March, 1909.] 



26l 



Edible ProductSi 



per acre, and the average near the Lake 

 Mary factory, where the figures should 

 be reliable, has been given as 9 and 10 

 tons ; but the average crop in the United 

 States is not believed to be more than 

 5 tons. In Pondicherry 5 metric tons is 

 the maximum crop expected from a 

 hectare of well-manured plants irrigated 

 five or six times a month during the dry 

 season- In Ceylon and Java the yield is 

 expected to be above 25 tons per hectare. 

 A test crop ten months old at Hue was 

 14-44 tons per hectare. There are no 

 data as to the actual yield of any con- 

 siderable acres in the Philippines, but 

 from the unanimous opinion of growers 

 and from our own observations and 

 weighings of apparently representative 

 plants, it is our opinion that 25 tons per 

 hectare is a very low estimate of the 

 yield fairly to be expected from a crop 

 a year 01 upward old. 



Many analyses showing the starch 

 content of manioc have been published, 

 but it is not worth while to copy them 

 here. Data on the best Jamaica varie- 

 ties have already been given. The 

 matter of business interest is not the 

 absolute starch content of the roots, but 

 the amount which can be obtained by 

 practical manufacturing methods. This 

 of course depends on how much is in the 

 roots, but to a greater exteut still upon 

 the method used in its extraction, and 

 the thoroughness with which it is worth 

 while to extract must be settled locally 

 according to the value of the starch and 

 the cost of producing the roots. For 

 this reason the same methods are by no 

 means to be recommended for all parts 

 of the world. The highest starch con- 

 tent I have seen published is 315-5 per 

 cent, in a Colombian variety grown in 

 Jamaica in 1903 and called " Governor 

 Hemming " ; four years later the same 

 vaiiety contained only 30*17 per cent. 

 By commercially feasible treatment we 

 obtained here more than 31 per cent, of 

 air-dry starch, in one test. 



Starch Manufacture. 



The first step in starch manufacture 

 is always and everywhere the same, the 

 roots are washed clean. If any dirt 

 finds its way into the mill, it will stay 

 with the starch through the whole 

 process of manufacture, and be in it at 

 the end. A very little dirt destroys the 

 perfect whiteness of the starch, and only 

 perfectly white starch can be sold at a 

 good price. The washing can be done 

 by hand or in a mechanical washer. 

 Mechanical washing of most roots and 

 in most places is much the cheaper, but 

 the great and irregular size, ot manioc 

 roots offers some difficulty. When the 



roots have been thoroughly washed, it 

 is customary to peel them removing the 

 bark and cortex which contain no starch 

 and at the same time to get rid of any 

 last particles of dirt. If the roots are 

 kept moist until decay begins the cortex 

 will slip off readily in the hands, but 

 this course is not to be recommended. 

 If the roots are perfectly clean the 

 removal of the bark and cortex is un- 

 necessary, for they contain nothing 

 which discolours or mixes with the 

 starch. 



The starch in the roots is contained in 

 the cells of the pith. The starch grains 

 are very minute, decidedly smaller than 

 those of potato starch, and the cells 

 containing them are also smaller. For 

 this reason and because the roots are 

 more woody than the tubers of potato, 

 it is mot e difficult. to extract the starch 

 completely. The walls must clearly be 

 broken to let the starch escape ; the 

 starch in any unbroken cell is lost with 

 the fibrous part of the roots. The walls 

 are broken by decay or by scraping, 

 rasping, or crushing the roots. 



The practice of letting the roots decay 

 has been in use among the poor Chinese 

 of the Malay States since about 1891 

 in making a low grade of tapioca, 

 but not in the manufacture of saleable 

 starch, A Dutch writer, de Kruijff 

 (Teysmania, 1900, No. 8), has recently 

 stated that by letting the walls decay it 

 is possible to secure the starch very 

 completely and of as good quality as is 

 obtained by other processes. Anyone 

 adopting this method will wisely try 

 it at first on a very small scale. 



The old Chinese method in the Straits 

 Settlements, when that district first 

 took a prominent place in the manu- 

 facture of starch and tapioca, was to 

 wash the roots, peel them, wash them 

 again, grind or crush them between 

 rollers, strain out the fibrous part of the 

 pulp with a sieve which permitted the 

 starch to pass through with the water, 

 let the starch settle, draw off the water, 

 and wash until clean and dry. This was 

 all done by hand. In the factories of 

 Europeans steam power is in use and the 

 roots are pulped by scraping. 



Elaborate machinery for starch man 

 facture has naturally been developed 

 special adaptation to corn aud potatoes. 

 The first attempts at manioc starch 

 mills on the same scale were made in 

 Florida about a decade ago. The Lake 

 Mary factory put in potato-starch 

 machinery, and with it was unable to 

 get more than 20 per cent, of the starch 

 from roots containing fully 27 per cent. 

 They have since made improvements— 



