Plants produce a variety of structural and chemical defences to deter herbivores, which can covary across every scale of biological organization. Along these lines, it has been suggested that young leaves differ from old leaves in their defence strategies, but studies have largely focused on woody plants and/or chemical traits. Thus, we lack an understanding of how structural and chemical defences ontogenetically covary across a large portion of plant diversity. We investigate the range of structural and chemical defences of thistles (tribe Cardueae), which comprise herbaceous species with a wide spectrum of defensive traits. We collected structural and chemical trait data across species and leaf ages from field collections, herbarium specimens and the literature. Species' young leaves tend to produce similar suites of traits, with greater interspecific variation in older leaves. Young leaves are dominated by chemical defences, while older leaves produce a broader range of both chemical and physical defence traits. Leaf spininess is associated with a distinct chemical profile that differs from that of non-spiny species. Our study demonstrates how defence strategies can vary across the leaves of herbaceous species and their ages. Expanding such efforts will be critical to further our macroevolutionary understanding of variation in plant defences.Read the free for this article on the Journal blog.
In this paper, we study thermodynamics and its applications of a family of static charged dilaton black holes in 2+1 dimensions found by Chan and Mann [Phys. Rev. D50, 6385 (1994) [Erratum-ibid. 52, 2600 (1995)]] and Xu [Eur. Phys. J. C79, 642 (2019)]. There is a dimensionless parameter N in the black hole solutions presented: It is related to the coupling constant for the dilaton with the electromagnetic field and the gravitational field. Black hole horizons exist only for 2/3 <= N<2. N=1 black hole is a solution to low-energy string theory. Thermodynamics is studied in the canonical ensemble where charge is constant as well as in grand canonical ensemble where the potential is constant. The cosmological constant is considered as a thermodynamical variable where the pressure P=-Lambda 8 pi. We computed the first law and the Smarr relations for the black hole and introduced two new thermodynamical parameters in order to satisfy the first law. We computed temperature, thermodynamic volume, specific heat capacities, Gibbs free energy and studied local and global stability of the black hole. Thermodynamic volume differs from the geometric volume. In the canonical ensemble, we noticed that thermodynamic behavior falls into two broad categories: For 23 <= N<1, small black holes are locally stable and large black holes are not. For 1 <= N<2, the black hole is locally and globally stable for all values of the horizon radius. In order to demonstrate the two broad categories, we have presented N=1,23 and N=67 black holes in detail. There were no phase transitions for the above values of N. In the grand canonical ensemble, we noticed that there is a Hawking-Page phase transition for the black hole with N=6/5. We have also studied the Joule-Thomson expansion and the Reverse Isoperimetric Inequality of these black holes. We made the observation that the charged dilaton black hole does not violate the Reverse Isoperimetric Inequality for certain values of the parameters of the theory. Finally, we have suggested future work.
Using a novel climate policy uncertainty (CPU) measure based on emissions legislation, climate protests, and presidential statements, we show that in response to climate policy risk, firms tend to strategically reduce their future innovation, measured by patent counts, citations, and innovation value. Green innovation, however, is less vulnerable than non-green innovation. We argue that CPU constrains innovation through a precautionary motive: firms under high uncertainty face higher external financing costs and shift from prospector (innovation-oriented) to defender (cost-minimizing) strategies. Results are robust to instrumental variable estimation, firm fixed effects, and a difference-in-differences design exploiting the Paris Agreement as an exogenous shock. These findings underscore the strategic and policy implications of uncertainty in the transition to a low-carbon economy. Overall, our results provide insights into policy implications.
We review results from ongoing monitoring campaigns (in the optical, UV, X-rays and radio) on some of the most highly variable AGN known, and the identification of new AGN in extreme flux or spectral states, including some of the highest-amplitude outbursts observed to date, deep low-states, unexpected long-term trends, and systems which exhibit extreme Seyfert-type transitions ("changing-look AGN"). Long-term lightcurves, densely covered for multiple years, and follow-up spectroscopy in different spectral bands are used to shed light on the underlying variability mechanisms including accretion disk and broad-line region physics. Remarkable differences are seen, for instance, in the optical spectral response to extreme outbursts, implying very different intrinsic variability mechanisms. If time allows, I will also review most recent results from multi-year projects to test binary supermassive black hole models of the highly variable blazar OJ 287.
Existing benchmarks measure capability – whether a model succeeds on a single attempt – but production deployments require reliability – consistent success across repeated attempts on tasks of varying duration. We show these properties diverge systematically as task duration grows, and that pass@1 on short tasks is structurally blind to this divergence. We introduce a reliability science framework for long-horizon LLM agents with four metrics: Reliability Decay Curve (RDC), Variance Amplification Factor (VAF), Graceful Degradation Score (GDS), and Meltdown Onset Point (MOP). We evaluate 10 models across 23,392 episodes on a 396-task benchmark spanning four duration buckets and three domains. Key findings: (1) reliability decay is domain-stratified – SE GDS drops from 0.90 to 0.44 while document processing is nearly flat (0.74 to 0.71); (2) VAF bifurcates by capability tier – high VAF is a capability signature, not an instability signal; (3) capability and reliability rankings diverge substantially, with multi-rank inversions at long horizons; (4) frontier models have the highest meltdown rates (up to 19 strategies that sometimes spiral; and (5) memory scaffolds universally hurt long-horizon performance across all 10 models. These results motivate reliability as a first-class evaluation dimension alongside capability.
