Molecular evolution, adaptive radiation, and geographic diversification in the amphiatlantic family Rapateaceae: evidence from ndhF sequences and morphology.

Rapateaceae (16 genera, approximately 100 species) is largely restricted to the tepuis and sandplains of the Guayana Shield in northern South America, with Maschalocephalus endemic to West Africa. The family has undergone extensive radiation in flower form, leaf shape, habit, and habitat. To analyze the evolution of these distributions and traits, we derived a molecular phylogeny for representatives of 14 genera, based on sequence variation in the chloroplast-encoded ndhF gene. The lowland subfamily Rapateoideae is paraphyletic and includes the largely montane subfamily Saxofridericioideae as a monophyletic subset. Overall, the morphological/anatomical data differ significantly from ndhF sequences in phylogenetic structure, but show a high degree of concordance with the molecular tree in three of four tribes. Branch lengths are consistent with the operation of a molecular clock. Maschalocephalus diverges only slightly from other Monotremae: it is the product of relatively recent, long-distance dispersal, not continental drift--only its habitat atop rifted, nutrient-poor sandstones is vicariant. The family appears to have originated approximately 65 Mya in inundated lowlands of the Guayana Shield, followed by: (1) wide geographic spread of lowland taxa along riverine corridors; (2) colonization of Amazonian white-sand savannas in the western Shield; (3) invasion of tepui habitats with frequent speciation, evolution of narrow endemism, and origin of hummingbird pollination in the western Shield; and (4) reinvasion of lowland white-sand savannas. The apparent timing of speciation in the Stegolepis alliance about 6-12 Mya occurred long after the tepuis began to be dissected from each other as the Atlantic rifted approximately 90 Mya. Given the narrow distributions of most montane taxa, this suggests that infrequent long-distance dispersal combined with vicariance accounts for speciation atop tepuis in the Stegolepis alliance.

Whole-plant nitrogen- and water-relations traits, and their associated trade-offs, in adjacent muskeg and upland boreal spruce species.

Black and white spruce (Picea mariana and P. glauca) exhibit a striking micro-geographic distribution pattern at the southern edge of the boreal forest. Black spruce grows in flooded nutrient-poor muskegs, while white spruce is found primarily on drier upland sites, and the two rarely form mixed stands. In an attempt to characterize the physiological and, hence, mechanistic basis of this pattern, we sampled five adjacent populations of black and white spruce from northern British Columbia and measured a suite of physiological and allocative characteristics, and associated trade-offs, that may be important to survival in habitats limited in nutrient or water availability. Two laboratory experiments were conducted: a greenhouse dry-down experiment to assess relative degree of drought tolerance; and a 2×2 nested factorial experiment in which seedlings were subjected to varying water and nitrogen regimes for approximately 16 weeks. White spruce was more drought-tolerant (i.e., maintained positive net photosynthesis at lower shoot water potential) and more efficient in water-use (as indicated by carbon isotopic composition) than black spruce. Black spruce was found to be significantly less sensitive to nitrogen stress, exhibited greater plasticity in nitrogen-use efficiency (measured as the carbon-to-nitrogen ratio in total plant tissue), and had a greater specific N absorption rate under high-N conditions than white spruce. Trade-offs hypothesized to be associated with these nitrogen and water relations traits were examined, but few were confirmed. Water-use efficiency and nitrogen-use efficiency did not trade-off between species, but did trade-off plastically (i.e., across treatments) within species. When exposed to simultaneous limitations of N and water both species were forced to utilize each resource with suboptimal efficiency. The change in isotopic composition per unit change in C/N ratio was not the same in the two species. This difference may reflect optimization of the trade-off, whereby each species maximizes the use efficiency of the most limiting resource (respective to its habitat), while minimizing the concomitant reduction in the use efficiency of the other resource.