Botany

Subfamily Caesalpinioideae is a paraphyletic grade of 171 genera that comprises the first branches of the Leguminosae and from which are derived the monophyletic subfamilies Mimosoideae and Papilionoideae. We have sequenced the chloroplast matK gene, and the trnL and 3′-trnK introns for 153 genera of caesalpinioid legumes. Parsimony and Bayesian phylogenetic analyses of these data support the monophyly of several major groups within the caesalpinioid legumes: the Cercideae, Detarieae, Detarieae s. str., Prioria , Amherstieae, Dialiinae, Cassia , Caesalpinia , Peltophorum , and Tachigali clades. Relationships among the first branching lineages of the legumes are not well supported, with Cercideae, Detarieae, and the genus Duparquetia alternatively resolved as sister group to all of the legumes. The division of certain large genera (e.g., Caesalpinia s. l., Bauhinia s. l.) into segregate genera generally is supported by our molecular data. Using 18 well-documented fossils as calibration points, fixing the stem node of the legumes at 65 Ma, and using the Penalised Likelihood method, we estimate the crown node of the Leguminosae at 64 Ma and the crown age of each of the major caesalpinioid lineages varying from 34 to 56 Ma. Fossil cross-validation suggests that none of the 18 fossil calibrations is internally inconsistent. Analyses done without fossil calibrations yield much younger divergence times. The age estimates for the Detarieae clade are more sensitive to the presence of calibration points than other caesalpinioid clades, a situation which we attribute to the slow rate of chloroplast DNA evolution in this group.

The aerial surface of plants, known as the phyllosphere, represents a widespread and diverse habitat for microbes, but the fungal communities colonizing the surface of leaves are not well characterized, and how these communities are assembled on hosts is unknown. We used high-throughput sequencing of fungal communities on the leaves of 51 tree species in a lowland tropical rainforest in Panama to examine the influence of host plant taxonomy and traits on the fungi colonizing the phyllosphere. Fungal communities on leaves were dominated by the phyla Ascomycota (79% of all sequences), Basidiomycota (11%), and Chytridiomycota (5%). Host plant taxonomic identity explained more than half of the variation in fungal community composition across trees, and numerous host functional traits related to leaf morphology, leaf chemistry, and plant growth and mortality were significantly associated with fungal community structure. Differences in fungal biodiversity among hosts suggest that individual tree species support unique fungal communities and that diverse tropical forests also support a large number of fungal species. Similarities between phyllosphere and decomposer communities suggest that fungi inhabiting living leaves may have significant roles in ecosystem functioning in tropical forests.

Saunders and McDevit recently reported their efforts to extract and amplify DNA by PCR in successively older red algal (Rhodophyta) herbarium specimens. They found that recent collections (4-11 years old) readily amplified but that archival material (decades to a century old) yielded contamination problems, diminished success correlated with age, or failed to amplify. As a solution, they proposed that epitypes be designated based on contemporary sequenced specimens. In response, we extracted and amplified in independent laboratories three loci (COI, ITS2, and rbcL) from the same 1836 Sparlingia pertusa specimen that Saunders and McDevit were unable to amplify. The use of Q-solution enhanced amplification success and likely is partly responsible for our achievements with archival specimens. These findings, along with data from the last 13 years in which we have sequenced over 100 historical and type specimens, indicate that with proper controls, amplifying DNA from red algal herbarium specimens of any age is practical and reproducible. The designation of contemporary epitypes should be a last resort, not an alternative to sequencing type material, and must be done with an understanding of the historical record of the species.

Species of the water fern Salvinia are well known for their extremely water-repellent floating leaves. The architecture of Salvinia surfaces is of great interest for biomimetic applications, because submersed in water, they retain air films for a long period. Knowledge of these surfaces is also important for pest control, since one species ( Salvinia molesta D.S. Mitch.) is a pantropic invasive aquatic weed. The micromorphology of the leaf surfaces of six representative species has been characterized by scanning electron microscopy. Based on their morphology, the trichomes are classified in four types, named after the typical species. Among the species examined, numbers, distribution, and sizes of the trichome types vary significantly. The simplest types, the Cucullata trichomes, are multicellular, uniseriate, and up to 800 µm high. Groups of two multicellular, uniseriate trichomes are described as the Oblongifolia trichomes. The Natans trichomes are grouped as four multicellular, uniseriate trichomes. The Molesta trichomes are composed of four trichomes, which are connected by the second last apical cells of the trichome. The ontogeny of the Molesta trichome groups was puzzling and resulted in various names being applied to them (“Krönchenhaare,” “egg-beater,” or “coroniform” hairs). Their unique development from four solitary uniseriate trichomes to groups of four connected trichomes is described in detail.

Quantifying the proportion of roots colonized by arbuscular mycorrhizal fungi (AMF) is routine work for researchers conducting AMF studies. However, in practice, the methods are always misused, with their adaptability to different conditions neglected. In this study, four frequently used methods (root segment ±, root segment estimation, grid-line intersect, and magnified intersections) were evaluated and compared. Using the light microscopy based staining technique, we assessed AMF colonization of the roots of five plant species (Trifolium repens Linn., Zea mays Linn., Robinia pseudoacacia Linn., Populus simonii Carr., and Caragana korshinskii Kom.). The results revealed that a root length of at least 150 cm (rather than the usual 30 or 50 cm or 100 to 150 intersections generally used when following these four methods) should be examined to represent a single root sample whatever the method used. All four methods had good reproducibility, even though there was a high level of divergence among the results obtained using the different methods to assess the same root sample. We concluded that when assessing the AMF colonization of roots from the same species, all methods except the root segment ± method can be used; however, when assessing root samples from different species, the root segment estimation and magnified intersections methods give more reliable results. We suggest that the root segment ± method is an effective method for revealing the uniformity of AMF distributed in host roots of a certain length.

4-Aminobutyrate (GABA) is a nonproteinogenic amino acid that functions in stress tolerance and signaling. Here, we report that salinity stress and elevated GABA levels coincided with the induction of glutamate decarboxylase 4 (GAD4) expression in Arabidopsis thaliana (L.) Heynh. ecotype Col-0. In-silico and microarray analysis revealed the over-representation of binding sites for WRKY and MYB transcription factors in the GAD4 promoter, as well as their co-expression with GAD4. Transcript profiling of liquid culture-grown, wild-type plantlets subjected to salinity stress for up to two days confirmed that GAD4 expression is associated with the inducible co-expression of WRKY28, WRKY30, WRKY40, MYB2, MYB15, and MYB108, as well as calmodulin-like 37 and aluminum-activated malate transporter 2, suggesting the involvement of gene regulation, protein activation, and anion transport in GABA accumulation. Transcript profiling of shoots from soil-grown, wild-type plants and corresponding single and double GAD mutants subjected to two days of salinity stress suggests that the GABA accumulation could involve post-translational activation of pre-existing GAD1 and GAD2 by elevated cytosolic calmodulin, as well as induction of GAD4 expression.

In the Rocky Mountain Forest District of British Columbia, a dramatic series of fires occurred during 2003, setting the stage for an abundant morel crop in the following year. During 2004, the abundance of post-fire morels ( Morchella spp.) was measured and the plant community associated with morel production was characterized. Morel production averaged 6473 ± 2721 morels·ha^–1in five burnt forests that were surveyed. Production ranged from 1702·ha^–1at Plumbob Mountain to a significantly higher 16827·ha^–1in the Kootenay National Park, where the highest level of duff consumption (71%) was also observed. Several plant species had high importance in morel habitat, and were also associated with above-average morel abundance: Chamerion angustifolium (L.) Holub, Arnica cordifolia Hook., Erythronium grandiflorum Pursh, Spiraea betulifolia Pallas subsp. lucida (Dougl. ex Greene) Taylor and MacBryde, Menziesia ferruginea Sm., Rosa acicularis Lindl. subsp. sayi (Schwein.) W.H. Lewis, Pinus contorta Dougl. ex Loud. var. latifolia Engelm., Abies lasiocarpa (Hook.) Nutt., and Picea glauca (Moench) Voss × engelmannii (Torr. & Gray ex Hook.) Brayshaw. Compositae and Vaccinium spp. were important species when considered as groups. Grass species, including Calamagrostis rubescens Buckl., were more proximate to morel-free plots. The characteristics of morel habitats observed in this study may be useful in future management of the resource, through conservation of habitat, management of prescribed burning, and postponement of salvage logging in potentially highly productive areas.