On this page...


Ichthyology is the study of fishes. There are estimated to be over 32,500 species of fishes worldwide and nearly 5,000 in Australia.

Fishes are cold-blooded (mostly) animals that have:

  • gills for breathing
  • a backbone or a notochord (a cartilaginous rod)
  • fins (most species but not all)
  • scales (most species but not all)

The Australian Museum Ichthyology collection consists of both adult and larval fish specimens. It is most likely the largest in the Southern Hemisphere and is the fourth largest collection of Primary Type species in the world. The overall size of the collection is estimated to be approximately 194,086 lots made up of 650,000 adult and 1,000,000 larvae specimens. The collection, of which 100 percent has been digitised, also consists of 2,500 Type specimens dating back to the 1800s, upon which the original scientific description of the relevant species was based.

The collection is housed across two Australian Museum storage sites, our main Sydney CBD location and the Castle Hill Discovery Centre. The majority of the collection has been fixed in 10% formaldehyde then transferred to 70% ethyl alcohol for long term storage. In recent years many frozen and alcohol-fixed tissues have been added to the collection for use in genetic studies.

Approximately 5000 species of marine and freshwater fishes that occur in Australian waters. Collection strengths reflect the research interests of retired Curators and Senior Fellows Dr John Paxton (Myctophidae and Deepsea taxa), Dr Doug Hoese (Gobiods) and Dr Leis (Indo-Pacific reef fish larvae).

The geographic coverage of the collection is primarily that of the Indo-Pacific region, however we have fish specimens from all continents and oceans. Particular strengths include, New South Wales, Great Barrier Reef, temperate rocky reefs, mesopelagic waters from the upper slope to 1000m depth and Indo-Pacific Islands.

Gilbert Percy Whitley (1903-1975) was a prolific author and collection builder. From 1925 to his retirement in 1964, Whitley registered some 37,000 specimens, at an annual average of almost 950 specimens, virtually doubling the size of the collection. Considering he personally identified, registered, and tagged all the specimens and concurrently wrote 500+ papers and described 451 species of fishes, the effort is indeed impressive.


Frequently asked questions

Why keep a fish collection?

A researcher wanting to study the anatomy, taxonomy, reproductive biology or feeding habits of a particular species, can go to the collection. This saves the time, expense, and conservation issues associated with capturing fresh specimens. For many species, capturing fresh specimens is often difficult or impossible, such as those which migrate, are found in the deep sea or are endangered.

The collection serves much as a library, with specimens being loaned and returned. Unlike a library however, the collection becomes more valuable after specimens have been studied and returned. The Australian Museum Fish Collection collection is heavily utilised. In the last ten years (1990-1999), more than 45,000 fishes in over 1,200 separate transactions have been sent out of the Fish Section. Most of these were sent to researchers overseas because many of the taxonomic problems of Australian species are being solved by the international scientific community. Many other specimens are used on the premises by staff and visiting researchers.

The collection provides an historic record for each species. It can be used to document distributions, both present and past. The collection is also a repository of specimens that are used in many scientific studies. If questions arise as to what species were actually dealt with in any study, specimens placed in the collection by the worker can be examined to answer those questions.

A small proportion of the specimens are types. These scientifically priceless specimens have been used to describe a new species and bear the name of the new species. The collection of fish types at the Australian Museum is on the list of the ten largest in the world (W. Eschmeyer pers. comm.) and is of immense scientific value. The collection houses over 10,000* type specimens (nearly 2000* type species).

*Figures current June 1999.

Further reading

  1. Paxton, J.R. & M. McGrouther. 1991. Why so many specimens? Muse (Australian Museum News & Events) Aug -Sept. 1991:4, 11, 2 figs.

Why keep so many specimens?

The Fish Collection attempts to cover the range of variation in structure, form, and distribution, for each fish species. The Sunfish images on this page show the major changes in structure that some species go through as they develop. In order to document the changes it is necessary to keep a developmental series from tiny larvae to adults of each fish species.

If one considers the normal variation of the human species - males and females, young and old, redheads and blonds, blue eyes and brown, short and tall, fat and thin - coupled with cultural and geographic variation, hundreds of humans from around the world would be required to give adequate representation of the species.

While few species of fishes display the variation found in the human species, we need a reasonable number, say 20-30 each from different areas throughout the distribution of the species, to guarantee we have the extremes of variation. In addition we keep examples of the largest and smallest (including eggs and larvae), as well as distributional extremes. With an estimated 5000 species of fishes in Australian waters (and around 25,000 species worldwide) it is easy to see how and why fish collections become so large.

The Australian Museum collections are part of a larger whole. The major fish collections of all seven Australian state (and Territory) natural history museums, and the CSIRO Fisheries collection in Hobart, use a common family numbering system with compatible data entry and retrieval. The total number of fishes in all Australian collections exceeds 1.8 million, and is surely part of our national heritage. With some U.S. fish collections numbering 10 million specimens, our total national collection is not large relative to our very rich fish fauna.

Further reading

  1. Liem, K.F. in Paxton, J.R. & W.N. Eschmeyer (Eds). 1994. Encyclopedia of Fishes. Sydney: New South Wales University Press; San Diego: Academic Press [1995]. Pp. 240.
  2. Paxton, J.R. & M. McGrouther. 1991. Why so many specimens? Muse (Australian Museum News & Events) Aug -Sept. 1991:4, 11, 2 figs.

Why is it necessary to collect a specimen?

For some well known common species of fishes such as the Comb Wrasse, a photographic record, and even a sight record can be very useful. Problems arise however if the validity of the record is ever questioned. A sight record is worth very little because it cannot be validated. A photograph is useful but often doesn't display important characteristics such as the number of gill rakers or any internal character that may be required to confirm the identification of the specimen.

Another major reason for collecting fish specimens rather than just recording observations, is that many species are never seen. They live in a burrow or deep in the rocks or coral. A good example is the Palespotted Podge, Pseudogramma polyacantha (see imgage). This is one of the most common fishes collected in Indo-Pacific coral reef fish surveys. It lives deep inside the coral but is never observed alive by divers. Unless intensive collecting techniques are used only a fraction of the true biodiversity is recorded (Ackerman & Bellwood, 2000).

Dr J. Williams of the Smithsonian Institution stated that:

  • on the Navassa Island biodiversity survey (see further reading below), we used visual censusing and underwater video footage, as well as dipnetting, line fishing and rotenone (plant root powder which affects fish) stations.
  • Of the five blennioid families (blennies and related fishes) we collected 19 species(and possibly a few more that couldn't be identified in the field), using rotenone. In terms of numbers these were among the most common fishes.
  • We searched the same and similar areas visually and could only find two of the 19 species during the visual censuses or on the video tapes. Without collecting the specimens using rotenone, we would have missed 17 of the 19 blennioid species at Navassa, including an undescribed species of Acanthemblemaria (family Chaenopsidae, the pikeblennies, tubeblennies and flagblennies).
  • A strict visual census would have missed not only a major chunk of the biodiversity, but much of the biomass as well. These small cryptic fish are the food of the commercially important species.

In his visual census of Vanuatu fishes, D. Williams (1990) recorded 469 species of fishes. These workers recommended that a more complete assessment of the Vanuatu fauna was required using intensive collecting techniques. "Such a study by experienced collectors such as those at the Australian or West Australian Museum is recommended.". The Australian Museum in collaboration with other museums and Vanuatu Fisheries collected fishes in Vanuatu on two trips in 1996 and 1997. These trips verified the comments of Williams. On these trips, fishes from 40 families not recorded in William's visual census were collected. These include many species recorded from Vanuatu for the first time and indeed a number were new species. These data have been supplied to Vanuatu Fisheries who now have a more complete picture of their fish fauna. This information can be used to make conservation and fisheries management decisions.

Further reading

  1. Ackerman, J.L. & D.R. Bellwood. 2000. Reef fish assemblages: a re-evaluation using enclosed rotenone stations. Marine Ecology Progress Series 206: 227-237.
  2. Collette, B.B., Williams, J.T., Thacker, C.E. & M.L. Smith. 2003. Shore fishes of Navassa Island, West Indies: a case study on the need for rotenone sampling in reef fish biodiversity studies. Aqua. 6(3):89-131.
  3. Paxton, J.R. & M. McGrouther. 1991. Why so many specimens? Muse (Australian Museum News & Events) Aug -Sept. 1991:4, 11, 2 figs.
  4. Smith-Vaniz, W.F. Jelks, H.L. & L.A. Rocha. 2006. Relevance of cryptic fishes in biodiversity assessments: a case study at Buck Island Reef National Monument, St. Croix. Bulletin of Marine Science. 79(1): 17-48.
  5. Williams, D.McB. 1990. Shallow-water Reef Fishes. in Done, T.J. & Navin, K.F. (eds) Vanuatu Marine Resources: Australian Institute of Marine Science, Townsville. Pp. 66-76.

What is a Lot?

The number of specimens count is an obvious way to assess the size of the collection. The number of lots is another measure of collection size that is often used in natural history collections.

A lot is any number of specimens of a single species collected at the same place and time, by the same collectors, using the same method. For example, the Sea Mullet in the top image to the right, which was caught at Hastings Point in December 2002, is a single lot (AMS I.41874-009).

The school of Striped Catfish (if collected) would also be one lot. The Sea Mullet and Striped Catfish together are two lots (containing numerous specimens).

A fictional collection of one million specimens in only 10 lots, would obviously contain an enormous number of specimens of only a few species. Conversely a collection such as the Australian Museum Ichthyology Collection, which contains well over a million specimens in over 210,000 lots, has a much higher diversity in number of species, collecting localities and dates of collection.


What is the future of collections?

There are immense areas of the environment that are unsampled. Almost half of the nine million square kilometres of the Australian Fishing Zone has depths over 2000m, but surprisingly few fishes have been collected below 1500m.

Collection size will increase as voucher specimens from other studies are added and additional frozen tissue and alcohol fixed tissues are registered. In recent years these latter specimens have been in great demand for use in genetic studies which are regularly contributing to our knowledge and understanding of fishes.

Advances in computer hardware and software over the last twenty years have seen spectacular advances in the methods of registering specimens, processing loans and utilising the collection and its associated data. We now can easily answer questions about which species occur at a particular locality or habitat. As the collections continue to be used to deal with the problems of biodiversity, endangered species and human impact on the environment, they will continue to grow, in both size and value.

Further reading

  1. Paxton, J.R. & M. McGrouther. 1991. Why so many specimens? Muse (Australian Museum News & Events) Aug -Sept. 1991:4, 11, 2 figs.


Natural Sciences collections contact form