8 CASSAV-A. 



decomposition, when the material is allowed to rest in the presence of water. As 

 yet, however, we do not appear to be in possession of the exact amount of poison i he 

 roots originally contain, for in the laboratory processes the method has been to extract 

 with water, causing a c ' cyanogenetic " process to be set up, which actually appears 

 to generate the acid. 



In the work of the Trinidad station (Carmody) it is noted that the 

 analyst in experimenting upon both grated and sliced samples obtained 

 after two hours' digestion only 0.003 per cent of hydrocyanic acid, and 

 that upon continued extraction he gets a total of 0.012 per cent. He 

 calls attention to the fact that the one extraction of but two hours made 

 by Francis gives higher results than his nine extractions, and suggests 

 that this may be due to the exceptionally dry season. From further 

 observations on Carmody's work it appears that on some samples he 

 obtained practically the total hydrocyanic acid on the first extraction. 



From these comparisons it is apparent that some of the results 

 obtained at these two stations may be correct and some may be erratic, 

 depending upon whether the production of hydrocyanic acid had been 

 completed at the time the estimations were made. From the reports 

 of the Trinidad and Jamaica stations it is not possible to ascertain 

 whether the two stations at an} T time worked on pedigreed varieties. 

 This feature, in addition to the possible source of error just mentioned, 

 renders it difficult to draw any conclusions from the work done. 



EXPERIMENTAL WORK ON CASSAVA CONDUCTED BY THE 

 DEPARTMENT OF AGRICULTURE. 



INTRODUCTION. 



The great value of cassava as a feeding stuff, could the possibility 

 of disastrous results from poisoning be entirely eliminated, seems to 

 justify an extended research on this point. Cassava may be grown in 

 any of the Gulf States, and in the introduction of new varieties some- 

 thing definite should be known of their habits and characteristics 

 before they are brought into general cultivation. 



It can not be assumed that the observations recorded in this report 

 will be verified another season, even at the same places. Only a 

 series of similar experiments continued for a number of years can 

 conclusively establish the points under consideration. Some of the 

 varieties studied undoubtedly have properties which make them supe- 

 rior to others, such as yield, form, composition, and resistance to 

 disease. Some preliminary experiments made it evident that certain 

 varieties yield a better grade of meal than others, some being so woody 

 that they can not be masticated by animals. 



It is also desirable to ascertain the actual yields of the varieties 

 under field culture, and to obtain definite data on the production of 

 meal and its value for starch making and as a feeding stuff. While 

 extensive experiments would be necessary to definitely establish con- 



