|BOTANICAL ELECTRONIC NEWS|
|No. 347 April 15, firstname.lastname@example.org||Victoria, B.C.|
Botany BC 2005 will take place from Thursday May 26th through Sunday May 29th. Lytton will be the home base with field trips to the Stein and Botanie Valleys and the Skwaha Lake Ecological Reserve. The Program and Registration are available on the Botany BC website at: http://members.shaw.ca/dmeidinger/botanybc/
Presumably, everyone has some idea of what he or she means when they use these words. Surely, you may say, a common fungus is one we encounter often, and a rare fungus is one we find only occasionally (or may never have seen). But in saying this we are merely moving the argument back a level. What do we mean by 'often', and 'occasionally'? 'Never' is unambiguous for any particular study, but proving actual total absence (as would be the case in extirpation or extinction) is extremely difficult.
Certainly, we can all think of macrofungi we consider rare. I have seen the stellar Collybia racemosa on only a few occasions (How many other agarics bear visible anamorphs on their sexual fructifications?) I have found Asterophora lycoperdoides, which parasitizes other agarics, even less often, and I have encountered its congener, Asterophora parasitica, only once or twice (Too bad, since these are amazing species in which the tissue of the mushroom cap becomes converted into asexual chlamydospores). At the other extreme, Stropharia ambigua and Pluteus cervinus are seen in the woodlands near my home on a regular annual basis, as is Russula brevipes.
This raises the issue of geographical distribution. Some species which I remember as being 'common' in Ontario are not found at all on the west coast (and vice versa). That is a valid biogeographical issue, but it is beyond the scope of this essay.
Here are a few other variables and factors not yet fully considered in studies of fungal occurrence in restricted areas:
Of course, even at the local level, we can only expect a fungus to occur in the appropriate habitat. Ectomycorrhizal mushrooms can be anticipated only where their host trees flourish. Saprobic fungi may be more widely distributed, though some of them, such as Strobilurus trullisatus, which grows almost exclusively on decaying Douglas fir cones, are highly substrate- specific. This essay does not deal with that aspect of mushroom distribution. The assumption being made is that we are seeking the fungi in their normal haunts.
The same kind of caveat applies to seasonality. No point in looking for most mushrooms in August along the east coast of southern Vancouver Island - it is simply too dry. So we must assume that the sampling is done at appropriate times of year.
Size of individual fruit bodies
The size of a mushroom may also be expected to have some impact on the number of basidiomata produced. For example, we are unlikely to find as many fruit bodies of Russula brevipes, a very large agaric with caps 80-200 mm in diameter, as of Mycena aurantiidisca, the caps of which are usually only about 10 mm wide, since the biomass of an individual basidioma of the former must be several hundred times that of the latter, therefore representing a much larger investment of energy on the part of the mycelium (though we may note that the Russula obtains most of its energy directly from a cooperative tree, while the Mycena must depend on its own enzymes to degrade tree litter into an assimilable form.)
Not only the size, but also the longevity or persistence of fruit bodies will have an influence on the frequency with which they are recorded. There is some information on this, but it has yet to be compiled and consolidated, let alone factored into the equation that apparently needs to be attached to each taxon in this kind of study. Egli et al. (1997) found that monthly surveys recorded 31 per cent fewer taxa than weekly surveys. This reduction clearly springs from the differential longevity of fruit bodies in different species.
The size of basidiomata may also affect the likelihood of their being recorded. While Russula brevipes is hard to miss, tiny grey or brown Mycena species, unless present in numbers, can easily be overlooked, and there are much smaller 'macrofungi' out there as well. Nevertheless, I have assumed that, for the purposes of the various studies quoted here, experienced eyes will miss very little.
It is clear that ectomycorrhizal fungi such as species of Russula persist for years or generations on and around the roots of their plant partners. This makes it obvious that in years when such fungi do not fruit they are not absent from the habitat. Nevertheless we must be allowed to make at least some judgments about rarity on the basis of what we find fruiting, and the prolonged reluctance or inability of many fungi to make fruiting bodies on an annual basis would seem to affect their potential for long-distance dispersal.
In light of the vehicle in which this essay is published, let me draw some perhaps instructive comparisons with plants. When looking for a plant, we can in many cases say that if after prolonged and extensive searching, we find it only in a few locations, it is rare. This will certainly be true of persistent plants such as trees or perennials. We cannot say exactly the same about ephemeral plants such as some small spring annuals which, if not sought at the proper time, will simply not be found. A majority of fungal fruit bodies are similarly fugacious, and will not be found unless sought during their evolutionarily determined fruiting season. There is one important difference between ephemeral plants and fungi. The plants often persist only as seeds, while in many cases the fungi live on as hidden, though extensive, mycelia, which can now be detected by molecular techniques. It is far easier to see and identify the fruiting part of the life cycle, when it occurs. Nevertheless, molecular approaches may ultimately revolutionize our concepts of rarity, though that potential revolution still lies in the future.
Now we have clarified the conditions under which we are operating, we can proceed with the aim of the exercise, to devise useful (because partially quantified) definitions of 'common', 'rare', and intermediate categories, as they apply to fungi. This may help us in communicating about our collections, and may yield one or two generalizations about these matters. Fortunately, I am aware of some publications and databases which can be examined, and this paper will undertake a limited meta-analysis of these data, as well as of some unpublished data I have been involved in gathering.
There may be many different ways of defining common vs. rare in a quantitative manner, but I will concentrate on only two. The first method, on which we have the most data, is based on the frequency of occurrence of fungal fruiting bodies over a period of several to many years. It does not consider the numbers of basidiomata found in any single year, but simply whether a fungus has been recorded in a particular year (that is the only information available from most such studies). The second method considers the numbers of fruiting bodies encountered in a single excursion or season, and although detailed studies in Japan have followed the occurrence of Matsutake basidiomata over many years, few if any such studies have apparently been carried out on an all-taxa basis.
Method 1 - Extended linear studies
One thing that becomes apparent as soon as we begin to analyse the available multi-year databases is that although they can give us a handle of sorts on the matter at hand, their extended duration, although far beyond that possible in most grant- supported studies, is nevertheless not long enough to ensure a full accounting of all the fungi that have the capacity to produce fruiting bodies in the various study locations.
I'll begin by providing a few numbers from a study in which I was involved: a five-year macrofungal survey of Clayoquot Sound (Roberts et al. 2004). We recorded a total of 551 species, but only 28 were found in all years, and 310 were seen in only one year. On average we found over 100 newly encountered taxa each year. Although the study lasted for only 5 years, the results already suggest:
These numbers - the continuing high level of novelty we encountered - led us to assume that this accretion of taxa could be expected to continue for many years. Other longer-term studies confirm this impression.
A fascinating study by Tofts and Orton (1998) points out that although they had collected agarics regularly in a particular woodland in Scotland for 21 years, and had recorded 502 species in that time, in each successive year they still found species they had never seen before. They collected for over twenty years, and still could not say that they had a proper handle on agaric biodiversity in that woodland. They suggested that at least 25 to 30 years of collecting, and possibly more, would be necessary before that goal could be attained. I think they were being conservative in this estimate. It also seems intuitively obvious that fungi which appear only after more than two decades must be regarded as rare, at least by some definitions. Yet it is entirely possible that such laggards will be locally abundant when, eventually, they do fruit (see method 2).
A more recent paper by Straatsma et al. (2001) emphasizes many of the same points. They collected basidiomata weekly for 21 years (1975-1979, and 1984-1999), and recorded over 400 species in a 1500 m2 plot. Yet only 8 (eight) species (2 per cent) were found every year. The number of species found per year ranged widely - from 18 to 194 - and even in the last year of the study, 19 species appeared which had not previously been found. Clearly, the authors had not seen the full diversity of macrofungi that existed in their plot. Significantly, 37 per cent of the taxa they recorded were found in only one year.
In Fall 2003 the Cascade Mycological Society held its 16th successive mushroom fair at the Mount Pisgah Arboretum just outside Eugene, Oregon. As a guest speaker for the Society I was fortunate enough to be invited to participate in the collecting trips leading up to the fair. The fair is an exciting survey of the larger fungi because over 300 species are usually on display - bespeaking a huge effort on the part of many members. I was also fortunate enough to get my hands on the statistics for all sixteen years. Over those years almost 700 species have been recorded from the extensive and diverse habitats sampled by C.M.S. collectors. When we arranged the data according to the number of years in which each species had been collected, an interesting picture emerged.
Let us begin with the extremes. Only 37 species (5.5 per cent per cent of the total number) had been found in all sixteen years. Equally thought-provoking was the fact that no fewer than 190 species (almost 30 per cent of the total) had been recorded only once in those 16 years, and almost one hundred more (nearly 14 per cent) in only two of the 16 years. Here is the list of species versus years recorded.
|Species||Years recorded||% of total species|
There are possible flaws in the data set. For example, some species may have been misidentified. But the general trends are obvious. A relatively small number of taxa will show up every year, or almost every year, while a larger number of taxa will be found much less often, and a very large number will be encountered only once every decade or so. How many more taxa will show up in the years to come? What is the full number of species that the Cascade group can expect to find if they keep at it long enough? If we may be allowed to take a quick look in the crystal ball, might we not forecast that after 50 years they will have found 1,000 species fruiting?
This data set (for which I am indebted to the hardworking collectors and record-keepers of the Cascade Mycological Society) points up the necessity for very long-term studies wherever the diversity of fungi is to be fully explored, and calls for the accumulation of much concurrent data on weather conditions and other ecological factors if we are to understand why some fungi are so notably reluctant to fruit.
How are we to calculate common and rare in this case? It seems that we have no alternative but to make a few arbitrary decisions. For example, can a species be regarded as common if it does not occur every year? If we can countenance that concept, how many years of absence could be accepted for a 'common' species? We must remember that seasons differ widely in the degree of encouragement they offer to the fruiting of mushrooms - too dry, too cold, even too wet, are well-known situations. So it might be necessary to temper our purely numerical concerns with an injection of weather data. However, I do not have that information for any of the linear studies, and must leave it to their authors to provide such input, if they see fit.
My tentative, arbitrary, and open-to-debate conclusions from the linear Cascade Mycological Society study are as follows.
To iterate: in the linear C.M.S. study: abundant - 5 per cent, common - 10 per cent, sporadic - 35 per cent, uncommon - 20 per cent, rare - 30 per cent.
I believe that if the sampling is continued for another decade or more, the number of uncommon and rare fungi will increase substantially, producing a chart more closely resembling that emerging from the mould study reported below. This is a mathematical certainty, since no further abundant or common fungi could possibly emerge, while uncommon and rare fungi would continue to be added to the list.
Method 2: Concurrent frequency of occurrence
For the past two years I have been involved in several expeditions aimed at compiling a macrofungal inventory of the islands of Gwaii Haanas National Park Reserve in the Queen Charlotte archipelago (Haida Gwaii), British Columbia, Canada. During the expedition of Fall 2004 I undertook to record the numbers of fruit bodies of every taxon we encountered on the various islands. Over a period of 9 days I recorded well over 4,500 fruit bodies found by our 4- person team. We recorded 161 taxa, the numbers of specimens ranging from 1 to well in excess of 100 - I simply stopped counting when we exceeded 100 fruit bodies on any given foray. Of the 161 taxa, 53 were represented by a single fruit body, while 14 species considerably exceeded the 100 mark. Between those extremes there were 13 species with 2 fruit bodies each, 8 species with 3 fruit bodies, 10 species with 4, 9 species with 5, 6 species with 6, 19 species with between 10 and 19, 14 species with between 22 and 45, 5 species with between 59 and 66. So there was a concentration of taxa at the lower and upper ends of the scale, with very low numbers in between. If we accept fewer than 10 fruit bodies as indicating a degree of scarcity, this distribution suggests that 109 of the 161 taxa - almost 68 per cent - are uncommon. If we accept the discovery of 50 or more fruit bodies as indicating a common species, there were 19 such species, or 12 per cent of the total. The other 20 per cent lay somewhere between those extremes. It is also apparent that the 14 species (8.7 per cent of the total taxa) of which more than 100 specimens were seen (in several cases many more than 100) represent approximately 50 per cent of all fruit bodies encountered in the survey. That seems to me to be an incontrovertible measure of commonness.
My suggestions for ranking our data are as follows:
To iterate: in the concurrent Gwaii Haanas study: abundant - 9 per cent, common - 3 per cent, sporadic - 20 per cent, uncommon - 34 per cent, rare - 33 per cent.
Before continuing I must state a number of reservations about our data.
I hope that other surveys of this kind will be done, perhaps on a larger scale, perhaps repetitively over a longer period, so that the kind of information we can derive from them will be placed on a sounder statistical basis.
Our study, carried out by boat over a limited period, was perforce more a snapshot than a total seasonal compilation. It would be interesting to examine the results of, for example, a weekly sampling over an entire season, to see how the grouping of taxa might change. This would be impossibly expensive in Gwaii Haanas, but could certainly be done elsewhere.
I realize that it is unwise to read too much into a single survey, but our study does seem to confirm that there is a large number of uncommon species, a larger group of intermediate occurrence, and a much smaller number of common species, which last provide the majority of all fruit bodies. However, I suspect that had the surveys been more intensive or extensive, we might well have found a much larger number of rare or uncommon taxa, making the chart look more like that produced in the mould study below.
These results lead inescapably to a consideration of the manner in which identification keys should be constructed. It seems to me that such keys, especially when dichotomous, and when designed for amateurs, should concentrate on the common species, and should imply that taxa not covered are probably rare and should be left to the experts. This would mean that such selective keys would be shorter and simpler, and therefore have a greater chance of conferring success on the user. Although the species that are common in one geographic location will be different from those seen in another area, keys can be compiled either with a specific area (such as those produced by the Pacific North West Key Council) or particular habitats in mind, or, if wider ranging, will still deal with fewer taxa than the usual dichotomous keys encountered in the various manuals. By reducing the number of dichotomies, this simple strategy could prevent much frustration on the part of users, and would make such keys more accessible.
Fungi on indoor substrates
A database (Thiagarajan et al. 2004) of 167 fungal taxa recorded on various indoor substrates during 76,000 examinations of various kinds of samples (bulk, tape and swab) shows that there are only a few really common fungi in this environment.
It is interesting to note that of the 167 taxa noted here, 124 were hyphomycetes, 20 ascomycetes (including 2 of the top 10), 9 zygomycetes, 4 basidiomycetes, 4 coelomycetes, 3 myxomycetes (protozoans rather than fungi), 2 yeasts and 1 oomycete (a chromistan fungus).
My arbitrary suggestion, having examined these data, is that we regard the three taxa with more than 10 per cent frequency of occurrence as abundant, those between 10 per cent and 1 per cent as common, those with between 1 per cent per cent and 0.1 per cent as sporadic, those between 0.1 per cent and 0.01 per cent as uncommon, and those below 0.01 per cent as rare.
This arrangement has no persuasive statistical underpinning, but it looks as if, no matter what scheme of partitioning was to be adopted, a similar pattern would emerge, sooner or later. The more samples one examined, the more collecting trips one took, the more years a study endured, the more likely seems this last kind of bar chart. So this is the generalization that seems to emerge from this analysis. Not a surprise to many of us, but, given the profusion of mushrooms that emerged in Fall of 2004 on the west coast of North America, not something that stared us in the face as we forayed so successfully.
Rarity is likely to increase as fungal habitats are destroyed by human agency. Red lists will become longer (if it is not too late to compile them - many species will not even get to be in a red list before they disappear). Those wishing to analyze the matter of rarity more deeply can find food for thought in fine papers by Stebbins (1980) and Fiedler and Ahouse (1992), though those essays were written from the botanical point of view.
I would also like to cite a paper by Hawksworth (2004) in which he revisits the matter of the probable numbers of extant fungi as estimated from the number already found and described, and from the relative numbers of higher plants. It would appear that about 100,000 distinct taxa of fungi have been described, from an estimated total of about 1.5 million. It is significant in the context of the present document that we appear to have described only about 7 per cent of the fungi in existence. Hawksworth speculates that: (1) many of the missing fungi are in tropical forests; (2) many are in unexplored habitats; and (3) many are hidden in already described taxa: for example, the mould, Fusarium graminearum, formerly thought to be well- understood, is now known to comprise a complex of nine species. None of these caveats can disguise the fact that most of the 'missing' fungi are almost certainly rare.
In conclusion, I want to share a few delightful lines from page 131 of Stephen Jay Gould's book 'The Flamingo's Smile' (1985) which might well be applicable to the present discussion. He is analyzing a much earlier paper by Lord Kelvin which purported to establish a (much too young) age for the Earth:
'Thus, although all three arguments had a quantitative patina, none was precise. All depended upon simplifying assumptions that Kelvin could not justify. All therefore yielded vague estimates with large margins of error.'
I would like to thank Adolf Ceska for invaluable input during the gestation of this piece.
The statistical phenomenon called Raunkiaer's rule is well-known to ecologists. According to this rule the distribution of species in occupancy (constancy, ubiquity) classes shows a characteristic form (Raunkiaer 1934, Preston 1948, Dahl 1956, McIntosh 1962, Collins and Glenn 1990; for further literature see e.g. Gotelli and Simberloff 1987, Hanski 1982, McGeoch and Gaston 2002).The number of species belonging to the lowest and the upper occupancy class or classes significantly exceeds the number of species in middle occupancy classes. An occupancy class [n,m] is here a frequency class of species. A species belongs to this frequency class, if it occurs in n or n+1 quadrats or similar units. The number of occupancy classes is usually chosen to be between five and ten. It is debatable whether the rule is basically biological or can be derived from formal statistical conditions. Data on the validity of the rule in other disciplines would support the latter hypothesis.
We carried out related investigations on epidemiological statistics. A formal analogy between ecological and epidemiological scenarios is straightforward. Species can be thought of as corresponding to diagnosis categories and individuals to established diagnoses. Sample quadrats or similar units can correspond e.g. to nearly equal sized counties of a state or to equal time intervals. Our investigations were also motivated by some of our previous work on the possible background of the rule. We found earlier that the truncated lognormal distribution fitted well numerous epidemiological data sets (Izsak and Juh sz- Nagy 1982). On the other hand, by using simulation method we found that the lognormal distribution of species abundances leads to a valid Raunkiaer' rule (Papp and Izs k 1995). This is why we thought the rule would be valid in epidemiology and decided to test it.
The data were Hungarian statistics on infectious diseases and congenital anomalies at birth, both classified by counties and months. We established five classes for occupancies in the 19 counties (Table 1) and four for months (Table 2).
|Occurrence class||Number of counties in which the disease occurred|
Table 1. Number of counties in which the disease occurred.
|Occurrence class||Number of months in which the disease occurred|
Table 2. Number of months in which the disease occurred.
The main results are the numbers of diagnosis categories falling into each occurrence class (Tables 3,4). The distribution in occupancy classes for counties is given in Table 3.
|Occurrence class||Infectious diseases||Congenital anomalies at birth|
Table 3. Number of diseases falling into the occurrence classes for counties.
For example, figure 8 in 1984 refers to the fact that eight infectious diseases were reported in 1-4 counties in that year. The distribution in occupancy classes for months is given in Table 4.
|Occurrence class||Infectious diseases||Congenital anomalies at birth|
Table 4. Distribution in occupancy classes for months.
Note that the congenital anomalies data make it possible to imagine the change in shape of occurrence class vs. number of diagnoses graphs got by increasing the sample size, c.f. Gaston (1994). The numbers demonstrate unambiguously the validity of the rule in epidemiology. Of course, with very small or large total diagnosis numbers the rule would appear distorted. As mentioned above, the origin of the rule is probably the (nearly) lognormal distribution of the diagnosis frequencies. Indeed we found that the truncated lognormal distribution fitted the diagnosis frequencies well. Computations were carried out using the program package DIVERSI 2.1 (Izs k 2003), containing programs for fitting and diversity calculations and available at no cost by e- mail from the author.
A general conclusion is that analogous statistical conditions lead to similar statistical phenomena in remote scientific fields. For example, processes affecting plant species and infectious diseases (but not congenital diseases) show similar traits from a dynamical point of view. A remarkable link between these processes can be the data on the validity of Raunkiaer's rule for anthropochorous plants observed in isolated old houses and small villages, cited by Hanski(1982). At any rate, our findings are warning informations for researchers searching for ecological background of the rule. Naturally, possible ecological consequences of the rule and, in turn, the near-lognormal distribution, remains a question.
A note: not mentioned in the article, we have observed signs of latitudinal decline of virus diversity moving north in England (Hunter and Izsak 1993). This finding can be another data arguing for a more wide occurrence of a phenomenon often mentioned in ecology.
I should express my gratitude to Professor Mark Williamson for his valuable remarks and numerous corrections in the text.
Local people in B.C. often refer to two varieties of slugs - "good" slugs, the native giant Banana Slug Ariolimax columbianus and "bad" slugs, the invasive giant Black Slug Arion spp.
In contrast, Land Snails of British Columbia covers the natural history of the entire terrestrial gastropod fauna including 92 species of snails and slugs. This book is the first comprehensive checklist and key to landsnails in western Canada, and clearly, it fills an important gap. I am personally impressed that Forsyth declines to cash in on the tremendous popularity of the Banana Slug (Ariolimax columbianus) and gives it only equal or less coverage than other species, many of which are very poorly studied. Every snail or slug has a one to two-page description including identification features, distribution and life history, and a selection of essential primary references. Forsyth uses contemporary classification based on recent work rather than relying on the classic Pilsbry definitions, but also includes common names for all taxa. The most attractive feature of the book is the identification key, for use in the field, presented in terminology that is precise (and impartial) but not intimidatingly technical. Copious black and white illustrations also help distinguish the animals. Since no gastropods are endemic to B.C., this publication will be of great use to all malacologists west of the Rocky Mountains.
The book is up to the high standards of the other Handbooks published by the Royal BC Museum, and a great addition to any natural history library.
[For Robert Forsyth's Key to the slugs of British Columbia see BEN # 320, Dec 30, 2003. - AC]
Your April Fool's BEN story on the Peruvian orchid must be the basis for a recent evening Krim" I watched when I was too tired to sitting and looking at lichens. It was in the German evening Krimi series Tatort, set at a huge, posh orchid show near Lake Konstanz, with a guy who recently described an illegally acquired red lady's slipper murdered in cold blood by a female orchid fanatic who claimed it was her right to describe. Like the Selby story, this was portrayed as the "orchid find of the century", though in this case from Vietnam. The customs inspector who was caught on the crime scene trying to steal the plant later confessed to selling exotic orchids illegally to an Oriental collector - two crimes nailed with one swipe. Remarkably, the orchid involved looked not unlike that on the St. Petersburg Times website, although it was much darker red. The type specimen, we were told, went to the herbarium at the University of Konstanz. I just found the episode online, under the name Der Name der Orchidee, at http://www.daserste.de/tatort/sendung.asp?datum=06.03.2005
All the best, Toby
[Der Name der Orchidee_ (SWR) Klara Blum (Eva Mattes) Sonntag, 6. M„rz 2005 im Ersten "Alle jagen den roten Frauenschuh" Guess the Chief Botanist's name in this story! - His name was Dr. Klaus Raven! - AC]
Send submissions to email@example.com
BEN is archived at http://www.ou.edu/cas/botany-micro/ben/