Tag Archives: species-area relationships

Ten papers that shook my world – watching empty islands fill up – Simberloff & Wilson (1969)

Sadly this is the tenth and last in my series of the ten papers that had a great influence on my life as an ecologist.  I’m going to cheat somewhat and actually discuss three papers. In my defence they are extremely closely linked and I am pretty certain that in today’s publishing world they would all have had to have been combined anyway.  That aside, I really liked this experiment the first time I read about it and still rate it very highly.  I would, however, love to be able to travel back in time and give them a couple of hints with the benefit of hind-sight, although as the authors are two of the greatest living ecologists, Dan Simberloff and E O Wilson, I might be a bit apprehensive doing so 🙂 In any case, much of what I would have said was addressed a few years later (Simberloff, 1976).

Wilson and Simberloff wanted to practically test the island biogeography theory famously described by McArthur and Wilson a few year earlier (MacArthur & Wilson, 1967).  To do this they travelled to the Florida Keys and after due reconnaissance decided that the many mangrove “tree islands” would be ideal study sites (Figure 1).  Then came the really cool bit.

simberloff1

Figure 1.  Two of the experimental ‘islands’ from Wilson & Simberloff (1969)

They set about removing the arthropod animal life from nine of the islands (Figure 2), or as much as they could, by fogging with methyl bromide; not something we could do now.  They then monitored the islands at frequent intervals for the next year.  They had of course surveyed the islands before they fumigated them.

simberloff2

Figure 2.  What a cool project; defaunation in progress – from Wilson & Simberloff (1969)

simberloff3

Figure 3. Island equilibria – from Simberloff & Wilson (1970)

The major finding from their study was that recolonization happened quite quickly and that a year later had pretty much reached an equilibrium position (Figure 3).  Another important finding and one that has important implications for restoration and conservation strategies was that two years after the defaunation event, although the islands were well populated, the species composition, except for one island was less than 40% similar to the original inhabitants (Simberloff & Wilson (1970).  Most species present were new to those islands.  The analysis of the data presented in the two data papers is rather basic, some of the key island biogeographical premises are not addressed at all and I wondered why they had not done so.  Their data are all shown in some detail so it is possible to do some more analysis, which I took the liberty of doing.  The extra analysis shows why they did not discuss area effects per se .  The only significant relationship that I could find was that between the number of species and the distance from the ‘mainland’ source (Figure 4), which as predicted by MacArthur & Wilson (1967) was negative. Sadly, island size did not correlate with species number (Figure5).   Finally, there was a positive, but not significant relationship between the initial number of species found on an island and the number a year later (Figure 6).

simberloff3

Figure 4.  Relationship between distance from ‘mainland’ source and the number of arthropod species present (R2 = 0.65, P <.0.05) Data from Simberloff & Wilson (1970).

simberloff4

Figure 5.  Island diameter and number of arthropod species (not statistically significant, r2 = 0.19, although I am sure many politicians would view this as a positive trend). Data from Simberloff & Wilson (1970)

simberloff5

Figure 6.  Initial number of species on an island and number of species present one year later. Although it looks convincing (r2 = 0.54), there are too few observations to reach statistical significance.  Data from Simberloff & Wilson (1970)

Although this work was extremely influential, (my Bracknell roundabouts study owes a lot to it), there were two major flaws in the original experimental design.  Firstly the number of islands was very low, but of course this is understandable, given the effort and complex logistics required to remove the arthropods safely (Figure 2).  The other flaw was that the islands did not cover a large enough range of sizes, thus making it less likely for the species-area pattern to be detected which was a great shame.

As I mentioned earlier, these short-comings were not ignored by the authors, and a few years later Sinberloff (1976) reported the results of an enhanced study, again in the Florida Keys, where he was able to convincingly demonstrate the species-area effect.   I guess that this was pretty satisfying as it tied up a number of loose strings.  He also managed to get the phrase “flogging a dead horse” into his introduction 🙂

Of the three papers, Simberloff & Wilson (1969) is the most highly cited (according to Google Scholar, 618 to date) and became a “citation classic”* in 1984 at which time it had accumulated 164 citations.  Simberloff & Wilson (1970) has attracted 252 cites with Wilson & Simberloff (1969) trailing in third with a mere 158 cites.  As a point of interest, Simberloff (1976) has so far received 313 cites.  To reiterate, the original mangrove island study, despite its flaws was a fantastic piece of work and Sinberloff and Wilson won the Mercer Award of the Ecological Society of America for this work in 1971.

I can think of no better person to explain why Simberloff & Wilson (1969) deserves its place in the Ecological Hall of Fame than Simberloff himself who in the commentary to the 1984 citation classic article wrote “I think the main reason it is cited, however, and its lasting contribution, is not so much that it supports the [equilibrium] theory, as that it reported a field experiment on ecological communities, and thus seemed dramatically different from the correlative approach that dominated this field

 

References

MacArthur, R.H. & Wilson, E.O. (1967) The Theory of Island Biogeography Princeton University Press, Princeton.

Simberloff, D. (1976)  Experimental zoogeography of islands: effects of island size.  Ecology, 57, 629-648.

Simberloff, D. & Wilson, E.O. (1969) Experimental zoogeography of islands: the colonization of empty islands. Ecology, 50, 278-296.

Simberloff, D. & Wilson, E.O. (1970) Experimental zoogeography of islands: a two-year record of colonization. Ecology, 51, 934-937.

Wilson, E.O. & Simberloff, D. (1969) Experimental zoogeography of islands: defaunation and monitoring techniques. Ecology, 51, 267-278.

 

Leave a comment

Filed under Ten Papers That Shook My World

Entomological Classics – Southwood 1961 – The number of insect species associated with various trees

 

Nineteen-Sixty-One  was a momentous  year for entomology and ecology, although at the time I suspect few realised it.  Skip forward to 2013 when The British Ecological Society published a slim volume celebrating  the 100 most influential papers published in the Society’s journals.  The papers included in the booklet were selected based on the opinions of 113 ecologists from around the world, who were then asked to write a short account of why they thought that paper influential.  I was disappointed not to be asked to write about my nomination but instead asked to write about Maurice Solomon’s 1949 paper in which he formalised the term functional response.

The paper I had wanted to write about was included, but John Lawton had the privilege of extolling its virtues, and given the word limits did a pretty good job.  I do, however, feel that given its importance to ecology and entomology it deserves a bit more exposure, so I am taking the opportunity to write about it here.  I could have included this post in a series I have planned, called Ten Papers that Shook My World, but given the impact that this paper has had on entomologists I felt it deserved an entry in my Entomological Classics series.

For those of you who haven’t come across this paper before, this was an astonishingly influential paper.  Basically, Southwood, who despite his later reputation as one of the ecological greats, was an excellent entomologist, (in fact he was a Hemipterist), wanted to explain why some tree species had more insect species associated with them than others.  He made comparisons between trees in Britain, Russia and Cyprus and demonstrated that those trees that were more common and had a wider range had more insect species associated with them (Figure 1).

Southwood 1961 Fig 1

From Southwood 1961.  I was surprised to see that he had committed the cardinal error in his Figure caption of describing it as Graph and also including the regression equation in the figure pane; two things that I constantly reprimand students about!

Importantly he also showed that introduced trees tended to have fewer insects than native species.  He thus hypothesised that the number of insects associated with a tree species was proportional to its recent history and abundance and was a result of encounter rates and evolutionary adaptation.  He then tested this hypothesis using data on the Quaternary records of plant remains from Godwin (1956) making the assumption that these were a proxy for range as well as evolutionary age.

He commented on the outliers above and below the line suggesting that those above the line were a result of having a large number of congeners and those below the line either as being taxonomically isolated and/or very well defended.

He then went on to test his ideas about the evolutionary nature of the relationship by looking at trees and insects in Hawaii, (ironically this appeared in print (Southwood, 1960), before the earlier piece of work (Journal of Animal Ecology obviously had a slower turnaround time in those days than they do now).

Hawaiin figure

Figure 2.  Relationship between tree abundance and number of insect species associated with them (drawn using data from Southwood 1960).

Considering the research that these two papers stimulated over the next couple of decades, what I find really odd, is that Southwood, despite the fact that he was dealing with data from islands and that Darlington (1943) had published a paper on carabids on islands and mountains in which he discussed species-area relationships and further elaborated on in his fantastic book (Darlington, 1957), did not seem to see the possibility of using the species-area concept to explain his results.  It was left to Dan Janzen who in 1968 wrote

It is unfortunate that the data on insect-host plant relationships have not in general been collected in a manner facilitating analysis by MacArthur and Wilson’s methods (as is the case as well with most island biogeographical data). What we seem to need are lists of the insect species on various related and unrelated host plants, similarity measures between these lists (just as in Holloway and Jardine’s 1968 numerical taxonomic study of Indo- Australian islands), knowledge of the rates of buildup of all phytophagous insect species on a host plant new to a region, where these species come from, etc. Obviously, the insect fauna must be well known for such an activity. The English countryside might be such a place; it has few “islands” (making replication difficult) but a very interesting “island” diversity, with such plants as oaks being like very large islands and beeches being like very small ones, if the equilibrium number of species on a host plant (Elton, 1966; Southwood, 1960) is any measure of island size.”

 

In 1973 Dan Janzen  returned to the subject of trees as islands and cited Paul Opler’s 1974 paper in relation to the fact that the number of  herbivorous insects associated with a plant increases with the size of the host plant population (Figure 3), and further reiterated

Opler Figure

Figure 3.  Opler’s 1974 graph showing relationship between range of oak trees in the USA and the number of herbivorous insect species associated with them.

 his point about being able to consider trees as ecological islands.  Opler’s 1974 paper is also interesting in that he suggested that this approach could be used for predicting pest problems in agricultural systems, something that Don Strong and colleagues did indeed do (Strong et al., 1977; Rey et al., 1981), and that the concept of habitat islands and the species-area relationship could be used when designing and evaluating nature reserves, something which indeed has come to pass.

Again in 1974 but I think that Strong has precedence because Opler cites him in his 1974 paper, Don Strong reanalysed Southwood’s 1961 data using tree range (based on the Atlas of the British Flora)  as the explanatory variable  (figure 4) to explain the patterns seen.

Strong Figure

Figure 4Strong’s reworking of Southwood’s 1961 insect data using the distribution of British trees as shown in Perring & Walters1 (1962).

The publication of this paper opened the floodgates, and papers examining the species-area relationships of different insect groups and plant communities proliferated (e.g  leafhoppers (Claridge & Wilson, 1976); bracken (Rigby & Lawton, 1981); leaf miners (Claridge & Wilson, 1982); rosebay willow herb (McGarvin, 1982), with even me making my own modest contribution in relation to Rosaceous plants   (Leather, 1985, 1986).

Although not nearly as popular a subject as it was in the 1980s, people are still extending and refining the concept  (e.g. Brändle & Brandl, 2001; Sugiura, 2010; Baje et al., 2014).

Southwood (1961) inspired at least two generations of entomologists and ecologists, including me, and is still relevant today.  It is truly an entomological (and ecological) classic.

References

Baje, L., Stewart, A.J.A. & Novotny, V. (2014)  Mesophyll cell-sucking herbivores (Cicadellidae: Typhlocybinae) on rainforest trees in Papua New Guinea: local and regional diversity of a taxonomically unexplored guild.  Ecological Entomology 39: 325-333

Brändle, M. &Brandl, R. (2001). Species richness of insects and mites on trees: expanding Southwood. Journal of Animal Ecology 70: 491-504.

Claridge, M. F. &Wilson, M. R. (1976). Diversity and distribution patterns of some mesophyll-feeding leafhoppers of temperate trees. Ecological Entomology 1: 231-250.

Claridge, M. F. &Wilson, M. R. (1982). Insect herbivore guilds and species-area relationships: leafminers on British trees. Ecological Entomology 7: 19-30.

Darlington, P. J. (1943). Carabidae of mountains and islands: data on the evolution of isolated faunas and on atrophy of wings. Ecological Monographs 13: 37-61.

Darlington, P. J. (1957). Zoogeography: The Geographical Distribution of Animals. New York: John Wiley & Sons Inc.

Elton, C. S. (1966). The Pattern of Animal Communities. Wiley, New York.

Holloway, J. D., & Jardine, N. (1968). Two approaches to zoogeography: a study based on the distributions of butterflies, birds and bats in the Indo-Australian area. Proceedings of the Linnaean Society. (London) 179:153-188.

MacArthur, R. H. & Wilson, E.O. (1967). The Theory of Island Biogeography. Princeton University Press, Princeton, N. J

Janzen, D. H. (1968). Host plants as islands in evolutionary and contemporary time. American Naturalist 102: 592-595.

Janzen, D. H. (1973). Host plants as islands II.  Competitive in evolutionary and contemporary time. American Naturalist 107: 786-790.

Kennedy, C.E.J. & Southwood, T.R.E. (1984) The number of species of insects associated with British trees: a re-analysis. Journal of Animal Ecology 53: 455-478.

Leather, S. R. (1985). Does the bird cherry have its ‘fair share’ of insect pests ? An appraisal of the species-area relationships of the phytophagous insects associated with British Prunus species. Ecological Entomology 10: 43-56.

Leather, S. R. (1986). Insect species richness of the British Rosaceae: the importance of hostrange, plant architecture, age of establishment, taxonomic isolation and species-area relationships. Journal of Animal Ecology 55: 841-860.

Macgarvin, M. (1982). Species-area relationships of insects on host plants: herbivores on rosebay willowherbs. Journal of Animal Ecology 51: 207-223.

Opler, P. A. (1974). Oaks as evolutionary islands for leaf-mining insects. American Scientist 62: 67-73.

Perring, F.J. & Walters, S.M. (1962) Atlas of the British Flora BSBI Nelson, London & Edinburgh.

Preston,  C.D., Pearman, D.A. & Tines, T.D. (2002) New Atlas of the British and Irish Flora: An Atlas of the Vascular Plants of Britain, Ireland, The Isle of Man and the Channel Islands. BSBI, Oxford University Press

Rigby, C. & Lawton, J. H. (1981). Species-area relationships of arthropods on host plants: herbivores on bracken. Journal of Biogeography 8: 125-133.

Solomon, M. E. (1949). The natural control of animal populations. Journal of Animal Ecology 18: 1-35

Southwood, T. R. E. (1960). The abundance of the Hawaiian trees and the number of their associated insect species. Proceedings of the Hawaiian Entomological Society 17: 299-303.

Southwood, T. R. E. (1961). The number of species of insect associated with various trees. Journal of Animal Ecology 30: 1-8.

Sugiura, S. (2010). Associations of leaf miners and leaf gallers with island plants of different residency histories.  Journal of Biogeograpgy 37: 237-244

Rey, J.R.M.E.D. & Strong, D.R. (1981) Herbivore pests, habitat islands, and the species area relation. American Naturalist 117: 611-622.

Strong, D. R. (1974). The insects of British trees: community equilibrium in ecological time. Annals of the Missouri Botanical Gardens 61: 692-701.

Strong, D.R., D., M.E., & Rey, J.R. (1977) Time and the number of herbivore species: the pests of sugarcane. Ecology 58: 167-175

 

1Postscript

The Atlas of the British Flora by Perring and Walters (1962) was an iconic piece of work, although not without its flaws.  As with many distribution atlases it is based on a pence or absence score of plant species within one kilometre squares.  So although it is a good proxy or range it does not necessarily give you an entirely reliable figure for abundance.  A dot could represent a single specimen or several thousand specimens.   Later authors attempted to correct for this by using more detailed local surveys e.g. tetrads.  It must have been particularly galling for  Southwood that the Atlas didn’t appear until after he had published his seminal papers, but he later made up for it by reanalysing and extending his data from that original 1961 paper (Kennedy et al., 1984).

Those of us working in this area using the original Atlas had to count the dots by hand, a real labour of love especially for those widely distributed species; the new edition (Preston et al., 2002) actually tells you how many dots there are so the task for the modern-day insect-plant species-area relationship worker is much easier 😉

5 Comments

Filed under Entomological classics, EntoNotes, Ten Papers That Shook My World