66 JOURNAL OF THE 



In experiment I the loss of the flower alone (geranium) is shown 

 to be much more rapid than that of the leaf. In experiment II the 

 flower and leaf (zinnia) loses a larger per centage in the same time 

 than the leaf, but the difference is slight. Besides, the flowers con- 

 tain about five or six per cent, more water, and there is then more 

 water to be transpired in the case of flowers and leaves, hence a 

 larger loss is to be expected. 



Hellriegel, as cited in Chemi^ al News, No. 1336, says that when a 

 plant begins to wilt it has already lost nearly half its water. This 

 cannot possibly be so. I have observed the wilting of leaves after 

 the loss of 6 per cent., and even less, from a total of 75 per cent, 

 water. So, too, where some leaves of a plant are more exposed than 

 others to the sun's action, they will wilt whilst the others are quite 

 fresh looking. The plant cannot have lost much water. The trans- 

 piration withdraws water faster than it can be supplied to the leaf 

 through its ducts. 



University of North Carolina. 



EXPERIMENTS TO DETERMINE THE EFFECT 

 OF A SOLUTION OF COMMON SALT, (NaCl), 

 AS A WASH IN DETERMINATIONS OF 

 "CITRATE INSOLUBLE" PHOSPHO- 

 RIC ACID, TO REPLACE PURE 

 WATER WASH. 



F. B. DANCY. 



The Association of Agricultural Chemists in convention assembled 

 at Atlanta, Ga., in May, 1884, took tardy notice of a fact that had 

 been doubtless for some time evident to most practical chemists, 

 and this was, that the use of a solution of half citrate of ammonia 

 solution (Sp. Gr. 1.09) and half water as a wash in determinations 

 of "insoluble" phosphoric acid being equivalent to further treat- 

 ment, was no longer admissible. The method then adopted by them 

 prescribed pure water, at ordinary temperature, as the proper wash 

 to be substituted for the citrate wash. This was eminently proper 

 theoretically, and in many cases it proved satisfactory in practice. 

 But at the same time when the method comes to be tested practi- 

 cally, it is found that in very many cases, if not in a majority, the 



