54 PRESIDENTIAL ADDRESS— SECTION 15. 



The original alcoholic extract will contain a very large 

 number of substances, many of which are of no interest from a 

 physiological point of view, such as carbohydrates, fats, oils and 

 waxes, phytosterols and phytosterolins ( ])h)-tosterol glucosides). 

 free acids, esters, essential oils and colouring matters. Flavonol 

 derivatives, such as k?empferol and (|uercetin, anthracjuinone 

 derivatives, as chrysophanol and emodin. glucosides, alkaloids, 

 bitter principles (amaroids). saponins, toxallnmiins. as ricin and 

 abrin, tannins and active resins, are usually the substances which 

 impart toxic or medicinal properties to the plants. With the 

 exception of a few grou])s which are always encountered at a 

 certain sta^e of the extraction j^rocess. it is impossible to i)redict 

 at which juncture each of these are likely to be isolated. Fats, 

 hydrocarbons and ifree alcohols are always found in the petro- 

 leum ether extract of the original resin ; essential oils are always 

 obtained in the distillate (^f the steam distillation process. But 

 the i)harmac(;logically and j^hysiologically interesting compounds 

 may be extracted by any one of the solvents previously referred 

 to. Saponins, for example, have been found in the ac|ueous 

 lif|uid as well as in the resin. For this reason it is desirable in 

 every case to carry out a com])lete investigation of the plant. 

 An additional difficulty lies in the fact that a very large number, 

 if not the majority, of the principles are resinous; and with our 

 im])erfect knowledge of the nature and composition of resins, 

 and in the absence of general methods of treatment, the ])ro- 

 blem becomes one of extraordinary intricacy. Moreover, with 

 the very large number of constituents in the original resin, with 

 ■the many processes to which it is subjected and the inevitable 

 losses in purification, the amount of material obtained at each 

 .stage is small, imless a very considerable weight of the i)lant 

 material is employed initially. Often, by the time a substance 

 is obtained in a ])ure homogenous state, the total quantity avail- 

 able is less than a gram, and it re(|uires very careful and skilful 

 manipulation to make so small an amount suffice for complete 

 characterisation of an unknown compound, combustion, deter- 

 mination of melting-point or boiling-point, of molecular weight, 

 optical rotation, acetyl or benzoyl derivatives, physiological 

 tests, etc. A careful study of the literature on the chemistry 

 of other plants of the same species is often an excellent guide 

 as to how to proceed, or what to expect. For example, I identi- 

 fied daphnin in Lasiosiphon polyccphahis ( Januariesbosje). and 

 the same glucoside has been found to occur in two Furopean 

 species of the same order ( Thymelace?e ) , Daphne ]nc::;crcnm 

 and Daphne alpina. 



Turning now to the physiological action of the plants, our 

 kowledge, although more advanced than with regard to the 

 chemistry of the constituents, is still sadly deficient. With 

 regard to ])lants eaten by stock, a number have, of course, been 

 known for a long time to be extremely poisonous, such as 

 .^langkop. Cape and Transvaal tulp, giftblaar, etc., and there 

 is no doubt about the sym])toms and post-mortem lesions. But 



