686 



acidulated with acetic acid, and acetate of phenyl hydrazin added. 

 The precipitated furfnrol-hydrazon is collected in a Soxhlet asbestos 

 filtering tube and dried at 50 to 60° 0. in a partial vacuum, rarefied air, 

 freed from moisture, being- drawn over the precipitate by means of a 

 suction pump. The weighed furfnrol-hydrazon is calculated to furfurol, 

 using the factor O.olG and making a constant correction of 0.0252 gram 

 for the furfurol remaining un precipitated in the solution. In a large 

 number of determinations the average yield of furfurol from arabinose 

 has been 48.72 per cent, and from xylose 56.25 per cent. In calcula- 

 ting to pentaglucoses in general the furfurol is multiplied by 52.2. 



The difference between parallel determinations in the same material 

 is said to be not over 0.0 per cent of furfurol. The authors believe the 

 gravimetric method to be more accurate, safer, and no more difficult of 

 execution than that by titration. 



Investigations by Tollens, E. Schulze and Steiger,* and Stone t hi've 

 shown the occurrence of the pentaglucoses in vegetable materials to be 

 widespread, and often in considerable quantities. 



The nature of the root tubercles of leguminous plants. — A review by 

 H. W. Conn. — For a long time it has been known that when certain 

 l)lants of the family Leguminosse are grown under normal conditions 

 there are formed on their roots small swellings known as root tubercles. 

 Their origin and significance has not been understood, but within the 

 last 3 years it has been quite definitely shown that they are intimately 

 associated with the power jjossessed by these plants of accumulating 

 nitrogen from the atmosphere. The great interest of this fact has led 

 to a further study of the tubercles, and through the work of Beyerinck, 

 Prazmowski, Frank, Laurent, and others we have obtained many facts 

 in regard to them. The object of the present paper is to review recent 

 investigations in regard to their origin and structure. The discussion 

 of their functions, especially in connection with the acquisition of atmos- 

 pheric nitrogen, is reserved for a later issue. 



Earlier obfiervations. — At first the tubercles were supposed to be insect 

 galls. This idea was soon abandoned, and they were then regarded as 

 buds of incomi)lete plants, or as rudimentary roots. In 1860 Woronin | 

 found in them numerous minute bodies which bore some resemblance 

 to bacteria. They were sometimes rod-shaped, but often slightly fork<'d 

 into T or Y-shaped bodies. On account of this irregularity in shape 

 the discoverer was unable to say whether they were true bacteria or 

 not. He therefore called them bacteroids, and regarded them as the cause 

 of the tubercles. In 1874 Erickson § found that in the early stages of the 

 development of the tubercle it was filled with long, branching threads 

 resembling the mycelium of fungi, and to these hyphie he attributed the 



» Ber. d. d. cbem. Gesell., 23, p. 3110. 



\Ibid., 33, p. 3791. 



tM6m. Acad. imp. des Sciences de St.-P6tersboiirg, t. x, 18G6, No. 6. 



§ Studier olVer Legiimiuosernas Rotkuolar, Luud. 



