10. Vanelslander, B. et al.

(2012) Daily bursts of biogenic cyanogen bromide (BrCN) control biolm formation around a
marine benthic diatom. Proc. Natl. Acad. Sci. U.S.A. 109,
24122417
11. Rajamani, S. et al. (2008) The vitamin riboavin and its
derivative lumichrome activate the LasR bacterial quorum-sensing receptor. Mol. Plant-Microbe Interact. 21,
11841192
12. Seyedsayamdost, M.R. et al. (2014) Hybrid biosynthesis of
roseobacticides from algal and bacterial precursor molecules. J. Am. Chem. Soc. 136, 1515015153
13. Wang, H. et al. (2014) A dual-species co-cultivation system
to study the interactions between roseobacters and dinoagellates. Front. Microbiol. 5, 311
14. Decelle, J. et al. (2012) An original mode of symbiosis in
open ocean plankton. Proc. Natl. Acad. Sci. U.S.A. 109,
1800018005
15. Hom, E.F.Y. and Murray, A.W. (2014) Niche engineering
demonstrates a latent capacity for fungalalgal mutualism.
Science 345, 9498
16. Scherlach, K. et al. (2012) Symbiotic cooperation in the
biosynthesis of a phytotoxin. Angew. Chem. Int. Ed. Engl.
51, 96159618
17. Kiers, E.T. et al. (2010) Mutualisms in a changing world: an
evolutionary perspective. Ecol. Lett. 13, 14591474
18. Grube, M. et al. (2015) Exploring functional contexts of
symbiotic sustain within lichen-associated bacteria by comparative omics. ISME J. 9, 412424
19. Zenobi, R. (2013) Single-cell metabolomics: analytical and
biological perspectives. Science 342, 1243259

Spotlight

The Origins of
Tropical Rainforest
Hyperdiversity
R. Toby Pennington,1,*
Mark Hughes,1 and
Peter W. Moonlight1

the 200 years since the Age of Exploration,

many questions about the origin of their
diversity remain unanswered. Two paradigms have dominated. The rst, attributed
to Alfred Russel Wallace (18231913), suggested that tropical rainforests are speciesrich because of their age and climatic
stability. Later characterised by Ledyard
Stebbins as the museum hypothesis,
the central idea is that rainforests are climatically stable, warm, wet places where
species accumulate because extinction
rates are low. This was later challenged
by theories that rainforests might be crucibles or cradles of diversication with
elevated speciation rates, an idea which
gained considerable traction in the 1970s
in the specic guise of refuge theory. Refuge theory suggested that more arid climates during recent glacial periods caused
speciation by repeated fragmentation of
the ranges of rainforest species.
The balance of old versus recent speciation
in rainforests has received considerable
attention, especially in the New World
tropics (Neotropics), and some consider
a consensus has been reached that the
museum and cradle models are not mutually exclusive and that neotropical diversity
represents a balance of both old and newer
species [2]. A new paper by Koenen et al.
[3] revisits this debate and re-casts this
consensus in a different light by providing
the novel and tantalising suggestion that

Traditional models for tropical spe- (A)

cies richness contrast rainforests
as museums of old species or
cradles of recent speciation. High
plant species diversity in rainforests may be more likely to reect
high episodic evolutionary turnover
of speciesa scenario implicating
high rates of both speciation and
extinction through geological time.

high plant species diversity in neotropical

rainforests may reect high episodic evolutionary turnover of species a scenario
implicating high rates of both speciation
and extinction through geological time.
The Eocene neotropical rainforests were
even more species-rich than modern rainforests [4]. Koenen et al. suggested that
this high Eocene diversity could be
explained by species radiations that
occurred rapidly during that time period,
but which are now extinct or largely so.
What is most interesting is their idea that
recent, rapid radiations may be reective
of episodic species turnover. They suggest
a model where rainforests are museums for
higher level taxa (orders, families or genera)
with diverse traits that provide the stock
for species diversications through time.
Their study constructs a time-calibrated
phylogeny for mahoganies (Meliaceae), a
family of 700 species of tropical trees
and shrubs (Figure 1). The distribution of
mahogany species richness is representative of rainforest diversity generally,
being markedly greater in the Neotropics
and tropical Asia, and suggests the family
is a good proxy for investigating tropical
diversication. The phylogeny was dated
using carefully interpreted fossils and produced considerably younger dates than
were found in previous studies, with age
estimates of the Meliaceae ranging
between 59.3 and 80.4 Ma. The origin
(B)

The forests of Amazonia contain 16 000 Figure 1. The Mahogany Family (Meliaceae). (A) Fruit of Cedrela weberbaurei Harms, a deciduous species
species of trees alone [1]. This remarkable restricted to seasonally dry forest in the Peruvian Andes (photo: R.T. Pennington). Koenen et al. [3] suggest that
deciduousness and seasonal or montane biomes are ancestral for mahoganies. (B) Flowers of Trichilia
diversity immediately struck the rst biolo- tomentosa Kunth (photo C.E. Hughes). Trichilia shows a signicant recent increase in diversication rate that
gists to explore tropical rainforests, but in occurred after its evolutionary entry into the rainforest biome [3].

Trends in Plant Science, November 2015, Vol. 20, No. 11

693

of the family lies in the Old World, but

dates for some subgroups of Meliaceae
species that are disjunct in the New and
Old Worlds are too young to be explained
by plate tectonics or migration via Laurasia
in warmer Eocene climates. The bulk of
the neotropical mahogany ora therefore
has its origins in serendipitous trans-oceanic
dispersal events.
Some previous phylogenetic studies of
tropical plants have shown more or less
constant lineage accumulation [5,6] and
hence potentially support the museum
hypothesis, but lacked the species-level
sampling required to pick up the most
recent diversication rate shifts. Studies
of lineage accumulation in palms by Couvreur et al. [5] with 8% of species sampled,
and in moonseeds (Menispermaceae) by
Wang et al. [6] with 17% of species sampled, give cut-off times of 24 Ma and
38 Ma, respectively, after which their
results on lineage accumulation become
unreliable due to missing taxon sampling.
Koenen et al. sampled 34% of mahogany
species, and estimated and corrected for
unsampled diversity in each genus based
on their taxonomic knowledge. They use
recent analytical advances by Rabosky [7]
to estimate per-branch diversication
rates and identify signicant diversication
rate shifts.
The Palaeocene age of the family would
support a museum hypothesis if there had
been a gradual accumulation of rainforest
taxa through the Cenozoic, but evidence
from species diversication rate estimates
through time and across lineages, trait
evolution and dispersal events show a
striking deviation from this. The Koenen
et al. study reconstructs the ancestral
ecology for mahoganies, and shows that
the evergreen species currently prevalent
in lowland tropical rainforests share a
deciduous most recent common ancestor
with origins in either seasonal or montane
tropical biomes. Following entry to the
rainforest biome in the late Oligocene or
early Miocene, three clades show recent
signicant increases in their diversication

694

rates, two of these rate shifts subtending

recent Amazonian rainforest radiations
within the large genera Trichilia and
Guarea. These ecologically similar rainforest radiations are contemporaneous, with
most species within them arising during
the Pleistocene.

species inferred from phylogenies such

as that presented by Koenen et al., coupled with fossil evidence for higher overall
tropical plant species-richness in the
Eocene [4], forces a re-casting of the
museum and cradle models into one
incorporating high episodic species turnover. Mahoganies are just one of many
tropical plant families and it seems likely
that the evolution of tropical rainforest
hyperdiversity has played out using the
whole cast of rainforest lineages. Particular constellations of extrinsic environmental conditions and intrinsic traits would
favour high diversication for certain lineages at certain times. Then, later extinction may reduce these lineages to perhaps
just a few species, the depauperons of
Donoghue and Sanderson [12]. Recent
geological times have been good for
mahoganies, but it may be another lineage's turn soon.

Hence, Koenen et al. provided further

convincing evidence of the youth of much
species-level plant diversity in the Neotropics (e.g., [8]), which has also been
demonstrated in Africa [9] and Asia [10].
It seems that entry into the rainforest
biome triggered the radiations in Meliaceae, but it would be interesting to investigate in more detail which biotic or abiotic
factors in the rainforest biome in combination with intrinsic plant traits may underlie
these diversications (e.g., using methods
outlined in [7]). Although Koenen et al. suggested such studies can only explain individual radiations in specic settings, future
meta-analyses of phylogenies may help Acknowledgments
develop more universal models of historical We thank Kyle Dexter and Colin Hughes for very
species assembly. In addition, similar phy- helpful comments
logenetic studies of diversication in groups
1
characteristic of different tropical biomes Royal Botanic Garden Edinburgh, 20a Inverleith Row,
Edinburgh EH3 5LR, UK
such as dry forests would add generality.
Although the model of rainforests being *Correspondence: t.pennington@rbge.ac.uk
museums for higher level taxa and cruci- (R.T. Pennington).
bles for species divergence comfortably ts http://dx.doi.org/10.1016/j.tplants.2015.10.005
the phylogenetic data for many tropical References
plant groups, the picture for birds and 1. Ter Steege, H. et al. (2013) Hyperdominance in the Amazonian tree ora. Science 342, 1243092
mammals may be different. Weir and
2. Rull, V. (2008) Speciation timing and neotropical biodiSchluter [11] show a stronger signature
versity: The Tertiary-Quaternary debate in the light of
molecular phylogenetic evidence. Mol. Ecol. 17, 2722
for recent (Pleistocene) species divergence
2729
at higher latitudes, but a spread of Pleisto- 3. Koenen, E.J.M. et al. (2015) Recently evolved diversity and
convergent radiations of rainforest mahoganies (Meliaceae)
cene and pre-Pleistocene divergence
shed new light on the origins of rainforest hyperdiversity.
towards the equator. Whether a contrast
New Phytol. 207, 327339
exists between the ages of temperate and 4. Jaramillo, C. et al. (2006) Cenozoic plant diversity in the
Neotropics. Science 311, 18931896
tropical plant species should be a priority
5. Couvreur, T.L.P. et al. (2011) Origin and global diversifor further research.
cation patterns of tropical rain forests: inferences from a
complete genus-level phylogeny of palms. BMC Biol. 9, 44

Concluding Remarks
Ultimately, reliable fossil data for mahoganies would be needed to conrm species
turnover through the Cenozoic, but the
fossil record for Meliaceae and other tropical plant families is fragmentary. However,
the now convincing evidence for the youth
of many neotropical rainforest plant

Trends in Plant Science, November 2015, Vol. 20, No. 11

6. Wang, W. et al. (2012) Menispermaceae and the diversication of tropical rainforests near the Cretaceous-Paleogene boundary. New Phytol. 195, 470478
7. Rabosky, D.L. (2014) Automatic detection of key innovations, rate shifts, and diversity-dependence on phylogenetic trees. PLoS ONE 9, e89543
8. Richardson, J.E. et al. (2001) Rapid diversication of a
species-rich genus of Neotropical rain forest trees. Science
293, 22422245
9. Harris, D.J. et al. (2000) Rapid radiation in Aframomum
(Zingiberaceae): evidence from nuclear ribosomal DNA

internal transcribed spacer (ITS) sequences. Edinb. J. Bot.

57, 377395
10. Thomas, D.C. et al. (2012) West to east dispersal and
subsequent rapid diversication of the mega-diverse genus
Begonia (Begoniaceae) in the Malesian archipelago. J. Biogeogr. 39, 98113
11. Weir, J.T. and Schluter, D. (2007) The latitudinal gradient in
recent speciation and extinction rates of birds and mammals. Science 315, 15741576
12. Donoghue, M.J. and Sanderson, M.J. (2015) Conuence,
synnovation, and depauperons in plant diversication. New
Phytol. 207, 260274

Spotlight

Does Microbial
Diversity Confound
General Predictions?
Marie Duhamel1,* and
Kabir G. Peay1,*,@
Microbes show more geographic
diversity than previously expected,
a serious challenge for ecological
prediction. However, a recent
study shows that microbial communities from a global grassland
plot network responded consistently to nutrient addition. These
results highlight the risks of nutrient deposition, but also hope for
generalized
understanding
of
microbial communities.
The explosion of DNA sequencing power
has nally enabled microbiologists to
begin the large task of mapping the diversity of microbial life across the planet.
While the high local diversity of microbes
has long been appreciated, recent largescale sequencing studies have revealed
much greater degrees of regional variation
than previously expected, even in similar
ecosystems [14]. For example, Talbot
et al. [1] found that there were almost
no species of fungi shared between similar
forest types in eastern and western North
America. The explanations for this variation are of course complex, and likely

involve barriers to dispersal, spatial variation in climate, environment, and resources, as well as competition with other
microorganisms. While understanding
the ecological and evolutionary factors
that generate this diversity is fascinating,
high complexity systems have always
been challenging for ecological prediction
and raise the specter that each regional
community may have its own unique
rules. As a result, the incredible regional
diversity of microbes raises concerns that
it will be impossible to develop general
theories and predictions about microbial
communities.
Making such predictions is critical in the
global change era. Microbial metabolic
activity and biomass turnover drive the
biogeochemical cycles that support
human agriculture and natural ecosystems [5]. In the last century, human activities have increased global phosphorus
mobilization by fourfold [6] and have doubled the amount of the annual nitrogen
xation in terrestrial ecosystems [7]. As
soil microorganisms are involved in critical
steps of N and P cycles, the anthropogenic addition of these nutrients in the last
decades is likely to impact the soil microbiome. However, individual studies in different parts of the world have shown
varying degrees of responsiveness of
microbial communities to anthropogenic
stressors [8,9]. In addition, most studies
tend to look at a single taxonomic component of the microbiome, either Bacteria,
Fungi, or Archaea. These groups have
both overlapping and unique functions in
the soil. For example, both bacteria and
fungi are involved in plant nutritional symbioses, but bacterial symbionts obtain N
through xation of atmospheric nitrogen,
while fungi scavenge nitrogen and phosphorous from organic and inorganic sources in the soil. All groups have free-living
taxa involved with organic matter decomposition, but often with different specializations. For example, anaerobic cellulose
decomposition is primarily carried out by
bacteria, while fungi are the primary
agents of lignin decomposition. Similar

patterns are found in key nutrient transformations, such as the mineralization of

nitrogen or phosphorus. Archaea are
often the most abundant ammonia oxidizers in terrestrial soils, and bacteria are the
most common denitriers. As a result of
these differences, the response of one
microbial group to anthropogenic perturbations may not be predictive of others.
For these reasons, it has been difcult to
predict whether microbial responses to
anthropogenic perturbations will be similar
across different geographic regions
(Figure 1A,B) or contingent upon local
community composition (Figure 1A,C).
Against this backdrop, a recent paper
from Leff et al. in PNAS [10] is particularly
noteworthy, because it addresses twin
challenges of generalizing microbial
responses to nutrient deposition both
across the globe, and across major microbial taxonomic groups. Using standardized sampling and DNA sequencing
methods, the authors characterized the
soil microbial community across a global
network of grassland research sites spanning four continents. For the last four
years, experimental plots within each
site have been subjected to N, P, or
N+P fertilization to mimic anthropogenic
nutrient deposition [3]. Despite distinct differences in soil physico-chemical characteristics and microbial composition across
these 25 grasslands, the authors found a
number of consistent responses in the
composition of microbial communities.
While the paper contains a wealth of
details and data that microbial ecologists
will mine for years, perhaps the primary
nding was that nutrient deposition
tended to shift the microbial community
toward fast-growing (or copiotrophic)
organisms. For bacteria this was indicated
by increased abundance of the phyla Actinobacteria, Alphaproteobacteria, and
Gammaproteobacteria, relative to Acidobacteria, Planctomycetes, and Deltaproteobacteria. This was accompanied by
reductions in expected genome size,
and a reduction of particular gene ontology categories, such as metabolism of

Trends in Plant Science, November 2015, Vol. 20, No. 11

695