202 COCOA 



From Table G (p. 209) it will be seen that 58 per cent, 

 of the trees have given less than 2 Ib. of dry cacao per 

 tree per annum. These trees are over 50 years old, and 

 the cacao from this estate fetches a fancy price in the 

 market. 



It will also be seen from Table A that some trees give 

 a yield of 267 to 290 pods per tree. This is equivalent 

 to- 133 and 145 bags per 1,000 trees respectively, which 

 is a very high yield. An appreciable proportion of trees 

 gave over 100 pods per tree, which is equivalent to over 

 50 bags per 1,000 trees; and there is reason to hope that 

 with a careful selection of seed a minimum yield of 

 100 pods per tree is not an impossible ambition. Trinidad 

 planters are impressed by these results and with the 

 advantages of careful seed selection, and many of them 

 now obtain seeds from these heavy-bearing trees on the 

 Government estate, and all plants sold by the Depart- 

 ment are raised from the seeds of these trees. 



In Trinidad, 20 bags per 1,000 trees is a good yield 

 under present conditions. 



Owing to this great variation in the yield of individual 

 cacao trees, it will be seen that it is not possible to make 

 comparative experiments in the way usually adopted for 

 other kinds of agricultural crops by taking as a basis for 

 comparison equal areas or an equal number of trees. 

 One acre of sugar-cane, or one stool of canes, is 

 approximately equal in yield to another acre or stool if 

 the variety of cane under comparison is the same; but 

 our investigations have shown clearly that one cacao 

 tree may equal in productive value as many as 12 or 

 14 cacao trees of the same variety grown under similar 

 conditions. 



Tables B and D (p. 210) clearly show the great 

 variations in yield that occur in several adjacent plots 

 in the same field, and it also indicates that, however 

 large the number of trees in a plot may be, the natural 

 yield from any plot may differ considerably from that of 

 another plot of equal area. The same plots and the 

 same trees do not differ from year to year to anything 

 approaching the same amount. 



Owing to these marked variations, the best course to 



