136 EXPERIMENT STATION RECORD. 



In a cooperative test of cotton variation at Waipahn the following lint yields 

 per acre were secured: Chinese upland 180.83, Egyptian 115, Sea Island 54.66, 

 and Caravonica 25.37 lbs. The location appeared unfavorable, as only about 40 

 per cent of a stand was reported and the bollworm took about 30 per cent of 

 that. The damage to the upland cotton -was less than that suffered by the 

 other varieties. In another test transplanted Caravonica seedlings gave better 

 results than those not transi)lanted. 



At the station Caravonica cotton yielded at the rate of 165.9 lbs. of lint per 

 acre and produced fibers ranging from 1t% to 11 in. in length. A brief progress 

 report is given of date of pruning and pinching back tests, together with a state- 

 ment of the theory on which these tests were conducted. The cotton was 

 picked as it matured from the branches that had been pruned. Strength tests 

 of the fibers showed that they had decreased from 8.99 gm, to 6.24 gm. in 

 breaking strength. The more immature the fiber at pruning time the lower 

 its breaking strength when tested. Individual plants yielded from 0.33 to 2.73 

 lbs. of lint per plant. 



The use of a lantern placed above a pan of water and kerosene in the field 

 at night and the regular cutting off and burning of infested bolls reduced the 

 percentage of bollworm infestation from 68 per cent at the first picking to 50 

 per cent at the second and 26 per cent at the third picking. 



Sea Island cotton yielded at the rate of 395 lbs. of lint per acre and produced 

 fibers ranging from Ig to lil in. in length in spite of a 20 per cent loss due to 

 insects. Poisoned bran placed in small circles about the hills prevented damage 

 from cutworms. 



On December 31 the plants were prunetl back to stumps, some high and some 

 low. The growth invariably appeared near the ground, indicating that Sea 

 Island cotton grown as a perennial should be pruned low. 



Immature fiber picked from plants that had been pulled w^as found to con 

 tain a large amount of nep, Immature fiber, and excessive waste. The yarn 

 was too weak for thread. 



A number of crojis were planted for the purpose of determining their profit- 

 ableness in Hawaii. Peanuts did well and were in one case planted between 

 cotton rows. Broom corn yieldc<l at the rate of from n.l.l to 501.6 lbs. of brush 

 per acre. Other crops tested were corn, saccharin and nonsaccharin sorghums, 

 and a new legume called guar. 



[Botation and fertilizer tests], H. J. Eustace and V. M. Shoesmith 

 {llicJiignn Sta. Rpt. 1911, pp. 183, 181/; 212-221). — This is mainly a report of 

 tests begun in 1891. 



During the first 5 years all plats were uniformly cropped to determine their 

 relative fertility. Tables present data from which the author concludes that 

 aside from the matter of profit a wheat-clover rotation is somewhat better 

 in effect on soil fertility than a wheat-clover-corn or a wheat-clover-potato 

 rotation, but the differences are not marked. Plats which had been continu- 

 ously planted to wheat and corn were inferior in producing power to those 

 on which clover had been grown in rotation. Beans grown each season with 

 rye seeded in the fall and turned under in the sirring were followed by 21 

 per cent greater yields than a continuous cultivation of corn and 29 per cent 

 greater than was secured from the plat planted continuously to wheat. Con- 

 tinued cropping to orchard grass was followed by a 21 per cent greater yield 

 than was a wheat-clover rotation. This is attributed to the heavy root system 

 and the plant food which it stored. 



In one series plats fallowed for 10 years produced more than any others 

 of the series except those seeded to orchard grass, but when the averages of 

 2 series were considered fallow was excelled by a wheat-clover rotation. 



