AN EMPIRICALLY BASED MODEL OF VARIATION IN ROTATIONAL ELEMENTS IN CENTRIC DIATOMS WITH COMMENTS ON RATIOS IN PHYCOLOGY 1
AN EMPIRICALLY BASED MODEL OF VARIATION IN ROTATIONAL ELEMENTS IN CENTRIC DIATOMS WITH COMMENTS ON RATIOS IN PHYCOLOGY 1
复制标题
DOI：
10.1111/j.1529-8817.1988.tb04483.x
复制
发表时间：
1988-09
影响因子：
2.9
通讯作者：
E. Theriot
中科院分区：
文献类型：
--
作者：
E. Theriot
比率通常用于将藻类细胞的分类、生理和生态特性与细胞形状和大小的变化联系起来。然而，比率可能比原始变量更难解释。给出了硅藻分类学中常用比率的一个详细例子。中心硅藻阀上所报告的旋转元件的丰度通常表示为密度（例如，每μm圆周的边缘棘数），而不是简单地表示为总数。经验证明，密度通常与直径呈曲线关系，而总数与直径呈线性关系（如果有关系的话）。这种线性关系是解释实际例子中密度变化的经验模型的基础。在这个模型下，密度将是一个比总数更差的分类特征，除非在极少数情况下（例如，当总元素对直径的斜率回归为零，分子的方差不依赖于分母时）。其他比率同样难以解释。长度/宽度（或宽度/长度）通常与宽度（或宽度/长度）呈曲线关系，因为变化是由几个来源复合的。表面积/体积被用来研究大小和形状对浮游植物下沉的相对贡献。然而，这一比率的性质还没有得到充分的研究，尺寸、形状和下沉速度之间的关系也没有得到充分的阐明。
Ratios are commonly used to relate taxonomic, physiological and ecological properties of algal cells to variation in cell shape and size. However, ratios can be more difficult to interpret than the original variables. A detailed example of a ratio commonly used in diatom taxonomy is presented. The abundance of a reported rotational element on a centric diatom valve is often expressed as density (e.g. number of marginal spines per μm of circumference), rather than simply as total number. It is empirically demonstrated that density if often curvilinearly related to diameter, whereas total number is linearly related (when related at all) to diameter. This linear relationship is the basis for an empirical model which accounts for variation in density in real examples. Under this model, densitywill be a poorer taxonomic character than total number except under rare occasions (e.g. when the slope of the total element versus diameter regression in zero and variance in the numerator is not dependent on the denominator). Other ratios are similarly difficult to interpret. Length/width (or width/length) is often curvilinearly related to width (or width/lenght) is often curvilinearly related to width (Length) because variation is compounded from several sources. Surface/volume has been employed to investigate the relative contributions of size and shape to phytoplankton sinking. However, the properties of this ratio have not been fully investigated, and the relationship between size, shape and sinking rate has not been fully elucidated for the data set in question.