榧树营养苗端的解剖学研究

本文对榧树(Torreya grandis)成熟胚、幼苗及成熟植株不同生长季节营养苗端进行了组织学、组织化学和超微结构的研究。并观察了核酸、蛋白质及不溶性多糖物质在苗端的分布及其变化规律等。 观察结果表明：榧树成熟胚中已形成顶端分生组织突起却无分区现象。胚根出种皮前顶端分生组织出现类似原套一原体的结构，并最后变化到细胞组织分区样结构。苗端蛋白质、RNA含量始终较高，出种皮后细胞器丰富，结构清晰，液泡较多。幼苗在第一年连续多次产生新枝，直到秋末形成冠区。 成熟植株的季节生长中，营养芽的生长可分四个阶段：休眠期、叶扩展期、芽鳞形成期、新枝条形成期。其中休眠期的苗端体积中等，整个芽积累多糖、RNA含量较丰富，但蛋白质在苗端的含量少于叶中。各区域细胞超微结构差别较明显。叶扩展期的苗端体积迅速减小;积累的淀粉粒按一定顺序从各区域消失;液泡减少。蛋白质在苗端含量增加。RNA含量很少。芽鳞形成期的苗端体积恢复中等大小，多糖再次积累，细胞质浓厚，蛋白质含量明显高于叶中，RNA含量仍较少。细胞中出现许多小液泡，细胞壁较薄。新枝条形成期的苗端体积迅速增大，续而又减小。多糖先消失，后期再次积累;蛋白质极为丰富，RNA含量也较高，各种细胞器丰富，液泡增大。在整个生长周期中，苗端始终保持一定的分区形式。即具顶端原始细胞，亚顶端细胞群，周缘分生组织和肋状分生组织。但在不同的发育阶段各区域繁简不同。同时，多糖、蛋白质和核酸的组化反应的分区特性与细胞组织学分区式样基本一致。 最后，本文重点讨论了榧树苗端的分区。实验表明：成熟胚及幼苗的苗端分区现象不稳定，成熟植株营养苗端具明显的细胞组织学分区，并为组织化学及超微结构的证据细胞壁、液泡和质体的变化等，为苗端分区较明显的特征。此外，还讨论了榧树营养芽发育的阶段划分，以及根据苗端结构，讨论红豆杉科的系统位置等问题。

Shortspine thornyhead (Sebastolobus alascanus) abundance and habitat associations in the Gulf of Alaska

Shortspine thornyhead (Sebastolobus alascanus) abundance was estimated from 107 video transects at 27 stations recorded from a research submersible in 1991 off southeast Alaska at depths ranging from 165 to 355 m. Numbers of invertebrates in seven major taxa were estimated, as was substrate type. Thornyhead abundance ranged from 0 to 7.5/100 m2, with a mean of 1.22/100 m2, and was positively correlated with depth and amount of hard substrate. Invertebrate abundances were not significantly correlated with numbers of thornyheads. Shortspine thornyhead abundance estimates from this study were several times higher than estimates produced by bottom trawl surveys off southeast Alaska in 1990 and 1993, the two years of survey that encompassed the submersible transects; however, the trend of increasing abundance with depth was similar in the trawl surveys and in the submersible transects, suggesting that trawl surveys systematically underestimate abundance of shortspine thornyheads

Monthly spatial occurrence of phytoplankton and zooplankton in River Ogun, Abeokuta, Ogun State, southwest Nigeria

The aim of this study was to investigate the monthly spatial occurrence of phytoplankton and zooplankton in River Ogun, Abeokuta, Southwest Nigeria. This was carried out for seven months between December, 2011 and June, 2012 in 4 stations. A total of 41 species of phytoplankton and 16 zooplankton species from 5 classes respectively were recorded. Zooplankton was dominated by Cladocera throughout the study period while phytoplankton was dominated by blue green algae (Cyanophyta or Cyanobacteria). The dominance of Cyanophyta in this river is similar to findings by Sekadende and co-workers, Ogato, Deng and co-worker, and Shakila and co-worker. The dominance of Cladocera in this river is similar to findings by Ude and co-workers, and Ogbuagu and co-worker.

人工神经网络在植被遥感影像分类中的应用研究

比起传统的统计方法，人工神经网络具有很好的非线性处理和并行计算能力，在植被遥感信息处理中得到广泛的应用。本研究系统地介绍了人工神经网络理论及其在植被遥感信息处理中的应用现状。并就如何提高人工神经网络的相干被遥感影像的分类能力进行了详细研究。首次提出了结合植被指数和组成分分析的神经网络分类方法。过去这方面的研究工作大都集中在通过选择一个合适的神经网络模型来提高植被分类精度，而我们认为：根据植被遥感自身的规律，结合统计方法，确定合适的网络输入模式的特征变量，也可以提高分类精度。 研究结果表明，尽管一般的神经网络分类器不需要对输入的模式做明显的特征提取，网络的隐层就具有特征提取的功能。但对TM影像七个波段和常用的五个植被指数（PVI、NDVI、WDVI、PVI、MSAVI2），分别做主成分分析，从而获得人工神经网络输入的特征变量，使用这样一种结合VI、PCA的神经网络对遥感TM多波段影像进行植被分类，能大大提高分类的精度。

Some perspectives on the late Quaternary paleoclimate of Beringia

EXTRACT (SEE PDF FOR FULL ABSTRACT): Analyses of the modern summer synoptic climatology of Beringia illustrate that the region cannot be treated as a homogenous climatic unit as a result of different circulation controls that operate over the region. GCM (general circulation model) simulations and information from the modern synoptic climatology were used to infer the summer paleosynoptic climatology of the region since the last glacial maximum. ... Variations in these climatic controls offer important implications in assessing the vegetation histories of western Beringia versus eastern Beringia.