19 



moderate when compared with those used in 19 14-15, were made weekly 

 during the season October to May. In 19 14-15 the flow was not so plenti- 

 ful, although noticeably greater than in the "check" flowers. In the for- 

 mer year, the nectar was a brownish liquid with a sweet and bitter taste, 

 miscible with water, while in the latter year it was a clear, colorless 

 liquid. It had a sweet taste, and was neutral to litmus and phenolphthal- 

 ein. It charred on ignition on a platinum foil, with the odor of burnt 

 sugar, leaving a small amount of ash which was alkaline to moist litmus 

 paper and to phenolphthalein. Sodium and potassium flame tests were 

 positive, calcium doubtful. No indication of tannin was given by tests with 

 neutral ferric chloride and with potassium ferricyanide and ammonia. 

 A solution made by washing off the nectar with distiUed water reduced 

 Fehling's solution. A heavy osazone precipitate of bright yellow color 

 was thrown down upon heating it in a boiling water bath with phenyl- 

 hydrazine, acetic acid and a crystal of sodium acetate, after three minutes' 

 boiling. Ten minutes' boiling increased the amount. A much heavier 

 osazone precipitate was given after a few minutes' boiling with hydro- 

 chloric acid, and a portion of the solution inverted by the Clerget method 

 gave a heavier osazone precipitate than a similar amount before inver- 

 sion. The rotation in a i dm. tube of 1.5° Ventzke was changed to 1.18° 

 V. after the Clerget inversion. Hence, glucose and sucrose were present. 

 The precipitate formed in the hot solution was filtered off and the filtrate 

 again boiled till no further precipitate separated. On cooling the fil- 

 trate a further precipitate of sodium acetate and osazone separated. 

 This osazone possessed a roset structure characteristic of maltosazone, 

 and was soluble in the boiling solution and reprecipitated from it on 

 cooling as is maltosazone. Not enough of the precipitate could be ob- 

 tained after recrystallization for a melting-point determination.* Tests** 

 made with a guaiacol solution and neutral hydrogen peroxide gave a nega- 

 tive test with the exudation, but an equally intensive color with sections 

 of petal, ovary, leaf and stem of both normal and affected plants. Neither 

 of the reagents used alone gave a reaction. Microscopic examination of 

 the lower, plasmolyzed portions of the petals showed the cell walls intact 

 and of normal thickness. It was concluded from this that the increased 

 amount of sugar was not due to breaking down of these cell walls, but was 

 an exudation. Experiments were then undertaken to compare the sugar 

 content of the sap expressed from the stems of the plants not fertilized and 

 of those receiving applications of potassium sulfate. Evidence that a 

 larger amount of sugars was present in the sap of the latter plants was 



* Brown and Morris* used 200 g. of leaf tissue in order to obtain enough for prep- 

 aration of maltosazone. 



** Gruss'® believed gummosis might be caused by an excess of diastatic enzyme 

 and used this reagent as a means of detecting it. 



