442 BANANA GEL 



These results are of importance in connection with the preparation 

 of banana extract and subsequent treatment of the latter. As de- 

 scribed previously, the banana extract was prepared with sodium 

 chloride solution for a number of experiments. It was found neces- 

 sary to purify the sodium chloride, since a number of samples of the 

 latter were found to contain small quantities of calcium salts. If 

 the calcium was not removed, on bringing the extract to pH 7.0 or 

 more alkaline condition for various purposes, gel formation would 

 take place very rapidly. Similarly, if sodium hydroxide which con- 

 tained calcium as impurity, was used for neutralization, gel formation 

 might occur. The results described above were therefore obtained 

 by neutralization with ammonium hydroxide. An added complica- 

 tion was found in the fact that the banana itself contains very small 

 amounts of calcium salts^ which may appear in the extracts. Long 

 standing (48 hours or more) in solutions of water extract brought to 

 pH 7.0 with ammonium hydroxide resulted in the formation of some 

 gel presumably from this cause. 



Sodium chloride extracts showed smaller gelling actions with cal- 

 cium salts than did water extracts. Thus, with two banana extracts, 

 one prepared with water, the other with sodium chloride solution, 

 treated with 0.02 mg. of calcium in the form of calcium chloride per 

 cc. under the same conditions, formed a firm gel in the first case in 

 10 minutes, and very little gel in the second. 



A comparative study of the gelling actions of certain salts was made. 

 Two series were run, one at pH 5.0, and the other at pH 7.5. To 30 cc. 

 portions of banana-water extract were added 0.5 cc, 1.0 cc, and 1.5 

 cc. of 0.044 M solutions of calcium chloride, strontium chloride, barium 

 chloride, magnesiimi chloride, and Lithium chloride. None of the 

 solutions at pH 5.0 gelled even after 48 hours. Of those at pH 7.5, 

 the calcium chloride showed the most marked gelling action, the 

 strontivun chloride showed somewhat less, and the barium chloride 

 still less. The magnesium chloride and lithium chloride gave no 

 indication of gel formation until after 7 hours, after 48 hours there 

 were soft gels formed but even less than that formed in the control, 

 water extract plus ammonium hydroxide to pH 7.5. The relations 



5 Colby, G. E., California Agric. Exp. Sta. Report, 1892-94, 275. 



