its appearance at the top of the stem being preceded by the evolution 

 of one or more leaves smaller than the rest." 



Observations on this point have been made lately at Kew. (1.) A 

 stem of M. sapientum, about 12 months old, was cut down before 

 flowering when nearly, but not quite, fully grown. The whole stem 

 was 15 feet high. Cut longitudinally, it showed at the base a conical 

 body rising in the centre about 8 inches above the attachment of the 

 outermost leaves. From the apex of the cone the flower bud had 

 already started. It was on a slender leafless stalk about an inch in 

 diameter. The bud itself was found about 4 inches higher. In this 

 case it had evidently just begun to grow. (2.) In a plant of M. Basjoo, 

 apparently fully grown, the conical axis was 10 inches in diameter at 

 the base, invested at that point by a few leaf -sheaths. A foot higher 

 with the stem of leaf -sheaths 9 inches in diameter, the flower stalk was 

 an inch and a half in diameter. By carefully following it, the top of 

 the flower bud was found at 3 feet from the base, forming a club- 

 shaped body easily recognised by a bulging out of the innermost 

 leaf-sheaths. Here the flower bud was found about one-third of the 

 way up the stem. (3.) Finally a stem of M. sapientum that had 

 already borne fruit was examined. It was about 12 feet high. At a 

 foot above the conical base the fruiting stalk, cut through, was one and 

 three-quarter inches in diameter. It preserved the same diameter and 

 was traced as a slender, white, fibrous body, over 12 feet high, closely 

 invested by the leaf -sheaths, until it emerged from amongst the petioles 

 of the leaves. It then became coloured green and curved downwards. 



These observations fully confirm what has just been quoted from 

 Dr. Graham. They show that the rate of growth of the flower bud 

 must be very rapid. In the tropics where the whole plant matures and 

 ripens its fruit within 12 months, the flower bud probably takes only 

 a few weeks to push its way from the base to the top of the stem. From 

 the time of flowering until the fruit is ripe takes about six to eight weeks. 

 In the non-stoloniferous species the whole plant is strictly monocarpic, 

 and reproduces itself not by shoots but by seeds.* The majority of 

 species are, however, reproduced by buds or shoots which spring from 

 the perennial rootstock. 



The shape of the pseudo-stem varies a good deal. In one section 

 the stems are bottle-shap?d, having a swollen distended appearance. 

 This is very noticeable in M. superba. Usually the stem is cylindrical, 

 gradually tapering from the base upwards. In M. Ensete the stem 

 is 6 to 10 feet high, andjvery stout. In M. Hillii it is about 30 feet high, 

 and moderately stout only. Again, in M. Mann if, it is only 2-3 feet high 

 and 1 inch in diameter. 



The stems of most species are green, with occasional blotches of 

 black, red, or purple. Those in the banana (M. sapientum) are often 

 described as green and "purple spotted" ; while in the plantain (variety 

 M. pai adisiaca) they are said to be wholly "green." These characters 

 are, however, not constant. The stem of M. {sapientum) d%cca is 

 " pruinose," and appears covered with glittering particles of fine dew. In 

 M. {sapientum) rubra the stem is dull-red, while the interesting M 

 Felvi has a greenish stem with violet juice. 



* Kurz, however, remarks that even in non-stoloniferous species shoots are 

 occasionally produced. "I remember," he says, "a plant of Musa glauca in the 

 Botanical Gardens at Java, which threw out two shoots ; and if accounts 

 be correct M. Ensete is said to make shoots if the whole plant is cut down before 

 flowering." 



