308 



THE AGEICULTUrtAL XEW.S. 



Septembek 26, 1914. 



COTTON, 



mVERS' TYPE OF SEA ISLAND COTTON. 



W'lifii tlif l!i\ri>" type of .Sea I.^laml loltoii .seed was 

 tirst introduced into the West Indies some discussion took 

 ])lace concerning the extent to ■which this seed was covered 

 with fuzz. An article on the subject appeared in the A'lri- 

 itiltural A^ei'% Vol. TV, p. 106, in which Avas reported an 

 analysis of a typical .sample of EiTer.s'seed as imported. Kased 

 on the proportion of fuzz on the seeds in the sample. The 

 londusions arrived at were stated in the following paragraph 

 of the article thus: 'That in the selection of Kivers' .--eed 

 caie has been taken to select for the character 'jf the lint: that 

 .^niall variations in the character of the seed have received 

 less attention, hence the seed is to a certain extent variable in 

 character; that the characters of the seed are n(:>t fixed, so that 

 extreme types, such as fuzzy seed.s, on the one hand, and 

 absolutely clean seeds, on the other, both tend to produce 

 average seed of the Eivers' type, i.e., having a .small amount 

 of fuzz; and, finally, that the character of the lint is fixed and 

 does not vary with the variations in the character of the seed.' 



It was thought that some interest might attach tC' an 

 examination of the cottons now grown in various West Indian 

 islands, most of which have had their origin in Eivers" .seed. 

 Accordingly samples have been examined and tlie residts are 

 recorded in the following table, and the results of the examina- 

 tion of the original Itivers' .seed are included for comparison. 



The samples from Montserrat include Elvers', .Stirling 

 an<l Heaton from the plots grown on a large commercial scale. 

 The Elvers' sample came from a field planted from seed 

 derived in 1908-9 from a plant selected at the ^Mont-serrat 

 JJotanic Station from which commercial lots of seed have 

 been obtained in this instance without further regular 

 .selections. The sample of Stirling seed came from a field 

 planted with Stirling seed derived from Stirling seed obtained 

 from Barbados in 1912 and cultivated commercially in Mont- 

 serrat for two .seasons. The Heaton seed had its origin in 

 a single plant, not of Kivers' type, selected in the Montserrat 



Botanic Station in 190>'y and grown commonly from that 

 j>friod. 



The Mont.serrat .samples, St. Kitts T'l and T'O and 

 St. N'incent 2'o are derived from single plants grown from 

 seed f)riginally sent from St. Kitts and St. Vincent respec- 

 ti\ely: they do not represent commercial cidtivations. 



The .samples from St. Kitts, St. Vincent and Barbados 

 represent what are lielieved to be ordinary commercial samples 

 taken Avithout special selection or special precautions. 



It has hitherto been held that there exi.sts a .somewhat 

 do.se correlation between smooth black seed and inferior lint, 

 and also between this type of seed and a cotton plant of 

 straggling, tall habit bearing pale or nearly white blos.soms, 

 and pains have been taken in most West Indian colonies to 

 jjick out the smooth lilack seed from .seed used for planting. 

 It is not surprising, therefore, to find that smi«ith lilack seed 

 occurs less frerpiently in the .samples now examined (1914) 

 than in the seed originally imported (1904). 



In .some instances the smooth black seed has been 

 entirely eliminated ami in others redttced to ;i very low 

 percentage. 



It is curious to observe that the .samples from St. N'incent 

 — a colony which produces a quality of cotttm renown- 

 ed for its excellence — show the largest amounts of smooth, 

 black seed, though even here the amotmt is less than in the 

 f>riginal Elvers'. 



In addition to the removal of .smooth black seed from 

 seed used for jjlanting, it is customary to rogue out from the 

 fields any plants recognized as bearing lint of impure type, and 

 also plants not com[)aring generally to type: two methods of 

 selection on a crnnmercial scale are thus in oiieration and both 

 must exert their iiiHuence on the character and ipiality of the 

 succeeding ciitton cro|)s. 



The presumption now arises that there may lie at least 

 two kinds of smooth black seed, one of which is correlated 

 with inferior lint and associated with the tall plants with jiale 

 l)lo.ssoms,and which has been largely eliminatetl l>y the methods 

 of seed selection and rogueing adopted, and the other which i.'S 

 not correlated with inferior lint. If this is the ca.se, the smooth 

 black seeds now met with will not produce the poor types of 



COMPOSITION OF SAMPLES OF 1,000 SEEDS. 



