. w: SITES,JR., D. G. BUTH, AND C. H. HAUFLER. 1996. Proteins I: Isozyme electrophoresis. Pp. 51-120 in D. M. Hillis, C. Morit, and
B. K Mable, editors. Molecular Systematics.Second edition. Sinauer Associates,
Inc., Sunderland, Mississippi.
PAABO,S., J. A. GIFFORD, AND A. C. WILSON. 1988. Mitochondrial
DNA
se-
quences from a 7000 year old brain. Nucleic Acids Research 16:9775-9787.
PAABO,S., R. WAYNE, AND R. THOMAS. 1992. On the use of museum collections
for molecular genetic studies. Ancient DNA Newsletter 1:4-5.
RICHARDSON,B. J., P. R. BAVERSTOCK,
AND M. ADAMS. 1986. Allozyme Electrophoresis. Academic Press, Inc., Orlando, Florida.
SElTTIN,G., B. N. WHITE, AND P.T. BOAG. 1991. Preservation of avian blood and
tissue samples for DNA analyses.Canadian Journal of Zoology 69:82-90.
SHARBEL,T. F., L. A. LoWCOCK, AND R. w: MURPHY. 1997. Flow cytometry:
a
precise and rapid method for genomic analyses of amphibian populations.
In D. M. Green, editor. Amphibians in Decline: Canadian Studies of a Global
Problem. Herpetological
Conservation
1:78-86.
THOMAS, w: K., S. PAABO,F. x. VILLABLANCA,AND A. C. WILSON. 1990. Spatial
and temporal continuity
of kangaroo rat populations shown by sequencing
mitochondrial DNA from museum specimens. Journal of Molecular Evolution
31:101-112.
WHITFIELD, J. B. AND S. A. CAMERON.1994. Museum policies concerning specimen loans for molecular systematic research. Molecular Phylogenetics and
Evolution
3:268-270.
WHITMER, J. w: AND E. BARRATT.1996. A non-lethal method of tissue sampling
for genetic studies of chiropterans. Bat Research News 37:1-3.
WOODWARD,S. M. AND w: E. HYLWKA. 1993. A database for frozen tissues and
karyotype slides. Collection
Forum 9:76-83.
.
.