484 Y. Sumiki and A. Kaiuarada 



and evaporated to a small volume under reduced pressure, and a 

 dense greenish liquid (pH 5.4) was obtained. It was adjusted to pH 8.0 

 with sodium bicarbonate, extracted five times with ethyl acetate 

 (total volume 1.5 1.) to remove the nonacidic substances. The aqueous 

 layer was then acidified to pH 3.0 with dilute sulfuric acid, and again 

 extracted five times with ethyl acetate (total volume 1.5 1.). Upon 

 evaporation of the solvent, a dark greenish syrup was obtained. This 

 did not show any physiological activity in the gibberellin bioassay. 



For the purification of gibberellin-like factor from this crude acidic 

 substance, the countercurrent distribution method was applied [14 

 plates, ethyl acetate: \M phosphate buffer (pH 5.2) ]. After develop- 

 ment, the buffer layer of each plate was acidified to pH 2.0 with sul- 

 furic acid, extracted three times with an equal volume of ethyl acetate 

 added to the upper layer, and dried with anhydrous sodium sulfate. 

 The results of this method and subsequent bioassay indicated that 

 the peak was near plate 8. 



After the fractions of plates 6 to 10 were combined and counter- 

 current distribution was repeated, the small portion of plate 8 was 

 spotted on Whatman No. 1 filter paper and developed by ascending 

 method with the solvent system of isopropanol-28 per cent ammonia- 

 water (10:1:1). At the same time, samples of gibberellins A^ and A3 

 were subjected to the same procedure. When the solvent front as- 

 cended 24 cm., the paper was dried, divided into twelve sections, and 

 assayed. The physiologically active zone from the water sprouts co- 

 incided closely with that of gibberellin A^ or A3; the latter were 

 detected also by spraying bromocresol green indicator. 



Another collection (7.2 kg.) of water sprouts, harvested in Oc- 

 tober, 1958, was treated with the same procedure, i.e., aqueous ace- 

 tone extraction, ethyl acetate extracion, and two countercurrent 

 distributions. The active fraction was poured onto a column of cellu- 

 lose powder (3 X 32 cm., Whatman No. 1), and eluted with the sol- 

 vent system of isopropanol-28 per cent ammonia-water (10:1:1) at the 

 flow rate of 1 drop per 2 sec. Ten ml. fractions of eluates Avere col- 

 lected and the fractions 4 to 7 were combined and yielded 98 mg. of a 

 colorless oily substance. The material was then spotted on sheets of 

 Whatman No. 1 filter paper (four sheets, 8 X 40 cm.) and developed 

 with the solvent system described above. The appropriate areas, de- 

 tected by guide spots of authentic specimen on both sides of paper 

 sheets, were cut out and eluted with methanol (5 ml. per area). 



Then, in order to change the above ammonium salt solution to 

 free acid, the eluate was passed through a short column (0.8 X 5 cm.) 

 of Amberlite IR 120 (H^ form). On evaporating the solvent under 

 reduced pressure, 12 mg. of the colorless amorphous powder was ob- 



