25 
Api Api (Sumatra) with the information that the tree itself 
had sprung up from a nut of a tree at Siak, which also bore 
unbranched spadices. 
These facts render it very probable that we have to do here 
with a hereditary variety. 
As to the appearance of a branch (fig. 28) it should be added 
that the Buitenzorg specimens produce it now and then and 
also the tree of Malang. The specimens of Siak and Bagansi 
Api Api on the other hand are devoid of such an appendage. 
6) Legit J. A. Boyes, Tebing Tiaggi, Deli. 
Transition of inflorescences to leafy branches, fig. .30. 
Dr. Smita has sent two inflorescences of an extraordinary 
length for examination and comparison with his own observa- 
tion '). One specimen has a length of 1.9 M., the other one 
was divided into several pieces of which the dimensions and 
peculiarities are the follow. The basal portion of this inflores- 
cence — 7 d.M. long — is covered with 23 sheaths which firmly 
embrace one another and show a length of about 1—1}4 M. and 
a width of about 1 d.M. at the base; from the base they taper 
gradually and terminate with a rather blunt knitting-needle. 
Some of them branch off linear lobes, indication of a pinnate. 
leaf. The next section, 5.5 d.M. long bears numerous sheaths, 
most of these ending in pinnate blades. The third portion is the 
inflorescence proper enveloped in a great many linear-partly 
coalescing-sheaths. 
That in this way the growth of the partly transformed inflores- 
cence was put to and end goes without saying. Forbes who as 
early as 1879 described such a metamorphose’) did not find ~ 
any flowers and we see from his figures that the branches con- 
tinue their growth. 
c) Legit W. F. G. Dens, Makasar. 
The @ flowers show foliar prolification, both terminal and 
lateral. Fig. 31 shows a clear effort to transform 9 flowers 
into leafy branches. The spike produces branches in various 
1) Teysmannia 1919, p. 291. 
2) Journal of botany XVII, p- 193 sqq- 
