THE ASSIMILATION OF NITROGEN IN AUTOTROPHIC PLANTS 139 



From a physical standpoint the colloidal nature of proteids stands out pre- 

 eminent. It is, doubtless, largely owing to the size of their molecules that 

 proteids are unable to diffuse through parchment or animal membranes. 

 Nevertheless they may be regarded as forming genuine solutions which have 

 the peculiarity of being not very stable. Proteids coagulate on very slight 

 provocation, and very often this coagulation is accompanied by considerable 

 alterations in character. This coagulation is permanent and renewed solution 

 is impossible without fundamental chemical change. Such coagulations are 

 induced by alcohol, by boiling water, by strong mineral acids, as well as by certain 

 so-called alkaloid reagents (phosphotungstic acid, tannic acid, &c.). On the 

 other hand, proteids are transformed by salting out (especially by ammo- 

 nium sulphate), into a sohd and often crystalline condition without being 

 chemically altered. This salted out proteid remains soluble. 



The reagents mentioned above may be used as tests for proteids, but 

 certain colour reactions may also be employed, of which the most important 

 are the following : — 



1. They give a blue-violet to red colour with caustic soda and a few drops 

 of weak copper sulphate solution {biuret test). 



2. Heating with concentrated nitric acid gives a yellow colour {xantho- 

 proteic reaction). 



3. Boiling with a solution of mercuric nitrate containing a trace of nitrous 

 acid gives a rose to dark red colour {Milton's reaction). 



4. Treatment with an alcoholic solution of a-napthol and concentrated 

 sulphuric acid gives a violet colour {Molisch's reaction). 



5. By boiling with caustic soda and a salt of lead a black precipitate is 

 produced {lead- sulphide reaction). 



Apart from the biuret test, the reactions described are effects produced, 

 not by the proteid molecule as a whole, but by constituent groupings in 

 it ; Millon's reagent, for example, acts on a different group in the proteid 

 molecule than does the lead-sulphide test, and that again on a group not acted 

 on by Molisch's reagent. One is thus able to differentiate in the proteid 

 molecule a number of constituent groups, with which a study of the 

 decomposition products of proteid has made us familiar. Hydrolytic de- 

 composition more particularly has furnished us with especially valuable 

 data, because obviously that method entails no very profound changes on the 

 products of decomposition. Hydrolytic decomposition may be effected by 

 boiling mineral acids as well as by enzymes (proteases ; compare Lectures 

 XII and XIII) ; the products are similar in each case and we shall confine 

 ourselves at present chiefly to the action of enzymes. By the action of 

 proteolytic enzymes proteid is broken down, in the first instance, into smaller 

 molecules, which still retain many of the characters of proteids ; there arise 

 first the albumoses, which are no longer coagulable but may be precipitated 

 by salting out. From these arise the peptones, which cannot be salted out, but 

 which still respond to the biuret test. All subsequent decomposition products 

 fail to show any biuret reaction, and thus are no longer proteid. Albumoses 

 and peptones may still be considered as proteids, although many peptones 

 contain no sulphur. Among the products of further decomposition we have 

 next to recognize a sulphur-containing group. In what form this arises through 

 the action of enzymes is not as yet fully understood ; cystin (CgHjjNjSjO J rarely 

 occurs in plants ; sulphates on the other hand, are apparently produced directly. 

 [Intermediate substances between peptones and amino-acids have been dis- 

 covered, coupled amino-acids or polypeptides, many of which have been 

 synthetically prepared by E. Fischer (Czapek, Biochemie, II, 45).] Under 

 the head of sulphur-free proteid groups we may recognize the following 

 (Hofmeister, 1902) : — 



