6 V O L . X J U L Y / D E C E M B E R , 1 9 5 9 N o s . 3 / 4 ANNUAL REPORT OF THE COCONUT RESEARCH INSTITUTE FOR J959 Editorial Board: Mr. M.M. KUMARAKULASINGHAM Dr. M.L.M. SALGADO (Chairman) (Director, Coconut Researcli Institute) Mr. E. MUTTUKUMARU Dr. D.V. LIYANAGE (Botanist, Coconut Research Institute) Mr. M.D.H. SENEVIRATNE (Secretary) Editor: Dr. M.L.M. SALGADO, Director Coconut Research Institute PUBLISHED BY THE COCONUT RESEARCH INSTITUTE OF CEYLON 3033—A ANNUAL REPORT 1959 JULY/DECEMBER NOTE BY DIRECTOR It has been the practice to submit the Annual Report and accounts~to the Hon. Minister of Agriculture, Lands, Irrigation and Power, in terras, of Section 8 of the Coconut Research Ordinance, No. 29, 1928. In view of the fact that the Report of the Auditor General is invariably delayed, even beyond the end of the year subsequent to the period reported, the publication of the Annual Report, as a Sessional Paper as has been the practice so far, is unduly delayed. It has therefore, been decided at the 183rd Meeting of the Coconut Research Board that the Report, consisting of the reports of the Chairman, Coconut Research Board; Director, Coconut Research Institute and of Research Officers, Chief Advisory Officer," Planting Officer and the Superintendents of the Institute's estates, should be immediately published as a number of the Ceylon Coconut Quarterly. M.L.M. SALGADO, Director, Coconut Research Institute and Editor, Ceylon Coconut Quarterly. 1st November, I960. ANNUAL REPORT OF THE COCONUT RESEARCH INSTITUTE FOR 1959 The present report is the 31st Annual Repor t of the Coconut Research Institute which was established by Ordinance N o . 29 of 1928 dated December, 1928. REPORT OF THE CHAIRMAN On January 1st 1959 the Coconut Research Board consisted of the following:— Ex-officio Members: Chai rman—The Director of Agriculture: Dr. M.F . Chandrara tne , Ph.D. , B.Sc. (Lond.) , D . l . C , M.B.E. Treasury Representative: Mr. E.B. Wiratunge. Coconut Rehabili tation Commissioner : Mr. B. Mahadeva, CCS. Chairman, L.C.P.A. : Mr. C.S. Samaraweera. Director, C.R.I . : Dr . M.L.M. Salgado, Ph .D. (Cantab.) , B.Sc. (Lond.) Dip.Agric. (Cantab.) . Nominated Members Nominated by the Honourable Minister from Senators and Members of Parliament: Mr. J .C.W. Munasinghe, M.P., and Parliamentary Secretary lo the Minister of Industries and Fishe­ ries, Mr. W.I. Hugh Fernando, M.P. Nominated by the Planters' Association of Ceylon: Mr. C.T. Van Geyzel, J.P.. and (Vacant). Nominated by the L.C.P.A.: Mr. C.A.M. de Silva and Senator Thomas Amarasuriya. Nominated by the Honourable Minister to represent the Small-Holders: Mr. E. Mut tukumaru. J.P. and Mrs. L.J. de S. Seneviratne. Mr. M.M. Kumarakulasingham was nominated by the Honourable Minister of Agriculture & Lands to represent the Small-Holders with effect from 7th January. 1959 in piace of Mr. E. Muttu­ kumaru. Mr. C.S. Samaraweera was re-elected Chairman of the L.C.P.A. Mr. E. Mut tukumaru was nominated to represent the Planters' Association of Ceylon with effect from 17th February, 1959. Mr. W.I. Hugh Fernando, M.P., Parliamentary Secretary to the Honourable Minister of Agriculture & Lands was re-nominated for a further term of 3 years with effect from 15th May, 1959. Mr. J .C.W. Munasinghe, M.P., ceased to be a member of the Coconut Research Board and was succeeded by Mr. M.M. Mustapha, M.P. with effect from 30th May, 1959. Mrs. L.J. de S. Seneviratne was granted 3 months leave of absence from Board Meetings. 2 Messrs. W.t . Hugh Fernando, M.P. and M.M. fviustapfia, M.P. , ceased to be members of1 the Coconut Research Board with the dissolution of Parliament on 5th December, 1959. Meetings.—Seven Meetings of the Coconut Research Board were held dur ing the y e a r — o n 12th January, 31st March, 27th April, 6th June, 1st August, 12th August and 17th October. Committees Administration Committee (Personnel a t 1st January, 1959): (1) D r . M.F . Chandrara tne , Chai rman. (2) Mr. E.B. Wiratunge. (3) Mr. W.I. Hugh Fernando , M.P. (4) Mr. C .A.M. de Silva. (5) Senator Thomas Amarasuriya, O.B.E. (6) Mr. C.S. Samaraweera. (7) Mr. B. Mahadeva. (8) Dr. M.L .M. Salgado, Director, C.R.I. The 25th, 26th, 27th and 28th Committee Meetings of the Administration Committee were held on 2nd March, 10th June, 25th August and 3rd October respectively. Extension Committee (Personnel a t 1st January, 1959): (1) Mr. C.T. Van Geyzel, •/./>. (2) Mr. C.A.M. dc Silva. (3) M r . B. Mahadeva. (4) Mr. W.I. Hugh Fernando, M.P. (5) Mr. E. Mut tukumaru . (6) Mrs . L.J. de S. Seneviratne. (7) D r . M.L .M. Salgado, Director, C.R.I . The 22nd and 23rd Meetings of the Extension Committee were held on 21st September and 24th October respectively. Research Committee (Personnel a t 1st January, 1959): (1) Mr . M . M . Kumarakulas ingham. (2) D r . M.F . Chandrara tne . (3) Mr . C.T. Van Geyzel, J.P. (4) Mr. E. Mut tukumaru . (5) Mr. J .C.W. Munasinghe, M.P. (6) Senator Thomas Amarasuriya, O.B.E. (7) D r . M .L .M. Salgado, Director, C.R.I . The 30th and 31st Meetings of the Research Committee were held on 16th April, 29th May and 29th September respectively. Buildings Sub-Committee (Personnel at 1st January, 1959): (1) Mr . B . Mahadeva. (2) Senator Thomas Amarasuriya, O.B.E. (3) Mr . C.S. Samaraweera. (4) Dr . M.L .M. Salgado, Director, C.R.I . A Meeting of the Chairman and Directors of the three Research Institutes was held on 11 th July, 1959. A.V. R I C H A R D S , Chairman, Coconut Research Board. 3 REPORT OF THE DIRECTOR 1. S T A F F The Staff of ths Coconut Research Institute at the end of 1959 was as follows: Administration Division Director — D r . M.L.M. Salgado, Ph.D. (Cantab.) , B.Sc. (Lond.) , Dip.Agric. (Cantab.) . Chief Administrative. Officer and Secretary to the Board — Mr. S.C. Kahawita, B.Com. (Lond.) . Assistant Secretary — Mr. T.T.A.J.C. Samarasinghe, LL.B. (Ceylon). Botany Division Botanist — D r . D.V. Liyanage, Ph .D. (Manch.) , B.Sc. (Lond.) . Research Assistant — Vacant. Chemistry Division Chemist — M r . W.R.N. Nathanael , M.Sc. (Lond.) , A.R. l .C. Research A s s i s t a n t — M r . T.S. Balakrishnamurthie, B.Sc. (Lond.) . Soil Chemistry Division Acting Soil Chemist — Dr. D.A. Nethsinghc, D.Phil. (Oxon.), B.Sc. (Ccy.), A.R. l .C, (Lond.). Research Assistant — Vacant. Agrostology Division Acting Agrostologist — Mr. K. Santhirascgaram, B.Sc. (Ceylon). Research Assistant — Miss. N . Ramalingam, B.Sc. (Ceylon). Planting Division Planting Officer — Mr. P.D.L. Fernando. Assistant Planting Officer — Mr. C.W.S. de Silva. Advisory Division Chief Advisory Officer — Mr. C A . Wickremasuriya, B.Sc. (Ceylon). Crop Protection Crop Protection Officer — M r . Hilary F . Goonewardena, B.Sc. (Sydney), B.Agric.Sc. (N.Z. ) : on overseas study leave. Officer-in-Charge — Mr. J .X .F . Kirthisinghe. Biometrv Biometrician — Mr. V. Abeywardena. Research (Assistant) Statistics — Mr. J .K.T. Fernando, B.Sc. (Ceylon). 4 Dr. D.V. Liyanage, B.Sc. (Lond.), Ph.D. (Manch.), Botanist was away six weeks from March 16th, I9S9 on a U.S.O.M. Scholarship visiting Philippines, Indonesia and Thailand. Appointments The following have been appointed to the Staff during the course of the year — Senior Staff—Nil. Intermediate Staff — Mr. J.K.T. Fernando, B.Sc. (Hons.) (Ceylon), as Research Assistant (Statistics). Assistant Staff Grade I — Nil. Assistant Staff Grade II — 4 Technical Assistants, 6 Coconut Instructors, 4 Clerk-Typists and 1 Field Assistant. Minor Staff—7 Nursery Attendants, 12 Lab. and Field Attendants, 1 Office Attendant and 1 Power House Attendant. Promotions The following were promoted during the course of the year — Senior Staff—Nil. Intermediate Staff—Dr. D.A. Nethsinghe, B.Sc. (Ceylon), D.Phil. (Oxon.), A.R.I.C., Research Assis­ tant to Soil Chemist as Acting Soil Chemist with effect from 19-1-59. Mr. CA. Wickremasuriya, B.Sc. (Ceylon), Understudy to Chief Advisory Officer as Chief Advisory Officer with effect from 14-12-59. Assistant Staff Grade I — Nil. Assistant Staff Grade 11 — Mr. W.V. Fernando, Advisory Field Officer to Assistant Staff Grade 1 as District Coconut Instructor under the Citronella Subsidy Scheme, with effect from 15-12-1959. Minor Staff—Mr. F.B. Perera, Field Attendant to the Soil Chemist was promoted to the Selection Grade. Resignations The following left the services of the Institute during the course of the year — Senior Staff — Muh. Edwin A. Peiris, Chief Advisory Officer, since 14th December, 1957, left the services of the Institute on 14th December, 1959 after the completion of his contract. Intermediate Staff — Mr. J.C.L. de Mel, B.A. (Cantab.), Research Assistant in Agricultural Economics. Assistant Staff Grade I — Nil. Assistant Staff Grade II — 2 Technical Assistants and 1 Field Assistant. Minor Staff—7 Members of the Minor Staff. 2. MEETINGS The Director attended the following meetings, besides the meetings of the Coconut Research Board and its Committees — The Research Committee, the Administrative Committee and the Extension Committee:— (i) Committee of Chairmen and Directors of Tea, Rubber arid Coconut Research Institutes. (ii) Department of Census and Statistics: Agricultural Census. (iii) Conference at the National Planning Council with the Delegates of the International Atomic Energy Commission (with Soil Chemist). (iv) Permanent Secretary to Minister of Agriculture and Lands and the Chairman regarding control of the Coconut Caterpillar Infestation at Ganewatte. 5 (v) Meeting of Heads of Departments convened by H o n . Minister of Agriculture and Lands a t Moneragala. (vi) Meeting of the Planning Committee of the H o n . Minister of Agriculture and Food. (vii) Land Commissioner 's Depar tment on the Supply of Planting Material (with Planting Officer). (viii) Symposium on Pasture organised by the Faculty of Agriculture, Peradeniya (with Acting Agrostologist). (ix) F A O Fertilizer Pool (at Ministry of Commerce) . 3. C O N F E R E N C E S Conferences were held to discuss the following subjects: (a) A Research Conference and Field Day a t Vanatavillu, where 2,000 acres of virgin jungle had been alienated to middle class allottees for planting with coconuts. Various aspects of the geology, soils, agro-climatology and systems of catch cropping were discussed. Dr. C.R. Panabokke of the Dry Farming Research Station, Maha Illuppalama and Mr. D.B. Pattiarachchi of the Department of Mineralogy addressed the Conference and participated in the discussions and in the field tour of the area. (b) Dr. H .H. Smith of the International Atomic Energy Commission delivered an Illustrated Lecture and participated in a discussion on a Radio Isotopes. (c) 'Rad io Active Isotopes in Agricultural Research' discussion led by Dr. D.A. Nethsinghe. (d) Pasture and Animal Husbandry with reference to Coconut Cultivation in which Prof. C M . Donald participated in the discussion. (, Pi K — Total Ki 1,167 1,231 1,199 1.199 1.470 1.449 1,441 1.453 K s 1,472 1,613 1,529 1.538 P — Total 1,369 1,431 1,390 1,397 Pn />, P, N —Total N 0 1.345 1,494 1,388 1,409 N r 1,432 1,414 1,453 1.433 N 2 1,332 1,385 1,328 1.348 P — Total 1.409 1.433 1,348 1,397 2. Manurial x Cultivation Experiment (Ratmalagara Estate) This factorial experiment compares the effect of potash, phosphoric acid, and ploughing on adult palms. It includes all combinat ions of the following treatments and is of the 3 X 2 X 2 type consisting of 6 blocks of 6 plots each. K„ = no potash ^ p _ „, . . ,. C n = no cultivation K, = 1 lb. K 2 0 / p a l m > £ - ^ P h o s P h a t ^ \ C = Ploughing once in 2 years K 2 = 2 lb. K 2 0 / p a l m J ^ 1 I D ' P n o s P n o n c a c i d J a t the time of manuring All plots are given a basic application of 3 lb. Sulphate of Ammonia per palm. The 16th year of this experiment and its 9th biennial manuring was concluded in June 1959. 12 The yield da ta for the main effects for the year 1958-1959 are given below in Table II («). TABLE II (a) Treatment ., . Calculated Lb. copral a s acre percentage Difference lb. copra) acre Copra out­ turn nutsf candy Difference i mils cam, Po 1,012 100 — — — P 1,850 183 + 838 1,219 + 74 C 0 1,307 100 — 1,210 — C 1,555 119 + 248 1,178 - 3 2 Significant difference P -05 = 156 lb. copra/acre. K 0 1,350 100 — 1,220 + 7 K , 1,407 104 + 57 1,203 - 1 7 K 2 1,537 114 + 187 1,159 -61 Significant difference P ' 05 = 192 lb . per acre. dy The high response to phosphate has been maintained and the increase of 838 lb. copra per acre has been the highest recorded for this experiment. Ploughing has produced a significant response for the second year in succession, al though from the 4th to the 14th year of the experiment no response had been obtained. I t is possible that the lack of response to ploughing in the previous years was due to the fact that the entire experimental area was disc-harrowed thrice o r four times a year to control weeds. Since 1956 this practice was stopped, and weed control done by slashing only. I t appears therefore that where regular harrowing is carried out to control weeds, further cultivation by ploughing may be superfluous. 3. 3 x 3 X 3 N .P .K. Manurial Experiment on Young Palms — (Ratmalagara Research Station) This experiment compares the effect of applying nitrogen, phosphoric acid and potash to young palms at three different levels in all combinations. I t consists of 6 blocks of 9 plots each, and is of factorial design where the higher order interactions are confounded with that of blocks. The 11 th year of the experiment was completed in 1959, and the annual manur ing carried out in November . The rates of fertiliser application, which were the same as in the previous year, are as follows:— N 0 = no nitrogen f N x = 1$ lb. Sulphate of Ammonia /pa lm f x N 2 = 3 lb. Sulphate of Ammonia /pa lm J P 0 = no phosphoric acid ") V1 = 1£ lb. Saphos phosphate/palm y x P 2 = 3 lb. Saphos phosphate/palm J K 0 = no potash K x = 1 j lb. Muriate of Potash (60%)/palm K 2 = 3 lb . Muriate of Potash per palm, t3 (a) Palms in flower: At the end of the 1 Ith year of this experiment (December 1959), 918 palms were in flower, out of a total of 972 experimental palms. The distribution of the palms in flower according to the main treatment effects are shown in Table III {a) below:— TABLE III (a) Palms in flower % % of Total N 0 . . 313 100 97 N , . . 309 99 95 N , 296 95 91 Pn 289 100 89 P, . . P;> 312 317 108 110 96 98 K 0 . . 297 100 92 K, . . 306 103 94 K 3 . . 315 106 97 s yield data in lb. copra per acre for the main effects is given in Table III (/>)• TABLE III (b) lb. copra/ Treatment acre (55 palms per acre) Calculated as percentage Difference in lb. copra per acre Copra out­ turn nuts/ candy Difference in nuts/candy N 0 950 100 — 1,214 — N , 1,091 115 + 141 1.202 - 1 2 N 2 1,000 105 + 50 1,243 + 29 P.. 655 100 — 1,190 — Pi 1,204 184 + 549 1,211 + 21 P2 1,182 180 + 527 1,244 + 54 K„ 895 100 — 1,254 — K, 979 109 + 84 1,219 - 3 5 K , 1,167 130 + 272 1,193 -61 Significant difference P - 0 5 — 82-5 lb. copra/acre Phosphoric acid has produced a marked effect on yield at the lower level (P,) , which however is not further increased at the higher level. Potash produces a significant difference in yield at the lower level, which however is markedly increased at the higher level. 14 The mean yields for 1959 (lb. copra/acre) for the various t reatment combinat ions are given in the 2-way tables below:— TABLE m (c) Treatment NQ *i iV2 K—Total K„ 808 1,036 842 895 895 1,061 983 979 K 2 1,149 1,176 1,176 1,167 N — T o t a l 950 1,091 1,000 1,014 ^0 P i K—Total K „ 600 993 1,092 895 Kx 578 1,172 1,189 979 K 2 788 1,448 1,266 1,167 P — T o t a l 655 1,204 1,182 1,014 Po Pi N— Total N 0 737 1,130 985 950 Nt 678 1,313 1,282 1,091 N 2 551 1,170 1,280 1,000 P — T o t a l 655 1,204 1,182 1,014 4. Manurial Experiment on Methods of Application (Marandawila Group, Bingiriya). This experiment compares the effect of broadcasting and ploughing in of fertilisers with that of circular t rench manur ing . I t consists of 3 blocks of 9 plots each and is of an unreplicated 3 x 3 x 3 factorial design compar ing all combinat ions of N . P . and K . applied in the following ways :— N o = N o nitrogen 1 N c = Ni t rogen applied in circular trenches j (3 lb. Sulphate of Ammonia /pa lm) )• X N B = Ni t rogen broadcast and ploughed | (3 lb . Sulphate of Ammonia per palm) J P o = N o Phosphoric acid 1 P c = Phosphoric acid applied in circular trenches I (2 lb. Saphos Phospha te per palm) }• X P B = Phosphor ic acid broadcast and ploughed | (2 lb . Saphos Phosphate per palm) J K 0 = N o Potash K c = Potash applied in circular trenches (2 lb. Muria te of Potash per palm) K B = Potash broadcast and ploughed (2 lb . Mur ia te of Potash per palm) 3 6 3 3 — B 15 The 10th year of this experiment was completed in June 1959. The biennial manuring was carried ou t in June 1959. The yield data for the 10th year is given in Table IV (a) and it is seen that even after the 10th year of this experiment there is no significant difference shown between the two methods of application, al though there is a response to nitrogen, phosphoric acid and potash. TABLE IV (a) (Yields adjusted by covariance analysis) Treatment lb. copra/ acre. Calculated Difference in as percentage lb. copra/acre N o 1,653 100 — N 1,855 112 + 205 N c 1,935 118 + 282 Po 1,575 100 P B 1,893 120 + 318 P c 1,898 120 + 323 K o 1,633 100 — K B 1,857 114 + 224 K C 1,913 117 + 280 Significant difference P ' 0 5 = 188 lb. copra/acre. 5. Manurial Experiment on Organics vs. Inorganics (Marandawila Group) Bingiriya. This experiment comparing the effects of annual and biennial application of inorganic and organic manures on adult palms was commenced in June 1958. The experiment is of randomised block design, consisting of 6 blocks of 6 plots each. The treatments are as follows:—(a) Control (b) Inorganic mixture applied annually, (c) Inorganic mixture applied biennially, (d) Organic mixture applied annually, (e) organic mixture applied biennially, and (f) Cattle manur ing and supplements applied biennially. The inorganic mixture consists of equal parts by weight of sulphate of ammonia , saphos phosphate, and muriate of potash (50 per cent). The rate of application is 5 lb. per palm annually, and 10 lb. bien­ nially. The organic mixture is composed of fish guano (4 lb.) bone meal (1J lb.) and muriate of potash 50 per cent ( I f lb.). The rate of application is 7 lb. per palm annually, and 14 lb. biennally. The cattle manur ing treatment consists of tethering a pair of neat cattle for 5 nights supplemented with saphos phosphate (2J lb.) and muria te of potash 50 per cent (2\ lb.) per palm. The 1st year of this experiment was completed in June 1959. It is still too premature to draw any conclusions from the yield data which do not show any significant differences between the various treat­ ments , 16 6. Observation Trials on Organics vs. Inorganic manures (Bandirippuwa Estate) . These simple trials which were initiated in 1950 were closed down in early 1959. Five plots, each consisting of about 90 adult palms were subject to the following t rea tments :— 1. Control . 2. Coir dust applied a t the rate of one double bullock cart load per square, and dug over with mammott ies . 3. All plant products , except coconut oil, returned to the soil. 4. Inorganic manures only. The following fertilizer mixture was applied biennally in circular tren­ ches :— Sulphate of Ammonia 3 lb. "1 Saphos Phosphate 2 lb . f Per palm. Muriate of Potash 2 lb. J 5. Organic manures only. Crushed poonac was used as the source of nitrogen, and ash to supply potash and phosphoric acid — the amounts applied being equivalent to 0*6 lb. N . 0 ' 6 lb. P 2 0 5 and 1 -2 lb. K 2 0 per palm (similar to the inorganic mixture). Statistical analysis of the yield data for the period 1950—1958 has shown the different treatment to have the following order of efficiency:— 1. Inorganics 2. Organics 3. All p lant products returned 4. Coir dust 5. Control . 7. Observation trials on Yellowing Palms, Walgama Estate , Rukmale and Mattegoda, Polgasowita. These trials were commenced in 1956 to ascertain whether the inexplicable yellowing of regularly manured palms in certain areas could be arrested by the application of magnesium or cattle manure . Magnesium has been applied either as dolomite or magnesium sulphate. The problem of yellowing has been largely associated with palms growing on highly acid lateritic soils of the Western and Southern Provinces. In 1959, a further t rea tment—liming — was introduced at Walgama Estate to find out whether a rise in soil p H would be helpful. Hitherto there has been no noticeable change in the colour of the foliage of palms receiving the different t reatments. C . L A B O R A T O R Y I N V E S T I G A T I O N S Considerable time was spent in examining the reliability of new and quicker methods for the chemical analysis of potash, calcium and magnesium in soils and plant materials. The use of the E E L flame photo­ meter for estimating potash in soil extracts, leaflet samples and nutwater was found to be much more reliable and quicker than the old precipitation method. The volumetric E .D.T.A. method for estimating calcium and magnesium has also been found to be more convenient than the s tandard precipitation techniques. 17 Analysis of nitrogen in soils a t Bandir ippuwa showed a significant accumulation of ammoniacal and nitrate nitrogen even a t depths of 8—10 ft. Chemical analysis of soils, leaflet samples and nu t water from the trials on yellowing palms a t Wal- gama and Mattegoda estates has hitherto not given any conclusive results regarding the cause of yellow­ ing. Studies on the effects of air drying on the potash status of a sample of soil from Bandir ippuwa Estate indicated tha t while drying caused no change to occur in its content of exchangeable potash, the concentrat ion of potash in a 10" 2 molar calcium chloride extract of the soil increased. Chemical analysis of soil samples taken from the Methods of Application experiment a t Marandawila in 1950 (premanurial) was carried ou t as a preliminary step in studying the effect of fertiliser placement on the distr ibution of nutrients in the soil. The results of potash and phosphate analysis on nutwater from the manuria l experiments on young palms a t Ratmalagara , and the N .P .K . experiment at Bandir ippuwa showed the same trends as in the previous years. Details of the above investigations are given below. ( 0 New methods of Chemical analysis. (a) Estimation of potash using the EEL flame photometer. The E E L flame photometer was tested for the direct determination of potash in nutwater, water and acid extracts of ashed plant materials and extracts of soil with normal ammonium acetate and 1 0 - 2 molar calcium chloride solutions. Its high sensetivity has proved to be particularly useful for the direct estimation of potash in soil extracts where the concentration of potash is in the very low region of 0 — 10 ppm. K 2 0 . Highly reproducible results were obtained and excellent recoveries made of s tandard amounts of added potash. I t was found that acid extracts if kept too long in contact with glassware dissolved sufficient potash from the glass to give significant errors in the measurement of low concentrations of potash. Compar ison of the flame photometr ic method with the usual sodium cobaltinitrite precipitation method (Treadwell and Ha l l : Analytical Chemistry Vol. I I . p . 69, Wright : Soil Analysis — p. 87), how­ ever failed to show any concordant results. A detailed investigation was therefore carried ou t to ascertain the cause of this discrepancy. It was found that the s tandard precipitation method is subject to the follo­ wing errors which must necessarily give rise to variable results:— (i) T h e potassium cobaltinitrite precipitate tends to dissolve in the various wash liquids recommen­ ded — 1 0 per cent acetic acid, 2 . 5 per cent sodium sulphate, or water, giving results lower than what is theoretically expected. Greatest solubility was shown in acetic acid, and least in water. The amoun t of precipitate dissolved increased with increasing amounts of wash liquid used. The prescribed methods do no t specify the amounts of wash liquid to be used. Variable results could be obtained by using different quantities of the wash liquid. The results given in Table I (a) show the amounts of precipitate lost by washing with 10 per cent acetic acid as determined by analy­ sing the wash liquid by the flame photometer . The quanti ty of potassium in the wash liquid however did no t reflect the changes in the actual weight of precipitate after each washing. This was apparently governed by other factors as well, which are 18 discussed below. These studies werec arried out with 10 mg. quantities of K z O (i.e. equivalent of about SO mg. potassium cobaltinitrite precipitate). TABLE I (a) Potash in aliquots of 10 per cent acetic acid washings of potassium, cobaltinitrite precipitate _ Potash in 101111 °J$9 Total Loss of Precipitate W a s h N o w a s h „ i d ascobaltim- J No. , J trite precipi- r , \ (mg.K.O) taJ{mgr («y.) 1 0.16 0.77 2 0.25 1.21 1 3 0.295 1.42 4 0.34 1.64 5 0.35 1.69 6.73 Weight of precipitate No. 1 before 1st wash = 50.6 mgms. Weight of precipitate No. 1 after 5th wash = 4 6 . 3 mgms. 1 0.09 0.43 2 0.21 0.99 3 0.34 1.64 4 0.30 1.45 5 0.43 2.07 Weight of precipitate No. 2 before 1st wash = 55.1 mg. Weight of precipitate No. 2 after 5th wash = 45.0 mg. 1 0.06 0.31 2 0.17 0.80 3 0.32 1.54 4 0.27 1.33 5 0.40 1.93 Weight of precipitate No. 3 before 1st wash = 52.1 mg. Weight of precipitate No. 3 after 5th wash = 42.8 mg. Similar results were obtained with 2.5 per cent sodium sulphate and water. With sodium sulphate, the situation was more complicated since the potash precipitate tends to increase in weight with washing due to the absorption of sodium sulphate by the precipitate, while at the same time potash was found to be dissolved in the wash liquid. Subsequent washing of the same precipitate with water confirmed this, for the absorped sodium sulphate was now washed off the precipitate, as was detected by the flame method. (ii) Incomplete washing away of the sodium cobaltinitrite reagent tends to give results higher than what is expected theoretically. (Hi) The use of tared filter paper for weighing precipitates gives rise to errors due to the creeping of precipitate from the inner to the outer paper. The use of sintered glass filter crucibles is sometimes subject to error due to loss of weight of the crucible itself on repeated washing, possibly due to a gradual breakdown of the sintered glass plate. 19 (iv) The composition of precipitate is liable to vary from that given by the accepted s tandard formula K 2 N a ( C O N O ( i ) H 2 0 ) . Various quantities of potassium cobaltinitrite previously precipitated, thoroughly washed with 10 per cent acetic acid and dried were accurately weighed ou t and dissolved in 10 per cent Analar hydrochloric acid. The potash in solution determined by the flame method was found to be distinctly higher than that which was theoretically expected from the above formula (see Table I (a 1)- According to Milton and Walkers (Methods of quantitative Micro — Analysis — p. 178), the precipitate would also consist of some K 3 C o ( N o 2 ) 6 , the amoun t of which is largely governed by the temperature and speed of precipitation. These investigations have clearly shown that the flame photometric method is superior in all aspects to the old precipitation method for estimating potash. The latter method is likely to give results of poor reproducibility subject to serious errors particularly in estimates of small quantities of potash. TABLE I (a 1 ) Comparison of K 2 0 theoretically expected from precipitate with that actually found by the flame method. Weight ofpot as- K.20 (mg.) expec- sium cobaltinitrite ted theoretically K20 (mg.) found precipitate dis- from K^Na Co (No.^ by flame method solved (mgms.) Hfi 70 14.52 17.0 92 19.08 20 .5 139 28.83 32 .3 178 36.91 4 0 . 0 235 48 .73 50 .5 327 67.81 70 .0 (b). E.D.T.A. method for estimating calcium and magnesium in soil extracts, leaflets samples and nut water. The volumetric method as described by Cheng and Bray (Soil Science 72 p. 449 — 58) offers a simpler and more rapid means of estimating calcium and magnesium when compared with the usual oxalate and pyrophosphate methods . Investigations were carried out for testing the suitability of the method for estimating exchangeable calcium and magnesium in N . ammonium acetate leachates of soils, and the total calcium and magnesium in leaflet samples and nut water. The ammonium acetate solutions were subject to the usual evaporation and extraction procedure after adding varying amounts of s tandard calcium and magnesium ranging from 1 — 10 mg. Ca and Mg. The method gave high reproducibility of results and good recoveries of added Ca and Mg. The method also proved suitable for estimating calcium and magnesium in leaflet samples which had been subject to the dry ashing technique. But wet digestion with sulphuric, perchloric and nitric acids was unsatisfactory as considerable difficulty was encountered in obtaining a clear end point in the titration — possibly owing to the presence of the acid radicals in high concentrat ion. The estimation of total C a a n d Mg. in nu twa te r by the E.D.T.A. method was not successful whether the dry ashing o r wet acid digestion technique was used to oxidise the nut water. The problem needs further study. (II) Soil Nitrogen Studies. Previous studies on soils from coconut estates and virgin jungles from different parts of Ceylon have shown the general pattern that there is a greater proport ion of ammoniacal nitrogen than nitrate nitrogen both in the top and sub-soils (vide Annual Repor ts C.R.I . 1954, 1955, 1956, 1957). Further investigations on Bandirippuwa soils have shown that this pattern is maintained right down the soil profi le—and that there is a considerable accumulation of both ammoniacal and nitrate nitrogen even up to depths of 10 feet. 20 Table II (a) gives the moisture, ammoniacal and nitrate nitrogen p H and total exchangeable base of soil samples taken at different depths from a pit in the Bracharia brizantha plot of Agrostologist 's Division. TABLE I I (a) Depth Moisture % Nitrate N ppm. Ammoniacal N ppm. pH Total exchangeable bases m.e.\ 100 gms. soil 0 — 9" 11.6 9 . 7 17.6 5 .8 0.91 9 — 18" 4 . 2 8 .0 15 .4 5 . 4 1.04 18 — 27" 13.1 6 .0 8 .8 5 .2 0 .71 27 — 36" 6 .2 5 .5 9 . 9 4 . 9 — 36 — 4 5 " 9 .2 5 .5 7 . 7 4 . 9 1.45 45 — 54" 10.3 7 . 0 6 .6 4 . 6 1.15 54 — 63" 11.9 5 .5 9 . 9 4 . 8 1.04 63 — 72" 9 . 4 6 .2 11.5 4 . 7 0 . 8 6 72 — 8 1 " 9 . 6 5 .5 11.5 4 . 7 0 .89 81 — 9 0 " 10.5 11.0 8 .8 4 . 9 0 .97 90 — 96" 10.9 12.0 4 . 9 4 . 6 1.02 Three soil profile pits were dug at Bandir ippuwa estate opposite the Chemist 's laboratory in a traverse due east a t intervals of about 200 yards. The land slopes in an easterly direction. Ni t ra te and ammoniacal nitrogen estimations on the soils at different depths show the accumulation of significant amounts of nitrogen even at 8 — 10 feet. The results are given in Table 11 (b). TABLE I I (b) Pit No. 5 Pit No. 6 Pit No. 7 Depth of Soil Nitrate Ammoniacal Nitrate Ammoniacal Nitrate Ammoniacal Profile Nppm. Nppm. N ppm. N ppm. Nppm. Nppm. 0 - 9" 7 .0 16.6 5 .5 2 2 . 4 6 .25 12.8 9 — 18" 6 . 0 13.9 2 . 5 16 .0 10.0 11.8 1 8 — 27" 1.25 16.0 2 . 5 16 .0 5 .5 10.1 2 7 — 36" 4 . 0 18 .0 6 .2 16 .0 4 . 0 10.7 3 6 — 45" 2 . 5 17.1 2 . 7 14 .4 9 . 0 2 6 . 2 45 - 54" 2 . 0 16.6 6 .2 14 .4 5 .5 26 .7 54 — 63" 2 . 7 16.0 2 . 5 19 .3 5 .5 19.8 63 — 72" 2 . 5 18.0 7 . 0 10.7 6 . 0 2 3 . 5 72 — 84" 4 . 0 17.1 2 . 0 18.0 2 . 5 2 8 . 9 84 — 96" 6 .2 21 .4 5 . 5 17.1 5 .6 2 8 . 3 96 — 108" 8.75 14.9 — — — — 108 — 120" 5 .5 17.1 — — — — 1 2 0 — 132" 6 . 0 2 2 . 4 — — — — (DT) Chemical studies of soil, leaflet and nut water samples from the trials on yellowing-palms at Walgama and Mattegoda Estates. (a) Analysis of soils from Walgama Estate. Two series of soil samples were taken from Walgama Estate from areas in which the palms were yellowing and in which they appeared to be normal and green. Each sample was composed of three borings. The first series was air dried and then subject to chemical analysis for N , P, K, Ca and Mg. The pH ' s were measured in 10. ' - molar calcium chloride. The results are given in Table I I I (a). The second series of soil samples were collected in polythene bags and treated with toluene. The soils were analysed in their d a m p state — without air drying — for total, nitrate, and ammoniacal ni tro­ gen. The results are shown in Table 111 (a 1 ) . 21 Chemical analysis of the soils has failed to show any consistent differences between the soils from the good and affected areas. (b) Chemical analysis of leaflet samples from adult yellow and green palms from trials at Mattegoda Estate- Leaflet sampling was carried out in December 1958 on groups of adult yellowing palms subject to the following t rea tments :— A — N o manure B — 5 lbs. Baur ' s special per pa lm C — 5 lbs. Baur ' s special per palm + 4 5 4 gms. M g . S 0 4 D — 5 lbs. Baur ' s special per p a l m + 4 0 8 gms. Dolomi te E — Cattle manure + lbs. saphos phosphate + 2 lbs. 50 per cent muriate of potash. Leaflet samples were also taken from normal healthy looking green palms receiving 5 lbs. Baur 's special only. T h e samples were taken from the 1st fully opened leaf, and also from a mature leaf — either from 12th — 15th leaf, o r 16th — 19th leaf. Each sample was composed of leaflets from three palms. The leaflets were chosen from the centre par t of the leaf, and only the middle portions of leaflets — free of mid ribs were taken for analysis after air drying. TABLE I U (a) (Series 1) Chemical analysis of soils from Walgama Estate (Series I sampled on 16-3-59) Good Area — (Green Palms) (a) M A N U R E CIRCLES Sample No. Layer pH Total N. ppm. Truog ppm. Total ex­ changeable bases me% Exch. KzO me% Exch. Ca. me% Exch. Mg. me% Exch. Mn mg/ 100 gm. Soil 2 0 — 9" 4 . 5 1,700 54 2 .54 0 .063 1.70 0 .44 0 .18 5 0 — 9" 4 . 3 350 6 5 . 6 1.29 0 .052 0 .73 0 .30 0 .13 6 9 — 18* 4 . 5 220 34 .4 0 .84 0 .070 0 .39 0 .20 0 .07 13 0 — 9* 4 . 8 60 277 2 .68 0 .091 1.95 0 .43 0 .04 15 0 — 9" 4 . 5 190 12 1.34 0 .099 0 .65 0 .24 0 .06 16 9 — 1 8 " 4 . 2 190 9 .6 0 .97 0 .070 0 .39 0 .10 0 .02 17 0 — 9" 3 . 8 90 180 1.64 0.165, 1.04 0 .33 0 .03 18 9 — 1 8 " 3 .3 70 4 6 . 4 1.55 0.066 1 0.91 0 .48 0 .04 (b) S Q U A R E C E N T R E S 1 0 — 9" 3 .7 1,030 20 0 .89 0 .130 0 .37 0 .18 0 .08 3 0 — 9" 4 . 5 690 54 0 .89 0 .093 0.281 0 . 2 0 0 .09 4 9 — 1 8 " 4 . 6 310 42 0 .79 0.201 0 .30 0 .15 0 .07 Affected Area — (Yellow Palms) (a) M A N U R E CIRCLES 9 0 — 9" 5 .2 240 108 1.82 0 .080 J . 2 2 0 .20 0 .06 10 9 — 18" 4 . 8 240 4 3 . 2 1.59 0 .048 0 . 9 0 0 .23 0 .06 11 0 — 9" 4 . 8 120 94 1.85 0 .059 1.34 0 .24 0 .08 12 9 — 1 8 " 4 . 0 80 34 1.34 0 .069 0 .92 0 .23 0 .06 14 0 — 9" 4 . 8 70 124 3 .44 0 .148 2.81 0 .42 0 .07 19 0 — 9 " 3 .5 50 110 2 .10 0 .134 1.27 0 .58 0 .04 20 9 — 1 8 " 3 .4 60 18.4 1.05 0.377 0 .64 0 .37 0 .04 (b) S Q U A R E C E N T R E S 7 0 — 9" 4 . 8 250 2 8 . 8 1.78 0 .064 1.03 0 .42 0 .07 8 9 — 1 8 " 5 .4 330 4 1 . 6 2 . 0 6 0 .063 1.05 0.21 0 . 0 6 22 TABLE ffl (a1) Nitrogen analysis of soils from Walgama Estate (Series II sampled on 12-5-S9) Good Area—(Green Palms) (a) M A N U R E CIRCLES (b) Sample No. Layer % Moisture Total N.ppm. Nitrate N. ppm. Ammoniacal N.ppm. 8 0 — 9" 12.4 980 7.1 40 .1 9 — 18" 10.9 650 9 . 0 2 6 . 4 ' 10 0 — 9" 12.8 1,040 8 . 0 32 .0 12 0 — 9" 14.9 1,020 9 . 4 3 6 . 0 S Q U A R E CENTRES 7 0 — 9" 13.8 1,320 16.0 3 1 . 8 9 — 18" 12.4 430 3.1 2 5 . 0 9 0 — 9" 15.2 1,220 6 .5 39 .5 11 0 — 9 " 15 .6 1,290 9 . 5 4 2 . 9 Affected Area—(Yellow Palms) (a) M A N U R E CIRCLES 2 0 — 9" 14.5 1,230 7 .3 24 .3 9 — 18" 10.4 330 7 . 0 6 .7 4 0 — 9" 15.9 1,150 13.4 24 .8 9 — 18" 13.9 870 10.2 8 . 2 6 0 — 9" 15 .0 1,020 11.8 2 3 . 2 9 — 1 8 " 12.1 660 11 .4 6 .9 R E C E N T R E S 1 0 — 9" 14.7 1,290 9 . 4 28 .3 9 — 18" 12.7 620 10 .0 2 1 . 6 3 0 — 9" 16.7 970 12.0 2 5 . 0 9 — 18" 13.2 480 9 . 2 13.3 5 0 — 9" 16.2 1,110 9 .5 29 .5 9 — 18" 15.6 640 9 . 5 2 6 . 0 23 TABLE II (b) Chemical analysis of leaflet Samples from Yellow and green palms of Mattegoda Estate. 1ST L E A F Sample No. %N %Ca. %Mg. ppm.Fe. ppm.Mil. Green Palms 1 1.45 0 .19 2 .93 0.241 0 .098 567 125 Yellow Palms Trea tment A — 2 1.67 0 .13 2 .23 0 .286 0.117 217 117 — 14 1.25 0.21 2 .00 0 .375 0 .135 192 75 B — 6 1.13 0 .15 2 .73 0 .268 0 .045 150 100 — 16 1.56 0.21 0.281 0.028 — 84 C — 5 1.48 0 .13 2 .90 0 .244 0 .089 117 84 — 18 1.02 0 .23 2 .58 0 .258 0 .110 567 73 D 4 1.54 0 .15 2 . 9 0 0 .255 0 .047 225 150 — 8 1.91 0.21 1.53 0 .390 0 .123 150 108 E 3 1.75 0.21 3 .00 0 .274 0 .023 217 117 — 13 1.64 0 .23 2 .59 0 .258 0 .084 108 50 16TH — 19TH L E A V E S Green Palms — 11 1.67 0.11 1.09 0 .475 0.061 152 225 Yellow Palms Trea tment B — 12 1.56 0 .16 1.95 0 .305 0 .090 134 85 C — 15 1.78 0 .18 1.54 0 .399 0.018 150 158 D — 10 1.33 0 .11 — 0.408 0 .067 242 142 E — 9 1.73 0 . 1 5 1.74 0 .293 0 .043 — 200 12TH — 14TH L E A V E S Green Palms — 7 2 .30 0 .18 1.40 0 .535 0 .084 192 142 Yellow Palms Treatment B — 17 1.08 0 .17 1.59 0 .386 0 .016 275 108 Chemical analysis of the samples for N . P . K . Ca. Mg. Fe and Mn. are given in Table HI (b). The results do no t show any marked differences between the yellow and green palms. 24 (c) Analysis of nut waterfront trials on yellowing palms. N u t water from the trials a t Mat tegoda and Walgama Estates were analysed for nitrogen, phos­ phoric acid and potash. Again, the result failed to show any noticeable differences between the green and the yellow palms. T h e investigations on the problem of yellowing have hitherto not given any conclusive results. Fur ther work is being carried out . (IV) Studies on the effect of drying on the nutrient status of soils. The object of this study is to determine the changes which occur in the availability of soil nutrients when soils are subject to drying. As a preliminary step, investigations were made to find out whether any changes occurred in the exchangeable potash content and the potash concentrat ions in 10"- mola. calcium chloride extracts of the soil when a sample of soil from Bandirippuwa Estate was subject to varying degrees of drying. lOgm. samples of the fresh soil from the field were shaken u p overnight in ISOmls. 10 ' 2 M . Ca. C l^ . The potash content of the solution was determined after filtration. lOgm. samples of the d a m p soil were subject to similar extraction after air drying, and oven drying. The exchangeable potash in the fresh air dry and oven dry samples were determined by leaching 25gms. soil (weighed fresh) with 500cc. N . ammonium acetate. From the figures given in Table IV it is seen that while the drying has caused a distinct rise in the potash content in 10~2 M. Ca . C I . 2 extracts of the soil, it has made no changes in the exchangeable potash, as indicated by the concentration of ICO in the ammonium acetate leachates. Schofield has suggested that the potash status of a soil can be better judged by finding the equi­ librium concentration of potassium ions in a 10" 2 molar Ca. C I . extract of the soil than from its total exchangeable potash. (Schofield—Proceedings of the 1 Ith International Congress of Pure and applied Chemistry, July 1947, 3 , 257—261). These results, together with the observations of Paltridge and San- thiresegaram (C.R.I . Bulletin N o . I I ) on the increased availability of soil potash to plants on drying, support Schofield's hypothesis. TABLE IV Effect of drying on soil potash (Bandirippuwa Soil) ppm. K.JO in Solution N. Ammonium acetate 10" 2 Af. Ca. C. 2 extract leachate Fresh Soil 2 3 2 . 8 0 3 .00 2 .95 1.60 1.60. 1.65 Air dry Soil I 2 3 4 . 3 5 4 . 5 5 4 . 5 5 1.65 1.65 1.80 Oven dry Soil 2 3 4 . 8 0 4 . 8 0 4 . 6 5 1.65 1.65 1.60 25 (V) Chemical analysis of soil samples taken from Methods of Application Experiment (Marandawila). This work was commenced to study the effects of fertiliser placement on the distribution of nutrients in the soil. The following analysis were completed on row and manure circle top and sub soil samples taken from all the experimental plots before the first differential manur ing was done in 19S0:— (i) Tota l exchangeable bases. (ii) Exchangeable potash. (iii) Tota l nitrogen. (iv) T ruog phosphate . (v) p H . (vi) Mechanical analysis (on soil samples from 12 plots). (VI) M I S C E L L A N E O U S (o) T w o series of pasture samples (a total of 30) from the Agrostology Division were analysed for nitrogen, total protein, potash and moisture content . (b) The total exchangeable bases, exchangeable calcium and magnesium were determined in some soil samples from Vannankerney Estate, Pallai. T h e Soil Chemist delivered the following papers during the course of the year :— ( 0 " T h e use of Radio-active Isotopes in Agricultural R e s e a r c h " to theChi law-Negombo Planters ' Association. (ii) " Some new ideas on the availability of soil phosphorus to plants " to the Chemical Society of Ceylon. (/II) " Investigations on manur ia l and cultivation problems of the Coconut Palm " — a t the C.R.I . Technical Conference. D . A . N E T H S I N G H E , Soil Chemist. 26 REPORT OF THE CHEMIST 1. Distribution of Macro-nutrients in the Components of the Coconut Apical Bud. I t is reckoned tha t before initiating studies on the technique of foliar diagnosis of the coconut pa lm it is impor tant to determine the distribution of the macro-nutrients in the components of the tender apical bud. I t is always possible to distinguish and isolate the following rudimentary components in the tender bud :— (a) Stem (b) Petioles (c) Leaflets a n d (d) Flower Spathes. Representative bulk samples of these four components were carefully drawn from 25 apical buds. These samples were then dried in a dehydra tor a n d the dried mater ial analysed. The analytical results obtained for the macro-nutrients nitrogen, phosphorus , potassium, calcium and magnesium are presented in Table I . Parallel analyses for the macro-nutr ients were also carried ou t on twelve composite samples off apical buds of comparable size and age, carefully drawn from twelve palms in normal growth and vigour. T h e analytical results are summarised in Table IT. Some figures have been reported earlier for the com­ position of composi te samples of apical buds , bu t those unlike the present ones were drawn at r andom from palms of different ages and vigour. Fur ther , the earlier samples were analysed only for nitrogen, phosphorus and potassium. TABLE I Analytical Data on the Components of the Coconut Apical Bud. (Powdered air-dry samples) I 2 3 4 5 6 7 8 Compo­ nent % Mois­ ture % Nitrogen (asN) % Phosphorus (as Pid) % Potassium (as KiO) % Calcium (as Ca) % Magnesium (as Mg) % Ash (Sulphated) Compo­ nent % Mois­ ture Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry Stem 11.93 2.49 2.83 1.28 1.45 3.74 4.25 0.38 0.43 0.63 0.72 11.11 12.61 Petiole 11.86 2.47 2.80 1.12 1.28 3.13 3.55 0.27 0.31 0.70 0.80 10.98 12.40 Leaflet 10.37 4.72 5.27 2.00 2.23 3.56 4.61 0.34 0.38 0.53 0.60 11.44 12.76 Spathe (1) 16.58 2.66 3.18 1.45 1.74 2.94 3.53 0.65 0.78 0.65 0.77 12.21 14.64 Spathe (2) 16.06 2.68 3.. 19 1.45 1.72 2.81 3.34 0.64 0.76 0.65 0.77 12.33 14.69 27 TABLE n Analytical Data on Composite Samples of Coconut Apical Bud. {Powdered air-dry samples) 1 2 3 4 5 6 7 8 Sample % Mois­ ture % Nitrogen (as N) % Phosphorus (as P*Oh) % Potassium (as K20) % Calcium (as Ca) % Magnesium (as Mg) V /o Ash (Sulphated) Sample % Mois­ ture Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry 1 9.95 2.59 2.88 1.53 1.70 2.88 3.20 0.350 0.389 0.637 0.707 12.12 13.45 2 10.19 3.51 3.91 1.62 1.81 2.87 3.19 0.410 0.457 0.635 0.707 12.20 13.59 3 12.86 2.60 2.98 1.57 1.80 2.43 2.79 0.523 0.600 0.562 0.646 10.19 11.69 4 15.56 2.90 3.43 1.35 1.59 2.64 3.13 0.338 0.400 0.555 0.657 9.16 10.86 5 13.51 2.88 3.33 1.52 1.75 1.97 2.28 0.420 0.485 0.550 0.636 8.48 9.80 6 9.94 2.89 3.21 1.69 1.87 2.25 2.50 0.615 0.682 0.611 0.678 9.83 10.91 7 17.67 2.72 3.30 1.32 1.60 2.16 2.62 0.333 0.405 0.585 0.710 9.30 11.29 8 18.60 2.63 3.23 1.27 1.57 2.01 2.47 0.350 0.430 0.568 0.698 8.70 10.68 9 18.67 2.61 3.21 1.31 1.60 1.66 2.04 0.416 0.512 0.564 0.693 8.46 10.40 10 14.26 2.66 3.10 1.39 1.62 2.93 3.41 0.327 0.382 0.585 0.682 10.19 11.88 11 12.66 3.12 3.56 1.85 2.13 2.41 2.76 0.546 0.625 0.611 0.699 11.03 12.64 12 14.96 2.62 3.08 1.29 1.52 3.07 3.61 0.322 0.378 0.493 0.579 10.15 11.94 Mean 14.07 2.81 3.27 1.48 1.71 2.44 2.83 0.412 0.479 0.580 0.674 9.98 11.59 Range 9.94 to 18.67 2.59 to 3.51 2.88 to 3.91 1.27 to 1.85 1.52 (0 2.13 1.66 to 3.07 2.04 to 3.61 0.322 to 0.615 0.378 to 0.682 0.493 to 0.637 0.579J 8.46 to j to i 0.710;12.20 9.80 to 13.59 S.D. 3.17 0.28 0.28 0.18 0.17 0.45 0.48 0.098 0.105 0.040 0.039J 1.28 1.18 10.18 CV. (%) 22.53 9.96 8.56 12.16 .9.94 18.44 16.96 23.79 21.92 6.90 5.79 12.83 1.18 10.18 S.E. 0.92 0.08 0.08 0.05 0.05 0.13 0.14 0.028 0.030l 0.012 O.OIll 0.37 0.34 S.D. = Standard Devisution S.E. = Standard Error CV. = Co-efficient of Variation The results charted in Tables I and II reveal the interesting fact that the distribution of the mineral consti tuents in the various components of the tender bud is not uniform. In other words, even during the early phases of development of the bud there are centres of accumulation of particular nutrients. It would doubtless be interesting to follow how the respective concentrations change during progressive stages of development of the different components of the bud and the unfurling of the fronds. It is proposed to carry out these studies prior to the application of the technique of foliar diagnosis, as it would afford a better understanding of the pattern of distribution of the macro-nutrients in the pro­ gressive developmental stages of the fronds and suspensory elements in the crown of the coconut palm. 28 2. Pot Culture Experiment. It has been reported last year that a preliminary sand pot-culture experiment using a subtraGtive intermittent flowing technique has been laid down. The design and experimental procedure has already been described. Fur ther particulars of this experiment a re as follows :— Crop from which nuts were drawn — Bandir ippuwa May/June 1958. Size g roup — 6 to 8 inches (short axis) — random. Selected drupes planted in Botanist 's nursery — 1,000. D a t e of planting in nursery — 20th June 1958. Number uprooted from nursery — 850. D a t e of uproot ing and husking — 11th August 1958. Seedlings damaged during h u s k i n g — 123. Rejections (oversize and undersize sprouts) — 294. Seedlings planted in pots (typical range) — 288. Seedlings (typical range) used for initial analyses — 145. Seedlings planted per pot — 36. N O T E — O u t of the 150 seedlings left behind in the nursery, 36 seedlings (comparable in physical characteristics and representative of the 36 planted in each pot) were selected and tagged to serve as a control for the experiment. The above experiment has been maintained satisfactorily during the year. Growth measurements were recorded a t monthly intervals on all seedlings. F o u r seedlings (two amputated and two non-amputated) were uprooted each mon th from every pot and the various components were sampled for analysis after recording weight characteristics. As the organization and maintenance of this experiment has involved a prodigious amount of mani­ pulative work, it has not been possible to clear up the analysis of all the samples prepared. Average Analytical and Weight Characters of the Experimental Seedlings, on the day of planting in the pots (11-8-58). The 145 selected seedlings representative of the 288 planted in the pots were carefully examined for the analytical and weight characters of the different components , viz. husk, shell, kernel, nut water, haustorium, shoot and roots. After weighing the components of each seedling individually, bulk samples of the respective compo­ nents were prepared by combining the 145 individual samples. After triplicate moisture estimations on each lot, the bulk samples were dried in a dehydrator , were ground u p and then analysed for N . P . K . C a . and Mg. The results obtained in this study are charted in Tables 111 and 1V. It is reckoned that the average figures represent the analytical and weight characteristics of the seedlings on the day of planting in the pots. N O T E — (It should be mentioned that though the husks were not removed from the 36 control seed, lings in the nursery the 288 seedlings in the pots were planted without the husks), 29 TABLE III Moisture Contents and Average Weights of the Component Parts of the Seedling (on day of planting in the pots —11-8-58). 1 2 3 4 Component % Moisture Average Wt. of Component 1 Nut Dry Weight of Component as % of Total Dry Matter/ Seedling Component % Moisture Wet Weight (Gms.) Dry Weight (Gms.) Dry Weight of Component as % of Total Dry Matter/ Seedling Husk 57.16 782 .0 335.0 51 .7 Shell 17.01 179.7 149.1 2 3 . 0 Kernel (by diff.) 41 .08 237.4 139.9 2 1 . 6 N u t water 96 .23 16.4 0 . 6 0 .1 Cotyledon 86 .45 120.5 16.3 2 . 5 Shoot 84 .83 37 .8 5 .7 0 . 9 Roo t s 82 .44 9 .2 1.6 0 . 2 Whole seedling 53.13 1,383.0 648 .2 100.0 Seedling minus husk 47 .89 601 .0 313 .2 4 8 . 3 TABLE IV Distribution of Mineral Nutrients in the Components of the Seeding on 11-8-58 (on the powdered samples as analysed) 1 2 3 4 5 6 7 8 9 Compo­ nent % Ash (Sulphated) % Nitrogen (as N) % Phosphorus (as P) % Potassium (as K) % Calcium (as Ca) 0/ Magnesium (as Mg) % Mois­ ture V /o Dry Matter Compo­ nent Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry % Mois­ ture V /o Dry Matter Husk 3.85 4.37 0.425 0.483 0.017 0.020 0.557 0.632 0.117 0.133 0.098 0.112 11.87 88.13 Shell 0.666 0.743 0.203 0.224 0.006 0.006 0.105 0.115 0.022 0.024 0.080 0.089 9.69 90.31 Kernel 1.56 2.64 0.83 1.41 0.131 0.223 0.307 0.515 0.011 0.018 0.064 0.108 41.08 58.92 Nut Water 0.496 13.16 0.009 0.239 0.008 0.198 0.074 1.95 0.019 0.512 0.014 0.366 96.23 3.77 Cotyle­ don 7.51 9.05 1.15 1.39 0.342 0.412 1.83 2.21 0.175 0.211 0.196 0.236 16.99 83.01 Shoot 7.93 9.35 1.53 1.81 0.250 0.295 2.04 2.41 0.107 0.126 0.284 0.334 15.15 84.85 Roots 8.82 10.57 1.23 1.48 0.253 0.303 1.98 2.37 0.057 0.068 0.206 0.247 16.52 83.48 T h e results given in Table IV for the kernel are calculated figures based on the actual analytical da ta obtained on the oil-free meal. The da ta obtained on the oil-free meal and the corresponding figures calculated for the original kernel are shown in Table V. 30 TABLE V Analytical Da ta on the Extracted Mea l and Calculated Da ta for the Kernel. 1 2 3 4 5 6 7 8 9 Compo­ nent % Ash % Nitrogen % Phosphorus % Potassium % Calcium % Magnesium '/.Mois­ ture Matter Compo­ nent Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry '/.Mois­ ture Matter Oil-free Meal 7.57 8.71 4.05 4.66 0.647 0.743 1.49 1.71 0.051 0.059 0.309 0.356 13.09 86.91 Kernel (Calcu­ lated) 1.56 2.64 0.83 1.41 0.131 0.223 0.307 0.515 0.011 0.018 0.064 0.108 41.08 58.92 F r o m the da ta presented in Tables III and IV it should now be possible to compute average values for the actual amounts of the macro-nutrients N . P . K . Ca and M g present in each of the components of the seedlings a t the t ime of planting in the experimental pots . In other words, we could obtain a quantitative assessment of the mineral reserves with which the seedlings start off a t the commencement of the experiment, before they are subjected to the different treatments. T h e calculated da ta are given in Table VI bellow. TABLE V I Distiib'Jtion of the Nutrients, in the Various Components of the Seedlings. (a) Grammes of Nutrient per Component and Seedling. 1 2 3 4 5 6 7 8 9 Nutrient Husk Shell Kernel Nut Water Shoot Roots Coty­ ledon Whole Seedling Ash 14.64 1.11 3.69 0.079 0.533 0.169 1.48 21.7010 Nitrogen (asN) 1.62 0.334 1.97 0.0014 0.103 0.0237 0.227 4.2791 Phosphorus (asP) 0.067 0.009 0.312 0.0012 0.017 0.0048 0.0672 0.4782 Potassium (asK) 2.12 0.171 0.720 0.0117 0.137 0.0379 0.360 3.5576 Calcium (asCa) 0.446 0.036 0.025 0.0031 0.0072 0.0011 0.0344 0.5528 Magnesium (asMg) 0.375 0.133 0.151 0.0022 0.0190 0.0040 0.0385 0.7227 (b) Nutrients in the Components as a Percentage of Nutrients/Seedling. Ash | 67.4 5.1 17.0 0.4 2.5 0.8 6.8 100.0 Nitrogen (as N) 37.9 7.8 46.0 0.1 2 .4 0.5 5.3 100.0 Phosphorus (as P) 14.0 1.9 65.2 0.3 3.6 1.0 14.0 100.0 Potassium (as K) 59.6 4.8 20.2 0.3 3.9 1.1 10.0 100.0 Calcium (as Ca) 80.7 6.5 4.5 0.6 1.3 0.2 6.2 100.0 Magnesium (as Mg) 51.9 18.4 20.9 0.3 2.6 0.6 5.3 100.0 Wet Weight (per Seedling) 782.0 179.7 237.4 16.4 37.8 9 .2 120.5 1383.0 Dry Weight (per Seedling) 335.0 149.1 139.9 0.6 5.7 1.6 16.3 648.2 31 3633—C On the basis of the results summarised in the above tabulations, we could now arrive a t the following figures for the actual quantities of the different macro-nutrients (expressed in order of decreasing magni­ tude) present per seedling (with and without husk) a t the initial stage of the experiment. Macro-Nutrients per Seedling Nutrient Seedling with husk (as planted in nursery) grammes per seedling Seedling without husk (as planted in pots) grammes per seedling Ash (Sulphated) 21 .70 7 .06 Nitrogen as N 4 .28 2 .66 f a s K Pr\ tacci i im J _ 3.56 1.44 F V J i u a o i u i i i s (^as K , 0 4 . 2 9 1.73 f as Mg K if n nnAnii m-% J 0 .72 0 .35 1V1 d e l 1CS1U111 s l̂ as M g O 1.20 0 .58 f as Ca 0 .55 0.11 l̂ as Ca .O 0 .77 0 .15 f a s P 0 .48 0.41 I as P.O.-, 1.09 0 .94 It can be said that the above quantities represent the actual nutrient reserves available to the growing seedling. It will be seen that the amounts of nitrogen and potassium present are comparatively high. It is also noteworthy that in the whole seedling the amounts of magnesium and calcium are higher than the phosphorus present, when calculated as Mg. Ca. and P. respectively. 3. Copra from varieties grown in Ceylon. The typica o r tall coconut palm which is cultivated on a plantation scale represents the variety that is propagated essentially for commercial purposes. Besides this there are also other varieties and forms which are grown to a limited extent in the island. The economic possibilities of evolving hybrids (with high oil contents in the kernel) by selective breeding is no doubt an interesting speculation well worth investigation. Fifteen fresh samples of copra (uncomplicated by the changes accompanying deterioration and germination) sent by the Botanist were carefully examined and analysed for moisture and oil contents. The results obtained in this study are presented in Table VII. It will at once be apparent from the analy­ tical results that the dwarf green ( 6 9 . 9 % ) , G o n Thambili (69 .2%) , Bodiri (69 .6 % ) , Nawasi (69 .5%) and Pora Pol ( 6 9 . 7 % ) , are the only ones if a t all which have given somewhat higher figures than the commercial tall palm for oil content. 32 T A B L E VII (Oil content of the varieties and forms of Coconut grown in Ceylon). 1 2 3 4 5 6 7 8 Sample Variety or Form % Mois­ ture %Oil No. of Nuis Wl. of copra lbs. Out­ turn nutsi candy Copra Quality Sample Variety or Form % Mois­ ture Wet basis Dry basis No. of Nuis Wl. of copra lbs. Out­ turn nutsi candy Copra Quality 1 Tall palm (compo­ site) (Variety—TY­ PICA) 6.8 63.7 68.3 2600 1133.96 1284 Uniformly good 2 Dwarf Green (Var­ i e ty—NANA) 6.2 65.6 69.9 102 24.00 2380 Fair (not uni­ form) 3 Dwarf Red (a form of N A N A ) 6.8 60.8 65.2 90 17.25 2922 Inferior 4 Dwarf Yellow (a form of N A N A ) 7.1 60.8 65.5 100 24.75 2263 Inferior 5 King Coconut (Variety) 7.4 60.8 65.6 100 31.25 1792 Fair 6 Ran Thambili (a form of TYPICA) 7.5 63.3 68.5 19 9.25 1150 Good 7 Gon Thambili (a form of TYPICA) 6.6 64.6 69.2 55 28.00 1100 Good 8 Bodiri (a form of TYPICA) 7.2 64.6 69.6 20 2.25 4978 Inferior 9 San Ramon (a form of TYPICA) 8.0 60.4 65.6 66 42.50 870 Good 10 Kamandala Giant (a form of TYPICA) 7.2 62.8 67.7 45 37.50 672 Good 11 Nawasi (a form of TYPICA) 6.9 64.7 69.5 19 7.50 1419 Uniformly good 12 Nawasi Thambili (a form of TYPICA) 7.0 63.5 68.1 20 6.39 1753 Fair 13 Pora Pol (a form of TYPICA) 7.3 64.6 69.7 20 5.95 1882 Fair 14 Tall Palm (Green) (Variety—TYPICA) 6.6 64.4 68.9 102 45.19 1264 Uniformly good IS Tall Palm (Red­ dish Brown) (Var­ iety—TYPICA) —(1 Candy — 560 lb). 7.1 63.9 68.8 102 41.67 1371 Good 4. Moisture content of the Coconut Kernel. It is generally reckoned, that in the fresh state, the kernel of the coconut contains about SO per cent of moisture. Though this figure may be employed for rough and ready calculations it is felt that the actual moisture content could vary with the maturi ty of the drupe itself, and also perhaps condit ions of storage. Further , it is likely that there is a moisture gradient within the kernel itself. Whether o r not this is true can only be established on the basis of proper quantitative experiments. 33 I t is hoped tha t in connection with certain proposed studies on quality factors of copra that all aspects of moisture in the kernel will be exhaustively investigated. In this context (certain) preliminary quanti tat ive estimations of moisture content of the kernel have been made on certain carefully drawn samples. T h e results are presented in Table VIII below. TABLE VTII Moisture Content of the Coconut Endosperm. 1 2 3 4 Picked ripe Green Nuts (Immediately on harvesting) Fallen nuts (Dead ripe nuts naturally falling from the palm) Picked Ripe Green Nuts Sample Picked ripe Green Nuts (Immediately on harvesting) Fallen nuts (Dead ripe nuts naturally falling from the palm) After 15 days seasoning on the field After 30 days seasoning on the field 1 44 .44 43 .93 43 .99 45 .92 2 44 .48 41 .98 43 .34 44 .98 3 4 3 . 5 0 42 .91 4 4 . 0 0 4 5 . 7 6 4 44 .54 43 .87 42 .55 44 .58 5 4 3 . 6 6 43 .48 43 .80 43 .58 6 43 .18 34.22 43 .46 44 .88 7 43 .74 43 .52 44.51 43 .74 8 43 .88 43 .62 44 .11 42 .70 9 43 .17 44 .38 43 .50 42 .50 10 43 .65 43 .62 43 .04 43 .73 Mean 43 .82 43 .45 43 .63 44 .24 S.D. 0 .5096 0.6532 0 .5720 1.1812 C V . (%) 1.16 1.50 1.31 2 .67 S.E. 0 .16 0-21 0 .18 0 .37 F o r each category of nuts in the experiment 250 nuts were taken and these were sampled and analysed for moisture content in ten lots of 25 nuts each. O n the basis of these studies it could be concluded that there is no significant difference between the moisture contents of the kernel from harvested ripe green nuts and that from dead ripe nuts naturally falling from the palm. I t could also be assumed that the moisture content in the freshly harvested nut does no t change appreciably on seasoning for periods of 15 or 30 days on the field. 34 5 . Changes in the Coconut Endosperm during Development. When we consider the floral biology and development of the coconut , it should be possible to identify a t least 25 distinct stages from the t ime the rudimentary inflorescence makes its appearance unti l the cluster of fully ripe nuts is ready for harvesting. The kernel (endosperm) first makes its appearance as a jelly like substance a t stage 18 in this series. This would correspond to an age of ab o u t 5 months from the t ime the flower spathe burs t open in to flowers. At stage 25 the kernel would be fully formed and mature and thus ideally suited for the preparat ion of quality copra. The age of the ripe d rupe corres­ ponding to stage 25 would be abou t 12 — 1 3 months from the time of opening of the spathe. The changes taking place in the oil and moisture contents of the endosperm were followed during the course of its progressive development between stages 20 and 25. I t has no t been possible to accumulate sufficient quantities of the dried endosperm a t stages 18 a n d 19 for analysis. I t is hoped that it would be possible to carry ou t the analysis when sufficient material becomes available. The results so far obtained on the average oil and moisture contents in the kernel during the course of its progressive development between stages 20 and 25 are chartered in Table I X below. Parallel da ta are also given showing the total wet and dry weights of endosperm per drupe. The results reveal the following interesting features:— TABLE IX Changes in the Moisture and Oil Contents of the Coconut Endosperm during Progressive Stages of Development. 1 2 3 4 5 Stage Total Moisture in Endos­ perm Weight of Endosperm per nut {grammes) %Oil Description Stage Total Moisture in Endos­ perm Weight of Endosperm per nut {grammes) Wet basis Dry basis Description Stage Total Moisture in Endos­ perm Wet Dry Wet basis Dry basis Description 1 — 17 No En­ dosperm N o pulp 18 — — — — — — 19 — — — — — — 20 9 4 . 6 22 1.2 0 . 9 16 .0 Very Tender K u - rumba 21 8 3 . 9 111 16 .9 7 . 7 4 7 . 5 Tender K u r u m b a 22 7 4 . 3 165 4 2 . 2 14 .8 5 7 . 3 K u r u m b a 23 6 5 . 2 216 7 5 . 0 2 3 . 2 6 6 . 6 Tender Kala t i 24 4 9 . 0 260 132.0 36 .1 70 .7 Kala t i 25 4 2 . 5 273 161.4 4 1 . 7 6 8 . 1 Ripe Green N u t (a) T h a t the moisture content steadily d rops from abou t 95 per cent in the very tender gelatinous kernel t o abou t 43 per cent in the ripe fruit. 35 (h) T h a t the oil contents (dry basis) unlike the moisture increases with the ripening of the fruit. The figures recorded show an increase from 16 per cent to 68 per cent. The high value (70.7 per cent) obtained in this study for stage 24 is rather interesting. (c) T h a t there is a progressive increase in the deposition of dry mat ter in the kernel — the results showing an increase from 1 gramme to 161 grammes between stages 20 and 25. 6. Changes in the Coconut Endosperm during Germination. I t has been reported earlier that a simple ad hoc germination experiment has been laid down to ascertain quantitatively the mineral reserves in the fruit components and also the changes in the compo­ sition of the leaves of the seedlings during progressive stages of germination. It should be appropriate to refer to certain observations made in this experiment on the changes that take place in the endosperm of the coconut during germination. These observations are reckoned to be of some significance since certain commercial grades of copra invariably include the kernel from germinated coconuts. The analytical data reviewed in Table X below represent the changes that take place in the kernel during the progress of germination u p to a period of 1\ months from planting. The distinctive features of the results presented in the tabulation may be summarised as follows:— (a) There is a decided tendency for a d rop in the moisture content of the kernel during germination. This becomes particularly marked after 20 weeks from planting, from which stage onwards there is a steady fall in moisture from 4 1 . 9 per cent to 33 .4 per cent. (h) Unlike moisture, the tendency for changes in the oil content is definitely in the reverse direction. It will be seen that after 20 weeks this becomes very marked — the oil percentage steadily rising from 72 .9 per cent to 7 6 . 7 per cent (dry basis). TABLE X Changes in the Moisture and Oil Content of the Coconut Endosperm during Germination. 1 2 3 4 5 No. of weeks in nursery % Total moisture in kernel Weight of kernel per nut (grammes) Weight of oiljnut (grammes) % on Wet Dry Wet basis Dry basis Nil 2 4 6 8 10 12 14 16 18 20 22 24 26 28 3 0 ( 7 . 5 months) 43 .7 4 3 . 2 45 .1 4 3 . 3 42 .1 4 2 . 3 4 2 . 9 4 2 . 6 4 2 . 2 4 1 . 5 4 1 . 9 40 .1 4 0 . 0 39 .2 37 .5 3 3 . 4 345 346 351 339 350 343 344 346 336 324 303 289 281 271 242 216 194 197 193 192 203 198 196 199 194 190 176 173 169 165 151 144 133.3 134.2 131.8 134.4 139.9 138.8 135.8 139.3 137.0 133.8 128.3 127.7 124.9 123.6 115.1 110.4 38 .7 38 .7 37 .5 39 .7 39 .9 4 0 . 4 3 9 . 6 4 0 . 2 4 0 . 8 4 1 . 2 4 2 . 4 4 4 . 2 44 .3 4 5 . 5 4 7 . 6 51.1 68 .7 68.1 68 .3 7 0 . 0 68 .9 70.1 69 .3 7 0 . 0 7 0 . 6 70 .4 72 .9 73 .8 73 .9 74 .9 76 .2 76 .7 36 (c) Regarding the question of actual oil recoveries per nut it will be seen from column 4, that there is a persistent d rop after 18 weeks from 133.8 grammes to 110.4 grammes. (d) Though the oil contents are high when expressed as a percentage, it will be seen that these high values actually correspond to lower recoveries of oil per nut . The inference can therefore be drawn that from the commercial point of view, on a weight basis the low grade copra prepared from well germinated coconuts will doubtless yield more oil on expression. Though this will be of value to the oil miller, yet it should be remembered tha t the quality of oil and also the poonac recovered is inferior and should entail refinement losses. F r o m the point of view of production economics however there is n o significance or advantage in preparing copra from germinated coconuts because the quanti ty of oil recovered per nu t is below average. T h a t this is so will be evident when we consider the fact that the average oil recovery per nut during the first 18 weeks of planting is 136 grammes and for the period 20-30 weeks is only 122 grammes. (Computed from Table X , Column 4). 7. Oil and Moisture Gradients in the Coconut Endosperm. In order to find a plausible explanation for the observations made in the germination experiment, an investigation was carried out to test the presence of moisture and oil gradients in the coconut kernel. The figures presented in Table X I below demonstrate conclusively that there are to be found in the coconut kernel definite oil and moisture gradients when sliced tangentially, parallel to the testa or brown integument. TABLE X I Oil and Moisture Gradients in the Coconut Endosperm. Region % Moisture % on {Wet basis) % Oil (Dry basis) T 29.1 4 4 . 2 6 2 . 4 R T 32 .9 5 0 . 6 7 5 . 4 I R 44 .1 37 .9 67 .8 RVV 59 .3 22 .9 56 .3 T = Testa I R = I n t e r m e d i a t e region R T = Region near testa R W = R e g i o n near water cavity Fo r the experiment, the fresh kernels from ungerminated coconuts alone were used. These were split into half cups in the usual way and 1/8" radial strips were taken from each cup by making longi­ tudinal cuts. Each of these strips was then sliced parallel to the testa so that besides the parings three sections of equal thickness were obtained from the white meat. The corresponding sections from each strip were then bulked, and analysed for moisture and oil content. The results show that the section least rich in oil (56 .3 per cent) is on the inside nearest the water cavity. The section richest in oil (75 .4 per cent) is nearest the testa, and the intermediate section fits in between with an oil content of 6 7 . 8 per cent. The testa itself contains 6 2 . 4 per cent oil. It will be seen that the moisture gradient is in the reverse direction. 37 With a knowledge of these facts it is indeed easy to understand the changes that have been observed to take place in the kernel during germination. In germination it would appear that the inside tissues least r ich in oil progressively break down as the haustor ium (or apple) develops leaving the layers which are richer in oil. Tha t this is plausible will be seen from the fact that the oil content of 7 6 . 7 per cent recorded for the very thin kernel wafer from drupes that have been in the nursery for 7£ months is not far different from 7 5 . 4 per cent the oil content registered for the region of the kernel nearest the testa. The explanation for the d rop in moisture content of the kernel with the progress of germination is similarly provided by the fact that the tissues on the inside which break down first are richer in moisture (59 .3 per cent) than the residual layers nearer the testa (29.1 per cent). 8. Moisture Changes during Copra Drying in the Standard Ceylon Dryer A n experiment was carried ou t to study the moisture changes in the coconut kernel when dried in the s tandard Ceylon Dryer in accordance with the operational schedule outlined in C.R.I . Leaflet N o . 15. Table X I I below gives a summary of the figures obtained during successive stages when the copra was cured in the form of (a) cut kernels and (b) half cups. TABLE XII Moisture Changes in Copra during Curing Procedure in Ceylon Kiln. 1 2 3 4 5 Stage No. ofhrs. since splitting % Moisture Total moisture lost since splitting Total moisture lost as % of original Stage No. ofhrs. since splitting Cut Kernels Half Cups Cut Kernels Half Cups Cut Kernels Half Cups O n splitting nil 4 3 . 5 4 3 . 3 nil nil nil nil After Sun- drying 10 38 .0 37 .8 5 .5 5 .5 12.6 12.7 After 1st Fire 19 25 .7 24 .8 17.8 18.5 40 .9 4 2 . 7 After 2nd Fire 34 17.1 16 .4 2 6 . 4 26 .9 60 .7 62 .1 After 3rd Fi re 43 11.9 10 .0 31 .6 33 .3 72 .6 76 .9 After 4 th Fi re 58 10.1 8 .0 33 .4 35 .3 76 .8 81 .5 After 5th Fire 67 8.5 7 . 4 35 .0 35 .9 80 .5 82 .9 After 6th Fi re 82 8.4 6 .5 35.1 36 .8 80 .7 85 .0 After 7th Fire 91 7 .8 6 .0 35 .7 37 .3 82 .1 86 .1 After 8th Fire 106 6 .6 5 . 4 36.9 3 7 . 9 84 .8 87 .5 T h e results show tha t under climatic conditions prevailing in Ceylon only about 12 percent of the total moisture is lost during 10 hours of initial sundrying. 38 Regarding essential drying principles it is generally agreed t h a t : — (1) The moisture content of the kernel has to be reduced from SO per cent to 35 per cent pre­ ferably within 24 hours . (2) Dur ing the second 24 hours the moisture content should be reduced to abou t 20 per cent. (3) In the next 24 hours the moisture content should be reduced to 5 to 6 per cent. I t will be seen from Table X I I column 3 tha t for the Ceylon Dryer the above requirements a re more than fulfilled during the first 48 hours which may be regarded as the critical period of drying. After 48 hours the dryage is definitely slow in terms of the requirements. I t would take a t least 96 hours for the moisture to be reduced to 5 . 6 per cent in the Ceylon kiln as against the limit of 72 hours . I t may be said tha t from the point of view of copra quality the slower drying towards the end is a desirable feature. When the bulk of the moisture in the kernel has been expelled, overheating would definitely tend to caramelize the sugars with resultant discolouration, decomposition and also hardening of the meat. The principal virtue in the Ceylon kiln is its adaptabili ty to both small and large scale processing. I t can be said that there should be n o difficulty in producing fairly uniform high grade copra with it, provided the s tandard operational procedure is rigidly adhered to . 9 . Experiment on Copra Storage. Consequent on a request m a d e by the F .A.O. , Working Party on Copra Quality and Grading, an experiment has been laid down to estimate economic losses to the producer resulting from loss of anhydrous copra dur ing storage due to mould action, insect at tack and general deterioration. The results of the experiment will be reported later on. 10. Miscellaneous Work. Analyses a n d reports were m a d e on a few samples of copra, poonac and vinegar. W . R . N . N A T H A N A E L , Chemist, Coconut Research Institute. 39 REPORT OF THE BOTANIST 1. Hybridisation between Varieties and Forms. Typica x Nana.—The performance of the first generation (F , ) palms of typica X nana (i.e., the tall variety x dwarf variety) during the 5th to 8th years after planting has been summarised in the Annual Repor t for 1958. Dur ing the year under review i.e., the 9th year, these palms have continued to give good yields (Table I). The mean yield per progeny per year from the 6th to the 9th year has been 68, 86, 65 and 103 nuts or 34, 39, 35 and 49 lb. copra respectively. A mean yield of 103 nuts equivalent to 49 lb. copra per palm in the 9th year after planting is indeed very satisfactory. TABLE 1 Mean yield per progeny of typica X nana in the 9th year Weight of Weight per Cross No. of No. of husked-nuts husked-nttt progenies nuts (lb.) (lb.) 360 X 1713 . . 4 97 144.0 1.48 218 X 1713 . . 5 104 203.8 1.64 360 X 1712 . . 5 112 168.7 1.50 139 X 2646 . . 4 83 116.4 1.41 778 X D 3 73 82 .7 1.13 273 X 2646 . . 1 144 198.5 1.38 Mean per pro­ geny — 103 147.0 1.48 It has been pointed ou t previously, that considerable and significant variations exist between families with at least 4 progenies each. In 9 year old palms yield has varied between 65 and 37 lb. of copra per palm. Thus in any programme to utilise vigour of F , palms of typica and nana, it is necessary to take cognizance of the variation between families and isolate as far as possible desirable combinations for seed multiplication. With that end in view, a crossing trial using 25 selected nana palms and 5 selected typica pa lms was carried out during the year. The main purpose of this trial is to identify nana palms that show general combining ability: T o study the performance of typica X nana hybrids in the different coconut growing districts of Ceylon, 70 plants including both hybrids and typica open pollinated were issued to planters in 1957 and 1958 to be planted in different parts of the country. These plants were examined during the year and their growth so far has been satisfactory. Pollinations for the product ion of seed was carried out at three stations and 9,800 female flowers were pollinated. 3,124 nuts from crosses done in 1958 were harvested and planted. 40 Typica x Typica.—An extensive p rogramme of paired crosses between the tall variety palms was carried ou t a t seven stations using selected palms as the female parents and a ' p r e p o t e n t ' palm as the male parent. Altogether 31,000 female flowers were pollinated. 4,500 nuts from the crosses done during the previous year were harvested and planted in the nursery. F o u r of the prepotent palms were selfed with a view to fix prepotency and also to continue their line since they are senile. A Pollen Bank has been developed with pollen collected from prepotent palms. Besides, using pol­ len samples for our crosses, 4,610 samples were issued to six private estates who are carrying ou t a pro­ gramme of controlled pollination for production of seed. These estates were further assisted by training personnel for pollination work and by checking work in progress whenever necessary. 2. Identification of Prepotent Pa lms . Dur ing the year, three more typica palms were provisionally classified as 'prepotent , ' bringing the total of such palms to seven. A progeny trial was planted a t Bandir ippuwa Esta te in December, 1959, to identify prepotent palms by the method of comparison of open-pollinated progenies. Seed-nuts of 125 selected high-yielding palms were planted in a nursery, and 9 seedlings taken a t r andom from each family were transplanted. The design of the trial is a cubic (5 3 ) lattice, with 3 progenies per plot and 3 replications. The progeny trial tha t was planted a t the Government F a r m , Walpi ta and which was subsequently abandoned due to various difficulties of management was handed back to the Insti tute in June 1959. Dur ing the year agricultural operat ions were under taken. to put the block in order . Yield recording of individual palms would be continued from January 1960. 3. Replantation. A field trial was initiated in 1950 to study the relative merits of three methods of underplanting senile coconut plantations, viz., (a) New clearing — planting after the removal of the old stand of palms completely, (b) Gradual thinning — under planting and removing the old stand of palms gradually, and (c) No thinning — under planting without removal of the old palms, the latter to be removed during the initial stages of bearing of the young palms. The design of the trial is a randomised block layout with 25 palms per plot and 7 replications. Under gradual thinning abou t 20 per cent of the old palms were removed before underplanting and thereafter 10 per cent every year. All the old palms in the no thinning treatment were uprooted at the end of the 8th year. With regard to leaf production and early flowering in the young palms, it has been shown that the three systems of underplanting could be placed in the following order of mer i t :—(i ) new clearing (if) gradual thinning and (Hi) no thinning (Annual Repor t for 1957). Flowering Period.—The flowering period of a palm has been taken as the period from transplanting a seedling to the emergence of the first spathe on the palm. The percentage of palms in flower, cumu­ lative for each year is given in Table 2. TABLE 2 Percentage of palms in flower in the Replantation Experiment Treatment Year after planting 5th 6//i 7//i 8//i 9l/i New clearing 12 .0 4 4 . 0 6 3 . 3 8 4 . 6 89 .7 Gradua l thinning 3 .4 25 .1 4 5 . 1 65 .7 74 .3 N o thinning 1.7 12.0 2 6 . 3 4 4 . 0 53.1 41 T h e variance ratio for each year is significant indicating that the type of treatment had an effect on the flowering period of the palm. The presence of the old stand of palms has retarded the growth of the second plantation considerably that only S3 per cent of the palms were in flower during the 9th year whereas over 74 per cent of the palms of the other two treatments have flowered. A n analysis of variance of the palms in flower in the 9th year is indicated in Table 3. TABLE 3 Analysis of variance of palms in flower in the 9th year Degrees of Sum of Mean freedom squares square F Blocks 6 152 25 .3 Treatments 2 295 147.5 8.38 *** Er ro r 12 211 17.6 Tota l 20 658 — ***Significant P < 0 .01 The new clearing and gradual thinning treatments have given significantly more palms in flower than in the no thinning treatment, bu t the differences between the first two treatments are not significant. Yield of Nuts.—Yield of nuts and copra during the 8th and 9th years after planting are given in Table 4. The yield of copra has been calculated as 32 per cent of the weight of husked-nuts. The ana­ lysis of variance for each factor for each year shows that the differences between treatments are statis­ tically significant for P < 0 . 0 5 . In the 8th year, palms in the new clearing treatment have given significantly more nuts and copra than those in the other two t rea tments ; differences between gradual thinning and n o thinning treat­ ments are no t significant, al though the former has given appreciably more nuts and copra than the latter (Table 4). TABLE 4 Yield of nuts and copra per acre in the Replantation Experiment 8th year 9th year Nuts Copra Nuts Copra (cwts.) (cwts.) New clearing 1,065 5.71 1,718 8.45 Gradua l thinning 538 2 .72 1,245 5.78 N o thinning 225 1.21 594 2 .97 Critical difference 521 2 .95 762 3 .96 In the 9th year the yield with respect to nuts and copra has been in the following order of merit :— (/) new clearing, (if) gradual thinning and (iii) n o thinning. The differences between treatments (i) and (ii), (if) and (Hi) a re no t significant, yet yields of (//') are appreciably more than that of (iii). The new clearing has given significantly more nuts and copra than the n o thinning treatment. 42 General.—On the basis of da ta collected from the replantation trial, certain provisional recom­ mendations could be made regarding underplanting of senile coconut plantations. The system of under- planting leaving the old s tand of palms completely and removing them when the young palms are 7 to 8 years old cannot be recommended as the presence of the old palms adversely affect the growth and flowering of the under plantation. Al though the new clearing method has shown the best results on an experimental scale, yet it cannot be recommended as the performance of the young palms relative to the gradual thinning method has no t been so remarkable so as to offset the loss in crop due to removal of the old s tand of palms completely. Thus , considering particularly the small holders, a feasible and efficient method of under planting is the gradual thinning system. The following procedure is provisionally recommended. After lining out the area to be underplanted, remove all the old palms lying within 6 feet of the new planting sites and the other unproductive palms. Under this category about 30 per cent of old stand of palms may have t o be removed. Thereafter, 20 per cent of the original stand of palms should be removed a t the end of 2nd, 4th and 6th years. At the second thinning of palms Le., at the end of the 2nd year, palms lying within 10 feet of the young palms should be uprooted. Subsequent thinnings could be done on a basis of yield of the old palms. All the remaining old palms should be removed a t the end of 8 years. 4. Planting Techniques. Size of Seed Hole—This field trial initiated in November 1955 is to study the relationship if any, between the size of seed hole used to transplant a coconut seedling and its subsequent growth. N u m b e r of leaves produced during the year and the total number of green leaves per plant were scored and the da ta are presented in Table 5. The analysis of variance indicates that there are no significant differences between the treatments with respect t o the two characters listed above. TABLE 5 Leaf production a t the end of the 4th year in the Size of Seed Hole Experiment Mean per plant No. of new leaves Total No. of Type of seed hole produced during the year green leaves A 1 x 1 X 1 feet 10.1 16.0 B 3 X 3 X 3 feet 9 . 9 16.1 C Cruciform 9 .7 16 .3 D Post hole borer 10.1 16.2 Depth of Planting.—The field trial laid down in November 1956 to study the relationship, if any, between depth of planting a coconut seedling and its subsequent growth has completed three years. Leaf characters were recorded as for the previous experiment (Table 6). Here again, the differences between treatments with respect to leaf product ion are no t significant. TABLE 6 Leaf production at the end of the 3rd year in the Depth of Planting Trial Mean per plant No. of new leaves Total No. of Treatment produced during the year green leaves 6 inches deep planting 8 .4 . . 11.7 12 inches deep planting 7 .7 . . 10.8 18 inches deep planting 8 .5 . . 11.5 24 inches deep planting 8 .3 . . 11.4 43 Hedge Planting.—The observation plots of this new system of planting have completed three years. It is yet premature to make any observations. 5. Miscellaneous. Rout ine observations and yield recording of the Latin Square Selection Experiment and other trials were continued. Isolated Seed Garden.—Five acres were planted with selected dwarf palm seedlings in December 1959 for the mass production of Dwarf X Tall hybrid seed. When the dwarf palms flower, it is proposed to emasculate them systematically, so that their female