Speices diversity and the systematics of some New World leaf litter Thysanoptera (Phlaeothripinae; Glyptothripini)

Abstract More than fifty species of closely related glyptothripine Thysanoptera
have been described from leaf litter at one site in southern Brazil. The ecological and evolutionary significance of this species swarm are discussed, and the relationships of the New World glyptothripine genera are reassessed. A key is produced to the world genera of the Glyptothripini, and a check list is provided to the fifteen genera and 119 species from the New World. Nomenclatural changes include thirty-five new combinations, five new generic synonyms and one new specific synonym.     

Phylogeny and classification of the Xeromelissinae (Hymenoptera: Apoidea, Colletidae) with special emphasis on the genus Chilicola

Abstract A phylogenetic analysis and classification are provided for the bee subfamily Xeromelissinae based on 248 morphological characters, many of which are novel and illustrated. A total of 47 ingroup species was included in the analysis, representing at least two divergent members from each described genus or subgenus and seven taxa which did not readily fall within previously described subgenera. Three most parsimonious trees were found (length, 1508; consistency index, 40; retention index, 70). The result shows that Xeromelissa renders Chilimelissa paraphyletic, and the 20 known Chilimelissa species are reassigned to Xeromelissa, giving the following new combinations: X. luisa (Toro & Moldenke) (this is the type species of Chilimelissa), X. mucar (Toro & Moldenke), X. xanthorhina (Toro), X. brevimalaris (Toro), X. rosie (Toro and Packer), X. laureli (Toro and Packer), X. chusmiza (Toro), X. longipalpa (Toro), X. pedroi (Toro & Moldenke), X. australis (Toro & Moldenke), X. chillan (Toro & Moldenke), X. farellones (Toro & Moldenke), X. machi (Toro), X. minuta (Toro & Moldenke), X. nortina (Toro & Moldenke), X. sielfeldi (Toro & Moldenke), X. obscura (Toro & Moldenke), X. irwini (Toro & Moldenke), X. nolanai (Toro & Moldenke) and X. rozeni (Toro & Moldenke). Group support was estimated using symmetric resampling and group supported/contradicted (GC) ratios, which compare the frequency of each most parsimoniously resolved clade with the alternative arrangement that was most commonly found in resampling. Relationships among subgenera for Chilicola are weakly supported. By contrast, when three previously synonymized subgenera (Stenoediscelis, Heteroediscelis and Oediscelisca) are resurrected, there is good support for all subgenera (GC ≥ 99), except two: a paraphyletic Oediscelis and a polyphyletic Anoediscelis. Both of these subgenera became monophyletic following successive approximations character weighting. Four distinctive new subgenera are described: Unicarinicola Packer subgen.n. , Obesicola Packer subgen.n. , Capitatiscopa Packer subgen.n. and Toroediscelis Packer subgen.n. , and a revised key to the subgenera of Chilicola is provided.     

Investigating the vulnerability of an African savanna tree (Sclerocarya birrea ssp. caffra) to fire and herbivory

Sclerocarya birrea ssp. caffra (marula), a typical savanna tree, is vulnerable to the effects of fire, herbivory and their combination. This paper investigated the relative importance of these agents of disturbance, at the level of the individual stem, by specifically focusing on the following questions: (i) What is the greatest cause of mortality in adult marula stems in conservation areas with both elephants and fire? (ii) Does fire interact with bark stripping to cause adult stem mortality and if so what is the dominant mechanism? (iii) At what stem diameter are marulas resistant to fire? Field surveys quantified the extent of damage in marula individuals in the Kruger National Park, South Africa, highlighting the high levels of extreme herbivory such as toppling (7%) and pollarding (8%), relative to bark stripping (only 6% with more than 50% of the circumference stripped). In addition to extreme herbivory, the progression from bark stripping through to invasion of the soft, exposed heartwood by wood borers, often facilitated by fire, through to toppling of the weakened stem after successive fires, appears to be the dominant mechanism by which fire interacts with herbivory to cause adult stem death. Bark stripping and fire manipulation experiments indicated that bark stripping failed to increase the vulnerability of stems to fire directly through transport tissue damage. However, the combination of bark stripping and fire reduced the ability of the stem to regrow bark, increasing the vulnerability of the exposed stem to boring insects and future fires. Fire manipulation experiments were used to identify the minimum stem diameter of resistance to fire. Marula resisted stem death when greater than 3.4 cm in basal diameter. This paper emphasizes the importance of both fire and herbivory in the development of woody plant population structure and by extension, the relative proportion of trees and grasses in savanna landscapes.     

Estimation from quasi life tables

A large number of identical components are simultaneously setinto operation at each of a regular sequence of times and, onfailure, a component is instantaneously replaced. The functionalform of the distribution of the component lifelength is assumedto be known and it is desired to estimate the parameters ofthis distribution. However, no observations of the lifelengthsof individual components have been recorded: the only informationavailable is the numbers of components which fail between successivetime points. It is shown how this information, in conjunctionwith the theory of recurrent events, may be used to constructa nonparametric estimator of the discretised lifelength distribution.The latter is then used to estimate the unknown parameters.     

The functional anatomy of the rodent larynx in relation to audible and ultrasonic cry production

Rodents are well known for the production of physically quite different audible and ultrasonic cries. Both of these cries are known to be produced in the larynx, but the anatomy of the larynx is practically unknown, it thus being difficult to analyse how the cries are produced. The anatomy of the larynx of Mus musculus is described in detail and is found to be essentially typical of that of other mammals that can fix the thorax and make independent use of the forelimbs. Unlike the larynx of microchiropteran bats, no modifications for ultrasound production are apparent, although the laryngeäl structure is ideal for the production of audible cries. However, the audible and ultrasonic cries differ so markedly that they are unlikely to be produced by the same mechanism. The lack of laryngeal specialization therefore makes the ultrasound production mechanism largely an enigma.