According to January 2013 figures on their websites, the American Museum of Natural History (AMNH) has more than 32 million specimens, the Smithsonian Institution in Washington over 126 million  and the Natural History Museum over 70 million. Do we really know how many specimens we have here at the Museum? Are these figures meaningful and does it matter if we haven't estimated correctly?

When I consider the collections in my care I often have a chuckle about these figures and wonder if I could ever even get close to estimating the actual number of specimens in my collection. Take the jars and bottles below for example; there are literally hundreds of thousands of microfossils in there.

In this post I take you through a recent calculation to estimate the number of items we have in our micropalaeontology collection, and conclude that understanding how these collection sizes have been estimated is essential in deciding how to manage them.

Bottles of microfossil residues containing literally uncountable numbers of specimens.

It is relatively easy to make a quick and accurate size calculation for some parts of the microfossil collection. Slides are housed in standard cabinets holding 105 drawers that each hold 55 slides. The 24 standard cabinets in the Heron-Allen Microfossil Library therefore contain roughly 138,600 slides if they are full. By similar calculations, the Former Aberystwyth University Microfossil Collection contains a total of about 60,000 slides and the Former BP Microfossil Collection 300,000.

However, some cabinets are not completely full so we estimated percentage of expansion space and scaled down the figures accordingly. The total number of microfossil slides in the entire collection is estimated to be about 550,000.

But this is an estimate for the number of slides, not specimens. One slide, like the residue bottle, may contain 10s, 100s or even 1000s of specimens. Is it worth counting all of these? Probably not. You'd be there forever. Obviously when calculating the 70 million specimen figure, these vast numbers of additional specimens have not been taken into consideration otherwise the microfossil collection would have accounted for a large percentage of the total 70 million figure and perhaps even surpassed it!

This slide was counted as a single item in our size estimate for the microfossil collection. Each square contains a different species and multiple specimens are present, so these counts could legitimately be added to our total number of specimens for the collection.

It would appear that by counting slides and not the specimens on them, we are making the microfossil collection and hence the Museum collection appear smaller than it is. So does this matter?

This really depends on how you use the information. I think it is fine to give estimated figures like we do on the Museum website as it gives members of the public an idea of the vast size of the collection. On the other hand, if you use these figures to make decisions on how to allocate resources to the collection, then it becomes really important to account for the way in which the data is generated.

It wouldn't be right for example, to decide how much funding to give a museum relative to another one based on figures like these, without knowing how they had been generated. It's probably unwise to take too much notice of website details of the relative sizes of collections at the AMNH, Smithsonian and Natural History Museum, as the data has almost certainly been gathered in a different way by each institution.

Another slide with multiple specimens that counts for a single item within the 550,000 slides in the microfossil collection. The story behind this slide can be found in my Microfossil Christmas Card post.

If we have 70 million specimens in the Museum, and just over half a million in the microfossil collection, which is looked after by one curator, it would, on average, suggest that we need 140 curators to manage the entire collection. The actual figure is closer to 100. Taking these figures literally would therefore suggest that I am doing well to only have to manage half a million specimens!

Of course it is not that simple. Data derived from other parts of the Museum collection are not comparable. A tray of 100 identical sharks teeth for example would have been counted as 100 individual specimens, whereas the squared microfossil slide shown above would have counted as an individual item. Other parts of the collection might appear to require more management resources, until they are compared on an equal basis by separating out curatorial units sometimes referred to as 'collection lots'. The tray of 100 sharks teeth in this instance would count as one collection lot.

It would be wrong to suggest that collection size estimates are the only factors taken into consideration when deciding how to allocate resources across a vast collection like ours. Monetary value, state of conservation, suitability for display, visitor and loan demand, educational, scientific and historical significance are also taken into account. 

I would say that 70 million is probably an under-estimate of the size of the Museum collection if you take into consideration the 'microfossil factor' of collections where there are simply uncountable numbers of specimens within collection lots. I don't think we will ever come to a meaningful total if we attempt to count individual specimens.

However, it is vital that we are consistent in how we interpret the figures derived from our own collection, especially if we use them to help make decisions on how to manage it in the future. An estimation of the number of lots rather than specimens would help towards this.