Bronson R T, Hedley-Whyte E T
J Comp Neurol. 1977 Dec 1;176(3):315-29. doi: 10.1002/cne.901760302.
The effects on rat cranial nerve growth of removing various amounts of extraocular muscle was studied using morphometric techniques. Growth in the third cranial nerves was found to be severely retarded when most of the muscle tissue was removed. By contrast, removal of the eye alone, leaving extraocular muscles relatively intact, was found to have little or no effect on the subsequent growth of third and sixth cranial nerve fibers and of extraocular muscle fibers. This conclusion could be drawn only through the application of statistical methods which take into account several generally unrecognized facts: frequency distributions of axon circumference and myelin sheath thickness are highly variable from nerve to nerve even in normal rats, which often have more large fibers in left than in right nerves. The bimodal nature of peripheral nerve fiber distributions precludes the use of such parametric tests as the commonly and inappropriately used t-test, but a non-parametric test such as the Kolmogorov-Smirnov test, extensively used in these studies, is inadequate for data comprising several sets of distributions to be compared. The application of the analysis of variance to some of the data and the merits of the procedure are discussed.
利用形态测量技术研究了切除不同量眼外肌对大鼠颅神经生长的影响。当大部分肌肉组织被切除时，发现第三颅神经的生长严重受阻。相比之下，单独摘除眼球，而使眼外肌相对完整，发现对第三和第六颅神经纤维以及眼外肌纤维的后续生长几乎没有影响。只有通过应用考虑到几个通常未被认识到的事实的统计方法才能得出这一结论：即使在正常大鼠中，轴突周长和髓鞘厚度的频率分布在不同神经之间也有很大差异，正常大鼠左神经中的大纤维通常比右神经中的多。外周神经纤维分布的双峰性质排除了使用诸如常用但不恰当的t检验等参数检验，但在这些研究中广泛使用的非参数检验，如柯尔莫哥洛夫-斯米尔诺夫检验，对于包含几组要比较的分布的数据是不够的。讨论了方差分析在一些数据中的应用以及该方法的优点。
