341 
as a rule, exceed four feet in length. Being a very accessible and at the 
same time a very common plant, no wanderer in the mountain forests 
can fail to observe its inflorescence. This consists of greenish or pur- 
plish flask-shaped organs, resembling huge unopened buds, growing 
upright in groups upon the rhizome, at the bases of the leafstalks. 
The flower case has about the size and very much the shape of a 
sixteen ounce Florence flask. When cut open and examined it is 
found to consist of a thick and leathery spathe wrapped in a spiral 
about an upright, cylindrical spadix. The enveloping spathe tightly 
clasps in its embrace the upper, pollen-producing portion of the spadix, 
but expanding below, leaves the fruiting portion free, in a cavity which 
is partially filled with a mucilaginous liquid. All evaporation^ pre- 
vented by the overlapping of the spathe, and the floral organs thus 
hermetically sealed within the flask would seem to be destined to self- 
fertilization most rigidly enforced. Indeed it is difficult to conceive 
how any fertilization could be accomplished by the plant itself, since 
the pollen tubes of the spadix, being tightly enrolled by the inner folds 
of the spathe, are unable to give forth their fertilizing grains. 
During a recent visit to the island of Moutserrat, one of the Leeward 
group of the Lesser Antilles, I had occasion to observe that the matur- 
ing flowers of this plant are infested with numerous larva? of sap- 
loving beetles and flies, which swarm in the flower-cases, feeding upon 
the envelope and breaking it down, until at last the ripening fruits sur- 
rounding the base of the spadix are stripped of their covering and stand 
exposed, to be carried away by birds and other agencies which aid in 
the dissemination of the seed. The immature inflorescence, however, 
contained at first no insects, but in every case, sooner or later, a 
brownish spot, caused apparently by a rot-fungus, made its appearance 
upon the exterior of the flower-case (spathe). It was remarked that 
the fungus spot invariably occupied the same position, occurring 
always at the extreme outer edge of the overlapping spathe, opposite a 
deep sinus which cuts into the margin. The accompanying diagrams, 
although they do not attempt to reproduce exactly the structure of the 
floral organs, will serve to illustrate the general features of the inflo- 
rescence. Fig. 33 a represents the flask-like spathe upon which the 
fungus spot will be seen in process of development. Fig. 33 b gives 
a section of the same, showing the internal cavity, and the spadix with 
its base immersed in liquid, the surface of which is indicated by a 
horizontal line.* 
* Fig. 33 b is reproduced from rough field notes, and does not present an actual 
section of the spathe, in which the upper male portion of the spadix is enrolled 
within the inner flap of the spathe. The upper portion of the inflexed spat In- goes 
more than once around the spadix, which does not come in contact with the outer 
walls as shown in the figure. Of the double envelope thus formed, the inner walls 
should descend around the spadix very nearly to the surface of the liquid. 
12283— No. 4 5 
