148 
Notes . 
Of 49 inflorescences from one tree, there were 30 normal, 15 with three female 
flowers per cupule, 2 with four female flowers per cupule, 4 androgynous. 
Of 106 inflorescences from another tree, 99 were normal, 1 with only one 
female flower, 5 androgynous, 1 transitional. So that of 355 inflorescences examined, 
17 per cent, were abnormal in some way. It will be noted that in the case where 
the laterals'were kept separate from the terminals, 25 per cent, of the former bore 
male dowers as compared with 8 per cent, in the latter. 
The extra female flowers of the first group are most commonly situated at, or 
slightly above, the level of the normal pair. They are of exactly the same structure 
Fig. 1. 1, normal female flower; 2, rudimentary hermaphrodite flower with inferior ovary; 
3, male flower with vestigial pistil, the latter having a superior ovary and two stylar arms ; 4, male 
flower with androecium only; 5, vertical section of androgynous inflorescence with rudimentary 
hermaphrodite flowers on inside lateral walls of cupule and male flowers terminally on same ; 
6, 7, 8, 9, diagrams of normal and abnormal female inflorescences showing relation of cupule 
segments to position and number of flowers (shaded) ; 10, vertical section of inflorescence whose 
diagram is represented in 9, taken through plane xy —vascular system as seen macroscopically, shaded ; 
11, diagram of theoretical inflorescence with seven possible flowers of dichasium present, see text. 
0. flower on upgrowth from centre of receptacle. 
as these, although usually rather smaller in size. Regularly associated with the 
presence of one or two of such extra flowers is the further segmentation of the 
cupule. If three flowers are present the cupule is always five-partite, or with four 
flowers six-partite, as compared with the normal four segments. 
From the diagrams 6, 7, 8, 9, Fig. 1, it will be seen that the place of fission of 
the cupule is to be correlated with the orientation of the flowers. It is usually the 
case that fission caused by the presence of an extra flower does not proceed so far 
towards the base of the cupule as do the normal four slits. In the normal inflorescence, 
however, the four fissures of the cupule do not extend to the same depth ; the two 
situated in the plane of the axis seem invariably to extend the deeper. Whether this 
extra splitting of the cupule is a result of pressure on the part of the developing 
flowers, or whether it is due to some process of dialysis, remains obscure. 
