Field-Cage Evaluation of Survival, Reproduction, and Feeding Behavior of Adult Scymnus coniferarum (Coleoptera: Coccinellidae), a Predator of Adelges tsugae (Hemiptera: Adelgidae).

The hemlock woolly adelgid, Adelges tsugae Annand, is an invasive pest of eastern (Tsuga canadensis (L.) Carrière) and Carolina hemlock (Tsuga caroliniana Engelmann) forests in the eastern United States. Scymnus (Pullus) coniferarum Crotch (Coleoptera: Coccinellidae) is a lady beetle that preys on A. tsugae in the western United States, where A. tsugae infestations on western hemlocks are not lethal. It is thought that S. coniferarum could be an important predator that helps keep A. tsugae populations from reaching damaging levels in this region. This study assesses the potential of this predator as a biological control agent for A. tsugae in the eastern United States. S. coniferarum predation, reproductive potential, and survival were evaluated in field-cages on adelgid-infested T. canadensis at two sites in southwestern Virginia. Sampling was conducted between December 2012 and June 2014 to evaluate the impact of S. coniferarum on both generations of A. tsuage (sistens and progrediens). Adult S. coniferarum fed on both generations and all life stages of A. tsugae during both field trials at rates comparable to other adelgid-specific predators. Evidence of S. coniferarum oviposition was minimal, and may be attributed to low temperatures and prey availability. S. coniferarum mortality was greatest when exposed to winter temperatures at the higher elevation site in 2013, and least throughout the 2014 spring sample period. S. coniferarum demonstrated a high predation rate on A. tsugae and survived for extended periods of time at sites in southwest Virginia, indicating that this species could be an effective predator of hemlock woolly adelgid in similar climates.

Host-range testing of Eucryptorrhynchus brandti (Coleoptera: Curculionidae), a candidate for biological control of tree-of-heaven, Ailanthus altissima.

Ailanthus altissima (Mill.) Swingle, tree-of-heaven, is an invasive species native to Asia. It first was introduced into the United States in the 1700 s and now is distributed throughout much of North America. Mechanical and chemical controls are current suppression tactics, however, implementation is costly. A weevil, Eucryptorrhynchus brandti (Harold), was identified in China and imported for quarantine testing in 2004 as a potential biological control agent. Host specificity tests on adult feeding, larval development, and oviposition of this weevil were conducted from 2007 to 2011 on A. altissima and 29 nontarget species. Eucryptorrhynchus brandti adults fed significantly more on A. altissima foliage when compared with all test species. Range of means for feeding on A. altissima was 32.5-106.5 mm(2)/adult/d. In no-choice tests, Simarouba glauca DC, Leitneria floridana Chapm., and Citrus limon (L.) Burm. F., had feeding rates of only 10, 49, and 10%, respectively, compared with the level of feeding on A. altissima. The mean range of adult feeding by E. brandti on all other test species was <7% of feeding on A. altissima (0.0-3.3 ± 5.0 mm(2)/adult/d). In the no-choice larval inoculation tests, larval development only occurred in two of 10 L. floridana seedlings compared with seven of 10 A. altissima seedlings. In the no-choice oviposition tests, oviposition and subsequent larval development did not occur in L. floridiana, whereas all seven A. altissima seedlings supported oviposition and subsequent larval development. The weevil did not appear to be a threat to L. floridana or any other nontarget species tested. Therefore, we conclude that Eucryptorrhynchus brandti is highly host specific to A. altissima.

A comparison of neural networks and partial least squares for deconvoluting fluorescence spectra.

This article compares backpropagation neural networks (BNN) with partial least squares (PLS) techniques in terms of their ability to deconvolute fluorescence spectra. Both actual experimental and simulated spectral data are studied for 2 binary systems. These systems consist of mixtures of tryptophan and tyrosine, and NADH and tryptophan over a total concentration range of 10(-7) to 10(-4) M. It is shown that BNN is superior to PLS for both systems.

Modeling human operators using neural networks.

Often in the process industries, human operators rather than mathematically based advanced algorithms are used to achieve product control. Typically, if there are three shifts of operators per day, one shift achieves superior control. If the expertise of the best operator can be captured easily and economically and made available to the other operators, significant economic benefits would accrue. This paper discusses a methodology that uses artificial neural networks for capturing the knowledge of process operators. For many operator tasks, only readily available information obtained from a process control computer is required. Once a converged network is available, a stripping technique can be employed to simplify the net and to gain knowledge about what a good operator is doing compared to a poor one. It is felt that the proposed approach is superior to traditional expert system techniques employing knowledge extraction approaches.

Comparison of four neural net learning methods for dynamic system identification.

Four types of neural net learning rules are discussed for dynamic system identification. It is shown that the feedforward network (FFN) pattern learning rule is a first-order approximation of the FFN-batch learning rule. As a result, pattern learning is valid for nonlinear activation networks provided the learning rate is small. For recurrent types of networks (RecNs), RecN-pattern learning is different from RecN-batch learning. However, the difference can be controlled by using small learning rates. While RecN-batch learning is strict in a mathematical sense, RecN-pattern learning is simple to implement and can be implemented in a real-time manner. Simulation results agree very well with the theorems derived. It is shown by simulation that for system identification problems, recurrent networks are less sensitive to noise.

Pharmacokinetic modeling of heparin and its clinical implications.

Experimental work on heparin has indicated that its half-life increases with dose. Two models to describe heparin's pharmacokinetic behavior are proposed, and the parameters in the models are fitted to experimental data. Both models exhibit an apparent first-order decay with a "half-life" that increases with dose. It is shown that, even though both models exhibit a bolus half-life of from 1 to 2 hr, over 2 days can be required for true steady-state conditions to be achieved in these models when a constant intravenous infusion of drug is given. The clinical implications of these models are discussed. Suggestions are made for further research on heparin kinetics.

The biologic half-life of heparin.

Half-lives (t1/2s) varying from 23 min to 2.48 hr have been reported for heparin. It is shown that much of the discrepancy between authors can be attributed to methods used to analyze data. It is suggested that three terms should be carefully used, the t1/2 of bioassayed heparin concentration, the t1/2 of the extension of the clotting time, and the t 1/2 of the clotting time to differentiate among the various methods of analyzing the raw data.