31 



and between 0.043 and 0.114% of chlorine, so that we cannot say where 

 the exact limit lies, above which the cane will be injured by chlorides and 

 other salts. But it is somewhere between the figures just given. It is 

 possible that all the figures for chlorine, etc., were a little higher than 

 the average for a growing season, as the samples were taken at the end 

 of a dry spell. Heavy rains would have reduced the salt to some extent, 

 at least for a time. 



While it is evident that the general condition of the cane is greatly 

 influenced by the salt content of the soil, there is no direct relation 

 between this latter and the actual amount of disease present. In places 

 where the salt content is very low T , considerable root disease may be 

 present, and in others, where the salt is high, root disease may not be 

 found. This fact is strikingly shown at location No. 1, where the 

 conditions of moisture etc. seem to be excellent for the development of the 

 root fungus. But it is more than probable that the salt content of that 

 soil is too high for the proper development of the fungus. 



From the first series of analyses reported on in the preceding pages 

 we may conclude that in the ordinary mineral soils of the region investi- 

 gated the danger point for chlorine is between 0.043 and 0.114% (0.142 

 and 0.198% of total acid radicles), and that it is probably quite near that 

 fixed for Hawaiian conditions by Maxwell at 0.09% (0.15% of sodium 

 chloride), and by Eckart at 0.06^ chlorine (0.1% of sodium chloride). 



SECOND SERIES OF ANALYSES. 



Having thus made a general survey of the conditions obtaining on 

 the estates controlled by the Plazuela Sugar Company, so far as the 

 salt content of the soils is concerned, and having established the limit 

 of chlorine that the cane will stand in the ordinary soil types, attention 

 was now centered on the marsh itself. The general working plan was the 

 same as in the first investigation: a number of locations were carefully 

 selected, soil samples taken and the condition of the cane noted at the 

 same time. The area studied in this second investigation lies along a 

 branch of the plantation railroad which runs across the Caho de Tiburo- 

 nes at a point between kilometers 12 and 13 of the main line of the same 

 railroad, from north to south, crossing both drainage canals. This part 

 of the estate had only recently been planted to cane for the first time 

 (Spring 1911 ; samples taken in June 1911). Samples of soil were taken 

 at five different points along the branch line, collecting separately ths 

 samples of the surface foot and that of the underlying second foot. At the 

 same time notes were taken on the condition of the cane and on the height 

 of the water level at each place, during high tide. 



All of the soil in this entire field is the very moist black vegetable 

 mould already fully described. The' exact depth of this material was not 

 ascertained, but the underlying stratum most probably consists of the 



