56 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



Nitrogen, II {Kjeldahl method).— 0.2G9o gram substance, dried at 110° C, gave ammonia=9.68 

 cc. HCI (I cc. HC1=0.00854 gram N) = lG.39per cent. 



The filtrate from 4 B was evaporated to dryness and yielded one gram of substance containing 

 7.5 per cent of nitrogen. 



6 evidently contained a large amount of sugar. It was first treated repeatedly with water, in 

 which it formed a pasty lump. This was dehydrated with absolute alcohol, and when dried over 

 sulphuric acid weighed 7.63 grams and contained 14.61 per cent of nitrogen="6 A." 



6 A was then treated with 100 cc. of water, with which it formed a stringy lump, which, after 

 having been thoroughly kneaded under water, assumed a semiliquid condition and could not be 

 separated from the solution either by boiling or freezing. The solution, after concentrating on the 

 water-bath to about 30 cc, was poured into 300 cc. of absolute alcohol. This threw down a curdy 

 white precipitate, which, after treating with absolute alcohol and drying' over sulphuric acid, 

 weighed 6 grams and contained 15.67 per cent of nitrogeu when dried at 110° C.="6 Al." 



6 Al was treated with 500 cc. of alcohol of 0.9 sp. gr. But little dissolved and the solution 

 was too slimy to filter. The whole was then concentrated to one-third and cooled, but no separation 

 took place which would allow of filtering or decanting the solution. The solution was, therefore, 

 diluted with water and boiled. After cooling, a curdy precipitate settled out, leaving a turbid 

 liquid. This liquid was decanted and poured into a large amount of absolute alcohol. Oji standing, 

 a precipitate separated which, when treated with absolute alcohol and di'ied over sulphuric acid, 

 weighed 0.155 gram and contained, when dried at 110° C, 16.40 per cent of nitrogen = "6 A2." 



There was thus obtained of the proteid soluble in alcohol a large number of fractional precipi- 

 tations with difierent nitrogen content. It will be noticed, however, that five of these preparations 

 have from 16.27 to 16.56 per cent of nitrogeu. No preparations were obtained having a higher 

 percentage of nitrogen, and none of the other preparations approximate to a constant nitrogen 

 content. 



4 had 16.27 per cent of nitrogen. When this was separated into two parts, one soluble in di- 

 lute alcohol — 4 B — and the other insoluble — 4 A — the two parts were found to contain the same pro- 

 poi'tion of nitrogen, and exactly agreed in content of carbon and hydrogen. This is what might be 

 ex]iected of a pure proteid, for some of these bodies become insoluble without change of composi- 

 tion, so far as can be detected by our methods of analysis. 1 C, without correction for ash, has 

 nearly the same composition as 4 A and 4B; a correction would tend to bring it into closer agree- 

 ment. The amount of substance, however, was insufficient for an ash determination. 



As all the preparations analyzed thus far were the insoluble modification of the proteid, a new 

 extract was prepared. Five pounds of freshly ground oats were boiled up with 1 liters of alcohol 

 of 0.915 sp. gr. The extract was i^ressed out and added to another 5 i^ounds of oats and again 

 heated to boiling. After filtering, a perfectly clear, deep, ruby-red solution was obtained. This 

 was concentrated by distillation on the water-bath to about one-third its original volume. On 

 cooling, a large amount of substance separated. This was thrown ui)on a filter, and after twenty- 

 four hours about seven-eiglits of the solution had passed through. The filter, with its contents, 

 was then boiled in 10 liters of 60 per cent alcohol, in a flask fitted with a reflux condenser, for four 

 hours. The substance entirely dissolved to a clear, deep-red solution. The solution was then 

 concentrated by distillation to about one-half, and cooled rapidly to 35° C, when it filtered 

 readily. The i)recipitate formed a tough, jeUy-like mass which did not adhere to the pafter. It 

 was boiled, as before, for several hours, with 10 liters of 60 per cent alcohol, until completely dis- 

 solved. The perfectly clear solution was concentrated to one-half its volume, during which opera- 

 tion a large part of the substance deposited on the interior of the flask as a thick leather-like 

 coating. This coating readily separated from the glass in large pieces of a dark-brown color. On 

 cooling the hquid, a finely-divided yellowish precipitate appeared which did not settle out on 

 standing. The muddy liquid was ])oured off and the leathery deposit was minced and treated 

 with absolute alcohol, in order, if possible, to dehydrate the substance and render it pulverizable 

 for analysis, but as it remained forty eight hours without change, the attempt was abandoned. 

 The substance suspended in the muddy liquid was now caused to separate by adding alcohol. 

 After filtering, washing first with strong, then with absolute alcohol, the precipitate was united 

 to the portion which had been digested with absolute alcohol, and the whole then boiled in 2 liters 



