MISCELLANEOUS 437 



gives higher ash content and better checking dupHcates than ashing in a 

 muffle furnace for the same time at 650° C (1202° F). Tliis method of 

 ashing leaves no residual carbon and volatilizes none of the chlorides except 

 magnesium chloride, whereas the higher temperatures necessary in the 

 muffle furnace volatilize chlorides of potassium and calcium and lead to 

 loss of the metals themselves by sublimation. This method also eliminates 

 the necessity of adding chemicals to further the ashing, such as sulfuric 

 acid, which gives too high an ash value by changing the chlorides and car- 

 bonates to sulfates. The authors describe the method in detail and give 

 the precautions necessary for avoiding explosions. 



Determination of reducing sugars. The permanganate method for 

 estimating reducing sugars is used extensivel}^ in biological work. It was 

 considered necessary to standardize the potassium permanganate against 

 cuprous oxide produced by a known quantity of pure glucose, since it was 

 claimed that the use of oxalic acid, sodium oxalate, or iron as a standard 

 gave results which were too low. 



Kraybill, Youden, and Sullivan ^^ standardized the permanganate with 

 Bureau of Standards sodium oxalate in acid solutions and also with weighed 

 samples of cuprous oxide prepared by reduction with invert sugar, and 

 obtained excellent agreement. The amounts of copper found by the per- 

 manganate method were closely checked by direct electrolytic determina- 

 tions of the copper. 



Capillary glass electrode. The complex and unstable nature of many 

 biological fluids, such as plant sap, milk, and blood, complicate the task of 

 determining the hydrogen ion concentration of these liquids. In many 

 cases it is a great convenience to use extremely small samples of Uquid. 

 This is true for work with insects or small localized regions in plants. The 

 glass electrode possesses the important advantage of eliminating the 

 possibility of altering the sample, as may happen by bubbling hydrogen 

 through the liquid with the hydrogen electrode or adding quinhydrone 

 for the quinhydrone electrode. The action of the glass electrode depends 

 on separating the test solution by a thin glass membrane from a reference 

 solution. A glass electrode ^°- in the form of a fine-bore capillary tube ^^^th 

 very thin walls was found to possess, besides the usual desirable features 

 of the glass electrode, the further advantages of employing less than 0.01 cc 

 of sample and exposing only a minute surface of the liquid to the air. 



Experimental Planning and Statistics 

 The variability of biological material has always been a source of diffi- 

 culty in biological experiments. In chemical and physical experimentation 

 the materials are usually more homogeneous and the environment more 

 easily controlled and reproduced. Various methods are open to the biolo- 

 gist in order to deal with this variability, which in itself is often a charac- 

 teristic of considerable interest. Among these methods the first to be used 

 was the employment of extensive replication of the samples and of the 



