fl. E. ANNETT. 387 



ed in which the composition of the juice was approximately the 

 same. After several experiments it was decided that the variations 

 in the quality of the juice between any two groups which might be 

 so obtained was so large as to make it difficult to obtain a numerical 

 value of the effect of formalin. 



This effect was very obvious however. Most of the experiments 

 were done by lightly washing the cut surface in the afternoon with 

 a piece of cotton wool dipped in 10% formalin solution (commercial 

 formalin diluted 4 times). The washing was done by the gachi him- 

 self in our presence. In almost every case the juice obtained from 

 the formalined surface was quite clear and always much clearer 

 than from untreated trees. Moreover in the morning the untreated 

 surfaces always have a certain amount of yeast growth, whereas the 

 surfaces treated with formalin were practically free from yeast. 

 In order to make certain that the effect was due to the formalin 

 and not to simply the cleaning of the surface, experiments were made 

 in which the surface was washed in a similar way with water. This 

 also lessened the yeast growth but not to such an extent as the for- 

 malin did, neither was the juice of such good quality. Perhaps the 

 best proof of the efficacy of formalin was the testimony of the gachi 

 himself. He was so pleased with its effect that he wished us to let 

 him treat the whole of his garden in a similar way. 



It has been shewn (page 340) that much of the inversion of the 

 sucrose takes place after the juice has left the tree and while it is 

 standing in the pot overnight. Hence it is reasonable to suppose 

 that it would be worth while putting formalin in each pot overnight. 

 In the cases where we tried this, very good juices were obtained. 

 In one case in which one c.c. of 40% formalin was placed in the pot 

 overnight and dokat juice collected, a juice was obtained containing 

 in 100 c.c. 13 gms. of sucrose and 0'35 gm. of reducing sugar. This 

 is a very good dokat juice indeed. 



It would be worth while carrying out a more extended series 

 of experiments on the effect of formalin. We should feel inclined 

 to recommend washing the cut surface with formalin once a week 



