— 103 — 



products which are closely allied, in chemical constitution and com- 

 position, may differ completely from this point of view. These are 

 all peculiarities which cannot possibly be compared with the conditions 

 prevailing for colouring matters. And for this reason it will hardly 

 ever be possible to speak of a theory of odoriferous substances, in 

 like manner as is now done of a theory of dye-stuffs. The paper 

 concludes with a discussion on the manner in which the sensation 

 of odour is accomplished, and on the methods by which odours are 

 tested and compared with each other. In the last-named work a good 

 memory for the individual odours is very useful, and this can be very 

 much perfected by constant practice. 



From Vandevelde 1 ) originates a biological method for the deter- 

 mination of the toxic action of chemical compounds, which has assisted 

 in the examination of alcohols and of essential oils and some of their 

 constituents 2 ). The principle of his method is based on this, that 

 essential oils in definite quantities bring about haematolysis in defibrin- 

 ated bullock's blood floating in the proportion of 5 per cent, on 

 normal solution of common salt. The test is made in this manner, 

 that to 2,5 cc. of a 5°/ suspension of defibrinated bullock's blood in 

 normal solution of common salt (0,9 °/ NaCl), are added together 

 2,5 cc. of a mixture of normal solution of common salt and of a 

 .solution of 0,05 gm. of the oil to be examined, in 100 cc. of 50 per cent, 

 alcohol (with 0,9 °/ Na CI). According to the content of essential oil 

 or of the 5O°/ aqueous-alcoholic solution, in the last-named 2,5 cc. 

 mixture, haematolysis occurs sooner or later in the total mixture (5 cc), 

 and by eliminating the known haematolytic action of the alcohol, the 

 share of the essential oil in the haematolysis can be determined 

 numerically, and its relation as isotoxic value to the unit of weight 

 of alcohol fixed. According to this, the so-called "fruit oils", straw- 

 berry, raspberry, apple and other oils, possess the most feeble haema- 

 tolytic action, 4,78 gm. as compared with 100 gm. absolute alcohol; 

 the next in the series are peach and bitter almond oils, and benzal- 

 dehyde, with 2,33 gm., nitrobenzene and carvone with 1,10 gm., whilst 

 finally anethol, anise oil, and star-anise oil reach the highest isotoxic 

 value with 0,20 gm. Vandevelde hopes to be able to base anew 

 bio-chemical method for the determination of essential oils in alcoholic 

 beverages on his process. 



We feel bound to qualify the name "fruit oils" as indistinct, 

 inasmuch as it is not clear whether it applies to the natural oils, or 



x ) Chem. Ztg. 29 (1905), 975. 



2 ) Bull, de 1' Assoc. Beige des chimistes 17 (1903), 269; 208. Report October 

 1904, 102. 



