硅酸盐-盐（硫酸盐）流体不混溶：蒙古南部Mushugai-Khuduk含火成碳酸岩杂岩体矿物中的熔体包裹体研究

Crystalline and melt inclusions were studied in garnet, diopside, potassium feldspar, and sphene from the garnet syenite porphyry of the carbonatite-bearing complex Mushugai-Khuduk, southern Mongolia. Phlogopite, clinopyroxene, albite, potassium feldspar, sphene, wollastonite, magnetite, Ca and Sr sulfates, fluorite, and apatite were identified among the crystalline inclusions. The melt inclusions were homogenized at 1010℃～1080℃ and analyzed on an electron microprobe. Silicate, salt, and combined silicate-salt melt inclusions were found. Silicate melts show considerable variations in SiO2 concentration （56 to 66wt% ）, high Na2O ＋ K20 （up to 17wt% ）, and elevated Zr, F, and Cl contents. In terms of bulk rock chemistry, the silicate melts are alkali syenites. During thermometric experiments, salt melt inclusions quenched into homogeneous glasses of predominantly sulfate compositions containing no more than 1.3wt% SiO2. These melts are enriched in alkalis, Ba, Sr, P, F, and Cl. The investigation of the silicate and salt melt inclusions in minerals of the garnet syenite porphyries indicate that these rocks were formed under influence of the processes of crystallization differentiation and magma separation into immiscible silicate and salt （sulfate）liquids.     

Subspace design of low-rank estimators for higher-order statistics

Higher-order statistics (HOS) are well known for their robustness to additive Gaussian noise and ability to preserve phase. HOS estimates, on the other hand, have been criticized for high complexity and the need for long data in order to maintain small variance. Since rank reduction offers a general principle for reduction of estimator variance and complexity, we consider the problem of designing low-rank estimators for HOS. We propose three methods for choosing the transformation matrix that reduces the mean-square error (MSE) associated with the low-rank HOS estimates. We also demonstrate the advantages of using low-rank third-order moment estimates for blind system estimation. Results indicate that the full rank MSE corresponding to some data length N can be attained by a low-rank estimator corresponding to a length significantly smaller than N.     

青海省布青山早古生代末期埃达克岩的发现及其构造意义

在青藏高原北部阿尼玛卿蛇绿岩带布青山段首次发现了埃达克岩,它们多呈不规则岩脉状侵入于布青山地区北部早古生代蛇绿混杂岩中.其岩性为花岗-英云闪长岩,主矿物有斜长石、石英、黑云母、角闪石、少量钾长石和辉石;副矿物有榍石、磷灰石、锆石和磁铁矿等.属钙碱系列,具典型（Ⅰ类、O型、HSA）埃达克岩的地球化学特征,表现为：高SiO：（63.65%～71.59%）,高Al2O3（14.81%～16.11%）,高Na2O（4.68%～5.33%）和Na2O/K2O（1.48～3.08,平均2.39）,高Sr（444.08×10^-6～560.41×10^-6）和Sr/Y（50.8～105.3）,HREE强烈亏损,低Yb（0.38×10^-6～0.74×10^-6）,高La/Yb（44.93～90.97）,弱负Eu异常（δEu=0.74～0.90）,富集大离子亲石元素（LILE）,亏损高场强元素（HFSE）,Ba、Pb正异常而Nb、P负异常.获得布青山埃达克岩的锆石TIMS U-Pb年龄为402±24Ma.布青山埃达克岩熔体来自俯冲的昆祁秦洋壳（含海洋沉积物）在石榴石角闪岩相条件下部分熔融,主要残留相为石榴石（±角闪石）.布青山埃达克岩的发现指示早古生代末期洋壳（昆祁秦多岛洋的一部分）俯冲作用的存在,俯冲方向朝北（现在方位）,此前,俯冲带发生长时间的向洋迁移,早古生代末期俯冲带已后退到阿尼玛卿蛇绿混杂岩带以南;暗示典型埃达克岩也可以形成于洋壳俯冲晚期;提示阿尼玛卿蛇绿岩带可能有与典型埃达克岩有关的Au、Ag、Cu、Mo等热液和斑岩矿床.     

On orthogonal polynomials

A study is made of the orthogonal polynomials on certain curves in the complex plane. Necessary and sufficient conditions for a set of polynomials to be orthogonal on the curves are obtained in terms of symmetric matrices. The relations of the symmetric matrices to Toeplitz matrices and innerwise matrices are shown.     

Preschoolers' quarantining of fantasy stories

Preschool‐aged children are exposed to fantasy stories with the expectation that they will learn messages in those stories that are applied to real‐world situations. We examined children’s transfer from fantastical and real stories. Over the course of 2 studies, 3½‐ to 5½‐year‐old children were less likely to transfer problem solutions from stories about fantasy characters than stories about real people. A combined analysis of the participants in the 2 studies revealed that the factors predicting transfer differed for the fantasy and real stories. These findings are discussed within the context of their implications for preschoolers’ developing boundaries between fantasy and real worlds.     

Videobased lesson analysis: Effective science PD for teacher and student learning

The Science Teachers Learning from Lesson Analysis (STeLLA) project is a videobased analysis‐of‐practice PD program aimed at improving teacher and student learning at the upper elementary level. The PD program developed and utilized two “lenses,” a Science Content Storyline Lens and a Student Thinking Lens, to help teachers analyze science teaching and learning and to improve teaching practices in this year‐long program. Participants included 48 teachers (n = 32 experimental, n = 16 control) and 1,490 students. The STeLLA program significantly improved teachers' science content knowledge and their ability to analyze science teaching. Notably, the STeLLA teachers further increased their classroom use of science teaching strategies associated with both lenses while their students increased their science content knowledge. Multi‐level HLM analyses linked higher average gains in student learning with teachers' science content knowledge, teachers' pedagogical content knowledge about student thinking, and teaching practices aimed at improving the coherence of the science content storyline. This paper highlights the importance of the science content storyline in the STeLLA program and discusses its potential significance in science teaching and professional development more broadly. © 2011 Wiley Periodicals, Inc., J Res Sci Teach 48: 117–148, 2011     

Peer collaboration: the relation of regulatory behaviors to learning with hypermedia

Peer collaboration is a pedagogical method currently used to facilitate learning in classrooms. Similarly, computer-learning environments (CLEs) are often used to promote student learning in science classrooms, in particular. However, students often have difficulty utilizing these environments effectively. Does peer collaboration help students learn with these environments? Little research looking closely at face-to-face peer collaboration with computer learning environments exists. Utilizing a social-cognitive theoretical framework, this study investigated the relation between the conceptual-knowledge learning and the collaborative regulatory behaviors of students working with a peer as they studied about the human circulatory system using a hypermedia CLE. Fifty-four high-school students from the East Coast of the United States were audiotaped to identify the collaborative regulatory behaviors they evidenced as they studied. Results revealed significant correlations among students’ proportion of categories of regulatory behaviors and their learning gains (from pretest to posttest). Moreover, qualitative analyses revealed particular behaviors that larger-gain collaborative pairs engaged in to a greater extent than smaller-gain pairs as they learned with the hypermedia environment.     

Analysis of the efficiency of applying artificial neuron networks for solving recognition,compression, and prediction problems

The paper summarizes the experience of developing the PS ANN image recognition software system based on artificial neural networks (ANNs). The real abilities of neural networks to solve image recognition, data compression, and extrapolation problems are demonstrated. The results of experimental studies on accelerating parallel task computations are presented.