S. H. Collins and B. Thomas 281 



equally promisintr as a subject of enquiry. It is worthy of note that 

 whereas the chief digestible carbohydrate of oat grain is starch, which 

 is condensed dextrose, the sugar of the straw is chiefly laevulose. In 

 other words, straw supplies the feeding deficiency of oat grain, if cane 

 sugar, which is a condensed form of both sugars together, is considered 

 the complete food sugar. 



Samjjles of oat straw from many parts of Great Britain have been 

 examined for their chief constituents and as far as possible the conditions 

 under which the straw was grown have been recorded. From the results 

 obtained, averages of groups of results are given in the following notes. 

 The general method of analysis follows common practice for oil, albu- 

 minoids, fibre and ash. The estimation of laevulose has been described 

 by one of us (2), the estimation of the other sugars being by Fehling's 

 volumetric method. At times the method of Ling (3) was used for an 

 end point but in most cases the solutions were sufficiently free from 

 colour to render Ling's indicator superfluous. At times the oat straws 

 were extracted with ether to remove some substances that prevented 

 the cuprous oxide from setthng(4) but in most cases these precautions 

 proved unnecessary. Nevertheless the authors consider that both 

 methods are very valuable at times when difficulties arise. As regards 

 Ling's indicator the authors found that a simpler recipe worked even 

 better. A small ball rolled up from about haK a yard of flower wire was 

 placed in a test tube with 2 c.c. hydrochloric acid and 2 c.c. water and 

 boiled for one minute. A small sjjoonful (about | gni.) of ammonium 

 sulpho-cyanide was added and allowed a minute or two to stand and 

 cool. This indicator kept, in its original test tube with the residue of 

 the iron wire, very well for a week or two, since fresh iron dissolved as 

 fast as oxidation took place. Also carbon oxysulphide is formed which 

 acts as a reducing agent. In practice it is better to make fresh indicator 

 daily. In early samples pentoses were looked for, but either none or 

 the merest traces could be found. From the polarimetric readings and 

 other properties the unidentified sugar appears to be mostly dextrose. 



The albuminoids are the usual N / %\ figures. All the early samples 

 were also treated so as to give the ammonia volatile with steam and 

 10 per cent, potash and the nitrogen insoluble in various "protein pre- 

 cipitants." In the end it was found that the N precipitated by basic 

 lead acetate and the N volatile as NH3 by steam and potash almost 

 exactly added up to the total N and that other N precipitants were 

 uncertain. Further the amount of "non-albuminoid" nitrogen as de- 

 duced from such estimations was too small in amount to have much 



