MIAMI, FLA., 1904. 23 



The results on the Lemon City cassava show that the content of 

 hydrocyanic acid in some varieties remained practically constant, 

 while others show ti marked increase. Until March 17 it was thought 

 an average increase was taking place, although there were many 

 fluctuations. On March 21-23, a second series of analyses was made 

 on the Miami tubers, and at intervals thereafter, the same being 

 interspersed among the data for Lemon City. 



From a general study of Table VI, it will be seen that the content 

 of hydrocyanic acid appears to reach a maximum about the middle of 

 March, and at this time the Miami results closely duplicate those 

 obtained for the corresponding numbers at Lemon City. Toward the 

 close of March the content of the acid seemed to decrease, though 

 specific cases can be found which show a definite increase, or a con- 

 stant amount of the poison present. In studying these fluctuations, it 

 must be borne in mind that no plant was analyzed a second time, and 

 it is possible that the change is not so great as appears between two 

 given periods, inasmuch as an individual plant which shows an increase 

 over another one at a successive period might have shown the same 

 difference had the two plants been examined on the same day. In 

 other words, the variation may be in part individual and not due to 

 the periodic change. 



The one known poison variety, No. 9699, was found to contain 

 0.018 per cent of prussic acid on the first analysis and only 0.004 

 when examined a month later. As there were only two of these 

 plants, the series could not be further extended. A common Florida 

 variety, nonpedigreed, from a field near Lemon Cit} r , was studied 

 from March 16, practically no fluctuation in hydrocyanic acid being 

 observed during the ten days of observation, the content of acid being 

 low at all times. 



The results in Table VI suggest that the content of hydrocyanic 

 acid may undergo a variety of changes during the growth of the plant, 

 and may even disappear entirely at times and again be present. If the 

 maximum and minimum results on the same varieties grown at Miami 

 and Lemon City be compared with the results on the same varieties 

 grown at Biloxi (see Tables V and VI), it will be seen that at some 

 period each variety grown at Miami and Lemon City was practically 

 free from hydrocyanic acid. Further, it will be noted that during a 

 second selected period more than half of each variety reached the 

 maximum limit for sweet varieties, and some numbers which were 

 distinctively sweet at Biloxi were semibitter when grown at the 

 Florida stations, while others that were semibitter in Florida were 

 dangerously so in Biloxi. None of the Florida-grown varieties 

 becomes more than semibitter, while in Biloxi two- thirds of the 

 samples are semibitter and about one -fourth are considerably more 

 than half bitter. 



