Scientific Agriculture. 



435 



[November, 1909, 



PLANTING EXPERIMENTS AT THE 

 AGRICULTURAL EXPERIMENTAL 

 STATION, ZIMBITI (MOZAMBIQUE). 



By W. H. Johnson, p.l.s. &c, 

 Director of Agriculture to the Companhia 

 de Mozambique. 



(From the Tropical Life, Vol. V., No. 6, 

 June, 1909.) 



The Zimbiti Station is the first of 

 the three Agricultural Experimental 

 Stations which the Companhia de 

 Mozambique has decided to establish 

 for the purpose of testing the agricul- 

 tural possibilities of the three different 

 zones in its territory. It is situated in 

 the low country, 35 kilometres distant 

 from Beira, and is adjacent to the Beira 

 and Mashonaland Railway. The soil in 

 this neighbourhood varies from a light, 

 sandy alluvial to a rich, heavy clay. 

 A stream flows through the centre of 

 the site selected for the station, and the 

 land in the vicinity of this was thickly 

 covered with large timber trees. 



The station is named after a timber 

 tree which is very common in the 

 neighbourhood and known to the natives 

 as Zimbiti. Botanical specimens of this 

 tree were forwarded for identification 

 to the Royal Botanic Gardens, Kevv, 

 which resulted in eliciting the fact that 

 it is new to science, and it has been 

 distinguished by the Kew authorities as 

 Androstachys Johnsonii, Prain. 



It was intended to clear about 20 hec- 

 tares for preliminary experiments, but 

 as the commencement of this work was 

 delayed until November 18th, 1907, by 

 which time the rainy season was well 

 advanced, it was only found possible to 

 prepare about half this area tor the first 

 season's cultivation. The remainder was, 

 however, cleared and prepared for 

 planting before the end of the year 

 (November, 1907, to November, 1908), the 

 period covered by this report, 



Demarcation of Land. 



The selected area was divided up into 

 twenty square blocks, each measuring 

 one hectare, and each of these was sub- 

 divided into square plots containing 

 l-16th of a hectare. 



The hectare blocks were numbered 

 consecutively from 1 to 20, and the small 

 plots were given a letter of the alpha- 

 bet from A to P, as shown in the plan 

 sent in with the report. 



This system of demarcation consider- 

 ably facilitates reference to any parti- 

 cular area, and will enable a correct re- 

 cord to be kept of the crops grown in 

 successive years on every portion of the 



station. Such a record is of considerable 

 importance in estimating the effect of 

 various crop rotations. 



Staff and Labour Force. 



The European staff consisted of the 

 Superintendent and his assistant, and 

 fifty native labourers were constantly 

 employed. 



Crops. 



The principal annual crops grown were 

 cotton, maize, and tobacco. Owing to 

 the delay which occurred in commencing 

 the clearing and ploughing operations, 

 all of these were sown from six weeks to 

 two months too late in the season. The 

 yields obtained cannot, therefore, be 

 accepted as a fair criterion of the pro- 

 bable yields from these crops when 

 grow r n under more favourable conditions. 

 Previous to planting, the land for each 

 crop was thoroughly cleared of all roots 

 and tree stumps, then hoe-ploughed and 

 levelled. 



The Superintendent's records and field 

 notes in regard to each crop were given 

 in schedule form in the appendix to this 

 report when sent in, 



Cotton.— Twenty different varieties of 

 cotton were tested, and the following 

 tables furnish details in regard to the 

 results obtained from each. 



The Sea Island (Cherinda) plot mea- 

 sured l-25th hectare, the area of all others 

 being J hectare. All the plots were 

 similarly treated in legard to weeding 

 and harrowing. 



All the plots were attacked by cater- 

 pillars and aphis, but these were prompt- 

 ly exterminated by sprayings of Paris 

 green and kerosine emulsion respectively. 

 Grasshoppers were troublesome to young 

 plants, and stainers appeared in each 

 plot as soon as the bolls matured. 

 Vacancies were resown in each plot until 

 a regular stand of plants was obtained. 



The yields given in the following sche- 

 dule still further corroborate the opinion 

 expressed by the writer in previous 

 reports in regard to the advisability of 

 substituting upland varieties of cotton, 

 and perferably the long staple types, for 

 the Egyptiau and Sea Island varieties 

 formerly grown iu this territory. 



Description of variety, 



Long Staple Upland 

 Allen's Long Staple .. 

 Allen's Hybrid 

 Allen (Cherinda) 

 Griffin U.S.A. 

 Griffin (Cherinda) 

 Cook 



Commander 



Yield 



of 

 seed- 

 cotton per 

 hectare. 



kilos. 

 261-04 

 . 118-40 

 . 472-80 

 . 417 -GO 

 . 406-00 

 , 284'80 

 . 463*20 



Yield 

 of 

 lint 



per 

 hectare 

 kilos. 



84 00 



36-00 

 152-80 

 137-6 i 

 100-00 



88-40 

 123-20 



Percent- 

 age of 

 lint to 

 seed- 

 cotton. 



3218 

 30 40 

 32-32 



32*;)5 



26-1 



31-03 



30-55 



