l8o INTERNAL SECRETION 



process by which adrenalin is formed, is very scanty. As far as 

 its preceding stages are concerned, Friedmann supposed that 

 adrenalin might be derived from an oxyphenylserin or an 

 oxyphenylmethylserin, which substances might possibly have 

 been previously formed in certain albuminoids. According to 

 Halle and Frankel, tyrosin and phenylalamin may possibly be 

 waste products. The conversion of tyrosin into adrenalin is 

 supposed to be a combination of four chemical processes (oxida- 

 tion, evolution of carBonic acid, methylization of nitrogen, and a 

 second process of oxidation of the lateral chain in the /3 position), 

 in the course of which para-oxiphenylethylamin is formed as an 

 intermediary product. This supposition is supported by the 

 results of experiments showing that the addition of tyrosin to an 

 emulsion of ox's or pig's suprarenals, is followed after six days 

 in an incubator by an increase in the adrenalin contents of from 

 14 to 33 per cent. 



Neuberg found that the extract from a melanotic tumour con- 

 verted adrenalin in a short time into a dark brown product ; and 

 he also discovered that a melanin ferment obtained from the ink- 

 bags of cuttle fish (sepia) gives a similar melanin reaction with 

 para-oxy-phenylethylamin and tryptophan as with adrenalin. By 

 means of the tyrosinase from the fungus Prussula delica, Abder- 

 halden and Guggenheim obtained the same colour-reactions with 

 tryptophan and the derivatives of tyrosin (d-alanyl-1-tyrosin and 

 1-leucyl-l-tyrosin) as with 1- and d-suprarenin. 



From these results it appears highly probable that the 

 products of the decomposition of albumin, and especially 

 tyrosin, are the matrix of adrenalin. 



The most important chemical test of the presence of adrenalin 

 in liquids and tissue secretions is the characteristic colour reaction, 

 namely, green with ferric chloride and pink with iodine water. 

 The same reactions have also been employed as quantitative tests 

 in colorimetric experiments ; ferric chloride by Battelli, and the 

 iodine reaction by Abelous, Soulie and Toujan. But the results 

 obtained by these methods cannot be regarded as by any means 

 exact, for they are dependent upon the operation of certain 

 extraneous conditions. The substance must be employed in a 

 state of chemical purity and with an almost neutral reaction, for the 

 presence of free acids and other bodies prevents the ferric chloride 

 reaction, which, in any case, is common to all derivatives of 

 orthodioxybenzol, and is not specific to adrenalin. Iodine stain- 

 ing is prejudiced by the fact that adrenalin solutions which are 

 exposed to the air and light take on a red colouring. 



In the last two years a number of chemical tests for adrenalin 

 have been published. 



The method described by G. Comessatti is based upon the 

 fact, first observed by Muhlmann, that the addition of sublimate 

 to suprarenal extract produces a red colour. A few drops of a 



