2 TORREYA 



rosia caudida, Calopogonium and Centrosema, not only to hold soil and prevent 

 erosion but we used it as green manure as well. In Java, the practice is to 

 terrace, and every meter or two dig a sump hole on the back end of the terrace 

 about three to four feet long, two feet wide and three feet deep, for aeration 

 and to catch the eroded soil and water, then dig these holes out when filled and 

 pile the soil on the terrace behind. This is good practice where labor is plentiful 

 and cheap. 



We then began to check on the alkaloid content of the bark, beginning with 

 two-year old transplants, and set an average content for the area planted, and 

 all plants which by occular survey looked weak or diseased we eliminated, and 

 with this eliminated material the chemists of the Bureau of Science experi- 

 mented with various methods of extraction. If trees of high alkaloid content 

 were spotted, they were used for bud and seed stock and new areas were 

 planted when seed was produced, and the same procedure of elimination and 

 testing gone through until we had an average alkaloid yield of 9.6% throughout 

 the commercial plantation which today averages better than 10% and some 

 running 14 to 16%. 



The trees were pulled as soon as they showed evidence of disease or weak- 

 ness from two years upward and the average age of final pulling was around 

 six years. Two men could pull a tree by its roots when four to five years old 

 and the bark collected from the roots and stems. Root bark ran higher in con- 

 tent than trunk and stem bark. This method of eliminating by pulling and 

 testing was found the most economical method we could devise. Our goal was 

 the maximum of alkaloid content for the minimum of cost. Some good yielding 

 trees were barked and left growing for bark renewals and seed production, but 

 we found that considerable cross-breeding took place and the seed did not run 

 true to type, so to insure pure type seed we isolated areas with virgin timber as 

 separators. 



We noticed that whenever we found early flowering it was an indication 

 that the trees were not growing in their proper environment and that disease, 

 especially cankers, became prevalent. At Impalutao in Bukidnon, at an elevation 

 of 3,800 feet and with a definite dry season, trees flowered at 2^2 years, while 

 at Kaatoan, elevation 4,500 feet and no definite dry season, the trees of the 

 same stock flowered at about five years, and disease incidence was greatly 

 reduced and the alkaloid content increased. Cinchona flowers about April, May 

 or early June after the dry spells. Flower buds open in 40 to 50 days and ma- 

 ture in 25 to 35 days thereafter. The average height growth per year at Kaatoan 

 was 1.12 meters, the best 1.42 meters. 



Our first quantity production of bark took place in about 1927 and some 

 of it sold to European manufacturers. Our analyses in the Bureau of Science at 

 Manila ran between 8.5 and 9.6%, but the Kina Bureau in Holland, which does 

 all the analyses for manufacturers, gave only 7.6%. There was some loss in 



