P l a n t a n d Soil 41, 459-470 (1974)

THE

Ms. 2325

MINERAL COMPOSITION OF THE ROBUSTA BANANA PLANT

II. T H E CONCENTRATION OF M I N E R A L CONSTITUENTS by I. T. T W Y F O R D and D. W A L M S L E Y Windward Islands Banana Research Scheme, St. Lucia and University of the West Indies, Trinidad SUMMARY The concentrations (per c e n t o v e n - d r y material) of nitrogen, phosphorus, potassium, calcium and m a g n e s i u m in the different organs of t h e R o b u s t a b a n a n a plant, at various stages in t h e g r o w t h cycle are presented. The plants were sampled at sites w i t h differing soil properties and where fertilizer practices varied. The p a t t e r n of n u t r i e n t concentrations was similar at all sites. The effect of t h e v a r i a b i l i t y within sites of organ c o n c e n t r a t i o n on the choice of a diagnostic tissue is discussed. The changes of n u t r i e n t concentrations in the organs which t a k e place during t h e life cycle are described in relation to physiological function. INTRODUCTION

Part I of this study 10 dealt with the growth and production of fresh and dry matter b y the Robusta plant at several locations in the Eastern Caribbean. This paper discusses the distribution and concentration of mineral constituents ill the same plants, the changes which take place during the growth cycle and the influence of site factors on these. A description of the sites, the stages of growth at which the bananas were sampled and the methods of analysis were given in Part I 10 RESULTS AND DISCUSSION

Nutrient concentrations and changes in nutrient concentrations with age Although the actual concentrations of nutrients varied from site to site the pattern of nutrient concentrations was similar at all sites.

I. T. TWYFORD AND D. WALMSLEY

460

As an e x a m p l e of this p a t t e r n the d a t a for G r e n a d a are p r e s e n t e d in Fig. 1. 1. N i t r o g e n . A t all sites there was m a r k e d similarity in the distribution p a t t e r n a m o n g s t t h e different p l a n t parts. A t t h e sucker stage, the s h o o t was richer in nitrogen t h a n the corm.

%N 0

I

i

i

0"5 0"4 %P

0"5 0"2 0"1 0 9 8 7 6

%K

5 4 :3 2 I 0

2°f 1.5

% ~ ll"0

rr S

UL P M UL P M C L Ps C L PsC Sucker Smoll Lorge

L Ps I P C IS Shooting

rm mTn-n

L Ps I ES P C IS F Shof

L Ps I ES P C IS F Horvest

Fig. 1. Nutrient concentrations (per cent oven dry material) in various tissues of the Robusta banana at different stages of growth at the Grenada site: S-shoot; UL-unemerged leaf tissue; L-emerged leaves; P-petioles; Ps-pseudostem; M-meristem; C-corm; I-inflorescence; IS-internal stalk; ES-external stalk; F-fruit.

COMPOSITION OF BANANA I I

461

Throughout the vegetative phase, the unemerged leaves had by far the highest concentration, followed b y leaves and meristem. During the fruiting phase, the inflorescence was richest in nitrogen; leaves and the other fruiting parts having the next highest concentrations. There was little change in nitrogen concentration with plant age in the unemerged leaf as would be expected in a tissue whose physiological age remains more or less constant. In most other vegetative organs concentrations tended to decrease with age up to shooting after which there was little change. In the corm however, behaviour varied at different sites; at some sites the level remained constant b u t at others it decreased. Inflorescence and internal fruit stalk levels tended to fall with age, as did those of the external fruit stalk except in Grenada plants. As the fruits matured nitrogen levels fell. It is interesting that this pattern is little affected b y site differences and is, in general, in agreement with the work of M o n t a g u t et al. v except that in those trials the corm concentration was remarkably constant both with plant age and from site to site. 2. P h o s p h o r u s . The distribution of phosphorus concentrations was remarkably similar from site to site. At the sucker stage, shoot had a higher concentration than corm. At the small and large stages, unemerged leaf had a very much higher concentration than other parts, followed b y meristem and then usually leaves. In the fruiting phase the inflorescence had a much higher phosphorus concentration than any other organ, followed b y the fruit stalk. At all stages of growth, corm and petiole concentrations were generally lowest. As for nitrogen, the age of plant did not affect unemerged leaf phosphorus levels but meristem concentration fell with age. The concentration in leaves tended to fall as the plant got older, particularly from the small to the large stages; after shooting there was little change. Petiole, pseudostem and corm concentrations changed little during the life of the plant but with some tendency to fall with increasing age. Concentration in inflorescence, internal fruit stalk and fruits decreased with age at 'all sites as it did in external fruit stalk except in Grenada. Site effects in this pattern were few, as was also observed for nitrogen and again this general pattern is in agreement with that reported by M o n t a g u t et al. 7

462

I. T. T W Y F O R D AND D. WALMSLEY

3. P o t a s s i u m . The distribution pattern of potassium concentrations in the various organs was consistent at all the sites. At the sucker stage, shoot was richer in potassium than the corm. In later vegetative stages, unemerged leaf was always highest in potassium; next were meristem, pseudostem, petioles, leaves and corm seriatim. In the fruiting phase the richest tissues were fruit stalk and inflorescence. At some sites (Roseau and Cul-de-Sac) fruits were quite rich in potassium but all other organs had low and variable concentrations with the corm always poorest. In the vegetative phase, the concentration of potassium in all organs dropped with age except that of the meristem in Grenada plants which tended to rise. After shooting, concentration in leaves, fruits, pseudostem and inflorescence continued to decrease although in the last only slightly. The potassium concentration in petioles generally decreased from shooting to shot but thereafter behaviour varied with site; sometimes there was no further change and sometimes a slight rise. However, the concentration in the corm and external fruit stalk increased at all sites with advancing maturity as it also did in the internal fruit stalk except in St. Vincent. In contrast, in the French experiments 7 it was found that potassium concentrations fell throughout the cycle and especially after shooting. 4. C a l c i u m . The calcium concentration pattern was substantially similar at all sites. At all stages of growth petioles, leaves and pseudostem were richest in calcium except at tile small stage where meristem usually was highest. All other tissues had low calcium concentrations. The effect of age on the calcium concentrations in the various organs was small and variable. After shooting, leaf concentration always rose somewhat, but before shooting there was little age effect. With increasing age concentrations in petioles, internal and external fruit stalk tended to rise and in meristem and in fruits to decrease. Inflorescence, corm and pseudostem gave variable behaviour with age at different sites but the changes were all very small. This is in general agreement with M 0 n t a g u t et al. 7 who state that calcium concentrations increase with age especially at the end of the

C O M P O S I T I O N OF B A N A N A II

463

cycle, calcium going to tissues of reduced activity where it replaces other cations especially potassium. 5. M a g n e s i u m . There were not very marked differences in concentration from one organ to another except in Trinidad. Meristem tended to be richest in magnesium in the vegetative phase and from the shot stage, pseudostem, internal fruit stalk, petioles and corm tended always to be richest and the fruits the poorest. As the plant got older, there was no substantial change in magnesium concentration in the unemerged leaves, leaves and the external fruit stalk. On the other hand concentrations in petioles, pseudostem and corm tended to rise, especially after shooting. This was also observed in the French experiments 7. There was a slight upward trend in internal fruit stalk but a downward tendency in inflorescence and fruits and no consistent pattern with meristem.

Concentration o/ nutrient elements in the plant organs in relation to diagnosis In selecting a part of any plant as a statable tissue for diagnostic purposes it is important that the concentration in the sample tissue should reflect the nutritional level of the whole plant. At any site which is assumed to be of uniform fertility there should not be wide variations of concentration in the tissue chosen for analysis. In the present series of experiments the variability of organ nutrient concentrations within sites differed greatly depending on the element. This can be seen in Table 1 which gives the coefficients of variation at the St. Vincent site. The concentrations in some organs were less variable than others, notably unemerged leaves and inflorescence. Leaves, which are usually used for sampling, were more variable which is unexpected since the samples here were composites of all the leaves. However, the usual diagnostic tissue used in the West Indies 9 is the fourth leaf lamina sampled at the large stage and this was less variable for all elements than composite leaves or any other plant part at this growth stage. At other growth stages this did not hold, particularly for cations. The richest tissues do not necessarily give an indication of the status of the whole plant. M a r t i n - P r 6 v e l and M o n t a g u t 5 have found that the lamina is the least sensitive organ b u t is it likely to be the most reliable along with the corm. However, in practice this

464

I. T. T W Y F O R D A N D D. W A L M S L E Y TABLE 1

Coefficients of v a r i a t i o n (per cent) of the n u t r i e n t concentration ii~ various parts of 10 Robusta b a n a n a plants sampled at the same site in St. Vincent Stage of growth

Organ

N

P

K

Ca

Mg

Small

4th leaf l a m i n a Unemerged leaf Leaves Petioles Pseudostem Meristem Corm

9 6 9 19 32 18 32

8 8 12 27 17 13 17

29 17 26 18 19 18 16

29 20 22 21 12 21 21

22 8 18 24 34 35 22

Large

4th leaf l a m i n a Unemerged leaf Leaves Petioles Pseudostem Meristem Corm

7 7 14 13 30 44 34

3 8 12 26 30 29 20

12 15 15 24 35 25 32

6 29 16 21 21 41 23

7 11 18 19 27 29 43

Shooting

4th leaf lamina Leaves Petioles Pseudostem Corm Inflorescence I n t e r n a l stalk

8 14 14 25 16 6 18

7 12 19 45 9 7 23

41 10 31 36 28 7 14

16 16 22 20 14 24 32

11 12 22 18 25 17 26

Shot

4th leaf l a m i n a Leaves Petioles Pseudostem Corm Inflorescence Internal stalk External stalk Fruit

6 11 13 t8 31 5 27 13 18

7 9 21 43 24 6 13 11 13

18 19 29 25 26 7 7 8 17

13 23 13 14 22 11 16 12 34

38 17 30 29 32 10 27 13 19

Harvest

4th leaf l a m i n a Leaves Petioles Pseudostem Corm Inflorescence Internal stalk E x t e r n a l stalk Fruit

12 6 3 16 12 11 23 23 6

8 3 60 69 14 11 10 10 12

16 8 25 21 19 5 15 9 13

24 17 10 8 11 11 15 8 20

19 8 25 41 31 17 40 5 9

COMPOSITION OF B A N A N A II

465

latter is difficult to sample. They conclude that for nitrogen and phosphorus the lamina is the best index but for tile cations tile organs of translocation, midrib and petioles, should also be sampled. At the sites studied it was possible to compare the concentration of nutrients in various organs with total plant content of that particular nutrient. In general it was found that concentrations in any of the organs did not reflect the nutrient status of the whole plant. For example, the concentrations of potassium in the leaves (3.0 per cent) or petioles (3.2 per cent) at the large stage were the same for both the St. Vincent and Grenada sites although the plants at the latter site had taken up twice as much potassium (209.9 g compared with 107.6 g).

The pattern o/ nutrient concentration in relation to plant growth and physiological/unction As the plant develops from initiation through the sucker stage to full vegetative growth, a pattern of nutrient concentration in the different tissues emerges which is remarkably consistent at the various sites and therefore m a y be taken to be representative of the fundamental vegetative growth pattern of the banana plant. It would seem that the plant, whatever its nutritional environment, at least over the range encountered, tends to develop in a uniform manner. S u c k e r . As was shown in Part I 10, the corm develops from the bud faster than the aerial parts only up to the sucker stage. Then the aerial parts burgeon and it was found that major nutrients concentrated there. S u c k e r t o s m a 11. At the small stage, tile most rapidly developing tissues are unemerged leaves and meristem and unemerged leaves were richest in nitrogen, phosphorus and potassium. In unemerged leaves the same concentration of magnesium was maintained, even though the magnesium status of the plants was very different from one site to another, implying that an essential magnesium concentration (about 0.4 per cent) is required before tile plant develops normal leaves. This is in accordance with the observations of M a r t i n - P r 6 v e l and M o n t a g u t 6. Meristematic tissue was relatively richest in magnesium, second richest in phosphorus and

466

I. T. T W Y F O R D A N D D. W A L M S L E Y

potassium, being exceeded by unemerged leaves and third richest in nitrogen being exceeded by both emerged and unemerged leaves. Therefore, in the formation of new leaf tissue from meristem, phosphorus, potassium and particularly nitrogen were concentrated whereas magnesium and especially calcium were diluted. Unemerged leaf tissue differentiates into leaf sheath (pseudostem), petioles and emerged leaves. During leaf formation nitrogen concentration fell less than that of phosphorus and very much less than that of potassium. Calcium however was concentrated, and magnesium changed but little. In petiole formation and development, calcium and potassium concentrations were relatively high. However, the potassium would be in solution and presumably therefore its concentration partly reflects the establishment of the processes of transport. The formation and development of pseudostem seems to be associated with low concentrations of nitrogen and phosphorus. Potassium concentration was somewhat higher again partly reflecting the transport process. Calcium was relatively high in pseudostem and presumably this element is important for cell strength in an organ which supports the plant. The differentiation of corm tissue is associated with low levels of all major nutrients and at this early stage little storage has begun. S m a l l to large. As the plant grows to the large stage, the pattern of nutrient concentrations remained the same for nitrogen, phosphorus and potassium. Calcium however, was becoming more clearly concentrated in leaves, pseudostem and petioles. Magnesium was still most Concentrated in meristem and least in unemerged leaf tissue although the concentration was maintained at a level adequate for production of new leaves. At this stage, in spite of low corm concentrations, due to the size of the organ, substantial storage had in fact occurred. Between the small and the large stages of growth, every organ, except in a sense the unemerged leaf tissue, increases greatly in size. In spite of this, magnesium concentration actually increased in petioles, pseudostem and corm; calcium increased in petioles and also in leaves at the sites of better growth. L a r g e to s h o o t i n g . At the large stage the meristem is in the

COMPOSITION OF BANANA n

467

process of differentiating fruit primordia s and this was associated with a dimunition of its nutrient concentration, except for magnesium. Presumably this was a reflection of the fact that the meristem had grown to almost its final size and had nearly completed its function in vegetative tissue production. At earlier stages, the meristem is not only producing the vegetative parts of the plant but is also developing its own size, which is crucial for later fruit formation. This complexity of function was associated with high concentrations of nitrogen, phosphorus, potassium and calcium. By the shooting stage the functions of the meristem and unemerged leaf are completed and no new leaf, petiole and pseudostem tissue is formed. At shooting the centre of the pseudostem is occupied by internal fruit stalk to which is attached the inflorescence. The pattern of nutrient concentrations which appeared at shooting was very similar to that at the large stage for comparable organs. The organs of rapid development and differentiation at this stage are now the inflorescence and to a lesser extent the fruit stalk; these then took over the rank of nutrient concentration formerly held by unemerged leaves and meristem respectively. The inflorescence is now the growing point of the plant and is associated with a high nutritional level. The internal fruit stalk is the organ through which the food supply moves and it too was rich in nutrients. In the organs which were present both before and after shooting, magnesium concentration increased. Calcium also increased in the petioles and leaves but for the latter only at the sites where growth was good. S h o o t i n g to s h o t . From shooting to shot, no new vegetative material, except corm, has been produced and the sites of new growth are external fruit stalk, fruits and inflorescence, the last two being actively differentiating tissues. The inflorescence was the richest in nitrogen, phosphorus and potassium but only moderate in calcium and low in magnesium. The external fruit stalk, which supplies nutrients to the inflorescence and fruits was also rich in nitrogen, phosphorus and potassium but now potassium predominated, followed by phosphorus and then nitrogen. The implication is that potassium is needed in substantial amounts for fruit development, phosphate somewhat less and nitrogen less again. For internal fruit stalk, the same pattern is seen but at lower concentrations.

468

I. T. T W Y F O R D A N D D. W A L M S L E Y

Calcium and magnesium concentrations were low in all these organs. It is notable that magnesium was high in concentration rank in the most actively differentiating tissues up to shooting but then fell very considerably. Fruits were relatively low ranking in nutrient concentration but this is hardly surprising since tile fruit as analysed contains a great deal of pulp which is very low in minerals s, only the skin of the fruit containing the vascular bundles. In the vegetative parts of the plant, the order of nitrogen, phosphorus and potassium concentration was similar to that at earlier growth stages except for nitrogen in the leaves. Clearly the leaves are important for active photosynthesis after shooting to supply protein to the fruiting organs, a function which would be interrupted if, as sometimes happens in commercial banana growing, leaves are lost after shooting, e.g. by disease or pruning. Calcium was concentrated in petioles, leaves and pseudostem and magnesium also increased its concentration rank in these organs. S h o t to h a r v e s t . The final stage of growth from shot to harvest does not involve fundamental changes of function of the various parts. Fruits have greatly increased in size and so has external fruit stalk. On tile other hand, the inflorescence, whilst still actively producing new tissue, is much smaller than before, due to loss of male flowers and bracts and its active life is now over. The major nutrient distribution pattern also showed little change. Inflorescence, as would be expected, was still richest in nitrogen and phosphorus and of the same rank in calcium and magnesium but its potassium status was now distinctly not the highest in the plant, this place being occupied by external fruit stalk, followed by internal fruit stalk. Clearly potassium is chiefly needed in fruit development at this stage. Concentrations of all nutrients in fruits remained as low as at the shot stage. Among the vegetative organs, the nutrient concentration pattern did not alter. Nitrogen was still high in leaves reflecting a continuance of leaf function to the end of tile plant's life. The final stage of tile life of the plant was accompanied by rises in potassium concentration in both parts of the fruit stalk and corm, calcium in leaves, petioles and both parts of tile fruit stalk and magnesium in petioles, pseudostem and corm.

C O M P O S I T I O N OF B A N A N A II

469

CONCLUSIONS

The concentration patterns in the organs of plants growing on all the sites chosen were very similar and comparable with those reported by M o n t a g u t et al. 7 from work in the French Antilles. The only case where there was not complete agreement was in the ranking for potassium concentrations in the vegetative phase. The French workers found that pseudostem was always richest followed by petioles > corm >~ leaves but they did not treat the meristematic tissue separately and this had the highest concentration in the plants analysed in this study. Other reported results of banana analyses 1 2 4 are not directly comparable since different varieties were used. Concentrations in the centre portions of the fourth leaf lamina sampled at the large stage were less variable than in any other plant part for all nutrient elements. This confirms its value as a diagnostic tissue. It clearly emerged that the rapidly differentiating tissues in the banana throughout its life cycle are always associated with, and therefore presumably dependent on, high concentrations of particular nutrients but these nutrients are not always the same ones. Potassium is outstanding in being abundantly involved in all of them, especially fruitings organs. Phosphorus and nitrogen seem to be of great importance in most tissues but not so much those involved in fruit development (nitrogen even less so than phosphorus). Calcium and magnesium appear at high levels only in organs in the vegetative phase, especially meristem. Relatively high concentrations of these two elements are found also, however, at all stages of growth in support, transport and storage organs, increasingly so as the plant ages. The importance of high nitrogen in leaves throughout the plant's life is clear. Received September 6, 1973

470

C O M P O S I T I O N OF B A N A N A II

REFERENCES 1

2 3 4 6 6 7 8 9

10

B a i l l o l l , A. F., H o l m e s , E., and L e w i s , A. H., The composition of a n d nutrient uptake b y the b a n a n a plant with special reference to the Canaries. Trop. Agric. (Trinidad) 10, 139-144 (1933). B o l a n d , D., B a n a n a Board Research Department, Jamaica. Ann. Rept 1961, p. 10 (1962). C h a m p i o n , J., Le Bananier. Maisonneuve et Larose, Paris (1963). J o se p h, K. T., Nutrient content a n d nutrient removal in b a n a n a s as an initial guide for assessing fertilizer needs. The Planter. 47, (No. 538), 7-10 (1971). M a r t i n - P r 6 v e l , P. a n d M o n t a g u t , G., Fonetions de divers organes dans l'assimilation de P, K, Ca et Mg. Fruits 21, 395-416 (1966). M a r t i n - P r 6 v el, P. and M o n t a g u t, G., Less interactions dans la nutrition min6rale du bananier. Fruits 21, 19-36 (1966). M o n t a g u t , G., M a r t i n - P r 6 v e l , P. and L a c o e u i l h e , J-J., Nutrition minerale compar6e dans six essais. Fruits 20, 398-410 (1965). S u m m e r v i l l e , W. A. T., Nutrition and development in the banana. Queensland J. Agr. Soc. 1, 1-127 (1944). T w y f o r d , I. T. and C o u l t e r , J. K., Foliar diagnosis in b a n a n a fertilizer trials. Proe. 4th Colloq. on Plant Analysis and Fertilizer Problems, pp. 357-370. Brussels. (1964). T w y f o r d , I. T. and W a l m s l e y , D., The mineral composition of the R o b u s t a b a n a n a plant. I. Methods and plant growth studies. Plant and Soil 39, 227-243 (1973).